From 6859a8402945cf1d74af75a2e1aa4e327a506ab4 Mon Sep 17 00:00:00 2001
From: Alan Mayer <ajm@sgi.com>
Date: Wed, 26 Mar 2008 16:11:31 -0500
Subject: x86: resize NR_IRQS for large machines

On machines with very large numbers of cpus, tables that are dimensioned
by NR_IRQS get very large, especially the irq_desc table.  They are also
very sparsely used.  When the cpu count is > MAX_IO_APICS, use MAX_IO_APICS
to set NR_IRQS, otherwise use NR_CPUS.

Signed-off-by: Alan Mayer <ajm@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/kernel_stat.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index e8ffce898bf9..cf9f40a91c9c 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -1,11 +1,11 @@
 #ifndef _LINUX_KERNEL_STAT_H
 #define _LINUX_KERNEL_STAT_H
 
-#include <asm/irq.h>
 #include <linux/smp.h>
 #include <linux/threads.h>
 #include <linux/percpu.h>
 #include <linux/cpumask.h>
+#include <asm/irq.h>
 #include <asm/cputime.h>
 
 /*
-- 
cgit v1.2.3


From 988f7b5789ccf5cfed14c72e28573a49f0cb4809 Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Tue, 18 Mar 2008 17:00:22 -0700
Subject: x86: PAT export resource_wc in pci sysfs

For the ranges with IORESOURCE_PREFETCH, export a new resource_wc interface in
pci /sysfs along with resource (which is uncached).

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/filesystems/sysfs-pci.txt |  1 +
 drivers/pci/pci-sysfs.c                 | 84 ++++++++++++++++++++++++---------
 include/linux/pci.h                     |  1 +
 3 files changed, 64 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/sysfs-pci.txt b/Documentation/filesystems/sysfs-pci.txt
index 5daa2aaec2c5..68ef48839c04 100644
--- a/Documentation/filesystems/sysfs-pci.txt
+++ b/Documentation/filesystems/sysfs-pci.txt
@@ -36,6 +36,7 @@ files, each with their own function.
        local_cpus	   nearby CPU mask (cpumask, ro)
        resource		   PCI resource host addresses (ascii, ro)
        resource0..N	   PCI resource N, if present (binary, mmap)
+       resource0_wc..N_wc  PCI WC map resource N, if prefetchable (binary, mmap)
        rom		   PCI ROM resource, if present (binary, ro)
        subsystem_device	   PCI subsystem device (ascii, ro)
        subsystem_vendor	   PCI subsystem vendor (ascii, ro)
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 271d41cc05ab..9ec7d3977a82 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -489,13 +489,14 @@ pci_mmap_legacy_mem(struct kobject *kobj, struct bin_attribute *attr,
  * @kobj: kobject for mapping
  * @attr: struct bin_attribute for the file being mapped
  * @vma: struct vm_area_struct passed into the mmap
+ * @write_combine: 1 for write_combine mapping
  *
  * Use the regular PCI mapping routines to map a PCI resource into userspace.
  * FIXME: write combining?  maybe automatic for prefetchable regions?
  */
 static int
 pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
-		  struct vm_area_struct *vma)
+		  struct vm_area_struct *vma, int write_combine)
 {
 	struct pci_dev *pdev = to_pci_dev(container_of(kobj,
 						       struct device, kobj));
@@ -518,7 +519,21 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
 	vma->vm_pgoff += start >> PAGE_SHIFT;
 	mmap_type = res->flags & IORESOURCE_MEM ? pci_mmap_mem : pci_mmap_io;
 
-	return pci_mmap_page_range(pdev, vma, mmap_type, 0);
+	return pci_mmap_page_range(pdev, vma, mmap_type, write_combine);
+}
+
+static int
+pci_mmap_resource_uc(struct kobject *kobj, struct bin_attribute *attr,
+		     struct vm_area_struct *vma)
+{
+	return pci_mmap_resource(kobj, attr, vma, 0);
+}
+
+static int
+pci_mmap_resource_wc(struct kobject *kobj, struct bin_attribute *attr,
+		     struct vm_area_struct *vma)
+{
+	return pci_mmap_resource(kobj, attr, vma, 1);
 }
 
 /**
@@ -541,9 +556,46 @@ pci_remove_resource_files(struct pci_dev *pdev)
 			sysfs_remove_bin_file(&pdev->dev.kobj, res_attr);
 			kfree(res_attr);
 		}
+
+		res_attr = pdev->res_attr_wc[i];
+		if (res_attr) {
+			sysfs_remove_bin_file(&pdev->dev.kobj, res_attr);
+			kfree(res_attr);
+		}
 	}
 }
 
+static int pci_create_attr(struct pci_dev *pdev, int num, int write_combine)
+{
+	/* allocate attribute structure, piggyback attribute name */
+	int name_len = write_combine ? 13 : 10;
+	struct bin_attribute *res_attr;
+	int retval;
+
+	res_attr = kzalloc(sizeof(*res_attr) + name_len, GFP_ATOMIC);
+	if (res_attr) {
+		char *res_attr_name = (char *)(res_attr + 1);
+
+		if (write_combine) {
+			pdev->res_attr_wc[num] = res_attr;
+			sprintf(res_attr_name, "resource%d_wc", num);
+			res_attr->mmap = pci_mmap_resource_wc;
+		} else {
+			pdev->res_attr[num] = res_attr;
+			sprintf(res_attr_name, "resource%d", num);
+			res_attr->mmap = pci_mmap_resource_uc;
+		}
+		res_attr->attr.name = res_attr_name;
+		res_attr->attr.mode = S_IRUSR | S_IWUSR;
+		res_attr->size = pci_resource_len(pdev, num);
+		res_attr->private = &pdev->resource[num];
+		retval = sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
+	} else
+		retval = -ENOMEM;
+
+	return retval;
+}
+
 /**
  * pci_create_resource_files - create resource files in sysfs for @dev
  * @dev: dev in question
@@ -557,31 +609,19 @@ static int pci_create_resource_files(struct pci_dev *pdev)
 
 	/* Expose the PCI resources from this device as files */
 	for (i = 0; i < PCI_ROM_RESOURCE; i++) {
-		struct bin_attribute *res_attr;
 
 		/* skip empty resources */
 		if (!pci_resource_len(pdev, i))
 			continue;
 
-		/* allocate attribute structure, piggyback attribute name */
-		res_attr = kzalloc(sizeof(*res_attr) + 10, GFP_ATOMIC);
-		if (res_attr) {
-			char *res_attr_name = (char *)(res_attr + 1);
-
-			pdev->res_attr[i] = res_attr;
-			sprintf(res_attr_name, "resource%d", i);
-			res_attr->attr.name = res_attr_name;
-			res_attr->attr.mode = S_IRUSR | S_IWUSR;
-			res_attr->size = pci_resource_len(pdev, i);
-			res_attr->mmap = pci_mmap_resource;
-			res_attr->private = &pdev->resource[i];
-			retval = sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
-			if (retval) {
-				pci_remove_resource_files(pdev);
-				return retval;
-			}
-		} else {
-			return -ENOMEM;
+		retval = pci_create_attr(pdev, i, 0);
+		/* for prefetchable resources, create a WC mappable file */
+		if (!retval && pdev->resource[i].flags & IORESOURCE_PREFETCH)
+			retval = pci_create_attr(pdev, i, 1);
+
+		if (retval) {
+			pci_remove_resource_files(pdev);
+			return retval;
 		}
 	}
 	return 0;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 509159bcd4e7..d18b1dd49fab 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -206,6 +206,7 @@ struct pci_dev {
 	struct bin_attribute *rom_attr; /* attribute descriptor for sysfs ROM entry */
 	int rom_attr_enabled;		/* has display of the rom attribute been enabled? */
 	struct bin_attribute *res_attr[DEVICE_COUNT_RESOURCE]; /* sysfs file for resources */
+	struct bin_attribute *res_attr_wc[DEVICE_COUNT_RESOURCE]; /* sysfs file for WC mapping of resources */
 #ifdef CONFIG_PCI_MSI
 	struct list_head msi_list;
 #endif
-- 
cgit v1.2.3


From 5a433b3ad497120d80f1045d37bd1a9ce897388f Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Mon, 21 Apr 2008 10:41:10 -0700
Subject: mac80211: remove unnecessary byteshifts in frame control testing

Byteswap the constants rather than the frame_control member.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h | 32 +++++++++++++++++---------------
 1 file changed, 17 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 0b5e03eae6d2..a9102bc78b61 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -552,16 +552,17 @@ enum ieee80211_back_parties {
  */
 static inline u8 *ieee80211_get_SA(struct ieee80211_hdr *hdr)
 {
-	u8 *raw = (u8 *) hdr;
-	u8 tofrom = (*(raw+1)) & 3; /* get the TODS and FROMDS bits */
-
-	switch (tofrom) {
-		case 2:
-			return hdr->addr3;
-		case 3:
-			return hdr->addr4;
+	__le16 fc = hdr->frame_control;
+	fc &= cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS);
+
+	switch (fc) {
+	case __constant_cpu_to_le16(IEEE80211_FCTL_FROMDS):
+		return hdr->addr3;
+	case __constant_cpu_to_le16(IEEE80211_FCTL_TODS|IEEE80211_FCTL_FROMDS):
+		return hdr->addr4;
+	default:
+		return hdr->addr2;
 	}
-	return hdr->addr2;
 }
 
 /**
@@ -577,12 +578,13 @@ static inline u8 *ieee80211_get_SA(struct ieee80211_hdr *hdr)
  */
 static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 {
-	u8 *raw = (u8 *) hdr;
-	u8 to_ds = (*(raw+1)) & 1; /* get the TODS bit */
+	__le16 fc = hdr->frame_control;
+	fc &= cpu_to_le16(IEEE80211_FCTL_TODS);
 
-	if (to_ds)
+	if (fc)
 		return hdr->addr3;
-	return hdr->addr1;
+	else
+		return hdr->addr1;
 }
 
 /**
@@ -595,8 +597,8 @@ static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
  */
 static inline int ieee80211_get_morefrag(struct ieee80211_hdr *hdr)
 {
-	return (le16_to_cpu(hdr->frame_control) &
-		IEEE80211_FCTL_MOREFRAGS) != 0;
+	__le16 fc = hdr->frame_control;
+	return !!(fc & cpu_to_le16(IEEE80211_FCTL_MOREFRAGS));
 }
 
 #endif /* IEEE80211_H */
-- 
cgit v1.2.3


From 09e67ca2c523544e6b38aa570a5f62a0cf20b87b Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Fri, 16 May 2008 11:57:45 +1000
Subject: [POWERPC] Move of_device_get_modalias to drivers/of

Commit 140b932f8cb6cced10b96860651a198b1b89cbb9 ("Create modalias file
in sysfs for of_platform bus") needs this to avoid breaking the sparc
builds.

Just move the code and add whitespace around some binary operators.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/of_device.c | 48 -----------------------------------------
 drivers/of/device.c             | 48 +++++++++++++++++++++++++++++++++++++++++
 include/asm-powerpc/of_device.h |  2 --
 include/linux/of_device.h       |  3 +++
 4 files changed, 51 insertions(+), 50 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/kernel/of_device.c b/arch/powerpc/kernel/of_device.c
index 5748ddb47d9f..e9be908f199b 100644
--- a/arch/powerpc/kernel/of_device.c
+++ b/arch/powerpc/kernel/of_device.c
@@ -89,54 +89,6 @@ struct of_device *of_device_alloc(struct device_node *np,
 }
 EXPORT_SYMBOL(of_device_alloc);
 
-ssize_t of_device_get_modalias(struct of_device *ofdev,
-				char *str, ssize_t len)
-{
-	const char *compat;
-	int cplen, i;
-	ssize_t tsize, csize, repend;
-
-	/* Name & Type */
-	csize = snprintf(str, len, "of:N%sT%s",
-				ofdev->node->name, ofdev->node->type);
-
-	/* Get compatible property if any */
-	compat = of_get_property(ofdev->node, "compatible", &cplen);
-	if (!compat)
-		return csize;
-
-	/* Find true end (we tolerate multiple \0 at the end */
-	for (i=(cplen-1); i>=0 && !compat[i]; i--)
-		cplen--;
-	if (!cplen)
-		return csize;
-	cplen++;
-
-	/* Check space (need cplen+1 chars including final \0) */
-	tsize = csize + cplen;
-	repend = tsize;
-
-	if (csize>=len)		/* @ the limit, all is already filled */
-		return tsize;
-
-	if (tsize>=len) {		/* limit compat list */
-		cplen = len-csize-1;
-		repend = len;
-	}
-
-	/* Copy and do char replacement */
-	memcpy(&str[csize+1], compat, cplen);
-	for (i=csize; i<repend; i++) {
-		char c = str[i];
-		if (c=='\0')
-			str[i] = 'C';
-		else if (c==' ')
-			str[i] = '_';
-	}
-
-	return tsize;
-}
-
 int of_device_uevent(struct device *dev, struct kobj_uevent_env *env)
 {
 	struct of_device *ofdev;
diff --git a/drivers/of/device.c b/drivers/of/device.c
index 8fbfeee53c1e..8a1d93a2bb81 100644
--- a/drivers/of/device.c
+++ b/drivers/of/device.c
@@ -104,3 +104,51 @@ void of_device_unregister(struct of_device *ofdev)
 	device_unregister(&ofdev->dev);
 }
 EXPORT_SYMBOL(of_device_unregister);
+
+ssize_t of_device_get_modalias(struct of_device *ofdev,
+				char *str, ssize_t len)
+{
+	const char *compat;
+	int cplen, i;
+	ssize_t tsize, csize, repend;
+
+	/* Name & Type */
+	csize = snprintf(str, len, "of:N%sT%s",
+				ofdev->node->name, ofdev->node->type);
+
+	/* Get compatible property if any */
+	compat = of_get_property(ofdev->node, "compatible", &cplen);
+	if (!compat)
+		return csize;
+
+	/* Find true end (we tolerate multiple \0 at the end */
+	for (i = (cplen - 1); i >= 0 && !compat[i]; i--)
+		cplen--;
+	if (!cplen)
+		return csize;
+	cplen++;
+
+	/* Check space (need cplen+1 chars including final \0) */
+	tsize = csize + cplen;
+	repend = tsize;
+
+	if (csize >= len)		/* @ the limit, all is already filled */
+		return tsize;
+
+	if (tsize >= len) {		/* limit compat list */
+		cplen = len - csize - 1;
+		repend = len;
+	}
+
+	/* Copy and do char replacement */
+	memcpy(&str[csize + 1], compat, cplen);
+	for (i = csize; i < repend; i++) {
+		char c = str[i];
+		if (c == '\0')
+			str[i] = 'C';
+		else if (c == ' ')
+			str[i] = '_';
+	}
+
+	return tsize;
+}
diff --git a/include/asm-powerpc/of_device.h b/include/asm-powerpc/of_device.h
index 6526e139a463..3c123990ca2e 100644
--- a/include/asm-powerpc/of_device.h
+++ b/include/asm-powerpc/of_device.h
@@ -21,8 +21,6 @@ extern struct of_device *of_device_alloc(struct device_node *np,
 					 const char *bus_id,
 					 struct device *parent);
 
-extern ssize_t of_device_get_modalias(struct of_device *ofdev,
-					char *str, ssize_t len);
 extern int of_device_uevent(struct device *dev,
 			    struct kobj_uevent_env *env);
 
diff --git a/include/linux/of_device.h b/include/linux/of_device.h
index afe338217d91..d3a74e00a3e1 100644
--- a/include/linux/of_device.h
+++ b/include/linux/of_device.h
@@ -24,4 +24,7 @@ static inline void of_device_free(struct of_device *dev)
 	of_release_dev(&dev->dev);
 }
 
+extern ssize_t of_device_get_modalias(struct of_device *ofdev,
+					char *str, ssize_t len);
+
 #endif /* _LINUX_OF_DEVICE_H */
-- 
cgit v1.2.3


From 3e582db0783872a645f5236ef142be6579b300ad Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: Tue, 6 May 2008 12:11:29 -0400
Subject: Input: gameport - remove rescan/reconnect facilities

They have never been used and are unlikely to be used in the future
so remove them altogether.

Reported-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/gameport/gameport.c | 22 ----------------------
 include/linux/gameport.h          |  1 -
 2 files changed, 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c
index c5600ac5feb3..078e4eed0894 100644
--- a/drivers/input/gameport/gameport.c
+++ b/drivers/input/gameport/gameport.c
@@ -36,7 +36,6 @@ EXPORT_SYMBOL(__gameport_register_driver);
 EXPORT_SYMBOL(gameport_unregister_driver);
 EXPORT_SYMBOL(gameport_open);
 EXPORT_SYMBOL(gameport_close);
-EXPORT_SYMBOL(gameport_rescan);
 EXPORT_SYMBOL(gameport_set_phys);
 EXPORT_SYMBOL(gameport_start_polling);
 EXPORT_SYMBOL(gameport_stop_polling);
@@ -230,8 +229,6 @@ static void gameport_find_driver(struct gameport *gameport)
  */
 
 enum gameport_event_type {
-	GAMEPORT_RESCAN,
-	GAMEPORT_RECONNECT,
 	GAMEPORT_REGISTER_PORT,
 	GAMEPORT_REGISTER_DRIVER,
 };
@@ -365,15 +362,6 @@ static void gameport_handle_event(void)
 				gameport_add_port(event->object);
 				break;
 
-			case GAMEPORT_RECONNECT:
-				gameport_reconnect_port(event->object);
-				break;
-
-			case GAMEPORT_RESCAN:
-				gameport_disconnect_port(event->object);
-				gameport_find_driver(event->object);
-				break;
-
 			case GAMEPORT_REGISTER_DRIVER:
 				gameport_add_driver(event->object);
 				break;
@@ -651,16 +639,6 @@ static void gameport_disconnect_port(struct gameport *gameport)
 	device_release_driver(&gameport->dev);
 }
 
-void gameport_rescan(struct gameport *gameport)
-{
-	gameport_queue_event(gameport, NULL, GAMEPORT_RESCAN);
-}
-
-void gameport_reconnect(struct gameport *gameport)
-{
-	gameport_queue_event(gameport, NULL, GAMEPORT_RECONNECT);
-}
-
 /*
  * Submits register request to kgameportd for subsequent execution.
  * Note that port registration is always asynchronous.
diff --git a/include/linux/gameport.h b/include/linux/gameport.h
index afad95272841..f64e29c0ef3f 100644
--- a/include/linux/gameport.h
+++ b/include/linux/gameport.h
@@ -68,7 +68,6 @@ struct gameport_driver {
 
 int gameport_open(struct gameport *gameport, struct gameport_driver *drv, int mode);
 void gameport_close(struct gameport *gameport);
-void gameport_rescan(struct gameport *gameport);
 
 #if defined(CONFIG_GAMEPORT) || (defined(MODULE) && defined(CONFIG_GAMEPORT_MODULE))
 
-- 
cgit v1.2.3


From ccd3474569a101914444dc37cc108eec22578999 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: Tue, 6 May 2008 12:11:39 -0400
Subject: Input: libps2 - remove delayed command execution

Delayed command execution is not used by anyone so let's remove it.

Reported-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/serio/libps2.c | 52 --------------------------------------------
 include/linux/libps2.h       |  1 -
 2 files changed, 53 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c
index b819239d74dc..2b304c22c200 100644
--- a/drivers/input/serio/libps2.c
+++ b/drivers/input/serio/libps2.c
@@ -26,15 +26,6 @@ MODULE_AUTHOR("Dmitry Torokhov <dtor@mail.ru>");
 MODULE_DESCRIPTION("PS/2 driver library");
 MODULE_LICENSE("GPL");
 
-/* Work structure to schedule execution of a command */
-struct ps2work {
-	struct work_struct work;
-	struct ps2dev *ps2dev;
-	int command;
-	unsigned char param[0];
-};
-
-
 /*
  * ps2_sendbyte() sends a byte to the device and waits for acknowledge.
  * It doesn't handle retransmission, though it could - because if there
@@ -245,49 +236,6 @@ int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command)
 }
 EXPORT_SYMBOL(ps2_command);
 
-/*
- * ps2_execute_scheduled_command() sends a command, previously scheduled by
- * ps2_schedule_command(), to a PS/2 device (keyboard, mouse, etc.)
- */
-
-static void ps2_execute_scheduled_command(struct work_struct *work)
-{
-	struct ps2work *ps2work = container_of(work, struct ps2work, work);
-
-	ps2_command(ps2work->ps2dev, ps2work->param, ps2work->command);
-	kfree(ps2work);
-}
-
-/*
- * ps2_schedule_command() allows to schedule delayed execution of a PS/2
- * command and can be used to issue a command from an interrupt or softirq
- * context.
- */
-
-int ps2_schedule_command(struct ps2dev *ps2dev, unsigned char *param, int command)
-{
-	struct ps2work *ps2work;
-	int send = (command >> 12) & 0xf;
-	int receive = (command >> 8) & 0xf;
-
-	if (!(ps2work = kmalloc(sizeof(struct ps2work) + max(send, receive), GFP_ATOMIC)))
-		return -1;
-
-	memset(ps2work, 0, sizeof(struct ps2work));
-	ps2work->ps2dev = ps2dev;
-	ps2work->command = command;
-	memcpy(ps2work->param, param, send);
-	INIT_WORK(&ps2work->work, ps2_execute_scheduled_command);
-
-	if (!schedule_work(&ps2work->work)) {
-		kfree(ps2work);
-		return -1;
-	}
-
-	return 0;
-}
-EXPORT_SYMBOL(ps2_schedule_command);
-
 /*
  * ps2_init() initializes ps2dev structure
  */
diff --git a/include/linux/libps2.h b/include/linux/libps2.h
index f6f301e2b0f5..afc413369101 100644
--- a/include/linux/libps2.h
+++ b/include/linux/libps2.h
@@ -43,7 +43,6 @@ void ps2_init(struct ps2dev *ps2dev, struct serio *serio);
 int ps2_sendbyte(struct ps2dev *ps2dev, unsigned char byte, int timeout);
 void ps2_drain(struct ps2dev *ps2dev, int maxbytes, int timeout);
 int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command);
-int ps2_schedule_command(struct ps2dev *ps2dev, unsigned char *param, int command);
 int ps2_handle_ack(struct ps2dev *ps2dev, unsigned char data);
 int ps2_handle_response(struct ps2dev *ps2dev, unsigned char data);
 void ps2_cmd_aborted(struct ps2dev *ps2dev);
-- 
cgit v1.2.3


From 253b00f1e6619055d524188f254ccb951bffce5d Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Wed, 7 May 2008 11:15:23 -0400
Subject: Input: remove unused defines

Remove unused to_dev, to_handler, to_handle from include/linux/input.h
Move to_handle_h from include/linux/input.h to drivers/char/keyboard.c

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/char/keyboard.c | 2 ++
 include/linux/input.h   | 5 -----
 2 files changed, 2 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/keyboard.c b/drivers/char/keyboard.c
index 7f7e798c1384..d0369e05490a 100644
--- a/drivers/char/keyboard.c
+++ b/drivers/char/keyboard.c
@@ -46,6 +46,8 @@
 
 extern void ctrl_alt_del(void);
 
+#define to_handle_h(n) container_of(n, struct input_handle, h_node)
+
 /*
  * Exported functions/variables
  */
diff --git a/include/linux/input.h b/include/linux/input.h
index 28a094fcfe20..eff711d8a459 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -1213,11 +1213,6 @@ struct input_handle {
 	struct list_head	h_node;
 };
 
-#define to_dev(n) container_of(n, struct input_dev, node)
-#define to_handler(n) container_of(n, struct input_handler, node)
-#define to_handle(n) container_of(n, struct input_handle, d_node)
-#define to_handle_h(n) container_of(n, struct input_handle, h_node)
-
 struct input_dev *input_allocate_device(void);
 void input_free_device(struct input_dev *dev);
 
-- 
cgit v1.2.3


From a33466e31213cd7c46696160d3db78680b58f6a3 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Wed, 7 May 2008 16:30:15 -0400
Subject: Input: gpio-keys debouncing support

Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/keyboard/gpio_keys.c | 89 +++++++++++++++++++++++++++++---------
 include/linux/gpio_keys.h          |  1 +
 2 files changed, 70 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c
index bbd00c3fe98c..be58730e636a 100644
--- a/drivers/input/keyboard/gpio_keys.c
+++ b/drivers/input/keyboard/gpio_keys.c
@@ -26,23 +26,54 @@
 
 #include <asm/gpio.h>
 
+struct gpio_button_data {
+	struct gpio_keys_button *button;
+	struct input_dev *input;
+	struct timer_list timer;
+};
+
+struct gpio_keys_drvdata {
+	struct input_dev *input;
+	struct gpio_button_data data[0];
+};
+
+static void gpio_keys_report_event(struct gpio_keys_button *button,
+				   struct input_dev *input)
+{
+	unsigned int type = button->type ?: EV_KEY;
+	int state = (gpio_get_value(button->gpio) ? 1 : 0) ^ button->active_low;
+
+	input_event(input, type, button->code, !!state);
+	input_sync(input);
+}
+
+static void gpio_check_button(unsigned long _data)
+{
+	struct gpio_button_data *data = (struct gpio_button_data *)_data;
+
+	gpio_keys_report_event(data->button, data->input);
+}
+
 static irqreturn_t gpio_keys_isr(int irq, void *dev_id)
 {
-	int i;
 	struct platform_device *pdev = dev_id;
 	struct gpio_keys_platform_data *pdata = pdev->dev.platform_data;
-	struct input_dev *input = platform_get_drvdata(pdev);
+	struct gpio_keys_drvdata *ddata = platform_get_drvdata(pdev);
+	int i;
 
 	for (i = 0; i < pdata->nbuttons; i++) {
 		struct gpio_keys_button *button = &pdata->buttons[i];
-		int gpio = button->gpio;
 
-		if (irq == gpio_to_irq(gpio)) {
-			unsigned int type = button->type ?: EV_KEY;
-			int state = (gpio_get_value(gpio) ? 1 : 0) ^ button->active_low;
+		if (irq == gpio_to_irq(button->gpio)) {
+			struct gpio_button_data *bdata = &ddata->data[i];
+
+			if (button->debounce_interval)
+				mod_timer(&bdata->timer,
+					  jiffies +
+					  msecs_to_jiffies(button->debounce_interval));
+			else
+				gpio_keys_report_event(button, bdata->input);
 
-			input_event(input, type, button->code, !!state);
-			input_sync(input);
 			return IRQ_HANDLED;
 		}
 	}
@@ -53,17 +84,21 @@ static irqreturn_t gpio_keys_isr(int irq, void *dev_id)
 static int __devinit gpio_keys_probe(struct platform_device *pdev)
 {
 	struct gpio_keys_platform_data *pdata = pdev->dev.platform_data;
+	struct gpio_keys_drvdata *ddata;
 	struct input_dev *input;
 	int i, error;
 	int wakeup = 0;
 
+	ddata = kzalloc(sizeof(struct gpio_keys_drvdata) +
+			pdata->nbuttons * sizeof(struct gpio_button_data),
+			GFP_KERNEL);
 	input = input_allocate_device();
-	if (!input)
-		return -ENOMEM;
-
-	platform_set_drvdata(pdev, input);
+	if (!ddata || !input) {
+		error = -ENOMEM;
+		goto fail1;
+	}
 
-	input->evbit[0] = BIT_MASK(EV_KEY);
+	platform_set_drvdata(pdev, ddata);
 
 	input->name = pdev->name;
 	input->phys = "gpio-keys/input0";
@@ -74,16 +109,23 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev)
 	input->id.product = 0x0001;
 	input->id.version = 0x0100;
 
+	ddata->input = input;
+
 	for (i = 0; i < pdata->nbuttons; i++) {
 		struct gpio_keys_button *button = &pdata->buttons[i];
+		struct gpio_button_data *bdata = &ddata->data[i];
 		int irq;
 		unsigned int type = button->type ?: EV_KEY;
 
+		bdata->input = input;
+		setup_timer(&bdata->timer,
+			    gpio_check_button, (unsigned long)bdata);
+
 		error = gpio_request(button->gpio, button->desc ?: "gpio_keys");
 		if (error < 0) {
 			pr_err("gpio-keys: failed to request GPIO %d,"
 				" error %d\n", button->gpio, error);
-			goto fail;
+			goto fail2;
 		}
 
 		error = gpio_direction_input(button->gpio);
@@ -92,7 +134,7 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev)
 				" direction for GPIO %d, error %d\n",
 				button->gpio, error);
 			gpio_free(button->gpio);
-			goto fail;
+			goto fail2;
 		}
 
 		irq = gpio_to_irq(button->gpio);
@@ -102,7 +144,7 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev)
 				" for GPIO %d, error %d\n",
 				button->gpio, error);
 			gpio_free(button->gpio);
-			goto fail;
+			goto fail2;
 		}
 
 		error = request_irq(irq, gpio_keys_isr,
@@ -114,7 +156,7 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev)
 			pr_err("gpio-keys: Unable to claim irq %d; error %d\n",
 				irq, error);
 			gpio_free(button->gpio);
-			goto fail;
+			goto fail2;
 		}
 
 		if (button->wakeup)
@@ -127,21 +169,25 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev)
 	if (error) {
 		pr_err("gpio-keys: Unable to register input device, "
 			"error: %d\n", error);
-		goto fail;
+		goto fail2;
 	}
 
 	device_init_wakeup(&pdev->dev, wakeup);
 
 	return 0;
 
- fail:
+ fail2:
 	while (--i >= 0) {
 		free_irq(gpio_to_irq(pdata->buttons[i].gpio), pdev);
+		if (pdata->buttons[i].debounce_interval)
+			del_timer_sync(&ddata->data[i].timer);
 		gpio_free(pdata->buttons[i].gpio);
 	}
 
 	platform_set_drvdata(pdev, NULL);
+ fail1:
 	input_free_device(input);
+	kfree(ddata);
 
 	return error;
 }
@@ -149,7 +195,8 @@ static int __devinit gpio_keys_probe(struct platform_device *pdev)
 static int __devexit gpio_keys_remove(struct platform_device *pdev)
 {
 	struct gpio_keys_platform_data *pdata = pdev->dev.platform_data;
-	struct input_dev *input = platform_get_drvdata(pdev);
+	struct gpio_keys_drvdata *ddata = platform_get_drvdata(pdev);
+	struct input_dev *input = ddata->input;
 	int i;
 
 	device_init_wakeup(&pdev->dev, 0);
@@ -157,6 +204,8 @@ static int __devexit gpio_keys_remove(struct platform_device *pdev)
 	for (i = 0; i < pdata->nbuttons; i++) {
 		int irq = gpio_to_irq(pdata->buttons[i].gpio);
 		free_irq(irq, pdev);
+		if (pdata->buttons[i].debounce_interval)
+			del_timer_sync(&ddata->data[i].timer);
 		gpio_free(pdata->buttons[i].gpio);
 	}
 
diff --git a/include/linux/gpio_keys.h b/include/linux/gpio_keys.h
index c6d3a9de5634..ec6ecd74781d 100644
--- a/include/linux/gpio_keys.h
+++ b/include/linux/gpio_keys.h
@@ -9,6 +9,7 @@ struct gpio_keys_button {
 	char *desc;
 	int type;		/* input event type (EV_KEY, EV_SW) */
 	int wakeup;		/* configure the button as a wake-up source */
+	int debounce_interval;	/* debounce ticks interval in msecs */
 };
 
 struct gpio_keys_platform_data {
-- 
cgit v1.2.3


From 8b09dee67f484e9b42114b1a1f068e080fd7aa56 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:21:05 +0200
Subject: rcupreempt: remove duplicate prototypes

rcu_batches_completed and rcu_patches_completed_bh are both declared
in rcuclassic.h and rcupreempt.h. This patch removes the extra
prototypes for them from rcupdate.h.

rcu_batches_completed_bh is defined as a static inline in the rcupreempt.h
header file. Trying to export this as EXPORT_SYMBOL_GPL causes linking problems
with the powerpc linker. There's no need to export a static inlined function.

Modules must be compiled with the same type of RCU implementation as the
kernel they are for.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/rcupdate.h | 2 --
 kernel/rcupreempt.c      | 2 --
 2 files changed, 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index d42dbec06083..ec2fc5b32646 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -224,8 +224,6 @@ extern void call_rcu_bh(struct rcu_head *head,
 /* Exported common interfaces */
 extern void synchronize_rcu(void);
 extern void rcu_barrier(void);
-extern long rcu_batches_completed(void);
-extern long rcu_batches_completed_bh(void);
 
 /* Internal to kernel */
 extern void rcu_init(void);
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index e1cdf196a515..5e02b7740702 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -217,8 +217,6 @@ long rcu_batches_completed(void)
 }
 EXPORT_SYMBOL_GPL(rcu_batches_completed);
 
-EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
-
 void __rcu_read_lock(void)
 {
 	int idx;
-- 
cgit v1.2.3


From 4446a36ff8c74ac3b32feb009b651048e129c6af Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Mon, 12 May 2008 21:21:05 +0200
Subject: rcu: add call_rcu_sched()

Fourth cut of patch to provide the call_rcu_sched().  This is again to
synchronize_sched() as call_rcu() is to synchronize_rcu().

Should be fine for experimental and -rt use, but not ready for inclusion.
With some luck, I will be able to tell Andrew to come out of hiding on
the next round.

Passes multi-day rcutorture sessions with concurrent CPU hotplugging.

Fixes since the first version include a bug that could result in
indefinite blocking (spotted by Gautham Shenoy), better resiliency
against CPU-hotplug operations, and other minor fixes.

Fixes since the second version include reworking grace-period detection
to avoid deadlocks that could happen when running concurrently with
CPU hotplug, adding Mathieu's fix to avoid the softlockup messages,
as well as Mathieu's fix to allow use earlier in boot.

Fixes since the third version include a wrong-CPU bug spotted by
Andrew, getting rid of the obsolete synchronize_kernel API that somehow
snuck back in, merging spin_unlock() and local_irq_restore() in a
few places, commenting the code that checks for quiescent states based
on interrupting from user-mode execution or the idle loop, removing
some inline attributes, and some code-style changes.

Known/suspected shortcomings:

o	I still do not entirely trust the sleep/wakeup logic.  Next step
	will be to use a private snapshot of the CPU online mask in
	rcu_sched_grace_period() -- if the CPU wasn't there at the start
	of the grace period, we don't need to hear from it.  And the
	bit about accounting for changes in online CPUs inside of
	rcu_sched_grace_period() is ugly anyway.

o	It might be good for rcu_sched_grace_period() to invoke
	resched_cpu() when a given CPU wasn't responding quickly,
	but resched_cpu() is declared static...

This patch also fixes a long-standing bug in the earlier preemptable-RCU
implementation of synchronize_rcu() that could result in loss of
concurrent external changes to a task's CPU affinity mask.  I still cannot
remember who reported this...

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/rcuclassic.h |   3 +
 include/linux/rcupdate.h   |  22 +++
 include/linux/rcupreempt.h |  42 ++++-
 init/main.c                |   1 +
 kernel/rcupdate.c          |  20 +--
 kernel/rcupreempt.c        | 414 ++++++++++++++++++++++++++++++++++++++++-----
 6 files changed, 434 insertions(+), 68 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h
index b3aa05baab8a..8c774905dcfe 100644
--- a/include/linux/rcuclassic.h
+++ b/include/linux/rcuclassic.h
@@ -151,7 +151,10 @@ extern struct lockdep_map rcu_lock_map;
 
 #define __synchronize_sched() synchronize_rcu()
 
+#define call_rcu_sched(head, func) call_rcu(head, func)
+
 extern void __rcu_init(void);
+#define rcu_init_sched()	do { } while (0)
 extern void rcu_check_callbacks(int cpu, int user);
 extern void rcu_restart_cpu(int cpu);
 
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index ec2fc5b32646..411969cb5243 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -40,6 +40,7 @@
 #include <linux/cpumask.h>
 #include <linux/seqlock.h>
 #include <linux/lockdep.h>
+#include <linux/completion.h>
 
 /**
  * struct rcu_head - callback structure for use with RCU
@@ -168,6 +169,27 @@ struct rcu_head {
 		(p) = (v); \
 	})
 
+/* Infrastructure to implement the synchronize_() primitives. */
+
+struct rcu_synchronize {
+	struct rcu_head head;
+	struct completion completion;
+};
+
+extern void wakeme_after_rcu(struct rcu_head  *head);
+
+#define synchronize_rcu_xxx(name, func) \
+void name(void) \
+{ \
+	struct rcu_synchronize rcu; \
+	\
+	init_completion(&rcu.completion); \
+	/* Will wake me after RCU finished. */ \
+	func(&rcu.head, wakeme_after_rcu); \
+	/* Wait for it. */ \
+	wait_for_completion(&rcu.completion); \
+}
+
 /**
  * synchronize_sched - block until all CPUs have exited any non-preemptive
  * kernel code sequences.
diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h
index 8a05c7e20bc4..f04b64eca636 100644
--- a/include/linux/rcupreempt.h
+++ b/include/linux/rcupreempt.h
@@ -40,10 +40,39 @@
 #include <linux/cpumask.h>
 #include <linux/seqlock.h>
 
-#define rcu_qsctr_inc(cpu)
+struct rcu_dyntick_sched {
+	int dynticks;
+	int dynticks_snap;
+	int sched_qs;
+	int sched_qs_snap;
+	int sched_dynticks_snap;
+};
+
+DECLARE_PER_CPU(struct rcu_dyntick_sched, rcu_dyntick_sched);
+
+static inline void rcu_qsctr_inc(int cpu)
+{
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
+
+	rdssp->sched_qs++;
+}
 #define rcu_bh_qsctr_inc(cpu)
 #define call_rcu_bh(head, rcu) call_rcu(head, rcu)
 
+/**
+ * call_rcu_sched - Queue RCU callback for invocation after sched grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full
+ * synchronize_sched()-style grace period elapses, in other words after
+ * all currently executing preempt-disabled sections of code (including
+ * hardirq handlers, NMI handlers, and local_irq_save() blocks) have
+ * completed.
+ */
+extern void call_rcu_sched(struct rcu_head *head,
+			   void (*func)(struct rcu_head *head));
+
 extern void __rcu_read_lock(void)	__acquires(RCU);
 extern void __rcu_read_unlock(void)	__releases(RCU);
 extern int rcu_pending(int cpu);
@@ -55,6 +84,7 @@ extern int rcu_needs_cpu(int cpu);
 extern void __synchronize_sched(void);
 
 extern void __rcu_init(void);
+extern void rcu_init_sched(void);
 extern void rcu_check_callbacks(int cpu, int user);
 extern void rcu_restart_cpu(int cpu);
 extern long rcu_batches_completed(void);
@@ -81,20 +111,20 @@ extern struct rcupreempt_trace *rcupreempt_trace_cpu(int cpu);
 struct softirq_action;
 
 #ifdef CONFIG_NO_HZ
-DECLARE_PER_CPU(long, dynticks_progress_counter);
+DECLARE_PER_CPU(struct rcu_dyntick_sched, rcu_dyntick_sched);
 
 static inline void rcu_enter_nohz(void)
 {
 	smp_mb(); /* CPUs seeing ++ must see prior RCU read-side crit sects */
-	__get_cpu_var(dynticks_progress_counter)++;
-	WARN_ON(__get_cpu_var(dynticks_progress_counter) & 0x1);
+	__get_cpu_var(rcu_dyntick_sched).dynticks++;
+	WARN_ON(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1);
 }
 
 static inline void rcu_exit_nohz(void)
 {
-	__get_cpu_var(dynticks_progress_counter)++;
 	smp_mb(); /* CPUs seeing ++ must see later RCU read-side crit sects */
-	WARN_ON(!(__get_cpu_var(dynticks_progress_counter) & 0x1));
+	__get_cpu_var(rcu_dyntick_sched).dynticks++;
+	WARN_ON(!(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1));
 }
 
 #else /* CONFIG_NO_HZ */
diff --git a/init/main.c b/init/main.c
index f7fb20021d48..a9cc3e0803de 100644
--- a/init/main.c
+++ b/init/main.c
@@ -758,6 +758,7 @@ static void __init do_initcalls(void)
  */
 static void __init do_basic_setup(void)
 {
+	rcu_init_sched(); /* needed by module_init stage. */
 	/* drivers will send hotplug events */
 	init_workqueues();
 	usermodehelper_init();
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index c09605f8d16c..a4e329d92883 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -39,18 +39,12 @@
 #include <linux/sched.h>
 #include <asm/atomic.h>
 #include <linux/bitops.h>
-#include <linux/completion.h>
 #include <linux/percpu.h>
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <linux/mutex.h>
 #include <linux/module.h>
 
-struct rcu_synchronize {
-	struct rcu_head head;
-	struct completion completion;
-};
-
 static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL};
 static atomic_t rcu_barrier_cpu_count;
 static DEFINE_MUTEX(rcu_barrier_mutex);
@@ -60,7 +54,7 @@ static struct completion rcu_barrier_completion;
  * Awaken the corresponding synchronize_rcu() instance now that a
  * grace period has elapsed.
  */
-static void wakeme_after_rcu(struct rcu_head  *head)
+void wakeme_after_rcu(struct rcu_head  *head)
 {
 	struct rcu_synchronize *rcu;
 
@@ -77,17 +71,7 @@ static void wakeme_after_rcu(struct rcu_head  *head)
  * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
  * and may be nested.
  */
-void synchronize_rcu(void)
-{
-	struct rcu_synchronize rcu;
-
-	init_completion(&rcu.completion);
-	/* Will wake me after RCU finished */
-	call_rcu(&rcu.head, wakeme_after_rcu);
-
-	/* Wait for it */
-	wait_for_completion(&rcu.completion);
-}
+synchronize_rcu_xxx(synchronize_rcu, call_rcu)
 EXPORT_SYMBOL_GPL(synchronize_rcu);
 
 static void rcu_barrier_callback(struct rcu_head *notused)
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index 5e02b7740702..aaa7976bd85f 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -46,6 +46,7 @@
 #include <asm/atomic.h>
 #include <linux/bitops.h>
 #include <linux/module.h>
+#include <linux/kthread.h>
 #include <linux/completion.h>
 #include <linux/moduleparam.h>
 #include <linux/percpu.h>
@@ -87,9 +88,14 @@ struct rcu_data {
 	struct rcu_head **nexttail;
 	struct rcu_head *waitlist[GP_STAGES];
 	struct rcu_head **waittail[GP_STAGES];
-	struct rcu_head *donelist;
+	struct rcu_head *donelist;	/* from waitlist & waitschedlist */
 	struct rcu_head **donetail;
 	long rcu_flipctr[2];
+	struct rcu_head *nextschedlist;
+	struct rcu_head **nextschedtail;
+	struct rcu_head *waitschedlist;
+	struct rcu_head **waitschedtail;
+	int rcu_sched_sleeping;
 #ifdef CONFIG_RCU_TRACE
 	struct rcupreempt_trace trace;
 #endif /* #ifdef CONFIG_RCU_TRACE */
@@ -131,11 +137,24 @@ enum rcu_try_flip_states {
 	rcu_try_flip_waitmb_state,
 };
 
+/*
+ * States for rcu_ctrlblk.rcu_sched_sleep.
+ */
+
+enum rcu_sched_sleep_states {
+	rcu_sched_not_sleeping,	/* Not sleeping, callbacks need GP.  */
+	rcu_sched_sleep_prep,	/* Thinking of sleeping, rechecking. */
+	rcu_sched_sleeping,	/* Sleeping, awaken if GP needed. */
+};
+
 struct rcu_ctrlblk {
 	spinlock_t	fliplock;	/* Protect state-machine transitions. */
 	long		completed;	/* Number of last completed batch. */
 	enum rcu_try_flip_states rcu_try_flip_state; /* The current state of
 							the rcu state machine */
+	spinlock_t	schedlock;	/* Protect rcu_sched sleep state. */
+	enum rcu_sched_sleep_states sched_sleep; /* rcu_sched state. */
+	wait_queue_head_t sched_wq;	/* Place for rcu_sched to sleep. */
 };
 
 static DEFINE_PER_CPU(struct rcu_data, rcu_data);
@@ -143,8 +162,12 @@ static struct rcu_ctrlblk rcu_ctrlblk = {
 	.fliplock = __SPIN_LOCK_UNLOCKED(rcu_ctrlblk.fliplock),
 	.completed = 0,
 	.rcu_try_flip_state = rcu_try_flip_idle_state,
+	.schedlock = __SPIN_LOCK_UNLOCKED(rcu_ctrlblk.schedlock),
+	.sched_sleep = rcu_sched_not_sleeping,
+	.sched_wq = __WAIT_QUEUE_HEAD_INITIALIZER(rcu_ctrlblk.sched_wq),
 };
 
+static struct task_struct *rcu_sched_grace_period_task;
 
 #ifdef CONFIG_RCU_TRACE
 static char *rcu_try_flip_state_names[] =
@@ -207,6 +230,8 @@ static DEFINE_PER_CPU_SHARED_ALIGNED(enum rcu_mb_flag_values, rcu_mb_flag)
  */
 #define RCU_TRACE_RDP(f, rdp) RCU_TRACE(f, &((rdp)->trace));
 
+#define RCU_SCHED_BATCH_TIME (HZ / 50)
+
 /*
  * Return the number of RCU batches processed thus far.  Useful
  * for debug and statistics.
@@ -411,32 +436,34 @@ static void __rcu_advance_callbacks(struct rcu_data *rdp)
 	}
 }
 
-#ifdef CONFIG_NO_HZ
+DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_dyntick_sched, rcu_dyntick_sched) = {
+	.dynticks = 1,
+};
 
-DEFINE_PER_CPU(long, dynticks_progress_counter) = 1;
-static DEFINE_PER_CPU(long, rcu_dyntick_snapshot);
+#ifdef CONFIG_NO_HZ
 static DEFINE_PER_CPU(int, rcu_update_flag);
 
 /**
  * rcu_irq_enter - Called from Hard irq handlers and NMI/SMI.
  *
  * If the CPU was idle with dynamic ticks active, this updates the
- * dynticks_progress_counter to let the RCU handling know that the
+ * rcu_dyntick_sched.dynticks to let the RCU handling know that the
  * CPU is active.
  */
 void rcu_irq_enter(void)
 {
 	int cpu = smp_processor_id();
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
 
 	if (per_cpu(rcu_update_flag, cpu))
 		per_cpu(rcu_update_flag, cpu)++;
 
 	/*
 	 * Only update if we are coming from a stopped ticks mode
-	 * (dynticks_progress_counter is even).
+	 * (rcu_dyntick_sched.dynticks is even).
 	 */
 	if (!in_interrupt() &&
-	    (per_cpu(dynticks_progress_counter, cpu) & 0x1) == 0) {
+	    (rdssp->dynticks & 0x1) == 0) {
 		/*
 		 * The following might seem like we could have a race
 		 * with NMI/SMIs. But this really isn't a problem.
@@ -459,12 +486,12 @@ void rcu_irq_enter(void)
 		 * RCU read-side critical sections on this CPU would
 		 * have already completed.
 		 */
-		per_cpu(dynticks_progress_counter, cpu)++;
+		rdssp->dynticks++;
 		/*
 		 * The following memory barrier ensures that any
 		 * rcu_read_lock() primitives in the irq handler
 		 * are seen by other CPUs to follow the above
-		 * increment to dynticks_progress_counter. This is
+		 * increment to rcu_dyntick_sched.dynticks. This is
 		 * required in order for other CPUs to correctly
 		 * determine when it is safe to advance the RCU
 		 * grace-period state machine.
@@ -472,7 +499,7 @@ void rcu_irq_enter(void)
 		smp_mb(); /* see above block comment. */
 		/*
 		 * Since we can't determine the dynamic tick mode from
-		 * the dynticks_progress_counter after this routine,
+		 * the rcu_dyntick_sched.dynticks after this routine,
 		 * we use a second flag to acknowledge that we came
 		 * from an idle state with ticks stopped.
 		 */
@@ -480,7 +507,7 @@ void rcu_irq_enter(void)
 		/*
 		 * If we take an NMI/SMI now, they will also increment
 		 * the rcu_update_flag, and will not update the
-		 * dynticks_progress_counter on exit. That is for
+		 * rcu_dyntick_sched.dynticks on exit. That is for
 		 * this IRQ to do.
 		 */
 	}
@@ -490,12 +517,13 @@ void rcu_irq_enter(void)
  * rcu_irq_exit - Called from exiting Hard irq context.
  *
  * If the CPU was idle with dynamic ticks active, update the
- * dynticks_progress_counter to put let the RCU handling be
+ * rcu_dyntick_sched.dynticks to put let the RCU handling be
  * aware that the CPU is going back to idle with no ticks.
  */
 void rcu_irq_exit(void)
 {
 	int cpu = smp_processor_id();
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
 
 	/*
 	 * rcu_update_flag is set if we interrupted the CPU
@@ -503,7 +531,7 @@ void rcu_irq_exit(void)
 	 * Once this occurs, we keep track of interrupt nesting
 	 * because a NMI/SMI could also come in, and we still
 	 * only want the IRQ that started the increment of the
-	 * dynticks_progress_counter to be the one that modifies
+	 * rcu_dyntick_sched.dynticks to be the one that modifies
 	 * it on exit.
 	 */
 	if (per_cpu(rcu_update_flag, cpu)) {
@@ -515,28 +543,29 @@ void rcu_irq_exit(void)
 
 		/*
 		 * If an NMI/SMI happens now we are still
-		 * protected by the dynticks_progress_counter being odd.
+		 * protected by the rcu_dyntick_sched.dynticks being odd.
 		 */
 
 		/*
 		 * The following memory barrier ensures that any
 		 * rcu_read_unlock() primitives in the irq handler
 		 * are seen by other CPUs to preceed the following
-		 * increment to dynticks_progress_counter. This
+		 * increment to rcu_dyntick_sched.dynticks. This
 		 * is required in order for other CPUs to determine
 		 * when it is safe to advance the RCU grace-period
 		 * state machine.
 		 */
 		smp_mb(); /* see above block comment. */
-		per_cpu(dynticks_progress_counter, cpu)++;
-		WARN_ON(per_cpu(dynticks_progress_counter, cpu) & 0x1);
+		rdssp->dynticks++;
+		WARN_ON(rdssp->dynticks & 0x1);
 	}
 }
 
 static void dyntick_save_progress_counter(int cpu)
 {
-	per_cpu(rcu_dyntick_snapshot, cpu) =
-		per_cpu(dynticks_progress_counter, cpu);
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
+
+	rdssp->dynticks_snap = rdssp->dynticks;
 }
 
 static inline int
@@ -544,9 +573,10 @@ rcu_try_flip_waitack_needed(int cpu)
 {
 	long curr;
 	long snap;
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
 
-	curr = per_cpu(dynticks_progress_counter, cpu);
-	snap = per_cpu(rcu_dyntick_snapshot, cpu);
+	curr = rdssp->dynticks;
+	snap = rdssp->dynticks_snap;
 	smp_mb(); /* force ordering with cpu entering/leaving dynticks. */
 
 	/*
@@ -580,9 +610,10 @@ rcu_try_flip_waitmb_needed(int cpu)
 {
 	long curr;
 	long snap;
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
 
-	curr = per_cpu(dynticks_progress_counter, cpu);
-	snap = per_cpu(rcu_dyntick_snapshot, cpu);
+	curr = rdssp->dynticks;
+	snap = rdssp->dynticks_snap;
 	smp_mb(); /* force ordering with cpu entering/leaving dynticks. */
 
 	/*
@@ -609,14 +640,86 @@ rcu_try_flip_waitmb_needed(int cpu)
 	return 1;
 }
 
+static void dyntick_save_progress_counter_sched(int cpu)
+{
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
+
+	rdssp->sched_dynticks_snap = rdssp->dynticks;
+}
+
+static int rcu_qsctr_inc_needed_dyntick(int cpu)
+{
+	long curr;
+	long snap;
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
+
+	curr = rdssp->dynticks;
+	snap = rdssp->sched_dynticks_snap;
+	smp_mb(); /* force ordering with cpu entering/leaving dynticks. */
+
+	/*
+	 * If the CPU remained in dynticks mode for the entire time
+	 * and didn't take any interrupts, NMIs, SMIs, or whatever,
+	 * then it cannot be in the middle of an rcu_read_lock(), so
+	 * the next rcu_read_lock() it executes must use the new value
+	 * of the counter.  Therefore, this CPU has been in a quiescent
+	 * state the entire time, and we don't need to wait for it.
+	 */
+
+	if ((curr == snap) && ((curr & 0x1) == 0))
+		return 0;
+
+	/*
+	 * If the CPU passed through or entered a dynticks idle phase with
+	 * no active irq handlers, then, as above, this CPU has already
+	 * passed through a quiescent state.
+	 */
+
+	if ((curr - snap) > 2 || (snap & 0x1) == 0)
+		return 0;
+
+	/* We need this CPU to go through a quiescent state. */
+
+	return 1;
+}
+
 #else /* !CONFIG_NO_HZ */
 
-# define dyntick_save_progress_counter(cpu)	do { } while (0)
-# define rcu_try_flip_waitack_needed(cpu)	(1)
-# define rcu_try_flip_waitmb_needed(cpu)	(1)
+# define dyntick_save_progress_counter(cpu)		do { } while (0)
+# define rcu_try_flip_waitack_needed(cpu)		(1)
+# define rcu_try_flip_waitmb_needed(cpu)		(1)
+
+# define dyntick_save_progress_counter_sched(cpu)	do { } while (0)
+# define rcu_qsctr_inc_needed_dyntick(cpu)		(1)
 
 #endif /* CONFIG_NO_HZ */
 
+static void save_qsctr_sched(int cpu)
+{
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
+
+	rdssp->sched_qs_snap = rdssp->sched_qs;
+}
+
+static inline int rcu_qsctr_inc_needed(int cpu)
+{
+	struct rcu_dyntick_sched *rdssp = &per_cpu(rcu_dyntick_sched, cpu);
+
+	/*
+	 * If there has been a quiescent state, no more need to wait
+	 * on this CPU.
+	 */
+
+	if (rdssp->sched_qs != rdssp->sched_qs_snap) {
+		smp_mb(); /* force ordering with cpu entering schedule(). */
+		return 0;
+	}
+
+	/* We need this CPU to go through a quiescent state. */
+
+	return 1;
+}
+
 /*
  * Get here when RCU is idle.  Decide whether we need to
  * move out of idle state, and return non-zero if so.
@@ -819,6 +922,26 @@ void rcu_check_callbacks(int cpu, int user)
 	unsigned long flags;
 	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
 
+	/*
+	 * If this CPU took its interrupt from user mode or from the
+	 * idle loop, and this is not a nested interrupt, then
+	 * this CPU has to have exited all prior preept-disable
+	 * sections of code.  So increment the counter to note this.
+	 *
+	 * The memory barrier is needed to handle the case where
+	 * writes from a preempt-disable section of code get reordered
+	 * into schedule() by this CPU's write buffer.  So the memory
+	 * barrier makes sure that the rcu_qsctr_inc() is seen by other
+	 * CPUs to happen after any such write.
+	 */
+
+	if (user ||
+	    (idle_cpu(cpu) && !in_softirq() &&
+	     hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
+		smp_mb();	/* Guard against aggressive schedule(). */
+	     	rcu_qsctr_inc(cpu);
+	}
+
 	rcu_check_mb(cpu);
 	if (rcu_ctrlblk.completed == rdp->completed)
 		rcu_try_flip();
@@ -869,6 +992,8 @@ void rcu_offline_cpu(int cpu)
 	struct rcu_head *list = NULL;
 	unsigned long flags;
 	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+	struct rcu_head *schedlist = NULL;
+	struct rcu_head **schedtail = &schedlist;
 	struct rcu_head **tail = &list;
 
 	/*
@@ -882,6 +1007,11 @@ void rcu_offline_cpu(int cpu)
 		rcu_offline_cpu_enqueue(rdp->waitlist[i], rdp->waittail[i],
 						list, tail);
 	rcu_offline_cpu_enqueue(rdp->nextlist, rdp->nexttail, list, tail);
+	rcu_offline_cpu_enqueue(rdp->waitschedlist, rdp->waitschedtail,
+				schedlist, schedtail);
+	rcu_offline_cpu_enqueue(rdp->nextschedlist, rdp->nextschedtail,
+				schedlist, schedtail);
+	rdp->rcu_sched_sleeping = 0;
 	spin_unlock_irqrestore(&rdp->lock, flags);
 	rdp->waitlistcount = 0;
 
@@ -916,22 +1046,40 @@ void rcu_offline_cpu(int cpu)
 	 * fix.
 	 */
 
-	local_irq_save(flags);
+	local_irq_save(flags);  /* disable preempt till we know what lock. */
 	rdp = RCU_DATA_ME();
 	spin_lock(&rdp->lock);
 	*rdp->nexttail = list;
 	if (list)
 		rdp->nexttail = tail;
+	*rdp->nextschedtail = schedlist;
+	if (schedlist)
+		rdp->nextschedtail = schedtail;
 	spin_unlock_irqrestore(&rdp->lock, flags);
 }
 
 void __devinit rcu_online_cpu(int cpu)
 {
 	unsigned long flags;
+	struct rcu_data *rdp;
 
 	spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags);
 	cpu_set(cpu, rcu_cpu_online_map);
 	spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags);
+
+	/*
+	 * The rcu_sched grace-period processing might have bypassed
+	 * this CPU, given that it was not in the rcu_cpu_online_map
+	 * when the grace-period scan started.  This means that the
+	 * grace-period task might sleep.  So make sure that if this
+	 * should happen, the first callback posted to this CPU will
+	 * wake up the grace-period task if need be.
+	 */
+
+	rdp = RCU_DATA_CPU(cpu);
+	spin_lock_irqsave(&rdp->lock, flags);
+	rdp->rcu_sched_sleeping = 1;
+	spin_unlock_irqrestore(&rdp->lock, flags);
 }
 
 #else /* #ifdef CONFIG_HOTPLUG_CPU */
@@ -986,31 +1134,196 @@ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
 	*rdp->nexttail = head;
 	rdp->nexttail = &head->next;
 	RCU_TRACE_RDP(rcupreempt_trace_next_add, rdp);
-	spin_unlock(&rdp->lock);
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&rdp->lock, flags);
 }
 EXPORT_SYMBOL_GPL(call_rcu);
 
+void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+	struct rcu_data *rdp;
+	int wake_gp = 0;
+
+	head->func = func;
+	head->next = NULL;
+	local_irq_save(flags);
+	rdp = RCU_DATA_ME();
+	spin_lock(&rdp->lock);
+	*rdp->nextschedtail = head;
+	rdp->nextschedtail = &head->next;
+	if (rdp->rcu_sched_sleeping) {
+
+		/* Grace-period processing might be sleeping... */
+
+		rdp->rcu_sched_sleeping = 0;
+		wake_gp = 1;
+	}
+	spin_unlock_irqrestore(&rdp->lock, flags);
+	if (wake_gp) {
+
+		/* Wake up grace-period processing, unless someone beat us. */
+
+		spin_lock_irqsave(&rcu_ctrlblk.schedlock, flags);
+		if (rcu_ctrlblk.sched_sleep != rcu_sched_sleeping)
+			wake_gp = 0;
+		rcu_ctrlblk.sched_sleep = rcu_sched_not_sleeping;
+		spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags);
+		if (wake_gp)
+			wake_up_interruptible(&rcu_ctrlblk.sched_wq);
+	}
+}
+EXPORT_SYMBOL_GPL(call_rcu_sched);
+
 /*
  * Wait until all currently running preempt_disable() code segments
  * (including hardware-irq-disable segments) complete.  Note that
  * in -rt this does -not- necessarily result in all currently executing
  * interrupt -handlers- having completed.
  */
-void __synchronize_sched(void)
+synchronize_rcu_xxx(__synchronize_sched, call_rcu_sched)
+EXPORT_SYMBOL_GPL(__synchronize_sched);
+
+/*
+ * kthread function that manages call_rcu_sched grace periods.
+ */
+static int rcu_sched_grace_period(void *arg)
 {
-	cpumask_t oldmask;
+	int couldsleep;		/* might sleep after current pass. */
+	int couldsleepnext = 0; /* might sleep after next pass. */
 	int cpu;
+	unsigned long flags;
+	struct rcu_data *rdp;
+	int ret;
 
-	if (sched_getaffinity(0, &oldmask) < 0)
-		oldmask = cpu_possible_map;
-	for_each_online_cpu(cpu) {
-		sched_setaffinity(0, &cpumask_of_cpu(cpu));
-		schedule();
-	}
-	sched_setaffinity(0, &oldmask);
+	/*
+	 * Each pass through the following loop handles one
+	 * rcu_sched grace period cycle.
+	 */
+	do {
+		/* Save each CPU's current state. */
+
+		for_each_online_cpu(cpu) {
+			dyntick_save_progress_counter_sched(cpu);
+			save_qsctr_sched(cpu);
+		}
+
+		/*
+		 * Sleep for about an RCU grace-period's worth to
+		 * allow better batching and to consume less CPU.
+		 */
+		schedule_timeout_interruptible(RCU_SCHED_BATCH_TIME);
+
+		/*
+		 * If there was nothing to do last time, prepare to
+		 * sleep at the end of the current grace period cycle.
+		 */
+		couldsleep = couldsleepnext;
+		couldsleepnext = 1;
+		if (couldsleep) {
+			spin_lock_irqsave(&rcu_ctrlblk.schedlock, flags);
+			rcu_ctrlblk.sched_sleep = rcu_sched_sleep_prep;
+			spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags);
+		}
+
+		/*
+		 * Wait on each CPU in turn to have either visited
+		 * a quiescent state or been in dynticks-idle mode.
+		 */
+		for_each_online_cpu(cpu) {
+			while (rcu_qsctr_inc_needed(cpu) &&
+			       rcu_qsctr_inc_needed_dyntick(cpu)) {
+				/* resched_cpu(cpu); @@@ */
+				schedule_timeout_interruptible(1);
+			}
+		}
+
+		/* Advance callbacks for each CPU.  */
+
+		for_each_online_cpu(cpu) {
+
+			rdp = RCU_DATA_CPU(cpu);
+			spin_lock_irqsave(&rdp->lock, flags);
+
+			/*
+			 * We are running on this CPU irq-disabled, so no
+			 * CPU can go offline until we re-enable irqs.
+			 * The current CPU might have already gone
+			 * offline (between the for_each_offline_cpu and
+			 * the spin_lock_irqsave), but in that case all its
+			 * callback lists will be empty, so no harm done.
+			 *
+			 * Advance the callbacks!  We share normal RCU's
+			 * donelist, since callbacks are invoked the
+			 * same way in either case.
+			 */
+			if (rdp->waitschedlist != NULL) {
+				*rdp->donetail = rdp->waitschedlist;
+				rdp->donetail = rdp->waitschedtail;
+
+				/*
+				 * Next rcu_check_callbacks() will
+				 * do the required raise_softirq().
+				 */
+			}
+			if (rdp->nextschedlist != NULL) {
+				rdp->waitschedlist = rdp->nextschedlist;
+				rdp->waitschedtail = rdp->nextschedtail;
+				couldsleep = 0;
+				couldsleepnext = 0;
+			} else {
+				rdp->waitschedlist = NULL;
+				rdp->waitschedtail = &rdp->waitschedlist;
+			}
+			rdp->nextschedlist = NULL;
+			rdp->nextschedtail = &rdp->nextschedlist;
+
+			/* Mark sleep intention. */
+
+			rdp->rcu_sched_sleeping = couldsleep;
+
+			spin_unlock_irqrestore(&rdp->lock, flags);
+		}
+
+		/* If we saw callbacks on the last scan, go deal with them. */
+
+		if (!couldsleep)
+			continue;
+
+		/* Attempt to block... */
+
+		spin_lock_irqsave(&rcu_ctrlblk.schedlock, flags);
+		if (rcu_ctrlblk.sched_sleep != rcu_sched_sleep_prep) {
+
+			/*
+			 * Someone posted a callback after we scanned.
+			 * Go take care of it.
+			 */
+			spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags);
+			couldsleepnext = 0;
+			continue;
+		}
+
+		/* Block until the next person posts a callback. */
+
+		rcu_ctrlblk.sched_sleep = rcu_sched_sleeping;
+		spin_unlock_irqrestore(&rcu_ctrlblk.schedlock, flags);
+		ret = 0;
+		__wait_event_interruptible(rcu_ctrlblk.sched_wq,
+			rcu_ctrlblk.sched_sleep != rcu_sched_sleeping,
+			ret);
+
+		/*
+		 * Signals would prevent us from sleeping, and we cannot
+		 * do much with them in any case.  So flush them.
+		 */
+		if (ret)
+			flush_signals(current);
+		couldsleepnext = 0;
+
+	} while (!kthread_should_stop());
+
+	return (0);
 }
-EXPORT_SYMBOL_GPL(__synchronize_sched);
 
 /*
  * Check to see if any future RCU-related work will need to be done
@@ -1027,7 +1340,9 @@ int rcu_needs_cpu(int cpu)
 
 	return (rdp->donelist != NULL ||
 		!!rdp->waitlistcount ||
-		rdp->nextlist != NULL);
+		rdp->nextlist != NULL ||
+		rdp->nextschedlist != NULL ||
+		rdp->waitschedlist != NULL);
 }
 
 int rcu_pending(int cpu)
@@ -1038,7 +1353,9 @@ int rcu_pending(int cpu)
 
 	if (rdp->donelist != NULL ||
 	    !!rdp->waitlistcount ||
-	    rdp->nextlist != NULL)
+	    rdp->nextlist != NULL ||
+	    rdp->nextschedlist != NULL ||
+	    rdp->waitschedlist != NULL)
 		return 1;
 
 	/* The RCU core needs an acknowledgement from this CPU. */
@@ -1105,6 +1422,11 @@ void __init __rcu_init(void)
 		rdp->donetail = &rdp->donelist;
 		rdp->rcu_flipctr[0] = 0;
 		rdp->rcu_flipctr[1] = 0;
+		rdp->nextschedlist = NULL;
+		rdp->nextschedtail = &rdp->nextschedlist;
+		rdp->waitschedlist = NULL;
+		rdp->waitschedtail = &rdp->waitschedlist;
+		rdp->rcu_sched_sleeping = 0;
 	}
 	register_cpu_notifier(&rcu_nb);
 
@@ -1127,11 +1449,15 @@ void __init __rcu_init(void)
 }
 
 /*
- * Deprecated, use synchronize_rcu() or synchronize_sched() instead.
+ * Late-boot-time RCU initialization that must wait until after scheduler
+ * has been initialized.
  */
-void synchronize_kernel(void)
+void __init rcu_init_sched(void)
 {
-	synchronize_rcu();
+	rcu_sched_grace_period_task = kthread_run(rcu_sched_grace_period,
+						  NULL,
+						  "rcu_sched_grace_period");
+	WARN_ON(IS_ERR(rcu_sched_grace_period_task));
 }
 
 #ifdef CONFIG_RCU_TRACE
-- 
cgit v1.2.3


From 70f12f848d3e981479b4f6f751e73c14f7c13e5b Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Mon, 12 May 2008 21:21:05 +0200
Subject: rcu: add rcu_barrier_sched() and rcu_barrier_bh()

Add rcu_barrier_sched() and rcu_barrier_bh().  With these in place,
rcutorture no longer gives the occasional oops when repeatedly starting
and stopping torturing rcu_bh.  Also adds the API needed to flush out
pre-existing call_rcu_sched() callbacks.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/rcupdate.h |  2 ++
 kernel/rcupdate.c        | 55 ++++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 51 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 411969cb5243..e8b4039cfb2f 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -246,6 +246,8 @@ extern void call_rcu_bh(struct rcu_head *head,
 /* Exported common interfaces */
 extern void synchronize_rcu(void);
 extern void rcu_barrier(void);
+extern void rcu_barrier_bh(void);
+extern void rcu_barrier_sched(void);
 
 /* Internal to kernel */
 extern void rcu_init(void);
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index a4e329d92883..4a74b8d48d90 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -45,6 +45,12 @@
 #include <linux/mutex.h>
 #include <linux/module.h>
 
+enum rcu_barrier {
+	RCU_BARRIER_STD,
+	RCU_BARRIER_BH,
+	RCU_BARRIER_SCHED,
+};
+
 static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL};
 static atomic_t rcu_barrier_cpu_count;
 static DEFINE_MUTEX(rcu_barrier_mutex);
@@ -83,19 +89,30 @@ static void rcu_barrier_callback(struct rcu_head *notused)
 /*
  * Called with preemption disabled, and from cross-cpu IRQ context.
  */
-static void rcu_barrier_func(void *notused)
+static void rcu_barrier_func(void *type)
 {
 	int cpu = smp_processor_id();
 	struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu);
 
 	atomic_inc(&rcu_barrier_cpu_count);
-	call_rcu(head, rcu_barrier_callback);
+	switch ((enum rcu_barrier)type) {
+	case RCU_BARRIER_STD:
+		call_rcu(head, rcu_barrier_callback);
+		break;
+	case RCU_BARRIER_BH:
+		call_rcu_bh(head, rcu_barrier_callback);
+		break;
+	case RCU_BARRIER_SCHED:
+		call_rcu_sched(head, rcu_barrier_callback);
+		break;
+	}
 }
 
-/**
- * rcu_barrier - Wait until all the in-flight RCUs are complete.
+/*
+ * Orchestrate the specified type of RCU barrier, waiting for all
+ * RCU callbacks of the specified type to complete.
  */
-void rcu_barrier(void)
+static void _rcu_barrier(enum rcu_barrier type)
 {
 	BUG_ON(in_interrupt());
 	/* Take cpucontrol mutex to protect against CPU hotplug */
@@ -111,13 +128,39 @@ void rcu_barrier(void)
 	 * until all the callbacks are queued.
 	 */
 	rcu_read_lock();
-	on_each_cpu(rcu_barrier_func, NULL, 0, 1);
+	on_each_cpu(rcu_barrier_func, (void *)type, 0, 1);
 	rcu_read_unlock();
 	wait_for_completion(&rcu_barrier_completion);
 	mutex_unlock(&rcu_barrier_mutex);
 }
+
+/**
+ * rcu_barrier - Wait until all in-flight call_rcu() callbacks complete.
+ */
+void rcu_barrier(void)
+{
+	_rcu_barrier(RCU_BARRIER_STD);
+}
 EXPORT_SYMBOL_GPL(rcu_barrier);
 
+/**
+ * rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete.
+ */
+void rcu_barrier_bh(void)
+{
+	_rcu_barrier(RCU_BARRIER_BH);
+}
+EXPORT_SYMBOL_GPL(rcu_barrier_bh);
+
+/**
+ * rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks.
+ */
+void rcu_barrier_sched(void)
+{
+	_rcu_barrier(RCU_BARRIER_SCHED);
+}
+EXPORT_SYMBOL_GPL(rcu_barrier_sched);
+
 void __init rcu_init(void)
 {
 	__rcu_init();
-- 
cgit v1.2.3


From 82524746c27fa418c250a56dd7606b9d3fc79826 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <fbuihuu@gmail.com>
Date: Mon, 12 May 2008 21:21:05 +0200
Subject: rcu: split list.h and move rcu-protected lists into rculist.h

Move rcu-protected lists from list.h into a new header file rculist.h.

This is done because list are a very used primitive structure all over the
kernel and it's currently impossible to include other header files in this
list.h without creating some circular dependencies.

For example, list.h implements rcu-protected list and uses rcu_dereference()
without including rcupdate.h.  It actually compiles because users of
rcu_dereference() are macros.  Others RCU functions could be used too but
aren't probably because of this.

Therefore this patch creates rculist.h which includes rcupdates without to
many changes/troubles.

Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Josh Triplett <josh@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/ia64/sn/kernel/irq.c                       |   1 +
 crypto/async_tx/async_tx.c                      |   1 +
 drivers/infiniband/hw/ipath/ipath_verbs.c       |   1 +
 drivers/infiniband/hw/ipath/ipath_verbs_mcast.c |   3 +-
 drivers/net/macvlan.c                           |   2 +-
 include/linux/dcache.h                          |   1 +
 include/linux/list.h                            | 367 ----------------------
 include/linux/rculist.h                         | 396 ++++++++++++++++++++++++
 kernel/pid.c                                    |   1 +
 lib/textsearch.c                                |   1 +
 net/802/psnap.c                                 |   1 +
 net/8021q/vlan.c                                |   1 +
 net/bridge/br_fdb.c                             |   1 +
 net/bridge/br_stp.c                             |   1 +
 net/netlabel/netlabel_domainhash.c              |   3 +-
 15 files changed, 409 insertions(+), 372 deletions(-)
 create mode 100644 include/linux/rculist.h

(limited to 'include/linux')

diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c
index 53351c3cd7b1..96c31b4180c3 100644
--- a/arch/ia64/sn/kernel/irq.c
+++ b/arch/ia64/sn/kernel/irq.c
@@ -11,6 +11,7 @@
 #include <linux/irq.h>
 #include <linux/spinlock.h>
 #include <linux/init.h>
+#include <linux/rculist.h>
 #include <asm/sn/addrs.h>
 #include <asm/sn/arch.h>
 #include <asm/sn/intr.h>
diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index c6e772fc5ccd..095c798d3170 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -23,6 +23,7 @@
  * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
  *
  */
+#include <linux/rculist.h>
 #include <linux/kernel.h>
 #include <linux/async_tx.h>
 
diff --git a/drivers/infiniband/hw/ipath/ipath_verbs.c b/drivers/infiniband/hw/ipath/ipath_verbs.c
index e0ec540042bf..5d830d87ebca 100644
--- a/drivers/infiniband/hw/ipath/ipath_verbs.c
+++ b/drivers/infiniband/hw/ipath/ipath_verbs.c
@@ -35,6 +35,7 @@
 #include <rdma/ib_user_verbs.h>
 #include <linux/io.h>
 #include <linux/utsname.h>
+#include <linux/rculist.h>
 
 #include "ipath_kernel.h"
 #include "ipath_verbs.h"
diff --git a/drivers/infiniband/hw/ipath/ipath_verbs_mcast.c b/drivers/infiniband/hw/ipath/ipath_verbs_mcast.c
index 9e5abf9c309d..d73e32232879 100644
--- a/drivers/infiniband/hw/ipath/ipath_verbs_mcast.c
+++ b/drivers/infiniband/hw/ipath/ipath_verbs_mcast.c
@@ -31,8 +31,7 @@
  * SOFTWARE.
  */
 
-#include <linux/list.h>
-#include <linux/rcupdate.h>
+#include <linux/rculist.h>
 
 #include "ipath_verbs.h"
 
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index c36a03ae9bfb..860d75d81f82 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -20,7 +20,7 @@
 #include <linux/errno.h>
 #include <linux/slab.h>
 #include <linux/string.h>
-#include <linux/list.h>
+#include <linux/rculist.h>
 #include <linux/notifier.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 2a6639407c80..1f5cebf10a23 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -3,6 +3,7 @@
 
 #include <asm/atomic.h>
 #include <linux/list.h>
+#include <linux/rculist.h>
 #include <linux/spinlock.h>
 #include <linux/cache.h>
 #include <linux/rcupdate.h>
diff --git a/include/linux/list.h b/include/linux/list.h
index 08cf4f651889..139ec41d9c2e 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -84,65 +84,6 @@ static inline void list_add_tail(struct list_head *new, struct list_head *head)
 	__list_add(new, head->prev, head);
 }
 
-/*
- * Insert a new entry between two known consecutive entries.
- *
- * This is only for internal list manipulation where we know
- * the prev/next entries already!
- */
-static inline void __list_add_rcu(struct list_head * new,
-		struct list_head * prev, struct list_head * next)
-{
-	new->next = next;
-	new->prev = prev;
-	smp_wmb();
-	next->prev = new;
-	prev->next = new;
-}
-
-/**
- * list_add_rcu - add a new entry to rcu-protected list
- * @new: new entry to be added
- * @head: list head to add it after
- *
- * Insert a new entry after the specified head.
- * This is good for implementing stacks.
- *
- * The caller must take whatever precautions are necessary
- * (such as holding appropriate locks) to avoid racing
- * with another list-mutation primitive, such as list_add_rcu()
- * or list_del_rcu(), running on this same list.
- * However, it is perfectly legal to run concurrently with
- * the _rcu list-traversal primitives, such as
- * list_for_each_entry_rcu().
- */
-static inline void list_add_rcu(struct list_head *new, struct list_head *head)
-{
-	__list_add_rcu(new, head, head->next);
-}
-
-/**
- * list_add_tail_rcu - add a new entry to rcu-protected list
- * @new: new entry to be added
- * @head: list head to add it before
- *
- * Insert a new entry before the specified head.
- * This is useful for implementing queues.
- *
- * The caller must take whatever precautions are necessary
- * (such as holding appropriate locks) to avoid racing
- * with another list-mutation primitive, such as list_add_tail_rcu()
- * or list_del_rcu(), running on this same list.
- * However, it is perfectly legal to run concurrently with
- * the _rcu list-traversal primitives, such as
- * list_for_each_entry_rcu().
- */
-static inline void list_add_tail_rcu(struct list_head *new,
-					struct list_head *head)
-{
-	__list_add_rcu(new, head->prev, head);
-}
-
 /*
  * Delete a list entry by making the prev/next entries
  * point to each other.
@@ -173,36 +114,6 @@ static inline void list_del(struct list_head *entry)
 extern void list_del(struct list_head *entry);
 #endif
 
-/**
- * list_del_rcu - deletes entry from list without re-initialization
- * @entry: the element to delete from the list.
- *
- * Note: list_empty() on entry does not return true after this,
- * the entry is in an undefined state. It is useful for RCU based
- * lockfree traversal.
- *
- * In particular, it means that we can not poison the forward
- * pointers that may still be used for walking the list.
- *
- * The caller must take whatever precautions are necessary
- * (such as holding appropriate locks) to avoid racing
- * with another list-mutation primitive, such as list_del_rcu()
- * or list_add_rcu(), running on this same list.
- * However, it is perfectly legal to run concurrently with
- * the _rcu list-traversal primitives, such as
- * list_for_each_entry_rcu().
- *
- * Note that the caller is not permitted to immediately free
- * the newly deleted entry.  Instead, either synchronize_rcu()
- * or call_rcu() must be used to defer freeing until an RCU
- * grace period has elapsed.
- */
-static inline void list_del_rcu(struct list_head *entry)
-{
-	__list_del(entry->prev, entry->next);
-	entry->prev = LIST_POISON2;
-}
-
 /**
  * list_replace - replace old entry by new one
  * @old : the element to be replaced
@@ -226,25 +137,6 @@ static inline void list_replace_init(struct list_head *old,
 	INIT_LIST_HEAD(old);
 }
 
-/**
- * list_replace_rcu - replace old entry by new one
- * @old : the element to be replaced
- * @new : the new element to insert
- *
- * The @old entry will be replaced with the @new entry atomically.
- * Note: @old should not be empty.
- */
-static inline void list_replace_rcu(struct list_head *old,
-				struct list_head *new)
-{
-	new->next = old->next;
-	new->prev = old->prev;
-	smp_wmb();
-	new->next->prev = new;
-	new->prev->next = new;
-	old->prev = LIST_POISON2;
-}
-
 /**
  * list_del_init - deletes entry from list and reinitialize it.
  * @entry: the element to delete from the list.
@@ -368,62 +260,6 @@ static inline void list_splice_init(struct list_head *list,
 	}
 }
 
-/**
- * list_splice_init_rcu - splice an RCU-protected list into an existing list.
- * @list:	the RCU-protected list to splice
- * @head:	the place in the list to splice the first list into
- * @sync:	function to sync: synchronize_rcu(), synchronize_sched(), ...
- *
- * @head can be RCU-read traversed concurrently with this function.
- *
- * Note that this function blocks.
- *
- * Important note: the caller must take whatever action is necessary to
- *	prevent any other updates to @head.  In principle, it is possible
- *	to modify the list as soon as sync() begins execution.
- *	If this sort of thing becomes necessary, an alternative version
- *	based on call_rcu() could be created.  But only if -really-
- *	needed -- there is no shortage of RCU API members.
- */
-static inline void list_splice_init_rcu(struct list_head *list,
-					struct list_head *head,
-					void (*sync)(void))
-{
-	struct list_head *first = list->next;
-	struct list_head *last = list->prev;
-	struct list_head *at = head->next;
-
-	if (list_empty(head))
-		return;
-
-	/* "first" and "last" tracking list, so initialize it. */
-
-	INIT_LIST_HEAD(list);
-
-	/*
-	 * At this point, the list body still points to the source list.
-	 * Wait for any readers to finish using the list before splicing
-	 * the list body into the new list.  Any new readers will see
-	 * an empty list.
-	 */
-
-	sync();
-
-	/*
-	 * Readers are finished with the source list, so perform splice.
-	 * The order is important if the new list is global and accessible
-	 * to concurrent RCU readers.  Note that RCU readers are not
-	 * permitted to traverse the prev pointers without excluding
-	 * this function.
-	 */
-
-	last->next = at;
-	smp_wmb();
-	head->next = first;
-	first->prev = head;
-	at->prev = last;
-}
-
 /**
  * list_entry - get the struct for this entry
  * @ptr:	the &struct list_head pointer.
@@ -629,57 +465,6 @@ static inline void list_splice_init_rcu(struct list_head *list,
 	     &pos->member != (head); 					\
 	     pos = n, n = list_entry(n->member.prev, typeof(*n), member))
 
-/**
- * list_for_each_rcu	-	iterate over an rcu-protected list
- * @pos:	the &struct list_head to use as a loop cursor.
- * @head:	the head for your list.
- *
- * This list-traversal primitive may safely run concurrently with
- * the _rcu list-mutation primitives such as list_add_rcu()
- * as long as the traversal is guarded by rcu_read_lock().
- */
-#define list_for_each_rcu(pos, head) \
-	for (pos = rcu_dereference((head)->next); \
-		prefetch(pos->next), pos != (head); \
-		pos = rcu_dereference(pos->next))
-
-#define __list_for_each_rcu(pos, head) \
-	for (pos = rcu_dereference((head)->next); \
-		pos != (head); \
-		pos = rcu_dereference(pos->next))
-
-/**
- * list_for_each_entry_rcu	-	iterate over rcu list of given type
- * @pos:	the type * to use as a loop cursor.
- * @head:	the head for your list.
- * @member:	the name of the list_struct within the struct.
- *
- * This list-traversal primitive may safely run concurrently with
- * the _rcu list-mutation primitives such as list_add_rcu()
- * as long as the traversal is guarded by rcu_read_lock().
- */
-#define list_for_each_entry_rcu(pos, head, member) \
-	for (pos = list_entry(rcu_dereference((head)->next), typeof(*pos), member); \
-		prefetch(pos->member.next), &pos->member != (head); \
-		pos = list_entry(rcu_dereference(pos->member.next), typeof(*pos), member))
-
-
-/**
- * list_for_each_continue_rcu
- * @pos:	the &struct list_head to use as a loop cursor.
- * @head:	the head for your list.
- *
- * Iterate over an rcu-protected list, continuing after current point.
- *
- * This list-traversal primitive may safely run concurrently with
- * the _rcu list-mutation primitives such as list_add_rcu()
- * as long as the traversal is guarded by rcu_read_lock().
- */
-#define list_for_each_continue_rcu(pos, head) \
-	for ((pos) = rcu_dereference((pos)->next); \
-		prefetch((pos)->next), (pos) != (head); \
-		(pos) = rcu_dereference((pos)->next))
-
 /*
  * Double linked lists with a single pointer list head.
  * Mostly useful for hash tables where the two pointer list head is
@@ -730,31 +515,6 @@ static inline void hlist_del(struct hlist_node *n)
 	n->pprev = LIST_POISON2;
 }
 
-/**
- * hlist_del_rcu - deletes entry from hash list without re-initialization
- * @n: the element to delete from the hash list.
- *
- * Note: list_unhashed() on entry does not return true after this,
- * the entry is in an undefined state. It is useful for RCU based
- * lockfree traversal.
- *
- * In particular, it means that we can not poison the forward
- * pointers that may still be used for walking the hash list.
- *
- * The caller must take whatever precautions are necessary
- * (such as holding appropriate locks) to avoid racing
- * with another list-mutation primitive, such as hlist_add_head_rcu()
- * or hlist_del_rcu(), running on this same list.
- * However, it is perfectly legal to run concurrently with
- * the _rcu list-traversal primitives, such as
- * hlist_for_each_entry().
- */
-static inline void hlist_del_rcu(struct hlist_node *n)
-{
-	__hlist_del(n);
-	n->pprev = LIST_POISON2;
-}
-
 static inline void hlist_del_init(struct hlist_node *n)
 {
 	if (!hlist_unhashed(n)) {
@@ -763,27 +523,6 @@ static inline void hlist_del_init(struct hlist_node *n)
 	}
 }
 
-/**
- * hlist_replace_rcu - replace old entry by new one
- * @old : the element to be replaced
- * @new : the new element to insert
- *
- * The @old entry will be replaced with the @new entry atomically.
- */
-static inline void hlist_replace_rcu(struct hlist_node *old,
-					struct hlist_node *new)
-{
-	struct hlist_node *next = old->next;
-
-	new->next = next;
-	new->pprev = old->pprev;
-	smp_wmb();
-	if (next)
-		new->next->pprev = &new->next;
-	*new->pprev = new;
-	old->pprev = LIST_POISON2;
-}
-
 static inline void hlist_add_head(struct hlist_node *n, struct hlist_head *h)
 {
 	struct hlist_node *first = h->first;
@@ -794,38 +533,6 @@ static inline void hlist_add_head(struct hlist_node *n, struct hlist_head *h)
 	n->pprev = &h->first;
 }
 
-
-/**
- * hlist_add_head_rcu
- * @n: the element to add to the hash list.
- * @h: the list to add to.
- *
- * Description:
- * Adds the specified element to the specified hlist,
- * while permitting racing traversals.
- *
- * The caller must take whatever precautions are necessary
- * (such as holding appropriate locks) to avoid racing
- * with another list-mutation primitive, such as hlist_add_head_rcu()
- * or hlist_del_rcu(), running on this same list.
- * However, it is perfectly legal to run concurrently with
- * the _rcu list-traversal primitives, such as
- * hlist_for_each_entry_rcu(), used to prevent memory-consistency
- * problems on Alpha CPUs.  Regardless of the type of CPU, the
- * list-traversal primitive must be guarded by rcu_read_lock().
- */
-static inline void hlist_add_head_rcu(struct hlist_node *n,
-					struct hlist_head *h)
-{
-	struct hlist_node *first = h->first;
-	n->next = first;
-	n->pprev = &h->first;
-	smp_wmb();
-	if (first)
-		first->pprev = &n->next;
-	h->first = n;
-}
-
 /* next must be != NULL */
 static inline void hlist_add_before(struct hlist_node *n,
 					struct hlist_node *next)
@@ -847,63 +554,6 @@ static inline void hlist_add_after(struct hlist_node *n,
 		next->next->pprev  = &next->next;
 }
 
-/**
- * hlist_add_before_rcu
- * @n: the new element to add to the hash list.
- * @next: the existing element to add the new element before.
- *
- * Description:
- * Adds the specified element to the specified hlist
- * before the specified node while permitting racing traversals.
- *
- * The caller must take whatever precautions are necessary
- * (such as holding appropriate locks) to avoid racing
- * with another list-mutation primitive, such as hlist_add_head_rcu()
- * or hlist_del_rcu(), running on this same list.
- * However, it is perfectly legal to run concurrently with
- * the _rcu list-traversal primitives, such as
- * hlist_for_each_entry_rcu(), used to prevent memory-consistency
- * problems on Alpha CPUs.
- */
-static inline void hlist_add_before_rcu(struct hlist_node *n,
-					struct hlist_node *next)
-{
-	n->pprev = next->pprev;
-	n->next = next;
-	smp_wmb();
-	next->pprev = &n->next;
-	*(n->pprev) = n;
-}
-
-/**
- * hlist_add_after_rcu
- * @prev: the existing element to add the new element after.
- * @n: the new element to add to the hash list.
- *
- * Description:
- * Adds the specified element to the specified hlist
- * after the specified node while permitting racing traversals.
- *
- * The caller must take whatever precautions are necessary
- * (such as holding appropriate locks) to avoid racing
- * with another list-mutation primitive, such as hlist_add_head_rcu()
- * or hlist_del_rcu(), running on this same list.
- * However, it is perfectly legal to run concurrently with
- * the _rcu list-traversal primitives, such as
- * hlist_for_each_entry_rcu(), used to prevent memory-consistency
- * problems on Alpha CPUs.
- */
-static inline void hlist_add_after_rcu(struct hlist_node *prev,
-				       struct hlist_node *n)
-{
-	n->next = prev->next;
-	n->pprev = &prev->next;
-	smp_wmb();
-	prev->next = n;
-	if (n->next)
-		n->next->pprev = &n->next;
-}
-
 #define hlist_entry(ptr, type, member) container_of(ptr,type,member)
 
 #define hlist_for_each(pos, head) \
@@ -964,21 +614,4 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev,
 		({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
 	     pos = n)
 
-/**
- * hlist_for_each_entry_rcu - iterate over rcu list of given type
- * @tpos:	the type * to use as a loop cursor.
- * @pos:	the &struct hlist_node to use as a loop cursor.
- * @head:	the head for your list.
- * @member:	the name of the hlist_node within the struct.
- *
- * This list-traversal primitive may safely run concurrently with
- * the _rcu list-mutation primitives such as hlist_add_head_rcu()
- * as long as the traversal is guarded by rcu_read_lock().
- */
-#define hlist_for_each_entry_rcu(tpos, pos, head, member)		 \
-	for (pos = rcu_dereference((head)->first);			 \
-	        pos && ({ prefetch(pos->next); 1;}) &&			 \
-		({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
-	     pos = rcu_dereference(pos->next))
-
 #endif
diff --git a/include/linux/rculist.h b/include/linux/rculist.h
new file mode 100644
index 000000000000..aa9b3eb15683
--- /dev/null
+++ b/include/linux/rculist.h
@@ -0,0 +1,396 @@
+#ifndef _LINUX_RCULIST_H
+#define _LINUX_RCULIST_H
+
+#ifdef __KERNEL__
+
+/*
+ * RCU-protected list version
+ */
+#include <linux/list.h>
+
+/*
+ * Insert a new entry between two known consecutive entries.
+ *
+ * This is only for internal list manipulation where we know
+ * the prev/next entries already!
+ */
+static inline void __list_add_rcu(struct list_head *new,
+		struct list_head *prev, struct list_head *next)
+{
+	new->next = next;
+	new->prev = prev;
+	smp_wmb();
+	next->prev = new;
+	prev->next = new;
+}
+
+/**
+ * list_add_rcu - add a new entry to rcu-protected list
+ * @new: new entry to be added
+ * @head: list head to add it after
+ *
+ * Insert a new entry after the specified head.
+ * This is good for implementing stacks.
+ *
+ * The caller must take whatever precautions are necessary
+ * (such as holding appropriate locks) to avoid racing
+ * with another list-mutation primitive, such as list_add_rcu()
+ * or list_del_rcu(), running on this same list.
+ * However, it is perfectly legal to run concurrently with
+ * the _rcu list-traversal primitives, such as
+ * list_for_each_entry_rcu().
+ */
+static inline void list_add_rcu(struct list_head *new, struct list_head *head)
+{
+	__list_add_rcu(new, head, head->next);
+}
+
+/**
+ * list_add_tail_rcu - add a new entry to rcu-protected list
+ * @new: new entry to be added
+ * @head: list head to add it before
+ *
+ * Insert a new entry before the specified head.
+ * This is useful for implementing queues.
+ *
+ * The caller must take whatever precautions are necessary
+ * (such as holding appropriate locks) to avoid racing
+ * with another list-mutation primitive, such as list_add_tail_rcu()
+ * or list_del_rcu(), running on this same list.
+ * However, it is perfectly legal to run concurrently with
+ * the _rcu list-traversal primitives, such as
+ * list_for_each_entry_rcu().
+ */
+static inline void list_add_tail_rcu(struct list_head *new,
+					struct list_head *head)
+{
+	__list_add_rcu(new, head->prev, head);
+}
+
+/**
+ * list_del_rcu - deletes entry from list without re-initialization
+ * @entry: the element to delete from the list.
+ *
+ * Note: list_empty() on entry does not return true after this,
+ * the entry is in an undefined state. It is useful for RCU based
+ * lockfree traversal.
+ *
+ * In particular, it means that we can not poison the forward
+ * pointers that may still be used for walking the list.
+ *
+ * The caller must take whatever precautions are necessary
+ * (such as holding appropriate locks) to avoid racing
+ * with another list-mutation primitive, such as list_del_rcu()
+ * or list_add_rcu(), running on this same list.
+ * However, it is perfectly legal to run concurrently with
+ * the _rcu list-traversal primitives, such as
+ * list_for_each_entry_rcu().
+ *
+ * Note that the caller is not permitted to immediately free
+ * the newly deleted entry.  Instead, either synchronize_rcu()
+ * or call_rcu() must be used to defer freeing until an RCU
+ * grace period has elapsed.
+ */
+static inline void list_del_rcu(struct list_head *entry)
+{
+	__list_del(entry->prev, entry->next);
+	entry->prev = LIST_POISON2;
+}
+
+/**
+ * list_replace_rcu - replace old entry by new one
+ * @old : the element to be replaced
+ * @new : the new element to insert
+ *
+ * The @old entry will be replaced with the @new entry atomically.
+ * Note: @old should not be empty.
+ */
+static inline void list_replace_rcu(struct list_head *old,
+				struct list_head *new)
+{
+	new->next = old->next;
+	new->prev = old->prev;
+	smp_wmb();
+	new->next->prev = new;
+	new->prev->next = new;
+	old->prev = LIST_POISON2;
+}
+
+/**
+ * list_splice_init_rcu - splice an RCU-protected list into an existing list.
+ * @list:	the RCU-protected list to splice
+ * @head:	the place in the list to splice the first list into
+ * @sync:	function to sync: synchronize_rcu(), synchronize_sched(), ...
+ *
+ * @head can be RCU-read traversed concurrently with this function.
+ *
+ * Note that this function blocks.
+ *
+ * Important note: the caller must take whatever action is necessary to
+ *	prevent any other updates to @head.  In principle, it is possible
+ *	to modify the list as soon as sync() begins execution.
+ *	If this sort of thing becomes necessary, an alternative version
+ *	based on call_rcu() could be created.  But only if -really-
+ *	needed -- there is no shortage of RCU API members.
+ */
+static inline void list_splice_init_rcu(struct list_head *list,
+					struct list_head *head,
+					void (*sync)(void))
+{
+	struct list_head *first = list->next;
+	struct list_head *last = list->prev;
+	struct list_head *at = head->next;
+
+	if (list_empty(head))
+		return;
+
+	/* "first" and "last" tracking list, so initialize it. */
+
+	INIT_LIST_HEAD(list);
+
+	/*
+	 * At this point, the list body still points to the source list.
+	 * Wait for any readers to finish using the list before splicing
+	 * the list body into the new list.  Any new readers will see
+	 * an empty list.
+	 */
+
+	sync();
+
+	/*
+	 * Readers are finished with the source list, so perform splice.
+	 * The order is important if the new list is global and accessible
+	 * to concurrent RCU readers.  Note that RCU readers are not
+	 * permitted to traverse the prev pointers without excluding
+	 * this function.
+	 */
+
+	last->next = at;
+	smp_wmb();
+	head->next = first;
+	first->prev = head;
+	at->prev = last;
+}
+
+/**
+ * list_for_each_rcu	-	iterate over an rcu-protected list
+ * @pos:	the &struct list_head to use as a loop cursor.
+ * @head:	the head for your list.
+ *
+ * This list-traversal primitive may safely run concurrently with
+ * the _rcu list-mutation primitives such as list_add_rcu()
+ * as long as the traversal is guarded by rcu_read_lock().
+ */
+#define list_for_each_rcu(pos, head) \
+	for (pos = (head)->next; \
+		prefetch(rcu_dereference(pos)->next), pos != (head); \
+		pos = pos->next)
+
+#define __list_for_each_rcu(pos, head) \
+	for (pos = (head)->next; \
+		rcu_dereference(pos) != (head); \
+		pos = pos->next)
+
+/**
+ * list_for_each_safe_rcu
+ * @pos:	the &struct list_head to use as a loop cursor.
+ * @n:		another &struct list_head to use as temporary storage
+ * @head:	the head for your list.
+ *
+ * Iterate over an rcu-protected list, safe against removal of list entry.
+ *
+ * This list-traversal primitive may safely run concurrently with
+ * the _rcu list-mutation primitives such as list_add_rcu()
+ * as long as the traversal is guarded by rcu_read_lock().
+ */
+#define list_for_each_safe_rcu(pos, n, head) \
+	for (pos = (head)->next; \
+		n = rcu_dereference(pos)->next, pos != (head); \
+		pos = n)
+
+/**
+ * list_for_each_entry_rcu	-	iterate over rcu list of given type
+ * @pos:	the type * to use as a loop cursor.
+ * @head:	the head for your list.
+ * @member:	the name of the list_struct within the struct.
+ *
+ * This list-traversal primitive may safely run concurrently with
+ * the _rcu list-mutation primitives such as list_add_rcu()
+ * as long as the traversal is guarded by rcu_read_lock().
+ */
+#define list_for_each_entry_rcu(pos, head, member) \
+	for (pos = list_entry((head)->next, typeof(*pos), member); \
+		prefetch(rcu_dereference(pos)->member.next), \
+			&pos->member != (head); \
+		pos = list_entry(pos->member.next, typeof(*pos), member))
+
+
+/**
+ * list_for_each_continue_rcu
+ * @pos:	the &struct list_head to use as a loop cursor.
+ * @head:	the head for your list.
+ *
+ * Iterate over an rcu-protected list, continuing after current point.
+ *
+ * This list-traversal primitive may safely run concurrently with
+ * the _rcu list-mutation primitives such as list_add_rcu()
+ * as long as the traversal is guarded by rcu_read_lock().
+ */
+#define list_for_each_continue_rcu(pos, head) \
+	for ((pos) = (pos)->next; \
+		prefetch(rcu_dereference((pos))->next), (pos) != (head); \
+		(pos) = (pos)->next)
+
+/**
+ * hlist_del_rcu - deletes entry from hash list without re-initialization
+ * @n: the element to delete from the hash list.
+ *
+ * Note: list_unhashed() on entry does not return true after this,
+ * the entry is in an undefined state. It is useful for RCU based
+ * lockfree traversal.
+ *
+ * In particular, it means that we can not poison the forward
+ * pointers that may still be used for walking the hash list.
+ *
+ * The caller must take whatever precautions are necessary
+ * (such as holding appropriate locks) to avoid racing
+ * with another list-mutation primitive, such as hlist_add_head_rcu()
+ * or hlist_del_rcu(), running on this same list.
+ * However, it is perfectly legal to run concurrently with
+ * the _rcu list-traversal primitives, such as
+ * hlist_for_each_entry().
+ */
+static inline void hlist_del_rcu(struct hlist_node *n)
+{
+	__hlist_del(n);
+	n->pprev = LIST_POISON2;
+}
+
+/**
+ * hlist_replace_rcu - replace old entry by new one
+ * @old : the element to be replaced
+ * @new : the new element to insert
+ *
+ * The @old entry will be replaced with the @new entry atomically.
+ */
+static inline void hlist_replace_rcu(struct hlist_node *old,
+					struct hlist_node *new)
+{
+	struct hlist_node *next = old->next;
+
+	new->next = next;
+	new->pprev = old->pprev;
+	smp_wmb();
+	if (next)
+		new->next->pprev = &new->next;
+	*new->pprev = new;
+	old->pprev = LIST_POISON2;
+}
+
+/**
+ * hlist_add_head_rcu
+ * @n: the element to add to the hash list.
+ * @h: the list to add to.
+ *
+ * Description:
+ * Adds the specified element to the specified hlist,
+ * while permitting racing traversals.
+ *
+ * The caller must take whatever precautions are necessary
+ * (such as holding appropriate locks) to avoid racing
+ * with another list-mutation primitive, such as hlist_add_head_rcu()
+ * or hlist_del_rcu(), running on this same list.
+ * However, it is perfectly legal to run concurrently with
+ * the _rcu list-traversal primitives, such as
+ * hlist_for_each_entry_rcu(), used to prevent memory-consistency
+ * problems on Alpha CPUs.  Regardless of the type of CPU, the
+ * list-traversal primitive must be guarded by rcu_read_lock().
+ */
+static inline void hlist_add_head_rcu(struct hlist_node *n,
+					struct hlist_head *h)
+{
+	struct hlist_node *first = h->first;
+	n->next = first;
+	n->pprev = &h->first;
+	smp_wmb();
+	if (first)
+		first->pprev = &n->next;
+	h->first = n;
+}
+
+/**
+ * hlist_add_before_rcu
+ * @n: the new element to add to the hash list.
+ * @next: the existing element to add the new element before.
+ *
+ * Description:
+ * Adds the specified element to the specified hlist
+ * before the specified node while permitting racing traversals.
+ *
+ * The caller must take whatever precautions are necessary
+ * (such as holding appropriate locks) to avoid racing
+ * with another list-mutation primitive, such as hlist_add_head_rcu()
+ * or hlist_del_rcu(), running on this same list.
+ * However, it is perfectly legal to run concurrently with
+ * the _rcu list-traversal primitives, such as
+ * hlist_for_each_entry_rcu(), used to prevent memory-consistency
+ * problems on Alpha CPUs.
+ */
+static inline void hlist_add_before_rcu(struct hlist_node *n,
+					struct hlist_node *next)
+{
+	n->pprev = next->pprev;
+	n->next = next;
+	smp_wmb();
+	next->pprev = &n->next;
+	*(n->pprev) = n;
+}
+
+/**
+ * hlist_add_after_rcu
+ * @prev: the existing element to add the new element after.
+ * @n: the new element to add to the hash list.
+ *
+ * Description:
+ * Adds the specified element to the specified hlist
+ * after the specified node while permitting racing traversals.
+ *
+ * The caller must take whatever precautions are necessary
+ * (such as holding appropriate locks) to avoid racing
+ * with another list-mutation primitive, such as hlist_add_head_rcu()
+ * or hlist_del_rcu(), running on this same list.
+ * However, it is perfectly legal to run concurrently with
+ * the _rcu list-traversal primitives, such as
+ * hlist_for_each_entry_rcu(), used to prevent memory-consistency
+ * problems on Alpha CPUs.
+ */
+static inline void hlist_add_after_rcu(struct hlist_node *prev,
+				       struct hlist_node *n)
+{
+	n->next = prev->next;
+	n->pprev = &prev->next;
+	smp_wmb();
+	prev->next = n;
+	if (n->next)
+		n->next->pprev = &n->next;
+}
+
+/**
+ * hlist_for_each_entry_rcu - iterate over rcu list of given type
+ * @tpos:	the type * to use as a loop cursor.
+ * @pos:	the &struct hlist_node to use as a loop cursor.
+ * @head:	the head for your list.
+ * @member:	the name of the hlist_node within the struct.
+ *
+ * This list-traversal primitive may safely run concurrently with
+ * the _rcu list-mutation primitives such as hlist_add_head_rcu()
+ * as long as the traversal is guarded by rcu_read_lock().
+ */
+#define hlist_for_each_entry_rcu(tpos, pos, head, member)		 \
+	for (pos = (head)->first;					 \
+	     rcu_dereference(pos) && ({ prefetch(pos->next); 1; }) &&	 \
+		({ tpos = hlist_entry(pos, typeof(*tpos), member); 1; }); \
+	     pos = pos->next)
+
+#endif	/* __KERNEL__ */
+#endif
diff --git a/kernel/pid.c b/kernel/pid.c
index 20d59fa2d493..30bd5d4b2ac7 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -30,6 +30,7 @@
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/init.h>
+#include <linux/rculist.h>
 #include <linux/bootmem.h>
 #include <linux/hash.h>
 #include <linux/pid_namespace.h>
diff --git a/lib/textsearch.c b/lib/textsearch.c
index be8bda3862f5..a3e500ad51d7 100644
--- a/lib/textsearch.c
+++ b/lib/textsearch.c
@@ -97,6 +97,7 @@
 #include <linux/types.h>
 #include <linux/string.h>
 #include <linux/init.h>
+#include <linux/rculist.h>
 #include <linux/rcupdate.h>
 #include <linux/err.h>
 #include <linux/textsearch.h>
diff --git a/net/802/psnap.c b/net/802/psnap.c
index 31128cb92a23..ea4643931446 100644
--- a/net/802/psnap.c
+++ b/net/802/psnap.c
@@ -20,6 +20,7 @@
 #include <linux/mm.h>
 #include <linux/in.h>
 #include <linux/init.h>
+#include <linux/rculist.h>
 
 static LIST_HEAD(snap_list);
 static DEFINE_SPINLOCK(snap_lock);
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 2a739adaa92b..e7ddbfa0e02f 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -27,6 +27,7 @@
 #include <linux/mm.h>
 #include <linux/in.h>
 #include <linux/init.h>
+#include <linux/rculist.h>
 #include <net/p8022.h>
 #include <net/arp.h>
 #include <linux/rtnetlink.h>
diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
index 72c5976a5ce3..142060f02054 100644
--- a/net/bridge/br_fdb.c
+++ b/net/bridge/br_fdb.c
@@ -15,6 +15,7 @@
 
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include <linux/rculist.h>
 #include <linux/spinlock.h>
 #include <linux/times.h>
 #include <linux/netdevice.h>
diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c
index e38034aa56f5..9e96ffcd29a3 100644
--- a/net/bridge/br_stp.c
+++ b/net/bridge/br_stp.c
@@ -13,6 +13,7 @@
  *	2 of the License, or (at your option) any later version.
  */
 #include <linux/kernel.h>
+#include <linux/rculist.h>
 
 #include "br_private.h"
 #include "br_private_stp.h"
diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index 02c2f7c0b255..643c032a3a57 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -30,8 +30,7 @@
  */
 
 #include <linux/types.h>
-#include <linux/rcupdate.h>
-#include <linux/list.h>
+#include <linux/rculist.h>
 #include <linux/skbuff.h>
 #include <linux/spinlock.h>
 #include <linux/string.h>
-- 
cgit v1.2.3


From 10aa9d2cf9878757b003023d33ff90a37aa3044b Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <fbuihuu@gmail.com>
Date: Mon, 12 May 2008 21:21:06 +0200
Subject: rculist.h: use the rcu API

Make almost all list mutation primitives use rcu_assign_pointer().

The main point of this being readability improvement.

Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Josh Triplett <josh@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/rculist.h | 23 +++++++++--------------
 1 file changed, 9 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index aa9b3eb15683..8d2c81fccfe5 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -7,6 +7,7 @@
  * RCU-protected list version
  */
 #include <linux/list.h>
+#include <linux/rcupdate.h>
 
 /*
  * Insert a new entry between two known consecutive entries.
@@ -19,9 +20,8 @@ static inline void __list_add_rcu(struct list_head *new,
 {
 	new->next = next;
 	new->prev = prev;
-	smp_wmb();
+	rcu_assign_pointer(prev->next, new);
 	next->prev = new;
-	prev->next = new;
 }
 
 /**
@@ -110,9 +110,8 @@ static inline void list_replace_rcu(struct list_head *old,
 {
 	new->next = old->next;
 	new->prev = old->prev;
-	smp_wmb();
+	rcu_assign_pointer(new->prev->next, new);
 	new->next->prev = new;
-	new->prev->next = new;
 	old->prev = LIST_POISON2;
 }
 
@@ -166,8 +165,7 @@ static inline void list_splice_init_rcu(struct list_head *list,
 	 */
 
 	last->next = at;
-	smp_wmb();
-	head->next = first;
+	rcu_assign_pointer(head->next, first);
 	first->prev = head;
 	at->prev = last;
 }
@@ -280,10 +278,9 @@ static inline void hlist_replace_rcu(struct hlist_node *old,
 
 	new->next = next;
 	new->pprev = old->pprev;
-	smp_wmb();
+	rcu_assign_pointer(*new->pprev, new);
 	if (next)
 		new->next->pprev = &new->next;
-	*new->pprev = new;
 	old->pprev = LIST_POISON2;
 }
 
@@ -310,12 +307,12 @@ static inline void hlist_add_head_rcu(struct hlist_node *n,
 					struct hlist_head *h)
 {
 	struct hlist_node *first = h->first;
+
 	n->next = first;
 	n->pprev = &h->first;
-	smp_wmb();
+	rcu_assign_pointer(h->first, n);
 	if (first)
 		first->pprev = &n->next;
-	h->first = n;
 }
 
 /**
@@ -341,9 +338,8 @@ static inline void hlist_add_before_rcu(struct hlist_node *n,
 {
 	n->pprev = next->pprev;
 	n->next = next;
-	smp_wmb();
+	rcu_assign_pointer(*(n->pprev), n);
 	next->pprev = &n->next;
-	*(n->pprev) = n;
 }
 
 /**
@@ -369,8 +365,7 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev,
 {
 	n->next = prev->next;
 	n->pprev = &prev->next;
-	smp_wmb();
-	prev->next = n;
+	rcu_assign_pointer(prev->next, n);
 	if (n->next)
 		n->next->pprev = &n->next;
 }
-- 
cgit v1.2.3


From 78b0e0e9b27b62c4b22f05a147f7a80fa58b1ae3 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Mon, 12 May 2008 21:21:06 +0200
Subject: RCU, rculist.h: fix list iterators

RCU list iterators: should prefetch ever be optimised out with no
side-effects, the current version will lose the barrier completely.

Pointed-out-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/rculist.h | 48 +++++++++++++++---------------------------------
 1 file changed, 15 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index 8d2c81fccfe5..b0f39be08b6c 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -180,31 +180,14 @@ static inline void list_splice_init_rcu(struct list_head *list,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_rcu(pos, head) \
-	for (pos = (head)->next; \
-		prefetch(rcu_dereference(pos)->next), pos != (head); \
-		pos = pos->next)
+	for (pos = rcu_dereference((head)->next); \
+		prefetch(pos->next), pos != (head); \
+		pos = rcu_dereference(pos->next))
 
 #define __list_for_each_rcu(pos, head) \
-	for (pos = (head)->next; \
-		rcu_dereference(pos) != (head); \
-		pos = pos->next)
-
-/**
- * list_for_each_safe_rcu
- * @pos:	the &struct list_head to use as a loop cursor.
- * @n:		another &struct list_head to use as temporary storage
- * @head:	the head for your list.
- *
- * Iterate over an rcu-protected list, safe against removal of list entry.
- *
- * This list-traversal primitive may safely run concurrently with
- * the _rcu list-mutation primitives such as list_add_rcu()
- * as long as the traversal is guarded by rcu_read_lock().
- */
-#define list_for_each_safe_rcu(pos, n, head) \
-	for (pos = (head)->next; \
-		n = rcu_dereference(pos)->next, pos != (head); \
-		pos = n)
+	for (pos = rcu_dereference((head)->next); \
+		pos != (head); \
+		pos = rcu_dereference(pos->next))
 
 /**
  * list_for_each_entry_rcu	-	iterate over rcu list of given type
@@ -217,10 +200,9 @@ static inline void list_splice_init_rcu(struct list_head *list,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_entry_rcu(pos, head, member) \
-	for (pos = list_entry((head)->next, typeof(*pos), member); \
-		prefetch(rcu_dereference(pos)->member.next), \
-			&pos->member != (head); \
-		pos = list_entry(pos->member.next, typeof(*pos), member))
+	for (pos = list_entry(rcu_dereference((head)->next), typeof(*pos), member); \
+		prefetch(pos->member.next), &pos->member != (head); \
+		pos = list_entry(rcu_dereference(pos->member.next), typeof(*pos), member))
 
 
 /**
@@ -235,9 +217,9 @@ static inline void list_splice_init_rcu(struct list_head *list,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_continue_rcu(pos, head) \
-	for ((pos) = (pos)->next; \
-		prefetch(rcu_dereference((pos))->next), (pos) != (head); \
-		(pos) = (pos)->next)
+	for ((pos) = rcu_dereference((pos)->next); \
+		prefetch((pos)->next), (pos) != (head); \
+		(pos) = rcu_dereference((pos)->next))
 
 /**
  * hlist_del_rcu - deletes entry from hash list without re-initialization
@@ -382,10 +364,10 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define hlist_for_each_entry_rcu(tpos, pos, head, member)		 \
-	for (pos = (head)->first;					 \
-	     rcu_dereference(pos) && ({ prefetch(pos->next); 1; }) &&	 \
+	for (pos = rcu_dereference((head)->first);			 \
+		pos && ({ prefetch(pos->next); 1; }) &&			 \
 		({ tpos = hlist_entry(pos, typeof(*tpos), member); 1; }); \
-	     pos = pos->next)
+		pos = rcu_dereference(pos->next))
 
 #endif	/* __KERNEL__ */
 #endif
-- 
cgit v1.2.3


From 1a189b97190d3f0f8cf0379a799d3555b2d648bb Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sun, 13 Apr 2008 21:41:55 +0100
Subject: [ARM] pxa: Add bare bones PWM API

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/Kconfig    |  3 +++
 include/linux/pwm.h | 31 +++++++++++++++++++++++++++++++
 2 files changed, 34 insertions(+)
 create mode 100644 include/linux/pwm.h

(limited to 'include/linux')

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index b786e68914d4..c274dbb89a83 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -22,6 +22,9 @@ config ARM
 	  Europe.  There is an ARM Linux project with a web page at
 	  <http://www.arm.linux.org.uk/>.
 
+config HAVE_PWM
+	bool
+
 config SYS_SUPPORTS_APM_EMULATION
 	bool
 
diff --git a/include/linux/pwm.h b/include/linux/pwm.h
new file mode 100644
index 000000000000..3945f803d514
--- /dev/null
+++ b/include/linux/pwm.h
@@ -0,0 +1,31 @@
+#ifndef __LINUX_PWM_H
+#define __LINUX_PWM_H
+
+struct pwm_device;
+
+/*
+ * pwm_request - request a PWM device
+ */
+struct pwm_device *pwm_request(int pwm_id, const char *label);
+
+/*
+ * pwm_free - free a PWM device
+ */
+void pwm_free(struct pwm_device *pwm);
+
+/*
+ * pwm_config - change a PWM device configuration
+ */
+int pwm_config(struct pwm_device *pwm, int duty_ns, int period_ns);
+
+/*
+ * pwm_enable - start a PWM output toggling
+ */
+int pwm_enable(struct pwm_device *pwm);
+
+/*
+ * pwm_disable - stop a PWM output toggling
+ */
+void pwm_disable(struct pwm_device *pwm);
+
+#endif /* __ASM_ARCH_PWM_H */
-- 
cgit v1.2.3


From 48adcf148c83faa41999fb0b3524299c4e160fd9 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 20 May 2008 01:03:16 +0300
Subject: [CPUFREQ] cpufreq: remove CVS keywords

This patch removes CVS keywords that weren't updated for a long time
from comments.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/powernow-k7.h | 1 -
 include/linux/cpufreq.h                   | 3 ---
 2 files changed, 4 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.h b/arch/x86/kernel/cpu/cpufreq/powernow-k7.h
index f8a63b3664e3..35fb4eaf6e1c 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.h
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.h
@@ -1,5 +1,4 @@
 /*
- *  $Id: powernow-k7.h,v 1.2 2003/02/10 18:26:01 davej Exp $
  *  (C) 2003 Dave Jones.
  *
  *  Licensed under the terms of the GNU GPL License version 2.
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index e7e91dbfde0f..2270ca5ec631 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -4,9 +4,6 @@
  *  Copyright (C) 2001 Russell King
  *            (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de>
  *            
- *
- * $Id: cpufreq.h,v 1.36 2003/01/20 17:31:48 db Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
-- 
cgit v1.2.3


From d1659fcc59b21ec442564fedb67a5ad371f82380 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 20 May 2008 12:17:39 -0400
Subject: Input: remove CVS keywords

This patch removes CVS keywords that weren't updated for a long time
from comments.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 Documentation/input/gameport-programming.txt   | 2 --
 Documentation/input/input.txt                  | 1 -
 Documentation/input/joystick-api.txt           | 2 --
 Documentation/input/joystick-parport.txt       | 1 -
 Documentation/input/joystick.txt               | 1 -
 drivers/input/evbug.c                          | 2 --
 drivers/input/gameport/emu10k1-gp.c            | 2 --
 drivers/input/gameport/lightning.c             | 2 --
 drivers/input/gameport/ns558.c                 | 2 --
 drivers/input/joystick/a3d.c                   | 2 --
 drivers/input/joystick/amijoy.c                | 2 --
 drivers/input/joystick/cobra.c                 | 2 --
 drivers/input/joystick/db9.c                   | 2 --
 drivers/input/joystick/gf2k.c                  | 2 --
 drivers/input/joystick/grip.c                  | 2 --
 drivers/input/joystick/grip_mp.c               | 2 --
 drivers/input/joystick/guillemot.c             | 2 --
 drivers/input/joystick/iforce/iforce-ff.c      | 2 --
 drivers/input/joystick/iforce/iforce-main.c    | 2 --
 drivers/input/joystick/iforce/iforce-packets.c | 2 --
 drivers/input/joystick/iforce/iforce-serio.c   | 2 --
 drivers/input/joystick/iforce/iforce-usb.c     | 2 --
 drivers/input/joystick/iforce/iforce.h         | 2 --
 drivers/input/joystick/interact.c              | 2 --
 drivers/input/joystick/joydump.c               | 2 --
 drivers/input/joystick/magellan.c              | 2 --
 drivers/input/joystick/spaceball.c             | 2 --
 drivers/input/joystick/spaceorb.c              | 2 --
 drivers/input/joystick/stinger.c               | 2 --
 drivers/input/joystick/tmdc.c                  | 2 --
 drivers/input/joystick/turbografx.c            | 2 --
 drivers/input/joystick/twidjoy.c               | 4 ----
 drivers/input/joystick/warrior.c               | 2 --
 drivers/input/keyboard/amikbd.c                | 2 --
 drivers/input/keyboard/sunkbd.c                | 2 --
 drivers/input/keyboard/xtkbd.c                 | 2 --
 drivers/input/mouse/inport.c                   | 2 --
 drivers/input/mouse/logibm.c                   | 2 --
 drivers/input/mouse/pc110pad.c                 | 2 --
 drivers/input/mouse/sermouse.c                 | 2 --
 drivers/input/serio/ct82c710.c                 | 2 --
 drivers/input/serio/q40kbd.c                   | 2 --
 drivers/input/serio/rpckbd.c                   | 2 --
 drivers/input/touchscreen/gunze.c              | 2 --
 drivers/input/touchscreen/h3600_ts_input.c     | 2 --
 include/linux/joystick.h                       | 2 --
 46 files changed, 91 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/input/gameport-programming.txt b/Documentation/input/gameport-programming.txt
index 14e0a8b70225..03a74fc3b496 100644
--- a/Documentation/input/gameport-programming.txt
+++ b/Documentation/input/gameport-programming.txt
@@ -1,5 +1,3 @@
-$Id: gameport-programming.txt,v 1.3 2001/04/24 13:51:37 vojtech Exp $
-
 Programming gameport drivers
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
diff --git a/Documentation/input/input.txt b/Documentation/input/input.txt
index ff8cea0225f9..686ee9932dff 100644
--- a/Documentation/input/input.txt
+++ b/Documentation/input/input.txt
@@ -1,7 +1,6 @@
 			  Linux Input drivers v1.0
 	       (c) 1999-2001 Vojtech Pavlik <vojtech@ucw.cz>
 			     Sponsored by SuSE
-	    $Id: input.txt,v 1.8 2002/05/29 03:15:01 bradleym Exp $
 ----------------------------------------------------------------------------
 
 0. Disclaimer
diff --git a/Documentation/input/joystick-api.txt b/Documentation/input/joystick-api.txt
index acbd32b88454..c507330740cd 100644
--- a/Documentation/input/joystick-api.txt
+++ b/Documentation/input/joystick-api.txt
@@ -5,8 +5,6 @@
 
 			      7 Aug 1998
 
-	$Id: joystick-api.txt,v 1.2 2001/05/08 21:21:23 vojtech Exp $
-
 1. Initialization
 ~~~~~~~~~~~~~~~~~
 
diff --git a/Documentation/input/joystick-parport.txt b/Documentation/input/joystick-parport.txt
index ede5f33daad3..1c856f32ff2c 100644
--- a/Documentation/input/joystick-parport.txt
+++ b/Documentation/input/joystick-parport.txt
@@ -2,7 +2,6 @@
 	       (c) 1998-2000 Vojtech Pavlik <vojtech@ucw.cz>
 	       (c) 1998 Andree Borrmann <a.borrmann@tu-bs.de>
 			     Sponsored by SuSE
-	$Id: joystick-parport.txt,v 1.6 2001/09/25 09:31:32 vojtech Exp $
 ----------------------------------------------------------------------------
 
 0. Disclaimer
diff --git a/Documentation/input/joystick.txt b/Documentation/input/joystick.txt
index 389de9bd9878..154d767b2acb 100644
--- a/Documentation/input/joystick.txt
+++ b/Documentation/input/joystick.txt
@@ -1,7 +1,6 @@
 		       Linux Joystick driver v2.0.0
 	       (c) 1996-2000 Vojtech Pavlik <vojtech@ucw.cz>
 			     Sponsored by SuSE
-	   $Id: joystick.txt,v 1.12 2002/03/03 12:13:07 jdeneux Exp $
 ----------------------------------------------------------------------------
 
 0. Disclaimer
diff --git a/drivers/input/evbug.c b/drivers/input/evbug.c
index c21f2f127234..67754c776d88 100644
--- a/drivers/input/evbug.c
+++ b/drivers/input/evbug.c
@@ -1,6 +1,4 @@
 /*
- * $Id: evbug.c,v 1.10 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/gameport/emu10k1-gp.c b/drivers/input/gameport/emu10k1-gp.c
index 9793ac36d17f..b04930f7ea7d 100644
--- a/drivers/input/gameport/emu10k1-gp.c
+++ b/drivers/input/gameport/emu10k1-gp.c
@@ -1,6 +1,4 @@
 /*
- * $Id: emu10k1-gp.c,v 1.8 2002/01/22 20:40:46 vojtech Exp $
- *
  *  Copyright (c) 2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/gameport/lightning.c b/drivers/input/gameport/lightning.c
index 6b4d4561d465..06ad36ed3483 100644
--- a/drivers/input/gameport/lightning.c
+++ b/drivers/input/gameport/lightning.c
@@ -1,6 +1,4 @@
 /*
- * $Id: lightning.c,v 1.20 2002/01/22 20:41:31 vojtech Exp $
- *
  *  Copyright (c) 1998-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/gameport/ns558.c b/drivers/input/gameport/ns558.c
index 7b7a546323cf..2b282cde4b89 100644
--- a/drivers/input/gameport/ns558.c
+++ b/drivers/input/gameport/ns558.c
@@ -1,6 +1,4 @@
 /*
- * $Id: ns558.c,v 1.43 2002/01/24 19:23:21 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  *  Copyright (c) 1999 Brian Gerst
  */
diff --git a/drivers/input/joystick/a3d.c b/drivers/input/joystick/a3d.c
index 52ba16f487c7..92498d470b1f 100644
--- a/drivers/input/joystick/a3d.c
+++ b/drivers/input/joystick/a3d.c
@@ -1,6 +1,4 @@
 /*
- * $Id: a3d.c,v 1.21 2002/01/22 20:11:50 vojtech Exp $
- *
  *  Copyright (c) 1998-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/amijoy.c b/drivers/input/joystick/amijoy.c
index deb9f825f92c..05022f07ec77 100644
--- a/drivers/input/joystick/amijoy.c
+++ b/drivers/input/joystick/amijoy.c
@@ -1,6 +1,4 @@
 /*
- * $Id: amijoy.c,v 1.13 2002/01/22 20:26:32 vojtech Exp $
- *
  *  Copyright (c) 1998-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/cobra.c b/drivers/input/joystick/cobra.c
index 55646a6d89f5..639b975a8ed7 100644
--- a/drivers/input/joystick/cobra.c
+++ b/drivers/input/joystick/cobra.c
@@ -1,6 +1,4 @@
 /*
- * $Id: cobra.c,v 1.19 2002/01/22 20:26:52 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/db9.c b/drivers/input/joystick/db9.c
index 960e501c60c8..523959484753 100644
--- a/drivers/input/joystick/db9.c
+++ b/drivers/input/joystick/db9.c
@@ -1,6 +1,4 @@
 /*
- * $Id: db9.c,v 1.13 2002/04/07 20:13:37 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/joystick/gf2k.c b/drivers/input/joystick/gf2k.c
index 1f6302c0eb3f..cb6eef1f2d99 100644
--- a/drivers/input/joystick/gf2k.c
+++ b/drivers/input/joystick/gf2k.c
@@ -1,6 +1,4 @@
 /*
- * $Id: gf2k.c,v 1.19 2002/01/22 20:27:43 vojtech Exp $
- *
  *  Copyright (c) 1998-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/grip.c b/drivers/input/joystick/grip.c
index fd3853ab1aad..684e07cfccc8 100644
--- a/drivers/input/joystick/grip.c
+++ b/drivers/input/joystick/grip.c
@@ -1,6 +1,4 @@
 /*
- * $Id: grip.c,v 1.21 2002/01/22 20:27:57 vojtech Exp $
- *
  *  Copyright (c) 1998-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/grip_mp.c b/drivers/input/joystick/grip_mp.c
index c57e21d68c00..8279481b16e7 100644
--- a/drivers/input/joystick/grip_mp.c
+++ b/drivers/input/joystick/grip_mp.c
@@ -1,6 +1,4 @@
 /*
- * $Id: grip_mp.c,v 1.9 2002/07/20 19:28:45 bonnland Exp $
- *
  *  Driver for the Gravis Grip Multiport, a gamepad "hub" that
  *  connects up to four 9-pin digital gamepads/joysticks.
  *  Driver tested on SMP and UP kernel versions 2.4.18-4 and 2.4.18-5.
diff --git a/drivers/input/joystick/guillemot.c b/drivers/input/joystick/guillemot.c
index aa6bfb3fb8cd..25ec3fad9f27 100644
--- a/drivers/input/joystick/guillemot.c
+++ b/drivers/input/joystick/guillemot.c
@@ -1,6 +1,4 @@
 /*
- * $Id: guillemot.c,v 1.10 2002/01/22 20:28:12 vojtech Exp $
- *
  *  Copyright (c) 2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/iforce/iforce-ff.c b/drivers/input/joystick/iforce/iforce-ff.c
index f2a4381d0ab8..7839b7b6fa96 100644
--- a/drivers/input/joystick/iforce/iforce-ff.c
+++ b/drivers/input/joystick/iforce/iforce-ff.c
@@ -1,6 +1,4 @@
 /*
- * $Id: iforce-ff.c,v 1.9 2002/02/02 19:28:35 jdeneux Exp $
- *
  *  Copyright (c) 2000-2002 Vojtech Pavlik <vojtech@ucw.cz>
  *  Copyright (c) 2001-2002, 2007 Johann Deneux <johann.deneux@gmail.com>
  *
diff --git a/drivers/input/joystick/iforce/iforce-main.c b/drivers/input/joystick/iforce/iforce-main.c
index a2517fa72eb8..61ee6e38739d 100644
--- a/drivers/input/joystick/iforce/iforce-main.c
+++ b/drivers/input/joystick/iforce/iforce-main.c
@@ -1,6 +1,4 @@
 /*
- * $Id: iforce-main.c,v 1.19 2002/07/07 10:22:50 jdeneux Exp $
- *
  *  Copyright (c) 2000-2002 Vojtech Pavlik <vojtech@ucw.cz>
  *  Copyright (c) 2001-2002, 2007 Johann Deneux <johann.deneux@gmail.com>
  *
diff --git a/drivers/input/joystick/iforce/iforce-packets.c b/drivers/input/joystick/iforce/iforce-packets.c
index 45c4939ced75..015b50aa76fc 100644
--- a/drivers/input/joystick/iforce/iforce-packets.c
+++ b/drivers/input/joystick/iforce/iforce-packets.c
@@ -1,6 +1,4 @@
 /*
- * $Id: iforce-packets.c,v 1.16 2002/07/07 10:22:50 jdeneux Exp $
- *
  *  Copyright (c) 2000-2002 Vojtech Pavlik <vojtech@ucw.cz>
  *  Copyright (c) 2001-2002, 2007 Johann Deneux <johann.deneux@gmail.com>
  *
diff --git a/drivers/input/joystick/iforce/iforce-serio.c b/drivers/input/joystick/iforce/iforce-serio.c
index 7b4bc19cef27..46d5041d2d9d 100644
--- a/drivers/input/joystick/iforce/iforce-serio.c
+++ b/drivers/input/joystick/iforce/iforce-serio.c
@@ -1,6 +1,4 @@
 /*
- * $Id: iforce-serio.c,v 1.4 2002/01/28 22:45:00 jdeneux Exp $
- *
  *  Copyright (c) 2000-2001 Vojtech Pavlik <vojtech@ucw.cz>
  *  Copyright (c) 2001, 2007 Johann Deneux <johann.deneux@gmail.com>
  *
diff --git a/drivers/input/joystick/iforce/iforce-usb.c b/drivers/input/joystick/iforce/iforce-usb.c
index 8f6a987ebb9f..851cc4087c2f 100644
--- a/drivers/input/joystick/iforce/iforce-usb.c
+++ b/drivers/input/joystick/iforce/iforce-usb.c
@@ -1,6 +1,4 @@
  /*
- * $Id: iforce-usb.c,v 1.16 2002/06/09 11:08:04 jdeneux Exp $
- *
  *  Copyright (c) 2000-2002 Vojtech Pavlik <vojtech@ucw.cz>
  *  Copyright (c) 2001-2002, 2007 Johann Deneux <johann.deneux@gmail.com>
  *
diff --git a/drivers/input/joystick/iforce/iforce.h b/drivers/input/joystick/iforce/iforce.h
index a964a7cfd210..f2d91f4028ca 100644
--- a/drivers/input/joystick/iforce/iforce.h
+++ b/drivers/input/joystick/iforce/iforce.h
@@ -1,6 +1,4 @@
 /*
- * $Id: iforce.h,v 1.13 2002/07/07 10:22:50 jdeneux Exp $
- *
  *  Copyright (c) 2000-2002 Vojtech Pavlik <vojtech@ucw.cz>
  *  Copyright (c) 2001-2002, 2007 Johann Deneux <johann.deneux@gmail.com>
  *
diff --git a/drivers/input/joystick/interact.c b/drivers/input/joystick/interact.c
index bc8ea95dfd0e..8c3290b68205 100644
--- a/drivers/input/joystick/interact.c
+++ b/drivers/input/joystick/interact.c
@@ -1,6 +1,4 @@
 /*
- * $Id: interact.c,v 1.16 2002/01/22 20:28:25 vojtech Exp $
- *
  *  Copyright (c) 2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/joystick/joydump.c b/drivers/input/joystick/joydump.c
index 88ec5a918f2e..2a1b82c8b31c 100644
--- a/drivers/input/joystick/joydump.c
+++ b/drivers/input/joystick/joydump.c
@@ -1,6 +1,4 @@
 /*
- * $Id: joydump.c,v 1.1 2002/01/23 06:56:16 jsimmons Exp $
- *
  *  Copyright (c) 1996-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/magellan.c b/drivers/input/joystick/magellan.c
index 54e676948ebb..40e40780747d 100644
--- a/drivers/input/joystick/magellan.c
+++ b/drivers/input/joystick/magellan.c
@@ -1,6 +1,4 @@
 /*
- * $Id: magellan.c,v 1.16 2002/01/22 20:28:39 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/joystick/spaceball.c b/drivers/input/joystick/spaceball.c
index d4087fd49656..0cd9b29356a8 100644
--- a/drivers/input/joystick/spaceball.c
+++ b/drivers/input/joystick/spaceball.c
@@ -1,6 +1,4 @@
 /*
- * $Id: spaceball.c,v 1.17 2002/01/22 20:29:03 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/joystick/spaceorb.c b/drivers/input/joystick/spaceorb.c
index f7ce4004f4ba..a694bf8e557b 100644
--- a/drivers/input/joystick/spaceorb.c
+++ b/drivers/input/joystick/spaceorb.c
@@ -1,6 +1,4 @@
 /*
- * $Id: spaceorb.c,v 1.15 2002/01/22 20:29:19 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/joystick/stinger.c b/drivers/input/joystick/stinger.c
index baa10b2f7ba1..e0db9f5e4b41 100644
--- a/drivers/input/joystick/stinger.c
+++ b/drivers/input/joystick/stinger.c
@@ -1,6 +1,4 @@
 /*
- * $Id: stinger.c,v 1.10 2002/01/22 20:29:31 vojtech Exp $
- *
  *  Copyright (c) 2000-2001 Vojtech Pavlik
  *  Copyright (c) 2000 Mark Fletcher
  */
diff --git a/drivers/input/joystick/tmdc.c b/drivers/input/joystick/tmdc.c
index 0feeb8acb532..60c37bcb938d 100644
--- a/drivers/input/joystick/tmdc.c
+++ b/drivers/input/joystick/tmdc.c
@@ -1,6 +1,4 @@
 /*
- * $Id: tmdc.c,v 1.31 2002/01/22 20:29:52 vojtech Exp $
- *
  *  Copyright (c) 1998-2001 Vojtech Pavlik
  *
  *   Based on the work of:
diff --git a/drivers/input/joystick/turbografx.c b/drivers/input/joystick/turbografx.c
index 989483f53160..b6f859869540 100644
--- a/drivers/input/joystick/turbografx.c
+++ b/drivers/input/joystick/turbografx.c
@@ -1,6 +1,4 @@
 /*
- * $Id: turbografx.c,v 1.14 2002/01/22 20:30:39 vojtech Exp $
- *
  *  Copyright (c) 1998-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/joystick/twidjoy.c b/drivers/input/joystick/twidjoy.c
index 1085c841fec4..3f4ec73c9553 100644
--- a/drivers/input/joystick/twidjoy.c
+++ b/drivers/input/joystick/twidjoy.c
@@ -1,8 +1,4 @@
 /*
- * $Id: twidjoy.c,v 1.5 2002/01/22 20:31:53 vojtech Exp $
- *
- *  derived from CVS-ID "stinger.c,v 1.5 2001/05/29 12:57:18 vojtech Exp"
- *
  *  Copyright (c) 2001 Arndt Schoenewald
  *  Copyright (c) 2000-2001 Vojtech Pavlik
  *  Copyright (c) 2000 Mark Fletcher
diff --git a/drivers/input/joystick/warrior.c b/drivers/input/joystick/warrior.c
index e928b6e3724a..f72c83e15e60 100644
--- a/drivers/input/joystick/warrior.c
+++ b/drivers/input/joystick/warrior.c
@@ -1,6 +1,4 @@
 /*
- * $Id: warrior.c,v 1.14 2002/01/22 20:32:10 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/keyboard/amikbd.c b/drivers/input/keyboard/amikbd.c
index 81bf7562aca0..35149ec455a9 100644
--- a/drivers/input/keyboard/amikbd.c
+++ b/drivers/input/keyboard/amikbd.c
@@ -1,6 +1,4 @@
 /*
- * $Id: amikbd.c,v 1.13 2002/02/01 16:02:24 vojtech Exp $
- *
  *  Copyright (c) 2000-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/keyboard/sunkbd.c b/drivers/input/keyboard/sunkbd.c
index be0f5d19d023..9fce6d1e29b2 100644
--- a/drivers/input/keyboard/sunkbd.c
+++ b/drivers/input/keyboard/sunkbd.c
@@ -1,6 +1,4 @@
 /*
- * $Id: sunkbd.c,v 1.14 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/keyboard/xtkbd.c b/drivers/input/keyboard/xtkbd.c
index 152a2c070508..37b01d777a4a 100644
--- a/drivers/input/keyboard/xtkbd.c
+++ b/drivers/input/keyboard/xtkbd.c
@@ -1,6 +1,4 @@
 /*
- * $Id: xtkbd.c,v 1.11 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/mouse/inport.c b/drivers/input/mouse/inport.c
index 06c35fc553c0..3827a22362de 100644
--- a/drivers/input/mouse/inport.c
+++ b/drivers/input/mouse/inport.c
@@ -1,6 +1,4 @@
 /*
- * $Id: inport.c,v 1.11 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/mouse/logibm.c b/drivers/input/mouse/logibm.c
index 9ea895593b27..e2413113df22 100644
--- a/drivers/input/mouse/logibm.c
+++ b/drivers/input/mouse/logibm.c
@@ -1,6 +1,4 @@
 /*
- * $Id: logibm.c,v 1.11 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/mouse/pc110pad.c b/drivers/input/mouse/pc110pad.c
index 61cff8374e6c..fd09c8df81f2 100644
--- a/drivers/input/mouse/pc110pad.c
+++ b/drivers/input/mouse/pc110pad.c
@@ -1,6 +1,4 @@
 /*
- * $Id: pc110pad.c,v 1.12 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 2000-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/mouse/sermouse.c b/drivers/input/mouse/sermouse.c
index ed917bfd086a..17ff137b9bd5 100644
--- a/drivers/input/mouse/sermouse.c
+++ b/drivers/input/mouse/sermouse.c
@@ -1,6 +1,4 @@
 /*
- * $Id: sermouse.c,v 1.17 2002/03/13 10:03:43 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/serio/ct82c710.c b/drivers/input/serio/ct82c710.c
index 0d35018c23a9..d1380fc72cc6 100644
--- a/drivers/input/serio/ct82c710.c
+++ b/drivers/input/serio/ct82c710.c
@@ -1,6 +1,4 @@
 /*
- * $Id: ct82c710.c,v 1.11 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 1999-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/serio/q40kbd.c b/drivers/input/serio/q40kbd.c
index cb89aff2e160..eca125c8d535 100644
--- a/drivers/input/serio/q40kbd.c
+++ b/drivers/input/serio/q40kbd.c
@@ -1,6 +1,4 @@
 /*
- * $Id: q40kbd.c,v 1.12 2002/02/02 22:26:44 vojtech Exp $
- *
  *  Copyright (c) 2000-2001 Vojtech Pavlik
  *
  *  Based on the work of:
diff --git a/drivers/input/serio/rpckbd.c b/drivers/input/serio/rpckbd.c
index 34c59d9c6205..1567b7782478 100644
--- a/drivers/input/serio/rpckbd.c
+++ b/drivers/input/serio/rpckbd.c
@@ -1,6 +1,4 @@
 /*
- * $Id: rpckbd.c,v 1.7 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 2000-2001 Vojtech Pavlik
  *  Copyright (c) 2002 Russell King
  */
diff --git a/drivers/input/touchscreen/gunze.c b/drivers/input/touchscreen/gunze.c
index a48a15868c4a..a54f90e02ab6 100644
--- a/drivers/input/touchscreen/gunze.c
+++ b/drivers/input/touchscreen/gunze.c
@@ -1,6 +1,4 @@
 /*
- * $Id: gunze.c,v 1.12 2001/09/25 10:12:07 vojtech Exp $
- *
  *  Copyright (c) 2000-2001 Vojtech Pavlik
  */
 
diff --git a/drivers/input/touchscreen/h3600_ts_input.c b/drivers/input/touchscreen/h3600_ts_input.c
index 28ae15ed12c5..4f86081dc7fc 100644
--- a/drivers/input/touchscreen/h3600_ts_input.c
+++ b/drivers/input/touchscreen/h3600_ts_input.c
@@ -1,6 +1,4 @@
 /*
- * $Id: h3600_ts_input.c,v 1.4 2002/01/23 06:39:37 jsimmons Exp $
- *
  *  Copyright (c) 2001 "Crazy" James Simmons jsimmons@transvirtual.com
  *
  *  Sponsored by Transvirtual Technology.
diff --git a/include/linux/joystick.h b/include/linux/joystick.h
index e2d3a18af456..b5e051295a67 100644
--- a/include/linux/joystick.h
+++ b/include/linux/joystick.h
@@ -2,8 +2,6 @@
 #define _LINUX_JOYSTICK_H
 
 /*
- * $Id: joystick.h,v 1.3 2000/11/30 11:07:05 vojtech Exp $
- *
  *  Copyright (C) 1996-2000 Vojtech Pavlik
  *
  *  Sponsored by SuSE
-- 
cgit v1.2.3


From 59f0c4523fdea865fab7d69d878269992a9d08dd Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@windriver.com>
Date: Wed, 21 May 2008 14:52:30 -0700
Subject: tipc: Fix skb_under_panic when configuring TIPC without privileges

This patch prevents a TIPC configuration command requiring network
administrator privileges from triggering an skbuff underrun if it
is issued by a process lacking those privileges.  The revised error
handling code avoids the use of a potentially uninitialized global
variable by transforming the unauthorized command into a new command,
then following the standard command processing path to generate the
required error message.

Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tipc_config.h | 10 +++++++++-
 net/tipc/config.c           |  6 +++++-
 net/tipc/netlink.c          | 16 +++++++++-------
 3 files changed, 23 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tipc_config.h b/include/linux/tipc_config.h
index b0c916d1f375..2bc6fa4adeb5 100644
--- a/include/linux/tipc_config.h
+++ b/include/linux/tipc_config.h
@@ -2,7 +2,7 @@
  * include/linux/tipc_config.h: Include file for TIPC configuration interface
  * 
  * Copyright (c) 2003-2006, Ericsson AB
- * Copyright (c) 2005, Wind River Systems
+ * Copyright (c) 2005-2007, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -135,6 +135,14 @@
 #define  TIPC_CMD_SET_MAX_SLAVES    0x800A    /* tx unsigned, rx none */
 #define  TIPC_CMD_SET_NETID         0x800B    /* tx unsigned, rx none */
 
+/*
+ * Reserved commands:
+ * May not be issued by any process.
+ * Used internally by TIPC.
+ */
+
+#define  TIPC_CMD_NOT_NET_ADMIN     0xC001    /* tx none, rx none */
+
 /*
  * TLV types defined for TIPC
  */
diff --git a/net/tipc/config.c b/net/tipc/config.c
index 91d56f8fee9f..16e7cb74969b 100644
--- a/net/tipc/config.c
+++ b/net/tipc/config.c
@@ -2,7 +2,7 @@
  * net/tipc/config.c: TIPC configuration management code
  *
  * Copyright (c) 2002-2006, Ericsson AB
- * Copyright (c) 2004-2006, Wind River Systems
+ * Copyright (c) 2004-2007, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -602,6 +602,10 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area
 	case TIPC_CMD_GET_NETID:
 		rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id);
 		break;
+	case TIPC_CMD_NOT_NET_ADMIN:
+		rep_tlv_buf =
+			tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN);
+		break;
 	default:
 		rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED
 							  " (unknown command)");
diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c
index 6a7f7b4c2595..c387217bb230 100644
--- a/net/tipc/netlink.c
+++ b/net/tipc/netlink.c
@@ -2,7 +2,7 @@
  * net/tipc/netlink.c: TIPC configuration handling
  *
  * Copyright (c) 2005-2006, Ericsson AB
- * Copyright (c) 2005, Wind River Systems
+ * Copyright (c) 2005-2007, Wind River Systems
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -45,15 +45,17 @@ static int handle_cmd(struct sk_buff *skb, struct genl_info *info)
 	struct nlmsghdr *req_nlh = info->nlhdr;
 	struct tipc_genlmsghdr *req_userhdr = info->userhdr;
 	int hdr_space = NLMSG_SPACE(GENL_HDRLEN + TIPC_GENL_HDRLEN);
+	u16 cmd;
 
 	if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN)))
-		rep_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN);
+		cmd = TIPC_CMD_NOT_NET_ADMIN;
 	else
-		rep_buf = tipc_cfg_do_cmd(req_userhdr->dest,
-					  req_userhdr->cmd,
-					  NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN,
-					  NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN),
-					  hdr_space);
+		cmd = req_userhdr->cmd;
+
+	rep_buf = tipc_cfg_do_cmd(req_userhdr->dest, cmd,
+			NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN,
+			NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN),
+			hdr_space);
 
 	if (rep_buf) {
 		skb_push(rep_buf, hdr_space);
-- 
cgit v1.2.3


From edcdf8b21ac920e06b4180246123fe43b022e020 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Thu, 15 May 2008 13:53:55 +0800
Subject: mac80211: separate Tx and Rx MCS when configuring HT

This patch follows the 11n spec in separation between Tx and Rx MCS
capabilities. Up until now, when configuring the HT possible set of Tx
MCS only Rx MCS were considered, assuming they are the same as the Tx MCS.
This patch fixed this by looking at low level driver Tx capabilities.

Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl4965-base.c | 14 +++--
 include/linux/ieee80211.h                   | 18 ++++--
 net/mac80211/main.c                         | 96 ++++++++++++++++++-----------
 3 files changed, 83 insertions(+), 45 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 1a2d3768867d..c713b27ec653 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -3958,6 +3958,13 @@ static void iwl4965_dealloc_ucode_pci(struct iwl_priv *priv)
 	iwl_free_fw_desc(priv->pci_dev, &priv->ucode_boot);
 }
 
+static void iwl4965_nic_start(struct iwl_priv *priv)
+{
+	/* Remove all resets to allow NIC to operate */
+	iwl_write32(priv, CSR_RESET, 0);
+}
+
+
 /**
  * iwl4965_read_ucode - Read uCode images from disk file.
  *
@@ -4447,8 +4454,7 @@ static int __iwl4965_up(struct iwl_priv *priv)
 		}
 
 		/* start card; "initialize" will load runtime ucode */
-		/* Remove all resets to allow NIC to operate */
-		iwl_write32(priv, CSR_RESET, 0);
+		iwl4965_nic_start(priv);
 
 		IWL_DEBUG_INFO(DRV_NAME " is coming up\n");
 
@@ -6842,10 +6848,6 @@ static int __init iwl4965_init(void)
 
 	return ret;
 
-
-#ifdef CONFIG_IWLWIFI_DEBUG
-	pci_unregister_driver(&iwl_driver);
-#endif
 error_register:
 	iwl4965_rate_control_unregister();
 	return ret;
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index a9102bc78b61..3c2ac0c37aa8 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -306,8 +306,18 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_CAP_SGI_40			0x0040
 #define IEEE80211_HT_CAP_DELAY_BA		0x0400
 #define IEEE80211_HT_CAP_MAX_AMSDU		0x0800
+/* 802.11n HT capability AMPDU settings */
 #define IEEE80211_HT_CAP_AMPDU_FACTOR		0x03
 #define IEEE80211_HT_CAP_AMPDU_DENSITY		0x1C
+/* 802.11n HT capability MSC set */
+#define IEEE80211_SUPP_MCS_SET_UEQM		4
+#define IEEE80211_HT_CAP_MAX_STREAMS		4
+#define IEEE80211_SUPP_MCS_SET_LEN		10
+/* maximum streams the spec allows */
+#define IEEE80211_HT_CAP_MCS_TX_DEFINED		0x01
+#define IEEE80211_HT_CAP_MCS_TX_RX_DIFF		0x02
+#define IEEE80211_HT_CAP_MCS_TX_STREAMS		0x0C
+#define IEEE80211_HT_CAP_MCS_TX_UEQM		0x10
 /* 802.11n HT IE masks */
 #define IEEE80211_HT_IE_CHA_SEC_OFFSET		0x03
 #define IEEE80211_HT_IE_CHA_WIDTH		0x04
@@ -316,10 +326,10 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_IE_NON_HT_STA_PRSNT	0x0010
 
 /* MIMO Power Save Modes */
-#define WLAN_HT_CAP_MIMO_PS_STATIC         0
-#define WLAN_HT_CAP_MIMO_PS_DYNAMIC        1
-#define WLAN_HT_CAP_MIMO_PS_INVALID        2
-#define WLAN_HT_CAP_MIMO_PS_DISABLED       3
+#define WLAN_HT_CAP_MIMO_PS_STATIC	0
+#define WLAN_HT_CAP_MIMO_PS_DYNAMIC	1
+#define WLAN_HT_CAP_MIMO_PS_INVALID	2
+#define WLAN_HT_CAP_MIMO_PS_DISABLED	3
 
 /* Authentication algorithms */
 #define WLAN_AUTH_OPEN 0
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 36016363d225..b0fddb7de549 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -35,8 +35,6 @@
 #include "debugfs.h"
 #include "debugfs_netdev.h"
 
-#define SUPP_MCS_SET_LEN 16
-
 /*
  * For seeing transmitted packets on monitor interfaces
  * we have a radiotap header too.
@@ -1068,56 +1066,84 @@ u32 ieee80211_handle_ht(struct ieee80211_local *local, int enable_ht,
 	struct ieee80211_supported_band *sband;
 	struct ieee80211_ht_info ht_conf;
 	struct ieee80211_ht_bss_info ht_bss_conf;
-	int i;
 	u32 changed = 0;
+	int i;
+	u8 max_tx_streams = IEEE80211_HT_CAP_MAX_STREAMS;
+	u8 tx_mcs_set_cap;
 
 	sband = local->hw.wiphy->bands[conf->channel->band];
 
+	memset(&ht_conf, 0, sizeof(struct ieee80211_ht_info));
+	memset(&ht_bss_conf, 0, sizeof(struct ieee80211_ht_bss_info));
+
 	/* HT is not supported */
 	if (!sband->ht_info.ht_supported) {
 		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
-		return 0;
+		goto out;
 	}
 
-	memset(&ht_conf, 0, sizeof(struct ieee80211_ht_info));
-	memset(&ht_bss_conf, 0, sizeof(struct ieee80211_ht_bss_info));
-
-	if (enable_ht) {
-		if (!(conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE))
+	/* disable HT */
+	if (!enable_ht) {
+		if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE)
 			changed |= BSS_CHANGED_HT;
+		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
+		conf->ht_conf.ht_supported = 0;
+		goto out;
+	}
 
-		conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE;
-		ht_conf.ht_supported = 1;
 
-		ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap;
-		ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
-		ht_conf.cap |= sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
+	if (!(conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE))
+		changed |= BSS_CHANGED_HT;
 
-		for (i = 0; i < SUPP_MCS_SET_LEN; i++)
-			ht_conf.supp_mcs_set[i] =
-					sband->ht_info.supp_mcs_set[i] &
-					req_ht_cap->supp_mcs_set[i];
+	conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE;
+	ht_conf.ht_supported = 1;
 
-		ht_bss_conf.primary_channel = req_bss_cap->primary_channel;
-		ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
-		ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
+	ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap;
+	ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS);
+	ht_conf.cap |= sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS;
+	ht_bss_conf.primary_channel = req_bss_cap->primary_channel;
+	ht_bss_conf.bss_cap = req_bss_cap->bss_cap;
+	ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode;
 
-		ht_conf.ampdu_factor = req_ht_cap->ampdu_factor;
-		ht_conf.ampdu_density = req_ht_cap->ampdu_density;
+	ht_conf.ampdu_factor = req_ht_cap->ampdu_factor;
+	ht_conf.ampdu_density = req_ht_cap->ampdu_density;
 
-		/* if bss configuration changed store the new one */
-		if (memcmp(&conf->ht_conf, &ht_conf, sizeof(ht_conf)) ||
-		    memcmp(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf))) {
-			changed |= BSS_CHANGED_HT;
-			memcpy(&conf->ht_conf, &ht_conf, sizeof(ht_conf));
-			memcpy(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf));
-		}
-	} else {
-		if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE)
-			changed |= BSS_CHANGED_HT;
-		conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE;
-	}
+	/* Bits 96-100 */
+	tx_mcs_set_cap = sband->ht_info.supp_mcs_set[12];
 
+	/* configure suppoerted Tx MCS according to requested MCS
+	 * (based in most cases on Rx capabilities of peer) and self
+	 * Tx MCS capabilities (as defined by low level driver HW
+	 * Tx capabilities) */
+	if (!(tx_mcs_set_cap & IEEE80211_HT_CAP_MCS_TX_DEFINED))
+		goto check_changed;
+
+	/* Counting from 0 therfore + 1 */
+	if (tx_mcs_set_cap & IEEE80211_HT_CAP_MCS_TX_RX_DIFF)
+		max_tx_streams = ((tx_mcs_set_cap &
+				IEEE80211_HT_CAP_MCS_TX_STREAMS) >> 2) + 1;
+
+	for (i = 0; i < max_tx_streams; i++)
+		ht_conf.supp_mcs_set[i] =
+			sband->ht_info.supp_mcs_set[i] &
+					req_ht_cap->supp_mcs_set[i];
+
+	if (tx_mcs_set_cap & IEEE80211_HT_CAP_MCS_TX_UEQM)
+		for (i = IEEE80211_SUPP_MCS_SET_UEQM;
+		     i < IEEE80211_SUPP_MCS_SET_LEN; i++)
+			ht_conf.supp_mcs_set[i] =
+				sband->ht_info.supp_mcs_set[i] &
+					req_ht_cap->supp_mcs_set[i];
+
+check_changed:
+	/* if bss configuration changed store the new one */
+	if (memcmp(&conf->ht_conf, &ht_conf, sizeof(ht_conf)) ||
+	    memcmp(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf))) {
+		changed |= BSS_CHANGED_HT;
+		memcpy(&conf->ht_conf, &ht_conf, sizeof(ht_conf));
+		memcpy(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf));
+	}
+out:
 	return changed;
 }
 
-- 
cgit v1.2.3


From 41df0d61c266998b8049df7fec119cd518a43aa1 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Mon, 12 May 2008 21:21:13 +0200
Subject: x86: Add performance variants of cpumask operators

  * Increase performance for systems with large count NR_CPUS by limiting
    the range of the cpumask operators that loop over the bits in a cpumask_t
    variable.  This removes a large amount of wasted cpu cycles.

  * Add performance variants of the cpumask operators:

    int cpus_weight_nr(mask)	     Same using nr_cpu_ids instead of NR_CPUS
    int first_cpu_nr(mask)	     Number lowest set bit, or nr_cpu_ids
    int next_cpu_nr(cpu, mask)	     Next cpu past 'cpu', or nr_cpu_ids
    for_each_cpu_mask_nr(cpu, mask)  for-loop cpu over mask using nr_cpu_ids

  * Modify following to use performance variants:

    #define num_online_cpus()	cpus_weight_nr(cpu_online_map)
    #define num_possible_cpus()	cpus_weight_nr(cpu_possible_map)
    #define num_present_cpus()	cpus_weight_nr(cpu_present_map)

    #define for_each_possible_cpu(cpu) for_each_cpu_mask_nr((cpu), ...)
    #define for_each_online_cpu(cpu)   for_each_cpu_mask_nr((cpu), ...)
    #define for_each_present_cpu(cpu)  for_each_cpu_mask_nr((cpu), ...)

  * Comment added to include/linux/cpumask.h:

    Note: The alternate operations with the suffix "_nr" are used
	  to limit the range of the loop to nr_cpu_ids instead of
	  NR_CPUS when NR_CPUS > 64 for performance reasons.
	  If NR_CPUS is <= 64 then most assembler bitmask
	  operators execute faster with a constant range, so
	  the operator will continue to use NR_CPUS.

	  Another consideration is that nr_cpu_ids is initialized
	  to NR_CPUS and isn't lowered until the possible cpus are
	  discovered (including any disabled cpus).  So early uses
	  will span the entire range of NR_CPUS.

    (The net effect is that for systems with 64 or less CPU's there are no
     functional changes.)

For inclusion into sched-devel/latest tree.

Based on:
	git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
    +   sched-devel/latest  .../mingo/linux-2.6-sched-devel.git

Cc: Paul Jackson <pj@sgi.com>
Cc: Christoph Lameter <clameter@sgi.com>
Reviewed-by: Paul Jackson <pj@sgi.com>
Reviewed-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/cpumask.h | 92 +++++++++++++++++++++++++++++++++----------------
 lib/cpumask.c           |  9 +++++
 2 files changed, 71 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 5df3db58fcc6..b49472d1af84 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -17,6 +17,20 @@
  * For details of cpus_onto(), see bitmap_onto in lib/bitmap.c.
  * For details of cpus_fold(), see bitmap_fold in lib/bitmap.c.
  *
+ * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
+ * Note: The alternate operations with the suffix "_nr" are used
+ *       to limit the range of the loop to nr_cpu_ids instead of
+ *       NR_CPUS when NR_CPUS > 64 for performance reasons.
+ *       If NR_CPUS is <= 64 then most assembler bitmask
+ *       operators execute faster with a constant range, so
+ *       the operator will continue to use NR_CPUS.
+ *
+ *       Another consideration is that nr_cpu_ids is initialized
+ *       to NR_CPUS and isn't lowered until the possible cpus are
+ *       discovered (including any disabled cpus).  So early uses
+ *       will span the entire range of NR_CPUS.
+ * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
+ *
  * The available cpumask operations are:
  *
  * void cpu_set(cpu, mask)		turn on bit 'cpu' in mask
@@ -38,12 +52,14 @@
  * int cpus_empty(mask)			Is mask empty (no bits sets)?
  * int cpus_full(mask)			Is mask full (all bits sets)?
  * int cpus_weight(mask)		Hamming weigh - number of set bits
+ * int cpus_weight_nr(mask)		Same using nr_cpu_ids instead of NR_CPUS
  *
  * void cpus_shift_right(dst, src, n)	Shift right
  * void cpus_shift_left(dst, src, n)	Shift left
  *
  * int first_cpu(mask)			Number lowest set bit, or NR_CPUS
  * int next_cpu(cpu, mask)		Next cpu past 'cpu', or NR_CPUS
+ * int next_cpu_nr(cpu, mask)		Next cpu past 'cpu', or nr_cpu_ids
  *
  * cpumask_t cpumask_of_cpu(cpu)	Return cpumask with bit 'cpu' set
  * CPU_MASK_ALL				Initializer - all bits set
@@ -59,7 +75,8 @@
  * void cpus_onto(dst, orig, relmap)	*dst = orig relative to relmap
  * void cpus_fold(dst, orig, sz)	dst bits = orig bits mod sz
  *
- * for_each_cpu_mask(cpu, mask)		for-loop cpu over mask
+ * for_each_cpu_mask(cpu, mask)		for-loop cpu over mask using NR_CPUS
+ * for_each_cpu_mask_nr(cpu, mask)	for-loop cpu over mask using nr_cpu_ids
  *
  * int num_online_cpus()		Number of online CPUs
  * int num_possible_cpus()		Number of all possible CPUs
@@ -216,15 +233,6 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 	bitmap_shift_left(dstp->bits, srcp->bits, n, nbits);
 }
 
-#ifdef CONFIG_SMP
-int __first_cpu(const cpumask_t *srcp);
-#define first_cpu(src) __first_cpu(&(src))
-int __next_cpu(int n, const cpumask_t *srcp);
-#define next_cpu(n, src) __next_cpu((n), &(src))
-#else
-#define first_cpu(src)		({ (void)(src); 0; })
-#define next_cpu(n, src)	({ (void)(src); 1; })
-#endif
 
 #ifdef CONFIG_HAVE_CPUMASK_OF_CPU_MAP
 extern cpumask_t *cpumask_of_cpu_map;
@@ -343,15 +351,48 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
 	bitmap_fold(dstp->bits, origp->bits, sz, nbits);
 }
 
-#if NR_CPUS > 1
+#if NR_CPUS == 1
+
+#define nr_cpu_ids		1
+#define first_cpu(src)		({ (void)(src); 0; })
+#define next_cpu(n, src)	({ (void)(src); 1; })
+#define any_online_cpu(mask)	0
+#define for_each_cpu_mask(cpu, mask)	\
+	for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask)
+
+#else /* NR_CPUS > 1 */
+
+extern int nr_cpu_ids;
+int __first_cpu(const cpumask_t *srcp);
+int __next_cpu(int n, const cpumask_t *srcp);
+int __any_online_cpu(const cpumask_t *mask);
+
+#define first_cpu(src)		__first_cpu(&(src))
+#define next_cpu(n, src)	__next_cpu((n), &(src))
+#define any_online_cpu(mask) __any_online_cpu(&(mask))
 #define for_each_cpu_mask(cpu, mask)		\
 	for ((cpu) = first_cpu(mask);		\
 		(cpu) < NR_CPUS;		\
 		(cpu) = next_cpu((cpu), (mask)))
-#else /* NR_CPUS == 1 */
-#define for_each_cpu_mask(cpu, mask)		\
-	for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask)
-#endif /* NR_CPUS */
+#endif
+
+#if NR_CPUS <= 64
+
+#define next_cpu_nr(n, src)		next_cpu(n, src)
+#define cpus_weight_nr(cpumask)		cpus_weight(cpumask)
+#define for_each_cpu_mask_nr(cpu, mask)	for_each_cpu_mask(cpu, mask)
+
+#else /* NR_CPUS > 64 */
+
+int __next_cpu_nr(int n, const cpumask_t *srcp);
+#define next_cpu_nr(n, src)	__next_cpu_nr((n), &(src))
+#define cpus_weight_nr(cpumask)	__cpus_weight(&(cpumask), nr_cpu_ids)
+#define for_each_cpu_mask_nr(cpu, mask)		\
+	for ((cpu) = first_cpu(mask);		\
+		(cpu) < nr_cpu_ids;		\
+		(cpu) = next_cpu_nr((cpu), (mask)))
+
+#endif /* NR_CPUS > 64 */
 
 /*
  * The following particular system cpumasks and operations manage
@@ -414,9 +455,9 @@ extern cpumask_t cpu_online_map;
 extern cpumask_t cpu_present_map;
 
 #if NR_CPUS > 1
-#define num_online_cpus()	cpus_weight(cpu_online_map)
-#define num_possible_cpus()	cpus_weight(cpu_possible_map)
-#define num_present_cpus()	cpus_weight(cpu_present_map)
+#define num_online_cpus()	cpus_weight_nr(cpu_online_map)
+#define num_possible_cpus()	cpus_weight_nr(cpu_possible_map)
+#define num_present_cpus()	cpus_weight_nr(cpu_present_map)
 #define cpu_online(cpu)		cpu_isset((cpu), cpu_online_map)
 #define cpu_possible(cpu)	cpu_isset((cpu), cpu_possible_map)
 #define cpu_present(cpu)	cpu_isset((cpu), cpu_present_map)
@@ -431,17 +472,8 @@ extern cpumask_t cpu_present_map;
 
 #define cpu_is_offline(cpu)	unlikely(!cpu_online(cpu))
 
-#ifdef CONFIG_SMP
-extern int nr_cpu_ids;
-#define any_online_cpu(mask) __any_online_cpu(&(mask))
-int __any_online_cpu(const cpumask_t *mask);
-#else
-#define nr_cpu_ids			1
-#define any_online_cpu(mask)		0
-#endif
-
-#define for_each_possible_cpu(cpu)  for_each_cpu_mask((cpu), cpu_possible_map)
-#define for_each_online_cpu(cpu)  for_each_cpu_mask((cpu), cpu_online_map)
-#define for_each_present_cpu(cpu) for_each_cpu_mask((cpu), cpu_present_map)
+#define for_each_possible_cpu(cpu) for_each_cpu_mask_nr((cpu), cpu_possible_map)
+#define for_each_online_cpu(cpu)   for_each_cpu_mask_nr((cpu), cpu_online_map)
+#define for_each_present_cpu(cpu)  for_each_cpu_mask_nr((cpu), cpu_present_map)
 
 #endif /* __LINUX_CPUMASK_H */
diff --git a/lib/cpumask.c b/lib/cpumask.c
index bb4f76d3c3e7..5f97dc25ef9c 100644
--- a/lib/cpumask.c
+++ b/lib/cpumask.c
@@ -15,6 +15,15 @@ int __next_cpu(int n, const cpumask_t *srcp)
 }
 EXPORT_SYMBOL(__next_cpu);
 
+#if NR_CPUS > 64
+int __next_cpu_nr(int n, const cpumask_t *srcp)
+{
+	return min_t(int, nr_cpu_ids,
+				find_next_bit(srcp->bits, nr_cpu_ids, n+1));
+}
+EXPORT_SYMBOL(__next_cpu_nr);
+#endif
+
 int __any_online_cpu(const cpumask_t *mask)
 {
 	int cpu;
-- 
cgit v1.2.3


From 7baac8b91f9871ba8cb09af84de4ae1d86d07812 Mon Sep 17 00:00:00 2001
From: Alexander van Heukelum <heukelum@mailshack.com>
Date: Tue, 13 May 2008 11:28:21 +0200
Subject: cpumask: make for_each_cpu_mask a bit smaller

The for_each_cpu_mask loop is used quite often in the kernel. It
makes use of two functions: first_cpu and next_cpu. This patch
changes for_each_cpu_mask to use only the latter. Because next_cpu
finds the next eligible cpu _after_ the given one, the iteration
variable has to be initialized to -1 and next_cpu has to be
called with this value before the first iteration. An x86_64
defconfig kernel (from sched/latest) is about 2500 bytes smaller
with this patch applied:

   text	   data	    bss	    dec	    hex	filename
6222517	 917952	 749932	7890401	 7865e1	vmlinux.orig
6219922	 917952	 749932	7887806	 785bbe	vmlinux

The same size reduction is seen for defconfig+MAXSMP

   text	   data	    bss	    dec	    hex	filename
6241772	2563968	1492716	10298456	 9d2458	vmlinux.orig
6239211	2563968	1492716	10295895	 9d1a57	vmlinux

Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/cpumask.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index b49472d1af84..80226e776143 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -370,10 +370,10 @@ int __any_online_cpu(const cpumask_t *mask);
 #define first_cpu(src)		__first_cpu(&(src))
 #define next_cpu(n, src)	__next_cpu((n), &(src))
 #define any_online_cpu(mask) __any_online_cpu(&(mask))
-#define for_each_cpu_mask(cpu, mask)		\
-	for ((cpu) = first_cpu(mask);		\
-		(cpu) < NR_CPUS;		\
-		(cpu) = next_cpu((cpu), (mask)))
+#define for_each_cpu_mask(cpu, mask)			\
+	for ((cpu) = -1;				\
+		(cpu) = next_cpu((cpu), (mask)),	\
+		(cpu) < NR_CPUS; )
 #endif
 
 #if NR_CPUS <= 64
@@ -387,10 +387,10 @@ int __any_online_cpu(const cpumask_t *mask);
 int __next_cpu_nr(int n, const cpumask_t *srcp);
 #define next_cpu_nr(n, src)	__next_cpu_nr((n), &(src))
 #define cpus_weight_nr(cpumask)	__cpus_weight(&(cpumask), nr_cpu_ids)
-#define for_each_cpu_mask_nr(cpu, mask)		\
-	for ((cpu) = first_cpu(mask);		\
-		(cpu) < nr_cpu_ids;		\
-		(cpu) = next_cpu_nr((cpu), (mask)))
+#define for_each_cpu_mask_nr(cpu, mask)			\
+	for ((cpu) = -1;				\
+		(cpu) = next_cpu_nr((cpu), (mask)),	\
+		(cpu) < nr_cpu_ids; )
 
 #endif /* NR_CPUS > 64 */
 
-- 
cgit v1.2.3


From bd3bff9e20f454b242d979ec2f9a4dca0d5fa06f Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:41 +0200
Subject: sched: add latency tracer callbacks to the scheduler

add 3 lightweight callbacks to the tracer backend.

zero impact if tracing is turned off.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h | 26 ++++++++++++++++++++++++++
 kernel/sched.c        |  3 +++
 2 files changed, 29 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5395a6176f4b..717cab8a0c83 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2117,6 +2117,32 @@ static inline void arch_pick_mmap_layout(struct mm_struct *mm)
 }
 #endif
 
+#ifdef CONFIG_CONTEXT_SWITCH_TRACER
+extern void
+ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next);
+#else
+static inline void
+ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next)
+{
+}
+#endif
+
+#ifdef CONFIG_SCHED_TRACER
+extern void
+ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr);
+extern void
+ftrace_wake_up_new_task(struct task_struct *wakee, struct task_struct *curr);
+#else
+static inline void
+ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr)
+{
+}
+static inline void
+ftrace_wake_up_new_task(struct task_struct *wakee, struct task_struct *curr)
+{
+}
+#endif
+
 extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask);
 extern long sched_getaffinity(pid_t pid, cpumask_t *mask);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index cfa222a91539..463dcdb36ef8 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2467,6 +2467,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 
 out_activate:
 #endif /* CONFIG_SMP */
+	ftrace_wake_up_task(p, rq->curr);
 	schedstat_inc(p, se.nr_wakeups);
 	if (sync)
 		schedstat_inc(p, se.nr_wakeups_sync);
@@ -2611,6 +2612,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 		p->sched_class->task_new(rq, p);
 		inc_nr_running(rq);
 	}
+	ftrace_wake_up_new_task(p, rq->curr);
 	check_preempt_curr(rq, p);
 #ifdef CONFIG_SMP
 	if (p->sched_class->task_wake_up)
@@ -2783,6 +2785,7 @@ context_switch(struct rq *rq, struct task_struct *prev,
 	struct mm_struct *mm, *oldmm;
 
 	prepare_task_switch(rq, prev, next);
+	ftrace_ctx_switch(prev, next);
 	mm = next->mm;
 	oldmm = prev->active_mm;
 	/*
-- 
cgit v1.2.3


From 7c731e0a495e25e79dc1e9e68772a67a55721a65 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:41 +0200
Subject: ftrace: make the task state char-string visible to all

The tracer wants to be able to convert the state number
into a user visible character. This patch pulls that conversion
string out the scheduler into the header. This way if it were to
ever change, other parts of the kernel will know.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h | 2 ++
 kernel/sched.c        | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 717cab8a0c83..6e26f1fdbfe2 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2237,6 +2237,8 @@ static inline void mm_init_owner(struct mm_struct *mm, struct task_struct *p)
 }
 #endif /* CONFIG_MM_OWNER */
 
+#define TASK_STATE_TO_CHAR_STR "RSDTtZX"
+
 #endif /* __KERNEL__ */
 
 #endif
diff --git a/kernel/sched.c b/kernel/sched.c
index 463dcdb36ef8..73e600852365 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5729,7 +5729,7 @@ out_unlock:
 	return retval;
 }
 
-static const char stat_nam[] = "RSDTtZX";
+static const char stat_nam[] = TASK_STATE_TO_CHAR_STR;
 
 void sched_show_task(struct task_struct *p)
 {
-- 
cgit v1.2.3


From 502825282e6f79c975a644afc124432ec1744de4 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:41 +0200
Subject: ftrace: add preempt_enable/disable notrace macros

The tracer may need to call preempt_enable and disable functions
for time keeping and such. The trace gets ugly when we see these
functions show up for all traces. To make the output cleaner
this patch adds preempt_enable_notrace and preempt_disable_notrace
to be used by tracer (and debugging) functions.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/preempt.h | 32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/preempt.h b/include/linux/preempt.h
index 23f0c54175cd..36b03d50bf40 100644
--- a/include/linux/preempt.h
+++ b/include/linux/preempt.h
@@ -52,6 +52,34 @@ do { \
 	preempt_check_resched(); \
 } while (0)
 
+/* For debugging and tracer internals only! */
+#define add_preempt_count_notrace(val)			\
+	do { preempt_count() += (val); } while (0)
+#define sub_preempt_count_notrace(val)			\
+	do { preempt_count() -= (val); } while (0)
+#define inc_preempt_count_notrace() add_preempt_count_notrace(1)
+#define dec_preempt_count_notrace() sub_preempt_count_notrace(1)
+
+#define preempt_disable_notrace() \
+do { \
+	inc_preempt_count_notrace(); \
+	barrier(); \
+} while (0)
+
+#define preempt_enable_no_resched_notrace() \
+do { \
+	barrier(); \
+	dec_preempt_count_notrace(); \
+} while (0)
+
+/* preempt_check_resched is OK to trace */
+#define preempt_enable_notrace() \
+do { \
+	preempt_enable_no_resched_notrace(); \
+	barrier(); \
+	preempt_check_resched(); \
+} while (0)
+
 #else
 
 #define preempt_disable()		do { } while (0)
@@ -59,6 +87,10 @@ do { \
 #define preempt_enable()		do { } while (0)
 #define preempt_check_resched()		do { } while (0)
 
+#define preempt_disable_notrace()		do { } while (0)
+#define preempt_enable_no_resched_notrace()	do { } while (0)
+#define preempt_enable_notrace()		do { } while (0)
+
 #endif
 
 #ifdef CONFIG_PREEMPT_NOTIFIERS
-- 
cgit v1.2.3


From ffdc1a09ae7e2cbd714a446ee38a27f625b5f1c8 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:41 +0200
Subject: tracing: add notrace to linkage.h

notrace signals that a function should not be traced. Most of the
time this is used by tracers to annotate code that cannot be
traced - it's in a volatile state (such as in user vdso context
or NMI context) or it's in the tracer internals.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/linkage.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index 2119610b24f8..14f329c64ba8 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -3,6 +3,8 @@
 
 #include <asm/linkage.h>
 
+#define notrace __attribute__((no_instrument_function))
+
 #ifdef __cplusplus
 #define CPP_ASMLINKAGE extern "C"
 #else
-- 
cgit v1.2.3


From 16444a8a40d4c7b4f6de34af0cae1f76a4f6c901 Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Date: Mon, 12 May 2008 21:20:42 +0200
Subject: ftrace: add basic support for gcc profiler instrumentation

If CONFIG_FTRACE is selected and /proc/sys/kernel/ftrace_enabled is
set to a non-zero value the ftrace routine will be called everytime
we enter a kernel function that is not marked with the "notrace"
attribute.

The ftrace routine will then call a registered function if a function
happens to be registered.

[ This code has been highly hacked by Steven Rostedt and Ingo Molnar,
  so don't blame Arnaldo for all of this ;-) ]

Update:
  It is now possible to register more than one ftrace function.
  If only one ftrace function is registered, that will be the
  function that ftrace calls directly. If more than one function
  is registered, then ftrace will call a function that will loop
  through the functions to call.

Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Makefile                   |   4 ++
 arch/x86/Kconfig           |   1 +
 arch/x86/kernel/entry_32.S |  27 +++++++++
 arch/x86/kernel/entry_64.S |  37 ++++++++++++
 include/linux/ftrace.h     |  38 +++++++++++++
 kernel/Makefile            |   1 +
 kernel/trace/Kconfig       |   5 ++
 kernel/trace/Makefile      |   3 +
 kernel/trace/ftrace.c      | 138 +++++++++++++++++++++++++++++++++++++++++++++
 lib/Kconfig.debug          |   2 +
 10 files changed, 256 insertions(+)
 create mode 100644 include/linux/ftrace.h
 create mode 100644 kernel/trace/Kconfig
 create mode 100644 kernel/trace/Makefile
 create mode 100644 kernel/trace/ftrace.c

(limited to 'include/linux')

diff --git a/Makefile b/Makefile
index 20b32351906b..b4a273f19b52 100644
--- a/Makefile
+++ b/Makefile
@@ -528,6 +528,10 @@ KBUILD_CFLAGS	+= -g
 KBUILD_AFLAGS	+= -gdwarf-2
 endif
 
+ifdef CONFIG_FTRACE
+KBUILD_CFLAGS	+= -pg
+endif
+
 # We trigger additional mismatches with less inlining
 ifdef CONFIG_DEBUG_SECTION_MISMATCH
 KBUILD_CFLAGS += $(call cc-option, -fno-inline-functions-called-once)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index fe361ae7ef2f..c742dfeb0dbe 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -23,6 +23,7 @@ config X86
 	select HAVE_OPROFILE
 	select HAVE_KPROBES
 	select HAVE_KRETPROBES
+	select HAVE_FTRACE
 	select HAVE_KVM if ((X86_32 && !X86_VOYAGER && !X86_VISWS && !X86_NUMAQ) || X86_64)
 	select HAVE_ARCH_KGDB if !X86_VOYAGER
 
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 2a609dc3271c..f47b9b5440d2 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -1109,6 +1109,33 @@ ENDPROC(xen_failsafe_callback)
 
 #endif	/* CONFIG_XEN */
 
+#ifdef CONFIG_FTRACE
+ENTRY(mcount)
+	cmpl $ftrace_stub, ftrace_trace_function
+	jnz trace
+
+.globl ftrace_stub
+ftrace_stub:
+	ret
+
+	/* taken from glibc */
+trace:
+	pushl %eax
+	pushl %ecx
+	pushl %edx
+	movl 0xc(%esp), %eax
+	movl 0x4(%ebp), %edx
+
+	call   *ftrace_trace_function
+
+	popl %edx
+	popl %ecx
+	popl %eax
+
+	jmp ftrace_stub
+END(mcount)
+#endif
+
 .section .rodata,"a"
 #include "syscall_table_32.S"
 
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 556a8df522a7..f046e0c64883 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -54,6 +54,43 @@
 
 	.code64
 
+#ifdef CONFIG_FTRACE
+ENTRY(mcount)
+	cmpq $ftrace_stub, ftrace_trace_function
+	jnz trace
+.globl ftrace_stub
+ftrace_stub:
+	retq
+
+trace:
+	/* taken from glibc */
+	subq $0x38, %rsp
+	movq %rax, (%rsp)
+	movq %rcx, 8(%rsp)
+	movq %rdx, 16(%rsp)
+	movq %rsi, 24(%rsp)
+	movq %rdi, 32(%rsp)
+	movq %r8, 40(%rsp)
+	movq %r9, 48(%rsp)
+
+	movq 0x38(%rsp), %rdi
+	movq 8(%rbp), %rsi
+
+	call   *ftrace_trace_function
+
+	movq 48(%rsp), %r9
+	movq 40(%rsp), %r8
+	movq 32(%rsp), %rdi
+	movq 24(%rsp), %rsi
+	movq 16(%rsp), %rdx
+	movq 8(%rsp), %rcx
+	movq (%rsp), %rax
+	addq $0x38, %rsp
+
+	jmp ftrace_stub
+END(mcount)
+#endif
+
 #ifndef CONFIG_PREEMPT
 #define retint_kernel retint_restore_args
 #endif	
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
new file mode 100644
index 000000000000..b96ef14c249a
--- /dev/null
+++ b/include/linux/ftrace.h
@@ -0,0 +1,38 @@
+#ifndef _LINUX_FTRACE_H
+#define _LINUX_FTRACE_H
+
+#ifdef CONFIG_FTRACE
+
+#include <linux/linkage.h>
+
+#define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
+#define CALLER_ADDR1 ((unsigned long)__builtin_return_address(1))
+#define CALLER_ADDR2 ((unsigned long)__builtin_return_address(2))
+
+typedef void (*ftrace_func_t)(unsigned long ip, unsigned long parent_ip);
+
+struct ftrace_ops {
+	ftrace_func_t	  func;
+	struct ftrace_ops *next;
+};
+
+/*
+ * The ftrace_ops must be a static and should also
+ * be read_mostly.  These functions do modify read_mostly variables
+ * so use them sparely. Never free an ftrace_op or modify the
+ * next pointer after it has been registered. Even after unregistering
+ * it, the next pointer may still be used internally.
+ */
+int register_ftrace_function(struct ftrace_ops *ops);
+int unregister_ftrace_function(struct ftrace_ops *ops);
+void clear_ftrace_function(void);
+
+extern void ftrace_stub(unsigned long a0, unsigned long a1);
+extern void mcount(void);
+
+#else /* !CONFIG_FTRACE */
+# define register_ftrace_function(ops) do { } while (0)
+# define unregister_ftrace_function(ops) do { } while (0)
+# define clear_ftrace_function(ops) do { } while (0)
+#endif /* CONFIG_FTRACE */
+#endif /* _LINUX_FTRACE_H */
diff --git a/kernel/Makefile b/kernel/Makefile
index 1c9938addb9d..fa05f6d8bdbf 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -69,6 +69,7 @@ obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
 obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o
 obj-$(CONFIG_MARKERS) += marker.o
 obj-$(CONFIG_LATENCYTOP) += latencytop.o
+obj-$(CONFIG_FTRACE) += trace/
 
 ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
 # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
new file mode 100644
index 000000000000..8185c91417bc
--- /dev/null
+++ b/kernel/trace/Kconfig
@@ -0,0 +1,5 @@
+#
+# Architectures that offer an FTRACE implementation should select HAVE_FTRACE:
+#
+config HAVE_FTRACE
+	bool
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
new file mode 100644
index 000000000000..bf4fd215a6a9
--- /dev/null
+++ b/kernel/trace/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_FTRACE) += libftrace.o
+
+libftrace-y := ftrace.o
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
new file mode 100644
index 000000000000..b6a80b98a3fb
--- /dev/null
+++ b/kernel/trace/ftrace.c
@@ -0,0 +1,138 @@
+/*
+ * Infrastructure for profiling code inserted by 'gcc -pg'.
+ *
+ * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
+ * Copyright (C) 2004-2008 Ingo Molnar <mingo@redhat.com>
+ *
+ * Originally ported from the -rt patch by:
+ *   Copyright (C) 2007 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Based on code in the latency_tracer, that is:
+ *
+ *  Copyright (C) 2004-2006 Ingo Molnar
+ *  Copyright (C) 2004 William Lee Irwin III
+ */
+
+#include <linux/module.h>
+#include <linux/ftrace.h>
+
+static DEFINE_SPINLOCK(ftrace_func_lock);
+static struct ftrace_ops ftrace_list_end __read_mostly =
+{
+	.func = ftrace_stub,
+};
+
+static struct ftrace_ops *ftrace_list __read_mostly = &ftrace_list_end;
+ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub;
+
+/* mcount is defined per arch in assembly */
+EXPORT_SYMBOL(mcount);
+
+notrace void ftrace_list_func(unsigned long ip, unsigned long parent_ip)
+{
+	struct ftrace_ops *op = ftrace_list;
+
+	/* in case someone actually ports this to alpha! */
+	read_barrier_depends();
+
+	while (op != &ftrace_list_end) {
+		/* silly alpha */
+		read_barrier_depends();
+		op->func(ip, parent_ip);
+		op = op->next;
+	};
+}
+
+/**
+ * register_ftrace_function - register a function for profiling
+ * @ops - ops structure that holds the function for profiling.
+ *
+ * Register a function to be called by all functions in the
+ * kernel.
+ *
+ * Note: @ops->func and all the functions it calls must be labeled
+ *       with "notrace", otherwise it will go into a
+ *       recursive loop.
+ */
+int register_ftrace_function(struct ftrace_ops *ops)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&ftrace_func_lock, flags);
+	ops->next = ftrace_list;
+	/*
+	 * We are entering ops into the ftrace_list but another
+	 * CPU might be walking that list. We need to make sure
+	 * the ops->next pointer is valid before another CPU sees
+	 * the ops pointer included into the ftrace_list.
+	 */
+	smp_wmb();
+	ftrace_list = ops;
+	/*
+	 * For one func, simply call it directly.
+	 * For more than one func, call the chain.
+	 */
+	if (ops->next == &ftrace_list_end)
+		ftrace_trace_function = ops->func;
+	else
+		ftrace_trace_function = ftrace_list_func;
+	spin_unlock_irqrestore(&ftrace_func_lock, flags);
+
+	return 0;
+}
+
+/**
+ * unregister_ftrace_function - unresgister a function for profiling.
+ * @ops - ops structure that holds the function to unregister
+ *
+ * Unregister a function that was added to be called by ftrace profiling.
+ */
+int unregister_ftrace_function(struct ftrace_ops *ops)
+{
+	unsigned long flags;
+	struct ftrace_ops **p;
+	int ret = 0;
+
+	spin_lock_irqsave(&ftrace_func_lock, flags);
+
+	/*
+	 * If we are the only function, then the ftrace pointer is
+	 * pointing directly to that function.
+	 */
+	if (ftrace_list == ops && ops->next == &ftrace_list_end) {
+		ftrace_trace_function = ftrace_stub;
+		ftrace_list = &ftrace_list_end;
+		goto out;
+	}
+
+	for (p = &ftrace_list; *p != &ftrace_list_end; p = &(*p)->next)
+		if (*p == ops)
+			break;
+
+	if (*p != ops) {
+		ret = -1;
+		goto out;
+	}
+
+	*p = (*p)->next;
+
+	/* If we only have one func left, then call that directly */
+	if (ftrace_list->next == &ftrace_list_end)
+		ftrace_trace_function = ftrace_list->func;
+
+ out:
+	spin_unlock_irqrestore(&ftrace_func_lock, flags);
+
+	return 0;
+}
+
+/**
+ * clear_ftrace_function - reset the ftrace function
+ *
+ * This NULLs the ftrace function and in essence stops
+ * tracing.  There may be lag
+ */
+void clear_ftrace_function(void)
+{
+	ftrace_trace_function = ftrace_stub;
+}
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index d2099f41aa1e..d8b6279a9b42 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -634,6 +634,8 @@ config LATENCYTOP
 	  Enable this option if you want to use the LatencyTOP tool
 	  to find out which userspace is blocking on what kernel operations.
 
+source kernel/trace/Kconfig
+
 config PROVIDE_OHCI1394_DMA_INIT
 	bool "Remote debugging over FireWire early on boot"
 	depends on PCI && X86
-- 
cgit v1.2.3


From 352ad25aa4a189c667cb2af333948d34692a2d27 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:42 +0200
Subject: ftrace: tracer for scheduler wakeup latency

This patch adds the tracer that tracks the wakeup latency of the
highest priority waking task.

  "wakeup" is added to /debugfs/tracing/available_tracers

Also added to /debugfs/tracing

  tracing_max_latency
     holds the current max latency for the wakeup

  wakeup_thresh
     if set to other than zero, a log will be recorded
     for every wakeup that takes longer than the number
     entered in here (usecs for all counters)
     (deletes previous trace)

Examples:

  (with ftrace_enabled = 0)

============
preemption latency trace v1.1.5 on 2.6.24-rc8
Signed-off-by: Ingo Molnar <mingo@elte.hu>
--------------------------------------------------------------------
 latency: 26 us, #2/2, CPU#1 | (M:rt VP:0, KP:0, SP:0 HP:0 #P:2)
    -----------------
    | task: migration/0-3 (uid:0 nice:-5 policy:1 rt_prio:99)
    -----------------

                 _------=> CPU#
                / _-----=> irqs-off
               | / _----=> need-resched
               || / _---=> hardirq/softirq
               ||| / _--=> preempt-depth
               |||| /
               |||||     delay
   cmd     pid ||||| time  |   caller
      \   /    |||||   \   |   /
   quilt-8551  0d..3    0us+: wake_up_process+0x15/0x17 <ffffffff80233e80> (sched_exec+0xc9/0x100 <ffffffff80235343>)
   quilt-8551  0d..4   26us : sched_switch_callback+0x73/0x81 <ffffffff80338d2f> (schedule+0x483/0x6d5 <ffffffff8048b3ee>)

vim:ft=help
============

  (with ftrace_enabled = 1)

============
preemption latency trace v1.1.5 on 2.6.24-rc8
--------------------------------------------------------------------
 latency: 36 us, #45/45, CPU#0 | (M:rt VP:0, KP:0, SP:0 HP:0 #P:2)
    -----------------
    | task: migration/1-5 (uid:0 nice:-5 policy:1 rt_prio:99)
    -----------------

                 _------=> CPU#
                / _-----=> irqs-off
               | / _----=> need-resched
               || / _---=> hardirq/softirq
               ||| / _--=> preempt-depth
               |||| /
               |||||     delay
   cmd     pid ||||| time  |   caller
      \   /    |||||   \   |   /
    bash-10653 1d..3    0us : wake_up_process+0x15/0x17 <ffffffff80233e80> (sched_exec+0xc9/0x100 <ffffffff80235343>)
    bash-10653 1d..3    1us : try_to_wake_up+0x271/0x2e7 <ffffffff80233dcf> (sub_preempt_count+0xc/0x7a <ffffffff8023309e>)
    bash-10653 1d..2    2us : try_to_wake_up+0x296/0x2e7 <ffffffff80233df4> (update_rq_clock+0x9/0x20 <ffffffff802303f3>)
    bash-10653 1d..2    2us : update_rq_clock+0x1e/0x20 <ffffffff80230408> (__update_rq_clock+0xc/0x90 <ffffffff80230366>)
    bash-10653 1d..2    3us : __update_rq_clock+0x1b/0x90 <ffffffff80230375> (sched_clock+0x9/0x29 <ffffffff80214529>)
    bash-10653 1d..2    4us : try_to_wake_up+0x2a6/0x2e7 <ffffffff80233e04> (activate_task+0xc/0x3f <ffffffff8022ffca>)
    bash-10653 1d..2    4us : activate_task+0x2d/0x3f <ffffffff8022ffeb> (enqueue_task+0xe/0x66 <ffffffff8022ff66>)
    bash-10653 1d..2    5us : enqueue_task+0x5b/0x66 <ffffffff8022ffb3> (enqueue_task_rt+0x9/0x3c <ffffffff80233351>)
    bash-10653 1d..2    6us : try_to_wake_up+0x2ba/0x2e7 <ffffffff80233e18> (check_preempt_wakeup+0x12/0x99 <ffffffff80234f84>)
[...]
    bash-10653 1d..5   33us : tracing_record_cmdline+0xcf/0xd4 <ffffffff80338aad> (_spin_unlock+0x9/0x33 <ffffffff8048d3ec>)
    bash-10653 1d..5   34us : _spin_unlock+0x19/0x33 <ffffffff8048d3fc> (sub_preempt_count+0xc/0x7a <ffffffff8023309e>)
    bash-10653 1d..4   35us : wakeup_sched_switch+0x65/0x2ff <ffffffff80339f66> (_spin_lock_irqsave+0xc/0xa9 <ffffffff8048d08b>)
    bash-10653 1d..4   35us : _spin_lock_irqsave+0x19/0xa9 <ffffffff8048d098> (add_preempt_count+0xe/0x77 <ffffffff8023311a>)
    bash-10653 1d..4   36us : sched_switch_callback+0x73/0x81 <ffffffff80338d2f> (schedule+0x483/0x6d5 <ffffffff8048b3ee>)

vim:ft=help
============

The [...] was added here to not waste your email box space.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h            |  23 ++-
 kernel/trace/Kconfig              |  13 ++
 kernel/trace/Makefile             |   1 +
 kernel/trace/trace_sched_wakeup.c | 310 ++++++++++++++++++++++++++++++++++++++
 4 files changed, 343 insertions(+), 4 deletions(-)
 create mode 100644 kernel/trace/trace_sched_wakeup.c

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index b96ef14c249a..db8a5e7abe41 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -5,10 +5,6 @@
 
 #include <linux/linkage.h>
 
-#define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
-#define CALLER_ADDR1 ((unsigned long)__builtin_return_address(1))
-#define CALLER_ADDR2 ((unsigned long)__builtin_return_address(2))
-
 typedef void (*ftrace_func_t)(unsigned long ip, unsigned long parent_ip);
 
 struct ftrace_ops {
@@ -35,4 +31,23 @@ extern void mcount(void);
 # define unregister_ftrace_function(ops) do { } while (0)
 # define clear_ftrace_function(ops) do { } while (0)
 #endif /* CONFIG_FTRACE */
+
+
+#ifdef CONFIG_FRAME_POINTER
+/* TODO: need to fix this for ARM */
+# define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
+# define CALLER_ADDR1 ((unsigned long)__builtin_return_address(1))
+# define CALLER_ADDR2 ((unsigned long)__builtin_return_address(2))
+# define CALLER_ADDR3 ((unsigned long)__builtin_return_address(3))
+# define CALLER_ADDR4 ((unsigned long)__builtin_return_address(4))
+# define CALLER_ADDR5 ((unsigned long)__builtin_return_address(5))
+#else
+# define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
+# define CALLER_ADDR1 0UL
+# define CALLER_ADDR2 0UL
+# define CALLER_ADDR3 0UL
+# define CALLER_ADDR4 0UL
+# define CALLER_ADDR5 0UL
+#endif
+
 #endif /* _LINUX_FTRACE_H */
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 5d6aa92866cd..892ecc94a82b 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -4,6 +4,9 @@
 config HAVE_FTRACE
 	bool
 
+config TRACER_MAX_TRACE
+	bool
+
 config TRACING
 	bool
 	select DEBUG_FS
@@ -23,6 +26,16 @@ config FTRACE
 	  (the bootup default), then the overhead of the instructions is very
 	  small and not measurable even in micro-benchmarks.
 
+config SCHED_TRACER
+	bool "Scheduling Latency Tracer"
+	depends on DEBUG_KERNEL
+	select TRACING
+	select CONTEXT_SWITCH_TRACER
+	select TRACER_MAX_TRACE
+	help
+	  This tracer tracks the latency of the highest priority task
+	  to be scheduled in, starting from the point it has woken up.
+
 config CONTEXT_SWITCH_TRACER
 	bool "Trace process context switches"
 	depends on DEBUG_KERNEL
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index 6b54ceb7f16e..5508cdb19aea 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -3,5 +3,6 @@ obj-$(CONFIG_FTRACE) += libftrace.o
 obj-$(CONFIG_TRACING) += trace.o
 obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FTRACE) += trace_functions.o
+obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o
 
 libftrace-y := ftrace.o
diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c
new file mode 100644
index 000000000000..7c3ccefcf4c3
--- /dev/null
+++ b/kernel/trace/trace_sched_wakeup.c
@@ -0,0 +1,310 @@
+/*
+ * trace task wakeup timings
+ *
+ * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
+ * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com>
+ *
+ * Based on code from the latency_tracer, that is:
+ *
+ *  Copyright (C) 2004-2006 Ingo Molnar
+ *  Copyright (C) 2004 William Lee Irwin III
+ */
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/debugfs.h>
+#include <linux/kallsyms.h>
+#include <linux/uaccess.h>
+#include <linux/ftrace.h>
+
+#include "trace.h"
+
+static struct trace_array	*wakeup_trace;
+static int __read_mostly	tracer_enabled;
+
+static struct task_struct	*wakeup_task;
+static int			wakeup_cpu;
+static unsigned			wakeup_prio = -1;
+
+static DEFINE_SPINLOCK(wakeup_lock);
+
+static void notrace __wakeup_reset(struct trace_array *tr);
+
+/*
+ * Should this new latency be reported/recorded?
+ */
+static int notrace report_latency(cycle_t delta)
+{
+	if (tracing_thresh) {
+		if (delta < tracing_thresh)
+			return 0;
+	} else {
+		if (delta <= tracing_max_latency)
+			return 0;
+	}
+	return 1;
+}
+
+void notrace
+wakeup_sched_switch(struct task_struct *prev, struct task_struct *next)
+{
+	unsigned long latency = 0, t0 = 0, t1 = 0;
+	struct trace_array *tr = wakeup_trace;
+	struct trace_array_cpu *data;
+	cycle_t T0, T1, delta;
+	unsigned long flags;
+	long disabled;
+	int cpu;
+
+	if (unlikely(!tracer_enabled))
+		return;
+
+	/*
+	 * When we start a new trace, we set wakeup_task to NULL
+	 * and then set tracer_enabled = 1. We want to make sure
+	 * that another CPU does not see the tracer_enabled = 1
+	 * and the wakeup_task with an older task, that might
+	 * actually be the same as next.
+	 */
+	smp_rmb();
+
+	if (next != wakeup_task)
+		return;
+
+	/* The task we are waitng for is waking up */
+	data = tr->data[wakeup_cpu];
+
+	/* disable local data, not wakeup_cpu data */
+	cpu = raw_smp_processor_id();
+	disabled = atomic_inc_return(&tr->data[cpu]->disabled);
+	if (likely(disabled != 1))
+		goto out;
+
+	spin_lock_irqsave(&wakeup_lock, flags);
+
+	/* We could race with grabbing wakeup_lock */
+	if (unlikely(!tracer_enabled || next != wakeup_task))
+		goto out_unlock;
+
+	ftrace(tr, data, CALLER_ADDR1, CALLER_ADDR2, flags);
+
+	/*
+	 * usecs conversion is slow so we try to delay the conversion
+	 * as long as possible:
+	 */
+	T0 = data->preempt_timestamp;
+	T1 = now(cpu);
+	delta = T1-T0;
+
+	if (!report_latency(delta))
+		goto out_unlock;
+
+	latency = nsecs_to_usecs(delta);
+
+	tracing_max_latency = delta;
+	t0 = nsecs_to_usecs(T0);
+	t1 = nsecs_to_usecs(T1);
+
+	update_max_tr(tr, wakeup_task, wakeup_cpu);
+
+	if (tracing_thresh) {
+		printk(KERN_INFO "(%16s-%-5d|#%d): %lu us wakeup latency "
+		       "violates %lu us threshold.\n"
+		       " => started at timestamp %lu: ",
+				wakeup_task->comm, wakeup_task->pid,
+				raw_smp_processor_id(),
+				latency, nsecs_to_usecs(tracing_thresh), t0);
+	} else {
+		printk(KERN_INFO "(%16s-%-5d|#%d): new %lu us maximum "
+		       "wakeup latency.\n => started at timestamp %lu: ",
+				wakeup_task->comm, wakeup_task->pid,
+				cpu, latency, t0);
+	}
+
+	printk(KERN_CONT "   ended at timestamp %lu: ", t1);
+	dump_stack();
+	t1 = nsecs_to_usecs(now(cpu));
+	printk(KERN_CONT "   dump-end timestamp %lu\n\n", t1);
+
+out_unlock:
+	__wakeup_reset(tr);
+	spin_unlock_irqrestore(&wakeup_lock, flags);
+out:
+	atomic_dec(&tr->data[cpu]->disabled);
+}
+
+static void notrace __wakeup_reset(struct trace_array *tr)
+{
+	struct trace_array_cpu *data;
+	int cpu;
+
+	assert_spin_locked(&wakeup_lock);
+
+	for_each_possible_cpu(cpu) {
+		data = tr->data[cpu];
+		tracing_reset(data);
+	}
+
+	wakeup_cpu = -1;
+	wakeup_prio = -1;
+
+	if (wakeup_task)
+		put_task_struct(wakeup_task);
+
+	wakeup_task = NULL;
+}
+
+static void notrace wakeup_reset(struct trace_array *tr)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&wakeup_lock, flags);
+	__wakeup_reset(tr);
+	spin_unlock_irqrestore(&wakeup_lock, flags);
+}
+
+static notrace void
+wakeup_check_start(struct trace_array *tr, struct task_struct *p,
+		   struct task_struct *curr)
+{
+	int cpu = smp_processor_id();
+	unsigned long flags;
+	long disabled;
+
+	if (likely(!rt_task(p)) ||
+			p->prio >= wakeup_prio ||
+			p->prio >= curr->prio)
+		return;
+
+	disabled = atomic_inc_return(&tr->data[cpu]->disabled);
+	if (unlikely(disabled != 1))
+		goto out;
+
+	/* interrupts should be off from try_to_wake_up */
+	spin_lock(&wakeup_lock);
+
+	/* check for races. */
+	if (!tracer_enabled || p->prio >= wakeup_prio)
+		goto out_locked;
+
+	/* reset the trace */
+	__wakeup_reset(tr);
+
+	wakeup_cpu = task_cpu(p);
+	wakeup_prio = p->prio;
+
+	wakeup_task = p;
+	get_task_struct(wakeup_task);
+
+	local_save_flags(flags);
+
+	tr->data[wakeup_cpu]->preempt_timestamp = now(cpu);
+	ftrace(tr, tr->data[wakeup_cpu], CALLER_ADDR1, CALLER_ADDR2, flags);
+
+out_locked:
+	spin_unlock(&wakeup_lock);
+out:
+	atomic_dec(&tr->data[cpu]->disabled);
+}
+
+notrace void
+ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr)
+{
+	if (likely(!tracer_enabled))
+		return;
+
+	wakeup_check_start(wakeup_trace, wakee, curr);
+}
+
+notrace void
+ftrace_wake_up_new_task(struct task_struct *wakee, struct task_struct *curr)
+{
+	if (likely(!tracer_enabled))
+		return;
+
+	wakeup_check_start(wakeup_trace, wakee, curr);
+}
+
+static notrace void start_wakeup_tracer(struct trace_array *tr)
+{
+	wakeup_reset(tr);
+
+	/*
+	 * Don't let the tracer_enabled = 1 show up before
+	 * the wakeup_task is reset. This may be overkill since
+	 * wakeup_reset does a spin_unlock after setting the
+	 * wakeup_task to NULL, but I want to be safe.
+	 * This is a slow path anyway.
+	 */
+	smp_wmb();
+
+	tracer_enabled = 1;
+
+	return;
+}
+
+static notrace void stop_wakeup_tracer(struct trace_array *tr)
+{
+	tracer_enabled = 0;
+}
+
+static notrace void wakeup_tracer_init(struct trace_array *tr)
+{
+	wakeup_trace = tr;
+
+	if (tr->ctrl)
+		start_wakeup_tracer(tr);
+}
+
+static notrace void wakeup_tracer_reset(struct trace_array *tr)
+{
+	if (tr->ctrl) {
+		stop_wakeup_tracer(tr);
+		/* make sure we put back any tasks we are tracing */
+		wakeup_reset(tr);
+	}
+}
+
+static void wakeup_tracer_ctrl_update(struct trace_array *tr)
+{
+	if (tr->ctrl)
+		start_wakeup_tracer(tr);
+	else
+		stop_wakeup_tracer(tr);
+}
+
+static void notrace wakeup_tracer_open(struct trace_iterator *iter)
+{
+	/* stop the trace while dumping */
+	if (iter->tr->ctrl)
+		stop_wakeup_tracer(iter->tr);
+}
+
+static void notrace wakeup_tracer_close(struct trace_iterator *iter)
+{
+	/* forget about any processes we were recording */
+	if (iter->tr->ctrl)
+		start_wakeup_tracer(iter->tr);
+}
+
+static struct tracer wakeup_tracer __read_mostly =
+{
+	.name		= "wakeup",
+	.init		= wakeup_tracer_init,
+	.reset		= wakeup_tracer_reset,
+	.open		= wakeup_tracer_open,
+	.close		= wakeup_tracer_close,
+	.ctrl_update	= wakeup_tracer_ctrl_update,
+	.print_max	= 1,
+};
+
+__init static int init_wakeup_tracer(void)
+{
+	int ret;
+
+	ret = register_tracer(&wakeup_tracer);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+device_initcall(init_wakeup_tracer);
-- 
cgit v1.2.3


From 81d68a96a39844853b37f20cc8282d9b65b78ef3 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:42 +0200
Subject: ftrace: trace irq disabled critical timings

This patch adds latency tracing for critical timings
(how long interrupts are disabled for).

 "irqsoff" is added to /debugfs/tracing/available_tracers

Note:
  tracing_max_latency
    also holds the max latency for irqsoff (in usecs).
   (default to large number so one must start latency tracing)

  tracing_thresh
    threshold (in usecs) to always print out if irqs off
    is detected to be longer than stated here.
    If irq_thresh is non-zero, then max_irq_latency
    is ignored.

Here's an example of a trace with ftrace_enabled = 0

=======
preemption latency trace v1.1.5 on 2.6.24-rc7
Signed-off-by: Ingo Molnar <mingo@elte.hu>
--------------------------------------------------------------------
 latency: 100 us, #3/3, CPU#1 | (M:rt VP:0, KP:0, SP:0 HP:0 #P:2)
    -----------------
    | task: swapper-0 (uid:0 nice:0 policy:0 rt_prio:0)
    -----------------
 => started at: _spin_lock_irqsave+0x2a/0xb7
 => ended at:   _spin_unlock_irqrestore+0x32/0x5f

                 _------=> CPU#
                / _-----=> irqs-off
               | / _----=> need-resched
               || / _---=> hardirq/softirq
               ||| / _--=> preempt-depth
               |||| /
               |||||     delay
   cmd     pid ||||| time  |   caller
      \   /    |||||   \   |   /
 swapper-0     1d.s3    0us+: _spin_lock_irqsave+0x2a/0xb7 (e1000_update_stats+0x47/0x64c [e1000])
 swapper-0     1d.s3  100us : _spin_unlock_irqrestore+0x32/0x5f (e1000_update_stats+0x641/0x64c [e1000])
 swapper-0     1d.s3  100us : trace_hardirqs_on_caller+0x75/0x89 (_spin_unlock_irqrestore+0x32/0x5f)

vim:ft=help
=======

And this is a trace with ftrace_enabled == 1

=======
preemption latency trace v1.1.5 on 2.6.24-rc7
--------------------------------------------------------------------
 latency: 102 us, #12/12, CPU#1 | (M:rt VP:0, KP:0, SP:0 HP:0 #P:2)
    -----------------
    | task: swapper-0 (uid:0 nice:0 policy:0 rt_prio:0)
    -----------------
 => started at: _spin_lock_irqsave+0x2a/0xb7
 => ended at:   _spin_unlock_irqrestore+0x32/0x5f

                 _------=> CPU#
                / _-----=> irqs-off
               | / _----=> need-resched
               || / _---=> hardirq/softirq
               ||| / _--=> preempt-depth
               |||| /
               |||||     delay
   cmd     pid ||||| time  |   caller
      \   /    |||||   \   |   /
 swapper-0     1dNs3    0us+: _spin_lock_irqsave+0x2a/0xb7 (e1000_update_stats+0x47/0x64c [e1000])
 swapper-0     1dNs3   46us : e1000_read_phy_reg+0x16/0x225 [e1000] (e1000_update_stats+0x5e2/0x64c [e1000])
 swapper-0     1dNs3   46us : e1000_swfw_sync_acquire+0x10/0x99 [e1000] (e1000_read_phy_reg+0x49/0x225 [e1000])
 swapper-0     1dNs3   46us : e1000_get_hw_eeprom_semaphore+0x12/0xa6 [e1000] (e1000_swfw_sync_acquire+0x36/0x99 [e1000])
 swapper-0     1dNs3   47us : __const_udelay+0x9/0x47 (e1000_read_phy_reg+0x116/0x225 [e1000])
 swapper-0     1dNs3   47us+: __delay+0x9/0x50 (__const_udelay+0x45/0x47)
 swapper-0     1dNs3   97us : preempt_schedule+0xc/0x84 (__delay+0x4e/0x50)
 swapper-0     1dNs3   98us : e1000_swfw_sync_release+0xc/0x55 [e1000] (e1000_read_phy_reg+0x211/0x225 [e1000])
 swapper-0     1dNs3   99us+: e1000_put_hw_eeprom_semaphore+0x9/0x35 [e1000] (e1000_swfw_sync_release+0x50/0x55 [e1000])
 swapper-0     1dNs3  101us : _spin_unlock_irqrestore+0xe/0x5f (e1000_update_stats+0x641/0x64c [e1000])
 swapper-0     1dNs3  102us : _spin_unlock_irqrestore+0x32/0x5f (e1000_update_stats+0x641/0x64c [e1000])
 swapper-0     1dNs3  102us : trace_hardirqs_on_caller+0x75/0x89 (_spin_unlock_irqrestore+0x32/0x5f)

vim:ft=help
=======

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_64.c |   3 +
 arch/x86/lib/Makefile        |   1 +
 arch/x86/lib/thunk_32.S      |  47 +++++
 arch/x86/lib/thunk_64.S      |  19 +-
 include/asm-x86/irqflags.h   |  24 +--
 include/linux/ftrace.h       |   8 +
 include/linux/irqflags.h     |  12 +-
 kernel/fork.c                |   2 +-
 kernel/lockdep.c             |  23 ++-
 kernel/printk.c              |   2 +
 kernel/trace/Kconfig         |  18 ++
 kernel/trace/Makefile        |   1 +
 kernel/trace/trace_irqsoff.c | 402 +++++++++++++++++++++++++++++++++++++++++++
 13 files changed, 531 insertions(+), 31 deletions(-)
 create mode 100644 arch/x86/lib/thunk_32.S
 create mode 100644 kernel/trace/trace_irqsoff.c

(limited to 'include/linux')

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index e2319f39988b..dd349c92f051 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -165,7 +165,10 @@ void cpu_idle(void)
 			 */
 			local_irq_disable();
 			enter_idle();
+			/* Don't trace irqs off for idle */
+			stop_critical_timings();
 			idle();
+			start_critical_timings();
 			/* In many cases the interrupt that ended idle
 			   has already called exit_idle. But some idle
 			   loops can be woken up without interrupt. */
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index 76f60f52a885..84aa2883fe15 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -5,6 +5,7 @@
 obj-$(CONFIG_SMP) := msr-on-cpu.o
 
 lib-y := delay_$(BITS).o
+lib-y += thunk_$(BITS).o
 lib-y += usercopy_$(BITS).o getuser_$(BITS).o putuser_$(BITS).o
 lib-y += memcpy_$(BITS).o
 
diff --git a/arch/x86/lib/thunk_32.S b/arch/x86/lib/thunk_32.S
new file mode 100644
index 000000000000..650b11e00ecc
--- /dev/null
+++ b/arch/x86/lib/thunk_32.S
@@ -0,0 +1,47 @@
+/*
+ * Trampoline to trace irqs off. (otherwise CALLER_ADDR1 might crash)
+ * Copyright 2008 by Steven Rostedt, Red Hat, Inc
+ *  (inspired by Andi Kleen's thunk_64.S)
+ * Subject to the GNU public license, v.2. No warranty of any kind.
+ */
+
+	#include <linux/linkage.h>
+
+#define ARCH_TRACE_IRQS_ON			\
+	pushl %eax;				\
+	pushl %ecx;				\
+	pushl %edx;				\
+	call trace_hardirqs_on;			\
+	popl %edx;				\
+	popl %ecx;				\
+	popl %eax;
+
+#define ARCH_TRACE_IRQS_OFF			\
+	pushl %eax;				\
+	pushl %ecx;				\
+	pushl %edx;				\
+	call trace_hardirqs_off;		\
+	popl %edx;				\
+	popl %ecx;				\
+	popl %eax;
+
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/* put return address in eax (arg1) */
+	.macro thunk_ra name,func
+	.globl \name
+\name:
+	pushl %eax
+	pushl %ecx
+	pushl %edx
+	/* Place EIP in the arg1 */
+	movl 3*4(%esp), %eax
+	call \func
+	popl %edx
+	popl %ecx
+	popl %eax
+	ret
+	.endm
+
+	thunk_ra trace_hardirqs_on_thunk,trace_hardirqs_on_caller
+	thunk_ra trace_hardirqs_off_thunk,trace_hardirqs_off_caller
+#endif
diff --git a/arch/x86/lib/thunk_64.S b/arch/x86/lib/thunk_64.S
index e009251d4e9f..bf9a7d5a5428 100644
--- a/arch/x86/lib/thunk_64.S
+++ b/arch/x86/lib/thunk_64.S
@@ -2,6 +2,7 @@
  * Save registers before calling assembly functions. This avoids
  * disturbance of register allocation in some inline assembly constructs.
  * Copyright 2001,2002 by Andi Kleen, SuSE Labs.
+ * Added trace_hardirqs callers - Copyright 2007 Steven Rostedt, Red Hat, Inc.
  * Subject to the GNU public license, v.2. No warranty of any kind.
  */
 
@@ -42,8 +43,22 @@
 #endif	
 	
 #ifdef CONFIG_TRACE_IRQFLAGS
-	thunk trace_hardirqs_on_thunk,trace_hardirqs_on
-	thunk trace_hardirqs_off_thunk,trace_hardirqs_off
+	/* put return address in rdi (arg1) */
+	.macro thunk_ra name,func
+	.globl \name
+\name:
+	CFI_STARTPROC
+	SAVE_ARGS
+	/* SAVE_ARGS pushs 9 elements */
+	/* the next element would be the rip */
+	movq 9*8(%rsp), %rdi
+	call \func
+	jmp  restore
+	CFI_ENDPROC
+	.endm
+
+	thunk_ra trace_hardirqs_on_thunk,trace_hardirqs_on_caller
+	thunk_ra trace_hardirqs_off_thunk,trace_hardirqs_off_caller
 #endif
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h
index c242527f970e..24d71b1eb189 100644
--- a/include/asm-x86/irqflags.h
+++ b/include/asm-x86/irqflags.h
@@ -179,8 +179,6 @@ static inline void trace_hardirqs_fixup(void)
  * have a reliable stack. x86_64 only.
  */
 #define SWAPGS_UNSAFE_STACK	swapgs
-#define ARCH_TRACE_IRQS_ON		call trace_hardirqs_on_thunk
-#define ARCH_TRACE_IRQS_OFF		call trace_hardirqs_off_thunk
 #define ARCH_LOCKDEP_SYS_EXIT		call lockdep_sys_exit_thunk
 #define ARCH_LOCKDEP_SYS_EXIT_IRQ	\
 	TRACE_IRQS_ON; \
@@ -192,24 +190,6 @@ static inline void trace_hardirqs_fixup(void)
 	TRACE_IRQS_OFF;
 
 #else
-#define ARCH_TRACE_IRQS_ON			\
-	pushl %eax;				\
-	pushl %ecx;				\
-	pushl %edx;				\
-	call trace_hardirqs_on;			\
-	popl %edx;				\
-	popl %ecx;				\
-	popl %eax;
-
-#define ARCH_TRACE_IRQS_OFF			\
-	pushl %eax;				\
-	pushl %ecx;				\
-	pushl %edx;				\
-	call trace_hardirqs_off;		\
-	popl %edx;				\
-	popl %ecx;				\
-	popl %eax;
-
 #define ARCH_LOCKDEP_SYS_EXIT			\
 	pushl %eax;				\
 	pushl %ecx;				\
@@ -223,8 +203,8 @@ static inline void trace_hardirqs_fixup(void)
 #endif
 
 #ifdef CONFIG_TRACE_IRQFLAGS
-#  define TRACE_IRQS_ON		ARCH_TRACE_IRQS_ON
-#  define TRACE_IRQS_OFF	ARCH_TRACE_IRQS_OFF
+#  define TRACE_IRQS_ON		call trace_hardirqs_on_thunk;
+#  define TRACE_IRQS_OFF	call trace_hardirqs_off_thunk;
 #else
 #  define TRACE_IRQS_ON
 #  define TRACE_IRQS_OFF
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index db8a5e7abe41..0a20445dcbcc 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -50,4 +50,12 @@ extern void mcount(void);
 # define CALLER_ADDR5 0UL
 #endif
 
+#ifdef CONFIG_IRQSOFF_TRACER
+  extern void notrace time_hardirqs_on(unsigned long a0, unsigned long a1);
+  extern void notrace time_hardirqs_off(unsigned long a0, unsigned long a1);
+#else
+# define time_hardirqs_on(a0, a1)		do { } while (0)
+# define time_hardirqs_off(a0, a1)		do { } while (0)
+#endif
+
 #endif /* _LINUX_FTRACE_H */
diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index e600c4e9b8c5..5b711d4e9fd9 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -12,10 +12,10 @@
 #define _LINUX_TRACE_IRQFLAGS_H
 
 #ifdef CONFIG_TRACE_IRQFLAGS
-  extern void trace_hardirqs_on(void);
-  extern void trace_hardirqs_off(void);
   extern void trace_softirqs_on(unsigned long ip);
   extern void trace_softirqs_off(unsigned long ip);
+  extern void trace_hardirqs_on(void);
+  extern void trace_hardirqs_off(void);
 # define trace_hardirq_context(p)	((p)->hardirq_context)
 # define trace_softirq_context(p)	((p)->softirq_context)
 # define trace_hardirqs_enabled(p)	((p)->hardirqs_enabled)
@@ -41,6 +41,14 @@
 # define INIT_TRACE_IRQFLAGS
 #endif
 
+#ifdef CONFIG_IRQSOFF_TRACER
+ extern void stop_critical_timings(void);
+ extern void start_critical_timings(void);
+#else
+# define stop_critical_timings() do { } while (0)
+# define start_critical_timings() do { } while (0)
+#endif
+
 #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
 
 #include <asm/irqflags.h>
diff --git a/kernel/fork.c b/kernel/fork.c
index 19908b26cf80..d66d676dc362 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -909,7 +909,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 
 	rt_mutex_init_task(p);
 
-#ifdef CONFIG_TRACE_IRQFLAGS
+#if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_LOCKDEP)
 	DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled);
 	DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
 #endif
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 81a4e4a3f087..e21924365ea3 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -39,6 +39,7 @@
 #include <linux/irqflags.h>
 #include <linux/utsname.h>
 #include <linux/hash.h>
+#include <linux/ftrace.h>
 
 #include <asm/sections.h>
 
@@ -982,7 +983,7 @@ check_noncircular(struct lock_class *source, unsigned int depth)
 	return 1;
 }
 
-#ifdef CONFIG_TRACE_IRQFLAGS
+#if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING)
 /*
  * Forwards and backwards subgraph searching, for the purposes of
  * proving that two subgraphs can be connected by a new dependency
@@ -1680,7 +1681,7 @@ valid_state(struct task_struct *curr, struct held_lock *this,
 static int mark_lock(struct task_struct *curr, struct held_lock *this,
 		     enum lock_usage_bit new_bit);
 
-#ifdef CONFIG_TRACE_IRQFLAGS
+#if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING)
 
 /*
  * print irq inversion bug:
@@ -2013,11 +2014,13 @@ void early_boot_irqs_on(void)
 /*
  * Hardirqs will be enabled:
  */
-void trace_hardirqs_on(void)
+void notrace trace_hardirqs_on_caller(unsigned long a0)
 {
 	struct task_struct *curr = current;
 	unsigned long ip;
 
+	time_hardirqs_on(CALLER_ADDR0, a0);
+
 	if (unlikely(!debug_locks || current->lockdep_recursion))
 		return;
 
@@ -2055,16 +2058,23 @@ void trace_hardirqs_on(void)
 	curr->hardirq_enable_event = ++curr->irq_events;
 	debug_atomic_inc(&hardirqs_on_events);
 }
+EXPORT_SYMBOL(trace_hardirqs_on_caller);
 
+void notrace trace_hardirqs_on(void)
+{
+	trace_hardirqs_on_caller(CALLER_ADDR0);
+}
 EXPORT_SYMBOL(trace_hardirqs_on);
 
 /*
  * Hardirqs were disabled:
  */
-void trace_hardirqs_off(void)
+void notrace trace_hardirqs_off_caller(unsigned long a0)
 {
 	struct task_struct *curr = current;
 
+	time_hardirqs_off(CALLER_ADDR0, a0);
+
 	if (unlikely(!debug_locks || current->lockdep_recursion))
 		return;
 
@@ -2082,7 +2092,12 @@ void trace_hardirqs_off(void)
 	} else
 		debug_atomic_inc(&redundant_hardirqs_off);
 }
+EXPORT_SYMBOL(trace_hardirqs_off_caller);
 
+void notrace trace_hardirqs_off(void)
+{
+	trace_hardirqs_off_caller(CALLER_ADDR0);
+}
 EXPORT_SYMBOL(trace_hardirqs_off);
 
 /*
diff --git a/kernel/printk.c b/kernel/printk.c
index 8fb01c32aa3b..ae7d5b9e535d 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -1041,7 +1041,9 @@ void release_console_sem(void)
 		_log_end = log_end;
 		con_start = log_end;		/* Flush */
 		spin_unlock(&logbuf_lock);
+		stop_critical_timings();	/* don't trace print latency */
 		call_console_drivers(_con_start, _log_end);
+		start_critical_timings();
 		local_irq_restore(flags);
 	}
 	console_locked = 0;
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 892ecc94a82b..896df1cf6adc 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -26,6 +26,24 @@ config FTRACE
 	  (the bootup default), then the overhead of the instructions is very
 	  small and not measurable even in micro-benchmarks.
 
+config IRQSOFF_TRACER
+	bool "Interrupts-off Latency Tracer"
+	default n
+	depends on TRACE_IRQFLAGS_SUPPORT
+	depends on GENERIC_TIME
+	select TRACE_IRQFLAGS
+	select TRACING
+	select TRACER_MAX_TRACE
+	help
+	  This option measures the time spent in irqs-off critical
+	  sections, with microsecond accuracy.
+
+	  The default measurement method is a maximum search, which is
+	  disabled by default and can be runtime (re-)started
+	  via:
+
+	      echo 0 > /debugfs/tracing/tracing_max_latency
+
 config SCHED_TRACER
 	bool "Scheduling Latency Tracer"
 	depends on DEBUG_KERNEL
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index 5508cdb19aea..46be8647fb65 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -3,6 +3,7 @@ obj-$(CONFIG_FTRACE) += libftrace.o
 obj-$(CONFIG_TRACING) += trace.o
 obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FTRACE) += trace_functions.o
+obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o
 
 libftrace-y := ftrace.o
diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
new file mode 100644
index 000000000000..a9131b0cf1a5
--- /dev/null
+++ b/kernel/trace/trace_irqsoff.c
@@ -0,0 +1,402 @@
+/*
+ * trace irqs off criticall timings
+ *
+ * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
+ * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com>
+ *
+ * From code in the latency_tracer, that is:
+ *
+ *  Copyright (C) 2004-2006 Ingo Molnar
+ *  Copyright (C) 2004 William Lee Irwin III
+ */
+#include <linux/kallsyms.h>
+#include <linux/debugfs.h>
+#include <linux/uaccess.h>
+#include <linux/module.h>
+#include <linux/ftrace.h>
+#include <linux/fs.h>
+
+#include "trace.h"
+
+static struct trace_array		*irqsoff_trace __read_mostly;
+static int				tracer_enabled __read_mostly;
+
+/*
+ * Sequence count - we record it when starting a measurement and
+ * skip the latency if the sequence has changed - some other section
+ * did a maximum and could disturb our measurement with serial console
+ * printouts, etc. Truly coinciding maximum latencies should be rare
+ * and what happens together happens separately as well, so this doesnt
+ * decrease the validity of the maximum found:
+ */
+static __cacheline_aligned_in_smp	unsigned long max_sequence;
+
+#ifdef CONFIG_FTRACE
+/*
+ * irqsoff uses its own tracer function to keep the overhead down:
+ */
+static void notrace
+irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip)
+{
+	struct trace_array *tr = irqsoff_trace;
+	struct trace_array_cpu *data;
+	unsigned long flags;
+	long disabled;
+	int cpu;
+
+	if (likely(!tracer_enabled))
+		return;
+
+	local_save_flags(flags);
+
+	if (!irqs_disabled_flags(flags))
+		return;
+
+	cpu = raw_smp_processor_id();
+	data = tr->data[cpu];
+	disabled = atomic_inc_return(&data->disabled);
+
+	if (likely(disabled == 1))
+		ftrace(tr, data, ip, parent_ip, flags);
+
+	atomic_dec(&data->disabled);
+}
+
+static struct ftrace_ops trace_ops __read_mostly =
+{
+	.func = irqsoff_tracer_call,
+};
+#endif /* CONFIG_FTRACE */
+
+/*
+ * Should this new latency be reported/recorded?
+ */
+static int notrace report_latency(cycle_t delta)
+{
+	if (tracing_thresh) {
+		if (delta < tracing_thresh)
+			return 0;
+	} else {
+		if (delta <= tracing_max_latency)
+			return 0;
+	}
+	return 1;
+}
+
+static void notrace
+check_critical_timing(struct trace_array *tr,
+		      struct trace_array_cpu *data,
+		      unsigned long parent_ip,
+		      int cpu)
+{
+	unsigned long latency, t0, t1;
+	cycle_t T0, T1, T2, delta;
+	unsigned long flags;
+
+	/*
+	 * usecs conversion is slow so we try to delay the conversion
+	 * as long as possible:
+	 */
+	T0 = data->preempt_timestamp;
+	T1 = now(cpu);
+	delta = T1-T0;
+
+	local_save_flags(flags);
+
+	if (!report_latency(delta))
+		goto out;
+
+	ftrace(tr, data, CALLER_ADDR0, parent_ip, flags);
+	/*
+	 * Update the timestamp, because the trace entry above
+	 * might change it (it can only get larger so the latency
+	 * is fair to be reported):
+	 */
+	T2 = now(cpu);
+
+	delta = T2-T0;
+
+	latency = nsecs_to_usecs(delta);
+
+	if (data->critical_sequence != max_sequence)
+		goto out;
+
+	tracing_max_latency = delta;
+	t0 = nsecs_to_usecs(T0);
+	t1 = nsecs_to_usecs(T1);
+
+	data->critical_end = parent_ip;
+
+	update_max_tr_single(tr, current, cpu);
+
+	if (tracing_thresh)
+		printk(KERN_INFO "(%16s-%-5d|#%d): %lu us critical section "
+		       "violates %lu us threshold.\n"
+		       " => started at timestamp %lu: ",
+				current->comm, current->pid,
+				raw_smp_processor_id(),
+				latency, nsecs_to_usecs(tracing_thresh), t0);
+	else
+		printk(KERN_INFO "(%16s-%-5d|#%d):"
+		       " new %lu us maximum-latency "
+		       "critical section.\n => started at timestamp %lu: ",
+				current->comm, current->pid,
+				raw_smp_processor_id(),
+				latency, t0);
+
+	print_symbol(KERN_CONT "<%s>\n", data->critical_start);
+	printk(KERN_CONT " =>   ended at timestamp %lu: ", t1);
+	print_symbol(KERN_CONT "<%s>\n", data->critical_end);
+	dump_stack();
+	t1 = nsecs_to_usecs(now(cpu));
+	printk(KERN_CONT " =>   dump-end timestamp %lu\n\n", t1);
+
+	max_sequence++;
+
+out:
+	data->critical_sequence = max_sequence;
+	data->preempt_timestamp = now(cpu);
+	tracing_reset(data);
+	ftrace(tr, data, CALLER_ADDR0, parent_ip, flags);
+}
+
+static inline void notrace
+start_critical_timing(unsigned long ip, unsigned long parent_ip)
+{
+	int cpu;
+	struct trace_array *tr = irqsoff_trace;
+	struct trace_array_cpu *data;
+	unsigned long flags;
+
+	if (likely(!tracer_enabled))
+		return;
+
+	cpu = raw_smp_processor_id();
+	data = tr->data[cpu];
+
+	if (unlikely(!data) || unlikely(!data->trace) ||
+	    data->critical_start || atomic_read(&data->disabled))
+		return;
+
+	atomic_inc(&data->disabled);
+
+	data->critical_sequence = max_sequence;
+	data->preempt_timestamp = now(cpu);
+	data->critical_start = parent_ip;
+	tracing_reset(data);
+
+	local_save_flags(flags);
+	ftrace(tr, data, ip, parent_ip, flags);
+
+	atomic_dec(&data->disabled);
+}
+
+static inline void notrace
+stop_critical_timing(unsigned long ip, unsigned long parent_ip)
+{
+	int cpu;
+	struct trace_array *tr = irqsoff_trace;
+	struct trace_array_cpu *data;
+	unsigned long flags;
+
+	if (likely(!tracer_enabled))
+		return;
+
+	cpu = raw_smp_processor_id();
+	data = tr->data[cpu];
+
+	if (unlikely(!data) || unlikely(!data->trace) ||
+	    !data->critical_start || atomic_read(&data->disabled))
+		return;
+
+	atomic_inc(&data->disabled);
+	local_save_flags(flags);
+	ftrace(tr, data, ip, parent_ip, flags);
+	check_critical_timing(tr, data, parent_ip, cpu);
+	data->critical_start = 0;
+	atomic_dec(&data->disabled);
+}
+
+void notrace start_critical_timings(void)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
+}
+
+void notrace stop_critical_timings(void)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
+}
+
+#ifdef CONFIG_PROVE_LOCKING
+void notrace time_hardirqs_on(unsigned long a0, unsigned long a1)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		stop_critical_timing(a0, a1);
+}
+
+void notrace time_hardirqs_off(unsigned long a0, unsigned long a1)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		start_critical_timing(a0, a1);
+}
+
+#else /* !CONFIG_PROVE_LOCKING */
+
+/*
+ * Stubs:
+ */
+
+void early_boot_irqs_off(void)
+{
+}
+
+void early_boot_irqs_on(void)
+{
+}
+
+void trace_softirqs_on(unsigned long ip)
+{
+}
+
+void trace_softirqs_off(unsigned long ip)
+{
+}
+
+inline void print_irqtrace_events(struct task_struct *curr)
+{
+}
+
+/*
+ * We are only interested in hardirq on/off events:
+ */
+void notrace trace_hardirqs_on(void)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
+}
+EXPORT_SYMBOL(trace_hardirqs_on);
+
+void notrace trace_hardirqs_off(void)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
+}
+EXPORT_SYMBOL(trace_hardirqs_off);
+
+void notrace trace_hardirqs_on_caller(unsigned long caller_addr)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		stop_critical_timing(CALLER_ADDR0, caller_addr);
+}
+EXPORT_SYMBOL(trace_hardirqs_on_caller);
+
+void notrace trace_hardirqs_off_caller(unsigned long caller_addr)
+{
+	unsigned long flags;
+
+	local_save_flags(flags);
+
+	if (irqs_disabled_flags(flags))
+		start_critical_timing(CALLER_ADDR0, caller_addr);
+}
+EXPORT_SYMBOL(trace_hardirqs_off_caller);
+
+#endif /* CONFIG_PROVE_LOCKING */
+
+static void start_irqsoff_tracer(struct trace_array *tr)
+{
+	tracer_enabled = 1;
+	register_ftrace_function(&trace_ops);
+}
+
+static void stop_irqsoff_tracer(struct trace_array *tr)
+{
+	unregister_ftrace_function(&trace_ops);
+	tracer_enabled = 0;
+}
+
+static void irqsoff_tracer_init(struct trace_array *tr)
+{
+	irqsoff_trace = tr;
+	/* make sure that the tracer is visibel */
+	smp_wmb();
+
+	if (tr->ctrl)
+		start_irqsoff_tracer(tr);
+}
+
+static void irqsoff_tracer_reset(struct trace_array *tr)
+{
+	if (tr->ctrl)
+		stop_irqsoff_tracer(tr);
+}
+
+static void irqsoff_tracer_ctrl_update(struct trace_array *tr)
+{
+	if (tr->ctrl)
+		start_irqsoff_tracer(tr);
+	else
+		stop_irqsoff_tracer(tr);
+}
+
+static void notrace irqsoff_tracer_open(struct trace_iterator *iter)
+{
+	/* stop the trace while dumping */
+	if (iter->tr->ctrl)
+		stop_irqsoff_tracer(iter->tr);
+}
+
+static void notrace irqsoff_tracer_close(struct trace_iterator *iter)
+{
+	if (iter->tr->ctrl)
+		start_irqsoff_tracer(iter->tr);
+}
+
+static struct tracer irqsoff_tracer __read_mostly =
+{
+	.name		= "irqsoff",
+	.init		= irqsoff_tracer_init,
+	.reset		= irqsoff_tracer_reset,
+	.open		= irqsoff_tracer_open,
+	.close		= irqsoff_tracer_close,
+	.ctrl_update	= irqsoff_tracer_ctrl_update,
+	.print_max	= 1,
+};
+
+__init static int init_irqsoff_tracer(void)
+{
+	register_tracer(&irqsoff_tracer);
+
+	return 0;
+}
+device_initcall(init_irqsoff_tracer);
-- 
cgit v1.2.3


From 6cd8a4bb2f97527a9ceb30bc77ea4e959c6a95e3 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:42 +0200
Subject: ftrace: trace preempt off critical timings

Add preempt off timings. A lot of kernel core code is taken from the RT patch
latency trace that was written by Ingo Molnar.

This adds "preemptoff" and "preemptirqsoff" to /debugfs/tracing/available_tracers

Now instead of just tracing irqs off, preemption off can be selected
to be recorded.

When this is selected, it shares the same files as irqs off timings.
One can either trace preemption off, irqs off, or one or the other off.

By echoing "preemptoff" into /debugfs/tracing/current_tracer, recording
of preempt off only is performed. "irqsoff" will only record the time
irqs are disabled, but "preemptirqsoff" will take the total time irqs
or preemption are disabled. Runtime switching of these options is now
supported by simpling echoing in the appropriate trace name into
/debugfs/tracing/current_tracer.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/process_32.c |   3 +
 include/linux/ftrace.h       |   8 ++
 include/linux/irqflags.h     |   3 +-
 include/linux/preempt.h      |   2 +-
 kernel/sched.c               |  24 +++++-
 kernel/trace/Kconfig         |  25 ++++++
 kernel/trace/Makefile        |   1 +
 kernel/trace/trace_irqsoff.c | 184 +++++++++++++++++++++++++++++++------------
 8 files changed, 197 insertions(+), 53 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index f8476dfbb60d..a30aa1f2607a 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -185,7 +185,10 @@ void cpu_idle(void)
 
 			local_irq_disable();
 			__get_cpu_var(irq_stat).idle_timestamp = jiffies;
+			/* Don't trace irqs off for idle */
+			stop_critical_timings();
 			idle();
+			start_critical_timings();
 		}
 		tick_nohz_restart_sched_tick();
 		preempt_enable_no_resched();
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 0a20445dcbcc..740c97dcf9cb 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -58,4 +58,12 @@ extern void mcount(void);
 # define time_hardirqs_off(a0, a1)		do { } while (0)
 #endif
 
+#ifdef CONFIG_PREEMPT_TRACER
+  extern void notrace trace_preempt_on(unsigned long a0, unsigned long a1);
+  extern void notrace trace_preempt_off(unsigned long a0, unsigned long a1);
+#else
+# define trace_preempt_on(a0, a1)		do { } while (0)
+# define trace_preempt_off(a0, a1)		do { } while (0)
+#endif
+
 #endif /* _LINUX_FTRACE_H */
diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index 5b711d4e9fd9..2b1c2e58566e 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -41,7 +41,8 @@
 # define INIT_TRACE_IRQFLAGS
 #endif
 
-#ifdef CONFIG_IRQSOFF_TRACER
+#if defined(CONFIG_IRQSOFF_TRACER) || \
+	defined(CONFIG_PREEMPT_TRACER)
  extern void stop_critical_timings(void);
  extern void start_critical_timings(void);
 #else
diff --git a/include/linux/preempt.h b/include/linux/preempt.h
index 36b03d50bf40..72b1a10a59b6 100644
--- a/include/linux/preempt.h
+++ b/include/linux/preempt.h
@@ -10,7 +10,7 @@
 #include <linux/linkage.h>
 #include <linux/list.h>
 
-#ifdef CONFIG_DEBUG_PREEMPT
+#if defined(CONFIG_DEBUG_PREEMPT) || defined(CONFIG_PREEMPT_TRACER)
   extern void add_preempt_count(int val);
   extern void sub_preempt_count(int val);
 #else
diff --git a/kernel/sched.c b/kernel/sched.c
index 73e600852365..328494e28df2 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -70,6 +70,7 @@
 #include <linux/bootmem.h>
 #include <linux/debugfs.h>
 #include <linux/ctype.h>
+#include <linux/ftrace.h>
 
 #include <asm/tlb.h>
 #include <asm/irq_regs.h>
@@ -4365,26 +4366,44 @@ void scheduler_tick(void)
 #endif
 }
 
-#if defined(CONFIG_PREEMPT) && defined(CONFIG_DEBUG_PREEMPT)
+#if defined(CONFIG_PREEMPT) && (defined(CONFIG_DEBUG_PREEMPT) || \
+				defined(CONFIG_PREEMPT_TRACER))
+
+static inline unsigned long get_parent_ip(unsigned long addr)
+{
+	if (in_lock_functions(addr)) {
+		addr = CALLER_ADDR2;
+		if (in_lock_functions(addr))
+			addr = CALLER_ADDR3;
+	}
+	return addr;
+}
 
 void __kprobes add_preempt_count(int val)
 {
+#ifdef CONFIG_DEBUG_PREEMPT
 	/*
 	 * Underflow?
 	 */
 	if (DEBUG_LOCKS_WARN_ON((preempt_count() < 0)))
 		return;
+#endif
 	preempt_count() += val;
+#ifdef CONFIG_DEBUG_PREEMPT
 	/*
 	 * Spinlock count overflowing soon?
 	 */
 	DEBUG_LOCKS_WARN_ON((preempt_count() & PREEMPT_MASK) >=
 				PREEMPT_MASK - 10);
+#endif
+	if (preempt_count() == val)
+		trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1));
 }
 EXPORT_SYMBOL(add_preempt_count);
 
 void __kprobes sub_preempt_count(int val)
 {
+#ifdef CONFIG_DEBUG_PREEMPT
 	/*
 	 * Underflow?
 	 */
@@ -4396,7 +4415,10 @@ void __kprobes sub_preempt_count(int val)
 	if (DEBUG_LOCKS_WARN_ON((val < PREEMPT_MASK) &&
 			!(preempt_count() & PREEMPT_MASK)))
 		return;
+#endif
 
+	if (preempt_count() == val)
+		trace_preempt_on(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1));
 	preempt_count() -= val;
 }
 EXPORT_SYMBOL(sub_preempt_count);
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 896df1cf6adc..6430016b98e8 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -44,6 +44,31 @@ config IRQSOFF_TRACER
 
 	      echo 0 > /debugfs/tracing/tracing_max_latency
 
+	  (Note that kernel size and overhead increases with this option
+	  enabled. This option and the preempt-off timing option can be
+	  used together or separately.)
+
+config PREEMPT_TRACER
+	bool "Preemption-off Latency Tracer"
+	default n
+	depends on GENERIC_TIME
+	depends on PREEMPT
+	select TRACING
+	select TRACER_MAX_TRACE
+	help
+	  This option measures the time spent in preemption off critical
+	  sections, with microsecond accuracy.
+
+	  The default measurement method is a maximum search, which is
+	  disabled by default and can be runtime (re-)started
+	  via:
+
+	      echo 0 > /debugfs/tracing/tracing_max_latency
+
+	  (Note that kernel size and overhead increases with this option
+	  enabled. This option and the irqs-off timing option can be
+	  used together or separately.)
+
 config SCHED_TRACER
 	bool "Scheduling Latency Tracer"
 	depends on DEBUG_KERNEL
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index 46be8647fb65..3fec653d6533 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -4,6 +4,7 @@ obj-$(CONFIG_TRACING) += trace.o
 obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FTRACE) += trace_functions.o
 obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
+obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o
 
 libftrace-y := ftrace.o
diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
index a9131b0cf1a5..8b1231633dc5 100644
--- a/kernel/trace/trace_irqsoff.c
+++ b/kernel/trace/trace_irqsoff.c
@@ -21,6 +21,36 @@
 static struct trace_array		*irqsoff_trace __read_mostly;
 static int				tracer_enabled __read_mostly;
 
+static DEFINE_PER_CPU(int, tracing_cpu);
+
+enum {
+	TRACER_IRQS_OFF		= (1 << 1),
+	TRACER_PREEMPT_OFF	= (1 << 2),
+};
+
+static int trace_type __read_mostly;
+
+#ifdef CONFIG_PREEMPT_TRACER
+static inline int notrace
+preempt_trace(void)
+{
+	return ((trace_type & TRACER_PREEMPT_OFF) && preempt_count());
+}
+#else
+# define preempt_trace() (0)
+#endif
+
+#ifdef CONFIG_IRQSOFF_TRACER
+static inline int notrace
+irq_trace(void)
+{
+	return ((trace_type & TRACER_IRQS_OFF) &&
+		irqs_disabled());
+}
+#else
+# define irq_trace() (0)
+#endif
+
 /*
  * Sequence count - we record it when starting a measurement and
  * skip the latency if the sequence has changed - some other section
@@ -44,14 +74,11 @@ irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip)
 	long disabled;
 	int cpu;
 
-	if (likely(!tracer_enabled))
+	if (likely(!__get_cpu_var(tracing_cpu)))
 		return;
 
 	local_save_flags(flags);
 
-	if (!irqs_disabled_flags(flags))
-		return;
-
 	cpu = raw_smp_processor_id();
 	data = tr->data[cpu];
 	disabled = atomic_inc_return(&data->disabled);
@@ -171,23 +198,29 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip)
 	if (likely(!tracer_enabled))
 		return;
 
+	if (__get_cpu_var(tracing_cpu))
+		return;
+
 	cpu = raw_smp_processor_id();
 	data = tr->data[cpu];
 
 	if (unlikely(!data) || unlikely(!data->trace) ||
-	    data->critical_start || atomic_read(&data->disabled))
+	    atomic_read(&data->disabled))
 		return;
 
 	atomic_inc(&data->disabled);
 
 	data->critical_sequence = max_sequence;
 	data->preempt_timestamp = now(cpu);
-	data->critical_start = parent_ip;
+	data->critical_start = parent_ip ? : ip;
 	tracing_reset(data);
 
 	local_save_flags(flags);
+
 	ftrace(tr, data, ip, parent_ip, flags);
 
+	__get_cpu_var(tracing_cpu) = 1;
+
 	atomic_dec(&data->disabled);
 }
 
@@ -199,7 +232,13 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip)
 	struct trace_array_cpu *data;
 	unsigned long flags;
 
-	if (likely(!tracer_enabled))
+	/* Always clear the tracing cpu on stopping the trace */
+	if (unlikely(__get_cpu_var(tracing_cpu)))
+		__get_cpu_var(tracing_cpu) = 0;
+	else
+		return;
+
+	if (!tracer_enabled)
 		return;
 
 	cpu = raw_smp_processor_id();
@@ -212,49 +251,35 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip)
 	atomic_inc(&data->disabled);
 	local_save_flags(flags);
 	ftrace(tr, data, ip, parent_ip, flags);
-	check_critical_timing(tr, data, parent_ip, cpu);
+	check_critical_timing(tr, data, parent_ip ? : ip, cpu);
 	data->critical_start = 0;
 	atomic_dec(&data->disabled);
 }
 
+/* start and stop critical timings used to for stoppage (in idle) */
 void notrace start_critical_timings(void)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (preempt_trace() || irq_trace())
 		start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
 
 void notrace stop_critical_timings(void)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (preempt_trace() || irq_trace())
 		stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
 
+#ifdef CONFIG_IRQSOFF_TRACER
 #ifdef CONFIG_PROVE_LOCKING
 void notrace time_hardirqs_on(unsigned long a0, unsigned long a1)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irq_trace())
 		stop_critical_timing(a0, a1);
 }
 
 void notrace time_hardirqs_off(unsigned long a0, unsigned long a1)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irq_trace())
 		start_critical_timing(a0, a1);
 }
 
@@ -289,49 +314,46 @@ inline void print_irqtrace_events(struct task_struct *curr)
  */
 void notrace trace_hardirqs_on(void)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irq_trace())
 		stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
 EXPORT_SYMBOL(trace_hardirqs_on);
 
 void notrace trace_hardirqs_off(void)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irq_trace())
 		start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
 EXPORT_SYMBOL(trace_hardirqs_off);
 
 void notrace trace_hardirqs_on_caller(unsigned long caller_addr)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irq_trace())
 		stop_critical_timing(CALLER_ADDR0, caller_addr);
 }
 EXPORT_SYMBOL(trace_hardirqs_on_caller);
 
 void notrace trace_hardirqs_off_caller(unsigned long caller_addr)
 {
-	unsigned long flags;
-
-	local_save_flags(flags);
-
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irq_trace())
 		start_critical_timing(CALLER_ADDR0, caller_addr);
 }
 EXPORT_SYMBOL(trace_hardirqs_off_caller);
 
 #endif /* CONFIG_PROVE_LOCKING */
+#endif /*  CONFIG_IRQSOFF_TRACER */
+
+#ifdef CONFIG_PREEMPT_TRACER
+void notrace trace_preempt_on(unsigned long a0, unsigned long a1)
+{
+	stop_critical_timing(a0, a1);
+}
+
+void notrace trace_preempt_off(unsigned long a0, unsigned long a1)
+{
+	start_critical_timing(a0, a1);
+}
+#endif /* CONFIG_PREEMPT_TRACER */
 
 static void start_irqsoff_tracer(struct trace_array *tr)
 {
@@ -345,7 +367,7 @@ static void stop_irqsoff_tracer(struct trace_array *tr)
 	tracer_enabled = 0;
 }
 
-static void irqsoff_tracer_init(struct trace_array *tr)
+static void __irqsoff_tracer_init(struct trace_array *tr)
 {
 	irqsoff_trace = tr;
 	/* make sure that the tracer is visibel */
@@ -382,6 +404,13 @@ static void notrace irqsoff_tracer_close(struct trace_iterator *iter)
 		start_irqsoff_tracer(iter->tr);
 }
 
+#ifdef CONFIG_IRQSOFF_TRACER
+static void irqsoff_tracer_init(struct trace_array *tr)
+{
+	trace_type = TRACER_IRQS_OFF;
+
+	__irqsoff_tracer_init(tr);
+}
 static struct tracer irqsoff_tracer __read_mostly =
 {
 	.name		= "irqsoff",
@@ -392,10 +421,65 @@ static struct tracer irqsoff_tracer __read_mostly =
 	.ctrl_update	= irqsoff_tracer_ctrl_update,
 	.print_max	= 1,
 };
+# define register_irqsoff(trace) register_tracer(&trace)
+#else
+# define register_irqsoff(trace) do { } while (0)
+#endif
+
+#ifdef CONFIG_PREEMPT_TRACER
+static void preemptoff_tracer_init(struct trace_array *tr)
+{
+	trace_type = TRACER_PREEMPT_OFF;
+
+	__irqsoff_tracer_init(tr);
+}
+
+static struct tracer preemptoff_tracer __read_mostly =
+{
+	.name		= "preemptoff",
+	.init		= preemptoff_tracer_init,
+	.reset		= irqsoff_tracer_reset,
+	.open		= irqsoff_tracer_open,
+	.close		= irqsoff_tracer_close,
+	.ctrl_update	= irqsoff_tracer_ctrl_update,
+	.print_max	= 1,
+};
+# define register_preemptoff(trace) register_tracer(&trace)
+#else
+# define register_preemptoff(trace) do { } while (0)
+#endif
+
+#if defined(CONFIG_IRQSOFF_TRACER) && \
+	defined(CONFIG_PREEMPT_TRACER)
+
+static void preemptirqsoff_tracer_init(struct trace_array *tr)
+{
+	trace_type = TRACER_IRQS_OFF | TRACER_PREEMPT_OFF;
+
+	__irqsoff_tracer_init(tr);
+}
+
+static struct tracer preemptirqsoff_tracer __read_mostly =
+{
+	.name		= "preemptirqsoff",
+	.init		= preemptirqsoff_tracer_init,
+	.reset		= irqsoff_tracer_reset,
+	.open		= irqsoff_tracer_open,
+	.close		= irqsoff_tracer_close,
+	.ctrl_update	= irqsoff_tracer_ctrl_update,
+	.print_max	= 1,
+};
+
+# define register_preemptirqsoff(trace) register_tracer(&trace)
+#else
+# define register_preemptirqsoff(trace) do { } while (0)
+#endif
 
 __init static int init_irqsoff_tracer(void)
 {
-	register_tracer(&irqsoff_tracer);
+	register_irqsoff(irqsoff_tracer);
+	register_preemptoff(preemptoff_tracer);
+	register_preemptirqsoff(preemptirqsoff_tracer);
 
 	return 0;
 }
-- 
cgit v1.2.3


From 3d0833953e1b98b79ddf491dd49229eef9baeac1 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:42 +0200
Subject: ftrace: dynamic enabling/disabling of function calls

This patch adds a feature to dynamically replace the ftrace code
with the jmps to allow a kernel with ftrace configured to run
as fast as it can without it configured.

The way this works, is on bootup (if ftrace is enabled), a ftrace
function is registered to record the instruction pointer of all
places that call the function.

Later, if there's still any code to patch, a kthread is awoken
(rate limited to at most once a second) that performs a stop_machine,
and replaces all the code that was called with a jmp over the call
to ftrace. It only replaces what was found the previous time. Typically
the system reaches equilibrium quickly after bootup and there's no code
patching needed at all.

e.g.

  call ftrace  /* 5 bytes */

is replaced with

  jmp 3f  /* jmp is 2 bytes and we jump 3 forward */
3:

When we want to enable ftrace for function tracing, the IP recording
is removed, and stop_machine is called again to replace all the locations
of that were recorded back to the call of ftrace.  When it is disabled,
we replace the code back to the jmp.

Allocation is done by the kthread. If the ftrace recording function is
called, and we don't have any record slots available, then we simply
skip that call. Once a second a new page (if needed) is allocated for
recording new ftrace function calls.  A large batch is allocated at
boot up to get most of the calls there.

Because we do this via stop_machine, we don't have to worry about another
CPU executing a ftrace call as we modify it. But we do need to worry
about NMI's so all functions that might be called via nmi must be
annotated with notrace_nmi. When this code is configured in, the NMI code
will not call notrace.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/Makefile |   1 +
 arch/x86/kernel/ftrace.c | 237 +++++++++++++++++++++++++++++++
 include/linux/ftrace.h   |  18 +++
 kernel/trace/Kconfig     |  17 +++
 kernel/trace/ftrace.c    | 356 ++++++++++++++++++++++++++++++++++++++++++-----
 5 files changed, 597 insertions(+), 32 deletions(-)
 create mode 100644 arch/x86/kernel/ftrace.c

(limited to 'include/linux')

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 5e618c3b4720..e142091524b0 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -56,6 +56,7 @@ obj-$(CONFIG_X86_MPPARSE)	+= mpparse.o
 obj-$(CONFIG_X86_LOCAL_APIC)	+= apic_$(BITS).o nmi_$(BITS).o
 obj-$(CONFIG_X86_IO_APIC)	+= io_apic_$(BITS).o
 obj-$(CONFIG_X86_REBOOTFIXUPS)	+= reboot_fixups_32.o
+obj-$(CONFIG_DYNAMIC_FTRACE)	+= ftrace.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec_$(BITS).o
 obj-$(CONFIG_KEXEC)		+= relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_$(BITS).o
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
new file mode 100644
index 000000000000..5dd58136ef02
--- /dev/null
+++ b/arch/x86/kernel/ftrace.c
@@ -0,0 +1,237 @@
+/*
+ * Code for replacing ftrace calls with jumps.
+ *
+ * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
+ *
+ * Thanks goes to Ingo Molnar, for suggesting the idea.
+ * Mathieu Desnoyers, for suggesting postponing the modifications.
+ * Arjan van de Ven, for keeping me straight, and explaining to me
+ * the dangers of modifying code on the run.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/hardirq.h>
+#include <linux/ftrace.h>
+#include <linux/percpu.h>
+#include <linux/init.h>
+#include <linux/list.h>
+
+#define CALL_BACK		5
+
+#define JMPFWD			0x03eb
+
+static unsigned short ftrace_jmp = JMPFWD;
+
+struct ftrace_record {
+	struct dyn_ftrace	rec;
+	int			failed;
+} __attribute__((packed));
+
+struct ftrace_page {
+	struct ftrace_page	*next;
+	int			index;
+	struct ftrace_record	records[];
+} __attribute__((packed));
+
+#define ENTRIES_PER_PAGE \
+  ((PAGE_SIZE - sizeof(struct ftrace_page)) / sizeof(struct ftrace_record))
+
+/* estimate from running different kernels */
+#define NR_TO_INIT		10000
+
+#define MCOUNT_ADDR ((long)(&mcount))
+
+union ftrace_code_union {
+	char code[5];
+	struct {
+		char e8;
+		int offset;
+	} __attribute__((packed));
+};
+
+static struct ftrace_page	*ftrace_pages_start;
+static struct ftrace_page	*ftrace_pages;
+
+notrace struct dyn_ftrace *ftrace_alloc_shutdown_node(unsigned long ip)
+{
+	struct ftrace_record *rec;
+	unsigned short save;
+
+	ip -= CALL_BACK;
+	save = *(short *)ip;
+
+	/* If this was already converted, skip it */
+	if (save == JMPFWD)
+		return NULL;
+
+	if (ftrace_pages->index == ENTRIES_PER_PAGE) {
+		if (!ftrace_pages->next)
+			return NULL;
+		ftrace_pages = ftrace_pages->next;
+	}
+
+	rec = &ftrace_pages->records[ftrace_pages->index++];
+
+	return &rec->rec;
+}
+
+static int notrace
+ftrace_modify_code(unsigned long ip, unsigned char *old_code,
+		   unsigned char *new_code)
+{
+	unsigned short old = *(unsigned short *)old_code;
+	unsigned short new = *(unsigned short *)new_code;
+	unsigned short replaced;
+	int faulted = 0;
+
+	/*
+	 * Note: Due to modules and __init, code can
+	 *  disappear and change, we need to protect against faulting
+	 *  as well as code changing.
+	 *
+	 * No real locking needed, this code is run through
+	 * kstop_machine.
+	 */
+	asm volatile (
+		"1: lock\n"
+		"   cmpxchg %w3, (%2)\n"
+		"2:\n"
+		".section .fixup, \"ax\"\n"
+		"	movl $1, %0\n"
+		"3:	jmp 2b\n"
+		".previous\n"
+		_ASM_EXTABLE(1b, 3b)
+		: "=r"(faulted), "=a"(replaced)
+		: "r"(ip), "r"(new), "0"(faulted), "a"(old)
+		: "memory");
+	sync_core();
+
+	if (replaced != old)
+		faulted = 2;
+
+	return faulted;
+}
+
+static int notrace ftrace_calc_offset(long ip)
+{
+	return (int)(MCOUNT_ADDR - ip);
+}
+
+notrace void ftrace_code_disable(struct dyn_ftrace *rec)
+{
+	unsigned long ip;
+	union ftrace_code_union save;
+	struct ftrace_record *r =
+		container_of(rec, struct ftrace_record, rec);
+
+	ip = rec->ip;
+
+	save.e8		= 0xe8;
+	save.offset 	= ftrace_calc_offset(ip);
+
+	/* move the IP back to the start of the call */
+	ip -= CALL_BACK;
+
+	r->failed = ftrace_modify_code(ip, save.code, (char *)&ftrace_jmp);
+}
+
+static void notrace ftrace_replace_code(int saved)
+{
+	unsigned char *new = NULL, *old = NULL;
+	struct ftrace_record *rec;
+	struct ftrace_page *pg;
+	unsigned long ip;
+	int i;
+
+	if (saved)
+		old = (char *)&ftrace_jmp;
+	else
+		new = (char *)&ftrace_jmp;
+
+	for (pg = ftrace_pages_start; pg; pg = pg->next) {
+		for (i = 0; i < pg->index; i++) {
+			union ftrace_code_union calc;
+			rec = &pg->records[i];
+
+			/* don't modify code that has already faulted */
+			if (rec->failed)
+				continue;
+
+			ip = rec->rec.ip;
+
+			calc.e8		= 0xe8;
+			calc.offset	= ftrace_calc_offset(ip);
+
+			if (saved)
+				new = calc.code;
+			else
+				old = calc.code;
+
+			ip -= CALL_BACK;
+
+			rec->failed = ftrace_modify_code(ip, old, new);
+		}
+	}
+
+}
+
+notrace void ftrace_startup_code(void)
+{
+	ftrace_replace_code(1);
+}
+
+notrace void ftrace_shutdown_code(void)
+{
+	ftrace_replace_code(0);
+}
+
+notrace void ftrace_shutdown_replenish(void)
+{
+	if (ftrace_pages->next)
+		return;
+
+	/* allocate another page */
+	ftrace_pages->next = (void *)get_zeroed_page(GFP_KERNEL);
+}
+
+notrace int ftrace_shutdown_arch_init(void)
+{
+	struct ftrace_page *pg;
+	int cnt;
+	int i;
+
+	/* allocate a few pages */
+	ftrace_pages_start = (void *)get_zeroed_page(GFP_KERNEL);
+	if (!ftrace_pages_start)
+		return -1;
+
+	/*
+	 * Allocate a few more pages.
+	 *
+	 * TODO: have some parser search vmlinux before
+	 *   final linking to find all calls to ftrace.
+	 *   Then we can:
+	 *    a) know how many pages to allocate.
+	 *     and/or
+	 *    b) set up the table then.
+	 *
+	 *  The dynamic code is still necessary for
+	 *  modules.
+	 */
+
+	pg = ftrace_pages = ftrace_pages_start;
+
+	cnt = NR_TO_INIT / ENTRIES_PER_PAGE;
+
+	for (i = 0; i < cnt; i++) {
+		pg->next = (void *)get_zeroed_page(GFP_KERNEL);
+
+		/* If we fail, we'll try later anyway */
+		if (!pg->next)
+			break;
+
+		pg = pg->next;
+	}
+
+	return 0;
+}
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 740c97dcf9cb..90dbc0ee2046 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -32,6 +32,24 @@ extern void mcount(void);
 # define clear_ftrace_function(ops) do { } while (0)
 #endif /* CONFIG_FTRACE */
 
+#ifdef CONFIG_DYNAMIC_FTRACE
+# define FTRACE_HASHBITS	10
+# define FTRACE_HASHSIZE	(1<<FTRACE_HASHBITS)
+
+struct dyn_ftrace {
+	struct hlist_node node;
+	unsigned long	  ip;
+};
+
+/* defined in arch */
+extern struct dyn_ftrace *
+ftrace_alloc_shutdown_node(unsigned long ip);
+extern int ftrace_shutdown_arch_init(void);
+extern void ftrace_code_disable(struct dyn_ftrace *rec);
+extern void ftrace_startup_code(void);
+extern void ftrace_shutdown_code(void);
+extern void ftrace_shutdown_replenish(void);
+#endif
 
 #ifdef CONFIG_FRAME_POINTER
 /* TODO: need to fix this for ARM */
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 6430016b98e8..cad9db1dee02 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -88,3 +88,20 @@ config CONTEXT_SWITCH_TRACER
 	  This tracer gets called from the context switch and records
 	  all switching of tasks.
 
+config DYNAMIC_FTRACE
+	bool "enable/disable ftrace tracepoints dynamically"
+	depends on FTRACE
+	default y
+	help
+         This option will modify all the calls to ftrace dynamically
+	 (will patch them out of the binary image and replaces them
+	 with a No-Op instruction) as they are called. A table is
+	 created to dynamically enable them again.
+
+	 This way a CONFIG_FTRACE kernel is slightly larger, but otherwise
+	 has native performance as long as no tracing is active.
+
+	 The changes to the code are done by a kernel thread that
+	 wakes up once a second and checks to see if any ftrace calls
+	 were made. If so, it runs stop_machine (stops all CPUS)
+	 and modifies the code to jump over the call to ftrace.
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index b6a80b98a3fb..d1ae2ba25274 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -13,10 +13,19 @@
  *  Copyright (C) 2004 William Lee Irwin III
  */
 
-#include <linux/module.h>
+#include <linux/stop_machine.h>
+#include <linux/clocksource.h>
+#include <linux/kallsyms.h>
+#include <linux/kthread.h>
+#include <linux/hardirq.h>
 #include <linux/ftrace.h>
+#include <linux/module.h>
+#include <linux/hash.h>
+#include <linux/list.h>
+
+#include "trace.h"
 
-static DEFINE_SPINLOCK(ftrace_func_lock);
+static DEFINE_SPINLOCK(ftrace_lock);
 static struct ftrace_ops ftrace_list_end __read_mostly =
 {
 	.func = ftrace_stub,
@@ -44,21 +53,21 @@ notrace void ftrace_list_func(unsigned long ip, unsigned long parent_ip)
 }
 
 /**
- * register_ftrace_function - register a function for profiling
- * @ops - ops structure that holds the function for profiling.
- *
- * Register a function to be called by all functions in the
- * kernel.
+ * clear_ftrace_function - reset the ftrace function
  *
- * Note: @ops->func and all the functions it calls must be labeled
- *       with "notrace", otherwise it will go into a
- *       recursive loop.
+ * This NULLs the ftrace function and in essence stops
+ * tracing.  There may be lag
  */
-int register_ftrace_function(struct ftrace_ops *ops)
+void clear_ftrace_function(void)
 {
-	unsigned long flags;
+	ftrace_trace_function = ftrace_stub;
+}
+
+static int notrace __register_ftrace_function(struct ftrace_ops *ops)
+{
+	/* Should never be called by interrupts */
+	spin_lock(&ftrace_lock);
 
-	spin_lock_irqsave(&ftrace_func_lock, flags);
 	ops->next = ftrace_list;
 	/*
 	 * We are entering ops into the ftrace_list but another
@@ -68,6 +77,7 @@ int register_ftrace_function(struct ftrace_ops *ops)
 	 */
 	smp_wmb();
 	ftrace_list = ops;
+
 	/*
 	 * For one func, simply call it directly.
 	 * For more than one func, call the chain.
@@ -76,28 +86,22 @@ int register_ftrace_function(struct ftrace_ops *ops)
 		ftrace_trace_function = ops->func;
 	else
 		ftrace_trace_function = ftrace_list_func;
-	spin_unlock_irqrestore(&ftrace_func_lock, flags);
+
+	spin_unlock(&ftrace_lock);
 
 	return 0;
 }
 
-/**
- * unregister_ftrace_function - unresgister a function for profiling.
- * @ops - ops structure that holds the function to unregister
- *
- * Unregister a function that was added to be called by ftrace profiling.
- */
-int unregister_ftrace_function(struct ftrace_ops *ops)
+static int notrace __unregister_ftrace_function(struct ftrace_ops *ops)
 {
-	unsigned long flags;
 	struct ftrace_ops **p;
 	int ret = 0;
 
-	spin_lock_irqsave(&ftrace_func_lock, flags);
+	spin_lock(&ftrace_lock);
 
 	/*
-	 * If we are the only function, then the ftrace pointer is
-	 * pointing directly to that function.
+	 * If we are removing the last function, then simply point
+	 * to the ftrace_stub.
 	 */
 	if (ftrace_list == ops && ops->next == &ftrace_list_end) {
 		ftrace_trace_function = ftrace_stub;
@@ -117,22 +121,310 @@ int unregister_ftrace_function(struct ftrace_ops *ops)
 	*p = (*p)->next;
 
 	/* If we only have one func left, then call that directly */
-	if (ftrace_list->next == &ftrace_list_end)
+	if (ftrace_list == &ftrace_list_end ||
+	    ftrace_list->next == &ftrace_list_end)
 		ftrace_trace_function = ftrace_list->func;
 
  out:
-	spin_unlock_irqrestore(&ftrace_func_lock, flags);
+	spin_unlock(&ftrace_lock);
+
+	return ret;
+}
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+
+static struct hlist_head ftrace_hash[FTRACE_HASHSIZE];
+
+static DEFINE_PER_CPU(int, ftrace_shutdown_disable_cpu);
+
+static DEFINE_SPINLOCK(ftrace_shutdown_lock);
+static DEFINE_MUTEX(ftraced_lock);
+
+static int ftraced_trigger;
+static int ftraced_suspend;
+
+static int ftrace_record_suspend;
+
+static inline int
+notrace ftrace_ip_in_hash(unsigned long ip, unsigned long key)
+{
+	struct dyn_ftrace *p;
+	struct hlist_node *t;
+	int found = 0;
+
+	hlist_for_each_entry(p, t, &ftrace_hash[key], node) {
+		if (p->ip == ip) {
+			found = 1;
+			break;
+		}
+	}
+
+	return found;
+}
+
+static inline void notrace
+ftrace_add_hash(struct dyn_ftrace *node, unsigned long key)
+{
+	hlist_add_head(&node->node, &ftrace_hash[key]);
+}
+
+static void notrace
+ftrace_record_ip(unsigned long ip, unsigned long parent_ip)
+{
+	struct dyn_ftrace *node;
+	unsigned long flags;
+	unsigned long key;
+	int resched;
+	int atomic;
+
+	resched = need_resched();
+	preempt_disable_notrace();
+
+	/* We simply need to protect against recursion */
+	__get_cpu_var(ftrace_shutdown_disable_cpu)++;
+	if (__get_cpu_var(ftrace_shutdown_disable_cpu) != 1)
+		goto out;
+
+	if (unlikely(ftrace_record_suspend))
+		goto out;
+
+	key = hash_long(ip, FTRACE_HASHBITS);
+
+	WARN_ON_ONCE(key >= FTRACE_HASHSIZE);
+
+	if (ftrace_ip_in_hash(ip, key))
+		goto out;
+
+	atomic = irqs_disabled();
+
+	spin_lock_irqsave(&ftrace_shutdown_lock, flags);
+
+	/* This ip may have hit the hash before the lock */
+	if (ftrace_ip_in_hash(ip, key))
+		goto out_unlock;
+
+	/*
+	 * There's a slight race that the ftraced will update the
+	 * hash and reset here. The arch alloc is responsible
+	 * for seeing if the IP has already changed, and if
+	 * it has, the alloc will fail.
+	 */
+	node = ftrace_alloc_shutdown_node(ip);
+	if (!node)
+		goto out_unlock;
+
+	node->ip = ip;
+
+	ftrace_add_hash(node, key);
+
+	ftraced_trigger = 1;
+
+ out_unlock:
+	spin_unlock_irqrestore(&ftrace_shutdown_lock, flags);
+ out:
+	__get_cpu_var(ftrace_shutdown_disable_cpu)--;
+
+	/* prevent recursion with scheduler */
+	if (resched)
+		preempt_enable_no_resched_notrace();
+	else
+		preempt_enable_notrace();
+}
+
+static struct ftrace_ops ftrace_shutdown_ops __read_mostly =
+{
+	.func = ftrace_record_ip,
+};
+
+
+static int notrace __ftrace_modify_code(void *data)
+{
+	void (*func)(void) = data;
+
+	func();
+	return 0;
+}
+
+static void notrace ftrace_run_startup_code(void)
+{
+	stop_machine_run(__ftrace_modify_code, ftrace_startup_code, NR_CPUS);
+}
+
+static void notrace ftrace_run_shutdown_code(void)
+{
+	stop_machine_run(__ftrace_modify_code, ftrace_shutdown_code, NR_CPUS);
+}
+
+static void notrace ftrace_startup(void)
+{
+	mutex_lock(&ftraced_lock);
+	ftraced_suspend++;
+	if (ftraced_suspend != 1)
+		goto out;
+	__unregister_ftrace_function(&ftrace_shutdown_ops);
+
+	ftrace_run_startup_code();
+ out:
+	mutex_unlock(&ftraced_lock);
+}
+
+static void notrace ftrace_shutdown(void)
+{
+	mutex_lock(&ftraced_lock);
+	ftraced_suspend--;
+	if (ftraced_suspend)
+		goto out;
+
+	ftrace_run_shutdown_code();
+
+	__register_ftrace_function(&ftrace_shutdown_ops);
+ out:
+	mutex_unlock(&ftraced_lock);
+}
+
+static cycle_t		ftrace_update_time;
+static unsigned long	ftrace_update_cnt;
+unsigned long		ftrace_update_tot_cnt;
+
+static int notrace __ftrace_update_code(void *ignore)
+{
+	struct dyn_ftrace *p;
+	struct hlist_head head;
+	struct hlist_node *t;
+	cycle_t start, stop;
+	int i;
+
+	/* Don't be calling ftrace ops now */
+	__unregister_ftrace_function(&ftrace_shutdown_ops);
+
+	start = now(raw_smp_processor_id());
+	ftrace_update_cnt = 0;
+
+	/* No locks needed, the machine is stopped! */
+	for (i = 0; i < FTRACE_HASHSIZE; i++) {
+		if (hlist_empty(&ftrace_hash[i]))
+			continue;
+
+		head = ftrace_hash[i];
+		INIT_HLIST_HEAD(&ftrace_hash[i]);
+
+		/* all CPUS are stopped, we are safe to modify code */
+		hlist_for_each_entry(p, t, &head, node) {
+			ftrace_code_disable(p);
+			ftrace_update_cnt++;
+		}
+
+	}
+
+	stop = now(raw_smp_processor_id());
+	ftrace_update_time = stop - start;
+	ftrace_update_tot_cnt += ftrace_update_cnt;
+
+	__register_ftrace_function(&ftrace_shutdown_ops);
 
 	return 0;
 }
 
+static void notrace ftrace_update_code(void)
+{
+	stop_machine_run(__ftrace_update_code, NULL, NR_CPUS);
+}
+
+static int notrace ftraced(void *ignore)
+{
+	unsigned long usecs;
+
+	set_current_state(TASK_INTERRUPTIBLE);
+
+	while (!kthread_should_stop()) {
+
+		/* check once a second */
+		schedule_timeout(HZ);
+
+		mutex_lock(&ftraced_lock);
+		if (ftraced_trigger && !ftraced_suspend) {
+			ftrace_record_suspend++;
+			ftrace_update_code();
+			usecs = nsecs_to_usecs(ftrace_update_time);
+			if (ftrace_update_tot_cnt > 100000) {
+				ftrace_update_tot_cnt = 0;
+				pr_info("hm, dftrace overflow: %lu change%s"
+					 " (%lu total) in %lu usec%s\n",
+					ftrace_update_cnt,
+					ftrace_update_cnt != 1 ? "s" : "",
+					ftrace_update_tot_cnt,
+					usecs, usecs != 1 ? "s" : "");
+				WARN_ON_ONCE(1);
+			}
+			ftraced_trigger = 0;
+			ftrace_record_suspend--;
+		}
+		mutex_unlock(&ftraced_lock);
+
+		ftrace_shutdown_replenish();
+
+		set_current_state(TASK_INTERRUPTIBLE);
+	}
+	__set_current_state(TASK_RUNNING);
+	return 0;
+}
+
+static int __init notrace ftrace_shutdown_init(void)
+{
+	struct task_struct *p;
+	int ret;
+
+	ret = ftrace_shutdown_arch_init();
+	if (ret)
+		return ret;
+
+	p = kthread_run(ftraced, NULL, "ftraced");
+	if (IS_ERR(p))
+		return -1;
+
+	__register_ftrace_function(&ftrace_shutdown_ops);
+
+	return 0;
+}
+
+core_initcall(ftrace_shutdown_init);
+#else
+# define ftrace_startup()	do { } while (0)
+# define ftrace_shutdown()	do { } while (0)
+#endif /* CONFIG_DYNAMIC_FTRACE */
+
 /**
- * clear_ftrace_function - reset the ftrace function
+ * register_ftrace_function - register a function for profiling
+ * @ops - ops structure that holds the function for profiling.
  *
- * This NULLs the ftrace function and in essence stops
- * tracing.  There may be lag
+ * Register a function to be called by all functions in the
+ * kernel.
+ *
+ * Note: @ops->func and all the functions it calls must be labeled
+ *       with "notrace", otherwise it will go into a
+ *       recursive loop.
  */
-void clear_ftrace_function(void)
+int register_ftrace_function(struct ftrace_ops *ops)
 {
-	ftrace_trace_function = ftrace_stub;
+	ftrace_startup();
+
+	return __register_ftrace_function(ops);
+}
+
+/**
+ * unregister_ftrace_function - unresgister a function for profiling.
+ * @ops - ops structure that holds the function to unregister
+ *
+ * Unregister a function that was added to be called by ftrace profiling.
+ */
+int unregister_ftrace_function(struct ftrace_ops *ops)
+{
+	int ret;
+
+	ret = __unregister_ftrace_function(ops);
+
+	if (ftrace_list == &ftrace_list_end)
+		ftrace_shutdown();
+
+	return ret;
 }
-- 
cgit v1.2.3


From b0fc494fae96a7089f3651cb451f461c7291244c Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:43 +0200
Subject: ftrace: add ftrace_enabled sysctl to disable mcount function

This patch adds back the sysctl ftrace_enabled. This time it is
defaulted to on, if DYNAMIC_FTRACE is configured. When ftrace_enabled
is disabled, the ftrace function is set to the stub return.

If DYNAMIC_FTRACE is also configured, on ftrace_enabled = 0,
the registered ftrace functions will all be set to jmps, but no more
new calls to ftrace recording (used to find the ftrace calling sites)
will be called.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h |   6 +++
 kernel/sysctl.c        |  11 +++++
 kernel/trace/ftrace.c  | 125 ++++++++++++++++++++++++++++++++++++++++++-------
 3 files changed, 124 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 90dbc0ee2046..ccd8537dbdb7 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -5,6 +5,12 @@
 
 #include <linux/linkage.h>
 
+extern int ftrace_enabled;
+extern int
+ftrace_enable_sysctl(struct ctl_table *table, int write,
+		     struct file *filp, void __user *buffer, size_t *lenp,
+		     loff_t *ppos);
+
 typedef void (*ftrace_func_t)(unsigned long ip, unsigned long parent_ip);
 
 struct ftrace_ops {
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 29116652dca8..efaf7c5500e9 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -46,6 +46,7 @@
 #include <linux/nfs_fs.h>
 #include <linux/acpi.h>
 #include <linux/reboot.h>
+#include <linux/ftrace.h>
 
 #include <asm/uaccess.h>
 #include <asm/processor.h>
@@ -455,6 +456,16 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+#ifdef CONFIG_FTRACE
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "ftrace_enabled",
+		.data		= &ftrace_enabled,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &ftrace_enable_sysctl,
+	},
+#endif
 #ifdef CONFIG_KMOD
 	{
 		.ctl_name	= KERN_MODPROBE,
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index d1ae2ba25274..d3de37299ba4 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -20,12 +20,24 @@
 #include <linux/hardirq.h>
 #include <linux/ftrace.h>
 #include <linux/module.h>
+#include <linux/sysctl.h>
 #include <linux/hash.h>
 #include <linux/list.h>
 
 #include "trace.h"
 
+#ifdef CONFIG_DYNAMIC_FTRACE
+# define FTRACE_ENABLED_INIT 1
+#else
+# define FTRACE_ENABLED_INIT 0
+#endif
+
+int ftrace_enabled = FTRACE_ENABLED_INIT;
+static int last_ftrace_enabled = FTRACE_ENABLED_INIT;
+
 static DEFINE_SPINLOCK(ftrace_lock);
+static DEFINE_MUTEX(ftrace_sysctl_lock);
+
 static struct ftrace_ops ftrace_list_end __read_mostly =
 {
 	.func = ftrace_stub,
@@ -78,14 +90,16 @@ static int notrace __register_ftrace_function(struct ftrace_ops *ops)
 	smp_wmb();
 	ftrace_list = ops;
 
-	/*
-	 * For one func, simply call it directly.
-	 * For more than one func, call the chain.
-	 */
-	if (ops->next == &ftrace_list_end)
-		ftrace_trace_function = ops->func;
-	else
-		ftrace_trace_function = ftrace_list_func;
+	if (ftrace_enabled) {
+		/*
+		 * For one func, simply call it directly.
+		 * For more than one func, call the chain.
+		 */
+		if (ops->next == &ftrace_list_end)
+			ftrace_trace_function = ops->func;
+		else
+			ftrace_trace_function = ftrace_list_func;
+	}
 
 	spin_unlock(&ftrace_lock);
 
@@ -120,10 +134,12 @@ static int notrace __unregister_ftrace_function(struct ftrace_ops *ops)
 
 	*p = (*p)->next;
 
-	/* If we only have one func left, then call that directly */
-	if (ftrace_list == &ftrace_list_end ||
-	    ftrace_list->next == &ftrace_list_end)
-		ftrace_trace_function = ftrace_list->func;
+	if (ftrace_enabled) {
+		/* If we only have one func left, then call that directly */
+		if (ftrace_list == &ftrace_list_end ||
+		    ftrace_list->next == &ftrace_list_end)
+			ftrace_trace_function = ftrace_list->func;
+	}
 
  out:
 	spin_unlock(&ftrace_lock);
@@ -263,7 +279,8 @@ static void notrace ftrace_startup(void)
 		goto out;
 	__unregister_ftrace_function(&ftrace_shutdown_ops);
 
-	ftrace_run_startup_code();
+	if (ftrace_enabled)
+		ftrace_run_startup_code();
  out:
 	mutex_unlock(&ftraced_lock);
 }
@@ -275,13 +292,32 @@ static void notrace ftrace_shutdown(void)
 	if (ftraced_suspend)
 		goto out;
 
-	ftrace_run_shutdown_code();
+	if (ftrace_enabled)
+		ftrace_run_shutdown_code();
 
 	__register_ftrace_function(&ftrace_shutdown_ops);
  out:
 	mutex_unlock(&ftraced_lock);
 }
 
+static void notrace ftrace_startup_sysctl(void)
+{
+	mutex_lock(&ftraced_lock);
+	/* ftraced_suspend is true if we want ftrace running */
+	if (ftraced_suspend)
+		ftrace_run_startup_code();
+	mutex_unlock(&ftraced_lock);
+}
+
+static void notrace ftrace_shutdown_sysctl(void)
+{
+	mutex_lock(&ftraced_lock);
+	/* ftraced_suspend is true if ftrace is running */
+	if (ftraced_suspend)
+		ftrace_run_shutdown_code();
+	mutex_unlock(&ftraced_lock);
+}
+
 static cycle_t		ftrace_update_time;
 static unsigned long	ftrace_update_cnt;
 unsigned long		ftrace_update_tot_cnt;
@@ -341,8 +377,9 @@ static int notrace ftraced(void *ignore)
 		/* check once a second */
 		schedule_timeout(HZ);
 
+		mutex_lock(&ftrace_sysctl_lock);
 		mutex_lock(&ftraced_lock);
-		if (ftraced_trigger && !ftraced_suspend) {
+		if (ftrace_enabled && ftraced_trigger && !ftraced_suspend) {
 			ftrace_record_suspend++;
 			ftrace_update_code();
 			usecs = nsecs_to_usecs(ftrace_update_time);
@@ -360,6 +397,7 @@ static int notrace ftraced(void *ignore)
 			ftrace_record_suspend--;
 		}
 		mutex_unlock(&ftraced_lock);
+		mutex_unlock(&ftrace_sysctl_lock);
 
 		ftrace_shutdown_replenish();
 
@@ -389,8 +427,10 @@ static int __init notrace ftrace_shutdown_init(void)
 
 core_initcall(ftrace_shutdown_init);
 #else
-# define ftrace_startup()	do { } while (0)
-# define ftrace_shutdown()	do { } while (0)
+# define ftrace_startup()	  do { } while (0)
+# define ftrace_shutdown()	  do { } while (0)
+# define ftrace_startup_sysctl()  do { } while (0)
+# define ftrace_shutdown_sysctl() do { } while (0)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 /**
@@ -406,9 +446,15 @@ core_initcall(ftrace_shutdown_init);
  */
 int register_ftrace_function(struct ftrace_ops *ops)
 {
+	int ret;
+
+	mutex_lock(&ftrace_sysctl_lock);
 	ftrace_startup();
 
-	return __register_ftrace_function(ops);
+	ret = __register_ftrace_function(ops);
+	mutex_unlock(&ftrace_sysctl_lock);
+
+	return ret;
 }
 
 /**
@@ -421,10 +467,53 @@ int unregister_ftrace_function(struct ftrace_ops *ops)
 {
 	int ret;
 
+	mutex_lock(&ftrace_sysctl_lock);
 	ret = __unregister_ftrace_function(ops);
 
 	if (ftrace_list == &ftrace_list_end)
 		ftrace_shutdown();
 
+	mutex_unlock(&ftrace_sysctl_lock);
+
+	return ret;
+}
+
+notrace int
+ftrace_enable_sysctl(struct ctl_table *table, int write,
+		     struct file *filp, void __user *buffer, size_t *lenp,
+		     loff_t *ppos)
+{
+	int ret;
+
+	mutex_lock(&ftrace_sysctl_lock);
+
+	ret  = proc_dointvec(table, write, filp, buffer, lenp, ppos);
+
+	if (ret || !write || (last_ftrace_enabled == ftrace_enabled))
+		goto out;
+
+	last_ftrace_enabled = ftrace_enabled;
+
+	if (ftrace_enabled) {
+
+		ftrace_startup_sysctl();
+
+		/* we are starting ftrace again */
+		if (ftrace_list != &ftrace_list_end) {
+			if (ftrace_list->next == &ftrace_list_end)
+				ftrace_trace_function = ftrace_list->func;
+			else
+				ftrace_trace_function = ftrace_list_func;
+		}
+
+	} else {
+		/* stopping ftrace calls (just send to ftrace_stub) */
+		ftrace_trace_function = ftrace_stub;
+
+		ftrace_shutdown_sysctl();
+	}
+
+ out:
+	mutex_unlock(&ftrace_sysctl_lock);
 	return ret;
 }
-- 
cgit v1.2.3


From 3c1720f00bb619302ba19d55986ab565e74d06db Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:43 +0200
Subject: ftrace: move memory management out of arch code

This patch moves the memory management of the ftrace
records out of the arch code and into the generic code
making the arch code simpler.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/ftrace.c | 183 ++++++++---------------------------------------
 include/linux/ftrace.h   |  18 +++--
 kernel/trace/ftrace.c    | 154 ++++++++++++++++++++++++++++++++++++++-
 3 files changed, 192 insertions(+), 163 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 2e060c58b860..b69795efa226 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -23,25 +23,6 @@
 /* Long is fine, even if it is only 4 bytes ;-) */
 static long *ftrace_nop;
 
-struct ftrace_record {
-	struct dyn_ftrace	rec;
-	int			failed;
-} __attribute__((packed));
-
-struct ftrace_page {
-	struct ftrace_page	*next;
-	int			index;
-	struct ftrace_record	records[];
-} __attribute__((packed));
-
-#define ENTRIES_PER_PAGE \
-  ((PAGE_SIZE - sizeof(struct ftrace_page)) / sizeof(struct ftrace_record))
-
-/* estimate from running different kernels */
-#define NR_TO_INIT		10000
-
-#define MCOUNT_ADDR ((long)(&mcount))
-
 union ftrace_code_union {
 	char code[5];
 	struct {
@@ -50,33 +31,41 @@ union ftrace_code_union {
 	} __attribute__((packed));
 };
 
-static struct ftrace_page	*ftrace_pages_start;
-static struct ftrace_page	*ftrace_pages;
-
-notrace struct dyn_ftrace *ftrace_alloc_shutdown_node(unsigned long ip)
+notrace int ftrace_ip_converted(unsigned long ip)
 {
-	struct ftrace_record *rec;
 	unsigned long save;
 
 	ip -= CALL_BACK;
 	save = *(long *)ip;
 
-	/* If this was already converted, skip it */
-	if (save == *ftrace_nop)
-		return NULL;
+	return save == *ftrace_nop;
+}
 
-	if (ftrace_pages->index == ENTRIES_PER_PAGE) {
-		if (!ftrace_pages->next)
-			return NULL;
-		ftrace_pages = ftrace_pages->next;
-	}
+static int notrace ftrace_calc_offset(long ip, long addr)
+{
+	return (int)(addr - ip);
+}
 
-	rec = &ftrace_pages->records[ftrace_pages->index++];
+notrace unsigned char *ftrace_nop_replace(void)
+{
+	return (char *)ftrace_nop;
+}
+
+notrace unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr)
+{
+	static union ftrace_code_union calc;
 
-	return &rec->rec;
+	calc.e8		= 0xe8;
+	calc.offset	= ftrace_calc_offset(ip, addr);
+
+	/*
+	 * No locking needed, this must be called via kstop_machine
+	 * which in essence is like running on a uniprocessor machine.
+	 */
+	return calc.code;
 }
 
-static int notrace
+notrace int
 ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 		   unsigned char *new_code)
 {
@@ -86,6 +75,9 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 	unsigned char newch = new_code[4];
 	int faulted = 0;
 
+	/* move the IP back to the start of the call */
+	ip -= CALL_BACK;
+
 	/*
 	 * Note: Due to modules and __init, code can
 	 *  disappear and change, we need to protect against faulting
@@ -117,129 +109,12 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 	return faulted;
 }
 
-static int notrace ftrace_calc_offset(long ip)
-{
-	return (int)(MCOUNT_ADDR - ip);
-}
-
-notrace void ftrace_code_disable(struct dyn_ftrace *rec)
-{
-	unsigned long ip;
-	union ftrace_code_union save;
-	struct ftrace_record *r =
-		container_of(rec, struct ftrace_record, rec);
-
-	ip = rec->ip;
-
-	save.e8		= 0xe8;
-	save.offset 	= ftrace_calc_offset(ip);
-
-	/* move the IP back to the start of the call */
-	ip -= CALL_BACK;
-
-	r->failed = ftrace_modify_code(ip, save.code, (char *)ftrace_nop);
-}
-
-static void notrace ftrace_replace_code(int saved)
-{
-	unsigned char *new = NULL, *old = NULL;
-	struct ftrace_record *rec;
-	struct ftrace_page *pg;
-	unsigned long ip;
-	int i;
-
-	if (saved)
-		old = (char *)ftrace_nop;
-	else
-		new = (char *)ftrace_nop;
-
-	for (pg = ftrace_pages_start; pg; pg = pg->next) {
-		for (i = 0; i < pg->index; i++) {
-			union ftrace_code_union calc;
-			rec = &pg->records[i];
-
-			/* don't modify code that has already faulted */
-			if (rec->failed)
-				continue;
-
-			ip = rec->rec.ip;
-
-			calc.e8		= 0xe8;
-			calc.offset	= ftrace_calc_offset(ip);
-
-			if (saved)
-				new = calc.code;
-			else
-				old = calc.code;
-
-			ip -= CALL_BACK;
-
-			rec->failed = ftrace_modify_code(ip, old, new);
-		}
-	}
-
-}
-
-notrace void ftrace_startup_code(void)
-{
-	ftrace_replace_code(1);
-}
-
-notrace void ftrace_shutdown_code(void)
-{
-	ftrace_replace_code(0);
-}
-
-notrace void ftrace_shutdown_replenish(void)
-{
-	if (ftrace_pages->next)
-		return;
-
-	/* allocate another page */
-	ftrace_pages->next = (void *)get_zeroed_page(GFP_KERNEL);
-}
-
-notrace int __init ftrace_shutdown_arch_init(void)
+int __init ftrace_dyn_arch_init(void)
 {
 	const unsigned char *const *noptable = find_nop_table();
-	struct ftrace_page *pg;
-	int cnt;
-	int i;
 
 	ftrace_nop = (unsigned long *)noptable[CALL_BACK];
 
-	/* allocate a few pages */
-	ftrace_pages_start = (void *)get_zeroed_page(GFP_KERNEL);
-	if (!ftrace_pages_start)
-		return -1;
-
-	/*
-	 * Allocate a few more pages.
-	 *
-	 * TODO: have some parser search vmlinux before
-	 *   final linking to find all calls to ftrace.
-	 *   Then we can:
-	 *    a) know how many pages to allocate.
-	 *     and/or
-	 *    b) set up the table then.
-	 *
-	 *  The dynamic code is still necessary for
-	 *  modules.
-	 */
-
-	pg = ftrace_pages = ftrace_pages_start;
-
-	cnt = NR_TO_INIT / ENTRIES_PER_PAGE;
-
-	for (i = 0; i < cnt; i++) {
-		pg->next = (void *)get_zeroed_page(GFP_KERNEL);
-
-		/* If we fail, we'll try later anyway */
-		if (!pg->next)
-			break;
-
-		pg = pg->next;
-	}
-
 	return 0;
 }
+
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index ccd8537dbdb7..d509ad6c9cb8 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -42,19 +42,23 @@ extern void mcount(void);
 # define FTRACE_HASHBITS	10
 # define FTRACE_HASHSIZE	(1<<FTRACE_HASHBITS)
 
+enum {
+	FTRACE_FL_FAILED	= (1<<0),
+};
+
 struct dyn_ftrace {
 	struct hlist_node node;
 	unsigned long	  ip;
+	unsigned long	  flags;
 };
 
 /* defined in arch */
-extern struct dyn_ftrace *
-ftrace_alloc_shutdown_node(unsigned long ip);
-extern int ftrace_shutdown_arch_init(void);
-extern void ftrace_code_disable(struct dyn_ftrace *rec);
-extern void ftrace_startup_code(void);
-extern void ftrace_shutdown_code(void);
-extern void ftrace_shutdown_replenish(void);
+extern int ftrace_ip_converted(unsigned long ip);
+extern unsigned char *ftrace_nop_replace(void);
+extern unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr);
+extern int ftrace_dyn_arch_init(void);
+extern int ftrace_modify_code(unsigned long ip, unsigned char *old_code,
+			      unsigned char *new_code);
 #endif
 
 #ifdef CONFIG_FRAME_POINTER
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index d3de37299ba4..f6d9af3bf66b 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -156,6 +156,21 @@ static DEFINE_PER_CPU(int, ftrace_shutdown_disable_cpu);
 static DEFINE_SPINLOCK(ftrace_shutdown_lock);
 static DEFINE_MUTEX(ftraced_lock);
 
+struct ftrace_page {
+	struct ftrace_page	*next;
+	int			index;
+	struct dyn_ftrace	records[];
+} __attribute__((packed));
+
+#define ENTRIES_PER_PAGE \
+  ((PAGE_SIZE - sizeof(struct ftrace_page)) / sizeof(struct dyn_ftrace))
+
+/* estimate from running different kernels */
+#define NR_TO_INIT		10000
+
+static struct ftrace_page	*ftrace_pages_start;
+static struct ftrace_page	*ftrace_pages;
+
 static int ftraced_trigger;
 static int ftraced_suspend;
 
@@ -184,6 +199,21 @@ ftrace_add_hash(struct dyn_ftrace *node, unsigned long key)
 	hlist_add_head(&node->node, &ftrace_hash[key]);
 }
 
+static notrace struct dyn_ftrace *ftrace_alloc_shutdown_node(unsigned long ip)
+{
+	/* If this was already converted, skip it */
+	if (ftrace_ip_converted(ip))
+		return NULL;
+
+	if (ftrace_pages->index == ENTRIES_PER_PAGE) {
+		if (!ftrace_pages->next)
+			return NULL;
+		ftrace_pages = ftrace_pages->next;
+	}
+
+	return &ftrace_pages->records[ftrace_pages->index++];
+}
+
 static void notrace
 ftrace_record_ip(unsigned long ip, unsigned long parent_ip)
 {
@@ -252,6 +282,62 @@ static struct ftrace_ops ftrace_shutdown_ops __read_mostly =
 	.func = ftrace_record_ip,
 };
 
+#define MCOUNT_ADDR ((long)(&mcount))
+
+static void notrace ftrace_replace_code(int saved)
+{
+	unsigned char *new = NULL, *old = NULL;
+	struct dyn_ftrace *rec;
+	struct ftrace_page *pg;
+	unsigned long ip;
+	int failed;
+	int i;
+
+	if (saved)
+		old = ftrace_nop_replace();
+	else
+		new = ftrace_nop_replace();
+
+	for (pg = ftrace_pages_start; pg; pg = pg->next) {
+		for (i = 0; i < pg->index; i++) {
+			rec = &pg->records[i];
+
+			/* don't modify code that has already faulted */
+			if (rec->flags & FTRACE_FL_FAILED)
+				continue;
+
+			ip = rec->ip;
+
+			if (saved)
+				new = ftrace_call_replace(ip, MCOUNT_ADDR);
+			else
+				old = ftrace_call_replace(ip, MCOUNT_ADDR);
+
+			failed = ftrace_modify_code(ip, old, new);
+			if (failed)
+				rec->flags |= FTRACE_FL_FAILED;
+		}
+	}
+}
+
+static notrace void ftrace_startup_code(void)
+{
+	ftrace_replace_code(1);
+}
+
+static notrace void ftrace_shutdown_code(void)
+{
+	ftrace_replace_code(0);
+}
+
+static notrace void ftrace_shutdown_replenish(void)
+{
+	if (ftrace_pages->next)
+		return;
+
+	/* allocate another page */
+	ftrace_pages->next = (void *)get_zeroed_page(GFP_KERNEL);
+}
 
 static int notrace __ftrace_modify_code(void *data)
 {
@@ -261,6 +347,23 @@ static int notrace __ftrace_modify_code(void *data)
 	return 0;
 }
 
+static notrace void
+ftrace_code_disable(struct dyn_ftrace *rec, unsigned long addr)
+{
+	unsigned long ip;
+	unsigned char *nop, *call;
+	int failed;
+
+	ip = rec->ip;
+
+	nop = ftrace_nop_replace();
+	call = ftrace_call_replace(ip, addr);
+
+	failed = ftrace_modify_code(ip, call, nop);
+	if (failed)
+		rec->flags |= FTRACE_FL_FAILED;
+}
+
 static void notrace ftrace_run_startup_code(void)
 {
 	stop_machine_run(__ftrace_modify_code, ftrace_startup_code, NR_CPUS);
@@ -346,7 +449,7 @@ static int notrace __ftrace_update_code(void *ignore)
 
 		/* all CPUS are stopped, we are safe to modify code */
 		hlist_for_each_entry(p, t, &head, node) {
-			ftrace_code_disable(p);
+			ftrace_code_disable(p, MCOUNT_ADDR);
 			ftrace_update_cnt++;
 		}
 
@@ -407,12 +510,59 @@ static int notrace ftraced(void *ignore)
 	return 0;
 }
 
+static int __init ftrace_dyn_table_alloc(void)
+{
+	struct ftrace_page *pg;
+	int cnt;
+	int i;
+	int ret;
+
+	ret = ftrace_dyn_arch_init();
+	if (ret)
+		return ret;
+
+	/* allocate a few pages */
+	ftrace_pages_start = (void *)get_zeroed_page(GFP_KERNEL);
+	if (!ftrace_pages_start)
+		return -1;
+
+	/*
+	 * Allocate a few more pages.
+	 *
+	 * TODO: have some parser search vmlinux before
+	 *   final linking to find all calls to ftrace.
+	 *   Then we can:
+	 *    a) know how many pages to allocate.
+	 *     and/or
+	 *    b) set up the table then.
+	 *
+	 *  The dynamic code is still necessary for
+	 *  modules.
+	 */
+
+	pg = ftrace_pages = ftrace_pages_start;
+
+	cnt = NR_TO_INIT / ENTRIES_PER_PAGE;
+
+	for (i = 0; i < cnt; i++) {
+		pg->next = (void *)get_zeroed_page(GFP_KERNEL);
+
+		/* If we fail, we'll try later anyway */
+		if (!pg->next)
+			break;
+
+		pg = pg->next;
+	}
+
+	return 0;
+}
+
 static int __init notrace ftrace_shutdown_init(void)
 {
 	struct task_struct *p;
 	int ret;
 
-	ret = ftrace_shutdown_arch_init();
+	ret = ftrace_dyn_table_alloc();
 	if (ret)
 		return ret;
 
-- 
cgit v1.2.3


From d61f82d06672f57fca410da6f7fffd15867db622 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:43 +0200
Subject: ftrace: use dynamic patching for updating mcount calls

This patch replaces the indirect call to the mcount function
pointer with a direct call that will be patched by the
dynamic ftrace routines.

On boot up, the mcount function calls the ftace_stub function.
When the dynamic ftrace code is initialized, the ftrace_stub
is replaced with a call to the ftrace_record_ip, which records
the instruction pointers of the locations that call it.

Later, the ftraced daemon will call kstop_machine and patch all
the locations to nops.

When a ftrace is enabled, the original calls to mcount will now
be set top call ftrace_caller, which will do a direct call
to the registered ftrace function. This direct call is also patched
when the function that should be called is updated.

All patching is performed by a kstop_machine routine to prevent any
type of race conditions that is associated with modifying code
on the fly.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/entry_32.S |  47 +++++++++++-
 arch/x86/kernel/entry_64.S |  67 ++++++++++++++++-
 arch/x86/kernel/ftrace.c   |  41 +++++++++-
 include/linux/ftrace.h     |   7 +-
 kernel/trace/ftrace.c      | 183 ++++++++++++++++++++++++++-------------------
 5 files changed, 261 insertions(+), 84 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index f47b9b5440d2..e6517ce0b824 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -1110,10 +1110,50 @@ ENDPROC(xen_failsafe_callback)
 #endif	/* CONFIG_XEN */
 
 #ifdef CONFIG_FTRACE
+#ifdef CONFIG_DYNAMIC_FTRACE
+
+ENTRY(mcount)
+	pushl %eax
+	pushl %ecx
+	pushl %edx
+	movl 0xc(%esp), %eax
+
+.globl mcount_call
+mcount_call:
+	call ftrace_stub
+
+	popl %edx
+	popl %ecx
+	popl %eax
+
+	ret
+END(mcount)
+
+ENTRY(ftrace_caller)
+	pushl %eax
+	pushl %ecx
+	pushl %edx
+	movl 0xc(%esp), %eax
+	movl 0x4(%ebp), %edx
+
+.globl ftrace_call
+ftrace_call:
+	call ftrace_stub
+
+	popl %edx
+	popl %ecx
+	popl %eax
+
+.globl ftrace_stub
+ftrace_stub:
+	ret
+END(ftrace_caller)
+
+#else /* ! CONFIG_DYNAMIC_FTRACE */
+
 ENTRY(mcount)
 	cmpl $ftrace_stub, ftrace_trace_function
 	jnz trace
-
 .globl ftrace_stub
 ftrace_stub:
 	ret
@@ -1126,7 +1166,7 @@ trace:
 	movl 0xc(%esp), %eax
 	movl 0x4(%ebp), %edx
 
-	call   *ftrace_trace_function
+	call *ftrace_trace_function
 
 	popl %edx
 	popl %ecx
@@ -1134,7 +1174,8 @@ trace:
 
 	jmp ftrace_stub
 END(mcount)
-#endif
+#endif /* CONFIG_DYNAMIC_FTRACE */
+#endif /* CONFIG_FTRACE */
 
 .section .rodata,"a"
 #include "syscall_table_32.S"
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index f046e0c64883..fe25e5febca3 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -55,6 +55,70 @@
 	.code64
 
 #ifdef CONFIG_FTRACE
+#ifdef CONFIG_DYNAMIC_FTRACE
+ENTRY(mcount)
+
+	subq $0x38, %rsp
+	movq %rax, (%rsp)
+	movq %rcx, 8(%rsp)
+	movq %rdx, 16(%rsp)
+	movq %rsi, 24(%rsp)
+	movq %rdi, 32(%rsp)
+	movq %r8, 40(%rsp)
+	movq %r9, 48(%rsp)
+
+	movq 0x38(%rsp), %rdi
+
+.globl mcount_call
+mcount_call:
+	call ftrace_stub
+
+	movq 48(%rsp), %r9
+	movq 40(%rsp), %r8
+	movq 32(%rsp), %rdi
+	movq 24(%rsp), %rsi
+	movq 16(%rsp), %rdx
+	movq 8(%rsp), %rcx
+	movq (%rsp), %rax
+	addq $0x38, %rsp
+
+	retq
+END(mcount)
+
+ENTRY(ftrace_caller)
+
+	/* taken from glibc */
+	subq $0x38, %rsp
+	movq %rax, (%rsp)
+	movq %rcx, 8(%rsp)
+	movq %rdx, 16(%rsp)
+	movq %rsi, 24(%rsp)
+	movq %rdi, 32(%rsp)
+	movq %r8, 40(%rsp)
+	movq %r9, 48(%rsp)
+
+	movq 0x38(%rsp), %rdi
+	movq 8(%rbp), %rsi
+
+.globl ftrace_call
+ftrace_call:
+	call ftrace_stub
+
+	movq 48(%rsp), %r9
+	movq 40(%rsp), %r8
+	movq 32(%rsp), %rdi
+	movq 24(%rsp), %rsi
+	movq 16(%rsp), %rdx
+	movq 8(%rsp), %rcx
+	movq (%rsp), %rax
+	addq $0x38, %rsp
+
+.globl ftrace_stub
+ftrace_stub:
+	retq
+END(ftrace_caller)
+
+#else /* ! CONFIG_DYNAMIC_FTRACE */
 ENTRY(mcount)
 	cmpq $ftrace_stub, ftrace_trace_function
 	jnz trace
@@ -89,7 +153,8 @@ trace:
 
 	jmp ftrace_stub
 END(mcount)
-#endif
+#endif /* CONFIG_DYNAMIC_FTRACE */
+#endif /* CONFIG_FTRACE */
 
 #ifndef CONFIG_PREEMPT
 #define retint_kernel retint_restore_args
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index b69795efa226..9f44623e0072 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -109,10 +109,49 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 	return faulted;
 }
 
-int __init ftrace_dyn_arch_init(void)
+notrace int ftrace_update_ftrace_func(ftrace_func_t func)
+{
+	unsigned long ip = (unsigned long)(&ftrace_call);
+	unsigned char old[5], *new;
+	int ret;
+
+	ip += CALL_BACK;
+
+	memcpy(old, &ftrace_call, 5);
+	new = ftrace_call_replace(ip, (unsigned long)func);
+	ret = ftrace_modify_code(ip, old, new);
+
+	return ret;
+}
+
+notrace int ftrace_mcount_set(unsigned long *data)
+{
+	unsigned long ip = (long)(&mcount_call);
+	unsigned long *addr = data;
+	unsigned char old[5], *new;
+
+	/* ip is at the location, but modify code will subtact this */
+	ip += CALL_BACK;
+
+	/*
+	 * Replace the mcount stub with a pointer to the
+	 * ip recorder function.
+	 */
+	memcpy(old, &mcount_call, 5);
+	new = ftrace_call_replace(ip, *addr);
+	*addr = ftrace_modify_code(ip, old, new);
+
+	return 0;
+}
+
+int __init ftrace_dyn_arch_init(void *data)
 {
 	const unsigned char *const *noptable = find_nop_table();
 
+	/* This is running in kstop_machine */
+
+	ftrace_mcount_set(data);
+
 	ftrace_nop = (unsigned long *)noptable[CALL_BACK];
 
 	return 0;
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index d509ad6c9cb8..b0dd0093058f 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -56,9 +56,14 @@ struct dyn_ftrace {
 extern int ftrace_ip_converted(unsigned long ip);
 extern unsigned char *ftrace_nop_replace(void);
 extern unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr);
-extern int ftrace_dyn_arch_init(void);
+extern int ftrace_dyn_arch_init(void *data);
+extern int ftrace_mcount_set(unsigned long *data);
 extern int ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 			      unsigned char *new_code);
+extern int ftrace_update_ftrace_func(ftrace_func_t func);
+extern void ftrace_caller(void);
+extern void ftrace_call(void);
+extern void mcount_call(void);
 #endif
 
 #ifdef CONFIG_FRAME_POINTER
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index f6d9af3bf66b..88544f9bc0ed 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -26,14 +26,8 @@
 
 #include "trace.h"
 
-#ifdef CONFIG_DYNAMIC_FTRACE
-# define FTRACE_ENABLED_INIT 1
-#else
-# define FTRACE_ENABLED_INIT 0
-#endif
-
-int ftrace_enabled = FTRACE_ENABLED_INIT;
-static int last_ftrace_enabled = FTRACE_ENABLED_INIT;
+int ftrace_enabled;
+static int last_ftrace_enabled;
 
 static DEFINE_SPINLOCK(ftrace_lock);
 static DEFINE_MUTEX(ftrace_sysctl_lock);
@@ -149,6 +143,14 @@ static int notrace __unregister_ftrace_function(struct ftrace_ops *ops)
 
 #ifdef CONFIG_DYNAMIC_FTRACE
 
+enum {
+	FTRACE_ENABLE_CALLS		= (1 << 0),
+	FTRACE_DISABLE_CALLS		= (1 << 1),
+	FTRACE_UPDATE_TRACE_FUNC	= (1 << 2),
+	FTRACE_ENABLE_MCOUNT		= (1 << 3),
+	FTRACE_DISABLE_MCOUNT		= (1 << 4),
+};
+
 static struct hlist_head ftrace_hash[FTRACE_HASHSIZE];
 
 static DEFINE_PER_CPU(int, ftrace_shutdown_disable_cpu);
@@ -199,12 +201,8 @@ ftrace_add_hash(struct dyn_ftrace *node, unsigned long key)
 	hlist_add_head(&node->node, &ftrace_hash[key]);
 }
 
-static notrace struct dyn_ftrace *ftrace_alloc_shutdown_node(unsigned long ip)
+static notrace struct dyn_ftrace *ftrace_alloc_dyn_node(unsigned long ip)
 {
-	/* If this was already converted, skip it */
-	if (ftrace_ip_converted(ip))
-		return NULL;
-
 	if (ftrace_pages->index == ENTRIES_PER_PAGE) {
 		if (!ftrace_pages->next)
 			return NULL;
@@ -215,7 +213,7 @@ static notrace struct dyn_ftrace *ftrace_alloc_shutdown_node(unsigned long ip)
 }
 
 static void notrace
-ftrace_record_ip(unsigned long ip, unsigned long parent_ip)
+ftrace_record_ip(unsigned long ip)
 {
 	struct dyn_ftrace *node;
 	unsigned long flags;
@@ -223,6 +221,9 @@ ftrace_record_ip(unsigned long ip, unsigned long parent_ip)
 	int resched;
 	int atomic;
 
+	if (!ftrace_enabled)
+		return;
+
 	resched = need_resched();
 	preempt_disable_notrace();
 
@@ -251,11 +252,12 @@ ftrace_record_ip(unsigned long ip, unsigned long parent_ip)
 
 	/*
 	 * There's a slight race that the ftraced will update the
-	 * hash and reset here. The arch alloc is responsible
-	 * for seeing if the IP has already changed, and if
-	 * it has, the alloc will fail.
+	 * hash and reset here. If it is already converted, skip it.
 	 */
-	node = ftrace_alloc_shutdown_node(ip);
+	if (ftrace_ip_converted(ip))
+		goto out_unlock;
+
+	node = ftrace_alloc_dyn_node(ip);
 	if (!node)
 		goto out_unlock;
 
@@ -277,11 +279,7 @@ ftrace_record_ip(unsigned long ip, unsigned long parent_ip)
 		preempt_enable_notrace();
 }
 
-static struct ftrace_ops ftrace_shutdown_ops __read_mostly =
-{
-	.func = ftrace_record_ip,
-};
-
+#define FTRACE_ADDR ((long)(&ftrace_caller))
 #define MCOUNT_ADDR ((long)(&mcount))
 
 static void notrace ftrace_replace_code(int saved)
@@ -309,9 +307,9 @@ static void notrace ftrace_replace_code(int saved)
 			ip = rec->ip;
 
 			if (saved)
-				new = ftrace_call_replace(ip, MCOUNT_ADDR);
+				new = ftrace_call_replace(ip, FTRACE_ADDR);
 			else
-				old = ftrace_call_replace(ip, MCOUNT_ADDR);
+				old = ftrace_call_replace(ip, FTRACE_ADDR);
 
 			failed = ftrace_modify_code(ip, old, new);
 			if (failed)
@@ -320,16 +318,6 @@ static void notrace ftrace_replace_code(int saved)
 	}
 }
 
-static notrace void ftrace_startup_code(void)
-{
-	ftrace_replace_code(1);
-}
-
-static notrace void ftrace_shutdown_code(void)
-{
-	ftrace_replace_code(0);
-}
-
 static notrace void ftrace_shutdown_replenish(void)
 {
 	if (ftrace_pages->next)
@@ -339,16 +327,8 @@ static notrace void ftrace_shutdown_replenish(void)
 	ftrace_pages->next = (void *)get_zeroed_page(GFP_KERNEL);
 }
 
-static int notrace __ftrace_modify_code(void *data)
-{
-	void (*func)(void) = data;
-
-	func();
-	return 0;
-}
-
 static notrace void
-ftrace_code_disable(struct dyn_ftrace *rec, unsigned long addr)
+ftrace_code_disable(struct dyn_ftrace *rec)
 {
 	unsigned long ip;
 	unsigned char *nop, *call;
@@ -357,67 +337,113 @@ ftrace_code_disable(struct dyn_ftrace *rec, unsigned long addr)
 	ip = rec->ip;
 
 	nop = ftrace_nop_replace();
-	call = ftrace_call_replace(ip, addr);
+	call = ftrace_call_replace(ip, MCOUNT_ADDR);
 
 	failed = ftrace_modify_code(ip, call, nop);
 	if (failed)
 		rec->flags |= FTRACE_FL_FAILED;
 }
 
-static void notrace ftrace_run_startup_code(void)
+static int notrace __ftrace_modify_code(void *data)
 {
-	stop_machine_run(__ftrace_modify_code, ftrace_startup_code, NR_CPUS);
+	unsigned long addr;
+	int *command = data;
+
+	if (*command & FTRACE_ENABLE_CALLS)
+		ftrace_replace_code(1);
+	else if (*command & FTRACE_DISABLE_CALLS)
+		ftrace_replace_code(0);
+
+	if (*command & FTRACE_UPDATE_TRACE_FUNC)
+		ftrace_update_ftrace_func(ftrace_trace_function);
+
+	if (*command & FTRACE_ENABLE_MCOUNT) {
+		addr = (unsigned long)ftrace_record_ip;
+		ftrace_mcount_set(&addr);
+	} else if (*command & FTRACE_DISABLE_MCOUNT) {
+		addr = (unsigned long)ftrace_stub;
+		ftrace_mcount_set(&addr);
+	}
+
+	return 0;
 }
 
-static void notrace ftrace_run_shutdown_code(void)
+static void notrace ftrace_run_update_code(int command)
 {
-	stop_machine_run(__ftrace_modify_code, ftrace_shutdown_code, NR_CPUS);
+	stop_machine_run(__ftrace_modify_code, &command, NR_CPUS);
 }
 
+static ftrace_func_t saved_ftrace_func;
+
 static void notrace ftrace_startup(void)
 {
+	int command = 0;
+
 	mutex_lock(&ftraced_lock);
 	ftraced_suspend++;
-	if (ftraced_suspend != 1)
+	if (ftraced_suspend == 1)
+		command |= FTRACE_ENABLE_CALLS;
+
+	if (saved_ftrace_func != ftrace_trace_function) {
+		saved_ftrace_func = ftrace_trace_function;
+		command |= FTRACE_UPDATE_TRACE_FUNC;
+	}
+
+	if (!command || !ftrace_enabled)
 		goto out;
-	__unregister_ftrace_function(&ftrace_shutdown_ops);
 
-	if (ftrace_enabled)
-		ftrace_run_startup_code();
+	ftrace_run_update_code(command);
  out:
 	mutex_unlock(&ftraced_lock);
 }
 
 static void notrace ftrace_shutdown(void)
 {
+	int command = 0;
+
 	mutex_lock(&ftraced_lock);
 	ftraced_suspend--;
-	if (ftraced_suspend)
-		goto out;
+	if (!ftraced_suspend)
+		command |= FTRACE_DISABLE_CALLS;
 
-	if (ftrace_enabled)
-		ftrace_run_shutdown_code();
+	if (saved_ftrace_func != ftrace_trace_function) {
+		saved_ftrace_func = ftrace_trace_function;
+		command |= FTRACE_UPDATE_TRACE_FUNC;
+	}
 
-	__register_ftrace_function(&ftrace_shutdown_ops);
+	if (!command || !ftrace_enabled)
+		goto out;
+
+	ftrace_run_update_code(command);
  out:
 	mutex_unlock(&ftraced_lock);
 }
 
 static void notrace ftrace_startup_sysctl(void)
 {
+	int command = FTRACE_ENABLE_MCOUNT;
+
 	mutex_lock(&ftraced_lock);
+	/* Force update next time */
+	saved_ftrace_func = NULL;
 	/* ftraced_suspend is true if we want ftrace running */
 	if (ftraced_suspend)
-		ftrace_run_startup_code();
+		command |= FTRACE_ENABLE_CALLS;
+
+	ftrace_run_update_code(command);
 	mutex_unlock(&ftraced_lock);
 }
 
 static void notrace ftrace_shutdown_sysctl(void)
 {
+	int command = FTRACE_DISABLE_MCOUNT;
+
 	mutex_lock(&ftraced_lock);
 	/* ftraced_suspend is true if ftrace is running */
 	if (ftraced_suspend)
-		ftrace_run_shutdown_code();
+		command |= FTRACE_DISABLE_CALLS;
+
+	ftrace_run_update_code(command);
 	mutex_unlock(&ftraced_lock);
 }
 
@@ -430,11 +456,13 @@ static int notrace __ftrace_update_code(void *ignore)
 	struct dyn_ftrace *p;
 	struct hlist_head head;
 	struct hlist_node *t;
+	int save_ftrace_enabled;
 	cycle_t start, stop;
 	int i;
 
-	/* Don't be calling ftrace ops now */
-	__unregister_ftrace_function(&ftrace_shutdown_ops);
+	/* Don't be recording funcs now */
+	save_ftrace_enabled = ftrace_enabled;
+	ftrace_enabled = 0;
 
 	start = now(raw_smp_processor_id());
 	ftrace_update_cnt = 0;
@@ -449,7 +477,7 @@ static int notrace __ftrace_update_code(void *ignore)
 
 		/* all CPUS are stopped, we are safe to modify code */
 		hlist_for_each_entry(p, t, &head, node) {
-			ftrace_code_disable(p, MCOUNT_ADDR);
+			ftrace_code_disable(p);
 			ftrace_update_cnt++;
 		}
 
@@ -459,7 +487,7 @@ static int notrace __ftrace_update_code(void *ignore)
 	ftrace_update_time = stop - start;
 	ftrace_update_tot_cnt += ftrace_update_cnt;
 
-	__register_ftrace_function(&ftrace_shutdown_ops);
+	ftrace_enabled = save_ftrace_enabled;
 
 	return 0;
 }
@@ -515,11 +543,6 @@ static int __init ftrace_dyn_table_alloc(void)
 	struct ftrace_page *pg;
 	int cnt;
 	int i;
-	int ret;
-
-	ret = ftrace_dyn_arch_init();
-	if (ret)
-		return ret;
 
 	/* allocate a few pages */
 	ftrace_pages_start = (void *)get_zeroed_page(GFP_KERNEL);
@@ -557,11 +580,19 @@ static int __init ftrace_dyn_table_alloc(void)
 	return 0;
 }
 
-static int __init notrace ftrace_shutdown_init(void)
+static int __init notrace ftrace_dynamic_init(void)
 {
 	struct task_struct *p;
+	unsigned long addr;
 	int ret;
 
+	addr = (unsigned long)ftrace_record_ip;
+	stop_machine_run(ftrace_dyn_arch_init, &addr, NR_CPUS);
+
+	/* ftrace_dyn_arch_init places the return code in addr */
+	if (addr)
+		return addr;
+
 	ret = ftrace_dyn_table_alloc();
 	if (ret)
 		return ret;
@@ -570,12 +601,12 @@ static int __init notrace ftrace_shutdown_init(void)
 	if (IS_ERR(p))
 		return -1;
 
-	__register_ftrace_function(&ftrace_shutdown_ops);
+	last_ftrace_enabled = ftrace_enabled = 1;
 
 	return 0;
 }
 
-core_initcall(ftrace_shutdown_init);
+core_initcall(ftrace_dynamic_init);
 #else
 # define ftrace_startup()	  do { } while (0)
 # define ftrace_shutdown()	  do { } while (0)
@@ -599,9 +630,8 @@ int register_ftrace_function(struct ftrace_ops *ops)
 	int ret;
 
 	mutex_lock(&ftrace_sysctl_lock);
-	ftrace_startup();
-
 	ret = __register_ftrace_function(ops);
+	ftrace_startup();
 	mutex_unlock(&ftrace_sysctl_lock);
 
 	return ret;
@@ -619,10 +649,7 @@ int unregister_ftrace_function(struct ftrace_ops *ops)
 
 	mutex_lock(&ftrace_sysctl_lock);
 	ret = __unregister_ftrace_function(ops);
-
-	if (ftrace_list == &ftrace_list_end)
-		ftrace_shutdown();
-
+	ftrace_shutdown();
 	mutex_unlock(&ftrace_sysctl_lock);
 
 	return ret;
-- 
cgit v1.2.3


From 5072c59fd45e9976d02ee6f18c7336ef97623cbc Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:43 +0200
Subject: ftrace: add filter select functions to trace

This patch adds two files to the debugfs system:

 /debugfs/tracing/available_filter_functions

and

 /debugfs/tracing/set_ftrace_filter

The available_filter_functions lists all functions that has been
recorded by the ftraced that has called the ftrace_record_ip function.
This is to allow users to see what functions have been converted
to nops and can be enabled for tracing.

To enable functions, simply echo the names (whitespace delimited)
into set_ftrace_filter. Simple wildcards are also allowed.

echo 'scheduler' > /debugfs/tracing/set_ftrace_filter

Will have only the scheduler be activated when tracing is enabled.

echo 'sched_*' > /debugfs/tracing/set_ftrace_filter

Will have only the functions starting with 'sched_' be activated.

echo '*lock' > /debugfs/tracing/set_ftrace_filter

Will have only functions ending with 'lock' be activated.

echo '*lock*' > /debugfs/tracing/set_ftrace_filter

Will have only functions with 'lock' in its name be activated.

Note: 'sched*lock' will not work. The only wildcards that are
allowed is an asterisk and the beginning and or end of the string
passed in.

Multiple names can be passed in with whitespace delimited:

echo 'scheduler *lock *acpi*' > /debugfs/tracing/set_ftrace_filter

is also the same as:

echo 'scheduler' > /debugfs/tracing/set_ftrace_filter
echo '*lock' >> /debugfs/tracing/set_ftrace_filter
echo '*acpi*' >> /debugfs/tracing/set_ftrace_filter

Appending does just that. It appends to the list.

To disable all filters simply echo an empty line in:

echo > /debugfs/tracing/set_ftrace_filter

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h |   4 +-
 kernel/trace/ftrace.c  | 527 +++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 513 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index b0dd0093058f..f5911d2d42c3 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -43,7 +43,9 @@ extern void mcount(void);
 # define FTRACE_HASHSIZE	(1<<FTRACE_HASHBITS)
 
 enum {
-	FTRACE_FL_FAILED	= (1<<0),
+	FTRACE_FL_FAILED	= (1 << 0),
+	FTRACE_FL_FILTER	= (1 << 1),
+	FTRACE_FL_ENABLED	= (1 << 2),
 };
 
 struct dyn_ftrace {
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 88544f9bc0ed..97d5cb7b7e75 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -16,12 +16,15 @@
 #include <linux/stop_machine.h>
 #include <linux/clocksource.h>
 #include <linux/kallsyms.h>
+#include <linux/seq_file.h>
+#include <linux/debugfs.h>
 #include <linux/kthread.h>
 #include <linux/hardirq.h>
 #include <linux/ftrace.h>
-#include <linux/module.h>
+#include <linux/uaccess.h>
 #include <linux/sysctl.h>
 #include <linux/hash.h>
+#include <linux/ctype.h>
 #include <linux/list.h>
 
 #include "trace.h"
@@ -151,12 +154,15 @@ enum {
 	FTRACE_DISABLE_MCOUNT		= (1 << 4),
 };
 
+static int ftrace_filtered;
+
 static struct hlist_head ftrace_hash[FTRACE_HASHSIZE];
 
 static DEFINE_PER_CPU(int, ftrace_shutdown_disable_cpu);
 
 static DEFINE_SPINLOCK(ftrace_shutdown_lock);
 static DEFINE_MUTEX(ftraced_lock);
+static DEFINE_MUTEX(ftrace_filter_lock);
 
 struct ftrace_page {
 	struct ftrace_page	*next;
@@ -282,16 +288,82 @@ ftrace_record_ip(unsigned long ip)
 #define FTRACE_ADDR ((long)(&ftrace_caller))
 #define MCOUNT_ADDR ((long)(&mcount))
 
-static void notrace ftrace_replace_code(int saved)
+static void notrace
+__ftrace_replace_code(struct dyn_ftrace *rec,
+		      unsigned char *old, unsigned char *new, int enable)
+{
+	unsigned long ip;
+	int failed;
+
+	ip = rec->ip;
+
+	if (ftrace_filtered && enable) {
+		unsigned long fl;
+		/*
+		 * If filtering is on:
+		 *
+		 * If this record is set to be filtered and
+		 * is enabled then do nothing.
+		 *
+		 * If this record is set to be filtered and
+		 * it is not enabled, enable it.
+		 *
+		 * If this record is not set to be filtered
+		 * and it is not enabled do nothing.
+		 *
+		 * If this record is not set to be filtered and
+		 * it is enabled, disable it.
+		 */
+		fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_ENABLED);
+
+		if ((fl ==  (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) ||
+		    (fl == 0))
+			return;
+
+		/*
+		 * If it is enabled disable it,
+		 * otherwise enable it!
+		 */
+		if (fl == FTRACE_FL_ENABLED) {
+			/* swap new and old */
+			new = old;
+			old = ftrace_call_replace(ip, FTRACE_ADDR);
+			rec->flags &= ~FTRACE_FL_ENABLED;
+		} else {
+			new = ftrace_call_replace(ip, FTRACE_ADDR);
+			rec->flags |= FTRACE_FL_ENABLED;
+		}
+	} else {
+
+		if (enable)
+			new = ftrace_call_replace(ip, FTRACE_ADDR);
+		else
+			old = ftrace_call_replace(ip, FTRACE_ADDR);
+
+		if (enable) {
+			if (rec->flags & FTRACE_FL_ENABLED)
+				return;
+			rec->flags |= FTRACE_FL_ENABLED;
+		} else {
+			if (!(rec->flags & FTRACE_FL_ENABLED))
+				return;
+			rec->flags &= ~FTRACE_FL_ENABLED;
+		}
+	}
+
+	failed = ftrace_modify_code(ip, old, new);
+	if (failed)
+		rec->flags |= FTRACE_FL_FAILED;
+}
+
+static void notrace ftrace_replace_code(int enable)
 {
 	unsigned char *new = NULL, *old = NULL;
 	struct dyn_ftrace *rec;
 	struct ftrace_page *pg;
-	unsigned long ip;
-	int failed;
 	int i;
 
-	if (saved)
+	if (enable)
 		old = ftrace_nop_replace();
 	else
 		new = ftrace_nop_replace();
@@ -304,16 +376,7 @@ static void notrace ftrace_replace_code(int saved)
 			if (rec->flags & FTRACE_FL_FAILED)
 				continue;
 
-			ip = rec->ip;
-
-			if (saved)
-				new = ftrace_call_replace(ip, FTRACE_ADDR);
-			else
-				old = ftrace_call_replace(ip, FTRACE_ADDR);
-
-			failed = ftrace_modify_code(ip, old, new);
-			if (failed)
-				rec->flags |= FTRACE_FL_FAILED;
+			__ftrace_replace_code(rec, old, new, enable);
 		}
 	}
 }
@@ -580,6 +643,436 @@ static int __init ftrace_dyn_table_alloc(void)
 	return 0;
 }
 
+enum {
+	FTRACE_ITER_FILTER	= (1 << 0),
+	FTRACE_ITER_CONT	= (1 << 1),
+};
+
+#define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */
+
+struct ftrace_iterator {
+	loff_t			pos;
+	struct ftrace_page	*pg;
+	unsigned		idx;
+	unsigned		flags;
+	unsigned char		buffer[FTRACE_BUFF_MAX+1];
+	unsigned		buffer_idx;
+	unsigned		filtered;
+};
+
+static void notrace *
+t_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	struct ftrace_iterator *iter = m->private;
+	struct dyn_ftrace *rec = NULL;
+
+	(*pos)++;
+
+ retry:
+	if (iter->idx >= iter->pg->index) {
+		if (iter->pg->next) {
+			iter->pg = iter->pg->next;
+			iter->idx = 0;
+			goto retry;
+		}
+	} else {
+		rec = &iter->pg->records[iter->idx++];
+		if ((rec->flags & FTRACE_FL_FAILED) ||
+		    ((iter->flags & FTRACE_ITER_FILTER) &&
+		     !(rec->flags & FTRACE_FL_FILTER))) {
+			rec = NULL;
+			goto retry;
+		}
+	}
+
+	iter->pos = *pos;
+
+	return rec;
+}
+
+static void *t_start(struct seq_file *m, loff_t *pos)
+{
+	struct ftrace_iterator *iter = m->private;
+	void *p = NULL;
+	loff_t l = -1;
+
+	if (*pos != iter->pos) {
+		for (p = t_next(m, p, &l); p && l < *pos; p = t_next(m, p, &l))
+			;
+	} else {
+		l = *pos;
+		p = t_next(m, p, &l);
+	}
+
+	return p;
+}
+
+static void t_stop(struct seq_file *m, void *p)
+{
+}
+
+static int t_show(struct seq_file *m, void *v)
+{
+	struct dyn_ftrace *rec = v;
+	char str[KSYM_SYMBOL_LEN];
+
+	if (!rec)
+		return 0;
+
+	kallsyms_lookup(rec->ip, NULL, NULL, NULL, str);
+
+	seq_printf(m, "%s\n", str);
+
+	return 0;
+}
+
+static struct seq_operations show_ftrace_seq_ops = {
+	.start = t_start,
+	.next = t_next,
+	.stop = t_stop,
+	.show = t_show,
+};
+
+static int notrace
+ftrace_avail_open(struct inode *inode, struct file *file)
+{
+	struct ftrace_iterator *iter;
+	int ret;
+
+	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
+	if (!iter)
+		return -ENOMEM;
+
+	iter->pg = ftrace_pages_start;
+	iter->pos = -1;
+
+	ret = seq_open(file, &show_ftrace_seq_ops);
+	if (!ret) {
+		struct seq_file *m = file->private_data;
+		m->private = iter;
+	} else
+		kfree(iter);
+
+	return ret;
+}
+
+int ftrace_avail_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *m = (struct seq_file *)file->private_data;
+	struct ftrace_iterator *iter = m->private;
+
+	seq_release(inode, file);
+	kfree(iter);
+	return 0;
+}
+
+static void notrace ftrace_filter_reset(void)
+{
+	struct ftrace_page *pg;
+	struct dyn_ftrace *rec;
+	unsigned i;
+
+	/* keep kstop machine from running */
+	preempt_disable();
+	ftrace_filtered = 0;
+	pg = ftrace_pages_start;
+	while (pg) {
+		for (i = 0; i < pg->index; i++) {
+			rec = &pg->records[i];
+			if (rec->flags & FTRACE_FL_FAILED)
+				continue;
+			rec->flags &= ~FTRACE_FL_FILTER;
+		}
+		pg = pg->next;
+	}
+	preempt_enable();
+}
+
+static int notrace
+ftrace_filter_open(struct inode *inode, struct file *file)
+{
+	struct ftrace_iterator *iter;
+	int ret = 0;
+
+	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
+	if (!iter)
+		return -ENOMEM;
+
+	mutex_lock(&ftrace_filter_lock);
+	if ((file->f_mode & FMODE_WRITE) &&
+	    !(file->f_flags & O_APPEND))
+		ftrace_filter_reset();
+
+	if (file->f_mode & FMODE_READ) {
+		iter->pg = ftrace_pages_start;
+		iter->pos = -1;
+		iter->flags = FTRACE_ITER_FILTER;
+
+		ret = seq_open(file, &show_ftrace_seq_ops);
+		if (!ret) {
+			struct seq_file *m = file->private_data;
+			m->private = iter;
+		} else
+			kfree(iter);
+	} else
+		file->private_data = iter;
+	mutex_unlock(&ftrace_filter_lock);
+
+	return ret;
+}
+
+static ssize_t notrace
+ftrace_filter_read(struct file *file, char __user *ubuf,
+		       size_t cnt, loff_t *ppos)
+{
+	if (file->f_mode & FMODE_READ)
+		return seq_read(file, ubuf, cnt, ppos);
+	else
+		return -EPERM;
+}
+
+static loff_t notrace
+ftrace_filter_lseek(struct file *file, loff_t offset, int origin)
+{
+	loff_t ret;
+
+	if (file->f_mode & FMODE_READ)
+		ret = seq_lseek(file, offset, origin);
+	else
+		file->f_pos = ret = 1;
+
+	return ret;
+}
+
+enum {
+	MATCH_FULL,
+	MATCH_FRONT_ONLY,
+	MATCH_MIDDLE_ONLY,
+	MATCH_END_ONLY,
+};
+
+static void notrace
+ftrace_match(unsigned char *buff, int len)
+{
+	char str[KSYM_SYMBOL_LEN];
+	char *search = NULL;
+	struct ftrace_page *pg;
+	struct dyn_ftrace *rec;
+	int type = MATCH_FULL;
+	unsigned i, match = 0, search_len = 0;
+
+	for (i = 0; i < len; i++) {
+		if (buff[i] == '*') {
+			if (!i) {
+				search = buff + i + 1;
+				type = MATCH_END_ONLY;
+				search_len = len - (i + 1);
+			} else {
+				if (type == MATCH_END_ONLY) {
+					type = MATCH_MIDDLE_ONLY;
+				} else {
+					match = i;
+					type = MATCH_FRONT_ONLY;
+				}
+				buff[i] = 0;
+				break;
+			}
+		}
+	}
+
+	/* keep kstop machine from running */
+	preempt_disable();
+	ftrace_filtered = 1;
+	pg = ftrace_pages_start;
+	while (pg) {
+		for (i = 0; i < pg->index; i++) {
+			int matched = 0;
+			char *ptr;
+
+			rec = &pg->records[i];
+			if (rec->flags & FTRACE_FL_FAILED)
+				continue;
+			kallsyms_lookup(rec->ip, NULL, NULL, NULL, str);
+			switch (type) {
+			case MATCH_FULL:
+				if (strcmp(str, buff) == 0)
+					matched = 1;
+				break;
+			case MATCH_FRONT_ONLY:
+				if (memcmp(str, buff, match) == 0)
+					matched = 1;
+				break;
+			case MATCH_MIDDLE_ONLY:
+				if (strstr(str, search))
+					matched = 1;
+				break;
+			case MATCH_END_ONLY:
+				ptr = strstr(str, search);
+				if (ptr && (ptr[search_len] == 0))
+					matched = 1;
+				break;
+			}
+			if (matched)
+				rec->flags |= FTRACE_FL_FILTER;
+		}
+		pg = pg->next;
+	}
+	preempt_enable();
+}
+
+static ssize_t notrace
+ftrace_filter_write(struct file *file, const char __user *ubuf,
+		    size_t cnt, loff_t *ppos)
+{
+	struct ftrace_iterator *iter;
+	char ch;
+	size_t read = 0;
+	ssize_t ret;
+
+	if (!cnt || cnt < 0)
+		return 0;
+
+	mutex_lock(&ftrace_filter_lock);
+
+	if (file->f_mode & FMODE_READ) {
+		struct seq_file *m = file->private_data;
+		iter = m->private;
+	} else
+		iter = file->private_data;
+
+	if (!*ppos) {
+		iter->flags &= ~FTRACE_ITER_CONT;
+		iter->buffer_idx = 0;
+	}
+
+	ret = get_user(ch, ubuf++);
+	if (ret)
+		goto out;
+	read++;
+	cnt--;
+
+	if (!(iter->flags & ~FTRACE_ITER_CONT)) {
+		/* skip white space */
+		while (cnt && isspace(ch)) {
+			ret = get_user(ch, ubuf++);
+			if (ret)
+				goto out;
+			read++;
+			cnt--;
+		}
+
+
+		if (isspace(ch)) {
+			file->f_pos += read;
+			ret = read;
+			goto out;
+		}
+
+		iter->buffer_idx = 0;
+	}
+
+	while (cnt && !isspace(ch)) {
+		if (iter->buffer_idx < FTRACE_BUFF_MAX)
+			iter->buffer[iter->buffer_idx++] = ch;
+		else {
+			ret = -EINVAL;
+			goto out;
+		}
+		ret = get_user(ch, ubuf++);
+		if (ret)
+			goto out;
+		read++;
+		cnt--;
+	}
+
+	if (isspace(ch)) {
+		iter->filtered++;
+		iter->buffer[iter->buffer_idx] = 0;
+		ftrace_match(iter->buffer, iter->buffer_idx);
+		iter->buffer_idx = 0;
+	} else
+		iter->flags |= FTRACE_ITER_CONT;
+
+
+	file->f_pos += read;
+
+	ret = read;
+ out:
+	mutex_unlock(&ftrace_filter_lock);
+
+	return ret;
+}
+
+static int notrace
+ftrace_filter_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *m = (struct seq_file *)file->private_data;
+	struct ftrace_iterator *iter;
+
+	mutex_lock(&ftrace_filter_lock);
+	if (file->f_mode & FMODE_READ) {
+		iter = m->private;
+
+		seq_release(inode, file);
+	} else
+		iter = file->private_data;
+
+	if (iter->buffer_idx) {
+		iter->filtered++;
+		iter->buffer[iter->buffer_idx] = 0;
+		ftrace_match(iter->buffer, iter->buffer_idx);
+	}
+
+	mutex_lock(&ftrace_sysctl_lock);
+	mutex_lock(&ftraced_lock);
+	if (iter->filtered && ftraced_suspend && ftrace_enabled)
+		ftrace_run_update_code(FTRACE_ENABLE_CALLS);
+	mutex_unlock(&ftraced_lock);
+	mutex_unlock(&ftrace_sysctl_lock);
+
+	kfree(iter);
+	mutex_unlock(&ftrace_filter_lock);
+	return 0;
+}
+
+static struct file_operations ftrace_avail_fops = {
+	.open = ftrace_avail_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = ftrace_avail_release,
+};
+
+static struct file_operations ftrace_filter_fops = {
+	.open = ftrace_filter_open,
+	.read = ftrace_filter_read,
+	.write = ftrace_filter_write,
+	.llseek = ftrace_filter_lseek,
+	.release = ftrace_filter_release,
+};
+
+static __init int ftrace_init_debugfs(void)
+{
+	struct dentry *d_tracer;
+	struct dentry *entry;
+
+	d_tracer = tracing_init_dentry();
+
+	entry = debugfs_create_file("available_filter_functions", 0444,
+				    d_tracer, NULL, &ftrace_avail_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs "
+			   "'available_filter_functions' entry\n");
+
+	entry = debugfs_create_file("set_ftrace_filter", 0644, d_tracer,
+				    NULL, &ftrace_filter_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs "
+			   "'set_ftrace_filter' entry\n");
+	return 0;
+}
+
+fs_initcall(ftrace_init_debugfs);
+
 static int __init notrace ftrace_dynamic_init(void)
 {
 	struct task_struct *p;
@@ -657,14 +1150,14 @@ int unregister_ftrace_function(struct ftrace_ops *ops)
 
 notrace int
 ftrace_enable_sysctl(struct ctl_table *table, int write,
-		     struct file *filp, void __user *buffer, size_t *lenp,
+		     struct file *file, void __user *buffer, size_t *lenp,
 		     loff_t *ppos)
 {
 	int ret;
 
 	mutex_lock(&ftrace_sysctl_lock);
 
-	ret  = proc_dointvec(table, write, filp, buffer, lenp, ppos);
+	ret  = proc_dointvec(table, write, file, buffer, lenp, ppos);
 
 	if (ret || !write || (last_ftrace_enabled == ftrace_enabled))
 		goto out;
-- 
cgit v1.2.3


From f43fdad8627fec2d21df92799b254dceb66c9c3c Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:43 +0200
Subject: ftrace: fix kexec

disable the tracer while kexec pulls the rug from under the old
kernel.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/machine_kexec_32.c | 4 ++++
 arch/x86/kernel/machine_kexec_64.c | 4 ++++
 include/linux/ftrace.h             | 7 +++++++
 3 files changed, 15 insertions(+)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index d0b234c9fc31..88923fd7a6fc 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -11,6 +11,8 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/numa.h>
+#include <linux/ftrace.h>
+
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
@@ -107,6 +109,8 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	unsigned long page_list[PAGES_NR];
 	void *control_page;
 
+	tracer_disable();
+
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c
index 576a03db4511..1558fdc174f9 100644
--- a/arch/x86/kernel/machine_kexec_64.c
+++ b/arch/x86/kernel/machine_kexec_64.c
@@ -11,6 +11,8 @@
 #include <linux/string.h>
 #include <linux/reboot.h>
 #include <linux/numa.h>
+#include <linux/ftrace.h>
+
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <asm/mmu_context.h>
@@ -184,6 +186,8 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	unsigned long page_list[PAGES_NR];
 	void *control_page;
 
+	tracer_disable();
+
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index f5911d2d42c3..a42390c1d6e1 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -68,6 +68,13 @@ extern void ftrace_call(void);
 extern void mcount_call(void);
 #endif
 
+static inline void tracer_disable(void)
+{
+#ifdef CONFIG_FTRACE
+	ftrace_enabled = 0;
+#endif
+}
+
 #ifdef CONFIG_FRAME_POINTER
 /* TODO: need to fix this for ARM */
 # define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
-- 
cgit v1.2.3


From e1c08bdd9fa73e44096e5a82c0d5928b04ab02c8 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:44 +0200
Subject: ftrace: force recording

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h |  4 ++++
 kernel/trace/ftrace.c  | 51 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 55 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index a42390c1d6e1..2c1670c65236 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -54,6 +54,8 @@ struct dyn_ftrace {
 	unsigned long	  flags;
 };
 
+int ftrace_force_update(void);
+
 /* defined in arch */
 extern int ftrace_ip_converted(unsigned long ip);
 extern unsigned char *ftrace_nop_replace(void);
@@ -66,6 +68,8 @@ extern int ftrace_update_ftrace_func(ftrace_func_t func);
 extern void ftrace_caller(void);
 extern void ftrace_call(void);
 extern void mcount_call(void);
+#else
+# define ftrace_force_update() do { } while (0)
 #endif
 
 static inline void tracer_disable(void)
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 97d5cb7b7e75..4facf5ceeb86 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -146,6 +146,10 @@ static int notrace __unregister_ftrace_function(struct ftrace_ops *ops)
 
 #ifdef CONFIG_DYNAMIC_FTRACE
 
+static struct task_struct *ftraced_task;
+static DECLARE_WAIT_QUEUE_HEAD(ftraced_waiters);
+static unsigned long ftraced_iteration_counter;
+
 enum {
 	FTRACE_ENABLE_CALLS		= (1 << 0),
 	FTRACE_DISABLE_CALLS		= (1 << 1),
@@ -590,9 +594,12 @@ static int notrace ftraced(void *ignore)
 			ftraced_trigger = 0;
 			ftrace_record_suspend--;
 		}
+		ftraced_iteration_counter++;
 		mutex_unlock(&ftraced_lock);
 		mutex_unlock(&ftrace_sysctl_lock);
 
+		wake_up_interruptible(&ftraced_waiters);
+
 		ftrace_shutdown_replenish();
 
 		set_current_state(TASK_INTERRUPTIBLE);
@@ -1050,6 +1057,49 @@ static struct file_operations ftrace_filter_fops = {
 	.release = ftrace_filter_release,
 };
 
+/**
+ * ftrace_force_update - force an update to all recording ftrace functions
+ *
+ * The ftrace dynamic update daemon only wakes up once a second.
+ * There may be cases where an update needs to be done immediately
+ * for tests or internal kernel tracing to begin. This function
+ * wakes the daemon to do an update and will not return until the
+ * update is complete.
+ */
+int ftrace_force_update(void)
+{
+	unsigned long last_counter;
+	DECLARE_WAITQUEUE(wait, current);
+	int ret = 0;
+
+	if (!ftraced_task)
+		return -ENODEV;
+
+	mutex_lock(&ftraced_lock);
+	last_counter = ftraced_iteration_counter;
+
+	set_current_state(TASK_INTERRUPTIBLE);
+	add_wait_queue(&ftraced_waiters, &wait);
+
+	do {
+		mutex_unlock(&ftraced_lock);
+		wake_up_process(ftraced_task);
+		schedule();
+		mutex_lock(&ftraced_lock);
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			break;
+		}
+		set_current_state(TASK_INTERRUPTIBLE);
+	} while (last_counter == ftraced_iteration_counter);
+
+	mutex_unlock(&ftraced_lock);
+	remove_wait_queue(&ftraced_waiters, &wait);
+	set_current_state(TASK_RUNNING);
+
+	return ret;
+}
+
 static __init int ftrace_init_debugfs(void)
 {
 	struct dentry *d_tracer;
@@ -1095,6 +1145,7 @@ static int __init notrace ftrace_dynamic_init(void)
 		return -1;
 
 	last_ftrace_enabled = ftrace_enabled = 1;
+	ftraced_task = p;
 
 	return 0;
 }
-- 
cgit v1.2.3


From c7aafc549766b87819285d3480648fc652a47bc4 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:45 +0200
Subject: ftrace: cleanups

factor out code and clean it up.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h            |   2 +-
 kernel/trace/ftrace.c             |   8 +--
 kernel/trace/trace.c              | 144 ++++++++++++++++++++++++--------------
 kernel/trace/trace.h              |   8 ++-
 kernel/trace/trace_irqsoff.c      |  32 ++++-----
 kernel/trace/trace_sched_wakeup.c |  18 ++---
 kernel/trace/trace_selftest.c     |  25 ++++---
 7 files changed, 134 insertions(+), 103 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 2c1670c65236..953a36d6a199 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -69,7 +69,7 @@ extern void ftrace_caller(void);
 extern void ftrace_call(void);
 extern void mcount_call(void);
 #else
-# define ftrace_force_update() do { } while (0)
+# define ftrace_force_update() ({ 0; })
 #endif
 
 static inline void tracer_disable(void)
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 4facf5ceeb86..6d4d2e86debc 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1152,10 +1152,10 @@ static int __init notrace ftrace_dynamic_init(void)
 
 core_initcall(ftrace_dynamic_init);
 #else
-# define ftrace_startup()	  do { } while (0)
-# define ftrace_shutdown()	  do { } while (0)
-# define ftrace_startup_sysctl()  do { } while (0)
-# define ftrace_shutdown_sysctl() do { } while (0)
+# define ftrace_startup()		do { } while (0)
+# define ftrace_shutdown()		do { } while (0)
+# define ftrace_startup_sysctl()	do { } while (0)
+# define ftrace_shutdown_sysctl()	do { } while (0)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 /**
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index f6d026f17dbb..61d2f0228866 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -142,12 +142,59 @@ __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
 	tracing_record_cmdline(current);
 }
 
+void check_pages(struct trace_array_cpu *data)
+{
+	struct page *page, *tmp;
+
+	BUG_ON(data->trace_pages.next->prev != &data->trace_pages);
+	BUG_ON(data->trace_pages.prev->next != &data->trace_pages);
+
+	list_for_each_entry_safe(page, tmp, &data->trace_pages, lru) {
+		BUG_ON(page->lru.next->prev != &page->lru);
+		BUG_ON(page->lru.prev->next != &page->lru);
+	}
+}
+
+void *head_page(struct trace_array_cpu *data)
+{
+	struct page *page;
+
+	check_pages(data);
+	if (list_empty(&data->trace_pages))
+		return NULL;
+
+	page = list_entry(data->trace_pages.next, struct page, lru);
+	BUG_ON(&page->lru == &data->trace_pages);
+
+	return page_address(page);
+}
+
+notrace static void
+flip_trace(struct trace_array_cpu *tr1, struct trace_array_cpu *tr2)
+{
+	struct list_head flip_pages;
+
+	INIT_LIST_HEAD(&flip_pages);
+
+	tr1->trace_current = NULL;
+	memcpy(&tr1->trace_current_idx, &tr2->trace_current_idx,
+		sizeof(struct trace_array_cpu) -
+		offsetof(struct trace_array_cpu, trace_current_idx));
+
+	check_pages(tr1);
+	check_pages(tr2);
+	list_splice_init(&tr1->trace_pages, &flip_pages);
+	list_splice_init(&tr2->trace_pages, &tr1->trace_pages);
+	list_splice_init(&flip_pages, &tr2->trace_pages);
+	BUG_ON(!list_empty(&flip_pages));
+	check_pages(tr1);
+	check_pages(tr2);
+}
+
 notrace void
 update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
 {
 	struct trace_array_cpu *data;
-	void *save_trace;
-	struct list_head save_pages;
 	int i;
 
 	WARN_ON_ONCE(!irqs_disabled());
@@ -155,11 +202,7 @@ update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
 	/* clear out all the previous traces */
 	for_each_possible_cpu(i) {
 		data = tr->data[i];
-		save_trace = max_tr.data[i]->trace;
-		save_pages = max_tr.data[i]->trace_pages;
-		memcpy(max_tr.data[i], data, sizeof(*data));
-		data->trace = save_trace;
-		data->trace_pages = save_pages;
+		flip_trace(max_tr.data[i], data);
 		tracing_reset(data);
 	}
 
@@ -177,8 +220,6 @@ notrace void
 update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
 {
 	struct trace_array_cpu *data = tr->data[cpu];
-	void *save_trace;
-	struct list_head save_pages;
 	int i;
 
 	WARN_ON_ONCE(!irqs_disabled());
@@ -186,11 +227,8 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
 	for_each_possible_cpu(i)
 		tracing_reset(max_tr.data[i]);
 
-	save_trace = max_tr.data[cpu]->trace;
-	save_pages = max_tr.data[cpu]->trace_pages;
-	memcpy(max_tr.data[cpu], data, sizeof(*data));
-	data->trace = save_trace;
-	data->trace_pages = save_pages;
+	flip_trace(max_tr.data[cpu], data);
+
 	tracing_reset(data);
 
 	__update_max_tr(tr, tsk, cpu);
@@ -234,9 +272,9 @@ int register_tracer(struct tracer *type)
 		 * If we fail, we do not register this tracer.
 		 */
 		for_each_possible_cpu(i) {
-			if (!data->trace)
-				continue;
 			data = tr->data[i];
+			if (!head_page(data))
+				continue;
 			tracing_reset(data);
 		}
 		current_trace = type;
@@ -298,7 +336,7 @@ void unregister_tracer(struct tracer *type)
 void notrace tracing_reset(struct trace_array_cpu *data)
 {
 	data->trace_idx = 0;
-	data->trace_current = data->trace;
+	data->trace_current = head_page(data);
 	data->trace_current_idx = 0;
 }
 
@@ -425,26 +463,31 @@ notrace void tracing_record_cmdline(struct task_struct *tsk)
 }
 
 static inline notrace struct trace_entry *
-tracing_get_trace_entry(struct trace_array *tr,
-			struct trace_array_cpu *data)
+tracing_get_trace_entry(struct trace_array *tr, struct trace_array_cpu *data)
 {
 	unsigned long idx, idx_next;
 	struct trace_entry *entry;
-	struct page *page;
 	struct list_head *next;
+	struct page *page;
 
 	data->trace_idx++;
 	idx = data->trace_current_idx;
 	idx_next = idx + 1;
 
+	BUG_ON(idx * TRACE_ENTRY_SIZE >= PAGE_SIZE);
+
 	entry = data->trace_current + idx * TRACE_ENTRY_SIZE;
 
 	if (unlikely(idx_next >= ENTRIES_PER_PAGE)) {
 		page = virt_to_page(data->trace_current);
-		if (unlikely(&page->lru == data->trace_pages.prev))
-			next = data->trace_pages.next;
-		else
-			next = page->lru.next;
+		/*
+		 * Roundrobin - but skip the head (which is not a real page):
+		 */
+		next = page->lru.next;
+		if (unlikely(next == &data->trace_pages))
+			next = next->next;
+		BUG_ON(next == &data->trace_pages);
+
 		page = list_entry(next, struct page, lru);
 		data->trace_current = page_address(page);
 		idx_next = 0;
@@ -456,18 +499,17 @@ tracing_get_trace_entry(struct trace_array *tr,
 }
 
 static inline notrace void
-tracing_generic_entry_update(struct trace_entry *entry,
-			     unsigned long flags)
+tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags)
 {
 	struct task_struct *tsk = current;
 	unsigned long pc;
 
 	pc = preempt_count();
 
-	entry->idx	= atomic_inc_return(&tracer_counter);
-	entry->preempt_count = pc & 0xff;
-	entry->pid	 = tsk->pid;
-	entry->t	 = now(raw_smp_processor_id());
+	entry->idx		= atomic_inc_return(&tracer_counter);
+	entry->preempt_count	= pc & 0xff;
+	entry->pid		= tsk->pid;
+	entry->t		= now(raw_smp_processor_id());
 	entry->flags = (irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) |
 		((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) |
 		((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
@@ -476,16 +518,15 @@ tracing_generic_entry_update(struct trace_entry *entry,
 
 notrace void
 ftrace(struct trace_array *tr, struct trace_array_cpu *data,
-		       unsigned long ip, unsigned long parent_ip,
-		       unsigned long flags)
+       unsigned long ip, unsigned long parent_ip, unsigned long flags)
 {
 	struct trace_entry *entry;
 
-	entry = tracing_get_trace_entry(tr, data);
+	entry			= tracing_get_trace_entry(tr, data);
 	tracing_generic_entry_update(entry, flags);
-	entry->type	    = TRACE_FN;
-	entry->fn.ip	    = ip;
-	entry->fn.parent_ip = parent_ip;
+	entry->type		= TRACE_FN;
+	entry->fn.ip		= ip;
+	entry->fn.parent_ip	= parent_ip;
 }
 
 notrace void
@@ -496,7 +537,7 @@ tracing_sched_switch_trace(struct trace_array *tr,
 {
 	struct trace_entry *entry;
 
-	entry = tracing_get_trace_entry(tr, data);
+	entry			= tracing_get_trace_entry(tr, data);
 	tracing_generic_entry_update(entry, flags);
 	entry->type		= TRACE_CTX;
 	entry->ctx.prev_pid	= prev->pid;
@@ -540,6 +581,8 @@ trace_entry_idx(struct trace_array *tr, struct trace_array_cpu *data,
 	}
 
 	page = list_entry(iter->next_page[cpu], struct page, lru);
+	BUG_ON(&data->trace_pages == &page->lru);
+
 	array = page_address(page);
 
 	return &array[iter->next_page_idx[cpu]];
@@ -554,7 +597,7 @@ find_next_entry(struct trace_iterator *iter, int *ent_cpu)
 	int cpu;
 
 	for_each_possible_cpu(cpu) {
-		if (!tr->data[cpu]->trace)
+		if (!head_page(tr->data[cpu]))
 			continue;
 		ent = trace_entry_idx(tr, tr->data[cpu], iter, cpu);
 		if (ent &&
@@ -762,7 +805,7 @@ print_trace_header(struct seq_file *m, struct trace_iterator *iter)
 		name = type->name;
 
 	for_each_possible_cpu(cpu) {
-		if (tr->data[cpu]->trace) {
+		if (head_page(tr->data[cpu])) {
 			total += tr->data[cpu]->trace_idx;
 			if (tr->data[cpu]->trace_idx > tr->entries)
 				entries += tr->entries;
@@ -975,8 +1018,7 @@ static int trace_empty(struct trace_iterator *iter)
 	for_each_possible_cpu(cpu) {
 		data = iter->tr->data[cpu];
 
-		if (data->trace &&
-		    data->trace_idx)
+		if (head_page(data) && data->trace_idx)
 			return 0;
 	}
 	return 1;
@@ -1576,9 +1618,9 @@ static struct tracer no_tracer __read_mostly =
 static int trace_alloc_page(void)
 {
 	struct trace_array_cpu *data;
-	void *array;
 	struct page *page, *tmp;
 	LIST_HEAD(pages);
+	void *array;
 	int i;
 
 	/* first allocate a page for each CPU */
@@ -1610,14 +1652,14 @@ static int trace_alloc_page(void)
 	for_each_possible_cpu(i) {
 		data = global_trace.data[i];
 		page = list_entry(pages.next, struct page, lru);
-		list_del(&page->lru);
+		list_del_init(&page->lru);
 		list_add_tail(&page->lru, &data->trace_pages);
 		ClearPageLRU(page);
 
 #ifdef CONFIG_TRACER_MAX_TRACE
 		data = max_tr.data[i];
 		page = list_entry(pages.next, struct page, lru);
-		list_del(&page->lru);
+		list_del_init(&page->lru);
 		list_add_tail(&page->lru, &data->trace_pages);
 		SetPageLRU(page);
 #endif
@@ -1628,7 +1670,7 @@ static int trace_alloc_page(void)
 
  free_pages:
 	list_for_each_entry_safe(page, tmp, &pages, lru) {
-		list_del(&page->lru);
+		list_del_init(&page->lru);
 		__free_page(page);
 	}
 	return -ENOMEM;
@@ -1654,7 +1696,6 @@ __init static int tracer_alloc_buffers(void)
 			       "for trace buffer!\n");
 			goto free_buffers;
 		}
-		data->trace = array;
 
 		/* set the array to the list */
 		INIT_LIST_HEAD(&data->trace_pages);
@@ -1671,7 +1712,6 @@ __init static int tracer_alloc_buffers(void)
 			       "for trace buffer!\n");
 			goto free_buffers;
 		}
-		max_tr.data[i]->trace = array;
 
 		INIT_LIST_HEAD(&max_tr.data[i]->trace_pages);
 		page = virt_to_page(array);
@@ -1716,24 +1756,22 @@ __init static int tracer_alloc_buffers(void)
 		struct page *page, *tmp;
 		struct trace_array_cpu *data = global_trace.data[i];
 
-		if (data && data->trace) {
+		if (data) {
 			list_for_each_entry_safe(page, tmp,
 						 &data->trace_pages, lru) {
-				list_del(&page->lru);
+				list_del_init(&page->lru);
 				__free_page(page);
 			}
-			data->trace = NULL;
 		}
 
 #ifdef CONFIG_TRACER_MAX_TRACE
 		data = max_tr.data[i];
-		if (data && data->trace) {
+		if (data) {
 			list_for_each_entry_safe(page, tmp,
 						 &data->trace_pages, lru) {
-				list_del(&page->lru);
+				list_del_init(&page->lru);
 				__free_page(page);
 			}
-			data->trace = NULL;
 		}
 #endif
 	}
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 88edbf1f6788..cc1d34b8b771 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -53,12 +53,12 @@ struct trace_entry {
  * the trace, etc.)
  */
 struct trace_array_cpu {
-	void			*trace;
 	void			*trace_current;
-	unsigned		trace_current_idx;
 	struct list_head	trace_pages;
-	unsigned long		trace_idx;
 	atomic_t		disabled;
+	/* these fields get copied into max-trace: */
+	unsigned		trace_current_idx;
+	unsigned long		trace_idx;
 	unsigned long		saved_latency;
 	unsigned long		critical_start;
 	unsigned long		critical_end;
@@ -216,4 +216,6 @@ extern int trace_selftest_startup_sched_switch(struct tracer *trace,
 #endif
 #endif /* CONFIG_FTRACE_STARTUP_TEST */
 
+extern void *head_page(struct trace_array_cpu *data);
+
 #endif /* _LINUX_KERNEL_TRACE_H */
diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
index 14183b8f79c5..2dfebb67fdfb 100644
--- a/kernel/trace/trace_irqsoff.c
+++ b/kernel/trace/trace_irqsoff.c
@@ -144,7 +144,7 @@ check_critical_timing(struct trace_array *tr,
 	if (!report_latency(delta))
 		goto out;
 
-	spin_lock(&max_trace_lock);
+	spin_lock_irqsave(&max_trace_lock, flags);
 
 	/* check if we are still the max latency */
 	if (!report_latency(delta))
@@ -165,32 +165,24 @@ check_critical_timing(struct trace_array *tr,
 
 	update_max_tr_single(tr, current, cpu);
 
-	if (tracing_thresh)
-		printk(KERN_INFO "(%16s-%-5d|#%d): %lu us critical section "
-		       "violates %lu us threshold.\n"
-		       " => started at timestamp %lu: ",
+	if (tracing_thresh) {
+		printk(KERN_INFO "(%16s-%-5d|#%d):"
+			" %lu us critical section violates %lu us threshold.\n",
 				current->comm, current->pid,
 				raw_smp_processor_id(),
-				latency, nsecs_to_usecs(tracing_thresh), t0);
-	else
+				latency, nsecs_to_usecs(tracing_thresh));
+	} else {
 		printk(KERN_INFO "(%16s-%-5d|#%d):"
-		       " new %lu us maximum-latency "
-		       "critical section.\n => started at timestamp %lu: ",
+		       " new %lu us maximum-latency critical section.\n",
 				current->comm, current->pid,
 				raw_smp_processor_id(),
-				latency, t0);
-
-	print_symbol(KERN_CONT "<%s>\n", data->critical_start);
-	printk(KERN_CONT " =>   ended at timestamp %lu: ", t1);
-	print_symbol(KERN_CONT "<%s>\n", data->critical_end);
-	dump_stack();
-	t1 = nsecs_to_usecs(now(cpu));
-	printk(KERN_CONT " =>   dump-end timestamp %lu\n\n", t1);
+				latency);
+	}
 
 	max_sequence++;
 
 out_unlock:
-	spin_unlock(&max_trace_lock);
+	spin_unlock_irqrestore(&max_trace_lock, flags);
 
 out:
 	data->critical_sequence = max_sequence;
@@ -216,7 +208,7 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip)
 	cpu = raw_smp_processor_id();
 	data = tr->data[cpu];
 
-	if (unlikely(!data) || unlikely(!data->trace) ||
+	if (unlikely(!data) || unlikely(!head_page(data)) ||
 	    atomic_read(&data->disabled))
 		return;
 
@@ -256,7 +248,7 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip)
 	cpu = raw_smp_processor_id();
 	data = tr->data[cpu];
 
-	if (unlikely(!data) || unlikely(!data->trace) ||
+	if (unlikely(!data) || unlikely(!head_page(data)) ||
 	    !data->critical_start || atomic_read(&data->disabled))
 		return;
 
diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c
index 3d10ff01f805..688df965f3f2 100644
--- a/kernel/trace/trace_sched_wakeup.c
+++ b/kernel/trace/trace_sched_wakeup.c
@@ -107,24 +107,18 @@ wakeup_sched_switch(struct task_struct *prev, struct task_struct *next)
 	update_max_tr(tr, wakeup_task, wakeup_cpu);
 
 	if (tracing_thresh) {
-		printk(KERN_INFO "(%16s-%-5d|#%d): %lu us wakeup latency "
-		       "violates %lu us threshold.\n"
-		       " => started at timestamp %lu: ",
+		printk(KERN_INFO "(%16s-%-5d|#%d):"
+			" %lu us wakeup latency violates %lu us threshold.\n",
 				wakeup_task->comm, wakeup_task->pid,
 				raw_smp_processor_id(),
-				latency, nsecs_to_usecs(tracing_thresh), t0);
+				latency, nsecs_to_usecs(tracing_thresh));
 	} else {
-		printk(KERN_INFO "(%16s-%-5d|#%d): new %lu us maximum "
-		       "wakeup latency.\n => started at timestamp %lu: ",
+		printk(KERN_INFO "(%16s-%-5d|#%d):"
+			" new %lu us maximum wakeup latency.\n",
 				wakeup_task->comm, wakeup_task->pid,
-				cpu, latency, t0);
+				cpu, latency);
 	}
 
-	printk(KERN_CONT "   ended at timestamp %lu: ", t1);
-	dump_stack();
-	t1 = nsecs_to_usecs(now(cpu));
-	printk(KERN_CONT "   dump-end timestamp %lu\n\n", t1);
-
 out_unlock:
 	__wakeup_reset(tr);
 	spin_unlock_irqrestore(&wakeup_lock, flags);
diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c
index ef4d3cc009f5..c01874c3b1f9 100644
--- a/kernel/trace/trace_selftest.c
+++ b/kernel/trace/trace_selftest.c
@@ -1,6 +1,7 @@
 /* Include in trace.c */
 
 #include <linux/kthread.h>
+#include <linux/delay.h>
 
 static inline int trace_valid_entry(struct trace_entry *entry)
 {
@@ -15,28 +16,29 @@ static inline int trace_valid_entry(struct trace_entry *entry)
 static int
 trace_test_buffer_cpu(struct trace_array *tr, struct trace_array_cpu *data)
 {
-	struct page *page;
 	struct trace_entry *entries;
+	struct page *page;
 	int idx = 0;
 	int i;
 
+	BUG_ON(list_empty(&data->trace_pages));
 	page = list_entry(data->trace_pages.next, struct page, lru);
 	entries = page_address(page);
 
-	if (data->trace != entries)
+	if (head_page(data) != entries)
 		goto failed;
 
 	/*
 	 * The starting trace buffer always has valid elements,
-	 * if any element exits.
+	 * if any element exists.
 	 */
-	entries = data->trace;
+	entries = head_page(data);
 
 	for (i = 0; i < tr->entries; i++) {
 
-		if (i < data->trace_idx &&
-		    !trace_valid_entry(&entries[idx])) {
-			printk(KERN_CONT ".. invalid entry %d ", entries[idx].type);
+		if (i < data->trace_idx && !trace_valid_entry(&entries[idx])) {
+			printk(KERN_CONT ".. invalid entry %d ",
+				entries[idx].type);
 			goto failed;
 		}
 
@@ -80,11 +82,10 @@ static int trace_test_buffer(struct trace_array *tr, unsigned long *count)
 	int ret = 0;
 
 	for_each_possible_cpu(cpu) {
-		if (!tr->data[cpu]->trace)
+		if (!head_page(tr->data[cpu]))
 			continue;
 
 		cnt += tr->data[cpu]->trace_idx;
-		printk("%d: count = %ld\n", cpu, cnt);
 
 		ret = trace_test_buffer_cpu(tr, tr->data[cpu]);
 		if (ret)
@@ -117,6 +118,8 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr)
 	}
 
 	/* start the tracing */
+	ftrace_enabled = 1;
+
 	tr->ctrl = 1;
 	trace->init(tr);
 	/* Sleep for a 1/10 of a second */
@@ -124,6 +127,8 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr)
 	/* stop the tracing. */
 	tr->ctrl = 0;
 	trace->ctrl_update(tr);
+	ftrace_enabled = 0;
+
 	/* check the trace buffer */
 	ret = trace_test_buffer(tr, &count);
 	trace->reset(tr);
@@ -328,7 +333,7 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr)
 
 	/* create a high prio thread */
 	p = kthread_run(trace_wakeup_test_thread, &isrt, "ftrace-test");
-	if (!IS_ERR(p)) {
+	if (IS_ERR(p)) {
 		printk(KERN_CONT "Failed to create ftrace wakeup test thread ");
 		return -1;
 	}
-- 
cgit v1.2.3


From 77a2b37d227483fe52aead242652aee406c25bf0 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:45 +0200
Subject: ftrace: startup tester on dynamic tracing.

This patch adds a startup self test on dynamic code modification
and filters. The test filters on a specific function, makes sure that
no other function is traced, exectutes the function, then makes sure that
the function is traced.

This patch also fixes a slight bug with the ftrace selftest, where
tracer_enabled was not being set.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h        |   2 +
 kernel/trace/ftrace.c         |  19 +++++++
 kernel/trace/trace_selftest.c | 113 ++++++++++++++++++++++++++++++++++++++++--
 3 files changed, 130 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 953a36d6a199..a842d96c6343 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -55,6 +55,7 @@ struct dyn_ftrace {
 };
 
 int ftrace_force_update(void);
+void ftrace_set_filter(unsigned char *buf, int len, int reset);
 
 /* defined in arch */
 extern int ftrace_ip_converted(unsigned long ip);
@@ -70,6 +71,7 @@ extern void ftrace_call(void);
 extern void mcount_call(void);
 #else
 # define ftrace_force_update() ({ 0; })
+# define ftrace_set_filter(buf, len, reset) do { } while (0)
 #endif
 
 static inline void tracer_disable(void)
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 6d4d2e86debc..5e9389faaf75 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1010,6 +1010,25 @@ ftrace_filter_write(struct file *file, const char __user *ubuf,
 	return ret;
 }
 
+/**
+ * ftrace_set_filter - set a function to filter on in ftrace
+ * @buf - the string that holds the function filter text.
+ * @len - the length of the string.
+ * @reset - non zero to reset all filters before applying this filter.
+ *
+ * Filters denote which functions should be enabled when tracing is enabled.
+ * If @buf is NULL and reset is set, all functions will be enabled for tracing.
+ */
+notrace void ftrace_set_filter(unsigned char *buf, int len, int reset)
+{
+	mutex_lock(&ftrace_filter_lock);
+	if (reset)
+		ftrace_filter_reset();
+	if (buf)
+		ftrace_match(buf, len);
+	mutex_unlock(&ftrace_filter_lock);
+}
+
 static int notrace
 ftrace_filter_release(struct inode *inode, struct file *file)
 {
diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c
index c01874c3b1f9..4c8a1b2d8231 100644
--- a/kernel/trace/trace_selftest.c
+++ b/kernel/trace/trace_selftest.c
@@ -99,6 +99,100 @@ static int trace_test_buffer(struct trace_array *tr, unsigned long *count)
 }
 
 #ifdef CONFIG_FTRACE
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+
+#define DYN_FTRACE_TEST_NAME trace_selftest_dynamic_test_func
+#define __STR(x) #x
+#define STR(x) __STR(x)
+static int DYN_FTRACE_TEST_NAME(void)
+{
+	/* used to call mcount */
+	return 0;
+}
+
+/* Test dynamic code modification and ftrace filters */
+int trace_selftest_startup_dynamic_tracing(struct tracer *trace,
+					   struct trace_array *tr,
+					   int (*func)(void))
+{
+	unsigned long count;
+	int ret;
+	int save_ftrace_enabled = ftrace_enabled;
+	int save_tracer_enabled = tracer_enabled;
+
+	/* The ftrace test PASSED */
+	printk(KERN_CONT "PASSED\n");
+	pr_info("Testing dynamic ftrace: ");
+
+	/* enable tracing, and record the filter function */
+	ftrace_enabled = 1;
+	tracer_enabled = 1;
+
+	/* passed in by parameter to fool gcc from optimizing */
+	func();
+
+	/* update the records */
+	ret = ftrace_force_update();
+	if (ret) {
+		printk(KERN_CONT ".. ftraced failed .. ");
+		return ret;
+	}
+
+	/* filter only on our function */
+	ftrace_set_filter(STR(DYN_FTRACE_TEST_NAME),
+			  sizeof(STR(DYN_FTRACE_TEST_NAME)), 1);
+
+	/* enable tracing */
+	tr->ctrl = 1;
+	trace->init(tr);
+	/* Sleep for a 1/10 of a second */
+	msleep(100);
+
+	/* we should have nothing in the buffer */
+	ret = trace_test_buffer(tr, &count);
+	if (ret)
+		goto out;
+
+	if (count) {
+		ret = -1;
+		printk(KERN_CONT ".. filter did not filter .. ");
+		goto out;
+	}
+
+	/* call our function again */
+	func();
+
+	/* sleep again */
+	msleep(100);
+
+	/* stop the tracing. */
+	tr->ctrl = 0;
+	trace->ctrl_update(tr);
+	ftrace_enabled = 0;
+
+	/* check the trace buffer */
+	ret = trace_test_buffer(tr, &count);
+	trace->reset(tr);
+
+	/* we should only have one item */
+	if (!ret && count != 1) {
+		printk(KERN_CONT ".. filter failed ..");
+		ret = -1;
+		goto out;
+	}
+ out:
+	ftrace_enabled = save_ftrace_enabled;
+	tracer_enabled = save_tracer_enabled;
+
+	/* Enable tracing on all functions again */
+	ftrace_set_filter(NULL, 0, 1);
+
+	return ret;
+}
+#else
+# define trace_selftest_startup_dynamic_tracing(trace, tr, func) ({ 0; })
+#endif /* CONFIG_DYNAMIC_FTRACE */
 /*
  * Simple verification test of ftrace function tracer.
  * Enable ftrace, sleep 1/10 second, and then read the trace
@@ -109,8 +203,13 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr)
 {
 	unsigned long count;
 	int ret;
+	int save_ftrace_enabled = ftrace_enabled;
+	int save_tracer_enabled = tracer_enabled;
 
-	/* make sure functions have been recorded */
+	/* make sure msleep has been recorded */
+	msleep(1);
+
+	/* force the recorded functions to be traced */
 	ret = ftrace_force_update();
 	if (ret) {
 		printk(KERN_CONT ".. ftraced failed .. ");
@@ -119,6 +218,7 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr)
 
 	/* start the tracing */
 	ftrace_enabled = 1;
+	tracer_enabled = 1;
 
 	tr->ctrl = 1;
 	trace->init(tr);
@@ -136,8 +236,16 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr)
 	if (!ret && !count) {
 		printk(KERN_CONT ".. no entries found ..");
 		ret = -1;
+		goto out;
 	}
 
+	ret = trace_selftest_startup_dynamic_tracing(trace, tr,
+						     DYN_FTRACE_TEST_NAME);
+
+ out:
+	ftrace_enabled = save_ftrace_enabled;
+	tracer_enabled = save_tracer_enabled;
+
 	return ret;
 }
 #endif /* CONFIG_FTRACE */
@@ -415,6 +523,3 @@ trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr
 	return ret;
 }
 #endif /* CONFIG_CONTEXT_SWITCH_TRACER */
-
-#ifdef CONFIG_DYNAMIC_FTRACE
-#endif /* CONFIG_DYNAMIC_FTRACE */
-- 
cgit v1.2.3


From 37ad508419f0fdfda7b378756eb1f35cfd26d96d Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:48 +0200
Subject: ftrace - fix dynamic ftrace memory leak

The ftrace dynamic function update allocates a record to store the
instruction pointers that are being modified. If the modified
instruction pointer fails to update, then the record is marked as
failed and nothing more is done.

Worse, if the modification fails, but the record ip function is still
called, it will allocate a new record and try again. In just a matter
of time, will this cause a serious memory leak and crash the system.

This patch plugs this memory leak. When a record fails, it is
included back into the pool of records to be used. Now a record may
fail over and over again, but the number of allocated records will
not increase.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h |  7 ++++---
 kernel/trace/ftrace.c  | 45 ++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 46 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index a842d96c6343..61e757bd2350 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -43,9 +43,10 @@ extern void mcount(void);
 # define FTRACE_HASHSIZE	(1<<FTRACE_HASHBITS)
 
 enum {
-	FTRACE_FL_FAILED	= (1 << 0),
-	FTRACE_FL_FILTER	= (1 << 1),
-	FTRACE_FL_ENABLED	= (1 << 2),
+	FTRACE_FL_FREE		= (1 << 0),
+	FTRACE_FL_FAILED	= (1 << 1),
+	FTRACE_FL_FILTER	= (1 << 2),
+	FTRACE_FL_ENABLED	= (1 << 3),
 };
 
 struct dyn_ftrace {
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index a15e068535f8..8e02aa690b2b 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -188,6 +188,8 @@ static int ftraced_suspend;
 
 static int ftrace_record_suspend;
 
+static struct dyn_ftrace *ftrace_free_records;
+
 static inline int
 notrace ftrace_ip_in_hash(unsigned long ip, unsigned long key)
 {
@@ -211,8 +213,35 @@ ftrace_add_hash(struct dyn_ftrace *node, unsigned long key)
 	hlist_add_head(&node->node, &ftrace_hash[key]);
 }
 
+static notrace void ftrace_free_rec(struct dyn_ftrace *rec)
+{
+	/* no locking, only called from kstop_machine */
+
+	rec->ip = (unsigned long)ftrace_free_records;
+	ftrace_free_records = rec;
+	rec->flags |= FTRACE_FL_FREE;
+}
+
 static notrace struct dyn_ftrace *ftrace_alloc_dyn_node(unsigned long ip)
 {
+	struct dyn_ftrace *rec;
+
+	/* First check for freed records */
+	if (ftrace_free_records) {
+		rec = ftrace_free_records;
+
+		/* todo, disable tracing altogether on this warning */
+		if (unlikely(!(rec->flags & FTRACE_FL_FREE))) {
+			WARN_ON_ONCE(1);
+			ftrace_free_records = NULL;
+			return NULL;
+		}
+
+		ftrace_free_records = (void *)rec->ip;
+		memset(rec, 0, sizeof(*rec));
+		return rec;
+	}
+
 	if (ftrace_pages->index == ENTRIES_PER_PAGE) {
 		if (!ftrace_pages->next)
 			return NULL;
@@ -356,8 +385,16 @@ __ftrace_replace_code(struct dyn_ftrace *rec,
 	}
 
 	failed = ftrace_modify_code(ip, old, new);
-	if (failed)
-		rec->flags |= FTRACE_FL_FAILED;
+	if (failed) {
+		unsigned long key;
+		/* It is possible that the function hasn't been converted yet */
+		key = hash_long(ip, FTRACE_HASHBITS);
+		if (!ftrace_ip_in_hash(ip, key)) {
+			rec->flags |= FTRACE_FL_FAILED;
+			ftrace_free_rec(rec);
+		}
+
+	}
 }
 
 static void notrace ftrace_replace_code(int enable)
@@ -407,8 +444,10 @@ ftrace_code_disable(struct dyn_ftrace *rec)
 	call = ftrace_call_replace(ip, MCOUNT_ADDR);
 
 	failed = ftrace_modify_code(ip, call, nop);
-	if (failed)
+	if (failed) {
 		rec->flags |= FTRACE_FL_FAILED;
+		ftrace_free_rec(rec);
+	}
 }
 
 static int notrace __ftrace_modify_code(void *data)
-- 
cgit v1.2.3


From 4eebcc81a33fbc45e28542b50197ed7b3c486d90 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 12 May 2008 21:20:48 +0200
Subject: ftrace: disable tracing on failure

Since ftrace touches practically every function. If we detect any
anomaly, we want to fully disable ftrace. This patch adds code
to try shutdown ftrace as much as possible without doing any more
harm is something is detected not quite correct.

This only kills ftrace, this patch does have checks for other parts of
the tracer (irqsoff, wakeup, etc.).

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h        |   3 ++
 kernel/trace/ftrace.c         | 112 ++++++++++++++++++++++++++++++++++++++----
 kernel/trace/trace_selftest.c |   4 ++
 3 files changed, 110 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 61e757bd2350..4650a3160b7f 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -58,6 +58,9 @@ struct dyn_ftrace {
 int ftrace_force_update(void);
 void ftrace_set_filter(unsigned char *buf, int len, int reset);
 
+/* totally disable ftrace - can not re-enable after this */
+void ftrace_kill(void);
+
 /* defined in arch */
 extern int ftrace_ip_converted(unsigned long ip);
 extern unsigned char *ftrace_nop_replace(void);
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 8e02aa690b2b..ff42345dd78e 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -29,9 +29,16 @@
 
 #include "trace.h"
 
-int ftrace_enabled;
+/* ftrace_enabled is a method to turn ftrace on or off */
+int ftrace_enabled __read_mostly;
 static int last_ftrace_enabled;
 
+/*
+ * ftrace_disabled is set when an anomaly is discovered.
+ * ftrace_disabled is much stronger than ftrace_enabled.
+ */
+static int ftrace_disabled __read_mostly;
+
 static DEFINE_SPINLOCK(ftrace_lock);
 static DEFINE_MUTEX(ftrace_sysctl_lock);
 
@@ -230,10 +237,11 @@ static notrace struct dyn_ftrace *ftrace_alloc_dyn_node(unsigned long ip)
 	if (ftrace_free_records) {
 		rec = ftrace_free_records;
 
-		/* todo, disable tracing altogether on this warning */
 		if (unlikely(!(rec->flags & FTRACE_FL_FREE))) {
 			WARN_ON_ONCE(1);
 			ftrace_free_records = NULL;
+			ftrace_disabled = 1;
+			ftrace_enabled = 0;
 			return NULL;
 		}
 
@@ -260,7 +268,7 @@ ftrace_record_ip(unsigned long ip)
 	int resched;
 	int atomic;
 
-	if (!ftrace_enabled)
+	if (!ftrace_enabled || ftrace_disabled)
 		return;
 
 	resched = need_resched();
@@ -485,6 +493,9 @@ static void notrace ftrace_startup(void)
 {
 	int command = 0;
 
+	if (unlikely(ftrace_disabled))
+		return;
+
 	mutex_lock(&ftraced_lock);
 	ftraced_suspend++;
 	if (ftraced_suspend == 1)
@@ -507,6 +518,9 @@ static void notrace ftrace_shutdown(void)
 {
 	int command = 0;
 
+	if (unlikely(ftrace_disabled))
+		return;
+
 	mutex_lock(&ftraced_lock);
 	ftraced_suspend--;
 	if (!ftraced_suspend)
@@ -529,6 +543,9 @@ static void notrace ftrace_startup_sysctl(void)
 {
 	int command = FTRACE_ENABLE_MCOUNT;
 
+	if (unlikely(ftrace_disabled))
+		return;
+
 	mutex_lock(&ftraced_lock);
 	/* Force update next time */
 	saved_ftrace_func = NULL;
@@ -544,6 +561,9 @@ static void notrace ftrace_shutdown_sysctl(void)
 {
 	int command = FTRACE_DISABLE_MCOUNT;
 
+	if (unlikely(ftrace_disabled))
+		return;
+
 	mutex_lock(&ftraced_lock);
 	/* ftraced_suspend is true if ftrace is running */
 	if (ftraced_suspend)
@@ -600,6 +620,9 @@ static int notrace __ftrace_update_code(void *ignore)
 
 static void notrace ftrace_update_code(void)
 {
+	if (unlikely(ftrace_disabled))
+		return;
+
 	stop_machine_run(__ftrace_update_code, NULL, NR_CPUS);
 }
 
@@ -614,6 +637,9 @@ static int notrace ftraced(void *ignore)
 		/* check once a second */
 		schedule_timeout(HZ);
 
+		if (unlikely(ftrace_disabled))
+			continue;
+
 		mutex_lock(&ftrace_sysctl_lock);
 		mutex_lock(&ftraced_lock);
 		if (ftrace_enabled && ftraced_trigger && !ftraced_suspend) {
@@ -628,6 +654,7 @@ static int notrace ftraced(void *ignore)
 					ftrace_update_cnt != 1 ? "s" : "",
 					ftrace_update_tot_cnt,
 					usecs, usecs != 1 ? "s" : "");
+				ftrace_disabled = 1;
 				WARN_ON_ONCE(1);
 			}
 			ftraced_trigger = 0;
@@ -785,6 +812,9 @@ ftrace_avail_open(struct inode *inode, struct file *file)
 	struct ftrace_iterator *iter;
 	int ret;
 
+	if (unlikely(ftrace_disabled))
+		return -ENODEV;
+
 	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
 	if (!iter)
 		return -ENOMEM;
@@ -843,6 +873,9 @@ ftrace_filter_open(struct inode *inode, struct file *file)
 	struct ftrace_iterator *iter;
 	int ret = 0;
 
+	if (unlikely(ftrace_disabled))
+		return -ENODEV;
+
 	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
 	if (!iter)
 		return -ENOMEM;
@@ -1063,6 +1096,9 @@ ftrace_filter_write(struct file *file, const char __user *ubuf,
  */
 notrace void ftrace_set_filter(unsigned char *buf, int len, int reset)
 {
+	if (unlikely(ftrace_disabled))
+		return;
+
 	mutex_lock(&ftrace_filter_lock);
 	if (reset)
 		ftrace_filter_reset();
@@ -1133,7 +1169,7 @@ int ftrace_force_update(void)
 	DECLARE_WAITQUEUE(wait, current);
 	int ret = 0;
 
-	if (!ftraced_task)
+	if (unlikely(ftrace_disabled))
 		return -ENODEV;
 
 	mutex_lock(&ftraced_lock);
@@ -1142,6 +1178,11 @@ int ftrace_force_update(void)
 	set_current_state(TASK_INTERRUPTIBLE);
 	add_wait_queue(&ftraced_waiters, &wait);
 
+	if (unlikely(!ftraced_task)) {
+		ret = -ENODEV;
+		goto out;
+	}
+
 	do {
 		mutex_unlock(&ftraced_lock);
 		wake_up_process(ftraced_task);
@@ -1154,6 +1195,7 @@ int ftrace_force_update(void)
 		set_current_state(TASK_INTERRUPTIBLE);
 	} while (last_counter == ftraced_iteration_counter);
 
+ out:
 	mutex_unlock(&ftraced_lock);
 	remove_wait_queue(&ftraced_waiters, &wait);
 	set_current_state(TASK_RUNNING);
@@ -1161,6 +1203,22 @@ int ftrace_force_update(void)
 	return ret;
 }
 
+static void ftrace_force_shutdown(void)
+{
+	struct task_struct *task;
+	int command = FTRACE_DISABLE_CALLS | FTRACE_UPDATE_TRACE_FUNC;
+
+	mutex_lock(&ftraced_lock);
+	task = ftraced_task;
+	ftraced_task = NULL;
+	ftraced_suspend = -1;
+	ftrace_run_update_code(command);
+	mutex_unlock(&ftraced_lock);
+
+	if (task)
+		kthread_stop(task);
+}
+
 static __init int ftrace_init_debugfs(void)
 {
 	struct dentry *d_tracer;
@@ -1194,21 +1252,29 @@ static int __init notrace ftrace_dynamic_init(void)
 	stop_machine_run(ftrace_dyn_arch_init, &addr, NR_CPUS);
 
 	/* ftrace_dyn_arch_init places the return code in addr */
-	if (addr)
-		return addr;
+	if (addr) {
+		ret = (int)addr;
+		goto failed;
+	}
 
 	ret = ftrace_dyn_table_alloc();
 	if (ret)
-		return ret;
+		goto failed;
 
 	p = kthread_run(ftraced, NULL, "ftraced");
-	if (IS_ERR(p))
-		return -1;
+	if (IS_ERR(p)) {
+		ret = -1;
+		goto failed;
+	}
 
 	last_ftrace_enabled = ftrace_enabled = 1;
 	ftraced_task = p;
 
 	return 0;
+
+ failed:
+	ftrace_disabled = 1;
+	return ret;
 }
 
 core_initcall(ftrace_dynamic_init);
@@ -1217,8 +1283,30 @@ core_initcall(ftrace_dynamic_init);
 # define ftrace_shutdown()		do { } while (0)
 # define ftrace_startup_sysctl()	do { } while (0)
 # define ftrace_shutdown_sysctl()	do { } while (0)
+# define ftrace_force_shutdown()	do { } while (0)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
+/**
+ * ftrace_kill - totally shutdown ftrace
+ *
+ * This is a safety measure. If something was detected that seems
+ * wrong, calling this function will keep ftrace from doing
+ * any more modifications, and updates.
+ * used when something went wrong.
+ */
+void ftrace_kill(void)
+{
+	mutex_lock(&ftrace_sysctl_lock);
+	ftrace_disabled = 1;
+	ftrace_enabled = 0;
+
+	clear_ftrace_function();
+	mutex_unlock(&ftrace_sysctl_lock);
+
+	/* Try to totally disable ftrace */
+	ftrace_force_shutdown();
+}
+
 /**
  * register_ftrace_function - register a function for profiling
  * @ops - ops structure that holds the function for profiling.
@@ -1234,6 +1322,9 @@ int register_ftrace_function(struct ftrace_ops *ops)
 {
 	int ret;
 
+	if (unlikely(ftrace_disabled))
+		return -1;
+
 	mutex_lock(&ftrace_sysctl_lock);
 	ret = __register_ftrace_function(ops);
 	ftrace_startup();
@@ -1267,6 +1358,9 @@ ftrace_enable_sysctl(struct ctl_table *table, int write,
 {
 	int ret;
 
+	if (unlikely(ftrace_disabled))
+		return -ENODEV;
+
 	mutex_lock(&ftrace_sysctl_lock);
 
 	ret  = proc_dointvec(table, write, file, buffer, lenp, ppos);
diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c
index a6f1ed75f836..85715b86a342 100644
--- a/kernel/trace/trace_selftest.c
+++ b/kernel/trace/trace_selftest.c
@@ -248,6 +248,10 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr)
 	ftrace_enabled = save_ftrace_enabled;
 	tracer_enabled = save_tracer_enabled;
 
+	/* kill ftrace totally if we failed */
+	if (ret)
+		ftrace_kill();
+
 	return ret;
 }
 #endif /* CONFIG_FTRACE */
-- 
cgit v1.2.3


From aeaee8a2c9cb4489f166ca0e39c568e8254faaa6 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:49 +0200
Subject: ftrace: build fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 4650a3160b7f..08fbef1744cc 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -58,9 +58,6 @@ struct dyn_ftrace {
 int ftrace_force_update(void);
 void ftrace_set_filter(unsigned char *buf, int len, int reset);
 
-/* totally disable ftrace - can not re-enable after this */
-void ftrace_kill(void);
-
 /* defined in arch */
 extern int ftrace_ip_converted(unsigned long ip);
 extern unsigned char *ftrace_nop_replace(void);
@@ -74,10 +71,13 @@ extern void ftrace_caller(void);
 extern void ftrace_call(void);
 extern void mcount_call(void);
 #else
-# define ftrace_force_update() ({ 0; })
-# define ftrace_set_filter(buf, len, reset) do { } while (0)
+# define ftrace_force_update()			({ 0; })
+# define ftrace_set_filter(buf, len, reset)	do { } while (0)
 #endif
 
+/* totally disable ftrace - can not re-enable after this */
+void ftrace_kill(void);
+
 static inline void tracer_disable(void)
 {
 #ifdef CONFIG_FTRACE
-- 
cgit v1.2.3


From 86387f7ee5d3273ff4859e2c64ce656639b6ca65 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:51 +0200
Subject: ftrace: add stack tracing

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h |   2 +
 kernel/trace/Kconfig   |   1 +
 kernel/trace/trace.c   | 103 ++++++++++++++++++++++++++++++++++++++++---------
 kernel/trace/trace.h   |  11 ++++++
 4 files changed, 99 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 08fbef1744cc..0d3714e7110b 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -93,6 +93,7 @@ static inline void tracer_disable(void)
 # define CALLER_ADDR3 ((unsigned long)__builtin_return_address(3))
 # define CALLER_ADDR4 ((unsigned long)__builtin_return_address(4))
 # define CALLER_ADDR5 ((unsigned long)__builtin_return_address(5))
+# define CALLER_ADDR6 ((unsigned long)__builtin_return_address(6))
 #else
 # define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
 # define CALLER_ADDR1 0UL
@@ -100,6 +101,7 @@ static inline void tracer_disable(void)
 # define CALLER_ADDR3 0UL
 # define CALLER_ADDR4 0UL
 # define CALLER_ADDR5 0UL
+# define CALLER_ADDR6 0UL
 #endif
 
 #ifdef CONFIG_IRQSOFF_TRACER
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 3f73a1710242..eb1988ed84b7 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -10,6 +10,7 @@ config TRACER_MAX_TRACE
 config TRACING
 	bool
 	select DEBUG_FS
+	select STACKTRACE
 
 config FTRACE
 	bool "Kernel Function Tracer"
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 192c1354a7e0..b4b1b4fe99fd 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -28,6 +28,8 @@
 #include <linux/gfp.h>
 #include <linux/fs.h>
 
+#include <linux/stacktrace.h>
+
 #include "trace.h"
 
 unsigned long __read_mostly	tracing_max_latency = (cycle_t)ULONG_MAX;
@@ -88,6 +90,7 @@ enum trace_type {
 	TRACE_FN,
 	TRACE_CTX,
 	TRACE_WAKE,
+	TRACE_STACK,
 	TRACE_SPECIAL,
 
 	__TRACE_LAST_TYPE
@@ -109,6 +112,7 @@ enum trace_iterator_flags {
 	TRACE_ITER_HEX			= 0x20,
 	TRACE_ITER_BIN			= 0x40,
 	TRACE_ITER_BLOCK		= 0x80,
+	TRACE_ITER_STACKTRACE		= 0x100,
 };
 
 #define TRACE_ITER_SYM_MASK \
@@ -124,10 +128,11 @@ static const char *trace_options[] = {
 	"hex",
 	"bin",
 	"block",
+	"stacktrace",
 	NULL
 };
 
-static unsigned trace_flags;
+static unsigned trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_STACKTRACE;
 
 static DEFINE_SPINLOCK(ftrace_max_lock);
 
@@ -657,7 +662,7 @@ trace_function(struct trace_array *tr, struct trace_array_cpu *data,
 	spin_unlock_irqrestore(&data->lock, irq_flags);
 
 	if (!(trace_flags & TRACE_ITER_BLOCK))
-		wake_up (&trace_wait);
+		wake_up(&trace_wait);
 }
 
 void
@@ -685,13 +690,39 @@ trace_special(struct trace_array *tr, struct trace_array_cpu *data,
 	spin_unlock_irqrestore(&data->lock, irq_flags);
 
 	if (!(trace_flags & TRACE_ITER_BLOCK))
-		wake_up (&trace_wait);
+		wake_up(&trace_wait);
+}
+
+void __trace_stack(struct trace_array *tr,
+		   struct trace_array_cpu *data,
+		   unsigned long flags,
+		   int skip)
+{
+	struct trace_entry *entry;
+	struct stack_trace trace;
+
+	if (!(trace_flags & TRACE_ITER_STACKTRACE))
+		return;
+
+	entry			= tracing_get_trace_entry(tr, data);
+	tracing_generic_entry_update(entry, flags);
+	entry->type		= TRACE_STACK;
+
+	memset(&entry->stack, 0, sizeof(entry->stack));
+
+	trace.nr_entries	= 0;
+	trace.max_entries	= FTRACE_STACK_ENTRIES;
+	trace.skip		= skip;
+	trace.entries		= entry->stack.caller;
+
+	save_stack_trace(&trace);
 }
 
 void
 tracing_sched_switch_trace(struct trace_array *tr,
 			   struct trace_array_cpu *data,
-			   struct task_struct *prev, struct task_struct *next,
+			   struct task_struct *prev,
+			   struct task_struct *next,
 			   unsigned long flags)
 {
 	struct trace_entry *entry;
@@ -706,16 +737,18 @@ tracing_sched_switch_trace(struct trace_array *tr,
 	entry->ctx.prev_state	= prev->state;
 	entry->ctx.next_pid	= next->pid;
 	entry->ctx.next_prio	= next->prio;
+	__trace_stack(tr, data, flags, 4);
 	spin_unlock_irqrestore(&data->lock, irq_flags);
 
 	if (!(trace_flags & TRACE_ITER_BLOCK))
-		wake_up (&trace_wait);
+		wake_up(&trace_wait);
 }
 
 void
 tracing_sched_wakeup_trace(struct trace_array *tr,
 			   struct trace_array_cpu *data,
-			   struct task_struct *wakee, struct task_struct *curr,
+			   struct task_struct *wakee,
+			   struct task_struct *curr,
 			   unsigned long flags)
 {
 	struct trace_entry *entry;
@@ -730,6 +763,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr,
 	entry->ctx.prev_state	= curr->state;
 	entry->ctx.next_pid	= wakee->pid;
 	entry->ctx.next_prio	= wakee->prio;
+	__trace_stack(tr, data, flags, 5);
 	spin_unlock_irqrestore(&data->lock, irq_flags);
 
 	if (!(trace_flags & TRACE_ITER_BLOCK))
@@ -1179,6 +1213,7 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu)
 	unsigned long rel_usecs;
 	char *comm;
 	int S;
+	int i;
 
 	if (!next_entry)
 		next_entry = entry;
@@ -1197,8 +1232,10 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu)
 				 abs_usecs % 1000, rel_usecs/1000,
 				 rel_usecs % 1000);
 	} else {
-		lat_print_generic(s, entry, cpu);
-		lat_print_timestamp(s, abs_usecs, rel_usecs);
+		if (entry->type != TRACE_STACK) {
+			lat_print_generic(s, entry, cpu);
+			lat_print_timestamp(s, abs_usecs, rel_usecs);
+		}
 	}
 	switch (entry->type) {
 	case TRACE_FN:
@@ -1226,6 +1263,14 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu)
 				 entry->special.arg2,
 				 entry->special.arg3);
 		break;
+	case TRACE_STACK:
+		for (i = 0; i < FTRACE_STACK_ENTRIES; i++) {
+			if (i)
+				trace_seq_puts(s, " <= ");
+			seq_print_ip_sym(s, entry->stack.caller[i], sym_flags);
+		}
+		trace_seq_puts(s, "\n");
+		break;
 	default:
 		trace_seq_printf(s, "Unknown type %d\n", entry->type);
 	}
@@ -1241,8 +1286,9 @@ static int print_trace_fmt(struct trace_iterator *iter)
 	unsigned long long t;
 	unsigned long secs;
 	char *comm;
-	int S;
 	int ret;
+	int S;
+	int i;
 
 	entry = iter->ent;
 
@@ -1252,15 +1298,17 @@ static int print_trace_fmt(struct trace_iterator *iter)
 	usec_rem = do_div(t, 1000000ULL);
 	secs = (unsigned long)t;
 
-	ret = trace_seq_printf(s, "%16s-%-5d ", comm, entry->pid);
-	if (!ret)
-		return 0;
-	ret = trace_seq_printf(s, "[%02d] ", iter->cpu);
-	if (!ret)
-		return 0;
-	ret = trace_seq_printf(s, "%5lu.%06lu: ", secs, usec_rem);
-	if (!ret)
-		return 0;
+	if (entry->type != TRACE_STACK) {
+		ret = trace_seq_printf(s, "%16s-%-5d ", comm, entry->pid);
+		if (!ret)
+			return 0;
+		ret = trace_seq_printf(s, "[%02d] ", iter->cpu);
+		if (!ret)
+			return 0;
+		ret = trace_seq_printf(s, "%5lu.%06lu: ", secs, usec_rem);
+		if (!ret)
+			return 0;
+	}
 
 	switch (entry->type) {
 	case TRACE_FN:
@@ -1303,6 +1351,22 @@ static int print_trace_fmt(struct trace_iterator *iter)
 		if (!ret)
 			return 0;
 		break;
+	case TRACE_STACK:
+		for (i = 0; i < FTRACE_STACK_ENTRIES; i++) {
+			if (i) {
+				ret = trace_seq_puts(s, " <= ");
+				if (!ret)
+					return 0;
+			}
+			ret = seq_print_ip_sym(s, entry->stack.caller[i],
+					       sym_flags);
+			if (!ret)
+				return 0;
+		}
+		ret = trace_seq_puts(s, "\n");
+		if (!ret)
+			return 0;
+		break;
 	}
 	return 1;
 }
@@ -1344,6 +1408,7 @@ static int print_raw_fmt(struct trace_iterator *iter)
 			return 0;
 		break;
 	case TRACE_SPECIAL:
+	case TRACE_STACK:
 		ret = trace_seq_printf(s, " %lx %lx %lx\n",
 				 entry->special.arg1,
 				 entry->special.arg2,
@@ -1399,6 +1464,7 @@ static int print_hex_fmt(struct trace_iterator *iter)
 		SEQ_PUT_HEX_FIELD_RET(s, entry->fn.parent_ip);
 		break;
 	case TRACE_SPECIAL:
+	case TRACE_STACK:
 		SEQ_PUT_HEX_FIELD_RET(s, entry->special.arg1);
 		SEQ_PUT_HEX_FIELD_RET(s, entry->special.arg2);
 		SEQ_PUT_HEX_FIELD_RET(s, entry->special.arg3);
@@ -1433,6 +1499,7 @@ static int print_bin_fmt(struct trace_iterator *iter)
 		SEQ_PUT_FIELD_RET(s, entry->ctx.next_prio);
 		break;
 	case TRACE_SPECIAL:
+	case TRACE_STACK:
 		SEQ_PUT_FIELD_RET(s, entry->special.arg1);
 		SEQ_PUT_FIELD_RET(s, entry->special.arg2);
 		SEQ_PUT_FIELD_RET(s, entry->special.arg3);
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 90e0ba0f6eba..387bdcf45e28 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -34,6 +34,16 @@ struct special_entry {
 	unsigned long		arg3;
 };
 
+/*
+ * Stack-trace entry:
+ */
+
+#define FTRACE_STACK_ENTRIES	5
+
+struct stack_entry {
+	unsigned long		caller[FTRACE_STACK_ENTRIES];
+};
+
 /*
  * The trace entry - the most basic unit of tracing. This is what
  * is printed in the end as a single line in the trace output, such as:
@@ -51,6 +61,7 @@ struct trace_entry {
 		struct ftrace_entry		fn;
 		struct ctx_switch_entry		ctx;
 		struct special_entry		special;
+		struct stack_entry		stack;
 	};
 };
 
-- 
cgit v1.2.3


From 8ac0fca4ccb355ce50471d7aa3f10f5900b28b95 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:51 +0200
Subject: ftrace: sched tracer fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h             |  6 ------
 kernel/sched.c                    |  2 +-
 kernel/trace/trace_sched_wakeup.c | 13 +++----------
 3 files changed, 4 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 6e26f1fdbfe2..05744f9cb096 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2130,17 +2130,11 @@ ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next)
 #ifdef CONFIG_SCHED_TRACER
 extern void
 ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr);
-extern void
-ftrace_wake_up_new_task(struct task_struct *wakee, struct task_struct *curr);
 #else
 static inline void
 ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr)
 {
 }
-static inline void
-ftrace_wake_up_new_task(struct task_struct *wakee, struct task_struct *curr)
-{
-}
 #endif
 
 extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask);
diff --git a/kernel/sched.c b/kernel/sched.c
index 328494e28df2..53ab1174664f 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2613,7 +2613,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 		p->sched_class->task_new(rq, p);
 		inc_nr_running(rq);
 	}
-	ftrace_wake_up_new_task(p, rq->curr);
+	ftrace_wake_up_task(p, rq->curr);
 	check_preempt_curr(rq, p);
 #ifdef CONFIG_SMP
 	if (p->sched_class->task_wake_up)
diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c
index 87fa7b253b57..2a012423f9d0 100644
--- a/kernel/trace/trace_sched_wakeup.c
+++ b/kernel/trace/trace_sched_wakeup.c
@@ -201,20 +201,13 @@ out:
 	atomic_dec(&tr->data[cpu]->disabled);
 }
 
-void
-wakeup_sched_wakeup(struct task_struct *wakee, struct task_struct *curr)
+void wakeup_sched_wakeup(struct task_struct *wakee, struct task_struct *curr)
 {
 	if (likely(!tracer_enabled))
 		return;
 
-	wakeup_check_start(wakeup_trace, wakee, curr);
-}
-
-void
-ftrace_wake_up_new_task(struct task_struct *wakee, struct task_struct *curr)
-{
-	if (likely(!tracer_enabled))
-		return;
+	tracing_record_cmdline(curr);
+	tracing_record_cmdline(wakee);
 
 	wakeup_check_start(wakeup_trace, wakee, curr);
 }
-- 
cgit v1.2.3


From 4e65551905fb0300ae7e667cbaa41ee2e3f29a13 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:52 +0200
Subject: ftrace: sched tracer, trace full rbtree

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h             | 32 ++++++++++++++++-------
 kernel/sched.c                    | 35 ++++++++++++++++++++++---
 kernel/trace/trace.c              | 55 ++++++++++++++++-----------------------
 kernel/trace/trace.h              | 14 ++++++++++
 kernel/trace/trace_sched_switch.c | 24 +++++++++++------
 5 files changed, 108 insertions(+), 52 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 05744f9cb096..652d380ae563 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2119,20 +2119,34 @@ static inline void arch_pick_mmap_layout(struct mm_struct *mm)
 
 #ifdef CONFIG_CONTEXT_SWITCH_TRACER
 extern void
-ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next);
+ftrace_ctx_switch(void *rq, struct task_struct *prev, struct task_struct *next);
+extern void
+ftrace_wake_up_task(void *rq, struct task_struct *wakee,
+		    struct task_struct *curr);
+extern void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data);
+extern void
+__trace_special(void *__tr, void *__data,
+		unsigned long arg1, unsigned long arg2, unsigned long arg3);
 #else
 static inline void
-ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next)
+ftrace_ctx_switch(void *rq, struct task_struct *prev, struct task_struct *next)
+{
+}
+static inline void
+sched_trace_special(unsigned long p1, unsigned long p2, unsigned long p3)
+{
+}
+static inline void
+ftrace_wake_up_task(void *rq, struct task_struct *wakee,
+		    struct task_struct *curr)
+{
+}
+static inline void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data)
 {
 }
-#endif
-
-#ifdef CONFIG_SCHED_TRACER
-extern void
-ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr);
-#else
 static inline void
-ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr)
+__trace_special(void *__tr, void *__data,
+		unsigned long arg1, unsigned long arg2, unsigned long arg3)
 {
 }
 #endif
diff --git a/kernel/sched.c b/kernel/sched.c
index 53ab1174664f..b9208a0e33a0 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2394,6 +2394,35 @@ static int sched_balance_self(int cpu, int flag)
 
 #endif /* CONFIG_SMP */
 
+#ifdef CONFIG_CONTEXT_SWITCH_TRACER
+
+void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data)
+{
+	struct sched_entity *se;
+	struct task_struct *p;
+	struct rb_node *curr;
+	struct rq *rq = __rq;
+
+	curr = first_fair(&rq->cfs);
+	if (!curr)
+		return;
+
+	while (curr) {
+		se = rb_entry(curr, struct sched_entity, run_node);
+		if (!entity_is_task(se))
+			continue;
+
+		p = task_of(se);
+
+		__trace_special(__tr, __data,
+			      p->pid, p->se.vruntime, p->se.sum_exec_runtime);
+
+		curr = rb_next(curr);
+	}
+}
+
+#endif
+
 /***
  * try_to_wake_up - wake up a thread
  * @p: the to-be-woken-up thread
@@ -2468,7 +2497,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 
 out_activate:
 #endif /* CONFIG_SMP */
-	ftrace_wake_up_task(p, rq->curr);
+	ftrace_wake_up_task(rq, p, rq->curr);
 	schedstat_inc(p, se.nr_wakeups);
 	if (sync)
 		schedstat_inc(p, se.nr_wakeups_sync);
@@ -2613,7 +2642,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 		p->sched_class->task_new(rq, p);
 		inc_nr_running(rq);
 	}
-	ftrace_wake_up_task(p, rq->curr);
+	ftrace_wake_up_task(rq, p, rq->curr);
 	check_preempt_curr(rq, p);
 #ifdef CONFIG_SMP
 	if (p->sched_class->task_wake_up)
@@ -2786,7 +2815,7 @@ context_switch(struct rq *rq, struct task_struct *prev,
 	struct mm_struct *mm, *oldmm;
 
 	prepare_task_switch(rq, prev, next);
-	ftrace_ctx_switch(prev, next);
+	ftrace_ctx_switch(rq, prev, next);
 	mm = next->mm;
 	oldmm = prev->active_mm;
 	/*
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 0e4b7119e263..65173b14b914 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -66,7 +66,18 @@ static struct tracer		*current_trace __read_mostly;
 static int			max_tracer_type_len;
 
 static DEFINE_MUTEX(trace_types_lock);
-static DECLARE_WAIT_QUEUE_HEAD (trace_wait);
+static DECLARE_WAIT_QUEUE_HEAD(trace_wait);
+
+unsigned long trace_flags = TRACE_ITER_PRINT_PARENT;
+
+/*
+ * FIXME: where should this be called?
+ */
+void trace_wake_up(void)
+{
+	if (!(trace_flags & TRACE_ITER_BLOCK))
+		wake_up(&trace_wait);
+}
 
 #define ENTRIES_PER_PAGE (PAGE_SIZE / sizeof(struct trace_entry))
 
@@ -103,18 +114,6 @@ enum trace_flag_type {
 	TRACE_FLAG_SOFTIRQ		= 0x08,
 };
 
-enum trace_iterator_flags {
-	TRACE_ITER_PRINT_PARENT		= 0x01,
-	TRACE_ITER_SYM_OFFSET		= 0x02,
-	TRACE_ITER_SYM_ADDR		= 0x04,
-	TRACE_ITER_VERBOSE		= 0x08,
-	TRACE_ITER_RAW			= 0x10,
-	TRACE_ITER_HEX			= 0x20,
-	TRACE_ITER_BIN			= 0x40,
-	TRACE_ITER_BLOCK		= 0x80,
-	TRACE_ITER_STACKTRACE		= 0x100,
-};
-
 #define TRACE_ITER_SYM_MASK \
 	(TRACE_ITER_PRINT_PARENT|TRACE_ITER_SYM_OFFSET|TRACE_ITER_SYM_ADDR)
 
@@ -132,8 +131,6 @@ static const char *trace_options[] = {
 	NULL
 };
 
-static unsigned trace_flags = TRACE_ITER_PRINT_PARENT;
-
 static DEFINE_SPINLOCK(ftrace_max_lock);
 
 /*
@@ -660,9 +657,6 @@ trace_function(struct trace_array *tr, struct trace_array_cpu *data,
 	entry->fn.ip		= ip;
 	entry->fn.parent_ip	= parent_ip;
 	spin_unlock_irqrestore(&data->lock, irq_flags);
-
-	if (!(trace_flags & TRACE_ITER_BLOCK))
-		wake_up(&trace_wait);
 }
 
 void
@@ -673,10 +667,14 @@ ftrace(struct trace_array *tr, struct trace_array_cpu *data,
 		trace_function(tr, data, ip, parent_ip, flags);
 }
 
+#ifdef CONFIG_CONTEXT_SWITCH_TRACER
+
 void
-trace_special(struct trace_array *tr, struct trace_array_cpu *data,
-	      unsigned long arg1, unsigned long arg2, unsigned long arg3)
+__trace_special(void *__tr, void *__data,
+		unsigned long arg1, unsigned long arg2, unsigned long arg3)
 {
+	struct trace_array_cpu *data = __data;
+	struct trace_array *tr = __tr;
 	struct trace_entry *entry;
 	unsigned long irq_flags;
 
@@ -688,11 +686,10 @@ trace_special(struct trace_array *tr, struct trace_array_cpu *data,
 	entry->special.arg2	= arg2;
 	entry->special.arg3	= arg3;
 	spin_unlock_irqrestore(&data->lock, irq_flags);
-
-	if (!(trace_flags & TRACE_ITER_BLOCK))
-		wake_up(&trace_wait);
 }
 
+#endif
+
 void __trace_stack(struct trace_array *tr,
 		   struct trace_array_cpu *data,
 		   unsigned long flags,
@@ -739,9 +736,6 @@ tracing_sched_switch_trace(struct trace_array *tr,
 	entry->ctx.next_prio	= next->prio;
 	__trace_stack(tr, data, flags, 4);
 	spin_unlock_irqrestore(&data->lock, irq_flags);
-
-	if (!(trace_flags & TRACE_ITER_BLOCK))
-		wake_up(&trace_wait);
 }
 
 void
@@ -765,9 +759,6 @@ tracing_sched_wakeup_trace(struct trace_array *tr,
 	entry->ctx.next_prio	= wakee->prio;
 	__trace_stack(tr, data, flags, 5);
 	spin_unlock_irqrestore(&data->lock, irq_flags);
-
-	if (!(trace_flags & TRACE_ITER_BLOCK))
-		wake_up(&trace_wait);
 }
 
 #ifdef CONFIG_FTRACE
@@ -1258,7 +1249,7 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu)
 				 comm);
 		break;
 	case TRACE_SPECIAL:
-		trace_seq_printf(s, " %lx %lx %lx\n",
+		trace_seq_printf(s, " %ld %ld %ld\n",
 				 entry->special.arg1,
 				 entry->special.arg2,
 				 entry->special.arg3);
@@ -1344,7 +1335,7 @@ static int print_trace_fmt(struct trace_iterator *iter)
 			return 0;
 		break;
 	case TRACE_SPECIAL:
-		ret = trace_seq_printf(s, " %lx %lx %lx\n",
+		ret = trace_seq_printf(s, " %ld %ld %ld\n",
 				 entry->special.arg1,
 				 entry->special.arg2,
 				 entry->special.arg3);
@@ -1409,7 +1400,7 @@ static int print_raw_fmt(struct trace_iterator *iter)
 		break;
 	case TRACE_SPECIAL:
 	case TRACE_STACK:
-		ret = trace_seq_printf(s, " %lx %lx %lx\n",
+		ret = trace_seq_printf(s, " %ld %ld %ld\n",
 				 entry->special.arg1,
 				 entry->special.arg2,
 				 entry->special.arg3);
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 387bdcf45e28..75e237475674 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -274,4 +274,18 @@ extern int trace_selftest_startup_sched_switch(struct tracer *trace,
 
 extern void *head_page(struct trace_array_cpu *data);
 
+extern unsigned long trace_flags;
+
+enum trace_iterator_flags {
+	TRACE_ITER_PRINT_PARENT		= 0x01,
+	TRACE_ITER_SYM_OFFSET		= 0x02,
+	TRACE_ITER_SYM_ADDR		= 0x04,
+	TRACE_ITER_VERBOSE		= 0x08,
+	TRACE_ITER_RAW			= 0x10,
+	TRACE_ITER_HEX			= 0x20,
+	TRACE_ITER_BIN			= 0x40,
+	TRACE_ITER_BLOCK		= 0x80,
+	TRACE_ITER_STACKTRACE		= 0x100,
+};
+
 #endif /* _LINUX_KERNEL_TRACE_H */
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index 8b1cf1a3aee0..12658b3f2b28 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -18,7 +18,7 @@ static struct trace_array	*ctx_trace;
 static int __read_mostly	tracer_enabled;
 
 static void
-ctx_switch_func(struct task_struct *prev, struct task_struct *next)
+ctx_switch_func(void *__rq, struct task_struct *prev, struct task_struct *next)
 {
 	struct trace_array *tr = ctx_trace;
 	struct trace_array_cpu *data;
@@ -34,14 +34,17 @@ ctx_switch_func(struct task_struct *prev, struct task_struct *next)
 	data = tr->data[cpu];
 	disabled = atomic_inc_return(&data->disabled);
 
-	if (likely(disabled == 1))
+	if (likely(disabled == 1)) {
 		tracing_sched_switch_trace(tr, data, prev, next, flags);
+		ftrace_all_fair_tasks(__rq, tr, data);
+	}
 
 	atomic_dec(&data->disabled);
 	local_irq_restore(flags);
 }
 
-static void wakeup_func(struct task_struct *wakee, struct task_struct *curr)
+static void
+wakeup_func(void *__rq, struct task_struct *wakee, struct task_struct *curr)
 {
 	struct trace_array *tr = ctx_trace;
 	struct trace_array_cpu *data;
@@ -57,14 +60,18 @@ static void wakeup_func(struct task_struct *wakee, struct task_struct *curr)
 	data = tr->data[cpu];
 	disabled = atomic_inc_return(&data->disabled);
 
-	if (likely(disabled == 1))
+	if (likely(disabled == 1)) {
 		tracing_sched_wakeup_trace(tr, data, wakee, curr, flags);
+		ftrace_all_fair_tasks(__rq, tr, data);
+	}
 
 	atomic_dec(&data->disabled);
 	local_irq_restore(flags);
 }
 
-void ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next)
+void
+ftrace_ctx_switch(void *__rq, struct task_struct *prev,
+		  struct task_struct *next)
 {
 	tracing_record_cmdline(prev);
 
@@ -72,7 +79,7 @@ void ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next)
 	 * If tracer_switch_func only points to the local
 	 * switch func, it still needs the ptr passed to it.
 	 */
-	ctx_switch_func(prev, next);
+	ctx_switch_func(__rq, prev, next);
 
 	/*
 	 * Chain to the wakeup tracer (this is a NOP if disabled):
@@ -81,11 +88,12 @@ void ftrace_ctx_switch(struct task_struct *prev, struct task_struct *next)
 }
 
 void
-ftrace_wake_up_task(struct task_struct *wakee, struct task_struct *curr)
+ftrace_wake_up_task(void *__rq, struct task_struct *wakee,
+		    struct task_struct *curr)
 {
 	tracing_record_cmdline(curr);
 
-	wakeup_func(wakee, curr);
+	wakeup_func(__rq, wakee, curr);
 
 	/*
 	 * Chain to the wakeup tracer (this is a NOP if disabled):
-- 
cgit v1.2.3


From 017730c11241e26577673eb9d957cfc66172ea91 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:52 +0200
Subject: ftrace: fix wakeups

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h |  2 ++
 kernel/sched.c        | 18 ++++++++++++++++++
 kernel/trace/trace.c  | 15 +++++++++++----
 3 files changed, 31 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 652d380ae563..a3970b563757 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -246,6 +246,8 @@ extern asmlinkage void schedule_tail(struct task_struct *prev);
 extern void init_idle(struct task_struct *idle, int cpu);
 extern void init_idle_bootup_task(struct task_struct *idle);
 
+extern int runqueue_is_locked(void);
+
 extern cpumask_t nohz_cpu_mask;
 #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ)
 extern int select_nohz_load_balancer(int cpu);
diff --git a/kernel/sched.c b/kernel/sched.c
index 673b588b713b..9ca4a2e6a236 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -642,6 +642,24 @@ static inline void update_rq_clock(struct rq *rq)
 # define const_debug static const
 #endif
 
+/**
+ * runqueue_is_locked
+ *
+ * Returns true if the current cpu runqueue is locked.
+ * This interface allows printk to be called with the runqueue lock
+ * held and know whether or not it is OK to wake up the klogd.
+ */
+int runqueue_is_locked(void)
+{
+	int cpu = get_cpu();
+	struct rq *rq = cpu_rq(cpu);
+	int ret;
+
+	ret = spin_is_locked(&rq->lock);
+	put_cpu();
+	return ret;
+}
+
 /*
  * Debugging: various feature bits
  */
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 65173b14b914..2ca9d66aa74e 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -70,12 +70,13 @@ static DECLARE_WAIT_QUEUE_HEAD(trace_wait);
 
 unsigned long trace_flags = TRACE_ITER_PRINT_PARENT;
 
-/*
- * FIXME: where should this be called?
- */
 void trace_wake_up(void)
 {
-	if (!(trace_flags & TRACE_ITER_BLOCK))
+	/*
+	 * The runqueue_is_locked() can fail, but this is the best we
+	 * have for now:
+	 */
+	if (!(trace_flags & TRACE_ITER_BLOCK) && !runqueue_is_locked())
 		wake_up(&trace_wait);
 }
 
@@ -657,6 +658,8 @@ trace_function(struct trace_array *tr, struct trace_array_cpu *data,
 	entry->fn.ip		= ip;
 	entry->fn.parent_ip	= parent_ip;
 	spin_unlock_irqrestore(&data->lock, irq_flags);
+
+	trace_wake_up();
 }
 
 void
@@ -686,6 +689,8 @@ __trace_special(void *__tr, void *__data,
 	entry->special.arg2	= arg2;
 	entry->special.arg3	= arg3;
 	spin_unlock_irqrestore(&data->lock, irq_flags);
+
+	trace_wake_up();
 }
 
 #endif
@@ -759,6 +764,8 @@ tracing_sched_wakeup_trace(struct trace_array *tr,
 	entry->ctx.next_prio	= wakee->prio;
 	__trace_stack(tr, data, flags, 5);
 	spin_unlock_irqrestore(&data->lock, irq_flags);
+
+	trace_wake_up();
 }
 
 #ifdef CONFIG_FTRACE
-- 
cgit v1.2.3


From 1a3c3034336320554a3342572dae98d69e054fc7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:52 +0200
Subject: ftrace: fix __trace_special()

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h | 20 ++++++++++++--------
 kernel/trace/trace.c  |  4 ----
 2 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index a3970b563757..5b186bed54bc 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2119,6 +2119,18 @@ static inline void arch_pick_mmap_layout(struct mm_struct *mm)
 }
 #endif
 
+#ifdef CONFIG_TRACING
+extern void
+__trace_special(void *__tr, void *__data,
+		unsigned long arg1, unsigned long arg2, unsigned long arg3);
+#else
+static inline void
+__trace_special(void *__tr, void *__data,
+		unsigned long arg1, unsigned long arg2, unsigned long arg3)
+{
+}
+#endif
+
 #ifdef CONFIG_CONTEXT_SWITCH_TRACER
 extern void
 ftrace_ctx_switch(void *rq, struct task_struct *prev, struct task_struct *next);
@@ -2126,9 +2138,6 @@ extern void
 ftrace_wake_up_task(void *rq, struct task_struct *wakee,
 		    struct task_struct *curr);
 extern void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data);
-extern void
-__trace_special(void *__tr, void *__data,
-		unsigned long arg1, unsigned long arg2, unsigned long arg3);
 #else
 static inline void
 ftrace_ctx_switch(void *rq, struct task_struct *prev, struct task_struct *next)
@@ -2146,11 +2155,6 @@ ftrace_wake_up_task(void *rq, struct task_struct *wakee,
 static inline void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data)
 {
 }
-static inline void
-__trace_special(void *__tr, void *__data,
-		unsigned long arg1, unsigned long arg2, unsigned long arg3)
-{
-}
 #endif
 
 extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask);
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 2ca9d66aa74e..65d2c0a61ed4 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -670,8 +670,6 @@ ftrace(struct trace_array *tr, struct trace_array_cpu *data,
 		trace_function(tr, data, ip, parent_ip, flags);
 }
 
-#ifdef CONFIG_CONTEXT_SWITCH_TRACER
-
 void
 __trace_special(void *__tr, void *__data,
 		unsigned long arg1, unsigned long arg2, unsigned long arg3)
@@ -693,8 +691,6 @@ __trace_special(void *__tr, void *__data,
 	trace_wake_up();
 }
 
-#endif
-
 void __trace_stack(struct trace_array *tr,
 		   struct trace_array_cpu *data,
 		   unsigned long flags,
-- 
cgit v1.2.3


From 88a4216c3ec4281fc7e6725cc3a3ccd01fb1aa14 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:20:53 +0200
Subject: ftrace: sched special

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h             |  6 ++++++
 kernel/sched_fair.c               |  3 +++
 kernel/trace/trace.c              |  6 +++---
 kernel/trace/trace_sched_switch.c | 24 ++++++++++++++++++++++++
 4 files changed, 36 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5b186bed54bc..360ca99033d2 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2138,6 +2138,8 @@ extern void
 ftrace_wake_up_task(void *rq, struct task_struct *wakee,
 		    struct task_struct *curr);
 extern void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data);
+extern void
+ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3);
 #else
 static inline void
 ftrace_ctx_switch(void *rq, struct task_struct *prev, struct task_struct *next)
@@ -2155,6 +2157,10 @@ ftrace_wake_up_task(void *rq, struct task_struct *wakee,
 static inline void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data)
 {
 }
+static inline void
+ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3)
+{
+}
 #endif
 
 extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask);
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index e24ecd39c4b8..dc1856f10795 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1061,6 +1061,8 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq,
 	if (!(this_sd->flags & SD_WAKE_AFFINE))
 		return 0;
 
+	ftrace_special(__LINE__, curr->se.avg_overlap, sync);
+	ftrace_special(__LINE__, p->se.avg_overlap, -1);
 	/*
 	 * If the currently running task will sleep within
 	 * a reasonable amount of time then attract this newly
@@ -1238,6 +1240,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p)
 	if (unlikely(se == pse))
 		return;
 
+	ftrace_special(__LINE__, p->pid, se->last_wakeup);
 	cfs_rq_of(pse)->next = pse;
 
 	/*
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 3a4032492fcb..b87a26414892 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -1251,7 +1251,7 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu)
 				 comm);
 		break;
 	case TRACE_SPECIAL:
-		trace_seq_printf(s, " %ld %ld %ld\n",
+		trace_seq_printf(s, "# %ld %ld %ld\n",
 				 entry->special.arg1,
 				 entry->special.arg2,
 				 entry->special.arg3);
@@ -1335,7 +1335,7 @@ static int print_trace_fmt(struct trace_iterator *iter)
 			return 0;
 		break;
 	case TRACE_SPECIAL:
-		ret = trace_seq_printf(s, " %ld %ld %ld\n",
+		ret = trace_seq_printf(s, "# %ld %ld %ld\n",
 				 entry->special.arg1,
 				 entry->special.arg2,
 				 entry->special.arg3);
@@ -1400,7 +1400,7 @@ static int print_raw_fmt(struct trace_iterator *iter)
 		break;
 	case TRACE_SPECIAL:
 	case TRACE_STACK:
-		ret = trace_seq_printf(s, " %ld %ld %ld\n",
+		ret = trace_seq_printf(s, "# %ld %ld %ld\n",
 				 entry->special.arg1,
 				 entry->special.arg2,
 				 entry->special.arg3);
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index 5a217e863586..bddf676914ed 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -103,6 +103,30 @@ ftrace_wake_up_task(void *__rq, struct task_struct *wakee,
 	wakeup_sched_wakeup(wakee, curr);
 }
 
+void
+ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3)
+{
+	struct trace_array *tr = ctx_trace;
+	struct trace_array_cpu *data;
+	unsigned long flags;
+	long disabled;
+	int cpu;
+
+	if (!tracer_enabled)
+		return;
+
+	local_irq_save(flags);
+	cpu = raw_smp_processor_id();
+	data = tr->data[cpu];
+	disabled = atomic_inc_return(&data->disabled);
+
+	if (likely(disabled == 1))
+		__trace_special(tr, data, arg1, arg2, arg3);
+
+	atomic_dec(&data->disabled);
+	local_irq_restore(flags);
+}
+
 static void sched_switch_reset(struct trace_array *tr)
 {
 	int cpu;
-- 
cgit v1.2.3


From 3eefae994d9224fb7771a3ddb683868363c23510 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Mon, 12 May 2008 21:21:04 +0200
Subject: ftrace: limit trace entries

Currently there is no protection from the root user to use up all of
memory for trace buffers. If the root user allocates too many entries,
the OOM killer might start kill off all tasks.

This patch adds an algorith to check the following condition:

 pages_requested > (freeable_memory + current_trace_buffer_pages) / 4

If the above is met then the allocation fails. The above prevents more
than 1/4th of freeable memory from being used by trace buffers.

To determine the freeable_memory, I made determine_dirtyable_memory in
mm/page-writeback.c global.

Special thanks goes to Peter Zijlstra for suggesting the above calculation.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/writeback.h |  2 ++
 kernel/trace/trace.c      | 38 ++++++++++++++++++++++++++++++++++++++
 mm/page-writeback.c       | 10 +++++++---
 3 files changed, 47 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index f462439cc288..bd91987c065f 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -105,6 +105,8 @@ extern int vm_highmem_is_dirtyable;
 extern int block_dump;
 extern int laptop_mode;
 
+extern unsigned long determine_dirtyable_memory(void);
+
 extern int dirty_ratio_handler(struct ctl_table *table, int write,
 		struct file *filp, void __user *buffer, size_t *lenp,
 		loff_t *ppos);
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 82ced406aacf..2824cf48cdca 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -27,6 +27,7 @@
 #include <linux/poll.h>
 #include <linux/gfp.h>
 #include <linux/fs.h>
+#include <linux/writeback.h>
 
 #include <linux/stacktrace.h>
 
@@ -51,6 +52,8 @@ static int trace_free_page(void);
 
 static int tracing_disabled = 1;
 
+static unsigned long tracing_pages_allocated;
+
 long
 ns2usecs(cycle_t nsec)
 {
@@ -2591,12 +2594,41 @@ tracing_entries_write(struct file *filp, const char __user *ubuf,
 	}
 
 	if (val > global_trace.entries) {
+		long pages_requested;
+		unsigned long freeable_pages;
+
+		/* make sure we have enough memory before mapping */
+		pages_requested =
+			(val + (ENTRIES_PER_PAGE-1)) / ENTRIES_PER_PAGE;
+
+		/* account for each buffer (and max_tr) */
+		pages_requested *= tracing_nr_buffers * 2;
+
+		/* Check for overflow */
+		if (pages_requested < 0) {
+			cnt = -ENOMEM;
+			goto out;
+		}
+
+		freeable_pages = determine_dirtyable_memory();
+
+		/* we only allow to request 1/4 of useable memory */
+		if (pages_requested >
+		    ((freeable_pages + tracing_pages_allocated) / 4)) {
+			cnt = -ENOMEM;
+			goto out;
+		}
+
 		while (global_trace.entries < val) {
 			if (trace_alloc_page()) {
 				cnt = -ENOMEM;
 				goto out;
 			}
+			/* double check that we don't go over the known pages */
+			if (tracing_pages_allocated > pages_requested)
+				break;
 		}
+
 	} else {
 		/* include the number of entries in val (inc of page entries) */
 		while (global_trace.entries > val + (ENTRIES_PER_PAGE - 1))
@@ -2776,6 +2808,7 @@ static int trace_alloc_page(void)
 	struct page *page, *tmp;
 	LIST_HEAD(pages);
 	void *array;
+	unsigned pages_allocated = 0;
 	int i;
 
 	/* first allocate a page for each CPU */
@@ -2787,6 +2820,7 @@ static int trace_alloc_page(void)
 			goto free_pages;
 		}
 
+		pages_allocated++;
 		page = virt_to_page(array);
 		list_add(&page->lru, &pages);
 
@@ -2798,6 +2832,7 @@ static int trace_alloc_page(void)
 			       "for trace buffer!\n");
 			goto free_pages;
 		}
+		pages_allocated++;
 		page = virt_to_page(array);
 		list_add(&page->lru, &pages);
 #endif
@@ -2819,6 +2854,7 @@ static int trace_alloc_page(void)
 		SetPageLRU(page);
 #endif
 	}
+	tracing_pages_allocated += pages_allocated;
 	global_trace.entries += ENTRIES_PER_PAGE;
 
 	return 0;
@@ -2853,6 +2889,8 @@ static int trace_free_page(void)
 		page = list_entry(p, struct page, lru);
 		ClearPageLRU(page);
 		list_del(&page->lru);
+		tracing_pages_allocated--;
+		tracing_pages_allocated--;
 		__free_page(page);
 
 		tracing_reset(data);
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 789b6adbef37..b38f700825fc 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -126,8 +126,6 @@ static void background_writeout(unsigned long _min_pages);
 static struct prop_descriptor vm_completions;
 static struct prop_descriptor vm_dirties;
 
-static unsigned long determine_dirtyable_memory(void);
-
 /*
  * couple the period to the dirty_ratio:
  *
@@ -347,7 +345,13 @@ static unsigned long highmem_dirtyable_memory(unsigned long total)
 #endif
 }
 
-static unsigned long determine_dirtyable_memory(void)
+/**
+ * determine_dirtyable_memory - amount of memory that may be used
+ *
+ * Returns the numebr of pages that can currently be freed and used
+ * by the kernel for direct mappings.
+ */
+unsigned long determine_dirtyable_memory(void)
 {
 	unsigned long x;
 
-- 
cgit v1.2.3


From dc102a8fae2d0d6bf5223fc549247f2e23959ae6 Mon Sep 17 00:00:00 2001
From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Date: Mon, 12 May 2008 21:21:09 +0200
Subject: Markers - remove extra format argument

Denys Vlasenko <vda.linux@googlemail.com> :

> Not in this patch, but I noticed:
>
> #define __trace_mark(name, call_private, format, args...)               \
>         do {                                                            \
>                 static const char __mstrtab_##name[]                    \
>                 __attribute__((section("__markers_strings")))           \
>                 = #name "\0" format;                                    \
>                 static struct marker __mark_##name                      \
>                 __attribute__((section("__markers"), aligned(8))) =     \
>                 { __mstrtab_##name, &__mstrtab_##name[sizeof(#name)],   \
>                 0, 0, marker_probe_cb,                                  \
>                 { __mark_empty_function, NULL}, NULL };                 \
>                 __mark_check_format(format, ## args);                   \
>                 if (unlikely(__mark_##name.state)) {                    \
>                         (*__mark_##name.call)                           \
>                                 (&__mark_##name, call_private,          \
>                                 format, ## args);                       \
>                 }                                                       \
>         } while (0)
>
> In this call:
>
>                         (*__mark_##name.call)                           \
>                                 (&__mark_##name, call_private,          \
>                                 format, ## args);                       \
>
> you make gcc allocate duplicate format string. You can use
> &__mstrtab_##name[sizeof(#name)] instead since it holds the same string,
> or drop ", format," above and "const char *fmt" from here:
>
>         void (*call)(const struct marker *mdata,        /* Probe wrapper */
>                 void *call_private, const char *fmt, ...);
>
> since mdata->format is the same and all callees which need it can take it there.

Very good point. I actually thought about dropping it, since it would
remove an unnecessary argument from the stack. And actually, since I now
have the marker_probe_cb sitting between the marker site and the
callbacks, there is no API change required. Thanks :)

Mathieu

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: Denys Vlasenko <vda.linux@googlemail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/marker.h | 11 +++++------
 kernel/marker.c        | 30 ++++++++++++++----------------
 2 files changed, 19 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/marker.h b/include/linux/marker.h
index 430f6adf9762..338533abb47b 100644
--- a/include/linux/marker.h
+++ b/include/linux/marker.h
@@ -44,8 +44,8 @@ struct marker {
 				 */
 	char state;		/* Marker state. */
 	char ptype;		/* probe type : 0 : single, 1 : multi */
-	void (*call)(const struct marker *mdata,	/* Probe wrapper */
-		void *call_private, const char *fmt, ...);
+				/* Probe wrapper */
+	void (*call)(const struct marker *mdata, void *call_private, ...);
 	struct marker_probe_closure single;
 	struct marker_probe_closure *multi;
 } __attribute__((aligned(8)));
@@ -72,8 +72,7 @@ struct marker {
 		__mark_check_format(format, ## args);			\
 		if (unlikely(__mark_##name.state)) {			\
 			(*__mark_##name.call)				\
-				(&__mark_##name, call_private,		\
-				format, ## args);			\
+				(&__mark_##name, call_private, ## args);\
 		}							\
 	} while (0)
 
@@ -117,9 +116,9 @@ static inline void __printf(1, 2) ___mark_check_format(const char *fmt, ...)
 extern marker_probe_func __mark_empty_function;
 
 extern void marker_probe_cb(const struct marker *mdata,
-	void *call_private, const char *fmt, ...);
+	void *call_private, ...);
 extern void marker_probe_cb_noarg(const struct marker *mdata,
-	void *call_private, const char *fmt, ...);
+	void *call_private, ...);
 
 /*
  * Connect a probe to a marker.
diff --git a/kernel/marker.c b/kernel/marker.c
index b5a9fe1d50d5..1abfb923b761 100644
--- a/kernel/marker.c
+++ b/kernel/marker.c
@@ -55,8 +55,8 @@ static DEFINE_MUTEX(markers_mutex);
 struct marker_entry {
 	struct hlist_node hlist;
 	char *format;
-	void (*call)(const struct marker *mdata,	/* Probe wrapper */
-		void *call_private, const char *fmt, ...);
+			/* Probe wrapper */
+	void (*call)(const struct marker *mdata, void *call_private, ...);
 	struct marker_probe_closure single;
 	struct marker_probe_closure *multi;
 	int refcount;	/* Number of times armed. 0 if disarmed. */
@@ -91,15 +91,13 @@ EXPORT_SYMBOL_GPL(__mark_empty_function);
  * marker_probe_cb Callback that prepares the variable argument list for probes.
  * @mdata: pointer of type struct marker
  * @call_private: caller site private data
- * @fmt: format string
  * @...:  Variable argument list.
  *
  * Since we do not use "typical" pointer based RCU in the 1 argument case, we
  * need to put a full smp_rmb() in this branch. This is why we do not use
  * rcu_dereference() for the pointer read.
  */
-void marker_probe_cb(const struct marker *mdata, void *call_private,
-	const char *fmt, ...)
+void marker_probe_cb(const struct marker *mdata, void *call_private, ...)
 {
 	va_list args;
 	char ptype;
@@ -120,8 +118,9 @@ void marker_probe_cb(const struct marker *mdata, void *call_private,
 		/* Must read the ptr before private data. They are not data
 		 * dependant, so we put an explicit smp_rmb() here. */
 		smp_rmb();
-		va_start(args, fmt);
-		func(mdata->single.probe_private, call_private, fmt, &args);
+		va_start(args, call_private);
+		func(mdata->single.probe_private, call_private, mdata->format,
+			&args);
 		va_end(args);
 	} else {
 		struct marker_probe_closure *multi;
@@ -136,9 +135,9 @@ void marker_probe_cb(const struct marker *mdata, void *call_private,
 		smp_read_barrier_depends();
 		multi = mdata->multi;
 		for (i = 0; multi[i].func; i++) {
-			va_start(args, fmt);
-			multi[i].func(multi[i].probe_private, call_private, fmt,
-				&args);
+			va_start(args, call_private);
+			multi[i].func(multi[i].probe_private, call_private,
+				mdata->format, &args);
 			va_end(args);
 		}
 	}
@@ -150,13 +149,11 @@ EXPORT_SYMBOL_GPL(marker_probe_cb);
  * marker_probe_cb Callback that does not prepare the variable argument list.
  * @mdata: pointer of type struct marker
  * @call_private: caller site private data
- * @fmt: format string
  * @...:  Variable argument list.
  *
  * Should be connected to markers "MARK_NOARGS".
  */
-void marker_probe_cb_noarg(const struct marker *mdata,
-	void *call_private, const char *fmt, ...)
+void marker_probe_cb_noarg(const struct marker *mdata, void *call_private, ...)
 {
 	va_list args;	/* not initialized */
 	char ptype;
@@ -172,7 +169,8 @@ void marker_probe_cb_noarg(const struct marker *mdata,
 		/* Must read the ptr before private data. They are not data
 		 * dependant, so we put an explicit smp_rmb() here. */
 		smp_rmb();
-		func(mdata->single.probe_private, call_private, fmt, &args);
+		func(mdata->single.probe_private, call_private, mdata->format,
+			&args);
 	} else {
 		struct marker_probe_closure *multi;
 		int i;
@@ -186,8 +184,8 @@ void marker_probe_cb_noarg(const struct marker *mdata,
 		smp_read_barrier_depends();
 		multi = mdata->multi;
 		for (i = 0; multi[i].func; i++)
-			multi[i].func(multi[i].probe_private, call_private, fmt,
-				&args);
+			multi[i].func(multi[i].probe_private, call_private,
+				mdata->format, &args);
 	}
 	preempt_enable();
 }
-- 
cgit v1.2.3


From 0aa977f592f17004f9d1d545f2e1bb9ea71896c3 Mon Sep 17 00:00:00 2001
From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Date: Mon, 12 May 2008 21:21:10 +0200
Subject: Markers - define non optimized marker

To support the forthcoming "immediate values" marker optimization, we must have
a way to declare markers in few code paths that does not use instruction
modification based enable. This will be the case of printk(), some traps and
eventually lockdep instrumentation.

Changelog :
- Fix reversed boolean logic of "generic".

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/marker.h | 29 ++++++++++++++++++++++++-----
 1 file changed, 24 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/marker.h b/include/linux/marker.h
index 338533abb47b..1290653f9241 100644
--- a/include/linux/marker.h
+++ b/include/linux/marker.h
@@ -58,8 +58,12 @@ struct marker {
  * Make sure the alignment of the structure in the __markers section will
  * not add unwanted padding between the beginning of the section and the
  * structure. Force alignment to the same alignment as the section start.
+ *
+ * The "generic" argument controls which marker enabling mechanism must be used.
+ * If generic is true, a variable read is used.
+ * If generic is false, immediate values are used.
  */
-#define __trace_mark(name, call_private, format, args...)		\
+#define __trace_mark(generic, name, call_private, format, args...)	\
 	do {								\
 		static const char __mstrtab_##name[]			\
 		__attribute__((section("__markers_strings")))		\
@@ -79,7 +83,7 @@ struct marker {
 extern void marker_update_probe_range(struct marker *begin,
 	struct marker *end);
 #else /* !CONFIG_MARKERS */
-#define __trace_mark(name, call_private, format, args...) \
+#define __trace_mark(generic, name, call_private, format, args...) \
 		__mark_check_format(format, ## args)
 static inline void marker_update_probe_range(struct marker *begin,
 	struct marker *end)
@@ -87,15 +91,30 @@ static inline void marker_update_probe_range(struct marker *begin,
 #endif /* CONFIG_MARKERS */
 
 /**
- * trace_mark - Marker
+ * trace_mark - Marker using code patching
  * @name: marker name, not quoted.
  * @format: format string
  * @args...: variable argument list
  *
- * Places a marker.
+ * Places a marker using optimized code patching technique (imv_read())
+ * to be enabled when immediate values are present.
  */
 #define trace_mark(name, format, args...) \
-	__trace_mark(name, NULL, format, ## args)
+	__trace_mark(0, name, NULL, format, ## args)
+
+/**
+ * _trace_mark - Marker using variable read
+ * @name: marker name, not quoted.
+ * @format: format string
+ * @args...: variable argument list
+ *
+ * Places a marker using a standard memory read (_imv_read()) to be
+ * enabled. Should be used for markers in code paths where instruction
+ * modification based enabling is not welcome. (__init and __exit functions,
+ * lockdep, some traps, printk).
+ */
+#define _trace_mark(name, format, args...) \
+	__trace_mark(1, name, NULL, format, ## args)
 
 /**
  * MARK_NOARGS - Format string for a marker with no argument.
-- 
cgit v1.2.3


From 5b82a1b08a00b2adca3d9dd9777efff40b7aaaa1 Mon Sep 17 00:00:00 2001
From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Date: Mon, 12 May 2008 21:21:10 +0200
Subject: Port ftrace to markers

Porting ftrace to the marker infrastructure.

Don't need to chain to the wakeup tracer from the sched tracer, because markers
support multiple probes connected.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h             |  32 -------
 kernel/sched.c                    |  14 +++-
 kernel/trace/trace.h              |  20 +----
 kernel/trace/trace_sched_switch.c | 171 +++++++++++++++++++++++++++++++-------
 kernel/trace/trace_sched_wakeup.c | 106 +++++++++++++++++++++--
 5 files changed, 255 insertions(+), 88 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 360ca99033d2..c0b1c69b55ce 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2131,38 +2131,6 @@ __trace_special(void *__tr, void *__data,
 }
 #endif
 
-#ifdef CONFIG_CONTEXT_SWITCH_TRACER
-extern void
-ftrace_ctx_switch(void *rq, struct task_struct *prev, struct task_struct *next);
-extern void
-ftrace_wake_up_task(void *rq, struct task_struct *wakee,
-		    struct task_struct *curr);
-extern void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data);
-extern void
-ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3);
-#else
-static inline void
-ftrace_ctx_switch(void *rq, struct task_struct *prev, struct task_struct *next)
-{
-}
-static inline void
-sched_trace_special(unsigned long p1, unsigned long p2, unsigned long p3)
-{
-}
-static inline void
-ftrace_wake_up_task(void *rq, struct task_struct *wakee,
-		    struct task_struct *curr)
-{
-}
-static inline void ftrace_all_fair_tasks(void *__rq, void *__tr, void *__data)
-{
-}
-static inline void
-ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3)
-{
-}
-#endif
-
 extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask);
 extern long sched_getaffinity(pid_t pid, cpumask_t *mask);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index ad95cca4e42e..e2e985eeee78 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2500,7 +2500,9 @@ out_activate:
 	success = 1;
 
 out_running:
-	ftrace_wake_up_task(rq, p, rq->curr);
+	trace_mark(kernel_sched_wakeup,
+		"pid %d state %ld ## rq %p task %p rq->curr %p",
+		p->pid, p->state, rq, p, rq->curr);
 	check_preempt_curr(rq, p);
 
 	p->state = TASK_RUNNING;
@@ -2631,7 +2633,9 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 		p->sched_class->task_new(rq, p);
 		inc_nr_running(rq);
 	}
-	ftrace_wake_up_task(rq, p, rq->curr);
+	trace_mark(kernel_sched_wakeup_new,
+		"pid %d state %ld ## rq %p task %p rq->curr %p",
+		p->pid, p->state, rq, p, rq->curr);
 	check_preempt_curr(rq, p);
 #ifdef CONFIG_SMP
 	if (p->sched_class->task_wake_up)
@@ -2804,7 +2808,11 @@ context_switch(struct rq *rq, struct task_struct *prev,
 	struct mm_struct *mm, *oldmm;
 
 	prepare_task_switch(rq, prev, next);
-	ftrace_ctx_switch(rq, prev, next);
+	trace_mark(kernel_sched_schedule,
+		"prev_pid %d next_pid %d prev_state %ld "
+		"## rq %p prev %p next %p",
+		prev->pid, next->pid, prev->state,
+		rq, prev, next);
 	mm = next->mm;
 	oldmm = prev->active_mm;
 	/*
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 8845033ab49d..f5de0601b408 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -234,25 +234,10 @@ void update_max_tr_single(struct trace_array *tr,
 
 extern cycle_t ftrace_now(int cpu);
 
-#ifdef CONFIG_SCHED_TRACER
-extern void
-wakeup_sched_switch(struct task_struct *prev, struct task_struct *next);
-extern void
-wakeup_sched_wakeup(struct task_struct *wakee, struct task_struct *curr);
-#else
-static inline void
-wakeup_sched_switch(struct task_struct *prev, struct task_struct *next)
-{
-}
-static inline void
-wakeup_sched_wakeup(struct task_struct *wakee, struct task_struct *curr)
-{
-}
-#endif
-
 #ifdef CONFIG_CONTEXT_SWITCH_TRACER
 typedef void
 (*tracer_switch_func_t)(void *private,
+			void *__rq,
 			struct task_struct *prev,
 			struct task_struct *next);
 
@@ -262,9 +247,6 @@ struct tracer_switch_ops {
 	struct tracer_switch_ops	*next;
 };
 
-extern int register_tracer_switch(struct tracer_switch_ops *ops);
-extern int unregister_tracer_switch(struct tracer_switch_ops *ops);
-
 #endif /* CONFIG_CONTEXT_SWITCH_TRACER */
 
 #ifdef CONFIG_DYNAMIC_FTRACE
diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c
index a3376478fc2c..d25ffa5eaf2b 100644
--- a/kernel/trace/trace_sched_switch.c
+++ b/kernel/trace/trace_sched_switch.c
@@ -16,11 +16,14 @@
 
 static struct trace_array	*ctx_trace;
 static int __read_mostly	tracer_enabled;
+static atomic_t			sched_ref;
 
 static void
-ctx_switch_func(void *__rq, struct task_struct *prev, struct task_struct *next)
+sched_switch_func(void *private, void *__rq, struct task_struct *prev,
+			struct task_struct *next)
 {
-	struct trace_array *tr = ctx_trace;
+	struct trace_array **ptr = private;
+	struct trace_array *tr = *ptr;
 	struct trace_array_cpu *data;
 	unsigned long flags;
 	long disabled;
@@ -41,10 +44,40 @@ ctx_switch_func(void *__rq, struct task_struct *prev, struct task_struct *next)
 	local_irq_restore(flags);
 }
 
+static notrace void
+sched_switch_callback(void *probe_data, void *call_data,
+		      const char *format, va_list *args)
+{
+	struct task_struct *prev;
+	struct task_struct *next;
+	struct rq *__rq;
+
+	if (!atomic_read(&sched_ref))
+		return;
+
+	/* skip prev_pid %d next_pid %d prev_state %ld */
+	(void)va_arg(*args, int);
+	(void)va_arg(*args, int);
+	(void)va_arg(*args, long);
+	__rq = va_arg(*args, typeof(__rq));
+	prev = va_arg(*args, typeof(prev));
+	next = va_arg(*args, typeof(next));
+
+	tracing_record_cmdline(prev);
+
+	/*
+	 * If tracer_switch_func only points to the local
+	 * switch func, it still needs the ptr passed to it.
+	 */
+	sched_switch_func(probe_data, __rq, prev, next);
+}
+
 static void
-wakeup_func(void *__rq, struct task_struct *wakee, struct task_struct *curr)
+wakeup_func(void *private, void *__rq, struct task_struct *wakee, struct
+			task_struct *curr)
 {
-	struct trace_array *tr = ctx_trace;
+	struct trace_array **ptr = private;
+	struct trace_array *tr = *ptr;
 	struct trace_array_cpu *data;
 	unsigned long flags;
 	long disabled;
@@ -67,35 +100,29 @@ wakeup_func(void *__rq, struct task_struct *wakee, struct task_struct *curr)
 	local_irq_restore(flags);
 }
 
-void
-ftrace_ctx_switch(void *__rq, struct task_struct *prev,
-		  struct task_struct *next)
+static notrace void
+wake_up_callback(void *probe_data, void *call_data,
+		 const char *format, va_list *args)
 {
-	if (unlikely(atomic_read(&trace_record_cmdline_enabled)))
-		tracing_record_cmdline(prev);
+	struct task_struct *curr;
+	struct task_struct *task;
+	struct rq *__rq;
 
-	/*
-	 * If tracer_switch_func only points to the local
-	 * switch func, it still needs the ptr passed to it.
-	 */
-	ctx_switch_func(__rq, prev, next);
+	if (likely(!tracer_enabled))
+		return;
 
-	/*
-	 * Chain to the wakeup tracer (this is a NOP if disabled):
-	 */
-	wakeup_sched_switch(prev, next);
-}
+	/* Skip pid %d state %ld */
+	(void)va_arg(*args, int);
+	(void)va_arg(*args, long);
+	/* now get the meat: "rq %p task %p rq->curr %p" */
+	__rq = va_arg(*args, typeof(__rq));
+	task = va_arg(*args, typeof(task));
+	curr = va_arg(*args, typeof(curr));
 
-void
-ftrace_wake_up_task(void *__rq, struct task_struct *wakee,
-		    struct task_struct *curr)
-{
-	wakeup_func(__rq, wakee, curr);
+	tracing_record_cmdline(task);
+	tracing_record_cmdline(curr);
 
-	/*
-	 * Chain to the wakeup tracer (this is a NOP if disabled):
-	 */
-	wakeup_sched_wakeup(wakee, curr);
+	wakeup_func(probe_data, __rq, task, curr);
 }
 
 void
@@ -132,15 +159,95 @@ static void sched_switch_reset(struct trace_array *tr)
 		tracing_reset(tr->data[cpu]);
 }
 
+static int tracing_sched_register(void)
+{
+	int ret;
+
+	ret = marker_probe_register("kernel_sched_wakeup",
+			"pid %d state %ld ## rq %p task %p rq->curr %p",
+			wake_up_callback,
+			&ctx_trace);
+	if (ret) {
+		pr_info("wakeup trace: Couldn't add marker"
+			" probe to kernel_sched_wakeup\n");
+		return ret;
+	}
+
+	ret = marker_probe_register("kernel_sched_wakeup_new",
+			"pid %d state %ld ## rq %p task %p rq->curr %p",
+			wake_up_callback,
+			&ctx_trace);
+	if (ret) {
+		pr_info("wakeup trace: Couldn't add marker"
+			" probe to kernel_sched_wakeup_new\n");
+		goto fail_deprobe;
+	}
+
+	ret = marker_probe_register("kernel_sched_schedule",
+		"prev_pid %d next_pid %d prev_state %ld "
+		"## rq %p prev %p next %p",
+		sched_switch_callback,
+		&ctx_trace);
+	if (ret) {
+		pr_info("sched trace: Couldn't add marker"
+			" probe to kernel_sched_schedule\n");
+		goto fail_deprobe_wake_new;
+	}
+
+	return ret;
+fail_deprobe_wake_new:
+	marker_probe_unregister("kernel_sched_wakeup_new",
+				wake_up_callback,
+				&ctx_trace);
+fail_deprobe:
+	marker_probe_unregister("kernel_sched_wakeup",
+				wake_up_callback,
+				&ctx_trace);
+	return ret;
+}
+
+static void tracing_sched_unregister(void)
+{
+	marker_probe_unregister("kernel_sched_schedule",
+				sched_switch_callback,
+				&ctx_trace);
+	marker_probe_unregister("kernel_sched_wakeup_new",
+				wake_up_callback,
+				&ctx_trace);
+	marker_probe_unregister("kernel_sched_wakeup",
+				wake_up_callback,
+				&ctx_trace);
+}
+
+void tracing_start_sched_switch(void)
+{
+	long ref;
+
+	ref = atomic_inc_return(&sched_ref);
+	if (ref == 1)
+		tracing_sched_register();
+}
+
+void tracing_stop_sched_switch(void)
+{
+	long ref;
+
+	ref = atomic_dec_and_test(&sched_ref);
+	if (ref)
+		tracing_sched_unregister();
+}
+
 static void start_sched_trace(struct trace_array *tr)
 {
 	sched_switch_reset(tr);
 	atomic_inc(&trace_record_cmdline_enabled);
 	tracer_enabled = 1;
+	tracing_start_sched_switch();
 }
 
 static void stop_sched_trace(struct trace_array *tr)
 {
+	tracing_stop_sched_switch();
 	atomic_dec(&trace_record_cmdline_enabled);
 	tracer_enabled = 0;
 }
@@ -181,6 +288,14 @@ static struct tracer sched_switch_trace __read_mostly =
 
 __init static int init_sched_switch_trace(void)
 {
+	int ret = 0;
+
+	if (atomic_read(&sched_ref))
+		ret = tracing_sched_register();
+	if (ret) {
+		pr_info("error registering scheduler trace\n");
+		return ret;
+	}
 	return register_tracer(&sched_switch_trace);
 }
 device_initcall(init_sched_switch_trace);
diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c
index 5948011006bc..5d2fb48e47f8 100644
--- a/kernel/trace/trace_sched_wakeup.c
+++ b/kernel/trace/trace_sched_wakeup.c
@@ -15,6 +15,7 @@
 #include <linux/kallsyms.h>
 #include <linux/uaccess.h>
 #include <linux/ftrace.h>
+#include <linux/marker.h>
 
 #include "trace.h"
 
@@ -44,11 +45,13 @@ static int report_latency(cycle_t delta)
 	return 1;
 }
 
-void
-wakeup_sched_switch(struct task_struct *prev, struct task_struct *next)
+static void notrace
+wakeup_sched_switch(void *private, void *rq, struct task_struct *prev,
+	struct task_struct *next)
 {
 	unsigned long latency = 0, t0 = 0, t1 = 0;
-	struct trace_array *tr = wakeup_trace;
+	struct trace_array **ptr = private;
+	struct trace_array *tr = *ptr;
 	struct trace_array_cpu *data;
 	cycle_t T0, T1, delta;
 	unsigned long flags;
@@ -113,6 +116,31 @@ out:
 	atomic_dec(&tr->data[cpu]->disabled);
 }
 
+static notrace void
+sched_switch_callback(void *probe_data, void *call_data,
+		      const char *format, va_list *args)
+{
+	struct task_struct *prev;
+	struct task_struct *next;
+	struct rq *__rq;
+
+	/* skip prev_pid %d next_pid %d prev_state %ld */
+	(void)va_arg(*args, int);
+	(void)va_arg(*args, int);
+	(void)va_arg(*args, long);
+	__rq = va_arg(*args, typeof(__rq));
+	prev = va_arg(*args, typeof(prev));
+	next = va_arg(*args, typeof(next));
+
+	tracing_record_cmdline(prev);
+
+	/*
+	 * If tracer_switch_func only points to the local
+	 * switch func, it still needs the ptr passed to it.
+	 */
+	wakeup_sched_switch(probe_data, __rq, prev, next);
+}
+
 static void __wakeup_reset(struct trace_array *tr)
 {
 	struct trace_array_cpu *data;
@@ -188,19 +216,68 @@ out:
 	atomic_dec(&tr->data[cpu]->disabled);
 }
 
-void wakeup_sched_wakeup(struct task_struct *wakee, struct task_struct *curr)
+static notrace void
+wake_up_callback(void *probe_data, void *call_data,
+		 const char *format, va_list *args)
 {
+	struct trace_array **ptr = probe_data;
+	struct trace_array *tr = *ptr;
+	struct task_struct *curr;
+	struct task_struct *task;
+	struct rq *__rq;
+
 	if (likely(!tracer_enabled))
 		return;
 
+	/* Skip pid %d state %ld */
+	(void)va_arg(*args, int);
+	(void)va_arg(*args, long);
+	/* now get the meat: "rq %p task %p rq->curr %p" */
+	__rq = va_arg(*args, typeof(__rq));
+	task = va_arg(*args, typeof(task));
+	curr = va_arg(*args, typeof(curr));
+
+	tracing_record_cmdline(task);
 	tracing_record_cmdline(curr);
-	tracing_record_cmdline(wakee);
 
-	wakeup_check_start(wakeup_trace, wakee, curr);
+	wakeup_check_start(tr, task, curr);
 }
 
 static void start_wakeup_tracer(struct trace_array *tr)
 {
+	int ret;
+
+	ret = marker_probe_register("kernel_sched_wakeup",
+			"pid %d state %ld ## rq %p task %p rq->curr %p",
+			wake_up_callback,
+			&wakeup_trace);
+	if (ret) {
+		pr_info("wakeup trace: Couldn't add marker"
+			" probe to kernel_sched_wakeup\n");
+		return;
+	}
+
+	ret = marker_probe_register("kernel_sched_wakeup_new",
+			"pid %d state %ld ## rq %p task %p rq->curr %p",
+			wake_up_callback,
+			&wakeup_trace);
+	if (ret) {
+		pr_info("wakeup trace: Couldn't add marker"
+			" probe to kernel_sched_wakeup_new\n");
+		goto fail_deprobe;
+	}
+
+	ret = marker_probe_register("kernel_sched_schedule",
+		"prev_pid %d next_pid %d prev_state %ld "
+		"## rq %p prev %p next %p",
+		sched_switch_callback,
+		&wakeup_trace);
+	if (ret) {
+		pr_info("sched trace: Couldn't add marker"
+			" probe to kernel_sched_schedule\n");
+		goto fail_deprobe_wake_new;
+	}
+
 	wakeup_reset(tr);
 
 	/*
@@ -215,11 +292,28 @@ static void start_wakeup_tracer(struct trace_array *tr)
 	tracer_enabled = 1;
 
 	return;
+fail_deprobe_wake_new:
+	marker_probe_unregister("kernel_sched_wakeup_new",
+				wake_up_callback,
+				&wakeup_trace);
+fail_deprobe:
+	marker_probe_unregister("kernel_sched_wakeup",
+				wake_up_callback,
+				&wakeup_trace);
 }
 
 static void stop_wakeup_tracer(struct trace_array *tr)
 {
 	tracer_enabled = 0;
+	marker_probe_unregister("kernel_sched_schedule",
+				sched_switch_callback,
+				&wakeup_trace);
+	marker_probe_unregister("kernel_sched_wakeup_new",
+				wake_up_callback,
+				&wakeup_trace);
+	marker_probe_unregister("kernel_sched_wakeup",
+				wake_up_callback,
+				&wakeup_trace);
 }
 
 static void wakeup_tracer_init(struct trace_array *tr)
-- 
cgit v1.2.3


From 74f4e369fc5b52433ad824cef32d3bf1304549be Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:21:15 +0200
Subject: ftrace: stacktrace fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h | 8 ++++++++
 kernel/semaphore.c     | 2 ++
 kernel/trace/trace.c   | 4 ++--
 kernel/trace/trace.h   | 2 +-
 4 files changed, 13 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 0d3714e7110b..017ab44d572a 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -120,4 +120,12 @@ static inline void tracer_disable(void)
 # define trace_preempt_off(a0, a1)		do { } while (0)
 #endif
 
+#ifdef CONFIG_CONTEXT_SWITCH_TRACER
+extern void
+ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3);
+#else
+static inline void
+ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) { }
+#endif
+
 #endif /* _LINUX_FTRACE_H */
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index 5c2942e768cd..1a064adab658 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -31,6 +31,7 @@
 #include <linux/sched.h>
 #include <linux/semaphore.h>
 #include <linux/spinlock.h>
+#include <linux/ftrace.h>
 
 static noinline void __down(struct semaphore *sem);
 static noinline int __down_interruptible(struct semaphore *sem);
@@ -53,6 +54,7 @@ void down(struct semaphore *sem)
 {
 	unsigned long flags;
 
+	ftrace_special(sem->count, 0, __LINE__);
 	spin_lock_irqsave(&sem->lock, flags);
 	if (likely(sem->count > 0))
 		sem->count--;
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 2824cf48cdca..3271916ff033 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -901,7 +901,7 @@ tracing_sched_switch_trace(struct trace_array *tr,
 	entry->ctx.next_pid	= next->pid;
 	entry->ctx.next_prio	= next->prio;
 	entry->ctx.next_state	= next->state;
-	__trace_stack(tr, data, flags, 4);
+	__trace_stack(tr, data, flags, 5);
 	__raw_spin_unlock(&data->lock);
 	raw_local_irq_restore(irq_flags);
 }
@@ -927,7 +927,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr,
 	entry->ctx.next_pid	= wakee->pid;
 	entry->ctx.next_prio	= wakee->prio;
 	entry->ctx.next_state	= wakee->state;
-	__trace_stack(tr, data, flags, 5);
+	__trace_stack(tr, data, flags, 6);
 	__raw_spin_unlock(&data->lock);
 	raw_local_irq_restore(irq_flags);
 
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index f5de0601b408..c460e85e94ed 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -51,7 +51,7 @@ struct special_entry {
  * Stack-trace entry:
  */
 
-#define FTRACE_STACK_ENTRIES	5
+#define FTRACE_STACK_ENTRIES	8
 
 struct stack_entry {
 	unsigned long		caller[FTRACE_STACK_ENTRIES];
-- 
cgit v1.2.3


From d49dbf33f0bf8748ee3662b973eb57e60525d622 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 16 May 2008 10:41:53 +0200
Subject: ftrace: fix include file dependency

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 017ab44d572a..911d5d80b49f 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -4,6 +4,7 @@
 #ifdef CONFIG_FTRACE
 
 #include <linux/linkage.h>
+#include <linux/fs.h>
 
 extern int ftrace_enabled;
 extern int
-- 
cgit v1.2.3


From 489f139614596cbc956a06f5e4bb41288e276fe3 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 25 Feb 2008 13:38:05 +0100
Subject: ftrace: fix build bug

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 911d5d80b49f..922e23d0196f 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -106,16 +106,16 @@ static inline void tracer_disable(void)
 #endif
 
 #ifdef CONFIG_IRQSOFF_TRACER
-  extern void notrace time_hardirqs_on(unsigned long a0, unsigned long a1);
-  extern void notrace time_hardirqs_off(unsigned long a0, unsigned long a1);
+  extern void time_hardirqs_on(unsigned long a0, unsigned long a1);
+  extern void time_hardirqs_off(unsigned long a0, unsigned long a1);
 #else
 # define time_hardirqs_on(a0, a1)		do { } while (0)
 # define time_hardirqs_off(a0, a1)		do { } while (0)
 #endif
 
 #ifdef CONFIG_PREEMPT_TRACER
-  extern void notrace trace_preempt_on(unsigned long a0, unsigned long a1);
-  extern void notrace trace_preempt_off(unsigned long a0, unsigned long a1);
+  extern void trace_preempt_on(unsigned long a0, unsigned long a1);
+  extern void trace_preempt_off(unsigned long a0, unsigned long a1);
 #else
 # define trace_preempt_on(a0, a1)		do { } while (0)
 # define trace_preempt_off(a0, a1)		do { } while (0)
-- 
cgit v1.2.3


From 8b7d89d02ef3c6a7c73d6596f28cea7632850af4 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:20:56 +0200
Subject: x86: mmiotrace - trace memory mapped IO

Mmiotrace is a tool for trapping memory mapped IO (MMIO) accesses within
the kernel. It is used for debugging and especially for reverse
engineering evil binary drivers.

Mmiotrace works by wrapping the ioremap family of kernel functions and
marking the returned pages as not present. Access to the IO memory
triggers a page fault, which will be handled by mmiotrace's custom page
fault handler. This will single-step the faulted instruction with the
MMIO page marked as present. Access logs are directed to user space via
relay and debug_fs.

This page fault approach is necessary, because binary drivers have
readl/writel etc. calls inlined and therefore extremely difficult to
trap with with e.g. kprobes.

This patch depends on the custom page fault handlers patch.

Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig.debug                    |  27 ++
 arch/x86/kernel/Makefile                  |   2 +
 arch/x86/kernel/init_task.c               |   1 +
 arch/x86/kernel/mmiotrace/Makefile        |   4 +
 arch/x86/kernel/mmiotrace/kmmio.c         | 391 ++++++++++++++++++++++
 arch/x86/kernel/mmiotrace/kmmio.h         |  58 ++++
 arch/x86/kernel/mmiotrace/mmio-mod.c      | 527 ++++++++++++++++++++++++++++++
 arch/x86/kernel/mmiotrace/pf_in.c         | 489 +++++++++++++++++++++++++++
 arch/x86/kernel/mmiotrace/pf_in.h         |  39 +++
 arch/x86/kernel/mmiotrace/testmmiotrace.c |  77 +++++
 include/linux/mmiotrace.h                 |  62 ++++
 11 files changed, 1677 insertions(+)
 create mode 100644 arch/x86/kernel/mmiotrace/Makefile
 create mode 100644 arch/x86/kernel/mmiotrace/kmmio.c
 create mode 100644 arch/x86/kernel/mmiotrace/kmmio.h
 create mode 100644 arch/x86/kernel/mmiotrace/mmio-mod.c
 create mode 100644 arch/x86/kernel/mmiotrace/pf_in.c
 create mode 100644 arch/x86/kernel/mmiotrace/pf_in.h
 create mode 100644 arch/x86/kernel/mmiotrace/testmmiotrace.c
 create mode 100644 include/linux/mmiotrace.h

(limited to 'include/linux')

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 9431a8399844..7c6496e2225e 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -176,6 +176,33 @@ config PAGE_FAULT_HANDLERS
 	  register a function that is called on every page fault. Custom
 	  handlers are used by some debugging and reverse engineering tools.
 
+config MMIOTRACE
+	tristate "Memory mapped IO tracing"
+	depends on DEBUG_KERNEL && PAGE_FAULT_HANDLERS && RELAY && DEBUG_FS
+	default n
+	help
+	  This will build a kernel module called mmiotrace.
+
+	  Mmiotrace traces Memory Mapped I/O access and is meant for debugging
+	  and reverse engineering. The kernel module offers wrapped
+	  versions of the ioremap family of functions. The driver to be traced
+	  must be modified to call these wrappers. A user space program is
+	  required to collect the MMIO data.
+
+	  See http://nouveau.freedesktop.org/wiki/MmioTrace
+	  If you are not helping to develop drivers, say N.
+
+config MMIOTRACE_TEST
+	tristate "Test module for mmiotrace"
+	depends on MMIOTRACE && m
+	default n
+	help
+	  This is a dumb module for testing mmiotrace. It is very dangerous
+	  as it will write garbage to IO memory starting at a given address.
+	  However, it should be safe to use on e.g. unused portion of VRAM.
+
+	  Say N, unless you absolutely know what you are doing.
+
 #
 # IO delay types:
 #
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 739d49acd2f1..a51ac153685e 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -79,6 +79,8 @@ obj-$(CONFIG_KGDB)		+= kgdb.o
 obj-$(CONFIG_VM86)		+= vm86_32.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 
+obj-$(CONFIG_MMIOTRACE)		+= mmiotrace/
+
 obj-$(CONFIG_HPET_TIMER) 	+= hpet.o
 
 obj-$(CONFIG_K8_NB)		+= k8.o
diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index a4f93b4120c1..027a5b6a12b2 100644
--- a/arch/x86/kernel/init_task.c
+++ b/arch/x86/kernel/init_task.c
@@ -15,6 +15,7 @@ static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
 EXPORT_UNUSED_SYMBOL(init_mm); /* will be removed in 2.6.26 */
+EXPORT_SYMBOL_GPL(init_mm);
 
 /*
  * Initial thread structure.
diff --git a/arch/x86/kernel/mmiotrace/Makefile b/arch/x86/kernel/mmiotrace/Makefile
new file mode 100644
index 000000000000..d6905f7f981b
--- /dev/null
+++ b/arch/x86/kernel/mmiotrace/Makefile
@@ -0,0 +1,4 @@
+obj-$(CONFIG_MMIOTRACE) += mmiotrace.o
+mmiotrace-objs := pf_in.o kmmio.o mmio-mod.o
+
+obj-$(CONFIG_MMIOTRACE_TEST) += testmmiotrace.o
diff --git a/arch/x86/kernel/mmiotrace/kmmio.c b/arch/x86/kernel/mmiotrace/kmmio.c
new file mode 100644
index 000000000000..8ba48f9c91b4
--- /dev/null
+++ b/arch/x86/kernel/mmiotrace/kmmio.c
@@ -0,0 +1,391 @@
+/* Support for MMIO probes.
+ * Benfit many code from kprobes
+ * (C) 2002 Louis Zhuang <louis.zhuang@intel.com>.
+ *     2007 Alexander Eichner
+ *     2008 Pekka Paalanen <pq@iki.fi>
+ */
+
+#include <linux/version.h>
+#include <linux/spinlock.h>
+#include <linux/hash.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/uaccess.h>
+#include <linux/ptrace.h>
+#include <linux/preempt.h>
+#include <asm/io.h>
+#include <asm/cacheflush.h>
+#include <asm/errno.h>
+#include <asm/tlbflush.h>
+
+#include "kmmio.h"
+
+#define KMMIO_HASH_BITS 6
+#define KMMIO_TABLE_SIZE (1 << KMMIO_HASH_BITS)
+#define KMMIO_PAGE_HASH_BITS 4
+#define KMMIO_PAGE_TABLE_SIZE (1 << KMMIO_PAGE_HASH_BITS)
+
+struct kmmio_context {
+	struct kmmio_fault_page *fpage;
+	struct kmmio_probe *probe;
+	unsigned long saved_flags;
+	int active;
+};
+
+static int kmmio_page_fault(struct pt_regs *regs, unsigned long error_code,
+						unsigned long address);
+static int kmmio_die_notifier(struct notifier_block *nb, unsigned long val,
+								void *args);
+
+static DEFINE_SPINLOCK(kmmio_lock);
+
+/* These are protected by kmmio_lock */
+unsigned int kmmio_count;
+static unsigned int handler_registered;
+static struct list_head kmmio_page_table[KMMIO_PAGE_TABLE_SIZE];
+static LIST_HEAD(kmmio_probes);
+
+static struct kmmio_context kmmio_ctx[NR_CPUS];
+
+static struct pf_handler kmmio_pf_hook = {
+	.handler = kmmio_page_fault
+};
+
+static struct notifier_block nb_die = {
+	.notifier_call = kmmio_die_notifier
+};
+
+int init_kmmio(void)
+{
+	int i;
+	for (i = 0; i < KMMIO_PAGE_TABLE_SIZE; i++)
+		INIT_LIST_HEAD(&kmmio_page_table[i]);
+
+	register_die_notifier(&nb_die);
+	return 0;
+}
+
+void cleanup_kmmio(void)
+{
+	/*
+	 * Assume the following have been already cleaned by calling
+	 * unregister_kmmio_probe() appropriately:
+	 * kmmio_page_table, kmmio_probes
+	 */
+	if (handler_registered) {
+		unregister_page_fault_handler(&kmmio_pf_hook);
+		synchronize_rcu();
+	}
+	unregister_die_notifier(&nb_die);
+}
+
+/*
+ * this is basically a dynamic stabbing problem:
+ * Could use the existing prio tree code or
+ * Possible better implementations:
+ * The Interval Skip List: A Data Structure for Finding All Intervals That
+ * Overlap a Point (might be simple)
+ * Space Efficient Dynamic Stabbing with Fast Queries - Mikkel Thorup
+ */
+/* Get the kmmio at this addr (if any). You must be holding kmmio_lock. */
+static struct kmmio_probe *get_kmmio_probe(unsigned long addr)
+{
+	struct kmmio_probe *p;
+	list_for_each_entry(p, &kmmio_probes, list) {
+		if (addr >= p->addr && addr <= (p->addr + p->len))
+			return p;
+	}
+	return NULL;
+}
+
+static struct kmmio_fault_page *get_kmmio_fault_page(unsigned long page)
+{
+	struct list_head *head, *tmp;
+
+	page &= PAGE_MASK;
+	head = &kmmio_page_table[hash_long(page, KMMIO_PAGE_HASH_BITS)];
+	list_for_each(tmp, head) {
+		struct kmmio_fault_page *p
+			= list_entry(tmp, struct kmmio_fault_page, list);
+		if (p->page == page)
+			return p;
+	}
+
+	return NULL;
+}
+
+static void arm_kmmio_fault_page(unsigned long page, int *large)
+{
+	unsigned long address = page & PAGE_MASK;
+	pgd_t *pgd = pgd_offset_k(address);
+	pud_t *pud = pud_offset(pgd, address);
+	pmd_t *pmd = pmd_offset(pud, address);
+	pte_t *pte = pte_offset_kernel(pmd, address);
+
+	if (pmd_large(*pmd)) {
+		set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_PRESENT));
+		if (large)
+			*large = 1;
+	} else {
+		set_pte(pte, __pte(pte_val(*pte) & ~_PAGE_PRESENT));
+	}
+
+	__flush_tlb_one(page);
+}
+
+static void disarm_kmmio_fault_page(unsigned long page, int *large)
+{
+	unsigned long address = page & PAGE_MASK;
+	pgd_t *pgd = pgd_offset_k(address);
+	pud_t *pud = pud_offset(pgd, address);
+	pmd_t *pmd = pmd_offset(pud, address);
+	pte_t *pte = pte_offset_kernel(pmd, address);
+
+	if (large && *large) {
+		set_pmd(pmd, __pmd(pmd_val(*pmd) | _PAGE_PRESENT));
+		*large = 0;
+	} else {
+		set_pte(pte, __pte(pte_val(*pte) | _PAGE_PRESENT));
+	}
+
+	__flush_tlb_one(page);
+}
+
+/*
+ * Interrupts are disabled on entry as trap3 is an interrupt gate
+ * and they remain disabled thorough out this function.
+ */
+static int kmmio_handler(struct pt_regs *regs, unsigned long addr)
+{
+	struct kmmio_context *ctx;
+	int cpu;
+
+	/*
+	 * Preemption is now disabled to prevent process switch during
+	 * single stepping. We can only handle one active kmmio trace
+	 * per cpu, so ensure that we finish it before something else
+	 * gets to run.
+	 *
+	 * XXX what if an interrupt occurs between returning from
+	 * do_page_fault() and entering the single-step exception handler?
+	 * And that interrupt triggers a kmmio trap?
+	 */
+	preempt_disable();
+	cpu = smp_processor_id();
+	ctx = &kmmio_ctx[cpu];
+
+	/* interrupts disabled and CPU-local data => atomicity guaranteed. */
+	if (ctx->active) {
+		/*
+		 * This avoids a deadlock with kmmio_lock.
+		 * If this page fault really was due to kmmio trap,
+		 * all hell breaks loose.
+		 */
+		printk(KERN_EMERG "mmiotrace: recursive probe hit on CPU %d, "
+					"for address %lu. Ignoring.\n",
+					cpu, addr);
+		goto no_kmmio;
+	}
+	ctx->active++;
+
+	/*
+	 * Acquire the kmmio lock to prevent changes affecting
+	 * get_kmmio_fault_page() and get_kmmio_probe(), since we save their
+	 * returned pointers.
+	 * The lock is released in post_kmmio_handler().
+	 * XXX: could/should get_kmmio_*() be using RCU instead of spinlock?
+	 */
+	spin_lock(&kmmio_lock);
+
+	ctx->fpage = get_kmmio_fault_page(addr);
+	if (!ctx->fpage) {
+		/* this page fault is not caused by kmmio */
+		goto no_kmmio_locked;
+	}
+
+	ctx->probe = get_kmmio_probe(addr);
+	ctx->saved_flags = (regs->flags & (TF_MASK|IF_MASK));
+
+	if (ctx->probe && ctx->probe->pre_handler)
+		ctx->probe->pre_handler(ctx->probe, regs, addr);
+
+	regs->flags |= TF_MASK;
+	regs->flags &= ~IF_MASK;
+
+	/* We hold lock, now we set present bit in PTE and single step. */
+	disarm_kmmio_fault_page(ctx->fpage->page, NULL);
+
+	return 1;
+
+no_kmmio_locked:
+	spin_unlock(&kmmio_lock);
+	ctx->active--;
+no_kmmio:
+	preempt_enable_no_resched();
+	/* page fault not handled by kmmio */
+	return 0;
+}
+
+/*
+ * Interrupts are disabled on entry as trap1 is an interrupt gate
+ * and they remain disabled thorough out this function.
+ * And we hold kmmio lock.
+ */
+static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
+{
+	int cpu = smp_processor_id();
+	struct kmmio_context *ctx = &kmmio_ctx[cpu];
+
+	if (!ctx->active)
+		return 0;
+
+	if (ctx->probe && ctx->probe->post_handler)
+		ctx->probe->post_handler(ctx->probe, condition, regs);
+
+	arm_kmmio_fault_page(ctx->fpage->page, NULL);
+
+	regs->flags &= ~TF_MASK;
+	regs->flags |= ctx->saved_flags;
+
+	/* These were acquired in kmmio_handler(). */
+	ctx->active--;
+	spin_unlock(&kmmio_lock);
+	preempt_enable_no_resched();
+
+	/*
+	 * if somebody else is singlestepping across a probe point, flags
+	 * will have TF set, in which case, continue the remaining processing
+	 * of do_debug, as if this is not a probe hit.
+	 */
+	if (regs->flags & TF_MASK)
+		return 0;
+
+	return 1;
+}
+
+static int add_kmmio_fault_page(unsigned long page)
+{
+	struct kmmio_fault_page *f;
+
+	page &= PAGE_MASK;
+	f = get_kmmio_fault_page(page);
+	if (f) {
+		f->count++;
+		return 0;
+	}
+
+	f = kmalloc(sizeof(*f), GFP_ATOMIC);
+	if (!f)
+		return -1;
+
+	f->count = 1;
+	f->page = page;
+	list_add(&f->list,
+		 &kmmio_page_table[hash_long(f->page, KMMIO_PAGE_HASH_BITS)]);
+
+	arm_kmmio_fault_page(f->page, NULL);
+
+	return 0;
+}
+
+static void release_kmmio_fault_page(unsigned long page)
+{
+	struct kmmio_fault_page *f;
+
+	page &= PAGE_MASK;
+	f = get_kmmio_fault_page(page);
+	if (!f)
+		return;
+
+	f->count--;
+	if (!f->count) {
+		disarm_kmmio_fault_page(f->page, NULL);
+		list_del(&f->list);
+	}
+}
+
+int register_kmmio_probe(struct kmmio_probe *p)
+{
+	int ret = 0;
+	unsigned long size = 0;
+
+	spin_lock_irq(&kmmio_lock);
+	kmmio_count++;
+	if (get_kmmio_probe(p->addr)) {
+		ret = -EEXIST;
+		goto out;
+	}
+	list_add(&p->list, &kmmio_probes);
+	/*printk("adding fault pages...\n");*/
+	while (size < p->len) {
+		if (add_kmmio_fault_page(p->addr + size))
+			printk(KERN_ERR "mmio: Unable to set page fault.\n");
+		size += PAGE_SIZE;
+	}
+
+	if (!handler_registered) {
+		register_page_fault_handler(&kmmio_pf_hook);
+		handler_registered++;
+	}
+
+out:
+	spin_unlock_irq(&kmmio_lock);
+	/*
+	 * XXX: What should I do here?
+	 * Here was a call to global_flush_tlb(), but it does not exist
+	 * anymore.
+	 */
+	return ret;
+}
+
+void unregister_kmmio_probe(struct kmmio_probe *p)
+{
+	unsigned long size = 0;
+
+	spin_lock_irq(&kmmio_lock);
+	while (size < p->len) {
+		release_kmmio_fault_page(p->addr + size);
+		size += PAGE_SIZE;
+	}
+	list_del(&p->list);
+	kmmio_count--;
+	spin_unlock_irq(&kmmio_lock);
+}
+
+/*
+ * According to 2.6.20, mainly x86_64 arch:
+ * This is being called from do_page_fault(), via the page fault notifier
+ * chain. The chain is called for both user space faults and kernel space
+ * faults (address >= TASK_SIZE64), except not on faults serviced by
+ * vmalloc_fault().
+ *
+ * We may be in an interrupt or a critical section. Also prefecthing may
+ * trigger a page fault. We may be in the middle of process switch.
+ * The page fault hook functionality has put us inside RCU read lock.
+ *
+ * Local interrupts are disabled, so preemption cannot happen.
+ * Do not enable interrupts, do not sleep, and watch out for other CPUs.
+ */
+static int kmmio_page_fault(struct pt_regs *regs, unsigned long error_code,
+						unsigned long address)
+{
+	if (is_kmmio_active())
+		if (kmmio_handler(regs, address) == 1)
+			return -1;
+	return 0;
+}
+
+static int kmmio_die_notifier(struct notifier_block *nb, unsigned long val,
+								void *args)
+{
+	struct die_args *arg = args;
+
+	if (val == DIE_DEBUG)
+		if (post_kmmio_handler(arg->err, arg->regs) == 1)
+			return NOTIFY_STOP;
+
+	return NOTIFY_DONE;
+}
diff --git a/arch/x86/kernel/mmiotrace/kmmio.h b/arch/x86/kernel/mmiotrace/kmmio.h
new file mode 100644
index 000000000000..85b7f68a3b8a
--- /dev/null
+++ b/arch/x86/kernel/mmiotrace/kmmio.h
@@ -0,0 +1,58 @@
+#ifndef _LINUX_KMMIO_H
+#define _LINUX_KMMIO_H
+
+#include <linux/list.h>
+#include <linux/notifier.h>
+#include <linux/smp.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/version.h>
+#include <linux/kdebug.h>
+
+struct kmmio_probe;
+struct kmmio_fault_page;
+struct pt_regs;
+
+typedef void (*kmmio_pre_handler_t)(struct kmmio_probe *,
+				struct pt_regs *, unsigned long addr);
+typedef void (*kmmio_post_handler_t)(struct kmmio_probe *,
+				unsigned long condition, struct pt_regs *);
+
+struct kmmio_probe {
+	struct list_head list;
+
+	/* start location of the probe point */
+	unsigned long addr;
+
+	/* length of the probe region */
+	unsigned long len;
+
+	 /* Called before addr is executed. */
+	kmmio_pre_handler_t pre_handler;
+
+	/* Called after addr is executed, unless... */
+	kmmio_post_handler_t post_handler;
+};
+
+struct kmmio_fault_page {
+	struct list_head list;
+
+	/* location of the fault page */
+	unsigned long page;
+
+	int count;
+};
+
+/* kmmio is active by some kmmio_probes? */
+static inline int is_kmmio_active(void)
+{
+	extern unsigned int kmmio_count;
+	return kmmio_count;
+}
+
+int init_kmmio(void);
+void cleanup_kmmio(void);
+int register_kmmio_probe(struct kmmio_probe *p);
+void unregister_kmmio_probe(struct kmmio_probe *p);
+
+#endif /* _LINUX_KMMIO_H */
diff --git a/arch/x86/kernel/mmiotrace/mmio-mod.c b/arch/x86/kernel/mmiotrace/mmio-mod.c
new file mode 100644
index 000000000000..73561fe85f03
--- /dev/null
+++ b/arch/x86/kernel/mmiotrace/mmio-mod.c
@@ -0,0 +1,527 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) IBM Corporation, 2005
+ *               Jeff Muizelaar, 2006, 2007
+ *               Pekka Paalanen, 2008 <pq@iki.fi>
+ *
+ * Derived from the read-mod example from relay-examples by Tom Zanussi.
+ */
+#include <linux/module.h>
+#include <linux/relay.h>
+#include <linux/debugfs.h>
+#include <linux/proc_fs.h>
+#include <asm/io.h>
+#include <linux/version.h>
+#include <linux/kallsyms.h>
+#include <asm/pgtable.h>
+#include <linux/mmiotrace.h>
+#include <asm/e820.h> /* for ISA_START_ADDRESS */
+
+#include "kmmio.h"
+#include "pf_in.h"
+
+/* This app's relay channel files will appear in /debug/mmio-trace */
+#define APP_DIR		"mmio-trace"
+/* the marker injection file in /proc */
+#define MARKER_FILE     "mmio-marker"
+
+#define MODULE_NAME     "mmiotrace"
+
+struct trap_reason {
+	unsigned long addr;
+	unsigned long ip;
+	enum reason_type type;
+	int active_traces;
+};
+
+static struct trap_reason pf_reason[NR_CPUS];
+static struct mm_io_header_rw cpu_trace[NR_CPUS];
+
+static struct file_operations mmio_fops = {
+	.owner = THIS_MODULE,
+};
+
+static const size_t subbuf_size = 256*1024;
+static struct rchan *chan;
+static struct dentry *dir;
+static int suspended;      /* XXX should this be per cpu? */
+static struct proc_dir_entry *proc_marker_file;
+
+/* module parameters */
+static unsigned int      n_subbufs = 32*4;
+static unsigned long filter_offset;
+static int                 nommiotrace;
+static int               ISA_trace;
+static int                trace_pc;
+
+module_param(n_subbufs, uint, 0);
+module_param(filter_offset, ulong, 0);
+module_param(nommiotrace, bool, 0);
+module_param(ISA_trace, bool, 0);
+module_param(trace_pc, bool, 0);
+
+MODULE_PARM_DESC(n_subbufs, "Number of 256kB buffers, default 128.");
+MODULE_PARM_DESC(filter_offset, "Start address of traced mappings.");
+MODULE_PARM_DESC(nommiotrace, "Disable actual MMIO tracing.");
+MODULE_PARM_DESC(ISA_trace, "Do not exclude the low ISA range.");
+MODULE_PARM_DESC(trace_pc, "Record address of faulting instructions.");
+
+static void record_timestamp(struct mm_io_header *header)
+{
+	struct timespec now;
+
+	getnstimeofday(&now);
+	header->sec = now.tv_sec;
+	header->nsec = now.tv_nsec;
+}
+
+/*
+ * Write callback for the /proc entry:
+ * Read a marker and write it to the mmio trace log
+ */
+static int write_marker(struct file *file, const char __user *buffer,
+					unsigned long count, void *data)
+{
+	char *event = NULL;
+	struct mm_io_header *headp;
+	int len = (count > 65535) ? 65535 : count;
+
+	event = kzalloc(sizeof(*headp) + len, GFP_KERNEL);
+	if (!event)
+		return -ENOMEM;
+
+	headp = (struct mm_io_header *)event;
+	headp->type = MMIO_MAGIC | (MMIO_MARKER << MMIO_OPCODE_SHIFT);
+	headp->data_len = len;
+	record_timestamp(headp);
+
+	if (copy_from_user(event + sizeof(*headp), buffer, len)) {
+		kfree(event);
+		return -EFAULT;
+	}
+
+	relay_write(chan, event, sizeof(*headp) + len);
+	kfree(event);
+	return len;
+}
+
+static void print_pte(unsigned long address)
+{
+	pgd_t *pgd = pgd_offset_k(address);
+	pud_t *pud = pud_offset(pgd, address);
+	pmd_t *pmd = pmd_offset(pud, address);
+	if (pmd_large(*pmd)) {
+		printk(KERN_EMERG MODULE_NAME ": 4MB pages are not "
+						"currently supported: %lx\n",
+						address);
+		BUG();
+	}
+	printk(KERN_DEBUG MODULE_NAME ": pte for 0x%lx: 0x%lx 0x%lx\n",
+		address,
+		pte_val(*pte_offset_kernel(pmd, address)),
+		pte_val(*pte_offset_kernel(pmd, address)) & _PAGE_PRESENT);
+}
+
+/*
+ * For some reason the pre/post pairs have been called in an
+ * unmatched order. Report and die.
+ */
+static void die_kmmio_nesting_error(struct pt_regs *regs, unsigned long addr)
+{
+	const unsigned long cpu = smp_processor_id();
+	printk(KERN_EMERG MODULE_NAME ": unexpected fault for address: %lx, "
+					"last fault for address: %lx\n",
+					addr, pf_reason[cpu].addr);
+	print_pte(addr);
+#ifdef __i386__
+	print_symbol(KERN_EMERG "faulting EIP is at %s\n", regs->ip);
+	print_symbol(KERN_EMERG "last faulting EIP was at %s\n",
+							pf_reason[cpu].ip);
+	printk(KERN_EMERG
+			"eax: %08lx   ebx: %08lx   ecx: %08lx   edx: %08lx\n",
+			regs->ax, regs->bx, regs->cx, regs->dx);
+	printk(KERN_EMERG
+			"esi: %08lx   edi: %08lx   ebp: %08lx   esp: %08lx\n",
+			regs->si, regs->di, regs->bp, regs->sp);
+#else
+	print_symbol(KERN_EMERG "faulting RIP is at %s\n", regs->ip);
+	print_symbol(KERN_EMERG "last faulting RIP was at %s\n",
+							pf_reason[cpu].ip);
+	printk(KERN_EMERG "rax: %016lx   rcx: %016lx   rdx: %016lx\n",
+					regs->ax, regs->cx, regs->dx);
+	printk(KERN_EMERG "rsi: %016lx   rdi: %016lx   "
+				"rbp: %016lx   rsp: %016lx\n",
+				regs->si, regs->di, regs->bp, regs->sp);
+#endif
+	BUG();
+}
+
+static void pre(struct kmmio_probe *p, struct pt_regs *regs,
+						unsigned long addr)
+{
+	const unsigned long cpu = smp_processor_id();
+	const unsigned long instptr = instruction_pointer(regs);
+	const enum reason_type type = get_ins_type(instptr);
+
+	/* it doesn't make sense to have more than one active trace per cpu */
+	if (pf_reason[cpu].active_traces)
+		die_kmmio_nesting_error(regs, addr);
+	else
+		pf_reason[cpu].active_traces++;
+
+	pf_reason[cpu].type = type;
+	pf_reason[cpu].addr = addr;
+	pf_reason[cpu].ip = instptr;
+
+	cpu_trace[cpu].header.type = MMIO_MAGIC;
+	cpu_trace[cpu].header.pid = 0;
+	cpu_trace[cpu].header.data_len = sizeof(struct mm_io_rw);
+	cpu_trace[cpu].rw.address = addr;
+
+	/*
+	 * Only record the program counter when requested.
+	 * It may taint clean-room reverse engineering.
+	 */
+	if (trace_pc)
+		cpu_trace[cpu].rw.pc = instptr;
+	else
+		cpu_trace[cpu].rw.pc = 0;
+
+	record_timestamp(&cpu_trace[cpu].header);
+
+	switch (type) {
+	case REG_READ:
+		cpu_trace[cpu].header.type |=
+			(MMIO_READ << MMIO_OPCODE_SHIFT) |
+			(get_ins_mem_width(instptr) << MMIO_WIDTH_SHIFT);
+		break;
+	case REG_WRITE:
+		cpu_trace[cpu].header.type |=
+			(MMIO_WRITE << MMIO_OPCODE_SHIFT) |
+			(get_ins_mem_width(instptr) << MMIO_WIDTH_SHIFT);
+		cpu_trace[cpu].rw.value = get_ins_reg_val(instptr, regs);
+		break;
+	case IMM_WRITE:
+		cpu_trace[cpu].header.type |=
+			(MMIO_WRITE << MMIO_OPCODE_SHIFT) |
+			(get_ins_mem_width(instptr) << MMIO_WIDTH_SHIFT);
+		cpu_trace[cpu].rw.value = get_ins_imm_val(instptr);
+		break;
+	default:
+		{
+			unsigned char *ip = (unsigned char *)instptr;
+			cpu_trace[cpu].header.type |=
+					(MMIO_UNKNOWN_OP << MMIO_OPCODE_SHIFT);
+			cpu_trace[cpu].rw.value = (*ip) << 16 |
+							*(ip + 1) << 8 |
+							*(ip + 2);
+		}
+	}
+}
+
+static void post(struct kmmio_probe *p, unsigned long condition,
+							struct pt_regs *regs)
+{
+	const unsigned long cpu = smp_processor_id();
+
+	/* this should always return the active_trace count to 0 */
+	pf_reason[cpu].active_traces--;
+	if (pf_reason[cpu].active_traces) {
+		printk(KERN_EMERG MODULE_NAME ": unexpected post handler");
+		BUG();
+	}
+
+	switch (pf_reason[cpu].type) {
+	case REG_READ:
+		cpu_trace[cpu].rw.value = get_ins_reg_val(pf_reason[cpu].ip,
+									regs);
+		break;
+	default:
+		break;
+	}
+	relay_write(chan, &cpu_trace[cpu], sizeof(struct mm_io_header_rw));
+}
+
+/*
+ * subbuf_start() relay callback.
+ *
+ * Defined so that we know when events are dropped due to the buffer-full
+ * condition.
+ */
+static int subbuf_start_handler(struct rchan_buf *buf, void *subbuf,
+					void *prev_subbuf, size_t prev_padding)
+{
+	if (relay_buf_full(buf)) {
+		if (!suspended) {
+			suspended = 1;
+			printk(KERN_ERR MODULE_NAME
+						": cpu %d buffer full!!!\n",
+						smp_processor_id());
+		}
+		return 0;
+	} else if (suspended) {
+		suspended = 0;
+		printk(KERN_ERR MODULE_NAME
+					": cpu %d buffer no longer full.\n",
+					smp_processor_id());
+	}
+
+	return 1;
+}
+
+/* file_create() callback.  Creates relay file in debugfs. */
+static struct dentry *create_buf_file_handler(const char *filename,
+						struct dentry *parent,
+						int mode,
+						struct rchan_buf *buf,
+						int *is_global)
+{
+	struct dentry *buf_file;
+
+	mmio_fops.read = relay_file_operations.read;
+	mmio_fops.open = relay_file_operations.open;
+	mmio_fops.poll = relay_file_operations.poll;
+	mmio_fops.mmap = relay_file_operations.mmap;
+	mmio_fops.release = relay_file_operations.release;
+	mmio_fops.splice_read = relay_file_operations.splice_read;
+
+	buf_file = debugfs_create_file(filename, mode, parent, buf,
+								&mmio_fops);
+
+	return buf_file;
+}
+
+/* file_remove() default callback.  Removes relay file in debugfs. */
+static int remove_buf_file_handler(struct dentry *dentry)
+{
+	debugfs_remove(dentry);
+	return 0;
+}
+
+static struct rchan_callbacks relay_callbacks = {
+	.subbuf_start = subbuf_start_handler,
+	.create_buf_file = create_buf_file_handler,
+	.remove_buf_file = remove_buf_file_handler,
+};
+
+/*
+ * create_channel - creates channel /debug/APP_DIR/cpuXXX
+ * Returns channel on success, NULL otherwise
+ */
+static struct rchan *create_channel(unsigned size, unsigned n)
+{
+	return relay_open("cpu", dir, size, n, &relay_callbacks, NULL);
+}
+
+/* destroy_channel - destroys channel /debug/APP_DIR/cpuXXX */
+static void destroy_channel(void)
+{
+	if (chan) {
+		relay_close(chan);
+		chan = NULL;
+	}
+}
+
+struct remap_trace {
+	struct list_head list;
+	struct kmmio_probe probe;
+};
+static LIST_HEAD(trace_list);
+static DEFINE_SPINLOCK(trace_list_lock);
+
+static void do_ioremap_trace_core(unsigned long offset, unsigned long size,
+							void __iomem *addr)
+{
+	struct remap_trace *trace = kmalloc(sizeof(*trace), GFP_KERNEL);
+	struct mm_io_header_map event = {
+		.header = {
+			.type = MMIO_MAGIC |
+					(MMIO_PROBE << MMIO_OPCODE_SHIFT),
+			.sec = 0,
+			.nsec = 0,
+			.pid = 0,
+			.data_len = sizeof(struct mm_io_map)
+		},
+		.map = {
+			.phys = offset,
+			.addr = (unsigned long)addr,
+			.len  = size,
+			.pc   = 0
+		}
+	};
+	record_timestamp(&event.header);
+
+	*trace = (struct remap_trace) {
+		.probe = {
+			.addr = (unsigned long)addr,
+			.len = size,
+			.pre_handler = pre,
+			.post_handler = post,
+		}
+	};
+
+	relay_write(chan, &event, sizeof(event));
+	spin_lock(&trace_list_lock);
+	list_add_tail(&trace->list, &trace_list);
+	spin_unlock(&trace_list_lock);
+	if (!nommiotrace)
+		register_kmmio_probe(&trace->probe);
+}
+
+static void ioremap_trace_core(unsigned long offset, unsigned long size,
+							void __iomem *addr)
+{
+	if ((filter_offset) && (offset != filter_offset))
+		return;
+
+	/* Don't trace the low PCI/ISA area, it's always mapped.. */
+	if (!ISA_trace && (offset < ISA_END_ADDRESS) &&
+					(offset + size > ISA_START_ADDRESS)) {
+		printk(KERN_NOTICE MODULE_NAME ": Ignoring map of low "
+						"PCI/ISA area (0x%lx-0x%lx)\n",
+						offset, offset + size);
+		return;
+	}
+	do_ioremap_trace_core(offset, size, addr);
+}
+
+void __iomem *ioremap_cache_trace(unsigned long offset, unsigned long size)
+{
+	void __iomem *p = ioremap_cache(offset, size);
+	printk(KERN_DEBUG MODULE_NAME ": ioremap_cache(0x%lx, 0x%lx) = %p\n",
+							offset, size, p);
+	ioremap_trace_core(offset, size, p);
+	return p;
+}
+EXPORT_SYMBOL(ioremap_cache_trace);
+
+void __iomem *ioremap_nocache_trace(unsigned long offset, unsigned long size)
+{
+	void __iomem *p = ioremap_nocache(offset, size);
+	printk(KERN_DEBUG MODULE_NAME ": ioremap_nocache(0x%lx, 0x%lx) = %p\n",
+							offset, size, p);
+	ioremap_trace_core(offset, size, p);
+	return p;
+}
+EXPORT_SYMBOL(ioremap_nocache_trace);
+
+void iounmap_trace(volatile void __iomem *addr)
+{
+	struct mm_io_header_map event = {
+		.header = {
+			.type = MMIO_MAGIC |
+				(MMIO_UNPROBE << MMIO_OPCODE_SHIFT),
+			.sec = 0,
+			.nsec = 0,
+			.pid = 0,
+			.data_len = sizeof(struct mm_io_map)
+		},
+		.map = {
+			.phys = 0,
+			.addr = (unsigned long)addr,
+			.len  = 0,
+			.pc   = 0
+		}
+	};
+	struct remap_trace *trace;
+	struct remap_trace *tmp;
+	printk(KERN_DEBUG MODULE_NAME ": Unmapping %p.\n", addr);
+	record_timestamp(&event.header);
+
+	spin_lock(&trace_list_lock);
+	list_for_each_entry_safe(trace, tmp, &trace_list, list) {
+		if ((unsigned long)addr == trace->probe.addr) {
+			if (!nommiotrace)
+				unregister_kmmio_probe(&trace->probe);
+			list_del(&trace->list);
+			kfree(trace);
+			break;
+		}
+	}
+	spin_unlock(&trace_list_lock);
+	relay_write(chan, &event, sizeof(event));
+	iounmap(addr);
+}
+EXPORT_SYMBOL(iounmap_trace);
+
+static void clear_trace_list(void)
+{
+	struct remap_trace *trace;
+	struct remap_trace *tmp;
+
+	spin_lock(&trace_list_lock);
+	list_for_each_entry_safe(trace, tmp, &trace_list, list) {
+		printk(KERN_WARNING MODULE_NAME ": purging non-iounmapped "
+					"trace @0x%08lx, size 0x%lx.\n",
+					trace->probe.addr, trace->probe.len);
+		if (!nommiotrace)
+			unregister_kmmio_probe(&trace->probe);
+		list_del(&trace->list);
+		kfree(trace);
+		break;
+	}
+	spin_unlock(&trace_list_lock);
+}
+
+static int __init init(void)
+{
+	if (n_subbufs < 2)
+		return -EINVAL;
+
+	dir = debugfs_create_dir(APP_DIR, NULL);
+	if (!dir) {
+		printk(KERN_ERR MODULE_NAME
+				": Couldn't create relay app directory.\n");
+		return -ENOMEM;
+	}
+
+	chan = create_channel(subbuf_size, n_subbufs);
+	if (!chan) {
+		debugfs_remove(dir);
+		printk(KERN_ERR MODULE_NAME
+				": relay app channel creation failed\n");
+		return -ENOMEM;
+	}
+
+	init_kmmio();
+
+	proc_marker_file = create_proc_entry(MARKER_FILE, 0, NULL);
+	if (proc_marker_file)
+		proc_marker_file->write_proc = write_marker;
+
+	printk(KERN_DEBUG MODULE_NAME ": loaded.\n");
+	if (nommiotrace)
+		printk(KERN_DEBUG MODULE_NAME ": MMIO tracing disabled.\n");
+	if (ISA_trace)
+		printk(KERN_WARNING MODULE_NAME
+				": Warning! low ISA range will be traced.\n");
+	return 0;
+}
+
+static void __exit cleanup(void)
+{
+	printk(KERN_DEBUG MODULE_NAME ": unload...\n");
+	clear_trace_list();
+	cleanup_kmmio();
+	remove_proc_entry(MARKER_FILE, NULL);
+	destroy_channel();
+	if (dir)
+		debugfs_remove(dir);
+}
+
+module_init(init);
+module_exit(cleanup);
+MODULE_LICENSE("GPL");
diff --git a/arch/x86/kernel/mmiotrace/pf_in.c b/arch/x86/kernel/mmiotrace/pf_in.c
new file mode 100644
index 000000000000..67ea520dde62
--- /dev/null
+++ b/arch/x86/kernel/mmiotrace/pf_in.c
@@ -0,0 +1,489 @@
+/*
+ *  Fault Injection Test harness (FI)
+ *  Copyright (C) Intel Crop.
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version 2
+ *  of the License, or (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307,
+ *  USA.
+ *
+ */
+
+/*  $Id: pf_in.c,v 1.1.1.1 2002/11/12 05:56:32 brlock Exp $
+ *  Copyright by Intel Crop., 2002
+ *  Louis Zhuang (louis.zhuang@intel.com)
+ *
+ *  Bjorn Steinbrink (B.Steinbrink@gmx.de), 2007
+ */
+
+#include <linux/module.h>
+#include <linux/ptrace.h> /* struct pt_regs */
+#include "pf_in.h"
+
+#ifdef __i386__
+/* IA32 Manual 3, 2-1 */
+static unsigned char prefix_codes[] = {
+	0xF0, 0xF2, 0xF3, 0x2E, 0x36, 0x3E, 0x26, 0x64,
+	0x65, 0x2E, 0x3E, 0x66, 0x67
+};
+/* IA32 Manual 3, 3-432*/
+static unsigned int reg_rop[] = {
+	0x8A, 0x8B, 0xB60F, 0xB70F, 0xBE0F, 0xBF0F
+};
+static unsigned int reg_wop[] = { 0x88, 0x89 };
+static unsigned int imm_wop[] = { 0xC6, 0xC7 };
+/* IA32 Manual 3, 3-432*/
+static unsigned int rw8[] = { 0x88, 0x8A, 0xC6 };
+static unsigned int rw32[] = {
+	0x89, 0x8B, 0xC7, 0xB60F, 0xB70F, 0xBE0F, 0xBF0F
+};
+static unsigned int mw8[] = { 0x88, 0x8A, 0xC6, 0xB60F, 0xBE0F };
+static unsigned int mw16[] = { 0xB70F, 0xBF0F };
+static unsigned int mw32[] = { 0x89, 0x8B, 0xC7 };
+static unsigned int mw64[] = {};
+#else /* not __i386__ */
+static unsigned char prefix_codes[] = {
+	0x66, 0x67, 0x2E, 0x3E, 0x26, 0x64, 0x65, 0x36,
+	0xF0, 0xF3, 0xF2,
+	/* REX Prefixes */
+	0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
+	0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f
+};
+/* AMD64 Manual 3, Appendix A*/
+static unsigned int reg_rop[] = {
+	0x8A, 0x8B, 0xB60F, 0xB70F, 0xBE0F, 0xBF0F
+};
+static unsigned int reg_wop[] = { 0x88, 0x89 };
+static unsigned int imm_wop[] = { 0xC6, 0xC7 };
+static unsigned int rw8[] = { 0xC6, 0x88, 0x8A };
+static unsigned int rw32[] = {
+	0xC7, 0x89, 0x8B, 0xB60F, 0xB70F, 0xBE0F, 0xBF0F
+};
+/* 8 bit only */
+static unsigned int mw8[] = { 0xC6, 0x88, 0x8A, 0xB60F, 0xBE0F };
+/* 16 bit only */
+static unsigned int mw16[] = { 0xB70F, 0xBF0F };
+/* 16 or 32 bit */
+static unsigned int mw32[] = { 0xC7 };
+/* 16, 32 or 64 bit */
+static unsigned int mw64[] = { 0x89, 0x8B };
+#endif /* not __i386__ */
+
+static int skip_prefix(unsigned char *addr, int *shorted, int *enlarged,
+								int *rexr)
+{
+	int i;
+	unsigned char *p = addr;
+	*shorted = 0;
+	*enlarged = 0;
+	*rexr = 0;
+
+restart:
+	for (i = 0; i < ARRAY_SIZE(prefix_codes); i++) {
+		if (*p == prefix_codes[i]) {
+			if (*p == 0x66)
+				*shorted = 1;
+#ifdef __amd64__
+			if ((*p & 0xf8) == 0x48)
+				*enlarged = 1;
+			if ((*p & 0xf4) == 0x44)
+				*rexr = 1;
+#endif
+			p++;
+			goto restart;
+		}
+	}
+
+	return (p - addr);
+}
+
+static int get_opcode(unsigned char *addr, unsigned int *opcode)
+{
+	int len;
+
+	if (*addr == 0x0F) {
+		/* 0x0F is extension instruction */
+		*opcode = *(unsigned short *)addr;
+		len = 2;
+	} else {
+		*opcode = *addr;
+		len = 1;
+	}
+
+	return len;
+}
+
+#define CHECK_OP_TYPE(opcode, array, type) \
+	for (i = 0; i < ARRAY_SIZE(array); i++) { \
+		if (array[i] == opcode) { \
+			rv = type; \
+			goto exit; \
+		} \
+	}
+
+enum reason_type get_ins_type(unsigned long ins_addr)
+{
+	unsigned int opcode;
+	unsigned char *p;
+	int shorted, enlarged, rexr;
+	int i;
+	enum reason_type rv = OTHERS;
+
+	p = (unsigned char *)ins_addr;
+	p += skip_prefix(p, &shorted, &enlarged, &rexr);
+	p += get_opcode(p, &opcode);
+
+	CHECK_OP_TYPE(opcode, reg_rop, REG_READ);
+	CHECK_OP_TYPE(opcode, reg_wop, REG_WRITE);
+	CHECK_OP_TYPE(opcode, imm_wop, IMM_WRITE);
+
+exit:
+	return rv;
+}
+#undef CHECK_OP_TYPE
+
+static unsigned int get_ins_reg_width(unsigned long ins_addr)
+{
+	unsigned int opcode;
+	unsigned char *p;
+	int i, shorted, enlarged, rexr;
+
+	p = (unsigned char *)ins_addr;
+	p += skip_prefix(p, &shorted, &enlarged, &rexr);
+	p += get_opcode(p, &opcode);
+
+	for (i = 0; i < ARRAY_SIZE(rw8); i++)
+		if (rw8[i] == opcode)
+			return 1;
+
+	for (i = 0; i < ARRAY_SIZE(rw32); i++)
+		if (rw32[i] == opcode)
+			return (shorted ? 2 : (enlarged ? 8 : 4));
+
+	printk(KERN_ERR "mmiotrace: Unknown opcode 0x%02x\n", opcode);
+	return 0;
+}
+
+unsigned int get_ins_mem_width(unsigned long ins_addr)
+{
+	unsigned int opcode;
+	unsigned char *p;
+	int i, shorted, enlarged, rexr;
+
+	p = (unsigned char *)ins_addr;
+	p += skip_prefix(p, &shorted, &enlarged, &rexr);
+	p += get_opcode(p, &opcode);
+
+	for (i = 0; i < ARRAY_SIZE(mw8); i++)
+		if (mw8[i] == opcode)
+			return 1;
+
+	for (i = 0; i < ARRAY_SIZE(mw16); i++)
+		if (mw16[i] == opcode)
+			return 2;
+
+	for (i = 0; i < ARRAY_SIZE(mw32); i++)
+		if (mw32[i] == opcode)
+			return shorted ? 2 : 4;
+
+	for (i = 0; i < ARRAY_SIZE(mw64); i++)
+		if (mw64[i] == opcode)
+			return shorted ? 2 : (enlarged ? 8 : 4);
+
+	printk(KERN_ERR "mmiotrace: Unknown opcode 0x%02x\n", opcode);
+	return 0;
+}
+
+/*
+ * Define register ident in mod/rm byte.
+ * Note: these are NOT the same as in ptrace-abi.h.
+ */
+enum {
+	arg_AL = 0,
+	arg_CL = 1,
+	arg_DL = 2,
+	arg_BL = 3,
+	arg_AH = 4,
+	arg_CH = 5,
+	arg_DH = 6,
+	arg_BH = 7,
+
+	arg_AX = 0,
+	arg_CX = 1,
+	arg_DX = 2,
+	arg_BX = 3,
+	arg_SP = 4,
+	arg_BP = 5,
+	arg_SI = 6,
+	arg_DI = 7,
+#ifdef __amd64__
+	arg_R8  = 8,
+	arg_R9  = 9,
+	arg_R10 = 10,
+	arg_R11 = 11,
+	arg_R12 = 12,
+	arg_R13 = 13,
+	arg_R14 = 14,
+	arg_R15 = 15
+#endif
+};
+
+static unsigned char *get_reg_w8(int no, struct pt_regs *regs)
+{
+	unsigned char *rv = NULL;
+
+	switch (no) {
+	case arg_AL:
+		rv = (unsigned char *)&regs->ax;
+		break;
+	case arg_BL:
+		rv = (unsigned char *)&regs->bx;
+		break;
+	case arg_CL:
+		rv = (unsigned char *)&regs->cx;
+		break;
+	case arg_DL:
+		rv = (unsigned char *)&regs->dx;
+		break;
+	case arg_AH:
+		rv = 1 + (unsigned char *)&regs->ax;
+		break;
+	case arg_BH:
+		rv = 1 + (unsigned char *)&regs->bx;
+		break;
+	case arg_CH:
+		rv = 1 + (unsigned char *)&regs->cx;
+		break;
+	case arg_DH:
+		rv = 1 + (unsigned char *)&regs->dx;
+		break;
+#ifdef __amd64__
+	case arg_R8:
+		rv = (unsigned char *)&regs->r8;
+		break;
+	case arg_R9:
+		rv = (unsigned char *)&regs->r9;
+		break;
+	case arg_R10:
+		rv = (unsigned char *)&regs->r10;
+		break;
+	case arg_R11:
+		rv = (unsigned char *)&regs->r11;
+		break;
+	case arg_R12:
+		rv = (unsigned char *)&regs->r12;
+		break;
+	case arg_R13:
+		rv = (unsigned char *)&regs->r13;
+		break;
+	case arg_R14:
+		rv = (unsigned char *)&regs->r14;
+		break;
+	case arg_R15:
+		rv = (unsigned char *)&regs->r15;
+		break;
+#endif
+	default:
+		printk(KERN_ERR "mmiotrace: Error reg no# %d\n", no);
+		break;
+	}
+	return rv;
+}
+
+static unsigned long *get_reg_w32(int no, struct pt_regs *regs)
+{
+	unsigned long *rv = NULL;
+
+	switch (no) {
+	case arg_AX:
+		rv = &regs->ax;
+		break;
+	case arg_BX:
+		rv = &regs->bx;
+		break;
+	case arg_CX:
+		rv = &regs->cx;
+		break;
+	case arg_DX:
+		rv = &regs->dx;
+		break;
+	case arg_SP:
+		rv = &regs->sp;
+		break;
+	case arg_BP:
+		rv = &regs->bp;
+		break;
+	case arg_SI:
+		rv = &regs->si;
+		break;
+	case arg_DI:
+		rv = &regs->di;
+		break;
+#ifdef __amd64__
+	case arg_R8:
+		rv = &regs->r8;
+		break;
+	case arg_R9:
+		rv = &regs->r9;
+		break;
+	case arg_R10:
+		rv = &regs->r10;
+		break;
+	case arg_R11:
+		rv = &regs->r11;
+		break;
+	case arg_R12:
+		rv = &regs->r12;
+		break;
+	case arg_R13:
+		rv = &regs->r13;
+		break;
+	case arg_R14:
+		rv = &regs->r14;
+		break;
+	case arg_R15:
+		rv = &regs->r15;
+		break;
+#endif
+	default:
+		printk(KERN_ERR "mmiotrace: Error reg no# %d\n", no);
+	}
+
+	return rv;
+}
+
+unsigned long get_ins_reg_val(unsigned long ins_addr, struct pt_regs *regs)
+{
+	unsigned int opcode;
+	unsigned char mod_rm;
+	int reg;
+	unsigned char *p;
+	int i, shorted, enlarged, rexr;
+	unsigned long rv;
+
+	p = (unsigned char *)ins_addr;
+	p += skip_prefix(p, &shorted, &enlarged, &rexr);
+	p += get_opcode(p, &opcode);
+	for (i = 0; i < ARRAY_SIZE(reg_rop); i++)
+		if (reg_rop[i] == opcode) {
+			rv = REG_READ;
+			goto do_work;
+		}
+
+	for (i = 0; i < ARRAY_SIZE(reg_wop); i++)
+		if (reg_wop[i] == opcode) {
+			rv = REG_WRITE;
+			goto do_work;
+		}
+
+	printk(KERN_ERR "mmiotrace: Not a register instruction, opcode "
+							"0x%02x\n", opcode);
+	goto err;
+
+do_work:
+	mod_rm = *p;
+	reg = ((mod_rm >> 3) & 0x7) | (rexr << 3);
+	switch (get_ins_reg_width(ins_addr)) {
+	case 1:
+		return *get_reg_w8(reg, regs);
+
+	case 2:
+		return *(unsigned short *)get_reg_w32(reg, regs);
+
+	case 4:
+		return *(unsigned int *)get_reg_w32(reg, regs);
+
+#ifdef __amd64__
+	case 8:
+		return *(unsigned long *)get_reg_w32(reg, regs);
+#endif
+
+	default:
+		printk(KERN_ERR "mmiotrace: Error width# %d\n", reg);
+	}
+
+err:
+	return 0;
+}
+
+unsigned long get_ins_imm_val(unsigned long ins_addr)
+{
+	unsigned int opcode;
+	unsigned char mod_rm;
+	unsigned char mod;
+	unsigned char *p;
+	int i, shorted, enlarged, rexr;
+	unsigned long rv;
+
+	p = (unsigned char *)ins_addr;
+	p += skip_prefix(p, &shorted, &enlarged, &rexr);
+	p += get_opcode(p, &opcode);
+	for (i = 0; i < ARRAY_SIZE(imm_wop); i++)
+		if (imm_wop[i] == opcode) {
+			rv = IMM_WRITE;
+			goto do_work;
+		}
+
+	printk(KERN_ERR "mmiotrace: Not an immediate instruction, opcode "
+							"0x%02x\n", opcode);
+	goto err;
+
+do_work:
+	mod_rm = *p;
+	mod = mod_rm >> 6;
+	p++;
+	switch (mod) {
+	case 0:
+		/* if r/m is 5 we have a 32 disp (IA32 Manual 3, Table 2-2)  */
+		/* AMD64: XXX Check for address size prefix? */
+		if ((mod_rm & 0x7) == 0x5)
+			p += 4;
+		break;
+
+	case 1:
+		p += 1;
+		break;
+
+	case 2:
+		p += 4;
+		break;
+
+	case 3:
+	default:
+		printk(KERN_ERR "mmiotrace: not a memory access instruction "
+						"at 0x%lx, rm_mod=0x%02x\n",
+						ins_addr, mod_rm);
+	}
+
+	switch (get_ins_reg_width(ins_addr)) {
+	case 1:
+		return *(unsigned char *)p;
+
+	case 2:
+		return *(unsigned short *)p;
+
+	case 4:
+		return *(unsigned int *)p;
+
+#ifdef __amd64__
+	case 8:
+		return *(unsigned long *)p;
+#endif
+
+	default:
+		printk(KERN_ERR "mmiotrace: Error: width.\n");
+	}
+
+err:
+	return 0;
+}
diff --git a/arch/x86/kernel/mmiotrace/pf_in.h b/arch/x86/kernel/mmiotrace/pf_in.h
new file mode 100644
index 000000000000..e05341a51a27
--- /dev/null
+++ b/arch/x86/kernel/mmiotrace/pf_in.h
@@ -0,0 +1,39 @@
+/*
+ *  Fault Injection Test harness (FI)
+ *  Copyright (C) Intel Crop.
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version 2
+ *  of the License, or (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307,
+ *  USA.
+ *
+ */
+
+#ifndef __PF_H_
+#define __PF_H_
+
+enum reason_type {
+	NOT_ME,	/* page fault is not in regions */
+	NOTHING,	/* access others point in regions */
+	REG_READ,	/* read from addr to reg */
+	REG_WRITE,	/* write from reg to addr */
+	IMM_WRITE,	/* write from imm to addr */
+	OTHERS	/* Other instructions can not intercept */
+};
+
+enum reason_type get_ins_type(unsigned long ins_addr);
+unsigned int get_ins_mem_width(unsigned long ins_addr);
+unsigned long get_ins_reg_val(unsigned long ins_addr, struct pt_regs *regs);
+unsigned long get_ins_imm_val(unsigned long ins_addr);
+
+#endif /* __PF_H_ */
diff --git a/arch/x86/kernel/mmiotrace/testmmiotrace.c b/arch/x86/kernel/mmiotrace/testmmiotrace.c
new file mode 100644
index 000000000000..40e66b0e6480
--- /dev/null
+++ b/arch/x86/kernel/mmiotrace/testmmiotrace.c
@@ -0,0 +1,77 @@
+/*
+ * Written by Pekka Paalanen, 2008 <pq@iki.fi>
+ */
+#include <linux/module.h>
+#include <asm/io.h>
+
+extern void __iomem *ioremap_nocache_trace(unsigned long offset,
+						unsigned long size);
+extern void iounmap_trace(volatile void __iomem *addr);
+
+#define MODULE_NAME "testmmiotrace"
+
+static unsigned long mmio_address;
+module_param(mmio_address, ulong, 0);
+MODULE_PARM_DESC(mmio_address, "Start address of the mapping of 16 kB.");
+
+static void do_write_test(void __iomem *p)
+{
+	unsigned int i;
+	for (i = 0; i < 256; i++)
+		iowrite8(i, p + i);
+	for (i = 1024; i < (5 * 1024); i += 2)
+		iowrite16(i * 12 + 7, p + i);
+	for (i = (5 * 1024); i < (16 * 1024); i += 4)
+		iowrite32(i * 212371 + 13, p + i);
+}
+
+static void do_read_test(void __iomem *p)
+{
+	unsigned int i;
+	volatile unsigned int v;
+	for (i = 0; i < 256; i++)
+		v = ioread8(p + i);
+	for (i = 1024; i < (5 * 1024); i += 2)
+		v = ioread16(p + i);
+	for (i = (5 * 1024); i < (16 * 1024); i += 4)
+		v = ioread32(p + i);
+}
+
+static void do_test(void)
+{
+	void __iomem *p = ioremap_nocache_trace(mmio_address, 0x4000);
+	if (!p) {
+		printk(KERN_ERR MODULE_NAME ": could not ioremap IO memory, "
+							"aborting.\n");
+		return;
+	}
+	do_write_test(p);
+	do_read_test(p);
+	iounmap_trace(p);
+}
+
+static int __init init(void)
+{
+	if (mmio_address == 0) {
+		printk(KERN_ERR MODULE_NAME ": you have to use the module "
+						"argument mmio_address.\n");
+		printk(KERN_ERR MODULE_NAME ": DO NOT LOAD THIS MODULE UNLESS"
+				" YOU REALLY KNOW WHAT YOU ARE DOING!\n");
+		return -ENXIO;
+	}
+
+	printk(KERN_WARNING MODULE_NAME ": WARNING: mapping 16 kB @ 0x%08lx "
+					"in PCI address space, and writing "
+					"rubbish in there.\n", mmio_address);
+	do_test();
+	return 0;
+}
+
+static void __exit cleanup(void)
+{
+	printk(KERN_DEBUG MODULE_NAME ": unloaded.\n");
+}
+
+module_init(init);
+module_exit(cleanup);
+MODULE_LICENSE("GPL");
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
new file mode 100644
index 000000000000..cb247825f3ec
--- /dev/null
+++ b/include/linux/mmiotrace.h
@@ -0,0 +1,62 @@
+#ifndef MMIOTRACE_H
+#define MMIOTRACE_H
+
+#include <asm/types.h>
+
+#define MMIO_VERSION 0x04
+
+/* mm_io_header.type */
+#define MMIO_OPCODE_MASK 0xff
+#define MMIO_OPCODE_SHIFT 0
+#define MMIO_WIDTH_MASK 0xff00
+#define MMIO_WIDTH_SHIFT 8
+#define MMIO_MAGIC (0x6f000000 | (MMIO_VERSION<<16))
+#define MMIO_MAGIC_MASK 0xffff0000
+
+enum mm_io_opcode {          /* payload type: */
+	MMIO_READ = 0x1,     /* struct mm_io_rw */
+	MMIO_WRITE = 0x2,    /* struct mm_io_rw */
+	MMIO_PROBE = 0x3,    /* struct mm_io_map */
+	MMIO_UNPROBE = 0x4,  /* struct mm_io_map */
+	MMIO_MARKER = 0x5,   /* raw char data */
+	MMIO_UNKNOWN_OP = 0x6, /* struct mm_io_rw */
+};
+
+struct mm_io_header {
+	__u32 type;
+	__u32 sec;      /* timestamp */
+	__u32 nsec;
+	__u32 pid;      /* PID of the process, or 0 for kernel core */
+	__u16 data_len; /* length of the following payload */
+};
+
+struct mm_io_rw {
+	__u64 address; /* virtual address of register */
+	__u64 value;
+	__u64 pc;      /* optional program counter */
+};
+
+struct mm_io_map {
+	__u64 phys;  /* base address in PCI space */
+	__u64 addr;  /* base virtual address */
+	__u64 len;   /* mapping size */
+	__u64 pc;    /* optional program counter */
+};
+
+
+/*
+ * These structures are used to allow a single relay_write()
+ * call to write a full packet.
+ */
+
+struct mm_io_header_rw {
+	struct mm_io_header header;
+	struct mm_io_rw rw;
+} __attribute__((packed));
+
+struct mm_io_header_map {
+	struct mm_io_header header;
+	struct mm_io_map map;
+} __attribute__((packed));
+
+#endif /* MMIOTRACE_H */
-- 
cgit v1.2.3


From 63ffa3e456c1a9884a3ebac997d91e3fdae18d78 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:20:57 +0200
Subject: x86 mmiotrace: comment about user space ABI

Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/mmiotrace.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index cb247825f3ec..6ec288f1fe24 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -3,6 +3,10 @@
 
 #include <asm/types.h>
 
+/*
+ * If you change anything here, you must bump MMIO_VERSION.
+ * This is the relay data format for user space.
+ */
 #define MMIO_VERSION 0x04
 
 /* mm_io_header.type */
@@ -23,7 +27,7 @@ enum mm_io_opcode {          /* payload type: */
 };
 
 struct mm_io_header {
-	__u32 type;
+	__u32 type;     /* see MMIO_* macros above */
 	__u32 sec;      /* timestamp */
 	__u32 nsec;
 	__u32 pid;      /* PID of the process, or 0 for kernel core */
-- 
cgit v1.2.3


From 0fd0e3da4557c479b820b9a4a7afa25b4637ddf2 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:20:57 +0200
Subject: x86: mmiotrace full patch, preview 1

kmmio.c handles the list of mmio probes with callbacks, list of traced
pages, and attaching into the page fault handler and die notifier. It
arms, traps and disarms the given pages, this is the core of mmiotrace.

mmio-mod.c is a user interface, hooking into ioremap functions and
registering the mmio probes. It also decodes the required information
from trapped mmio accesses via the pre and post callbacks in each probe.
Currently, hooking into ioremap functions works by redefining the symbols
of the target (binary) kernel module, so that it calls the traced
versions of the functions.

The most notable changes done since the last discussion are:
- kmmio.c is a built-in, not part of the module
- direct call from fault.c to kmmio.c, removing all dynamic hooks
- prepare for unregistering probes at any time
- make kmmio re-initializable and accessible to more than one user
- rewrite kmmio locking to remove all spinlocks from page fault path

Can I abuse call_rcu() like I do in kmmio.c:unregister_kmmio_probe()
or is there a better way?

The function called via call_rcu() itself calls call_rcu() again,
will this work or break? There I need a second grace period for RCU
after the first grace period for page faults.

Mmiotrace itself (mmio-mod.c) is still a module, I am going to attack
that next. At some point I will start looking into how to make mmiotrace
a tracer component of ftrace (thanks for the hint, Ingo). Ftrace should
make the user space part of mmiotracing as simple as
'cat /debug/trace/mmio > dump.txt'.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/init_task.c               |   1 -
 arch/x86/kernel/mmiotrace/Makefile        |   8 +-
 arch/x86/kernel/mmiotrace/kmmio.c         | 349 ++++++++++++++++++++----------
 arch/x86/kernel/mmiotrace/kmmio.h         |  58 -----
 arch/x86/kernel/mmiotrace/mmio-mod.c      |  81 ++++---
 arch/x86/kernel/mmiotrace/pf_in.c         |   2 +-
 arch/x86/kernel/mmiotrace/testmmiotrace.c |  13 +-
 arch/x86/mm/fault.c                       |  59 +----
 include/asm-x86/kdebug.h                  |   7 -
 include/linux/mmiotrace.h                 |  38 ++++
 10 files changed, 335 insertions(+), 281 deletions(-)
 delete mode 100644 arch/x86/kernel/mmiotrace/kmmio.h

(limited to 'include/linux')

diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index 027a5b6a12b2..a4f93b4120c1 100644
--- a/arch/x86/kernel/init_task.c
+++ b/arch/x86/kernel/init_task.c
@@ -15,7 +15,6 @@ static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
 EXPORT_UNUSED_SYMBOL(init_mm); /* will be removed in 2.6.26 */
-EXPORT_SYMBOL_GPL(init_mm);
 
 /*
  * Initial thread structure.
diff --git a/arch/x86/kernel/mmiotrace/Makefile b/arch/x86/kernel/mmiotrace/Makefile
index d6905f7f981b..cf1e747b463e 100644
--- a/arch/x86/kernel/mmiotrace/Makefile
+++ b/arch/x86/kernel/mmiotrace/Makefile
@@ -1,4 +1,4 @@
-obj-$(CONFIG_MMIOTRACE) += mmiotrace.o
-mmiotrace-objs := pf_in.o kmmio.o mmio-mod.o
-
-obj-$(CONFIG_MMIOTRACE_TEST) += testmmiotrace.o
+obj-$(CONFIG_MMIOTRACE_HOOKS)	+= kmmio.o
+obj-$(CONFIG_MMIOTRACE)		+= mmiotrace.o
+mmiotrace-objs			:= pf_in.o mmio-mod.o
+obj-$(CONFIG_MMIOTRACE_TEST)	+= testmmiotrace.o
diff --git a/arch/x86/kernel/mmiotrace/kmmio.c b/arch/x86/kernel/mmiotrace/kmmio.c
index 5e239d0b8467..539a9b19588f 100644
--- a/arch/x86/kernel/mmiotrace/kmmio.c
+++ b/arch/x86/kernel/mmiotrace/kmmio.c
@@ -6,6 +6,7 @@
  */
 
 #include <linux/version.h>
+#include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/hash.h>
 #include <linux/init.h>
@@ -17,70 +18,119 @@
 #include <linux/ptrace.h>
 #include <linux/preempt.h>
 #include <linux/percpu.h>
+#include <linux/kdebug.h>
 #include <asm/io.h>
 #include <asm/cacheflush.h>
 #include <asm/errno.h>
 #include <asm/tlbflush.h>
 #include <asm/pgtable.h>
 
-#include "kmmio.h"
+#include <linux/mmiotrace.h>
 
-#define KMMIO_HASH_BITS 6
-#define KMMIO_TABLE_SIZE (1 << KMMIO_HASH_BITS)
 #define KMMIO_PAGE_HASH_BITS 4
 #define KMMIO_PAGE_TABLE_SIZE (1 << KMMIO_PAGE_HASH_BITS)
 
+struct kmmio_fault_page {
+	struct list_head list;
+	struct kmmio_fault_page *release_next;
+	unsigned long page; /* location of the fault page */
+
+	/*
+	 * Number of times this page has been registered as a part
+	 * of a probe. If zero, page is disarmed and this may be freed.
+	 * Used only by writers (RCU).
+	 */
+	int count;
+};
+
+struct kmmio_delayed_release {
+	struct rcu_head rcu;
+	struct kmmio_fault_page *release_list;
+};
+
 struct kmmio_context {
 	struct kmmio_fault_page *fpage;
 	struct kmmio_probe *probe;
 	unsigned long saved_flags;
+	unsigned long addr;
 	int active;
 };
 
-static int kmmio_page_fault(struct pt_regs *regs, unsigned long error_code,
-						unsigned long address);
 static int kmmio_die_notifier(struct notifier_block *nb, unsigned long val,
 								void *args);
 
+static DECLARE_MUTEX(kmmio_init_mutex);
 static DEFINE_SPINLOCK(kmmio_lock);
 
 /* These are protected by kmmio_lock */
+static int kmmio_initialized;
 unsigned int kmmio_count;
-static unsigned int handler_registered;
+
+/* Read-protected by RCU, write-protected by kmmio_lock. */
 static struct list_head kmmio_page_table[KMMIO_PAGE_TABLE_SIZE];
 static LIST_HEAD(kmmio_probes);
 
+static struct list_head *kmmio_page_list(unsigned long page)
+{
+	return &kmmio_page_table[hash_long(page, KMMIO_PAGE_HASH_BITS)];
+}
+
 /* Accessed per-cpu */
 static DEFINE_PER_CPU(struct kmmio_context, kmmio_ctx);
 
+/* protected by kmmio_init_mutex */
 static struct notifier_block nb_die = {
 	.notifier_call = kmmio_die_notifier
 };
 
-int init_kmmio(void)
+/**
+ * Makes sure kmmio is initialized and usable.
+ * This must be called before any other kmmio function defined here.
+ * May sleep.
+ */
+void reference_kmmio(void)
 {
-	int i;
-	for (i = 0; i < KMMIO_PAGE_TABLE_SIZE; i++)
-		INIT_LIST_HEAD(&kmmio_page_table[i]);
-
-	register_die_notifier(&nb_die);
-	return 0;
+	down(&kmmio_init_mutex);
+	spin_lock_irq(&kmmio_lock);
+	if (!kmmio_initialized) {
+		int i;
+		for (i = 0; i < KMMIO_PAGE_TABLE_SIZE; i++)
+			INIT_LIST_HEAD(&kmmio_page_table[i]);
+		if (register_die_notifier(&nb_die))
+			BUG();
+	}
+	kmmio_initialized++;
+	spin_unlock_irq(&kmmio_lock);
+	up(&kmmio_init_mutex);
 }
+EXPORT_SYMBOL_GPL(reference_kmmio);
 
-void cleanup_kmmio(void)
+/**
+ * Clean up kmmio after use. This must be called for every call to
+ * reference_kmmio(). All probes registered after the corresponding
+ * reference_kmmio() must have been unregistered when calling this.
+ * May sleep.
+ */
+void unreference_kmmio(void)
 {
-	/*
-	 * Assume the following have been already cleaned by calling
-	 * unregister_kmmio_probe() appropriately:
-	 * kmmio_page_table, kmmio_probes
-	 */
-	if (handler_registered) {
-		if (mmiotrace_unregister_pf(&kmmio_page_fault))
-			BUG();
-		synchronize_rcu();
+	bool unreg = false;
+
+	down(&kmmio_init_mutex);
+	spin_lock_irq(&kmmio_lock);
+
+	if (kmmio_initialized == 1) {
+		BUG_ON(is_kmmio_active());
+		unreg = true;
 	}
-	unregister_die_notifier(&nb_die);
+	kmmio_initialized--;
+	BUG_ON(kmmio_initialized < 0);
+	spin_unlock_irq(&kmmio_lock);
+
+	if (unreg)
+		unregister_die_notifier(&nb_die); /* calls sync_rcu() */
+	up(&kmmio_init_mutex);
 }
+EXPORT_SYMBOL(unreference_kmmio);
 
 /*
  * this is basically a dynamic stabbing problem:
@@ -90,33 +140,33 @@ void cleanup_kmmio(void)
  * Overlap a Point (might be simple)
  * Space Efficient Dynamic Stabbing with Fast Queries - Mikkel Thorup
  */
-/* Get the kmmio at this addr (if any). You must be holding kmmio_lock. */
+/* Get the kmmio at this addr (if any). You must be holding RCU read lock. */
 static struct kmmio_probe *get_kmmio_probe(unsigned long addr)
 {
 	struct kmmio_probe *p;
-	list_for_each_entry(p, &kmmio_probes, list) {
+	list_for_each_entry_rcu(p, &kmmio_probes, list) {
 		if (addr >= p->addr && addr <= (p->addr + p->len))
 			return p;
 	}
 	return NULL;
 }
 
+/* You must be holding RCU read lock. */
 static struct kmmio_fault_page *get_kmmio_fault_page(unsigned long page)
 {
-	struct list_head *head, *tmp;
+	struct list_head *head;
+	struct kmmio_fault_page *p;
 
 	page &= PAGE_MASK;
-	head = &kmmio_page_table[hash_long(page, KMMIO_PAGE_HASH_BITS)];
-	list_for_each(tmp, head) {
-		struct kmmio_fault_page *p
-			= list_entry(tmp, struct kmmio_fault_page, list);
+	head = kmmio_page_list(page);
+	list_for_each_entry_rcu(p, head, list) {
 		if (p->page == page)
 			return p;
 	}
-
 	return NULL;
 }
 
+/** Mark the given page as not present. Access to it will trigger a fault. */
 static void arm_kmmio_fault_page(unsigned long page, int *page_level)
 {
 	unsigned long address = page & PAGE_MASK;
@@ -124,8 +174,8 @@ static void arm_kmmio_fault_page(unsigned long page, int *page_level)
 	pte_t *pte = lookup_address(address, &level);
 
 	if (!pte) {
-		printk(KERN_ERR "Error in %s: no pte for page 0x%08lx\n",
-						__FUNCTION__, page);
+		pr_err("kmmio: Error in %s: no pte for page 0x%08lx\n",
+							__func__, page);
 		return;
 	}
 
@@ -143,6 +193,7 @@ static void arm_kmmio_fault_page(unsigned long page, int *page_level)
 	__flush_tlb_one(page);
 }
 
+/** Mark the given page as present. */
 static void disarm_kmmio_fault_page(unsigned long page, int *page_level)
 {
 	unsigned long address = page & PAGE_MASK;
@@ -150,8 +201,8 @@ static void disarm_kmmio_fault_page(unsigned long page, int *page_level)
 	pte_t *pte = lookup_address(address, &level);
 
 	if (!pte) {
-		printk(KERN_ERR "Error in %s: no pte for page 0x%08lx\n",
-						__FUNCTION__, page);
+		pr_err("kmmio: Error in %s: no pte for page 0x%08lx\n",
+							__func__, page);
 		return;
 	}
 
@@ -169,13 +220,25 @@ static void disarm_kmmio_fault_page(unsigned long page, int *page_level)
 	__flush_tlb_one(page);
 }
 
+/*
+ * This is being called from do_page_fault().
+ *
+ * We may be in an interrupt or a critical section. Also prefecthing may
+ * trigger a page fault. We may be in the middle of process switch.
+ * We cannot take any locks, because we could be executing especially
+ * within a kmmio critical section.
+ *
+ * Local interrupts are disabled, so preemption cannot happen.
+ * Do not enable interrupts, do not sleep, and watch out for other CPUs.
+ */
 /*
  * Interrupts are disabled on entry as trap3 is an interrupt gate
  * and they remain disabled thorough out this function.
  */
-static int kmmio_handler(struct pt_regs *regs, unsigned long addr)
+int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 {
-	struct kmmio_context *ctx = &get_cpu_var(kmmio_ctx);
+	struct kmmio_context *ctx;
+	struct kmmio_fault_page *faultpage;
 
 	/*
 	 * Preemption is now disabled to prevent process switch during
@@ -186,40 +249,40 @@ static int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 	 * XXX what if an interrupt occurs between returning from
 	 * do_page_fault() and entering the single-step exception handler?
 	 * And that interrupt triggers a kmmio trap?
+	 * XXX If we tracing an interrupt service routine or whatever, is
+	 * this enough to keep it on the current cpu?
 	 */
 	preempt_disable();
 
-	/* interrupts disabled and CPU-local data => atomicity guaranteed. */
+	rcu_read_lock();
+	faultpage = get_kmmio_fault_page(addr);
+	if (!faultpage) {
+		/*
+		 * Either this page fault is not caused by kmmio, or
+		 * another CPU just pulled the kmmio probe from under
+		 * our feet. In the latter case all hell breaks loose.
+		 */
+		goto no_kmmio;
+	}
+
+	ctx = &get_cpu_var(kmmio_ctx);
 	if (ctx->active) {
 		/*
-		 * This avoids a deadlock with kmmio_lock.
+		 * Prevent overwriting already in-flight context.
 		 * If this page fault really was due to kmmio trap,
 		 * all hell breaks loose.
 		 */
-		printk(KERN_EMERG "mmiotrace: recursive probe hit on CPU %d, "
-					"for address %lu. Ignoring.\n",
+		pr_emerg("kmmio: recursive probe hit on CPU %d, "
+					"for address 0x%08lx. Ignoring.\n",
 					smp_processor_id(), addr);
-		goto no_kmmio;
+		goto no_kmmio_ctx;
 	}
 	ctx->active++;
 
-	/*
-	 * Acquire the kmmio lock to prevent changes affecting
-	 * get_kmmio_fault_page() and get_kmmio_probe(), since we save their
-	 * returned pointers.
-	 * The lock is released in post_kmmio_handler().
-	 * XXX: could/should get_kmmio_*() be using RCU instead of spinlock?
-	 */
-	spin_lock(&kmmio_lock);
-
-	ctx->fpage = get_kmmio_fault_page(addr);
-	if (!ctx->fpage) {
-		/* this page fault is not caused by kmmio */
-		goto no_kmmio_locked;
-	}
-
+	ctx->fpage = faultpage;
 	ctx->probe = get_kmmio_probe(addr);
 	ctx->saved_flags = (regs->flags & (TF_MASK|IF_MASK));
+	ctx->addr = addr;
 
 	if (ctx->probe && ctx->probe->pre_handler)
 		ctx->probe->pre_handler(ctx->probe, regs, addr);
@@ -227,46 +290,62 @@ static int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 	regs->flags |= TF_MASK;
 	regs->flags &= ~IF_MASK;
 
-	/* We hold lock, now we set present bit in PTE and single step. */
+	/* Now we set present bit in PTE and single step. */
 	disarm_kmmio_fault_page(ctx->fpage->page, NULL);
 
 	put_cpu_var(kmmio_ctx);
+	rcu_read_unlock();
 	return 1;
 
-no_kmmio_locked:
-	spin_unlock(&kmmio_lock);
-	ctx->active--;
+no_kmmio_ctx:
+	put_cpu_var(kmmio_ctx);
 no_kmmio:
+	rcu_read_unlock();
 	preempt_enable_no_resched();
-	put_cpu_var(kmmio_ctx);
-	/* page fault not handled by kmmio */
-	return 0;
+	return 0; /* page fault not handled by kmmio */
 }
 
 /*
  * Interrupts are disabled on entry as trap1 is an interrupt gate
  * and they remain disabled thorough out this function.
- * And we hold kmmio lock.
+ * This must always get called as the pair to kmmio_handler().
  */
 static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
 {
 	int ret = 0;
+	struct kmmio_probe *probe;
+	struct kmmio_fault_page *faultpage;
 	struct kmmio_context *ctx = &get_cpu_var(kmmio_ctx);
 
 	if (!ctx->active)
 		goto out;
 
+	rcu_read_lock();
+
+	faultpage = get_kmmio_fault_page(ctx->addr);
+	probe = get_kmmio_probe(ctx->addr);
+	if (faultpage != ctx->fpage || probe != ctx->probe) {
+		/*
+		 * The trace setup changed after kmmio_handler() and before
+		 * running this respective post handler. User does not want
+		 * the result anymore.
+		 */
+		ctx->probe = NULL;
+		ctx->fpage = NULL;
+	}
+
 	if (ctx->probe && ctx->probe->post_handler)
 		ctx->probe->post_handler(ctx->probe, condition, regs);
 
-	arm_kmmio_fault_page(ctx->fpage->page, NULL);
+	if (ctx->fpage)
+		arm_kmmio_fault_page(ctx->fpage->page, NULL);
 
 	regs->flags &= ~TF_MASK;
 	regs->flags |= ctx->saved_flags;
 
 	/* These were acquired in kmmio_handler(). */
 	ctx->active--;
-	spin_unlock(&kmmio_lock);
+	BUG_ON(ctx->active);
 	preempt_enable_no_resched();
 
 	/*
@@ -277,11 +356,13 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
 	if (!(regs->flags & TF_MASK))
 		ret = 1;
 
+	rcu_read_unlock();
 out:
 	put_cpu_var(kmmio_ctx);
 	return ret;
 }
 
+/* You must be holding kmmio_lock. */
 static int add_kmmio_fault_page(unsigned long page)
 {
 	struct kmmio_fault_page *f;
@@ -289,6 +370,8 @@ static int add_kmmio_fault_page(unsigned long page)
 	page &= PAGE_MASK;
 	f = get_kmmio_fault_page(page);
 	if (f) {
+		if (!f->count)
+			arm_kmmio_fault_page(f->page, NULL);
 		f->count++;
 		return 0;
 	}
@@ -299,15 +382,16 @@ static int add_kmmio_fault_page(unsigned long page)
 
 	f->count = 1;
 	f->page = page;
-	list_add(&f->list,
-		 &kmmio_page_table[hash_long(f->page, KMMIO_PAGE_HASH_BITS)]);
+	list_add_rcu(&f->list, kmmio_page_list(f->page));
 
 	arm_kmmio_fault_page(f->page, NULL);
 
 	return 0;
 }
 
-static void release_kmmio_fault_page(unsigned long page)
+/* You must be holding kmmio_lock. */
+static void release_kmmio_fault_page(unsigned long page,
+				struct kmmio_fault_page **release_list)
 {
 	struct kmmio_fault_page *f;
 
@@ -317,9 +401,11 @@ static void release_kmmio_fault_page(unsigned long page)
 		return;
 
 	f->count--;
+	BUG_ON(f->count < 0);
 	if (!f->count) {
 		disarm_kmmio_fault_page(f->page, NULL);
-		list_del(&f->list);
+		f->release_next = *release_list;
+		*release_list = f;
 	}
 }
 
@@ -334,68 +420,113 @@ int register_kmmio_probe(struct kmmio_probe *p)
 		ret = -EEXIST;
 		goto out;
 	}
-	list_add(&p->list, &kmmio_probes);
-	/*printk("adding fault pages...\n");*/
+	list_add_rcu(&p->list, &kmmio_probes);
 	while (size < p->len) {
 		if (add_kmmio_fault_page(p->addr + size))
-			printk(KERN_ERR "mmio: Unable to set page fault.\n");
+			pr_err("kmmio: Unable to set page fault.\n");
 		size += PAGE_SIZE;
 	}
-
-	if (!handler_registered) {
-		if (mmiotrace_register_pf(&kmmio_page_fault))
-			printk(KERN_ERR "mmiotrace: Cannot register page "
-					"fault handler.\n");
-		else
-			handler_registered++;
-	}
-
 out:
 	spin_unlock_irq(&kmmio_lock);
 	/*
 	 * XXX: What should I do here?
 	 * Here was a call to global_flush_tlb(), but it does not exist
-	 * anymore.
+	 * anymore. It seems it's not needed after all.
 	 */
 	return ret;
 }
+EXPORT_SYMBOL(register_kmmio_probe);
 
+static void rcu_free_kmmio_fault_pages(struct rcu_head *head)
+{
+	struct kmmio_delayed_release *dr = container_of(
+						head,
+						struct kmmio_delayed_release,
+						rcu);
+	struct kmmio_fault_page *p = dr->release_list;
+	while (p) {
+		struct kmmio_fault_page *next = p->release_next;
+		BUG_ON(p->count);
+		kfree(p);
+		p = next;
+	}
+	kfree(dr);
+}
+
+static void remove_kmmio_fault_pages(struct rcu_head *head)
+{
+	struct kmmio_delayed_release *dr = container_of(
+						head,
+						struct kmmio_delayed_release,
+						rcu);
+	struct kmmio_fault_page *p = dr->release_list;
+	struct kmmio_fault_page **prevp = &dr->release_list;
+	unsigned long flags;
+	spin_lock_irqsave(&kmmio_lock, flags);
+	while (p) {
+		if (!p->count)
+			list_del_rcu(&p->list);
+		else
+			*prevp = p->release_next;
+		prevp = &p->release_next;
+		p = p->release_next;
+	}
+	spin_unlock_irqrestore(&kmmio_lock, flags);
+	/* This is the real RCU destroy call. */
+	call_rcu(&dr->rcu, rcu_free_kmmio_fault_pages);
+}
+
+/*
+ * Remove a kmmio probe. You have to synchronize_rcu() before you can be
+ * sure that the callbacks will not be called anymore.
+ *
+ * Unregistering a kmmio fault page has three steps:
+ * 1. release_kmmio_fault_page()
+ *    Disarm the page, wait a grace period to let all faults finish.
+ * 2. remove_kmmio_fault_pages()
+ *    Remove the pages from kmmio_page_table.
+ * 3. rcu_free_kmmio_fault_pages()
+ *    Actally free the kmmio_fault_page structs as with RCU.
+ */
 void unregister_kmmio_probe(struct kmmio_probe *p)
 {
 	unsigned long size = 0;
+	struct kmmio_fault_page *release_list = NULL;
+	struct kmmio_delayed_release *drelease;
 
 	spin_lock_irq(&kmmio_lock);
 	while (size < p->len) {
-		release_kmmio_fault_page(p->addr + size);
+		release_kmmio_fault_page(p->addr + size, &release_list);
 		size += PAGE_SIZE;
 	}
-	list_del(&p->list);
+	list_del_rcu(&p->list);
 	kmmio_count--;
 	spin_unlock_irq(&kmmio_lock);
-}
 
-/*
- * According to 2.6.20, mainly x86_64 arch:
- * This is being called from do_page_fault(), via the page fault notifier
- * chain. The chain is called for both user space faults and kernel space
- * faults (address >= TASK_SIZE64), except not on faults serviced by
- * vmalloc_fault().
- *
- * We may be in an interrupt or a critical section. Also prefecthing may
- * trigger a page fault. We may be in the middle of process switch.
- * The page fault hook functionality has put us inside RCU read lock.
- *
- * Local interrupts are disabled, so preemption cannot happen.
- * Do not enable interrupts, do not sleep, and watch out for other CPUs.
- */
-static int kmmio_page_fault(struct pt_regs *regs, unsigned long error_code,
-						unsigned long address)
-{
-	if (is_kmmio_active())
-		if (kmmio_handler(regs, address) == 1)
-			return -1;
-	return 0;
+	drelease = kmalloc(sizeof(*drelease), GFP_ATOMIC);
+	if (!drelease) {
+		pr_crit("kmmio: leaking kmmio_fault_page objects.\n");
+		return;
+	}
+	drelease->release_list = release_list;
+
+	/*
+	 * This is not really RCU here. We have just disarmed a set of
+	 * pages so that they cannot trigger page faults anymore. However,
+	 * we cannot remove the pages from kmmio_page_table,
+	 * because a probe hit might be in flight on another CPU. The
+	 * pages are collected into a list, and they will be removed from
+	 * kmmio_page_table when it is certain that no probe hit related to
+	 * these pages can be in flight. RCU grace period sounds like a
+	 * good choice.
+	 *
+	 * If we removed the pages too early, kmmio page fault handler might
+	 * not find the respective kmmio_fault_page and determine it's not
+	 * a kmmio fault, when it actually is. This would lead to madness.
+	 */
+	call_rcu(&drelease->rcu, remove_kmmio_fault_pages);
 }
+EXPORT_SYMBOL(unregister_kmmio_probe);
 
 static int kmmio_die_notifier(struct notifier_block *nb, unsigned long val,
 								void *args)
diff --git a/arch/x86/kernel/mmiotrace/kmmio.h b/arch/x86/kernel/mmiotrace/kmmio.h
deleted file mode 100644
index 85b7f68a3b8a..000000000000
--- a/arch/x86/kernel/mmiotrace/kmmio.h
+++ /dev/null
@@ -1,58 +0,0 @@
-#ifndef _LINUX_KMMIO_H
-#define _LINUX_KMMIO_H
-
-#include <linux/list.h>
-#include <linux/notifier.h>
-#include <linux/smp.h>
-#include <linux/types.h>
-#include <linux/ptrace.h>
-#include <linux/version.h>
-#include <linux/kdebug.h>
-
-struct kmmio_probe;
-struct kmmio_fault_page;
-struct pt_regs;
-
-typedef void (*kmmio_pre_handler_t)(struct kmmio_probe *,
-				struct pt_regs *, unsigned long addr);
-typedef void (*kmmio_post_handler_t)(struct kmmio_probe *,
-				unsigned long condition, struct pt_regs *);
-
-struct kmmio_probe {
-	struct list_head list;
-
-	/* start location of the probe point */
-	unsigned long addr;
-
-	/* length of the probe region */
-	unsigned long len;
-
-	 /* Called before addr is executed. */
-	kmmio_pre_handler_t pre_handler;
-
-	/* Called after addr is executed, unless... */
-	kmmio_post_handler_t post_handler;
-};
-
-struct kmmio_fault_page {
-	struct list_head list;
-
-	/* location of the fault page */
-	unsigned long page;
-
-	int count;
-};
-
-/* kmmio is active by some kmmio_probes? */
-static inline int is_kmmio_active(void)
-{
-	extern unsigned int kmmio_count;
-	return kmmio_count;
-}
-
-int init_kmmio(void);
-void cleanup_kmmio(void);
-int register_kmmio_probe(struct kmmio_probe *p);
-void unregister_kmmio_probe(struct kmmio_probe *p);
-
-#endif /* _LINUX_KMMIO_H */
diff --git a/arch/x86/kernel/mmiotrace/mmio-mod.c b/arch/x86/kernel/mmiotrace/mmio-mod.c
index f9c609266d83..e1a508588f03 100644
--- a/arch/x86/kernel/mmiotrace/mmio-mod.c
+++ b/arch/x86/kernel/mmiotrace/mmio-mod.c
@@ -32,7 +32,6 @@
 #include <asm/atomic.h>
 #include <linux/percpu.h>
 
-#include "kmmio.h"
 #include "pf_in.h"
 
 /* This app's relay channel files will appear in /debug/mmio-trace */
@@ -129,18 +128,17 @@ static void print_pte(unsigned long address)
 	pte_t *pte = lookup_address(address, &level);
 
 	if (!pte) {
-		printk(KERN_ERR "Error in %s: no pte for page 0x%08lx\n",
-						__FUNCTION__, address);
+		pr_err(MODULE_NAME ": Error in %s: no pte for page 0x%08lx\n",
+							__func__, address);
 		return;
 	}
 
 	if (level == PG_LEVEL_2M) {
-		printk(KERN_EMERG MODULE_NAME ": 4MB pages are not "
-						"currently supported: %lx\n",
-						address);
+		pr_emerg(MODULE_NAME ": 4MB pages are not currently "
+						"supported: %lx\n", address);
 		BUG();
 	}
-	printk(KERN_DEBUG MODULE_NAME ": pte for 0x%lx: 0x%lx 0x%lx\n",
+	pr_info(MODULE_NAME ": pte for 0x%lx: 0x%lx 0x%lx\n",
 					address, pte_val(*pte),
 					pte_val(*pte) & _PAGE_PRESENT);
 }
@@ -152,7 +150,7 @@ static void print_pte(unsigned long address)
 static void die_kmmio_nesting_error(struct pt_regs *regs, unsigned long addr)
 {
 	const struct trap_reason *my_reason = &get_cpu_var(pf_reason);
-	printk(KERN_EMERG MODULE_NAME ": unexpected fault for address: %lx, "
+	pr_emerg(MODULE_NAME ": unexpected fault for address: %lx, "
 					"last fault for address: %lx\n",
 					addr, my_reason->addr);
 	print_pte(addr);
@@ -160,20 +158,17 @@ static void die_kmmio_nesting_error(struct pt_regs *regs, unsigned long addr)
 	print_symbol(KERN_EMERG "faulting EIP is at %s\n", regs->ip);
 	print_symbol(KERN_EMERG "last faulting EIP was at %s\n",
 							my_reason->ip);
-	printk(KERN_EMERG
-			"eax: %08lx   ebx: %08lx   ecx: %08lx   edx: %08lx\n",
+	pr_emerg("eax: %08lx   ebx: %08lx   ecx: %08lx   edx: %08lx\n",
 			regs->ax, regs->bx, regs->cx, regs->dx);
-	printk(KERN_EMERG
-			"esi: %08lx   edi: %08lx   ebp: %08lx   esp: %08lx\n",
+	pr_emerg("esi: %08lx   edi: %08lx   ebp: %08lx   esp: %08lx\n",
 			regs->si, regs->di, regs->bp, regs->sp);
 #else
 	print_symbol(KERN_EMERG "faulting RIP is at %s\n", regs->ip);
 	print_symbol(KERN_EMERG "last faulting RIP was at %s\n",
 							my_reason->ip);
-	printk(KERN_EMERG "rax: %016lx   rcx: %016lx   rdx: %016lx\n",
+	pr_emerg("rax: %016lx   rcx: %016lx   rdx: %016lx\n",
 					regs->ax, regs->cx, regs->dx);
-	printk(KERN_EMERG "rsi: %016lx   rdi: %016lx   "
-				"rbp: %016lx   rsp: %016lx\n",
+	pr_emerg("rsi: %016lx   rdi: %016lx   rbp: %016lx   rsp: %016lx\n",
 				regs->si, regs->di, regs->bp, regs->sp);
 #endif
 	put_cpu_var(pf_reason);
@@ -251,10 +246,15 @@ static void post(struct kmmio_probe *p, unsigned long condition,
 	struct trap_reason *my_reason = &get_cpu_var(pf_reason);
 	struct mm_io_header_rw *my_trace = &get_cpu_var(cpu_trace);
 
+	/*
+	 * XXX: This might not get called, if the probe is removed while
+	 * trace hit is on flight.
+	 */
+
 	/* this should always return the active_trace count to 0 */
 	my_reason->active_traces--;
 	if (my_reason->active_traces) {
-		printk(KERN_EMERG MODULE_NAME ": unexpected post handler");
+		pr_emerg(MODULE_NAME ": unexpected post handler");
 		BUG();
 	}
 
@@ -283,16 +283,15 @@ static int subbuf_start_handler(struct rchan_buf *buf, void *subbuf,
 	atomic_t *drop = &per_cpu(dropped, cpu);
 	int count;
 	if (relay_buf_full(buf)) {
-		if (atomic_inc_return(drop) == 1) {
-			printk(KERN_ERR MODULE_NAME ": cpu %d buffer full!\n",
-									cpu);
-		}
+		if (atomic_inc_return(drop) == 1)
+			pr_err(MODULE_NAME ": cpu %d buffer full!\n", cpu);
 		return 0;
-	} else if ((count = atomic_read(drop))) {
-		printk(KERN_ERR MODULE_NAME
-					": cpu %d buffer no longer full, "
-					"missed %d events.\n",
-					cpu, count);
+	}
+	count = atomic_read(drop);
+	if (count) {
+		pr_err(MODULE_NAME ": cpu %d buffer no longer full, "
+						"missed %d events.\n",
+						cpu, count);
 		atomic_sub(count, drop);
 	}
 
@@ -407,8 +406,8 @@ static void ioremap_trace_core(unsigned long offset, unsigned long size,
 	/* Don't trace the low PCI/ISA area, it's always mapped.. */
 	if (!ISA_trace && (offset < ISA_END_ADDRESS) &&
 					(offset + size > ISA_START_ADDRESS)) {
-		printk(KERN_NOTICE MODULE_NAME ": Ignoring map of low "
-						"PCI/ISA area (0x%lx-0x%lx)\n",
+		pr_notice(MODULE_NAME ": Ignoring map of low PCI/ISA area "
+						"(0x%lx-0x%lx)\n",
 						offset, offset + size);
 		return;
 	}
@@ -418,7 +417,7 @@ static void ioremap_trace_core(unsigned long offset, unsigned long size,
 void __iomem *ioremap_cache_trace(unsigned long offset, unsigned long size)
 {
 	void __iomem *p = ioremap_cache(offset, size);
-	printk(KERN_DEBUG MODULE_NAME ": ioremap_cache(0x%lx, 0x%lx) = %p\n",
+	pr_debug(MODULE_NAME ": ioremap_cache(0x%lx, 0x%lx) = %p\n",
 							offset, size, p);
 	ioremap_trace_core(offset, size, p);
 	return p;
@@ -428,7 +427,7 @@ EXPORT_SYMBOL(ioremap_cache_trace);
 void __iomem *ioremap_nocache_trace(unsigned long offset, unsigned long size)
 {
 	void __iomem *p = ioremap_nocache(offset, size);
-	printk(KERN_DEBUG MODULE_NAME ": ioremap_nocache(0x%lx, 0x%lx) = %p\n",
+	pr_debug(MODULE_NAME ": ioremap_nocache(0x%lx, 0x%lx) = %p\n",
 							offset, size, p);
 	ioremap_trace_core(offset, size, p);
 	return p;
@@ -455,7 +454,7 @@ void iounmap_trace(volatile void __iomem *addr)
 	};
 	struct remap_trace *trace;
 	struct remap_trace *tmp;
-	printk(KERN_DEBUG MODULE_NAME ": Unmapping %p.\n", addr);
+	pr_debug(MODULE_NAME ": Unmapping %p.\n", addr);
 	record_timestamp(&event.header);
 
 	spin_lock(&trace_list_lock);
@@ -481,7 +480,7 @@ static void clear_trace_list(void)
 
 	spin_lock(&trace_list_lock);
 	list_for_each_entry_safe(trace, tmp, &trace_list, list) {
-		printk(KERN_WARNING MODULE_NAME ": purging non-iounmapped "
+		pr_warning(MODULE_NAME ": purging non-iounmapped "
 					"trace @0x%08lx, size 0x%lx.\n",
 					trace->probe.addr, trace->probe.len);
 		if (!nommiotrace)
@@ -500,39 +499,37 @@ static int __init init(void)
 
 	dir = debugfs_create_dir(APP_DIR, NULL);
 	if (!dir) {
-		printk(KERN_ERR MODULE_NAME
-				": Couldn't create relay app directory.\n");
+		pr_err(MODULE_NAME ": Couldn't create relay app directory.\n");
 		return -ENOMEM;
 	}
 
 	chan = create_channel(subbuf_size, n_subbufs);
 	if (!chan) {
 		debugfs_remove(dir);
-		printk(KERN_ERR MODULE_NAME
-				": relay app channel creation failed\n");
+		pr_err(MODULE_NAME ": relay app channel creation failed\n");
 		return -ENOMEM;
 	}
 
-	init_kmmio();
+	reference_kmmio();
 
 	proc_marker_file = create_proc_entry(MARKER_FILE, 0, NULL);
 	if (proc_marker_file)
 		proc_marker_file->write_proc = write_marker;
 
-	printk(KERN_DEBUG MODULE_NAME ": loaded.\n");
+	pr_debug(MODULE_NAME ": loaded.\n");
 	if (nommiotrace)
-		printk(KERN_DEBUG MODULE_NAME ": MMIO tracing disabled.\n");
+		pr_info(MODULE_NAME ": MMIO tracing disabled.\n");
 	if (ISA_trace)
-		printk(KERN_WARNING MODULE_NAME
-				": Warning! low ISA range will be traced.\n");
+		pr_warning(MODULE_NAME ": Warning! low ISA range will be "
+								"traced.\n");
 	return 0;
 }
 
 static void __exit cleanup(void)
 {
-	printk(KERN_DEBUG MODULE_NAME ": unload...\n");
+	pr_debug(MODULE_NAME ": unload...\n");
 	clear_trace_list();
-	cleanup_kmmio();
+	unreference_kmmio();
 	remove_proc_entry(MARKER_FILE, NULL);
 	destroy_channel();
 	if (dir)
diff --git a/arch/x86/kernel/mmiotrace/pf_in.c b/arch/x86/kernel/mmiotrace/pf_in.c
index 67ea520dde62..efa1911e20ca 100644
--- a/arch/x86/kernel/mmiotrace/pf_in.c
+++ b/arch/x86/kernel/mmiotrace/pf_in.c
@@ -19,7 +19,7 @@
  *
  */
 
-/*  $Id: pf_in.c,v 1.1.1.1 2002/11/12 05:56:32 brlock Exp $
+/*  Id: pf_in.c,v 1.1.1.1 2002/11/12 05:56:32 brlock Exp
  *  Copyright by Intel Crop., 2002
  *  Louis Zhuang (louis.zhuang@intel.com)
  *
diff --git a/arch/x86/kernel/mmiotrace/testmmiotrace.c b/arch/x86/kernel/mmiotrace/testmmiotrace.c
index 40e66b0e6480..5ecff578672b 100644
--- a/arch/x86/kernel/mmiotrace/testmmiotrace.c
+++ b/arch/x86/kernel/mmiotrace/testmmiotrace.c
@@ -41,8 +41,7 @@ static void do_test(void)
 {
 	void __iomem *p = ioremap_nocache_trace(mmio_address, 0x4000);
 	if (!p) {
-		printk(KERN_ERR MODULE_NAME ": could not ioremap IO memory, "
-							"aborting.\n");
+		pr_err(MODULE_NAME ": could not ioremap, aborting.\n");
 		return;
 	}
 	do_write_test(p);
@@ -53,14 +52,14 @@ static void do_test(void)
 static int __init init(void)
 {
 	if (mmio_address == 0) {
-		printk(KERN_ERR MODULE_NAME ": you have to use the module "
-						"argument mmio_address.\n");
-		printk(KERN_ERR MODULE_NAME ": DO NOT LOAD THIS MODULE UNLESS"
+		pr_err(MODULE_NAME ": you have to use the module argument "
+							"mmio_address.\n");
+		pr_err(MODULE_NAME ": DO NOT LOAD THIS MODULE UNLESS"
 				" YOU REALLY KNOW WHAT YOU ARE DOING!\n");
 		return -ENXIO;
 	}
 
-	printk(KERN_WARNING MODULE_NAME ": WARNING: mapping 16 kB @ 0x%08lx "
+	pr_warning(MODULE_NAME ": WARNING: mapping 16 kB @ 0x%08lx "
 					"in PCI address space, and writing "
 					"rubbish in there.\n", mmio_address);
 	do_test();
@@ -69,7 +68,7 @@ static int __init init(void)
 
 static void __exit cleanup(void)
 {
-	printk(KERN_DEBUG MODULE_NAME ": unloaded.\n");
+	pr_debug(MODULE_NAME ": unloaded.\n");
 }
 
 module_init(init);
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index e9a086a1a9ff..8c828a68d3b6 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -10,6 +10,7 @@
 #include <linux/string.h>
 #include <linux/types.h>
 #include <linux/ptrace.h>
+#include <linux/mmiotrace.h>
 #include <linux/mman.h>
 #include <linux/mm.h>
 #include <linux/smp.h>
@@ -49,60 +50,14 @@
 #define PF_RSVD		(1<<3)
 #define PF_INSTR	(1<<4)
 
-#ifdef CONFIG_MMIOTRACE_HOOKS
-static pf_handler_func mmiotrace_pf_handler; /* protected by RCU */
-static DEFINE_SPINLOCK(mmiotrace_handler_lock);
-
-int mmiotrace_register_pf(pf_handler_func new_pfh)
-{
-	int ret = 0;
-	unsigned long flags;
-	spin_lock_irqsave(&mmiotrace_handler_lock, flags);
-	if (mmiotrace_pf_handler)
-		ret = -EBUSY;
-	else
-		mmiotrace_pf_handler = new_pfh;
-	spin_unlock_irqrestore(&mmiotrace_handler_lock, flags);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(mmiotrace_register_pf);
-
-/**
- * mmiotrace_unregister_pf:
- * The caller must ensure @old_pfh is not in use anymore before freeing it.
- * This function does not guarantee it. The handler function pointer is
- * protected by RCU, so you can do this by e.g. calling synchronize_rcu().
- */
-int mmiotrace_unregister_pf(pf_handler_func old_pfh)
-{
-	int ret = 0;
-	unsigned long flags;
-	spin_lock_irqsave(&mmiotrace_handler_lock, flags);
-	if (mmiotrace_pf_handler != old_pfh)
-		ret = -EPERM;
-	else
-		mmiotrace_pf_handler = NULL;
-	spin_unlock_irqrestore(&mmiotrace_handler_lock, flags);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(mmiotrace_unregister_pf);
-#endif /* CONFIG_MMIOTRACE_HOOKS */
-
-/* returns non-zero if do_page_fault() should return */
-static inline int call_mmiotrace(struct pt_regs *regs,
-					unsigned long error_code,
-					unsigned long address)
+static inline int kmmio_fault(struct pt_regs *regs, unsigned long addr)
 {
 #ifdef CONFIG_MMIOTRACE_HOOKS
-	int ret = 0;
-	rcu_read_lock();
-	if (mmiotrace_pf_handler)
-		ret = mmiotrace_pf_handler(regs, error_code, address);
-	rcu_read_unlock();
-	return ret;
-#else
-	return 0;
+	if (unlikely(is_kmmio_active()))
+		if (kmmio_handler(regs, addr) == 1)
+			return -1;
 #endif
+	return 0;
 }
 
 static inline int notify_page_fault(struct pt_regs *regs)
@@ -657,7 +612,7 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 
 	if (notify_page_fault(regs))
 		return;
-	if (call_mmiotrace(regs, error_code, address))
+	if (unlikely(kmmio_fault(regs, address)))
 		return;
 
 	/*
diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index 7063281040da..96651bb59ba1 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -35,11 +35,4 @@ extern void show_regs(struct pt_regs *regs);
 extern unsigned long oops_begin(void);
 extern void oops_end(unsigned long, struct pt_regs *, int signr);
 
-typedef int (*pf_handler_func)(struct pt_regs *regs,
-				unsigned long error_code,
-				unsigned long address);
-
-extern int mmiotrace_register_pf(pf_handler_func new_pfh);
-extern int mmiotrace_unregister_pf(pf_handler_func old_pfh);
-
 #endif
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index 6ec288f1fe24..d87a6cd8b686 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -3,6 +3,44 @@
 
 #include <asm/types.h>
 
+#ifdef __KERNEL__
+
+#include <linux/list.h>
+
+struct kmmio_probe;
+struct pt_regs;
+
+typedef void (*kmmio_pre_handler_t)(struct kmmio_probe *,
+				struct pt_regs *, unsigned long addr);
+typedef void (*kmmio_post_handler_t)(struct kmmio_probe *,
+				unsigned long condition, struct pt_regs *);
+
+struct kmmio_probe {
+	struct list_head list;
+	unsigned long addr; /* start location of the probe point */
+	unsigned long len; /* length of the probe region */
+	kmmio_pre_handler_t pre_handler; /* Called before addr is executed. */
+	kmmio_post_handler_t post_handler; /* Called after addr is executed */
+};
+
+/* kmmio is active by some kmmio_probes? */
+static inline int is_kmmio_active(void)
+{
+	extern unsigned int kmmio_count;
+	return kmmio_count;
+}
+
+extern void reference_kmmio(void);
+extern void unreference_kmmio(void);
+extern int register_kmmio_probe(struct kmmio_probe *p);
+extern void unregister_kmmio_probe(struct kmmio_probe *p);
+
+/* Called from page fault handler. */
+extern int kmmio_handler(struct pt_regs *regs, unsigned long addr);
+
+#endif /* __KERNEL__ */
+
+
 /*
  * If you change anything here, you must bump MMIO_VERSION.
  * This is the relay data format for user space.
-- 
cgit v1.2.3


From d61fc44853f46fb002228b18aa5f30db21fcd4ac Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:20:57 +0200
Subject: x86: mmiotrace, preview 2

Kconfig.debug, Makefile and testmmiotrace.c style fixes.
Use real mutex instead of mutex.
Fix failure path in register probe func.
kmmio: RCU read-locked over single stepping.
Generate mapping id's.
Make mmio-mod.c built-in and rewrite its locking.
Add debugfs file to enable/disable mmiotracing.
kmmio: use irqsave spinlocks.
Lots of cleanups in mmio-mod.c
Marker file moved from /proc into debugfs.
Call mmiotrace entrypoints directly from ioremap.c.

Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig.debug                    |  20 +-
 arch/x86/kernel/mmiotrace/Makefile        |   2 +-
 arch/x86/kernel/mmiotrace/kmmio.c         |  72 +++---
 arch/x86/kernel/mmiotrace/mmio-mod.c      | 397 ++++++++++++++++++++----------
 arch/x86/kernel/mmiotrace/testmmiotrace.c |  15 +-
 arch/x86/mm/ioremap.c                     |   9 +-
 include/linux/mmiotrace.h                 |  18 +-
 7 files changed, 332 insertions(+), 201 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 9491c0ae03a3..aa0d6462b1fc 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -170,22 +170,19 @@ config IOMMU_LEAK
 
 config MMIOTRACE_HOOKS
 	bool
-	default n
 
 config MMIOTRACE
-	tristate "Memory mapped IO tracing"
+	bool "Memory mapped IO tracing"
 	depends on DEBUG_KERNEL && RELAY && DEBUG_FS
 	select MMIOTRACE_HOOKS
-	default n
+	default y
 	help
-	  This will build a kernel module called mmiotrace.
-	  Making this a built-in is heavily discouraged.
-
-	  Mmiotrace traces Memory Mapped I/O access and is meant for debugging
-	  and reverse engineering. The kernel module offers wrapped
-	  versions of the ioremap family of functions. The driver to be traced
-	  must be modified to call these wrappers. A user space program is
-	  required to collect the MMIO data.
+	  Mmiotrace traces Memory Mapped I/O access and is meant for
+	  debugging and reverse engineering. It is called from the ioremap
+	  implementation and works via page faults. A user space program is
+	  required to collect the MMIO data from debugfs files.
+	  Tracing is disabled by default and can be enabled from a debugfs
+	  file.
 
 	  See http://nouveau.freedesktop.org/wiki/MmioTrace
 	  If you are not helping to develop drivers, say N.
@@ -193,7 +190,6 @@ config MMIOTRACE
 config MMIOTRACE_TEST
 	tristate "Test module for mmiotrace"
 	depends on MMIOTRACE && m
-	default n
 	help
 	  This is a dumb module for testing mmiotrace. It is very dangerous
 	  as it will write garbage to IO memory starting at a given address.
diff --git a/arch/x86/kernel/mmiotrace/Makefile b/arch/x86/kernel/mmiotrace/Makefile
index cf1e747b463e..dbcd8d50fb8d 100644
--- a/arch/x86/kernel/mmiotrace/Makefile
+++ b/arch/x86/kernel/mmiotrace/Makefile
@@ -1,4 +1,4 @@
 obj-$(CONFIG_MMIOTRACE_HOOKS)	+= kmmio.o
 obj-$(CONFIG_MMIOTRACE)		+= mmiotrace.o
-mmiotrace-objs			:= pf_in.o mmio-mod.o
+mmiotrace-y			:= pf_in.o mmio-mod.o
 obj-$(CONFIG_MMIOTRACE_TEST)	+= testmmiotrace.o
diff --git a/arch/x86/kernel/mmiotrace/kmmio.c b/arch/x86/kernel/mmiotrace/kmmio.c
index 539a9b19588f..efb467933087 100644
--- a/arch/x86/kernel/mmiotrace/kmmio.c
+++ b/arch/x86/kernel/mmiotrace/kmmio.c
@@ -19,6 +19,7 @@
 #include <linux/preempt.h>
 #include <linux/percpu.h>
 #include <linux/kdebug.h>
+#include <linux/mutex.h>
 #include <asm/io.h>
 #include <asm/cacheflush.h>
 #include <asm/errno.h>
@@ -59,7 +60,7 @@ struct kmmio_context {
 static int kmmio_die_notifier(struct notifier_block *nb, unsigned long val,
 								void *args);
 
-static DECLARE_MUTEX(kmmio_init_mutex);
+static DEFINE_MUTEX(kmmio_init_mutex);
 static DEFINE_SPINLOCK(kmmio_lock);
 
 /* These are protected by kmmio_lock */
@@ -90,7 +91,7 @@ static struct notifier_block nb_die = {
  */
 void reference_kmmio(void)
 {
-	down(&kmmio_init_mutex);
+	mutex_lock(&kmmio_init_mutex);
 	spin_lock_irq(&kmmio_lock);
 	if (!kmmio_initialized) {
 		int i;
@@ -101,7 +102,7 @@ void reference_kmmio(void)
 	}
 	kmmio_initialized++;
 	spin_unlock_irq(&kmmio_lock);
-	up(&kmmio_init_mutex);
+	mutex_unlock(&kmmio_init_mutex);
 }
 EXPORT_SYMBOL_GPL(reference_kmmio);
 
@@ -115,7 +116,7 @@ void unreference_kmmio(void)
 {
 	bool unreg = false;
 
-	down(&kmmio_init_mutex);
+	mutex_lock(&kmmio_init_mutex);
 	spin_lock_irq(&kmmio_lock);
 
 	if (kmmio_initialized == 1) {
@@ -128,7 +129,7 @@ void unreference_kmmio(void)
 
 	if (unreg)
 		unregister_die_notifier(&nb_die); /* calls sync_rcu() */
-	up(&kmmio_init_mutex);
+	mutex_unlock(&kmmio_init_mutex);
 }
 EXPORT_SYMBOL(unreference_kmmio);
 
@@ -244,17 +245,13 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 	 * Preemption is now disabled to prevent process switch during
 	 * single stepping. We can only handle one active kmmio trace
 	 * per cpu, so ensure that we finish it before something else
-	 * gets to run.
-	 *
-	 * XXX what if an interrupt occurs between returning from
-	 * do_page_fault() and entering the single-step exception handler?
-	 * And that interrupt triggers a kmmio trap?
-	 * XXX If we tracing an interrupt service routine or whatever, is
-	 * this enough to keep it on the current cpu?
+	 * gets to run. We also hold the RCU read lock over single
+	 * stepping to avoid looking up the probe and kmmio_fault_page
+	 * again.
 	 */
 	preempt_disable();
-
 	rcu_read_lock();
+
 	faultpage = get_kmmio_fault_page(addr);
 	if (!faultpage) {
 		/*
@@ -287,14 +284,24 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 	if (ctx->probe && ctx->probe->pre_handler)
 		ctx->probe->pre_handler(ctx->probe, regs, addr);
 
+	/*
+	 * Enable single-stepping and disable interrupts for the faulting
+	 * context. Local interrupts must not get enabled during stepping.
+	 */
 	regs->flags |= TF_MASK;
 	regs->flags &= ~IF_MASK;
 
 	/* Now we set present bit in PTE and single step. */
 	disarm_kmmio_fault_page(ctx->fpage->page, NULL);
 
+	/*
+	 * If another cpu accesses the same page while we are stepping,
+	 * the access will not be caught. It will simply succeed and the
+	 * only downside is we lose the event. If this becomes a problem,
+	 * the user should drop to single cpu before tracing.
+	 */
+
 	put_cpu_var(kmmio_ctx);
-	rcu_read_unlock();
 	return 1;
 
 no_kmmio_ctx:
@@ -313,32 +320,15 @@ no_kmmio:
 static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
 {
 	int ret = 0;
-	struct kmmio_probe *probe;
-	struct kmmio_fault_page *faultpage;
 	struct kmmio_context *ctx = &get_cpu_var(kmmio_ctx);
 
 	if (!ctx->active)
 		goto out;
 
-	rcu_read_lock();
-
-	faultpage = get_kmmio_fault_page(ctx->addr);
-	probe = get_kmmio_probe(ctx->addr);
-	if (faultpage != ctx->fpage || probe != ctx->probe) {
-		/*
-		 * The trace setup changed after kmmio_handler() and before
-		 * running this respective post handler. User does not want
-		 * the result anymore.
-		 */
-		ctx->probe = NULL;
-		ctx->fpage = NULL;
-	}
-
 	if (ctx->probe && ctx->probe->post_handler)
 		ctx->probe->post_handler(ctx->probe, condition, regs);
 
-	if (ctx->fpage)
-		arm_kmmio_fault_page(ctx->fpage->page, NULL);
+	arm_kmmio_fault_page(ctx->fpage->page, NULL);
 
 	regs->flags &= ~TF_MASK;
 	regs->flags |= ctx->saved_flags;
@@ -346,6 +336,7 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
 	/* These were acquired in kmmio_handler(). */
 	ctx->active--;
 	BUG_ON(ctx->active);
+	rcu_read_unlock();
 	preempt_enable_no_resched();
 
 	/*
@@ -355,8 +346,6 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
 	 */
 	if (!(regs->flags & TF_MASK))
 		ret = 1;
-
-	rcu_read_unlock();
 out:
 	put_cpu_var(kmmio_ctx);
 	return ret;
@@ -411,15 +400,16 @@ static void release_kmmio_fault_page(unsigned long page,
 
 int register_kmmio_probe(struct kmmio_probe *p)
 {
+	unsigned long flags;
 	int ret = 0;
 	unsigned long size = 0;
 
-	spin_lock_irq(&kmmio_lock);
-	kmmio_count++;
+	spin_lock_irqsave(&kmmio_lock, flags);
 	if (get_kmmio_probe(p->addr)) {
 		ret = -EEXIST;
 		goto out;
 	}
+	kmmio_count++;
 	list_add_rcu(&p->list, &kmmio_probes);
 	while (size < p->len) {
 		if (add_kmmio_fault_page(p->addr + size))
@@ -427,7 +417,7 @@ int register_kmmio_probe(struct kmmio_probe *p)
 		size += PAGE_SIZE;
 	}
 out:
-	spin_unlock_irq(&kmmio_lock);
+	spin_unlock_irqrestore(&kmmio_lock, flags);
 	/*
 	 * XXX: What should I do here?
 	 * Here was a call to global_flush_tlb(), but it does not exist
@@ -478,7 +468,8 @@ static void remove_kmmio_fault_pages(struct rcu_head *head)
 
 /*
  * Remove a kmmio probe. You have to synchronize_rcu() before you can be
- * sure that the callbacks will not be called anymore.
+ * sure that the callbacks will not be called anymore. Only after that
+ * you may actually release your struct kmmio_probe.
  *
  * Unregistering a kmmio fault page has three steps:
  * 1. release_kmmio_fault_page()
@@ -490,18 +481,19 @@ static void remove_kmmio_fault_pages(struct rcu_head *head)
  */
 void unregister_kmmio_probe(struct kmmio_probe *p)
 {
+	unsigned long flags;
 	unsigned long size = 0;
 	struct kmmio_fault_page *release_list = NULL;
 	struct kmmio_delayed_release *drelease;
 
-	spin_lock_irq(&kmmio_lock);
+	spin_lock_irqsave(&kmmio_lock, flags);
 	while (size < p->len) {
 		release_kmmio_fault_page(p->addr + size, &release_list);
 		size += PAGE_SIZE;
 	}
 	list_del_rcu(&p->list);
 	kmmio_count--;
-	spin_unlock_irq(&kmmio_lock);
+	spin_unlock_irqrestore(&kmmio_lock, flags);
 
 	drelease = kmalloc(sizeof(*drelease), GFP_ATOMIC);
 	if (!drelease) {
diff --git a/arch/x86/kernel/mmiotrace/mmio-mod.c b/arch/x86/kernel/mmiotrace/mmio-mod.c
index e1a508588f03..738644061e4e 100644
--- a/arch/x86/kernel/mmiotrace/mmio-mod.c
+++ b/arch/x86/kernel/mmiotrace/mmio-mod.c
@@ -19,6 +19,8 @@
  *
  * Derived from the read-mod example from relay-examples by Tom Zanussi.
  */
+#define DEBUG 1
+
 #include <linux/module.h>
 #include <linux/relay.h>
 #include <linux/debugfs.h>
@@ -34,12 +36,12 @@
 
 #include "pf_in.h"
 
-/* This app's relay channel files will appear in /debug/mmio-trace */
-#define APP_DIR		"mmio-trace"
-/* the marker injection file in /proc */
-#define MARKER_FILE     "mmio-marker"
+#define NAME "mmiotrace: "
 
-#define MODULE_NAME     "mmiotrace"
+/* This app's relay channel files will appear in /debug/mmio-trace */
+static const char APP_DIR[] = "mmio-trace";
+/* the marker injection file in /debug/APP_DIR */
+static const char MARKER_FILE[] = "mmio-marker";
 
 struct trap_reason {
 	unsigned long addr;
@@ -48,6 +50,15 @@ struct trap_reason {
 	int active_traces;
 };
 
+struct remap_trace {
+	struct list_head list;
+	struct kmmio_probe probe;
+	unsigned long phys;
+	unsigned long id;
+};
+
+static const size_t subbuf_size = 256*1024;
+
 /* Accessed per-cpu. */
 static DEFINE_PER_CPU(struct trap_reason, pf_reason);
 static DEFINE_PER_CPU(struct mm_io_header_rw, cpu_trace);
@@ -55,33 +66,53 @@ static DEFINE_PER_CPU(struct mm_io_header_rw, cpu_trace);
 /* Access to this is not per-cpu. */
 static DEFINE_PER_CPU(atomic_t, dropped);
 
-static struct file_operations mmio_fops = {
-	.owner = THIS_MODULE,
-};
+static struct dentry *dir;
+static struct dentry *enabled_file;
+static struct dentry *marker_file;
 
-static const size_t subbuf_size = 256*1024;
+static DEFINE_MUTEX(mmiotrace_mutex);
+static DEFINE_SPINLOCK(trace_lock);
+static atomic_t mmiotrace_enabled;
+static LIST_HEAD(trace_list);		/* struct remap_trace */
 static struct rchan *chan;
-static struct dentry *dir;
-static struct proc_dir_entry *proc_marker_file;
+
+/*
+ * Locking in this file:
+ * - mmiotrace_mutex enforces enable/disable_mmiotrace() critical sections.
+ * - mmiotrace_enabled may be modified only when holding mmiotrace_mutex
+ *   and trace_lock.
+ * - Routines depending on is_enabled() must take trace_lock.
+ * - trace_list users must hold trace_lock.
+ * - is_enabled() guarantees that chan is valid.
+ * - pre/post callbacks assume the effect of is_enabled() being true.
+ */
 
 /* module parameters */
-static unsigned int      n_subbufs = 32*4;
-static unsigned long filter_offset;
-static int                 nommiotrace;
-static int               ISA_trace;
-static int                trace_pc;
+static unsigned int	n_subbufs = 32*4;
+static unsigned long	filter_offset;
+static int		nommiotrace;
+static int		ISA_trace;
+static int		trace_pc;
+static int		enable_now;
 
 module_param(n_subbufs, uint, 0);
 module_param(filter_offset, ulong, 0);
 module_param(nommiotrace, bool, 0);
 module_param(ISA_trace, bool, 0);
 module_param(trace_pc, bool, 0);
+module_param(enable_now, bool, 0);
 
 MODULE_PARM_DESC(n_subbufs, "Number of 256kB buffers, default 128.");
 MODULE_PARM_DESC(filter_offset, "Start address of traced mappings.");
 MODULE_PARM_DESC(nommiotrace, "Disable actual MMIO tracing.");
 MODULE_PARM_DESC(ISA_trace, "Do not exclude the low ISA range.");
 MODULE_PARM_DESC(trace_pc, "Record address of faulting instructions.");
+MODULE_PARM_DESC(enable_now, "Start mmiotracing immediately on module load.");
+
+static bool is_enabled(void)
+{
+	return atomic_read(&mmiotrace_enabled);
+}
 
 static void record_timestamp(struct mm_io_header *header)
 {
@@ -93,15 +124,15 @@ static void record_timestamp(struct mm_io_header *header)
 }
 
 /*
- * Write callback for the /proc entry:
+ * Write callback for the debugfs entry:
  * Read a marker and write it to the mmio trace log
  */
-static int write_marker(struct file *file, const char __user *buffer,
-					unsigned long count, void *data)
+static ssize_t write_marker(struct file *file, const char __user *buffer,
+						size_t count, loff_t *ppos)
 {
 	char *event = NULL;
 	struct mm_io_header *headp;
-	int len = (count > 65535) ? 65535 : count;
+	ssize_t len = (count > 65535) ? 65535 : count;
 
 	event = kzalloc(sizeof(*headp) + len, GFP_KERNEL);
 	if (!event)
@@ -117,7 +148,12 @@ static int write_marker(struct file *file, const char __user *buffer,
 		return -EFAULT;
 	}
 
-	relay_write(chan, event, sizeof(*headp) + len);
+	spin_lock_irq(&trace_lock);
+	if (is_enabled())
+		relay_write(chan, event, sizeof(*headp) + len);
+	else
+		len = -EINVAL;
+	spin_unlock_irq(&trace_lock);
 	kfree(event);
 	return len;
 }
@@ -128,19 +164,18 @@ static void print_pte(unsigned long address)
 	pte_t *pte = lookup_address(address, &level);
 
 	if (!pte) {
-		pr_err(MODULE_NAME ": Error in %s: no pte for page 0x%08lx\n",
+		pr_err(NAME "Error in %s: no pte for page 0x%08lx\n",
 							__func__, address);
 		return;
 	}
 
 	if (level == PG_LEVEL_2M) {
-		pr_emerg(MODULE_NAME ": 4MB pages are not currently "
-						"supported: %lx\n", address);
+		pr_emerg(NAME "4MB pages are not currently supported: "
+							"0x%08lx\n", address);
 		BUG();
 	}
-	pr_info(MODULE_NAME ": pte for 0x%lx: 0x%lx 0x%lx\n",
-					address, pte_val(*pte),
-					pte_val(*pte) & _PAGE_PRESENT);
+	pr_info(NAME "pte for 0x%lx: 0x%lx 0x%lx\n", address, pte_val(*pte),
+						pte_val(*pte) & _PAGE_PRESENT);
 }
 
 /*
@@ -150,22 +185,18 @@ static void print_pte(unsigned long address)
 static void die_kmmio_nesting_error(struct pt_regs *regs, unsigned long addr)
 {
 	const struct trap_reason *my_reason = &get_cpu_var(pf_reason);
-	pr_emerg(MODULE_NAME ": unexpected fault for address: %lx, "
-					"last fault for address: %lx\n",
+	pr_emerg(NAME "unexpected fault for address: 0x%08lx, "
+					"last fault for address: 0x%08lx\n",
 					addr, my_reason->addr);
 	print_pte(addr);
+	print_symbol(KERN_EMERG "faulting IP is at %s\n", regs->ip);
+	print_symbol(KERN_EMERG "last faulting IP was at %s\n", my_reason->ip);
 #ifdef __i386__
-	print_symbol(KERN_EMERG "faulting EIP is at %s\n", regs->ip);
-	print_symbol(KERN_EMERG "last faulting EIP was at %s\n",
-							my_reason->ip);
 	pr_emerg("eax: %08lx   ebx: %08lx   ecx: %08lx   edx: %08lx\n",
 			regs->ax, regs->bx, regs->cx, regs->dx);
 	pr_emerg("esi: %08lx   edi: %08lx   ebp: %08lx   esp: %08lx\n",
 			regs->si, regs->di, regs->bp, regs->sp);
 #else
-	print_symbol(KERN_EMERG "faulting RIP is at %s\n", regs->ip);
-	print_symbol(KERN_EMERG "last faulting RIP was at %s\n",
-							my_reason->ip);
 	pr_emerg("rax: %016lx   rcx: %016lx   rdx: %016lx\n",
 					regs->ax, regs->cx, regs->dx);
 	pr_emerg("rsi: %016lx   rdi: %016lx   rbp: %016lx   rsp: %016lx\n",
@@ -197,6 +228,10 @@ static void pre(struct kmmio_probe *p, struct pt_regs *regs,
 	my_trace->header.pid = 0;
 	my_trace->header.data_len = sizeof(struct mm_io_rw);
 	my_trace->rw.address = addr;
+	/*
+	 * struct remap_trace *trace = p->user_data;
+	 * phys = addr - trace->probe.addr + trace->phys;
+	 */
 
 	/*
 	 * Only record the program counter when requested.
@@ -246,15 +281,10 @@ static void post(struct kmmio_probe *p, unsigned long condition,
 	struct trap_reason *my_reason = &get_cpu_var(pf_reason);
 	struct mm_io_header_rw *my_trace = &get_cpu_var(cpu_trace);
 
-	/*
-	 * XXX: This might not get called, if the probe is removed while
-	 * trace hit is on flight.
-	 */
-
 	/* this should always return the active_trace count to 0 */
 	my_reason->active_traces--;
 	if (my_reason->active_traces) {
-		pr_emerg(MODULE_NAME ": unexpected post handler");
+		pr_emerg(NAME "unexpected post handler");
 		BUG();
 	}
 
@@ -284,20 +314,23 @@ static int subbuf_start_handler(struct rchan_buf *buf, void *subbuf,
 	int count;
 	if (relay_buf_full(buf)) {
 		if (atomic_inc_return(drop) == 1)
-			pr_err(MODULE_NAME ": cpu %d buffer full!\n", cpu);
+			pr_err(NAME "cpu %d buffer full!\n", cpu);
 		return 0;
 	}
 	count = atomic_read(drop);
 	if (count) {
-		pr_err(MODULE_NAME ": cpu %d buffer no longer full, "
-						"missed %d events.\n",
-						cpu, count);
+		pr_err(NAME "cpu %d buffer no longer full, missed %d events.\n",
+								cpu, count);
 		atomic_sub(count, drop);
 	}
 
 	return 1;
 }
 
+static struct file_operations mmio_fops = {
+	.owner = THIS_MODULE,
+};
+
 /* file_create() callback.  Creates relay file in debugfs. */
 static struct dentry *create_buf_file_handler(const char *filename,
 						struct dentry *parent,
@@ -333,34 +366,10 @@ static struct rchan_callbacks relay_callbacks = {
 	.remove_buf_file = remove_buf_file_handler,
 };
 
-/*
- * create_channel - creates channel /debug/APP_DIR/cpuXXX
- * Returns channel on success, NULL otherwise
- */
-static struct rchan *create_channel(unsigned size, unsigned n)
-{
-	return relay_open("cpu", dir, size, n, &relay_callbacks, NULL);
-}
-
-/* destroy_channel - destroys channel /debug/APP_DIR/cpuXXX */
-static void destroy_channel(void)
-{
-	if (chan) {
-		relay_close(chan);
-		chan = NULL;
-	}
-}
-
-struct remap_trace {
-	struct list_head list;
-	struct kmmio_probe probe;
-};
-static LIST_HEAD(trace_list);
-static DEFINE_SPINLOCK(trace_list_lock);
-
-static void do_ioremap_trace_core(unsigned long offset, unsigned long size,
+static void ioremap_trace_core(unsigned long offset, unsigned long size,
 							void __iomem *addr)
 {
+	static atomic_t next_id;
 	struct remap_trace *trace = kmalloc(sizeof(*trace), GFP_KERNEL);
 	struct mm_io_header_map event = {
 		.header = {
@@ -380,61 +389,49 @@ static void do_ioremap_trace_core(unsigned long offset, unsigned long size,
 	};
 	record_timestamp(&event.header);
 
+	if (!trace) {
+		pr_err(NAME "kmalloc failed in ioremap\n");
+		return;
+	}
+
 	*trace = (struct remap_trace) {
 		.probe = {
 			.addr = (unsigned long)addr,
 			.len = size,
 			.pre_handler = pre,
 			.post_handler = post,
-		}
+			.user_data = trace
+		},
+		.phys = offset,
+		.id = atomic_inc_return(&next_id)
 	};
 
+	spin_lock_irq(&trace_lock);
+	if (!is_enabled())
+		goto not_enabled;
+
 	relay_write(chan, &event, sizeof(event));
-	spin_lock(&trace_list_lock);
 	list_add_tail(&trace->list, &trace_list);
-	spin_unlock(&trace_list_lock);
 	if (!nommiotrace)
 		register_kmmio_probe(&trace->probe);
+
+not_enabled:
+	spin_unlock_irq(&trace_lock);
 }
 
-static void ioremap_trace_core(unsigned long offset, unsigned long size,
-							void __iomem *addr)
+void
+mmiotrace_ioremap(unsigned long offset, unsigned long size, void __iomem *addr)
 {
-	if ((filter_offset) && (offset != filter_offset))
+	if (!is_enabled()) /* recheck and proper locking in *_core() */
 		return;
 
-	/* Don't trace the low PCI/ISA area, it's always mapped.. */
-	if (!ISA_trace && (offset < ISA_END_ADDRESS) &&
-					(offset + size > ISA_START_ADDRESS)) {
-		pr_notice(MODULE_NAME ": Ignoring map of low PCI/ISA area "
-						"(0x%lx-0x%lx)\n",
-						offset, offset + size);
+	pr_debug(NAME "ioremap_*(0x%lx, 0x%lx) = %p\n", offset, size, addr);
+	if ((filter_offset) && (offset != filter_offset))
 		return;
-	}
-	do_ioremap_trace_core(offset, size, addr);
-}
-
-void __iomem *ioremap_cache_trace(unsigned long offset, unsigned long size)
-{
-	void __iomem *p = ioremap_cache(offset, size);
-	pr_debug(MODULE_NAME ": ioremap_cache(0x%lx, 0x%lx) = %p\n",
-							offset, size, p);
-	ioremap_trace_core(offset, size, p);
-	return p;
+	ioremap_trace_core(offset, size, addr);
 }
-EXPORT_SYMBOL(ioremap_cache_trace);
 
-void __iomem *ioremap_nocache_trace(unsigned long offset, unsigned long size)
-{
-	void __iomem *p = ioremap_nocache(offset, size);
-	pr_debug(MODULE_NAME ": ioremap_nocache(0x%lx, 0x%lx) = %p\n",
-							offset, size, p);
-	ioremap_trace_core(offset, size, p);
-	return p;
-}
-EXPORT_SYMBOL(ioremap_nocache_trace);
-
-void iounmap_trace(volatile void __iomem *addr)
+static void iounmap_trace_core(volatile void __iomem *addr)
 {
 	struct mm_io_header_map event = {
 		.header = {
@@ -454,84 +451,212 @@ void iounmap_trace(volatile void __iomem *addr)
 	};
 	struct remap_trace *trace;
 	struct remap_trace *tmp;
-	pr_debug(MODULE_NAME ": Unmapping %p.\n", addr);
+	struct remap_trace *found_trace = NULL;
+
+	pr_debug(NAME "Unmapping %p.\n", addr);
 	record_timestamp(&event.header);
 
-	spin_lock(&trace_list_lock);
+	spin_lock_irq(&trace_lock);
+	if (!is_enabled())
+		goto not_enabled;
+
 	list_for_each_entry_safe(trace, tmp, &trace_list, list) {
 		if ((unsigned long)addr == trace->probe.addr) {
 			if (!nommiotrace)
 				unregister_kmmio_probe(&trace->probe);
 			list_del(&trace->list);
-			kfree(trace);
+			found_trace = trace;
 			break;
 		}
 	}
-	spin_unlock(&trace_list_lock);
 	relay_write(chan, &event, sizeof(event));
-	iounmap(addr);
+
+not_enabled:
+	spin_unlock_irq(&trace_lock);
+	if (found_trace) {
+		synchronize_rcu(); /* unregister_kmmio_probe() requirement */
+		kfree(found_trace);
+	}
+}
+
+void mmiotrace_iounmap(volatile void __iomem *addr)
+{
+	might_sleep();
+	if (is_enabled()) /* recheck and proper locking in *_core() */
+		iounmap_trace_core(addr);
 }
-EXPORT_SYMBOL(iounmap_trace);
 
 static void clear_trace_list(void)
 {
 	struct remap_trace *trace;
 	struct remap_trace *tmp;
 
-	spin_lock(&trace_list_lock);
-	list_for_each_entry_safe(trace, tmp, &trace_list, list) {
-		pr_warning(MODULE_NAME ": purging non-iounmapped "
+	/*
+	 * No locking required, because the caller ensures we are in a
+	 * critical section via mutex, and is_enabled() is false,
+	 * i.e. nothing can traverse or modify this list.
+	 * Caller also ensures is_enabled() cannot change.
+	 */
+	list_for_each_entry(trace, &trace_list, list) {
+		pr_notice(NAME "purging non-iounmapped "
 					"trace @0x%08lx, size 0x%lx.\n",
 					trace->probe.addr, trace->probe.len);
 		if (!nommiotrace)
 			unregister_kmmio_probe(&trace->probe);
+	}
+	synchronize_rcu(); /* unregister_kmmio_probe() requirement */
+
+	list_for_each_entry_safe(trace, tmp, &trace_list, list) {
 		list_del(&trace->list);
 		kfree(trace);
+	}
+}
+
+static ssize_t read_enabled_file_bool(struct file *file,
+		char __user *user_buf, size_t count, loff_t *ppos)
+{
+	char buf[3];
+
+	if (is_enabled())
+		buf[0] = '1';
+	else
+		buf[0] = '0';
+	buf[1] = '\n';
+	buf[2] = '\0';
+	return simple_read_from_buffer(user_buf, count, ppos, buf, 2);
+}
+
+static void enable_mmiotrace(void);
+static void disable_mmiotrace(void);
+
+static ssize_t write_enabled_file_bool(struct file *file,
+		const char __user *user_buf, size_t count, loff_t *ppos)
+{
+	char buf[32];
+	int buf_size = min(count, (sizeof(buf)-1));
+
+	if (copy_from_user(buf, user_buf, buf_size))
+		return -EFAULT;
+
+	switch (buf[0]) {
+	case 'y':
+	case 'Y':
+	case '1':
+		enable_mmiotrace();
+		break;
+	case 'n':
+	case 'N':
+	case '0':
+		disable_mmiotrace();
 		break;
 	}
-	spin_unlock(&trace_list_lock);
+
+	return count;
+}
+
+/* this ripped from kernel/kprobes.c */
+static struct file_operations fops_enabled = {
+	.owner =	THIS_MODULE,
+	.read =		read_enabled_file_bool,
+	.write =	write_enabled_file_bool
+};
+
+static struct file_operations fops_marker = {
+	.owner =	THIS_MODULE,
+	.write =	write_marker
+};
+
+static void enable_mmiotrace(void)
+{
+	mutex_lock(&mmiotrace_mutex);
+	if (is_enabled())
+		goto out;
+
+	chan = relay_open("cpu", dir, subbuf_size, n_subbufs,
+						&relay_callbacks, NULL);
+	if (!chan) {
+		pr_err(NAME "relay app channel creation failed.\n");
+		goto out;
+	}
+
+	reference_kmmio();
+
+	marker_file = debugfs_create_file("marker", 0660, dir, NULL,
+								&fops_marker);
+	if (!marker_file)
+		pr_err(NAME "marker file creation failed.\n");
+
+	if (nommiotrace)
+		pr_info(NAME "MMIO tracing disabled.\n");
+	if (ISA_trace)
+		pr_warning(NAME "Warning! low ISA range will be traced.\n");
+	spin_lock_irq(&trace_lock);
+	atomic_inc(&mmiotrace_enabled);
+	spin_unlock_irq(&trace_lock);
+	pr_info(NAME "enabled.\n");
+out:
+	mutex_unlock(&mmiotrace_mutex);
+}
+
+static void disable_mmiotrace(void)
+{
+	mutex_lock(&mmiotrace_mutex);
+	if (!is_enabled())
+		goto out;
+
+	spin_lock_irq(&trace_lock);
+	atomic_dec(&mmiotrace_enabled);
+	BUG_ON(is_enabled());
+	spin_unlock_irq(&trace_lock);
+
+	clear_trace_list(); /* guarantees: no more kmmio callbacks */
+	unreference_kmmio();
+	if (marker_file) {
+		debugfs_remove(marker_file);
+		marker_file = NULL;
+	}
+	if (chan) {
+		relay_close(chan);
+		chan = NULL;
+	}
+
+	pr_info(NAME "disabled.\n");
+out:
+	mutex_unlock(&mmiotrace_mutex);
 }
 
 static int __init init(void)
 {
+	pr_debug(NAME "load...\n");
 	if (n_subbufs < 2)
 		return -EINVAL;
 
 	dir = debugfs_create_dir(APP_DIR, NULL);
 	if (!dir) {
-		pr_err(MODULE_NAME ": Couldn't create relay app directory.\n");
+		pr_err(NAME "Couldn't create relay app directory.\n");
 		return -ENOMEM;
 	}
 
-	chan = create_channel(subbuf_size, n_subbufs);
-	if (!chan) {
+	enabled_file = debugfs_create_file("enabled", 0600, dir, NULL,
+								&fops_enabled);
+	if (!enabled_file) {
+		pr_err(NAME "Couldn't create enabled file.\n");
 		debugfs_remove(dir);
-		pr_err(MODULE_NAME ": relay app channel creation failed\n");
 		return -ENOMEM;
 	}
 
-	reference_kmmio();
-
-	proc_marker_file = create_proc_entry(MARKER_FILE, 0, NULL);
-	if (proc_marker_file)
-		proc_marker_file->write_proc = write_marker;
+	if (enable_now)
+		enable_mmiotrace();
 
-	pr_debug(MODULE_NAME ": loaded.\n");
-	if (nommiotrace)
-		pr_info(MODULE_NAME ": MMIO tracing disabled.\n");
-	if (ISA_trace)
-		pr_warning(MODULE_NAME ": Warning! low ISA range will be "
-								"traced.\n");
 	return 0;
 }
 
 static void __exit cleanup(void)
 {
-	pr_debug(MODULE_NAME ": unload...\n");
-	clear_trace_list();
-	unreference_kmmio();
-	remove_proc_entry(MARKER_FILE, NULL);
-	destroy_channel();
+	pr_debug(NAME "unload...\n");
+	if (enabled_file)
+		debugfs_remove(enabled_file);
+	disable_mmiotrace();
 	if (dir)
 		debugfs_remove(dir);
 }
diff --git a/arch/x86/kernel/mmiotrace/testmmiotrace.c b/arch/x86/kernel/mmiotrace/testmmiotrace.c
index 5ecff578672b..cfa60b227c8d 100644
--- a/arch/x86/kernel/mmiotrace/testmmiotrace.c
+++ b/arch/x86/kernel/mmiotrace/testmmiotrace.c
@@ -4,10 +4,6 @@
 #include <linux/module.h>
 #include <asm/io.h>
 
-extern void __iomem *ioremap_nocache_trace(unsigned long offset,
-						unsigned long size);
-extern void iounmap_trace(volatile void __iomem *addr);
-
 #define MODULE_NAME "testmmiotrace"
 
 static unsigned long mmio_address;
@@ -28,25 +24,24 @@ static void do_write_test(void __iomem *p)
 static void do_read_test(void __iomem *p)
 {
 	unsigned int i;
-	volatile unsigned int v;
 	for (i = 0; i < 256; i++)
-		v = ioread8(p + i);
+		ioread8(p + i);
 	for (i = 1024; i < (5 * 1024); i += 2)
-		v = ioread16(p + i);
+		ioread16(p + i);
 	for (i = (5 * 1024); i < (16 * 1024); i += 4)
-		v = ioread32(p + i);
+		ioread32(p + i);
 }
 
 static void do_test(void)
 {
-	void __iomem *p = ioremap_nocache_trace(mmio_address, 0x4000);
+	void __iomem *p = ioremap_nocache(mmio_address, 0x4000);
 	if (!p) {
 		pr_err(MODULE_NAME ": could not ioremap, aborting.\n");
 		return;
 	}
 	do_write_test(p);
 	do_read_test(p);
-	iounmap_trace(p);
+	iounmap(p);
 }
 
 static int __init init(void)
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 71bb3159031a..8927c878544d 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -12,6 +12,7 @@
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
+#include <linux/mmiotrace.h>
 
 #include <asm/cacheflush.h>
 #include <asm/e820.h>
@@ -126,6 +127,7 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr,
 	unsigned long new_prot_val;
 	pgprot_t prot;
 	int retval;
+	void __iomem *ret_addr;
 
 	/* Don't allow wraparound or zero size */
 	last_addr = phys_addr + size - 1;
@@ -233,7 +235,10 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr,
 		return NULL;
 	}
 
-	return (void __iomem *) (vaddr + offset);
+	ret_addr = (void __iomem *) (vaddr + offset);
+	mmiotrace_ioremap(phys_addr, size, ret_addr);
+
+	return ret_addr;
 }
 
 /**
@@ -325,6 +330,8 @@ void iounmap(volatile void __iomem *addr)
 	addr = (volatile void __iomem *)
 		(PAGE_MASK & (unsigned long __force)addr);
 
+	mmiotrace_iounmap(addr);
+
 	/* Use the vm area unlocked, assuming the caller
 	   ensures there isn't another iounmap for the same address
 	   in parallel. Reuse of the virtual address is prevented by
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index d87a6cd8b686..cb5efd0c7f51 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -16,11 +16,12 @@ typedef void (*kmmio_post_handler_t)(struct kmmio_probe *,
 				unsigned long condition, struct pt_regs *);
 
 struct kmmio_probe {
-	struct list_head list;
+	struct list_head list; /* kmmio internal list */
 	unsigned long addr; /* start location of the probe point */
 	unsigned long len; /* length of the probe region */
 	kmmio_pre_handler_t pre_handler; /* Called before addr is executed. */
 	kmmio_post_handler_t post_handler; /* Called after addr is executed */
+	void *user_data;
 };
 
 /* kmmio is active by some kmmio_probes? */
@@ -38,6 +39,21 @@ extern void unregister_kmmio_probe(struct kmmio_probe *p);
 /* Called from page fault handler. */
 extern int kmmio_handler(struct pt_regs *regs, unsigned long addr);
 
+/* Called from ioremap.c */
+#ifdef CONFIG_MMIOTRACE
+extern void
+mmiotrace_ioremap(unsigned long offset, unsigned long size, void __iomem *addr);
+extern void mmiotrace_iounmap(volatile void __iomem *addr);
+#else
+static inline void
+mmiotrace_ioremap(unsigned long offset, unsigned long size, void __iomem *addr)
+{
+}
+static inline void mmiotrace_iounmap(volatile void __iomem *addr)
+{
+}
+#endif /* CONFIG_MMIOTRACE_HOOKS */
+
 #endif /* __KERNEL__ */
 
 
-- 
cgit v1.2.3


From f984b51e0779a6dd30feedc41404013ca54e5d05 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:20:57 +0200
Subject: ftrace: add mmiotrace plugin

On Sat, 22 Mar 2008 13:07:47 +0100
Ingo Molnar <mingo@elte.hu> wrote:

> > > i'd suggest the following: pull x86.git and sched-devel.git into a
> > > single tree [the two will combine without rejects]. Then try to add a
> > > kernel/tracing/trace_mmiotrace.c ftrace plugin. The trace_sysprof.c
> > > plugin might be a good example.
> >
> > I did this and now I have mmiotrace enabled/disabled via the tracing
> > framework (what do we call this, since ftrace is one of the tracers?).
>
> cool! could you send the patches for that? (even if they are not fully
> functional yet)

Patch attached in the end. Nice to see how much code disappeared. I tried
to mark all the features I had to break with XXX-comments.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig.debug               |   3 +-
 arch/x86/kernel/mmiotrace/mmio-mod.c | 208 +++++------------------------------
 include/linux/mmiotrace.h            |   6 +
 kernel/trace/Makefile                |   1 +
 kernel/trace/trace_mmiotrace.c       |  84 ++++++++++++++
 5 files changed, 123 insertions(+), 179 deletions(-)
 create mode 100644 kernel/trace/trace_mmiotrace.c

(limited to 'include/linux')

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index aa0d6462b1fc..7e4b8494078e 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -173,7 +173,8 @@ config MMIOTRACE_HOOKS
 
 config MMIOTRACE
 	bool "Memory mapped IO tracing"
-	depends on DEBUG_KERNEL && RELAY && DEBUG_FS
+	depends on DEBUG_KERNEL && RELAY
+	select TRACING
 	select MMIOTRACE_HOOKS
 	default y
 	help
diff --git a/arch/x86/kernel/mmiotrace/mmio-mod.c b/arch/x86/kernel/mmiotrace/mmio-mod.c
index 738644061e4e..c7a67d7e482b 100644
--- a/arch/x86/kernel/mmiotrace/mmio-mod.c
+++ b/arch/x86/kernel/mmiotrace/mmio-mod.c
@@ -22,9 +22,8 @@
 #define DEBUG 1
 
 #include <linux/module.h>
-#include <linux/relay.h>
 #include <linux/debugfs.h>
-#include <linux/proc_fs.h>
+#include <linux/uaccess.h>
 #include <asm/io.h>
 #include <linux/version.h>
 #include <linux/kallsyms.h>
@@ -63,18 +62,18 @@ static const size_t subbuf_size = 256*1024;
 static DEFINE_PER_CPU(struct trap_reason, pf_reason);
 static DEFINE_PER_CPU(struct mm_io_header_rw, cpu_trace);
 
+#if 0 /* XXX: no way gather this info anymore */
 /* Access to this is not per-cpu. */
 static DEFINE_PER_CPU(atomic_t, dropped);
+#endif
 
 static struct dentry *dir;
-static struct dentry *enabled_file;
 static struct dentry *marker_file;
 
 static DEFINE_MUTEX(mmiotrace_mutex);
 static DEFINE_SPINLOCK(trace_lock);
 static atomic_t mmiotrace_enabled;
 static LIST_HEAD(trace_list);		/* struct remap_trace */
-static struct rchan *chan;
 
 /*
  * Locking in this file:
@@ -93,36 +92,24 @@ static unsigned long	filter_offset;
 static int		nommiotrace;
 static int		ISA_trace;
 static int		trace_pc;
-static int		enable_now;
 
 module_param(n_subbufs, uint, 0);
 module_param(filter_offset, ulong, 0);
 module_param(nommiotrace, bool, 0);
 module_param(ISA_trace, bool, 0);
 module_param(trace_pc, bool, 0);
-module_param(enable_now, bool, 0);
 
 MODULE_PARM_DESC(n_subbufs, "Number of 256kB buffers, default 128.");
 MODULE_PARM_DESC(filter_offset, "Start address of traced mappings.");
 MODULE_PARM_DESC(nommiotrace, "Disable actual MMIO tracing.");
 MODULE_PARM_DESC(ISA_trace, "Do not exclude the low ISA range.");
 MODULE_PARM_DESC(trace_pc, "Record address of faulting instructions.");
-MODULE_PARM_DESC(enable_now, "Start mmiotracing immediately on module load.");
 
 static bool is_enabled(void)
 {
 	return atomic_read(&mmiotrace_enabled);
 }
 
-static void record_timestamp(struct mm_io_header *header)
-{
-	struct timespec now;
-
-	getnstimeofday(&now);
-	header->sec = now.tv_sec;
-	header->nsec = now.tv_nsec;
-}
-
 /*
  * Write callback for the debugfs entry:
  * Read a marker and write it to the mmio trace log
@@ -141,7 +128,6 @@ static ssize_t write_marker(struct file *file, const char __user *buffer,
 	headp = (struct mm_io_header *)event;
 	headp->type = MMIO_MAGIC | (MMIO_MARKER << MMIO_OPCODE_SHIFT);
 	headp->data_len = len;
-	record_timestamp(headp);
 
 	if (copy_from_user(event + sizeof(*headp), buffer, len)) {
 		kfree(event);
@@ -149,9 +135,11 @@ static ssize_t write_marker(struct file *file, const char __user *buffer,
 	}
 
 	spin_lock_irq(&trace_lock);
+#if 0 /* XXX: convert this to use tracing */
 	if (is_enabled())
 		relay_write(chan, event, sizeof(*headp) + len);
 	else
+#endif
 		len = -EINVAL;
 	spin_unlock_irq(&trace_lock);
 	kfree(event);
@@ -242,7 +230,11 @@ static void pre(struct kmmio_probe *p, struct pt_regs *regs,
 	else
 		my_trace->rw.pc = 0;
 
-	record_timestamp(&my_trace->header);
+	/*
+	 * XXX: the timestamp recorded will be *after* the tracing has been
+	 * done, not at the time we hit the instruction. SMP implications
+	 * on event ordering?
+	 */
 
 	switch (type) {
 	case REG_READ:
@@ -295,77 +287,19 @@ static void post(struct kmmio_probe *p, unsigned long condition,
 	default:
 		break;
 	}
-	relay_write(chan, my_trace, sizeof(*my_trace));
+
+	/*
+	 * XXX: Several required values are ignored:
+	 * - mapping id
+	 * - program counter
+	 * Also the address should be physical, not virtual.
+	 */
+	mmio_trace_record(my_trace->header.type, my_trace->rw.address,
+							my_trace->rw.value);
 	put_cpu_var(cpu_trace);
 	put_cpu_var(pf_reason);
 }
 
-/*
- * subbuf_start() relay callback.
- *
- * Defined so that we know when events are dropped due to the buffer-full
- * condition.
- */
-static int subbuf_start_handler(struct rchan_buf *buf, void *subbuf,
-					void *prev_subbuf, size_t prev_padding)
-{
-	unsigned int cpu = buf->cpu;
-	atomic_t *drop = &per_cpu(dropped, cpu);
-	int count;
-	if (relay_buf_full(buf)) {
-		if (atomic_inc_return(drop) == 1)
-			pr_err(NAME "cpu %d buffer full!\n", cpu);
-		return 0;
-	}
-	count = atomic_read(drop);
-	if (count) {
-		pr_err(NAME "cpu %d buffer no longer full, missed %d events.\n",
-								cpu, count);
-		atomic_sub(count, drop);
-	}
-
-	return 1;
-}
-
-static struct file_operations mmio_fops = {
-	.owner = THIS_MODULE,
-};
-
-/* file_create() callback.  Creates relay file in debugfs. */
-static struct dentry *create_buf_file_handler(const char *filename,
-						struct dentry *parent,
-						int mode,
-						struct rchan_buf *buf,
-						int *is_global)
-{
-	struct dentry *buf_file;
-
-	mmio_fops.read = relay_file_operations.read;
-	mmio_fops.open = relay_file_operations.open;
-	mmio_fops.poll = relay_file_operations.poll;
-	mmio_fops.mmap = relay_file_operations.mmap;
-	mmio_fops.release = relay_file_operations.release;
-	mmio_fops.splice_read = relay_file_operations.splice_read;
-
-	buf_file = debugfs_create_file(filename, mode, parent, buf,
-								&mmio_fops);
-
-	return buf_file;
-}
-
-/* file_remove() default callback.  Removes relay file in debugfs. */
-static int remove_buf_file_handler(struct dentry *dentry)
-{
-	debugfs_remove(dentry);
-	return 0;
-}
-
-static struct rchan_callbacks relay_callbacks = {
-	.subbuf_start = subbuf_start_handler,
-	.create_buf_file = create_buf_file_handler,
-	.remove_buf_file = remove_buf_file_handler,
-};
-
 static void ioremap_trace_core(unsigned long offset, unsigned long size,
 							void __iomem *addr)
 {
@@ -387,7 +321,6 @@ static void ioremap_trace_core(unsigned long offset, unsigned long size,
 			.pc   = 0
 		}
 	};
-	record_timestamp(&event.header);
 
 	if (!trace) {
 		pr_err(NAME "kmalloc failed in ioremap\n");
@@ -410,7 +343,10 @@ static void ioremap_trace_core(unsigned long offset, unsigned long size,
 	if (!is_enabled())
 		goto not_enabled;
 
-	relay_write(chan, &event, sizeof(event));
+	/*
+	 * XXX: Insufficient data recorded!
+	 */
+	mmio_trace_record(event.header.type, event.map.addr, event.map.len);
 	list_add_tail(&trace->list, &trace_list);
 	if (!nommiotrace)
 		register_kmmio_probe(&trace->probe);
@@ -454,7 +390,6 @@ static void iounmap_trace_core(volatile void __iomem *addr)
 	struct remap_trace *found_trace = NULL;
 
 	pr_debug(NAME "Unmapping %p.\n", addr);
-	record_timestamp(&event.header);
 
 	spin_lock_irq(&trace_lock);
 	if (!is_enabled())
@@ -469,7 +404,8 @@ static void iounmap_trace_core(volatile void __iomem *addr)
 			break;
 		}
 	}
-	relay_write(chan, &event, sizeof(event));
+	mmio_trace_record(event.header.type, event.map.addr,
+					found_trace ? found_trace->id : -1);
 
 not_enabled:
 	spin_unlock_irq(&trace_lock);
@@ -512,77 +448,23 @@ static void clear_trace_list(void)
 	}
 }
 
-static ssize_t read_enabled_file_bool(struct file *file,
-		char __user *user_buf, size_t count, loff_t *ppos)
-{
-	char buf[3];
-
-	if (is_enabled())
-		buf[0] = '1';
-	else
-		buf[0] = '0';
-	buf[1] = '\n';
-	buf[2] = '\0';
-	return simple_read_from_buffer(user_buf, count, ppos, buf, 2);
-}
-
-static void enable_mmiotrace(void);
-static void disable_mmiotrace(void);
-
-static ssize_t write_enabled_file_bool(struct file *file,
-		const char __user *user_buf, size_t count, loff_t *ppos)
-{
-	char buf[32];
-	int buf_size = min(count, (sizeof(buf)-1));
-
-	if (copy_from_user(buf, user_buf, buf_size))
-		return -EFAULT;
-
-	switch (buf[0]) {
-	case 'y':
-	case 'Y':
-	case '1':
-		enable_mmiotrace();
-		break;
-	case 'n':
-	case 'N':
-	case '0':
-		disable_mmiotrace();
-		break;
-	}
-
-	return count;
-}
-
-/* this ripped from kernel/kprobes.c */
-static struct file_operations fops_enabled = {
-	.owner =	THIS_MODULE,
-	.read =		read_enabled_file_bool,
-	.write =	write_enabled_file_bool
-};
-
 static struct file_operations fops_marker = {
 	.owner =	THIS_MODULE,
 	.write =	write_marker
 };
 
-static void enable_mmiotrace(void)
+void enable_mmiotrace(void)
 {
 	mutex_lock(&mmiotrace_mutex);
 	if (is_enabled())
 		goto out;
 
-	chan = relay_open("cpu", dir, subbuf_size, n_subbufs,
-						&relay_callbacks, NULL);
-	if (!chan) {
-		pr_err(NAME "relay app channel creation failed.\n");
-		goto out;
-	}
-
 	reference_kmmio();
 
+#if 0 /* XXX: tracing does not support text entries */
 	marker_file = debugfs_create_file("marker", 0660, dir, NULL,
 								&fops_marker);
+#endif
 	if (!marker_file)
 		pr_err(NAME "marker file creation failed.\n");
 
@@ -598,7 +480,7 @@ out:
 	mutex_unlock(&mmiotrace_mutex);
 }
 
-static void disable_mmiotrace(void)
+void disable_mmiotrace(void)
 {
 	mutex_lock(&mmiotrace_mutex);
 	if (!is_enabled())
@@ -615,17 +497,13 @@ static void disable_mmiotrace(void)
 		debugfs_remove(marker_file);
 		marker_file = NULL;
 	}
-	if (chan) {
-		relay_close(chan);
-		chan = NULL;
-	}
 
 	pr_info(NAME "disabled.\n");
 out:
 	mutex_unlock(&mmiotrace_mutex);
 }
 
-static int __init init(void)
+int __init init_mmiotrace(void)
 {
 	pr_debug(NAME "load...\n");
 	if (n_subbufs < 2)
@@ -636,31 +514,5 @@ static int __init init(void)
 		pr_err(NAME "Couldn't create relay app directory.\n");
 		return -ENOMEM;
 	}
-
-	enabled_file = debugfs_create_file("enabled", 0600, dir, NULL,
-								&fops_enabled);
-	if (!enabled_file) {
-		pr_err(NAME "Couldn't create enabled file.\n");
-		debugfs_remove(dir);
-		return -ENOMEM;
-	}
-
-	if (enable_now)
-		enable_mmiotrace();
-
 	return 0;
 }
-
-static void __exit cleanup(void)
-{
-	pr_debug(NAME "unload...\n");
-	if (enabled_file)
-		debugfs_remove(enabled_file);
-	disable_mmiotrace();
-	if (dir)
-		debugfs_remove(dir);
-}
-
-module_init(init);
-module_exit(cleanup);
-MODULE_LICENSE("GPL");
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index cb5efd0c7f51..579b3b06c90e 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -54,6 +54,12 @@ static inline void mmiotrace_iounmap(volatile void __iomem *addr)
 }
 #endif /* CONFIG_MMIOTRACE_HOOKS */
 
+/* in kernel/trace/trace_mmiotrace.c */
+extern int __init init_mmiotrace(void);
+extern void enable_mmiotrace(void);
+extern void disable_mmiotrace(void);
+extern void mmio_trace_record(u32 type, unsigned long addr, unsigned long arg);
+
 #endif /* __KERNEL__ */
 
 
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index d9efbbfa2bdf..c44a7dce9086 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -18,5 +18,6 @@ obj-$(CONFIG_FTRACE) += trace_functions.o
 obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o
+obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o
 
 libftrace-y := ftrace.o
diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c
new file mode 100644
index 000000000000..e4dd03cc5aa6
--- /dev/null
+++ b/kernel/trace/trace_mmiotrace.c
@@ -0,0 +1,84 @@
+/*
+ * Memory mapped I/O tracing
+ *
+ * Copyright (C) 2008 Pekka Paalanen <pq@iki.fi>
+ */
+
+#define DEBUG 1
+
+#include <linux/kernel.h>
+#include <linux/mmiotrace.h>
+
+#include "trace.h"
+
+extern void
+__trace_special(void *__tr, void *__data,
+		unsigned long arg1, unsigned long arg2, unsigned long arg3);
+
+static struct trace_array *mmio_trace_array;
+
+
+static void mmio_trace_init(struct trace_array *tr)
+{
+	pr_debug("in %s\n", __func__);
+	mmio_trace_array = tr;
+	if (tr->ctrl)
+		enable_mmiotrace();
+}
+
+static void mmio_trace_reset(struct trace_array *tr)
+{
+	pr_debug("in %s\n", __func__);
+	if (tr->ctrl)
+		disable_mmiotrace();
+}
+
+static void mmio_trace_ctrl_update(struct trace_array *tr)
+{
+	pr_debug("in %s\n", __func__);
+	if (tr->ctrl)
+		enable_mmiotrace();
+	else
+		disable_mmiotrace();
+}
+
+static struct tracer mmio_tracer __read_mostly =
+{
+	.name		= "mmiotrace",
+	.init		= mmio_trace_init,
+	.reset		= mmio_trace_reset,
+	.ctrl_update	= mmio_trace_ctrl_update,
+};
+
+__init static int init_mmio_trace(void)
+{
+	int ret = init_mmiotrace();
+	if (ret)
+		return ret;
+	return register_tracer(&mmio_tracer);
+}
+device_initcall(init_mmio_trace);
+
+void mmio_trace_record(u32 type, unsigned long addr, unsigned long arg)
+{
+	struct trace_array *tr = mmio_trace_array;
+	struct trace_array_cpu *data = tr->data[smp_processor_id()];
+
+	if (!current || current->pid == 0) {
+		/*
+		 * XXX: This is a problem. We need to able to record, no
+		 * matter what. tracing_generic_entry_update() would crash.
+		 */
+		static unsigned limit;
+		if (limit++ < 12)
+			pr_err("Error in %s: no current.\n", __func__);
+		return;
+	}
+	if (!tr || !data) {
+		static unsigned limit;
+		if (limit++ < 12)
+			pr_err("%s: no tr or data\n", __func__);
+		return;
+	}
+	__trace_special(tr, data, type, addr, arg);
+}
-- 
cgit v1.2.3


From bd8ac686c73c7e925fcfe0b02dc4e7b947127864 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:20:57 +0200
Subject: ftrace: mmiotrace, updates

here is a patch that makes mmiotrace work almost well within the tracing
framework. The patch applies on top of my previous patch. I have my own
output formatting in place now.

Summary of changes:
- fix the NULL dereference that was due to not calling tracing_reset()
- add print_line() callback into struct tracer
- implement print_line() for mmiotrace, producing up-to-spec text
- add my output header, but that is not really called in the right place
- rewrote the main structs in mmiotrace
- added two new trace entry types: TRACE_MMIO_RW and TRACE_MMIO_MAP
- made some functions in trace.c non-static
- check current==NULL in tracing_generic_entry_update()
- fix(?) comparison in trace_seq_printf()

Things seem to work fine except a few issues. Markers (text lines injected
into mmiotrace log) are missing, I did not feel hacking them in before we
have variable length entries. My output header is printed only for 'trace'
file, but not 'trace_pipe'. For some reason, despite my quick fix,
iter->trace is NULL in print_trace_line() when called from 'trace_pipe'
file, which means I don't get proper output formatting.

I only tried by loading nouveau.ko, which just detects the card, and that
is traced fine. I didn't try further. Map, two reads and unmap. Works
perfectly.

I am missing the information about overflows, I'd prefer to have a
counter for lost events. I didn't try, but I guess currently there is no
way of knowning when it overflows?

So, not too far from being fully operational, it seems :-)
And looking at the diffstat, there also is some 700-900 lines of user space
code that just became obsolete.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/Kconfig.debug               |   2 +-
 arch/x86/kernel/mmiotrace/mmio-mod.c | 140 ++++++++++----------------------
 include/linux/mmiotrace.h            |  85 ++++++--------------
 kernel/trace/trace.c                 |  34 ++++++++
 kernel/trace/trace.h                 |  14 ++++
 kernel/trace/trace_mmiotrace.c       | 151 ++++++++++++++++++++++++++++-------
 6 files changed, 238 insertions(+), 188 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 7e4b8494078e..1d6de0d67f99 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -173,7 +173,7 @@ config MMIOTRACE_HOOKS
 
 config MMIOTRACE
 	bool "Memory mapped IO tracing"
-	depends on DEBUG_KERNEL && RELAY
+	depends on DEBUG_KERNEL
 	select TRACING
 	select MMIOTRACE_HOOKS
 	default y
diff --git a/arch/x86/kernel/mmiotrace/mmio-mod.c b/arch/x86/kernel/mmiotrace/mmio-mod.c
index c7a67d7e482b..62abc281a512 100644
--- a/arch/x86/kernel/mmiotrace/mmio-mod.c
+++ b/arch/x86/kernel/mmiotrace/mmio-mod.c
@@ -37,11 +37,6 @@
 
 #define NAME "mmiotrace: "
 
-/* This app's relay channel files will appear in /debug/mmio-trace */
-static const char APP_DIR[] = "mmio-trace";
-/* the marker injection file in /debug/APP_DIR */
-static const char MARKER_FILE[] = "mmio-marker";
-
 struct trap_reason {
 	unsigned long addr;
 	unsigned long ip;
@@ -56,18 +51,15 @@ struct remap_trace {
 	unsigned long id;
 };
 
-static const size_t subbuf_size = 256*1024;
-
 /* Accessed per-cpu. */
 static DEFINE_PER_CPU(struct trap_reason, pf_reason);
-static DEFINE_PER_CPU(struct mm_io_header_rw, cpu_trace);
+static DEFINE_PER_CPU(struct mmiotrace_rw, cpu_trace);
 
 #if 0 /* XXX: no way gather this info anymore */
 /* Access to this is not per-cpu. */
 static DEFINE_PER_CPU(atomic_t, dropped);
 #endif
 
-static struct dentry *dir;
 static struct dentry *marker_file;
 
 static DEFINE_MUTEX(mmiotrace_mutex);
@@ -82,24 +74,21 @@ static LIST_HEAD(trace_list);		/* struct remap_trace */
  *   and trace_lock.
  * - Routines depending on is_enabled() must take trace_lock.
  * - trace_list users must hold trace_lock.
- * - is_enabled() guarantees that chan is valid.
+ * - is_enabled() guarantees that mmio_trace_record is allowed.
  * - pre/post callbacks assume the effect of is_enabled() being true.
  */
 
 /* module parameters */
-static unsigned int	n_subbufs = 32*4;
 static unsigned long	filter_offset;
 static int		nommiotrace;
 static int		ISA_trace;
 static int		trace_pc;
 
-module_param(n_subbufs, uint, 0);
 module_param(filter_offset, ulong, 0);
 module_param(nommiotrace, bool, 0);
 module_param(ISA_trace, bool, 0);
 module_param(trace_pc, bool, 0);
 
-MODULE_PARM_DESC(n_subbufs, "Number of 256kB buffers, default 128.");
 MODULE_PARM_DESC(filter_offset, "Start address of traced mappings.");
 MODULE_PARM_DESC(nommiotrace, "Disable actual MMIO tracing.");
 MODULE_PARM_DESC(ISA_trace, "Do not exclude the low ISA range.");
@@ -110,6 +99,7 @@ static bool is_enabled(void)
 	return atomic_read(&mmiotrace_enabled);
 }
 
+#if 0 /* XXX: needs rewrite */
 /*
  * Write callback for the debugfs entry:
  * Read a marker and write it to the mmio trace log
@@ -145,6 +135,7 @@ static ssize_t write_marker(struct file *file, const char __user *buffer,
 	kfree(event);
 	return len;
 }
+#endif
 
 static void print_pte(unsigned long address)
 {
@@ -198,9 +189,10 @@ static void pre(struct kmmio_probe *p, struct pt_regs *regs,
 						unsigned long addr)
 {
 	struct trap_reason *my_reason = &get_cpu_var(pf_reason);
-	struct mm_io_header_rw *my_trace = &get_cpu_var(cpu_trace);
+	struct mmiotrace_rw *my_trace = &get_cpu_var(cpu_trace);
 	const unsigned long instptr = instruction_pointer(regs);
 	const enum reason_type type = get_ins_type(instptr);
+	struct remap_trace *trace = p->user_data;
 
 	/* it doesn't make sense to have more than one active trace per cpu */
 	if (my_reason->active_traces)
@@ -212,23 +204,17 @@ static void pre(struct kmmio_probe *p, struct pt_regs *regs,
 	my_reason->addr = addr;
 	my_reason->ip = instptr;
 
-	my_trace->header.type = MMIO_MAGIC;
-	my_trace->header.pid = 0;
-	my_trace->header.data_len = sizeof(struct mm_io_rw);
-	my_trace->rw.address = addr;
-	/*
-	 * struct remap_trace *trace = p->user_data;
-	 * phys = addr - trace->probe.addr + trace->phys;
-	 */
+	my_trace->phys = addr - trace->probe.addr + trace->phys;
+	my_trace->map_id = trace->id;
 
 	/*
 	 * Only record the program counter when requested.
 	 * It may taint clean-room reverse engineering.
 	 */
 	if (trace_pc)
-		my_trace->rw.pc = instptr;
+		my_trace->pc = instptr;
 	else
-		my_trace->rw.pc = 0;
+		my_trace->pc = 0;
 
 	/*
 	 * XXX: the timestamp recorded will be *after* the tracing has been
@@ -238,28 +224,25 @@ static void pre(struct kmmio_probe *p, struct pt_regs *regs,
 
 	switch (type) {
 	case REG_READ:
-		my_trace->header.type |=
-			(MMIO_READ << MMIO_OPCODE_SHIFT) |
-			(get_ins_mem_width(instptr) << MMIO_WIDTH_SHIFT);
+		my_trace->opcode = MMIO_READ;
+		my_trace->width = get_ins_mem_width(instptr);
 		break;
 	case REG_WRITE:
-		my_trace->header.type |=
-			(MMIO_WRITE << MMIO_OPCODE_SHIFT) |
-			(get_ins_mem_width(instptr) << MMIO_WIDTH_SHIFT);
-		my_trace->rw.value = get_ins_reg_val(instptr, regs);
+		my_trace->opcode = MMIO_WRITE;
+		my_trace->width = get_ins_mem_width(instptr);
+		my_trace->value = get_ins_reg_val(instptr, regs);
 		break;
 	case IMM_WRITE:
-		my_trace->header.type |=
-			(MMIO_WRITE << MMIO_OPCODE_SHIFT) |
-			(get_ins_mem_width(instptr) << MMIO_WIDTH_SHIFT);
-		my_trace->rw.value = get_ins_imm_val(instptr);
+		my_trace->opcode = MMIO_WRITE;
+		my_trace->width = get_ins_mem_width(instptr);
+		my_trace->value = get_ins_imm_val(instptr);
 		break;
 	default:
 		{
 			unsigned char *ip = (unsigned char *)instptr;
-			my_trace->header.type |=
-					(MMIO_UNKNOWN_OP << MMIO_OPCODE_SHIFT);
-			my_trace->rw.value = (*ip) << 16 | *(ip + 1) << 8 |
+			my_trace->opcode = MMIO_UNKNOWN_OP;
+			my_trace->width = 0;
+			my_trace->value = (*ip) << 16 | *(ip + 1) << 8 |
 								*(ip + 2);
 		}
 	}
@@ -271,7 +254,7 @@ static void post(struct kmmio_probe *p, unsigned long condition,
 							struct pt_regs *regs)
 {
 	struct trap_reason *my_reason = &get_cpu_var(pf_reason);
-	struct mm_io_header_rw *my_trace = &get_cpu_var(cpu_trace);
+	struct mmiotrace_rw *my_trace = &get_cpu_var(cpu_trace);
 
 	/* this should always return the active_trace count to 0 */
 	my_reason->active_traces--;
@@ -282,20 +265,13 @@ static void post(struct kmmio_probe *p, unsigned long condition,
 
 	switch (my_reason->type) {
 	case REG_READ:
-		my_trace->rw.value = get_ins_reg_val(my_reason->ip, regs);
+		my_trace->value = get_ins_reg_val(my_reason->ip, regs);
 		break;
 	default:
 		break;
 	}
 
-	/*
-	 * XXX: Several required values are ignored:
-	 * - mapping id
-	 * - program counter
-	 * Also the address should be physical, not virtual.
-	 */
-	mmio_trace_record(my_trace->header.type, my_trace->rw.address,
-							my_trace->rw.value);
+	mmio_trace_rw(my_trace);
 	put_cpu_var(cpu_trace);
 	put_cpu_var(pf_reason);
 }
@@ -305,21 +281,11 @@ static void ioremap_trace_core(unsigned long offset, unsigned long size,
 {
 	static atomic_t next_id;
 	struct remap_trace *trace = kmalloc(sizeof(*trace), GFP_KERNEL);
-	struct mm_io_header_map event = {
-		.header = {
-			.type = MMIO_MAGIC |
-					(MMIO_PROBE << MMIO_OPCODE_SHIFT),
-			.sec = 0,
-			.nsec = 0,
-			.pid = 0,
-			.data_len = sizeof(struct mm_io_map)
-		},
-		.map = {
-			.phys = offset,
-			.addr = (unsigned long)addr,
-			.len  = size,
-			.pc   = 0
-		}
+	struct mmiotrace_map map = {
+		.phys = offset,
+		.virt = (unsigned long)addr,
+		.len = size,
+		.opcode = MMIO_PROBE
 	};
 
 	if (!trace) {
@@ -338,15 +304,13 @@ static void ioremap_trace_core(unsigned long offset, unsigned long size,
 		.phys = offset,
 		.id = atomic_inc_return(&next_id)
 	};
+	map.map_id = trace->id;
 
 	spin_lock_irq(&trace_lock);
 	if (!is_enabled())
 		goto not_enabled;
 
-	/*
-	 * XXX: Insufficient data recorded!
-	 */
-	mmio_trace_record(event.header.type, event.map.addr, event.map.len);
+	mmio_trace_mapping(&map);
 	list_add_tail(&trace->list, &trace_list);
 	if (!nommiotrace)
 		register_kmmio_probe(&trace->probe);
@@ -369,21 +333,11 @@ mmiotrace_ioremap(unsigned long offset, unsigned long size, void __iomem *addr)
 
 static void iounmap_trace_core(volatile void __iomem *addr)
 {
-	struct mm_io_header_map event = {
-		.header = {
-			.type = MMIO_MAGIC |
-				(MMIO_UNPROBE << MMIO_OPCODE_SHIFT),
-			.sec = 0,
-			.nsec = 0,
-			.pid = 0,
-			.data_len = sizeof(struct mm_io_map)
-		},
-		.map = {
-			.phys = 0,
-			.addr = (unsigned long)addr,
-			.len  = 0,
-			.pc   = 0
-		}
+	struct mmiotrace_map map = {
+		.phys = 0,
+		.virt = (unsigned long)addr,
+		.len = 0,
+		.opcode = MMIO_UNPROBE
 	};
 	struct remap_trace *trace;
 	struct remap_trace *tmp;
@@ -404,8 +358,8 @@ static void iounmap_trace_core(volatile void __iomem *addr)
 			break;
 		}
 	}
-	mmio_trace_record(event.header.type, event.map.addr,
-					found_trace ? found_trace->id : -1);
+	map.map_id = (found_trace) ? found_trace->id : -1;
+	mmio_trace_mapping(&map);
 
 not_enabled:
 	spin_unlock_irq(&trace_lock);
@@ -448,10 +402,12 @@ static void clear_trace_list(void)
 	}
 }
 
+#if 0 /* XXX: out of order */
 static struct file_operations fops_marker = {
 	.owner =	THIS_MODULE,
 	.write =	write_marker
 };
+#endif
 
 void enable_mmiotrace(void)
 {
@@ -464,9 +420,9 @@ void enable_mmiotrace(void)
 #if 0 /* XXX: tracing does not support text entries */
 	marker_file = debugfs_create_file("marker", 0660, dir, NULL,
 								&fops_marker);
-#endif
 	if (!marker_file)
 		pr_err(NAME "marker file creation failed.\n");
+#endif
 
 	if (nommiotrace)
 		pr_info(NAME "MMIO tracing disabled.\n");
@@ -502,17 +458,3 @@ void disable_mmiotrace(void)
 out:
 	mutex_unlock(&mmiotrace_mutex);
 }
-
-int __init init_mmiotrace(void)
-{
-	pr_debug(NAME "load...\n");
-	if (n_subbufs < 2)
-		return -EINVAL;
-
-	dir = debugfs_create_dir(APP_DIR, NULL);
-	if (!dir) {
-		pr_err(NAME "Couldn't create relay app directory.\n");
-		return -ENOMEM;
-	}
-	return 0;
-}
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index 579b3b06c90e..c88a9c197d22 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -54,73 +54,38 @@ static inline void mmiotrace_iounmap(volatile void __iomem *addr)
 }
 #endif /* CONFIG_MMIOTRACE_HOOKS */
 
-/* in kernel/trace/trace_mmiotrace.c */
-extern int __init init_mmiotrace(void);
-extern void enable_mmiotrace(void);
-extern void disable_mmiotrace(void);
-extern void mmio_trace_record(u32 type, unsigned long addr, unsigned long arg);
-
-#endif /* __KERNEL__ */
-
-
-/*
- * If you change anything here, you must bump MMIO_VERSION.
- * This is the relay data format for user space.
- */
-#define MMIO_VERSION 0x04
-
-/* mm_io_header.type */
-#define MMIO_OPCODE_MASK 0xff
-#define MMIO_OPCODE_SHIFT 0
-#define MMIO_WIDTH_MASK 0xff00
-#define MMIO_WIDTH_SHIFT 8
-#define MMIO_MAGIC (0x6f000000 | (MMIO_VERSION<<16))
-#define MMIO_MAGIC_MASK 0xffff0000
-
-enum mm_io_opcode {          /* payload type: */
-	MMIO_READ = 0x1,     /* struct mm_io_rw */
-	MMIO_WRITE = 0x2,    /* struct mm_io_rw */
-	MMIO_PROBE = 0x3,    /* struct mm_io_map */
-	MMIO_UNPROBE = 0x4,  /* struct mm_io_map */
+enum mm_io_opcode {
+	MMIO_READ = 0x1,     /* struct mmiotrace_rw */
+	MMIO_WRITE = 0x2,    /* struct mmiotrace_rw */
+	MMIO_PROBE = 0x3,    /* struct mmiotrace_map */
+	MMIO_UNPROBE = 0x4,  /* struct mmiotrace_map */
 	MMIO_MARKER = 0x5,   /* raw char data */
-	MMIO_UNKNOWN_OP = 0x6, /* struct mm_io_rw */
+	MMIO_UNKNOWN_OP = 0x6, /* struct mmiotrace_rw */
 };
 
-struct mm_io_header {
-	__u32 type;     /* see MMIO_* macros above */
-	__u32 sec;      /* timestamp */
-	__u32 nsec;
-	__u32 pid;      /* PID of the process, or 0 for kernel core */
-	__u16 data_len; /* length of the following payload */
+struct mmiotrace_rw {
+	unsigned long phys;	/* PCI address of register */
+	unsigned long value;
+	unsigned long pc;	/* optional program counter */
+	int map_id;
+	unsigned char opcode;	/* one of MMIO_{READ,WRITE,UNKNOWN_OP} */
+	unsigned char width;	/* size of register access in bytes */
 };
 
-struct mm_io_rw {
-	__u64 address; /* virtual address of register */
-	__u64 value;
-	__u64 pc;      /* optional program counter */
+struct mmiotrace_map {
+	unsigned long phys;	/* base address in PCI space */
+	unsigned long virt;	/* base virtual address */
+	unsigned long len;	/* mapping size */
+	int map_id;
+	unsigned char opcode;	/* MMIO_PROBE or MMIO_UNPROBE */
 };
 
-struct mm_io_map {
-	__u64 phys;  /* base address in PCI space */
-	__u64 addr;  /* base virtual address */
-	__u64 len;   /* mapping size */
-	__u64 pc;    /* optional program counter */
-};
-
-
-/*
- * These structures are used to allow a single relay_write()
- * call to write a full packet.
- */
-
-struct mm_io_header_rw {
-	struct mm_io_header header;
-	struct mm_io_rw rw;
-} __attribute__((packed));
+/* in kernel/trace/trace_mmiotrace.c */
+extern void enable_mmiotrace(void);
+extern void disable_mmiotrace(void);
+extern void mmio_trace_rw(struct mmiotrace_rw *rw);
+extern void mmio_trace_mapping(struct mmiotrace_map *map);
 
-struct mm_io_header_map {
-	struct mm_io_header header;
-	struct mm_io_map map;
-} __attribute__((packed));
+#endif /* __KERNEL__ */
 
 #endif /* MMIOTRACE_H */
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 3271916ff033..d14fe49e9638 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -831,6 +831,40 @@ ftrace(struct trace_array *tr, struct trace_array_cpu *data,
 		trace_function(tr, data, ip, parent_ip, flags);
 }
 
+#ifdef CONFIG_MMIOTRACE
+void __trace_mmiotrace_rw(struct trace_array *tr, struct trace_array_cpu *data,
+						struct mmiotrace_rw *rw)
+{
+	struct trace_entry *entry;
+	unsigned long irq_flags;
+
+	spin_lock_irqsave(&data->lock, irq_flags);
+	entry			= tracing_get_trace_entry(tr, data);
+	tracing_generic_entry_update(entry, 0);
+	entry->type		= TRACE_MMIO_RW;
+	entry->mmiorw		= *rw;
+	spin_unlock_irqrestore(&data->lock, irq_flags);
+
+	trace_wake_up();
+}
+
+void __trace_mmiotrace_map(struct trace_array *tr, struct trace_array_cpu *data,
+						struct mmiotrace_map *map)
+{
+	struct trace_entry *entry;
+	unsigned long irq_flags;
+
+	spin_lock_irqsave(&data->lock, irq_flags);
+	entry			= tracing_get_trace_entry(tr, data);
+	tracing_generic_entry_update(entry, 0);
+	entry->type		= TRACE_MMIO_MAP;
+	entry->mmiomap		= *map;
+	spin_unlock_irqrestore(&data->lock, irq_flags);
+
+	trace_wake_up();
+}
+#endif
+
 void __trace_stack(struct trace_array *tr,
 		   struct trace_array_cpu *data,
 		   unsigned long flags,
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index c460e85e94ed..0ef9ef74c806 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -5,6 +5,7 @@
 #include <asm/atomic.h>
 #include <linux/sched.h>
 #include <linux/clocksource.h>
+#include <linux/mmiotrace.h>
 
 enum trace_type {
 	__TRACE_FIRST_TYPE = 0,
@@ -14,6 +15,8 @@ enum trace_type {
 	TRACE_WAKE,
 	TRACE_STACK,
 	TRACE_SPECIAL,
+	TRACE_MMIO_RW,
+	TRACE_MMIO_MAP,
 
 	__TRACE_LAST_TYPE
 };
@@ -75,6 +78,8 @@ struct trace_entry {
 		struct ctx_switch_entry		ctx;
 		struct special_entry		special;
 		struct stack_entry		stack;
+		struct mmiotrace_rw		mmiorw;
+		struct mmiotrace_map		mmiomap;
 	};
 };
 
@@ -255,6 +260,15 @@ extern unsigned long ftrace_update_tot_cnt;
 extern int DYN_FTRACE_TEST_NAME(void);
 #endif
 
+#ifdef CONFIG_MMIOTRACE
+extern void __trace_mmiotrace_rw(struct trace_array *tr,
+				struct trace_array_cpu *data,
+				struct mmiotrace_rw *rw);
+extern void __trace_mmiotrace_map(struct trace_array *tr,
+				struct trace_array_cpu *data,
+				struct mmiotrace_map *map);
+#endif
+
 #ifdef CONFIG_FTRACE_STARTUP_TEST
 #ifdef CONFIG_FTRACE
 extern int trace_selftest_startup_function(struct tracer *trace,
diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c
index e4dd03cc5aa6..3a12b1ad0c63 100644
--- a/kernel/trace/trace_mmiotrace.c
+++ b/kernel/trace/trace_mmiotrace.c
@@ -11,19 +11,26 @@
 
 #include "trace.h"
 
-extern void
-__trace_special(void *__tr, void *__data,
-		unsigned long arg1, unsigned long arg2, unsigned long arg3);
-
 static struct trace_array *mmio_trace_array;
 
+static void mmio_reset_data(struct trace_array *tr)
+{
+	int cpu;
+
+	tr->time_start = ftrace_now(tr->cpu);
+
+	for_each_online_cpu(cpu)
+		tracing_reset(tr->data[cpu]);
+}
 
 static void mmio_trace_init(struct trace_array *tr)
 {
 	pr_debug("in %s\n", __func__);
 	mmio_trace_array = tr;
-	if (tr->ctrl)
+	if (tr->ctrl) {
+		mmio_reset_data(tr);
 		enable_mmiotrace();
+	}
 }
 
 static void mmio_trace_reset(struct trace_array *tr)
@@ -31,15 +38,110 @@ static void mmio_trace_reset(struct trace_array *tr)
 	pr_debug("in %s\n", __func__);
 	if (tr->ctrl)
 		disable_mmiotrace();
+	mmio_reset_data(tr);
+	mmio_trace_array = NULL;
 }
 
 static void mmio_trace_ctrl_update(struct trace_array *tr)
 {
 	pr_debug("in %s\n", __func__);
-	if (tr->ctrl)
+	if (tr->ctrl) {
+		mmio_reset_data(tr);
 		enable_mmiotrace();
-	else
+	} else {
 		disable_mmiotrace();
+	}
+}
+
+/* XXX: This is not called for trace_pipe file! */
+void mmio_print_header(struct trace_iterator *iter)
+{
+	struct trace_seq *s = &iter->seq;
+	trace_seq_printf(s, "VERSION broken 20070824\n");
+	/* TODO: print /proc/bus/pci/devices contents as PCIDEV lines */
+}
+
+static int mmio_print_rw(struct trace_iterator *iter)
+{
+	struct trace_entry *entry = iter->ent;
+	struct mmiotrace_rw *rw	= &entry->mmiorw;
+	struct trace_seq *s	= &iter->seq;
+	unsigned long long t	= ns2usecs(entry->t);
+	unsigned long usec_rem	= do_div(t, 1000000ULL);
+	unsigned secs		= (unsigned long)t;
+	int ret = 1;
+
+	switch (entry->mmiorw.opcode) {
+	case MMIO_READ:
+		ret = trace_seq_printf(s,
+			"R %d %lu.%06lu %d 0x%lx 0x%lx 0x%lx %d\n",
+			rw->width, secs, usec_rem, rw->map_id, rw->phys,
+			rw->value, rw->pc, entry->pid);
+		break;
+	case MMIO_WRITE:
+		ret = trace_seq_printf(s,
+			"W %d %lu.%06lu %d 0x%lx 0x%lx 0x%lx %d\n",
+			rw->width, secs, usec_rem, rw->map_id, rw->phys,
+			rw->value, rw->pc, entry->pid);
+		break;
+	case MMIO_UNKNOWN_OP:
+		ret = trace_seq_printf(s,
+			"UNKNOWN %lu.%06lu %d 0x%lx %02x,%02x,%02x 0x%lx %d\n",
+			secs, usec_rem, rw->map_id, rw->phys,
+			(rw->value >> 16) & 0xff, (rw->value >> 8) & 0xff,
+			(rw->value >> 0) & 0xff, rw->pc, entry->pid);
+		break;
+	default:
+		ret = trace_seq_printf(s, "rw what?\n");
+		break;
+	}
+	if (ret)
+		return 1;
+	return 0;
+}
+
+static int mmio_print_map(struct trace_iterator *iter)
+{
+	struct trace_entry *entry = iter->ent;
+	struct mmiotrace_map *m	= &entry->mmiomap;
+	struct trace_seq *s	= &iter->seq;
+	unsigned long long t	= ns2usecs(entry->t);
+	unsigned long usec_rem	= do_div(t, 1000000ULL);
+	unsigned secs		= (unsigned long)t;
+	int ret = 1;
+
+	switch (entry->mmiorw.opcode) {
+	case MMIO_PROBE:
+		ret = trace_seq_printf(s,
+			"MAP %lu.%06lu %d 0x%lx 0x%lx 0x%lx 0x%lx %d\n",
+			secs, usec_rem, m->map_id, m->phys, m->virt, m->len,
+			0UL, entry->pid);
+		break;
+	case MMIO_UNPROBE:
+		ret = trace_seq_printf(s,
+			"UNMAP %lu.%06lu %d 0x%lx %d\n",
+			secs, usec_rem, m->map_id, 0UL, entry->pid);
+		break;
+	default:
+		ret = trace_seq_printf(s, "map what?\n");
+		break;
+	}
+	if (ret)
+		return 1;
+	return 0;
+}
+
+/* return 0 to abort printing without consuming current entry in pipe mode */
+static int mmio_print_line(struct trace_iterator *iter)
+{
+	switch (iter->ent->type) {
+	case TRACE_MMIO_RW:
+		return mmio_print_rw(iter);
+	case TRACE_MMIO_MAP:
+		return mmio_print_map(iter);
+	default:
+		return 1; /* ignore unknown entries */
+	}
 }
 
 static struct tracer mmio_tracer __read_mostly =
@@ -47,38 +149,31 @@ static struct tracer mmio_tracer __read_mostly =
 	.name		= "mmiotrace",
 	.init		= mmio_trace_init,
 	.reset		= mmio_trace_reset,
+	.open		= mmio_print_header,
 	.ctrl_update	= mmio_trace_ctrl_update,
+	.print_line	= mmio_print_line,
 };
 
 __init static int init_mmio_trace(void)
 {
-	int ret = init_mmiotrace();
-	if (ret)
-		return ret;
 	return register_tracer(&mmio_tracer);
 }
 device_initcall(init_mmio_trace);
 
-void mmio_trace_record(u32 type, unsigned long addr, unsigned long arg)
+void mmio_trace_rw(struct mmiotrace_rw *rw)
 {
 	struct trace_array *tr = mmio_trace_array;
 	struct trace_array_cpu *data = tr->data[smp_processor_id()];
+	__trace_mmiotrace_rw(tr, data, rw);
+}
 
-	if (!current || current->pid == 0) {
-		/*
-		 * XXX: This is a problem. We need to able to record, no
-		 * matter what. tracing_generic_entry_update() would crash.
-		 */
-		static unsigned limit;
-		if (limit++ < 12)
-			pr_err("Error in %s: no current.\n", __func__);
-		return;
-	}
-	if (!tr || !data) {
-		static unsigned limit;
-		if (limit++ < 12)
-			pr_err("%s: no tr or data\n", __func__);
-		return;
-	}
-	__trace_special(tr, data, type, addr, arg);
+void mmio_trace_mapping(struct mmiotrace_map *map)
+{
+	struct trace_array *tr = mmio_trace_array;
+	struct trace_array_cpu *data;
+
+	preempt_disable();
+	data = tr->data[smp_processor_id()];
+	__trace_mmiotrace_map(tr, data, map);
+	preempt_enable();
 }
-- 
cgit v1.2.3


From 138295373ccf7625fcb0218dfea114837983bc39 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:20:58 +0200
Subject: ftrace: mmiotrace update, #2

another weekend, another patch. This should apply on top of my previous patch
from March 23rd.

Summary of changes:
- Print PCI device list in output header
- work around recursive probe hits on SMP
- refactor dis/arm_kmmio_fault_page() and add check for page levels
- remove un/reference_kmmio(), the die notifier hook is registered
permanently into the list
- explicitly check for single stepping in die notifier callback

I have tested this version on my UP Athlon64 desktop with Nouveau, and
SMP Core 2 Duo laptop with the proprietary nvidia driver. Both systems
are 64-bit. One previously unknown bug crept into daylight: the ftrace
framework's output routines print the first entry last after buffer has
wrapped around.

The most important regressions compared to non-ftrace mmiotrace at this
time are:
- failure of trace_pipe file
- illegal lines in output file
- unaware of losing data due to buffer full

Personally I'd like to see these three solved before submitting to
mainline. Other issues may come up once we know when we lose events.

Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/mmiotrace/kmmio.c    | 186 ++++++++++++++---------------------
 arch/x86/kernel/mmiotrace/mmio-mod.c |   3 -
 include/linux/mmiotrace.h            |   2 -
 kernel/trace/trace_mmiotrace.c       |  47 ++++++++-
 4 files changed, 120 insertions(+), 118 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/mmiotrace/kmmio.c b/arch/x86/kernel/mmiotrace/kmmio.c
index efb467933087..cd0d95fe4fe6 100644
--- a/arch/x86/kernel/mmiotrace/kmmio.c
+++ b/arch/x86/kernel/mmiotrace/kmmio.c
@@ -5,15 +5,12 @@
  *     2008 Pekka Paalanen <pq@iki.fi>
  */
 
-#include <linux/version.h>
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/slab.h>
 #include <linux/kernel.h>
-#include <linux/mm.h>
 #include <linux/uaccess.h>
 #include <linux/ptrace.h>
 #include <linux/preempt.h>
@@ -22,10 +19,9 @@
 #include <linux/mutex.h>
 #include <asm/io.h>
 #include <asm/cacheflush.h>
-#include <asm/errno.h>
 #include <asm/tlbflush.h>
-#include <asm/pgtable.h>
-
+#include <asm/errno.h>
+#include <asm/debugreg.h>
 #include <linux/mmiotrace.h>
 
 #define KMMIO_PAGE_HASH_BITS 4
@@ -57,14 +53,9 @@ struct kmmio_context {
 	int active;
 };
 
-static int kmmio_die_notifier(struct notifier_block *nb, unsigned long val,
-								void *args);
-
-static DEFINE_MUTEX(kmmio_init_mutex);
 static DEFINE_SPINLOCK(kmmio_lock);
 
-/* These are protected by kmmio_lock */
-static int kmmio_initialized;
+/* Protected by kmmio_lock */
 unsigned int kmmio_count;
 
 /* Read-protected by RCU, write-protected by kmmio_lock. */
@@ -79,60 +70,6 @@ static struct list_head *kmmio_page_list(unsigned long page)
 /* Accessed per-cpu */
 static DEFINE_PER_CPU(struct kmmio_context, kmmio_ctx);
 
-/* protected by kmmio_init_mutex */
-static struct notifier_block nb_die = {
-	.notifier_call = kmmio_die_notifier
-};
-
-/**
- * Makes sure kmmio is initialized and usable.
- * This must be called before any other kmmio function defined here.
- * May sleep.
- */
-void reference_kmmio(void)
-{
-	mutex_lock(&kmmio_init_mutex);
-	spin_lock_irq(&kmmio_lock);
-	if (!kmmio_initialized) {
-		int i;
-		for (i = 0; i < KMMIO_PAGE_TABLE_SIZE; i++)
-			INIT_LIST_HEAD(&kmmio_page_table[i]);
-		if (register_die_notifier(&nb_die))
-			BUG();
-	}
-	kmmio_initialized++;
-	spin_unlock_irq(&kmmio_lock);
-	mutex_unlock(&kmmio_init_mutex);
-}
-EXPORT_SYMBOL_GPL(reference_kmmio);
-
-/**
- * Clean up kmmio after use. This must be called for every call to
- * reference_kmmio(). All probes registered after the corresponding
- * reference_kmmio() must have been unregistered when calling this.
- * May sleep.
- */
-void unreference_kmmio(void)
-{
-	bool unreg = false;
-
-	mutex_lock(&kmmio_init_mutex);
-	spin_lock_irq(&kmmio_lock);
-
-	if (kmmio_initialized == 1) {
-		BUG_ON(is_kmmio_active());
-		unreg = true;
-	}
-	kmmio_initialized--;
-	BUG_ON(kmmio_initialized < 0);
-	spin_unlock_irq(&kmmio_lock);
-
-	if (unreg)
-		unregister_die_notifier(&nb_die); /* calls sync_rcu() */
-	mutex_unlock(&kmmio_init_mutex);
-}
-EXPORT_SYMBOL(unreference_kmmio);
-
 /*
  * this is basically a dynamic stabbing problem:
  * Could use the existing prio tree code or
@@ -167,58 +104,56 @@ static struct kmmio_fault_page *get_kmmio_fault_page(unsigned long page)
 	return NULL;
 }
 
-/** Mark the given page as not present. Access to it will trigger a fault. */
-static void arm_kmmio_fault_page(unsigned long page, int *page_level)
+static void set_page_present(unsigned long addr, bool present, int *pglevel)
 {
-	unsigned long address = page & PAGE_MASK;
+	pteval_t pteval;
+	pmdval_t pmdval;
 	int level;
-	pte_t *pte = lookup_address(address, &level);
+	pmd_t *pmd;
+	pte_t *pte = lookup_address(addr, &level);
 
 	if (!pte) {
-		pr_err("kmmio: Error in %s: no pte for page 0x%08lx\n",
-							__func__, page);
+		pr_err("kmmio: no pte for page 0x%08lx\n", addr);
 		return;
 	}
 
-	if (level == PG_LEVEL_2M) {
-		pmd_t *pmd = (pmd_t *)pte;
-		set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_PRESENT));
-	} else {
-		/* PG_LEVEL_4K */
-		set_pte(pte, __pte(pte_val(*pte) & ~_PAGE_PRESENT));
+	if (pglevel)
+		*pglevel = level;
+
+	switch (level) {
+	case PG_LEVEL_2M:
+		pmd = (pmd_t *)pte;
+		pmdval = pmd_val(*pmd) & ~_PAGE_PRESENT;
+		if (present)
+			pmdval |= _PAGE_PRESENT;
+		set_pmd(pmd, __pmd(pmdval));
+		break;
+
+	case PG_LEVEL_4K:
+		pteval = pte_val(*pte) & ~_PAGE_PRESENT;
+		if (present)
+			pteval |= _PAGE_PRESENT;
+		set_pte_atomic(pte, __pte(pteval));
+		break;
+
+	default:
+		pr_err("kmmio: unexpected page level 0x%x.\n", level);
+		return;
 	}
 
-	if (page_level)
-		*page_level = level;
+	__flush_tlb_one(addr);
+}
 
-	__flush_tlb_one(page);
+/** Mark the given page as not present. Access to it will trigger a fault. */
+static void arm_kmmio_fault_page(unsigned long page, int *page_level)
+{
+	set_page_present(page & PAGE_MASK, false, page_level);
 }
 
 /** Mark the given page as present. */
 static void disarm_kmmio_fault_page(unsigned long page, int *page_level)
 {
-	unsigned long address = page & PAGE_MASK;
-	int level;
-	pte_t *pte = lookup_address(address, &level);
-
-	if (!pte) {
-		pr_err("kmmio: Error in %s: no pte for page 0x%08lx\n",
-							__func__, page);
-		return;
-	}
-
-	if (level == PG_LEVEL_2M) {
-		pmd_t *pmd = (pmd_t *)pte;
-		set_pmd(pmd, __pmd(pmd_val(*pmd) | _PAGE_PRESENT));
-	} else {
-		/* PG_LEVEL_4K */
-		set_pte(pte, __pte(pte_val(*pte) | _PAGE_PRESENT));
-	}
-
-	if (page_level)
-		*page_level = level;
-
-	__flush_tlb_one(page);
+	set_page_present(page & PAGE_MASK, true, page_level);
 }
 
 /*
@@ -240,6 +175,7 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 {
 	struct kmmio_context *ctx;
 	struct kmmio_fault_page *faultpage;
+	int ret = 0; /* default to fault not handled */
 
 	/*
 	 * Preemption is now disabled to prevent process switch during
@@ -257,21 +193,35 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 		/*
 		 * Either this page fault is not caused by kmmio, or
 		 * another CPU just pulled the kmmio probe from under
-		 * our feet. In the latter case all hell breaks loose.
+		 * our feet. The latter case should not be possible.
 		 */
 		goto no_kmmio;
 	}
 
 	ctx = &get_cpu_var(kmmio_ctx);
 	if (ctx->active) {
+		disarm_kmmio_fault_page(faultpage->page, NULL);
+		if (addr == ctx->addr) {
+			/*
+			 * On SMP we sometimes get recursive probe hits on the
+			 * same address. Context is already saved, fall out.
+			 */
+			pr_debug("kmmio: duplicate probe hit on CPU %d, for "
+						"address 0x%08lx.\n",
+						smp_processor_id(), addr);
+			ret = 1;
+			goto no_kmmio_ctx;
+		}
 		/*
 		 * Prevent overwriting already in-flight context.
-		 * If this page fault really was due to kmmio trap,
-		 * all hell breaks loose.
+		 * This should not happen, let's hope disarming at least
+		 * prevents a panic.
 		 */
 		pr_emerg("kmmio: recursive probe hit on CPU %d, "
 					"for address 0x%08lx. Ignoring.\n",
 					smp_processor_id(), addr);
+		pr_emerg("kmmio: previous hit was at 0x%08lx.\n",
+					ctx->addr);
 		goto no_kmmio_ctx;
 	}
 	ctx->active++;
@@ -302,14 +252,14 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr)
 	 */
 
 	put_cpu_var(kmmio_ctx);
-	return 1;
+	return 1; /* fault handled */
 
 no_kmmio_ctx:
 	put_cpu_var(kmmio_ctx);
 no_kmmio:
 	rcu_read_unlock();
 	preempt_enable_no_resched();
-	return 0; /* page fault not handled by kmmio */
+	return ret;
 }
 
 /*
@@ -322,8 +272,11 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs)
 	int ret = 0;
 	struct kmmio_context *ctx = &get_cpu_var(kmmio_ctx);
 
-	if (!ctx->active)
+	if (!ctx->active) {
+		pr_debug("kmmio: spurious debug trap on CPU %d.\n",
+							smp_processor_id());
 		goto out;
+	}
 
 	if (ctx->probe && ctx->probe->post_handler)
 		ctx->probe->post_handler(ctx->probe, condition, regs);
@@ -525,9 +478,22 @@ static int kmmio_die_notifier(struct notifier_block *nb, unsigned long val,
 {
 	struct die_args *arg = args;
 
-	if (val == DIE_DEBUG)
+	if (val == DIE_DEBUG && (arg->err & DR_STEP))
 		if (post_kmmio_handler(arg->err, arg->regs) == 1)
 			return NOTIFY_STOP;
 
 	return NOTIFY_DONE;
 }
+
+static struct notifier_block nb_die = {
+	.notifier_call = kmmio_die_notifier
+};
+
+static int __init init_kmmio(void)
+{
+	int i;
+	for (i = 0; i < KMMIO_PAGE_TABLE_SIZE; i++)
+		INIT_LIST_HEAD(&kmmio_page_table[i]);
+	return register_die_notifier(&nb_die);
+}
+fs_initcall(init_kmmio); /* should be before device_initcall() */
diff --git a/arch/x86/kernel/mmiotrace/mmio-mod.c b/arch/x86/kernel/mmiotrace/mmio-mod.c
index 62abc281a512..8256546d49bf 100644
--- a/arch/x86/kernel/mmiotrace/mmio-mod.c
+++ b/arch/x86/kernel/mmiotrace/mmio-mod.c
@@ -415,8 +415,6 @@ void enable_mmiotrace(void)
 	if (is_enabled())
 		goto out;
 
-	reference_kmmio();
-
 #if 0 /* XXX: tracing does not support text entries */
 	marker_file = debugfs_create_file("marker", 0660, dir, NULL,
 								&fops_marker);
@@ -448,7 +446,6 @@ void disable_mmiotrace(void)
 	spin_unlock_irq(&trace_lock);
 
 	clear_trace_list(); /* guarantees: no more kmmio callbacks */
-	unreference_kmmio();
 	if (marker_file) {
 		debugfs_remove(marker_file);
 		marker_file = NULL;
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index c88a9c197d22..dd6b64b160fc 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -31,8 +31,6 @@ static inline int is_kmmio_active(void)
 	return kmmio_count;
 }
 
-extern void reference_kmmio(void);
-extern void unreference_kmmio(void);
 extern int register_kmmio_probe(struct kmmio_probe *p);
 extern void unregister_kmmio_probe(struct kmmio_probe *p);
 
diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c
index 3a12b1ad0c63..361472b5788c 100644
--- a/kernel/trace/trace_mmiotrace.c
+++ b/kernel/trace/trace_mmiotrace.c
@@ -8,6 +8,7 @@
 
 #include <linux/kernel.h>
 #include <linux/mmiotrace.h>
+#include <linux/pci.h>
 
 #include "trace.h"
 
@@ -53,12 +54,52 @@ static void mmio_trace_ctrl_update(struct trace_array *tr)
 	}
 }
 
+static int mmio_print_pcidev(struct trace_seq *s, const struct pci_dev *dev)
+{
+	int ret = 0;
+	int i;
+	resource_size_t start, end;
+	const struct pci_driver *drv = pci_dev_driver(dev);
+
+	/* XXX: incomplete checks for trace_seq_printf() return value */
+	ret += trace_seq_printf(s, "PCIDEV %02x%02x %04x%04x %x",
+				dev->bus->number, dev->devfn,
+				dev->vendor, dev->device, dev->irq);
+	/*
+	 * XXX: is pci_resource_to_user() appropriate, since we are
+	 * supposed to interpret the __ioremap() phys_addr argument based on
+	 * these printed values?
+	 */
+	for (i = 0; i < 7; i++) {
+		pci_resource_to_user(dev, i, &dev->resource[i], &start, &end);
+		ret += trace_seq_printf(s, " %llx",
+			(unsigned long long)(start |
+			(dev->resource[i].flags & PCI_REGION_FLAG_MASK)));
+	}
+	for (i = 0; i < 7; i++) {
+		pci_resource_to_user(dev, i, &dev->resource[i], &start, &end);
+		ret += trace_seq_printf(s, " %llx",
+			dev->resource[i].start < dev->resource[i].end ?
+			(unsigned long long)(end - start) + 1 : 0);
+	}
+	if (drv)
+		ret += trace_seq_printf(s, " %s\n", drv->name);
+	else
+		ret += trace_seq_printf(s, " \n");
+	return ret;
+}
+
 /* XXX: This is not called for trace_pipe file! */
-void mmio_print_header(struct trace_iterator *iter)
+static void mmio_print_header(struct trace_iterator *iter)
 {
 	struct trace_seq *s = &iter->seq;
-	trace_seq_printf(s, "VERSION broken 20070824\n");
-	/* TODO: print /proc/bus/pci/devices contents as PCIDEV lines */
+	struct pci_dev *dev = NULL;
+
+	trace_seq_printf(s, "VERSION 20070824\n");
+
+	for_each_pci_dev(dev)
+		mmio_print_pcidev(s, dev);
+	/* XXX: return value? What if header is very long? */
 }
 
 static int mmio_print_rw(struct trace_iterator *iter)
-- 
cgit v1.2.3


From 970e6fa03885f32cc43e42cb08c73a5f54cd8bd9 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:21:03 +0200
Subject: mmiotrace: code style cleanups

From c2da03771e29159627c5c7b9509ec70bce9f91ee Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 28 Apr 2008 21:25:22 +0300

Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/kmmio.c         | 4 ++--
 arch/x86/mm/mmio-mod.c      | 7 +++----
 arch/x86/mm/testmmiotrace.c | 2 +-
 include/linux/mmiotrace.h   | 6 +-----
 4 files changed, 7 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/mm/kmmio.c b/arch/x86/mm/kmmio.c
index 3ad27b8504a5..6a92d9111b64 100644
--- a/arch/x86/mm/kmmio.c
+++ b/arch/x86/mm/kmmio.c
@@ -17,10 +17,10 @@
 #include <linux/percpu.h>
 #include <linux/kdebug.h>
 #include <linux/mutex.h>
-#include <asm/io.h>
+#include <linux/io.h>
 #include <asm/cacheflush.h>
 #include <asm/tlbflush.h>
-#include <asm/errno.h>
+#include <linux/errno.h>
 #include <asm/debugreg.h>
 #include <linux/mmiotrace.h>
 
diff --git a/arch/x86/mm/mmio-mod.c b/arch/x86/mm/mmio-mod.c
index 1f77d8532037..a8d2a0019da4 100644
--- a/arch/x86/mm/mmio-mod.c
+++ b/arch/x86/mm/mmio-mod.c
@@ -24,7 +24,7 @@
 #include <linux/module.h>
 #include <linux/debugfs.h>
 #include <linux/uaccess.h>
-#include <asm/io.h>
+#include <linux/io.h>
 #include <linux/version.h>
 #include <linux/kallsyms.h>
 #include <asm/pgtable.h>
@@ -418,11 +418,10 @@ static void enter_uniprocessor(void)
 
 	for_each_cpu_mask(cpu, downed_cpus) {
 		err = cpu_down(cpu);
-		if (!err) {
+		if (!err)
 			pr_info(NAME "CPU%d is down.\n", cpu);
-		} else {
+		else
 			pr_err(NAME "Error taking CPU%d down: %d\n", cpu, err);
-		}
 	}
 	if (num_online_cpus() > 1)
 		pr_warning(NAME "multiple CPUs still online, "
diff --git a/arch/x86/mm/testmmiotrace.c b/arch/x86/mm/testmmiotrace.c
index cfa60b227c8d..d877c5b423ef 100644
--- a/arch/x86/mm/testmmiotrace.c
+++ b/arch/x86/mm/testmmiotrace.c
@@ -2,7 +2,7 @@
  * Written by Pekka Paalanen, 2008 <pq@iki.fi>
  */
 #include <linux/module.h>
-#include <asm/io.h>
+#include <linux/io.h>
 
 #define MODULE_NAME "testmmiotrace"
 
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index dd6b64b160fc..de8e91258da7 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -1,9 +1,7 @@
 #ifndef MMIOTRACE_H
 #define MMIOTRACE_H
 
-#include <asm/types.h>
-
-#ifdef __KERNEL__
+#include <linux/types.h>
 
 #include <linux/list.h>
 
@@ -84,6 +82,4 @@ extern void disable_mmiotrace(void);
 extern void mmio_trace_rw(struct mmiotrace_rw *rw);
 extern void mmio_trace_mapping(struct mmiotrace_map *map);
 
-#endif /* __KERNEL__ */
-
 #endif /* MMIOTRACE_H */
-- 
cgit v1.2.3


From dee310d0adf41019aca476052ac3085ff286d9be Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:21:03 +0200
Subject: x86 mmiotrace: use resource_size_t for phys addresses

Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/mmio-mod.c         | 11 ++++++-----
 include/linux/mmiotrace.h      | 14 +++++++-------
 kernel/trace/trace_mmiotrace.c | 20 ++++++++++++--------
 3 files changed, 25 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/mm/mmio-mod.c b/arch/x86/mm/mmio-mod.c
index 278998c1998f..3b04a0126121 100644
--- a/arch/x86/mm/mmio-mod.c
+++ b/arch/x86/mm/mmio-mod.c
@@ -48,7 +48,7 @@ struct trap_reason {
 struct remap_trace {
 	struct list_head list;
 	struct kmmio_probe probe;
-	unsigned long phys;
+	resource_size_t phys;
 	unsigned long id;
 };
 
@@ -275,7 +275,7 @@ static void post(struct kmmio_probe *p, unsigned long condition,
 	put_cpu_var(pf_reason);
 }
 
-static void ioremap_trace_core(unsigned long offset, unsigned long size,
+static void ioremap_trace_core(resource_size_t offset, unsigned long size,
 							void __iomem *addr)
 {
 	static atomic_t next_id;
@@ -319,13 +319,14 @@ not_enabled:
 	spin_unlock_irq(&trace_lock);
 }
 
-void
-mmiotrace_ioremap(unsigned long offset, unsigned long size, void __iomem *addr)
+void mmiotrace_ioremap(resource_size_t offset, unsigned long size,
+						void __iomem *addr)
 {
 	if (!is_enabled()) /* recheck and proper locking in *_core() */
 		return;
 
-	pr_debug(NAME "ioremap_*(0x%lx, 0x%lx) = %p\n", offset, size, addr);
+	pr_debug(NAME "ioremap_*(0x%llx, 0x%lx) = %p\n",
+				(unsigned long long)offset, size, addr);
 	if ((filter_offset) && (offset != filter_offset))
 		return;
 	ioremap_trace_core(offset, size, addr);
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index de8e91258da7..5cbbc374e945 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -2,7 +2,6 @@
 #define MMIOTRACE_H
 
 #include <linux/types.h>
-
 #include <linux/list.h>
 
 struct kmmio_probe;
@@ -37,14 +36,15 @@ extern int kmmio_handler(struct pt_regs *regs, unsigned long addr);
 
 /* Called from ioremap.c */
 #ifdef CONFIG_MMIOTRACE
-extern void
-mmiotrace_ioremap(unsigned long offset, unsigned long size, void __iomem *addr);
+extern void mmiotrace_ioremap(resource_size_t offset, unsigned long size,
+							void __iomem *addr);
 extern void mmiotrace_iounmap(volatile void __iomem *addr);
 #else
-static inline void
-mmiotrace_ioremap(unsigned long offset, unsigned long size, void __iomem *addr)
+static inline void mmiotrace_ioremap(resource_size_t offset,
+					unsigned long size, void __iomem *addr)
 {
 }
+
 static inline void mmiotrace_iounmap(volatile void __iomem *addr)
 {
 }
@@ -60,7 +60,7 @@ enum mm_io_opcode {
 };
 
 struct mmiotrace_rw {
-	unsigned long phys;	/* PCI address of register */
+	resource_size_t phys;	/* PCI address of register */
 	unsigned long value;
 	unsigned long pc;	/* optional program counter */
 	int map_id;
@@ -69,7 +69,7 @@ struct mmiotrace_rw {
 };
 
 struct mmiotrace_map {
-	unsigned long phys;	/* base address in PCI space */
+	resource_size_t phys;	/* base address in PCI space */
 	unsigned long virt;	/* base virtual address */
 	unsigned long len;	/* mapping size */
 	int map_id;
diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c
index 3c1dacdc2d85..b13dc19dcbb4 100644
--- a/kernel/trace/trace_mmiotrace.c
+++ b/kernel/trace/trace_mmiotrace.c
@@ -184,20 +184,23 @@ static int mmio_print_rw(struct trace_iterator *iter)
 	switch (entry->mmiorw.opcode) {
 	case MMIO_READ:
 		ret = trace_seq_printf(s,
-			"R %d %lu.%06lu %d 0x%lx 0x%lx 0x%lx %d\n",
-			rw->width, secs, usec_rem, rw->map_id, rw->phys,
+			"R %d %lu.%06lu %d 0x%llx 0x%lx 0x%lx %d\n",
+			rw->width, secs, usec_rem, rw->map_id,
+			(unsigned long long)rw->phys,
 			rw->value, rw->pc, 0);
 		break;
 	case MMIO_WRITE:
 		ret = trace_seq_printf(s,
-			"W %d %lu.%06lu %d 0x%lx 0x%lx 0x%lx %d\n",
-			rw->width, secs, usec_rem, rw->map_id, rw->phys,
+			"W %d %lu.%06lu %d 0x%llx 0x%lx 0x%lx %d\n",
+			rw->width, secs, usec_rem, rw->map_id,
+			(unsigned long long)rw->phys,
 			rw->value, rw->pc, 0);
 		break;
 	case MMIO_UNKNOWN_OP:
 		ret = trace_seq_printf(s,
-			"UNKNOWN %lu.%06lu %d 0x%lx %02x,%02x,%02x 0x%lx %d\n",
-			secs, usec_rem, rw->map_id, rw->phys,
+			"UNKNOWN %lu.%06lu %d 0x%llx %02x,%02x,%02x 0x%lx %d\n",
+			secs, usec_rem, rw->map_id,
+			(unsigned long long)rw->phys,
 			(rw->value >> 16) & 0xff, (rw->value >> 8) & 0xff,
 			(rw->value >> 0) & 0xff, rw->pc, 0);
 		break;
@@ -223,8 +226,9 @@ static int mmio_print_map(struct trace_iterator *iter)
 	switch (entry->mmiorw.opcode) {
 	case MMIO_PROBE:
 		ret = trace_seq_printf(s,
-			"MAP %lu.%06lu %d 0x%lx 0x%lx 0x%lx 0x%lx %d\n",
-			secs, usec_rem, m->map_id, m->phys, m->virt, m->len,
+			"MAP %lu.%06lu %d 0x%llx 0x%lx 0x%lx 0x%lx %d\n",
+			secs, usec_rem, m->map_id,
+			(unsigned long long)m->phys, m->virt, m->len,
 			0UL, 0);
 		break;
 	case MMIO_UNPROBE:
-- 
cgit v1.2.3


From a50445d76c22a34ae149704ea5adaef171c8acb7 Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Mon, 12 May 2008 21:21:03 +0200
Subject: mmiotrace: rename kmmio_probe::user_data to :private.

Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/mmio-mod.c    | 4 ++--
 include/linux/mmiotrace.h | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/mm/mmio-mod.c b/arch/x86/mm/mmio-mod.c
index 3b04a0126121..ed0e0e90b3ef 100644
--- a/arch/x86/mm/mmio-mod.c
+++ b/arch/x86/mm/mmio-mod.c
@@ -191,7 +191,7 @@ static void pre(struct kmmio_probe *p, struct pt_regs *regs,
 	struct mmiotrace_rw *my_trace = &get_cpu_var(cpu_trace);
 	const unsigned long instptr = instruction_pointer(regs);
 	const enum reason_type type = get_ins_type(instptr);
-	struct remap_trace *trace = p->user_data;
+	struct remap_trace *trace = p->private;
 
 	/* it doesn't make sense to have more than one active trace per cpu */
 	if (my_reason->active_traces)
@@ -299,7 +299,7 @@ static void ioremap_trace_core(resource_size_t offset, unsigned long size,
 			.len = size,
 			.pre_handler = pre,
 			.post_handler = post,
-			.user_data = trace
+			.private = trace
 		},
 		.phys = offset,
 		.id = atomic_inc_return(&next_id)
diff --git a/include/linux/mmiotrace.h b/include/linux/mmiotrace.h
index 5cbbc374e945..61d19e1b7a0b 100644
--- a/include/linux/mmiotrace.h
+++ b/include/linux/mmiotrace.h
@@ -18,7 +18,7 @@ struct kmmio_probe {
 	unsigned long len; /* length of the probe region */
 	kmmio_pre_handler_t pre_handler; /* Called before addr is executed. */
 	kmmio_post_handler_t post_handler; /* Called after addr is executed */
-	void *user_data;
+	void *private;
 };
 
 /* kmmio is active by some kmmio_probes? */
-- 
cgit v1.2.3


From 42fdfa238a23643226910acf922ea930b3286032 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 24 May 2008 23:14:51 +0200
Subject: namespacecheck: more kernel/printk.c fixes

[ Stephen Rothwell <sfr@canb.auug.org.au>: build fix ]

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/kernel.h |  6 ------
 kernel/printk.c        | 13 -------------
 2 files changed, 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 792bf0aa779b..f2a668c195bf 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -184,9 +184,6 @@ asmlinkage int vprintk(const char *fmt, va_list args)
 	__attribute__ ((format (printf, 1, 0)));
 asmlinkage int printk(const char * fmt, ...)
 	__attribute__ ((format (printf, 1, 2))) __cold;
-extern int log_buf_get_len(void);
-extern int log_buf_read(int idx);
-extern int log_buf_copy(char *dest, int idx, int len);
 
 extern int printk_ratelimit_jiffies;
 extern int printk_ratelimit_burst;
@@ -202,9 +199,6 @@ static inline int vprintk(const char *s, va_list args) { return 0; }
 static inline int printk(const char *s, ...)
 	__attribute__ ((format (printf, 1, 2)));
 static inline int __cold printk(const char *s, ...) { return 0; }
-static inline int log_buf_get_len(void) { return 0; }
-static inline int log_buf_read(int idx) { return 0; }
-static inline int log_buf_copy(char *dest, int idx, int len) { return 0; }
 static inline int printk_ratelimit(void) { return 0; }
 static inline int __printk_ratelimit(int ratelimit_jiffies, \
 				     int ratelimit_burst) { return 0; }
diff --git a/kernel/printk.c b/kernel/printk.c
index b620e3d96136..55d16e57499a 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -267,19 +267,6 @@ int log_buf_copy(char *dest, int idx, int len)
 	return ret;
 }
 
-/*
- * Extract a single character from the log buffer.
- */
-static int log_buf_read(int idx)
-{
-	char ret;
-
-	if (log_buf_copy(&ret, idx, 1) == 1)
-		return ret;
-	else
-		return -1;
-}
-
 /*
  * Commands to do_syslog:
  *
-- 
cgit v1.2.3


From 9c44bc03fff44ff04237a7d92e35304a0e50c331 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:21:04 +0200
Subject: softlockup: allow panic on lockup

allow users to configure the softlockup detector to generate a panic
instead of a warning message.

high-availability systems might opt for this strict method (combined
with panic_timeout= boot option/sysctl), instead of generating
softlockup warnings ad infinitum.

also, automated tests work better if the system reboots reliably (into
a safe kernel) in case of a lockup.

The full spectrum of configurability is supported: boot option, sysctl
option and Kconfig option.

it's default-disabled.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/kernel-parameters.txt |  3 +++
 include/linux/sched.h               |  3 ++-
 kernel/softlockup.c                 | 21 +++++++++++++++++++++
 kernel/sysctl.c                     | 11 +++++++++++
 lib/Kconfig.debug                   | 26 +++++++++++++++++++++++++-
 5 files changed, 62 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index e07c432c731f..042588fa12e5 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1971,6 +1971,9 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	snd-ymfpci=	[HW,ALSA]
 
+	softlockup_panic=
+			[KNL] Should the soft-lockup detector generate panics.
+
 	sonypi.*=	[HW] Sony Programmable I/O Control Device driver
 			See Documentation/sonypi.txt
 
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5395a6176f4b..71f5972dc48e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -294,7 +294,8 @@ extern void softlockup_tick(void);
 extern void spawn_softlockup_task(void);
 extern void touch_softlockup_watchdog(void);
 extern void touch_all_softlockup_watchdogs(void);
-extern unsigned long  softlockup_thresh;
+extern unsigned int  softlockup_panic;
+extern unsigned long softlockup_thresh;
 extern unsigned long sysctl_hung_task_check_count;
 extern unsigned long sysctl_hung_task_timeout_secs;
 extern unsigned long sysctl_hung_task_warnings;
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 01b6522fd92b..78e0ad21cb0c 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -27,6 +27,21 @@ static DEFINE_PER_CPU(struct task_struct *, watchdog_task);
 static int __read_mostly did_panic;
 unsigned long __read_mostly softlockup_thresh = 60;
 
+/*
+ * Should we panic (and reboot, if panic_timeout= is set) when a
+ * soft-lockup occurs:
+ */
+unsigned int __read_mostly softlockup_panic =
+				CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE;
+
+static int __init softlockup_panic_setup(char *str)
+{
+	softlockup_panic = simple_strtoul(str, NULL, 0);
+
+	return 1;
+}
+__setup("softlockup_panic=", softlockup_panic_setup);
+
 static int
 softlock_panic(struct notifier_block *this, unsigned long event, void *ptr)
 {
@@ -120,6 +135,9 @@ void softlockup_tick(void)
 	else
 		dump_stack();
 	spin_unlock(&print_lock);
+
+	if (softlockup_panic)
+		panic("softlockup: hung tasks");
 }
 
 /*
@@ -172,6 +190,9 @@ static void check_hung_task(struct task_struct *t, unsigned long now)
 
 	t->last_switch_timestamp = now;
 	touch_nmi_watchdog();
+
+	if (softlockup_panic)
+		panic("softlockup: blocked tasks");
 }
 
 /*
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 29116652dca8..2d3b388c402d 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -727,6 +727,17 @@ static struct ctl_table kern_table[] = {
 	},
 #endif
 #ifdef CONFIG_DETECT_SOFTLOCKUP
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "softlockup_panic",
+		.data		= &softlockup_panic,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_doulongvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero,
+		.extra2		= &one,
+	},
 	{
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "softlockup_thresh",
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index d2099f41aa1e..509ae35a9ef5 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -147,7 +147,7 @@ config DETECT_SOFTLOCKUP
 	help
 	  Say Y here to enable the kernel to detect "soft lockups",
 	  which are bugs that cause the kernel to loop in kernel
-	  mode for more than 10 seconds, without giving other tasks a
+	  mode for more than 60 seconds, without giving other tasks a
 	  chance to run.
 
 	  When a soft-lockup is detected, the kernel will print the
@@ -159,6 +159,30 @@ config DETECT_SOFTLOCKUP
 	   can be detected via the NMI-watchdog, on platforms that
 	   support it.)
 
+config BOOTPARAM_SOFTLOCKUP_PANIC
+	bool "Panic (Reboot) On Soft Lockups"
+	depends on DETECT_SOFTLOCKUP
+	help
+	  Say Y here to enable the kernel to panic on "soft lockups",
+	  which are bugs that cause the kernel to loop in kernel
+	  mode for more than 60 seconds, without giving other tasks a
+	  chance to run.
+
+	  The panic can be used in combination with panic_timeout,
+	  to cause the system to reboot automatically after a
+	  lockup has been detected. This feature is useful for
+	  high-availability systems that have uptime guarantees and
+	  where a lockup must be resolved ASAP.
+
+	  Say N if unsure.
+
+config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
+	int
+	depends on DETECT_SOFTLOCKUP
+	range 0 1
+	default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
+	default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
+
 config SCHED_DEBUG
 	bool "Collect scheduler debugging info"
 	depends on DEBUG_KERNEL && PROC_FS
-- 
cgit v1.2.3


From 9383d9679056e6cc4e7ff70f31da945a268238f4 Mon Sep 17 00:00:00 2001
From: Dimitri Sivanich <sivanich@sgi.com>
Date: Mon, 12 May 2008 21:21:14 +0200
Subject: softlockup: fix softlockup_thresh unaligned access and disable
 detection at runtime

Fix unaligned access errors when setting softlockup_thresh on
64 bit platforms.

Allow softlockup detection to be disabled by setting
softlockup_thresh <= 0.

Detect that boot time softlockup detection has been disabled
earlier in softlockup_tick.

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h |  2 +-
 kernel/softlockup.c   | 12 ++++++++++--
 kernel/sysctl.c       |  9 +++++----
 3 files changed, 16 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 71f5972dc48e..ea26221644e2 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -295,10 +295,10 @@ extern void spawn_softlockup_task(void);
 extern void touch_softlockup_watchdog(void);
 extern void touch_all_softlockup_watchdogs(void);
 extern unsigned int  softlockup_panic;
-extern unsigned long softlockup_thresh;
 extern unsigned long sysctl_hung_task_check_count;
 extern unsigned long sysctl_hung_task_timeout_secs;
 extern unsigned long sysctl_hung_task_warnings;
+extern int softlockup_thresh;
 #else
 static inline void softlockup_tick(void)
 {
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 78e0ad21cb0c..a3a0b239b7f7 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -25,7 +25,7 @@ static DEFINE_PER_CPU(unsigned long, print_timestamp);
 static DEFINE_PER_CPU(struct task_struct *, watchdog_task);
 
 static int __read_mostly did_panic;
-unsigned long __read_mostly softlockup_thresh = 60;
+int __read_mostly softlockup_thresh = 60;
 
 /*
  * Should we panic (and reboot, if panic_timeout= is set) when a
@@ -94,6 +94,14 @@ void softlockup_tick(void)
 	struct pt_regs *regs = get_irq_regs();
 	unsigned long now;
 
+	/* Is detection switched off? */
+	if (!per_cpu(watchdog_task, this_cpu) || softlockup_thresh <= 0) {
+		/* Be sure we don't false trigger if switched back on */
+		if (touch_timestamp)
+			per_cpu(touch_timestamp, this_cpu) = 0;
+		return;
+	}
+
 	if (touch_timestamp == 0) {
 		touch_softlockup_watchdog();
 		return;
@@ -104,7 +112,7 @@ void softlockup_tick(void)
 	/* report at most once a second */
 	if ((print_timestamp >= touch_timestamp &&
 			print_timestamp < (touch_timestamp + 1)) ||
-			did_panic || !per_cpu(watchdog_task, this_cpu)) {
+			did_panic) {
 		return;
 	}
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 2d3b388c402d..31c19a79738d 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -84,12 +84,13 @@ extern int latencytop_enabled;
 extern int sysctl_nr_open_min, sysctl_nr_open_max;
 
 /* Constants used for minimum and  maximum */
-#if defined(CONFIG_DETECT_SOFTLOCKUP) || defined(CONFIG_HIGHMEM)
+#ifdef CONFIG_HIGHMEM
 static int one = 1;
 #endif
 
 #ifdef CONFIG_DETECT_SOFTLOCKUP
 static int sixty = 60;
+static int neg_one = -1;
 #endif
 
 #ifdef CONFIG_MMU
@@ -742,11 +743,11 @@ static struct ctl_table kern_table[] = {
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "softlockup_thresh",
 		.data		= &softlockup_thresh,
-		.maxlen		= sizeof(unsigned long),
+		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_doulongvec_minmax,
+		.proc_handler	= &proc_dointvec_minmax,
 		.strategy	= &sysctl_intvec,
-		.extra1		= &one,
+		.extra1		= &neg_one,
 		.extra2		= &sixty,
 	},
 	{
-- 
cgit v1.2.3


From 63687a528c39a67c1a213cdffa09feb0e6af9dbe Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Mon, 12 May 2008 15:44:41 +0200
Subject: x86: move tracedata to RODATA

.. allowing it to be write-protected just as other read-only data
under CONFIG_DEBUG_RODATA.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/vmlinux_32.lds.S  |  7 -------
 arch/x86/kernel/vmlinux_64.lds.S  |  7 -------
 drivers/base/power/trace.c        |  2 +-
 include/asm-generic/vmlinux.lds.h | 14 ++++++++++++++
 include/asm-x86/resume-trace.h    |  2 +-
 include/linux/resume-trace.h      |  2 +-
 6 files changed, 17 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S
index ce5ed083a1e9..2674f5796275 100644
--- a/arch/x86/kernel/vmlinux_32.lds.S
+++ b/arch/x86/kernel/vmlinux_32.lds.S
@@ -60,13 +60,6 @@ SECTIONS
 
   BUG_TABLE :text
 
-  . = ALIGN(4);
-  .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) {
-  	__tracedata_start = .;
-	*(.tracedata)
-  	__tracedata_end = .;
-  }
-
   RODATA
 
   /* writeable */
diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S
index fad3674b06a5..687041bfbae5 100644
--- a/arch/x86/kernel/vmlinux_64.lds.S
+++ b/arch/x86/kernel/vmlinux_64.lds.S
@@ -53,13 +53,6 @@ SECTIONS
 
   RODATA
 
-  . = ALIGN(4);
-  .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) {
-  	__tracedata_start = .;
-	*(.tracedata)
-  	__tracedata_end = .;
-  }
-
   . = ALIGN(PAGE_SIZE);		/* Align data segment to page size boundary */
 				/* Data */
   .data : AT(ADDR(.data) - LOAD_OFFSET) {
diff --git a/drivers/base/power/trace.c b/drivers/base/power/trace.c
index 2b4b392dcbc1..87a7f1d02578 100644
--- a/drivers/base/power/trace.c
+++ b/drivers/base/power/trace.c
@@ -153,7 +153,7 @@ EXPORT_SYMBOL(set_trace_device);
  * it's not any guarantee, but it's a high _likelihood_ that
  * the match is valid).
  */
-void generate_resume_trace(void *tracedata, unsigned int user)
+void generate_resume_trace(const void *tracedata, unsigned int user)
 {
 	unsigned short lineno = *(unsigned short *)tracedata;
 	const char *file = *(const char **)(tracedata + 2);
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index f054778e916c..f1992dc5c424 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -93,6 +93,8 @@
 		VMLINUX_SYMBOL(__end_rio_route_ops) = .;		\
 	}								\
 									\
+	TRACEDATA							\
+									\
 	/* Kernel symbol table: Normal symbols */			\
 	__ksymtab         : AT(ADDR(__ksymtab) - LOAD_OFFSET) {		\
 		VMLINUX_SYMBOL(__start___ksymtab) = .;			\
@@ -318,6 +320,18 @@
 		__stop___bug_table = .;					\
 	}
 
+#ifdef CONFIG_PM_TRACE
+#define TRACEDATA							\
+	. = ALIGN(4);							\
+	.tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) {		\
+	  	__tracedata_start = .;					\
+		*(.tracedata)						\
+	  	__tracedata_end = .;					\
+	}
+#else
+#define TRACEDATA
+#endif
+
 #define NOTES								\
 	.notes : AT(ADDR(.notes) - LOAD_OFFSET) {			\
 		VMLINUX_SYMBOL(__start_notes) = .;			\
diff --git a/include/asm-x86/resume-trace.h b/include/asm-x86/resume-trace.h
index 2557514d7ef6..8d9f0b41ee86 100644
--- a/include/asm-x86/resume-trace.h
+++ b/include/asm-x86/resume-trace.h
@@ -6,7 +6,7 @@
 #define TRACE_RESUME(user)					\
 do {								\
 	if (pm_trace_enabled) {					\
-		void *tracedata;				\
+		const void *tracedata;				\
 		asm volatile(_ASM_MOV_UL " $1f,%0\n"		\
 			     ".section .tracedata,\"a\"\n"	\
 			     "1:\t.word %c1\n\t"		\
diff --git a/include/linux/resume-trace.h b/include/linux/resume-trace.h
index f3f4f28c6960..c9ba2fdf807d 100644
--- a/include/linux/resume-trace.h
+++ b/include/linux/resume-trace.h
@@ -8,7 +8,7 @@ extern int pm_trace_enabled;
 
 struct device;
 extern void set_trace_device(struct device *);
-extern void generate_resume_trace(void *tracedata, unsigned int user);
+extern void generate_resume_trace(const void *tracedata, unsigned int user);
 
 #define TRACE_DEVICE(dev) do { \
 	if (pm_trace_enabled) \
-- 
cgit v1.2.3


From 962cf36c5bf6d2840b8d66ee9a606fae2f540bbd Mon Sep 17 00:00:00 2001
From: "Carlos R. Mafra" <crmafra2@gmail.com>
Date: Thu, 15 May 2008 11:15:37 -0300
Subject: Remove argument from open_softirq which is always NULL

As git-grep shows, open_softirq() is always called with the last argument
being NULL

block/blk-core.c:       open_softirq(BLOCK_SOFTIRQ, blk_done_softirq, NULL);
kernel/hrtimer.c:       open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq, NULL);
kernel/rcuclassic.c:    open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
kernel/rcupreempt.c:    open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
kernel/sched.c: open_softirq(SCHED_SOFTIRQ, run_rebalance_domains, NULL);
kernel/softirq.c:       open_softirq(TASKLET_SOFTIRQ, tasklet_action, NULL);
kernel/softirq.c:       open_softirq(HI_SOFTIRQ, tasklet_hi_action, NULL);
kernel/timer.c: open_softirq(TIMER_SOFTIRQ, run_timer_softirq, NULL);
net/core/dev.c: open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL);
net/core/dev.c: open_softirq(NET_RX_SOFTIRQ, net_rx_action, NULL);

This observation has already been made by Matthew Wilcox in June 2002
(http://www.cs.helsinki.fi/linux/linux-kernel/2002-25/0687.html)

"I notice that none of the current softirq routines use the data element
passed to them."

and the situation hasn't changed since them. So it appears we can safely
remove that extra argument to save 128 (54) bytes of kernel data (text).

Signed-off-by: Carlos R. Mafra <crmafra@ift.unesp.br>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 block/blk-core.c          | 2 +-
 include/linux/interrupt.h | 3 +--
 kernel/hrtimer.c          | 2 +-
 kernel/rcuclassic.c       | 2 +-
 kernel/rcupreempt.c       | 2 +-
 kernel/sched.c            | 2 +-
 kernel/softirq.c          | 7 +++----
 kernel/timer.c            | 2 +-
 net/core/dev.c            | 4 ++--
 9 files changed, 12 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index 6a9cc0d22a61..75fdc65136e8 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2048,7 +2048,7 @@ int __init blk_dev_init(void)
 	for_each_possible_cpu(i)
 		INIT_LIST_HEAD(&per_cpu(blk_cpu_done, i));
 
-	open_softirq(BLOCK_SOFTIRQ, blk_done_softirq, NULL);
+	open_softirq(BLOCK_SOFTIRQ, blk_done_softirq);
 	register_hotcpu_notifier(&blk_cpu_notifier);
 
 	return 0;
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index f1fc7470d26c..a86186dd0474 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -285,12 +285,11 @@ enum
 struct softirq_action
 {
 	void	(*action)(struct softirq_action *);
-	void	*data;
 };
 
 asmlinkage void do_softirq(void);
 asmlinkage void __do_softirq(void);
-extern void open_softirq(int nr, void (*action)(struct softirq_action*), void *data);
+extern void open_softirq(int nr, void (*action)(struct softirq_action *));
 extern void softirq_init(void);
 #define __raise_softirq_irqoff(nr) do { or_softirq_pending(1UL << (nr)); } while (0)
 extern void raise_softirq_irqoff(unsigned int nr);
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 421be5fe5cc7..861b4088092a 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -1669,7 +1669,7 @@ void __init hrtimers_init(void)
 			  (void *)(long)smp_processor_id());
 	register_cpu_notifier(&hrtimers_nb);
 #ifdef CONFIG_HIGH_RES_TIMERS
-	open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq, NULL);
+	open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq);
 #endif
 }
 
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index f4ffbd0f306f..f6e01f3ae9c6 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -529,7 +529,7 @@ static void __cpuinit rcu_online_cpu(int cpu)
 
 	rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp);
 	rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp);
-	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
+	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
 }
 
 static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index e1cdf196a515..9dd827db359f 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -1125,7 +1125,7 @@ void __init __rcu_init(void)
 	for_each_online_cpu(cpu)
 		rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,	(void *)(long) cpu);
 
-	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
+	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
 }
 
 /*
diff --git a/kernel/sched.c b/kernel/sched.c
index cfa222a91539..56ea3a203a5a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -8154,7 +8154,7 @@ void __init sched_init(void)
 #endif
 
 #ifdef CONFIG_SMP
-	open_softirq(SCHED_SOFTIRQ, run_rebalance_domains, NULL);
+	open_softirq(SCHED_SOFTIRQ, run_rebalance_domains);
 #endif
 
 #ifdef CONFIG_RT_MUTEXES
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 36e061740047..059256874e9b 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -347,9 +347,8 @@ void raise_softirq(unsigned int nr)
 	local_irq_restore(flags);
 }
 
-void open_softirq(int nr, void (*action)(struct softirq_action*), void *data)
+void open_softirq(int nr, void (*action)(struct softirq_action *))
 {
-	softirq_vec[nr].data = data;
 	softirq_vec[nr].action = action;
 }
 
@@ -503,8 +502,8 @@ void __init softirq_init(void)
 			&per_cpu(tasklet_hi_vec, cpu).head;
 	}
 
-	open_softirq(TASKLET_SOFTIRQ, tasklet_action, NULL);
-	open_softirq(HI_SOFTIRQ, tasklet_hi_action, NULL);
+	open_softirq(TASKLET_SOFTIRQ, tasklet_action);
+	open_softirq(HI_SOFTIRQ, tasklet_hi_action);
 }
 
 static int ksoftirqd(void * __bind_cpu)
diff --git a/kernel/timer.c b/kernel/timer.c
index ceacc6626572..b4da888497fa 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1502,7 +1502,7 @@ void __init init_timers(void)
 
 	BUG_ON(err == NOTIFY_BAD);
 	register_cpu_notifier(&timers_nb);
-	open_softirq(TIMER_SOFTIRQ, run_timer_softirq, NULL);
+	open_softirq(TIMER_SOFTIRQ, run_timer_softirq);
 }
 
 /**
diff --git a/net/core/dev.c b/net/core/dev.c
index 582963077877..cf0e16731dc7 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4563,8 +4563,8 @@ static int __init net_dev_init(void)
 
 	dev_boot_phase = 0;
 
-	open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL);
-	open_softirq(NET_RX_SOFTIRQ, net_rx_action, NULL);
+	open_softirq(NET_TX_SOFTIRQ, net_tx_action);
+	open_softirq(NET_RX_SOFTIRQ, net_rx_action);
 
 	hotcpu_notifier(dev_cpu_callback, 0);
 	dst_init();
-- 
cgit v1.2.3


From e9197bf0114661195bee35e7795cfc42164d9b2c Mon Sep 17 00:00:00 2001
From: Paul Jackson <pj@sgi.com>
Date: Wed, 14 May 2008 08:15:10 -0700
Subject: x86 boot: remove some unused extern function declarations

Remove three extern declarations for routines
that don't exist.  Fix a typo in a comment.

Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/numa_64.c | 2 +-
 include/linux/efi.h   | 4 ----
 2 files changed, 1 insertion(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index c5066d519e5d..afb07ffb931d 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -233,7 +233,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start,
 	else
 		bootmap_start = round_up(start, PAGE_SIZE);
 	/*
-	 * SMP_CAHCE_BYTES could be enough, but init_bootmem_node like
+	 * SMP_CACHE_BYTES could be enough, but init_bootmem_node like
 	 * to use that to align to PAGE_SIZE
 	 */
 	bootmap = early_node_mem(nodeid, bootmap_start, end,
diff --git a/include/linux/efi.h b/include/linux/efi.h
index a5f359a7ad0e..807373d467f7 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -287,7 +287,6 @@ efi_guid_unparse(efi_guid_t *guid, char *out)
 extern void efi_init (void);
 extern void *efi_get_pal_addr (void);
 extern void efi_map_pal_code (void);
-extern void efi_map_memmap(void);
 extern void efi_memmap_walk (efi_freemem_callback_t callback, void *arg);
 extern void efi_gettimeofday (struct timespec *ts);
 extern void efi_enter_virtual_mode (void);	/* switch EFI to virtual mode, if possible */
@@ -295,14 +294,11 @@ extern u64 efi_get_iobase (void);
 extern u32 efi_mem_type (unsigned long phys_addr);
 extern u64 efi_mem_attributes (unsigned long phys_addr);
 extern u64 efi_mem_attribute (unsigned long phys_addr, unsigned long size);
-extern int efi_mem_attribute_range (unsigned long phys_addr, unsigned long size,
-				    u64 attr);
 extern int __init efi_uart_console_only (void);
 extern void efi_initialize_iomem_resources(struct resource *code_resource,
 		struct resource *data_resource, struct resource *bss_resource);
 extern unsigned long efi_get_time(void);
 extern int efi_set_rtc_mmss(unsigned long nowtime);
-extern int is_available_memory(efi_memory_desc_t * md);
 extern struct efi_memory_map memmap;
 
 /**
-- 
cgit v1.2.3


From c801ed3860fe2f84281d4cae4c3e6ca87e81e890 Mon Sep 17 00:00:00 2001
From: Paul Jackson <pj@sgi.com>
Date: Wed, 14 May 2008 08:15:23 -0700
Subject: x86 boot: simplify pageblock_bits enum declaration

The use of #defines with '##' pre-processor concatenation is a useful
way to form several symbol names with a common pattern.  But when there
is just a single name obtained from that #define, it's just obfuscation.
Better to just write the plain symbol name, as is.

The following patch is a result of my wasting ten minutes looking through
the kernel to figure out what 'PB_migrate_end' meant, and forgetting what
I came to do, by the time I figured out that the #define PB_range macro
defined it.

Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/pageblock-flags.h | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pageblock-flags.h b/include/linux/pageblock-flags.h
index e875905f7b12..e8c06122be36 100644
--- a/include/linux/pageblock-flags.h
+++ b/include/linux/pageblock-flags.h
@@ -25,13 +25,11 @@
 
 #include <linux/types.h>
 
-/* Macro to aid the definition of ranges of bits */
-#define PB_range(name, required_bits) \
-	name, name ## _end = (name + required_bits) - 1
-
 /* Bit indices that affect a whole block of pages */
 enum pageblock_bits {
-	PB_range(PB_migrate, 3), /* 3 bits required for migrate types */
+	PB_migrate,
+	PB_migrate_end = PB_migrate + 3 - 1,
+			/* 3 bits required for migrate types */
 	NR_PAGEBLOCK_BITS
 };
 
-- 
cgit v1.2.3


From 41c52c0db9607e59f90da7da5309489fa06e887f Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Thu, 22 May 2008 11:46:33 -0400
Subject: ftrace: set_ftrace_notrace feature

While debugging latencies in the RT kernel, I found that it would be nice
to be able to filter away functions from the trace than just to filter
on functions.

I added a new interface to the debugfs tracing directory called

  set_ftrace_notrace

When dynamic frace is enabled, this lets you filter away functions that will
not be recorded in the trace. It is similar to adding 'notrace' to those
functions but by doing it without recompiling the kernel.

Here's how set_ftrace_filter and set_ftrace_notrace interact. Remember, if
set_ftrace_filter is set, it removes all functions from the trace execpt for
those listed in the set_ftrace_filter. set_ftrace_notrace will prevent those
functions from being traced.

If you were to set one function in both set_ftrace_filter and
set_ftrace_notrace and that function was the same, then you would end up
with an empty trace.

the set of functions to trace is:

  set_ftrace_filter == empty then

     all functions not in set_ftrace_notrace

  else

     set of the set_ftrace_filter and not in set of set_ftrace_notrace.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/ftrace.h |   1 +
 kernel/trace/ftrace.c  | 170 +++++++++++++++++++++++++++++++++++++------------
 2 files changed, 131 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 922e23d0196f..ffbbd54a720e 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -48,6 +48,7 @@ enum {
 	FTRACE_FL_FAILED	= (1 << 1),
 	FTRACE_FL_FILTER	= (1 << 2),
 	FTRACE_FL_ENABLED	= (1 << 3),
+	FTRACE_FL_NOTRACE	= (1 << 4),
 };
 
 struct dyn_ftrace {
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 89bd9a6f52ec..2552454609cf 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -170,7 +170,7 @@ static DEFINE_PER_CPU(int, ftrace_shutdown_disable_cpu);
 
 static DEFINE_SPINLOCK(ftrace_shutdown_lock);
 static DEFINE_MUTEX(ftraced_lock);
-static DEFINE_MUTEX(ftrace_filter_lock);
+static DEFINE_MUTEX(ftrace_regex_lock);
 
 struct ftrace_page {
 	struct ftrace_page	*next;
@@ -337,13 +337,12 @@ static void
 __ftrace_replace_code(struct dyn_ftrace *rec,
 		      unsigned char *old, unsigned char *new, int enable)
 {
-	unsigned long ip;
+	unsigned long ip, fl;
 	int failed;
 
 	ip = rec->ip;
 
 	if (ftrace_filtered && enable) {
-		unsigned long fl;
 		/*
 		 * If filtering is on:
 		 *
@@ -356,13 +355,16 @@ __ftrace_replace_code(struct dyn_ftrace *rec,
 		 * If this record is not set to be filtered
 		 * and it is not enabled do nothing.
 		 *
+		 * If this record is set not to trace then
+		 * do nothing.
+		 *
 		 * If this record is not set to be filtered and
 		 * it is enabled, disable it.
 		 */
 		fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_ENABLED);
 
 		if ((fl ==  (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) ||
-		    (fl == 0))
+		    (fl == 0) || (rec->flags & FTRACE_FL_NOTRACE))
 			return;
 
 		/*
@@ -380,9 +382,17 @@ __ftrace_replace_code(struct dyn_ftrace *rec,
 		}
 	} else {
 
-		if (enable)
+		if (enable) {
+			/*
+			 * If this record is set not to trace and is
+			 * not enabled, do nothing.
+			 */
+			fl = rec->flags & (FTRACE_FL_NOTRACE | FTRACE_FL_ENABLED);
+			if (fl == FTRACE_FL_NOTRACE)
+				return;
+
 			new = ftrace_call_replace(ip, FTRACE_ADDR);
-		else
+		} else
 			old = ftrace_call_replace(ip, FTRACE_ADDR);
 
 		if (enable) {
@@ -721,6 +731,7 @@ static int __init ftrace_dyn_table_alloc(void)
 enum {
 	FTRACE_ITER_FILTER	= (1 << 0),
 	FTRACE_ITER_CONT	= (1 << 1),
+	FTRACE_ITER_NOTRACE	= (1 << 2),
 };
 
 #define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */
@@ -754,7 +765,9 @@ t_next(struct seq_file *m, void *v, loff_t *pos)
 		rec = &iter->pg->records[iter->idx++];
 		if ((rec->flags & FTRACE_FL_FAILED) ||
 		    ((iter->flags & FTRACE_ITER_FILTER) &&
-		     !(rec->flags & FTRACE_FL_FILTER))) {
+		     !(rec->flags & FTRACE_FL_FILTER)) ||
+		    ((iter->flags & FTRACE_ITER_NOTRACE) &&
+		     !(rec->flags & FTRACE_FL_NOTRACE))) {
 			rec = NULL;
 			goto retry;
 		}
@@ -847,22 +860,24 @@ int ftrace_avail_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static void ftrace_filter_reset(void)
+static void ftrace_filter_reset(int enable)
 {
 	struct ftrace_page *pg;
 	struct dyn_ftrace *rec;
+	unsigned long type = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE;
 	unsigned i;
 
 	/* keep kstop machine from running */
 	preempt_disable();
-	ftrace_filtered = 0;
+	if (enable)
+		ftrace_filtered = 0;
 	pg = ftrace_pages_start;
 	while (pg) {
 		for (i = 0; i < pg->index; i++) {
 			rec = &pg->records[i];
 			if (rec->flags & FTRACE_FL_FAILED)
 				continue;
-			rec->flags &= ~FTRACE_FL_FILTER;
+			rec->flags &= ~type;
 		}
 		pg = pg->next;
 	}
@@ -870,7 +885,7 @@ static void ftrace_filter_reset(void)
 }
 
 static int
-ftrace_filter_open(struct inode *inode, struct file *file)
+ftrace_regex_open(struct inode *inode, struct file *file, int enable)
 {
 	struct ftrace_iterator *iter;
 	int ret = 0;
@@ -882,15 +897,16 @@ ftrace_filter_open(struct inode *inode, struct file *file)
 	if (!iter)
 		return -ENOMEM;
 
-	mutex_lock(&ftrace_filter_lock);
+	mutex_lock(&ftrace_regex_lock);
 	if ((file->f_mode & FMODE_WRITE) &&
 	    !(file->f_flags & O_APPEND))
-		ftrace_filter_reset();
+		ftrace_filter_reset(enable);
 
 	if (file->f_mode & FMODE_READ) {
 		iter->pg = ftrace_pages_start;
 		iter->pos = -1;
-		iter->flags = FTRACE_ITER_FILTER;
+		iter->flags = enable ? FTRACE_ITER_FILTER :
+			FTRACE_ITER_NOTRACE;
 
 		ret = seq_open(file, &show_ftrace_seq_ops);
 		if (!ret) {
@@ -900,13 +916,25 @@ ftrace_filter_open(struct inode *inode, struct file *file)
 			kfree(iter);
 	} else
 		file->private_data = iter;
-	mutex_unlock(&ftrace_filter_lock);
+	mutex_unlock(&ftrace_regex_lock);
 
 	return ret;
 }
 
+static int
+ftrace_filter_open(struct inode *inode, struct file *file)
+{
+	return ftrace_regex_open(inode, file, 1);
+}
+
+static int
+ftrace_notrace_open(struct inode *inode, struct file *file)
+{
+	return ftrace_regex_open(inode, file, 0);
+}
+
 static ssize_t
-ftrace_filter_read(struct file *file, char __user *ubuf,
+ftrace_regex_read(struct file *file, char __user *ubuf,
 		       size_t cnt, loff_t *ppos)
 {
 	if (file->f_mode & FMODE_READ)
@@ -916,7 +944,7 @@ ftrace_filter_read(struct file *file, char __user *ubuf,
 }
 
 static loff_t
-ftrace_filter_lseek(struct file *file, loff_t offset, int origin)
+ftrace_regex_lseek(struct file *file, loff_t offset, int origin)
 {
 	loff_t ret;
 
@@ -936,13 +964,14 @@ enum {
 };
 
 static void
-ftrace_match(unsigned char *buff, int len)
+ftrace_match(unsigned char *buff, int len, int enable)
 {
 	char str[KSYM_SYMBOL_LEN];
 	char *search = NULL;
 	struct ftrace_page *pg;
 	struct dyn_ftrace *rec;
 	int type = MATCH_FULL;
+	unsigned long flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE;
 	unsigned i, match = 0, search_len = 0;
 
 	for (i = 0; i < len; i++) {
@@ -966,7 +995,8 @@ ftrace_match(unsigned char *buff, int len)
 
 	/* keep kstop machine from running */
 	preempt_disable();
-	ftrace_filtered = 1;
+	if (enable)
+		ftrace_filtered = 1;
 	pg = ftrace_pages_start;
 	while (pg) {
 		for (i = 0; i < pg->index; i++) {
@@ -997,7 +1027,7 @@ ftrace_match(unsigned char *buff, int len)
 				break;
 			}
 			if (matched)
-				rec->flags |= FTRACE_FL_FILTER;
+				rec->flags |= flag;
 		}
 		pg = pg->next;
 	}
@@ -1005,8 +1035,8 @@ ftrace_match(unsigned char *buff, int len)
 }
 
 static ssize_t
-ftrace_filter_write(struct file *file, const char __user *ubuf,
-		    size_t cnt, loff_t *ppos)
+ftrace_regex_write(struct file *file, const char __user *ubuf,
+		   size_t cnt, loff_t *ppos, int enable)
 {
 	struct ftrace_iterator *iter;
 	char ch;
@@ -1016,7 +1046,7 @@ ftrace_filter_write(struct file *file, const char __user *ubuf,
 	if (!cnt || cnt < 0)
 		return 0;
 
-	mutex_lock(&ftrace_filter_lock);
+	mutex_lock(&ftrace_regex_lock);
 
 	if (file->f_mode & FMODE_READ) {
 		struct seq_file *m = file->private_data;
@@ -1045,7 +1075,6 @@ ftrace_filter_write(struct file *file, const char __user *ubuf,
 			cnt--;
 		}
 
-
 		if (isspace(ch)) {
 			file->f_pos += read;
 			ret = read;
@@ -1072,7 +1101,7 @@ ftrace_filter_write(struct file *file, const char __user *ubuf,
 	if (isspace(ch)) {
 		iter->filtered++;
 		iter->buffer[iter->buffer_idx] = 0;
-		ftrace_match(iter->buffer, iter->buffer_idx);
+		ftrace_match(iter->buffer, iter->buffer_idx, enable);
 		iter->buffer_idx = 0;
 	} else
 		iter->flags |= FTRACE_ITER_CONT;
@@ -1082,11 +1111,39 @@ ftrace_filter_write(struct file *file, const char __user *ubuf,
 
 	ret = read;
  out:
-	mutex_unlock(&ftrace_filter_lock);
+	mutex_unlock(&ftrace_regex_lock);
 
 	return ret;
 }
 
+static ssize_t
+ftrace_filter_write(struct file *file, const char __user *ubuf,
+		    size_t cnt, loff_t *ppos)
+{
+	return ftrace_regex_write(file, ubuf, cnt, ppos, 1);
+}
+
+static ssize_t
+ftrace_notrace_write(struct file *file, const char __user *ubuf,
+		     size_t cnt, loff_t *ppos)
+{
+	return ftrace_regex_write(file, ubuf, cnt, ppos, 0);
+}
+
+static void
+ftrace_set_regex(unsigned char *buf, int len, int reset, int enable)
+{
+	if (unlikely(ftrace_disabled))
+		return;
+
+	mutex_lock(&ftrace_regex_lock);
+	if (reset)
+		ftrace_filter_reset(enable);
+	if (buf)
+		ftrace_match(buf, len, enable);
+	mutex_unlock(&ftrace_regex_lock);
+}
+
 /**
  * ftrace_set_filter - set a function to filter on in ftrace
  * @buf - the string that holds the function filter text.
@@ -1098,24 +1155,31 @@ ftrace_filter_write(struct file *file, const char __user *ubuf,
  */
 void ftrace_set_filter(unsigned char *buf, int len, int reset)
 {
-	if (unlikely(ftrace_disabled))
-		return;
+	ftrace_set_regex(buf, len, reset, 1);
+}
 
-	mutex_lock(&ftrace_filter_lock);
-	if (reset)
-		ftrace_filter_reset();
-	if (buf)
-		ftrace_match(buf, len);
-	mutex_unlock(&ftrace_filter_lock);
+/**
+ * ftrace_set_notrace - set a function to not trace in ftrace
+ * @buf - the string that holds the function notrace text.
+ * @len - the length of the string.
+ * @reset - non zero to reset all filters before applying this filter.
+ *
+ * Notrace Filters denote which functions should not be enabled when tracing
+ * is enabled. If @buf is NULL and reset is set, all functions will be enabled
+ * for tracing.
+ */
+void ftrace_set_notrace(unsigned char *buf, int len, int reset)
+{
+	ftrace_set_regex(buf, len, reset, 0);
 }
 
 static int
-ftrace_filter_release(struct inode *inode, struct file *file)
+ftrace_regex_release(struct inode *inode, struct file *file, int enable)
 {
 	struct seq_file *m = (struct seq_file *)file->private_data;
 	struct ftrace_iterator *iter;
 
-	mutex_lock(&ftrace_filter_lock);
+	mutex_lock(&ftrace_regex_lock);
 	if (file->f_mode & FMODE_READ) {
 		iter = m->private;
 
@@ -1126,7 +1190,7 @@ ftrace_filter_release(struct inode *inode, struct file *file)
 	if (iter->buffer_idx) {
 		iter->filtered++;
 		iter->buffer[iter->buffer_idx] = 0;
-		ftrace_match(iter->buffer, iter->buffer_idx);
+		ftrace_match(iter->buffer, iter->buffer_idx, enable);
 	}
 
 	mutex_lock(&ftrace_sysctl_lock);
@@ -1137,10 +1201,22 @@ ftrace_filter_release(struct inode *inode, struct file *file)
 	mutex_unlock(&ftrace_sysctl_lock);
 
 	kfree(iter);
-	mutex_unlock(&ftrace_filter_lock);
+	mutex_unlock(&ftrace_regex_lock);
 	return 0;
 }
 
+static int
+ftrace_filter_release(struct inode *inode, struct file *file)
+{
+	return ftrace_regex_release(inode, file, 1);
+}
+
+static int
+ftrace_notrace_release(struct inode *inode, struct file *file)
+{
+	return ftrace_regex_release(inode, file, 0);
+}
+
 static struct file_operations ftrace_avail_fops = {
 	.open = ftrace_avail_open,
 	.read = seq_read,
@@ -1150,12 +1226,20 @@ static struct file_operations ftrace_avail_fops = {
 
 static struct file_operations ftrace_filter_fops = {
 	.open = ftrace_filter_open,
-	.read = ftrace_filter_read,
+	.read = ftrace_regex_read,
 	.write = ftrace_filter_write,
-	.llseek = ftrace_filter_lseek,
+	.llseek = ftrace_regex_lseek,
 	.release = ftrace_filter_release,
 };
 
+static struct file_operations ftrace_notrace_fops = {
+	.open = ftrace_notrace_open,
+	.read = ftrace_regex_read,
+	.write = ftrace_notrace_write,
+	.llseek = ftrace_regex_lseek,
+	.release = ftrace_notrace_release,
+};
+
 /**
  * ftrace_force_update - force an update to all recording ftrace functions
  *
@@ -1239,6 +1323,12 @@ static __init int ftrace_init_debugfs(void)
 	if (!entry)
 		pr_warning("Could not create debugfs "
 			   "'set_ftrace_filter' entry\n");
+
+	entry = debugfs_create_file("set_ftrace_notrace", 0644, d_tracer,
+				    NULL, &ftrace_notrace_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs "
+			   "'set_ftrace_notrace' entry\n");
 	return 0;
 }
 
-- 
cgit v1.2.3


From 9e124fe16ff24746d6de5a2ad685266d7bce0e08 Mon Sep 17 00:00:00 2001
From: Markus Armbruster <armbru@redhat.com>
Date: Mon, 26 May 2008 23:31:07 +0100
Subject: xen: Enable console tty by default in domU if it's not a dummy

Without console= arguments on the kernel command line, the first
console to register becomes enabled and the preferred console (the one
behind /dev/console).  This is normally tty (assuming
CONFIG_VT_CONSOLE is enabled, which it commonly is).

This is okay as long tty is a useful console.  But unless we have the
PV framebuffer, and it is enabled for this domain, tty0 in domU is
merely a dummy.  In that case, we want the preferred console to be the
Xen console hvc0, and we want it without having to fiddle with the
kernel command line.  Commit b8c2d3dfbc117dff26058fbac316b8acfc2cb5f7
did that for us.

Since we now have the PV framebuffer, we want to enable and prefer tty
again, but only when PVFB is enabled.  But even then we still want to
enable the Xen console as well.

Problem: when tty registers, we can't yet know whether the PVFB is
enabled.  By the time we can know (xenstore is up), the console setup
game is over.

Solution: enable console tty by default, but keep hvc as the preferred
console.  Change the preferred console to tty when PVFB probes
successfully, unless we've been given console kernel parameters.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/xen/enlighten.c    |  4 +++-
 drivers/video/xen-fbfront.c | 25 +++++++++++++++++++++++++
 include/linux/console.h     |  2 ++
 kernel/printk.c             |  3 +++
 4 files changed, 33 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 1b4b5fa498b3..6cfb708408e9 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1256,8 +1256,10 @@ asmlinkage void __init xen_start_kernel(void)
 		? __pa(xen_start_info->mod_start) : 0;
 	boot_params.hdr.ramdisk_size = xen_start_info->mod_len;
 
-	if (!is_initial_xendomain())
+	if (!is_initial_xendomain()) {
+		add_preferred_console("tty", 0, NULL);
 		add_preferred_console("hvc", 0, NULL);
+	}
 
 	/* Start the world */
 	start_kernel();
diff --git a/drivers/video/xen-fbfront.c b/drivers/video/xen-fbfront.c
index 619a6f8d65a2..4e10876e62fc 100644
--- a/drivers/video/xen-fbfront.c
+++ b/drivers/video/xen-fbfront.c
@@ -18,6 +18,7 @@
  * frame buffer.
  */
 
+#include <linux/console.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
 #include <linux/fb.h>
@@ -48,6 +49,7 @@ struct xenfb_info {
 
 static u32 xenfb_mem_len = XENFB_WIDTH * XENFB_HEIGHT * XENFB_DEPTH / 8;
 
+static void xenfb_make_preferred_console(void);
 static int xenfb_remove(struct xenbus_device *);
 static void xenfb_init_shared_page(struct xenfb_info *);
 static int xenfb_connect_backend(struct xenbus_device *, struct xenfb_info *);
@@ -348,6 +350,7 @@ static int __devinit xenfb_probe(struct xenbus_device *dev,
 	if (ret < 0)
 		goto error;
 
+	xenfb_make_preferred_console();
 	return 0;
 
  error_nomem:
@@ -358,6 +361,28 @@ static int __devinit xenfb_probe(struct xenbus_device *dev,
 	return ret;
 }
 
+static __devinit void
+xenfb_make_preferred_console(void)
+{
+	struct console *c;
+
+	if (console_set_on_cmdline)
+		return;
+
+	acquire_console_sem();
+	for (c = console_drivers; c; c = c->next) {
+		if (!strcmp(c->name, "tty") && c->index == 0)
+			break;
+	}
+	release_console_sem();
+	if (c) {
+		unregister_console(c);
+		c->flags |= CON_CONSDEV;
+		c->flags &= ~CON_PRINTBUFFER; /* don't print again */
+		register_console(c);
+	}
+}
+
 static int xenfb_resume(struct xenbus_device *dev)
 {
 	struct xenfb_info *info = dev->dev.driver_data;
diff --git a/include/linux/console.h b/include/linux/console.h
index a4f27fbdf549..248e6e3b9b73 100644
--- a/include/linux/console.h
+++ b/include/linux/console.h
@@ -108,6 +108,8 @@ struct console {
 	struct	 console *next;
 };
 
+extern int console_set_on_cmdline;
+
 extern int add_preferred_console(char *name, int idx, char *options);
 extern int update_console_cmdline(char *name, int idx, char *name_new, int idx_new, char *options);
 extern void register_console(struct console *);
diff --git a/kernel/printk.c b/kernel/printk.c
index 8fb01c32aa3b..028ed75d4864 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -121,6 +121,8 @@ struct console_cmdline
 static struct console_cmdline console_cmdline[MAX_CMDLINECONSOLES];
 static int selected_console = -1;
 static int preferred_console = -1;
+int console_set_on_cmdline;
+EXPORT_SYMBOL(console_set_on_cmdline);
 
 /* Flag: console code may call schedule() */
 static int console_may_schedule;
@@ -890,6 +892,7 @@ static int __init console_setup(char *str)
 	*s = 0;
 
 	__add_preferred_console(buf, idx, options, brl_options);
+	console_set_on_cmdline = 1;
 	return 1;
 }
 __setup("console=", console_setup);
-- 
cgit v1.2.3


From 0e91398f2a5d4eb6b07df8115917d0d1cf3e9b58 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Mon, 26 May 2008 23:31:27 +0100
Subject: xen: implement save/restore

This patch implements Xen save/restore and migration.

Saving is triggered via xenbus, which is polled in
drivers/xen/manage.c.  When a suspend request comes in, the kernel
prepares itself for saving by:

1 - Freeze all processes.  This is primarily to prevent any
    partially-completed pagetable updates from confusing the suspend
    process.  If CONFIG_PREEMPT isn't defined, then this isn't necessary.

2 - Suspend xenbus and other devices

3 - Stop_machine, to make sure all the other vcpus are quiescent.  The
    Xen tools require the domain to run its save off vcpu0.

4 - Within the stop_machine state, it pins any unpinned pgds (under
    construction or destruction), performs canonicalizes various other
    pieces of state (mostly converting mfns to pfns), and finally

5 - Suspend the domain

Restore reverses the steps used to save the domain, ending when all
the frozen processes are thawed.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/xen/Makefile      |   2 +-
 arch/x86/xen/enlighten.c   |   6 +--
 arch/x86/xen/mmu.c         |  46 +++++++++++++++++
 arch/x86/xen/smp.c         |   2 +-
 arch/x86/xen/suspend.c     |  42 +++++++++++++++
 arch/x86/xen/time.c        |   8 +++
 arch/x86/xen/xen-ops.h     |   4 ++
 drivers/xen/events.c       |  83 +++++++++++++++++++++++++++++
 drivers/xen/grant-table.c  |   4 +-
 drivers/xen/manage.c       | 126 ++++++++++++++++++++++++++++++++++++++++-----
 include/linux/page-flags.h |   1 +
 include/xen/events.h       |   3 ++
 include/xen/grant_table.h  |   3 ++
 include/xen/xen-ops.h      |   9 ++++
 14 files changed, 318 insertions(+), 21 deletions(-)
 create mode 100644 arch/x86/xen/suspend.c

(limited to 'include/linux')

diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile
index 40b119b6b103..2ba2d1649131 100644
--- a/arch/x86/xen/Makefile
+++ b/arch/x86/xen/Makefile
@@ -1,4 +1,4 @@
 obj-y		:= enlighten.o setup.o multicalls.o mmu.o \
-			time.o xen-asm.o grant-table.o
+			time.o xen-asm.o grant-table.o suspend.o
 
 obj-$(CONFIG_SMP)	+= smp.o
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index ce67dc8f36af..b94f63ac228b 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -857,7 +857,7 @@ static __init void xen_pagetable_setup_start(pgd_t *base)
 			  PFN_DOWN(__pa(xen_start_info->pt_base)));
 }
 
-static __init void setup_shared_info(void)
+void xen_setup_shared_info(void)
 {
 	if (!xen_feature(XENFEAT_auto_translated_physmap)) {
 		unsigned long addr = fix_to_virt(FIX_PARAVIRT_BOOTMAP);
@@ -894,7 +894,7 @@ static __init void xen_pagetable_setup_done(pgd_t *base)
 	pv_mmu_ops.release_pmd = xen_release_pmd;
 	pv_mmu_ops.set_pte = xen_set_pte;
 
-	setup_shared_info();
+	xen_setup_shared_info();
 
 	/* Actually pin the pagetable down, but we can't set PG_pinned
 	   yet because the page structures don't exist yet. */
@@ -902,7 +902,7 @@ static __init void xen_pagetable_setup_done(pgd_t *base)
 }
 
 /* This is called once we have the cpu_possible_map */
-void __init xen_setup_vcpu_info_placement(void)
+void xen_setup_vcpu_info_placement(void)
 {
 	int cpu;
 
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 4740cda36563..e95955968ba3 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -560,6 +560,29 @@ void xen_pgd_pin(pgd_t *pgd)
 	xen_mc_issue(0);
 }
 
+/*
+ * On save, we need to pin all pagetables to make sure they get their
+ * mfns turned into pfns.  Search the list for any unpinned pgds and pin
+ * them (unpinned pgds are not currently in use, probably because the
+ * process is under construction or destruction).
+ */
+void xen_mm_pin_all(void)
+{
+	unsigned long flags;
+	struct page *page;
+
+	spin_lock_irqsave(&pgd_lock, flags);
+
+	list_for_each_entry(page, &pgd_list, lru) {
+		if (!PagePinned(page)) {
+			xen_pgd_pin((pgd_t *)page_address(page));
+			SetPageSavePinned(page);
+		}
+	}
+
+	spin_unlock_irqrestore(&pgd_lock, flags);
+}
+
 /* The init_mm pagetable is really pinned as soon as its created, but
    that's before we have page structures to store the bits.  So do all
    the book-keeping now. */
@@ -617,6 +640,29 @@ static void xen_pgd_unpin(pgd_t *pgd)
 	xen_mc_issue(0);
 }
 
+/*
+ * On resume, undo any pinning done at save, so that the rest of the
+ * kernel doesn't see any unexpected pinned pagetables.
+ */
+void xen_mm_unpin_all(void)
+{
+	unsigned long flags;
+	struct page *page;
+
+	spin_lock_irqsave(&pgd_lock, flags);
+
+	list_for_each_entry(page, &pgd_list, lru) {
+		if (PageSavePinned(page)) {
+			BUG_ON(!PagePinned(page));
+			printk("unpinning pinned %p\n", page_address(page));
+			xen_pgd_unpin((pgd_t *)page_address(page));
+			ClearPageSavePinned(page);
+		}
+	}
+
+	spin_unlock_irqrestore(&pgd_lock, flags);
+}
+
 void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next)
 {
 	spin_lock(&next->page_table_lock);
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 74ab8968c525..d2e3c20127d7 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -35,7 +35,7 @@
 #include "xen-ops.h"
 #include "mmu.h"
 
-static cpumask_t xen_cpu_initialized_map;
+cpumask_t xen_cpu_initialized_map;
 static DEFINE_PER_CPU(int, resched_irq) = -1;
 static DEFINE_PER_CPU(int, callfunc_irq) = -1;
 static DEFINE_PER_CPU(int, debug_irq) = -1;
diff --git a/arch/x86/xen/suspend.c b/arch/x86/xen/suspend.c
new file mode 100644
index 000000000000..7620a16fe535
--- /dev/null
+++ b/arch/x86/xen/suspend.c
@@ -0,0 +1,42 @@
+#include <linux/types.h>
+
+#include <xen/interface/xen.h>
+#include <xen/grant_table.h>
+#include <xen/events.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+#include "xen-ops.h"
+#include "mmu.h"
+
+void xen_pre_suspend(void)
+{
+	xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn);
+	xen_start_info->console.domU.mfn =
+		mfn_to_pfn(xen_start_info->console.domU.mfn);
+
+	BUG_ON(!irqs_disabled());
+
+	HYPERVISOR_shared_info = &xen_dummy_shared_info;
+	if (HYPERVISOR_update_va_mapping(fix_to_virt(FIX_PARAVIRT_BOOTMAP),
+					 __pte_ma(0), 0))
+		BUG();
+}
+
+void xen_post_suspend(int suspend_cancelled)
+{
+	if (suspend_cancelled) {
+		xen_start_info->store_mfn =
+			pfn_to_mfn(xen_start_info->store_mfn);
+		xen_start_info->console.domU.mfn =
+			pfn_to_mfn(xen_start_info->console.domU.mfn);
+	} else {
+#ifdef CONFIG_SMP
+		xen_cpu_initialized_map = cpu_online_map;
+#endif
+	}
+
+	xen_setup_shared_info();
+}
+
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index c39e1a5aa241..0bef256e5f2d 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -572,6 +572,14 @@ void xen_setup_cpu_clockevents(void)
 	clockevents_register_device(&__get_cpu_var(xen_clock_events));
 }
 
+void xen_time_suspend(void)
+{
+}
+
+void xen_time_resume(void)
+{
+}
+
 __init void xen_time_init(void)
 {
 	int cpu = smp_processor_id();
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index a1bc89a8f169..a0503acad664 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -9,6 +9,7 @@
 extern const char xen_hypervisor_callback[];
 extern const char xen_failsafe_callback[];
 
+struct trap_info;
 void xen_copy_trap_info(struct trap_info *traps);
 
 DECLARE_PER_CPU(unsigned long, xen_cr3);
@@ -19,6 +20,7 @@ extern struct shared_info xen_dummy_shared_info;
 extern struct shared_info *HYPERVISOR_shared_info;
 
 void xen_setup_mfn_list_list(void);
+void xen_setup_shared_info(void);
 
 char * __init xen_memory_setup(void);
 void __init xen_arch_setup(void);
@@ -59,6 +61,8 @@ int xen_smp_call_function_single(int cpu, void (*func) (void *info), void *info,
 int xen_smp_call_function_mask(cpumask_t mask, void (*func)(void *),
 			       void *info, int wait);
 
+extern cpumask_t xen_cpu_initialized_map;
+
 
 /* Declare an asm function, along with symbols needed to make it
    inlineable */
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 70375a690761..73d78dc9b875 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -674,6 +674,89 @@ static int retrigger_dynirq(unsigned int irq)
 	return ret;
 }
 
+static void restore_cpu_virqs(unsigned int cpu)
+{
+	struct evtchn_bind_virq bind_virq;
+	int virq, irq, evtchn;
+
+	for (virq = 0; virq < NR_VIRQS; virq++) {
+		if ((irq = per_cpu(virq_to_irq, cpu)[virq]) == -1)
+			continue;
+
+		BUG_ON(irq_info[irq].type != IRQT_VIRQ);
+		BUG_ON(irq_info[irq].index != virq);
+
+		/* Get a new binding from Xen. */
+		bind_virq.virq = virq;
+		bind_virq.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
+						&bind_virq) != 0)
+			BUG();
+		evtchn = bind_virq.port;
+
+		/* Record the new mapping. */
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_irq_info(IRQT_VIRQ, virq, evtchn);
+		bind_evtchn_to_cpu(evtchn, cpu);
+
+		/* Ready for use. */
+		unmask_evtchn(evtchn);
+	}
+}
+
+static void restore_cpu_ipis(unsigned int cpu)
+{
+	struct evtchn_bind_ipi bind_ipi;
+	int ipi, irq, evtchn;
+
+	for (ipi = 0; ipi < XEN_NR_IPIS; ipi++) {
+		if ((irq = per_cpu(ipi_to_irq, cpu)[ipi]) == -1)
+			continue;
+
+		BUG_ON(irq_info[irq].type != IRQT_IPI);
+		BUG_ON(irq_info[irq].index != ipi);
+
+		/* Get a new binding from Xen. */
+		bind_ipi.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_ipi,
+						&bind_ipi) != 0)
+			BUG();
+		evtchn = bind_ipi.port;
+
+		/* Record the new mapping. */
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_irq_info(IRQT_IPI, ipi, evtchn);
+		bind_evtchn_to_cpu(evtchn, cpu);
+
+		/* Ready for use. */
+		unmask_evtchn(evtchn);
+
+	}
+}
+
+void xen_irq_resume(void)
+{
+	unsigned int cpu, irq, evtchn;
+
+	init_evtchn_cpu_bindings();
+
+	/* New event-channel space is not 'live' yet. */
+	for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
+		mask_evtchn(evtchn);
+
+	/* No IRQ <-> event-channel mappings. */
+	for (irq = 0; irq < NR_IRQS; irq++)
+		irq_info[irq].evtchn = 0; /* zap event-channel binding */
+
+	for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
+		evtchn_to_irq[evtchn] = -1;
+
+	for_each_possible_cpu(cpu) {
+		restore_cpu_virqs(cpu);
+		restore_cpu_ipis(cpu);
+	}
+}
+
 static struct irq_chip xen_dynamic_chip __read_mostly = {
 	.name		= "xen-dyn",
 	.mask		= disable_dynirq,
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 52b6b41b909d..e9e11168616a 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -471,14 +471,14 @@ static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
 	return 0;
 }
 
-static int gnttab_resume(void)
+int gnttab_resume(void)
 {
 	if (max_nr_grant_frames() < nr_grant_frames)
 		return -ENOSYS;
 	return gnttab_map(0, nr_grant_frames - 1);
 }
 
-static int gnttab_suspend(void)
+int gnttab_suspend(void)
 {
 	arch_gnttab_unmap_shared(shared, nr_grant_frames);
 	return 0;
diff --git a/drivers/xen/manage.c b/drivers/xen/manage.c
index aa7af9e6abc0..ba85fa2cff33 100644
--- a/drivers/xen/manage.c
+++ b/drivers/xen/manage.c
@@ -5,21 +5,113 @@
 #include <linux/err.h>
 #include <linux/reboot.h>
 #include <linux/sysrq.h>
+#include <linux/stop_machine.h>
+#include <linux/freezer.h>
 
 #include <xen/xenbus.h>
-
-#define SHUTDOWN_INVALID  -1
-#define SHUTDOWN_POWEROFF  0
-#define SHUTDOWN_SUSPEND   2
-/* Code 3 is SHUTDOWN_CRASH, which we don't use because the domain can only
- * report a crash, not be instructed to crash!
- * HALT is the same as POWEROFF, as far as we're concerned.  The tools use
- * the distinction when we return the reason code to them.
- */
-#define SHUTDOWN_HALT      4
+#include <xen/grant_table.h>
+#include <xen/events.h>
+#include <xen/hvc-console.h>
+#include <xen/xen-ops.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+enum shutdown_state {
+	SHUTDOWN_INVALID = -1,
+	SHUTDOWN_POWEROFF = 0,
+	SHUTDOWN_SUSPEND = 2,
+	/* Code 3 is SHUTDOWN_CRASH, which we don't use because the domain can only
+	   report a crash, not be instructed to crash!
+	   HALT is the same as POWEROFF, as far as we're concerned.  The tools use
+	   the distinction when we return the reason code to them.  */
+	 SHUTDOWN_HALT = 4,
+};
 
 /* Ignore multiple shutdown requests. */
-static int shutting_down = SHUTDOWN_INVALID;
+static enum shutdown_state shutting_down = SHUTDOWN_INVALID;
+
+static int xen_suspend(void *data)
+{
+	int *cancelled = data;
+
+	BUG_ON(!irqs_disabled());
+
+	load_cr3(swapper_pg_dir);
+
+	xen_mm_pin_all();
+	gnttab_suspend();
+	xen_time_suspend();
+	xen_pre_suspend();
+
+	/*
+	 * This hypercall returns 1 if suspend was cancelled
+	 * or the domain was merely checkpointed, and 0 if it
+	 * is resuming in a new domain.
+	 */
+	*cancelled = HYPERVISOR_suspend(virt_to_mfn(xen_start_info));
+
+	xen_post_suspend(*cancelled);
+	xen_time_resume();
+	gnttab_resume();
+	xen_mm_unpin_all();
+
+	if (!*cancelled) {
+		xen_irq_resume();
+		xen_console_resume();
+	}
+
+	return 0;
+}
+
+static void do_suspend(void)
+{
+	int err;
+	int cancelled = 1;
+
+	shutting_down = SHUTDOWN_SUSPEND;
+
+#ifdef CONFIG_PREEMPT
+	/* If the kernel is preemptible, we need to freeze all the processes
+	   to prevent them from being in the middle of a pagetable update
+	   during suspend. */
+	err = freeze_processes();
+	if (err) {
+		printk(KERN_ERR "xen suspend: freeze failed %d\n", err);
+		return;
+	}
+#endif
+
+	err = device_suspend(PMSG_SUSPEND);
+	if (err) {
+		printk(KERN_ERR "xen suspend: device_suspend %d\n", err);
+		goto out;
+	}
+
+	printk("suspending xenbus...\n");
+	/* XXX use normal device tree? */
+	xenbus_suspend();
+
+	err = stop_machine_run(xen_suspend, &cancelled, 0);
+	if (err) {
+		printk(KERN_ERR "failed to start xen_suspend: %d\n", err);
+		goto out;
+	}
+
+	if (!cancelled)
+		xenbus_resume();
+	else
+		xenbus_suspend_cancel();
+
+	device_resume();
+
+
+out:
+#ifdef CONFIG_PREEMPT
+	thaw_processes();
+#endif
+	shutting_down = SHUTDOWN_INVALID;
+}
 
 static void shutdown_handler(struct xenbus_watch *watch,
 			     const char **vec, unsigned int len)
@@ -52,11 +144,17 @@ static void shutdown_handler(struct xenbus_watch *watch,
 	}
 
 	if (strcmp(str, "poweroff") == 0 ||
-	    strcmp(str, "halt") == 0)
+	    strcmp(str, "halt") == 0) {
+		shutting_down = SHUTDOWN_POWEROFF;
 		orderly_poweroff(false);
-	else if (strcmp(str, "reboot") == 0)
+	} else if (strcmp(str, "reboot") == 0) {
+		shutting_down = SHUTDOWN_POWEROFF; /* ? */
 		ctrl_alt_del();
-	else {
+#ifdef CONFIG_PM_SLEEP
+	} else if (strcmp(str, "suspend") == 0) {
+		do_suspend();
+#endif
+	} else {
 		printk(KERN_INFO "Ignoring shutdown request: %s\n", str);
 		shutting_down = SHUTDOWN_INVALID;
 	}
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 590cff32415d..02955a1c3d76 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -157,6 +157,7 @@ PAGEFLAG(Active, active) __CLEARPAGEFLAG(Active, active)
 __PAGEFLAG(Slab, slab)
 PAGEFLAG(Checked, owner_priv_1)		/* Used by some filesystems */
 PAGEFLAG(Pinned, owner_priv_1) TESTSCFLAG(Pinned, owner_priv_1) /* Xen */
+PAGEFLAG(SavePinned, dirty);					/* Xen */
 PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved)
 PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private)
 	__SETPAGEFLAG(Private, private)
diff --git a/include/xen/events.h b/include/xen/events.h
index a82ec0c45c38..67c4436554a9 100644
--- a/include/xen/events.h
+++ b/include/xen/events.h
@@ -41,4 +41,7 @@ static inline void notify_remote_via_evtchn(int port)
 }
 
 extern void notify_remote_via_irq(int irq);
+
+extern void xen_irq_resume(void);
+
 #endif	/* _XEN_EVENTS_H */
diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h
index 466204846121..a40f1cd91be1 100644
--- a/include/xen/grant_table.h
+++ b/include/xen/grant_table.h
@@ -51,6 +51,9 @@ struct gnttab_free_callback {
 	u16 count;
 };
 
+int gnttab_suspend(void);
+int gnttab_resume(void);
+
 int gnttab_grant_foreign_access(domid_t domid, unsigned long frame,
 				int readonly);
 
diff --git a/include/xen/xen-ops.h b/include/xen/xen-ops.h
index 10ddfe0142d0..5d7a6db54a8c 100644
--- a/include/xen/xen-ops.h
+++ b/include/xen/xen-ops.h
@@ -5,4 +5,13 @@
 
 DECLARE_PER_CPU(struct vcpu_info *, xen_vcpu);
 
+void xen_pre_suspend(void);
+void xen_post_suspend(int suspend_cancelled);
+
+void xen_mm_pin_all(void);
+void xen_mm_unpin_all(void);
+
+void xen_time_suspend(void);
+void xen_time_resume(void);
+
 #endif /* INCLUDE_XEN_OPS_H */
-- 
cgit v1.2.3


From b1829d2705daa7cb72eb1e08bdc8b7e9fad34266 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 28 May 2008 01:22:08 +0200
Subject: ftrace: fix merge

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/ftrace.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index ffbbd54a720e..b482fe88bc04 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -122,7 +122,7 @@ static inline void tracer_disable(void)
 # define trace_preempt_off(a0, a1)		do { } while (0)
 #endif
 
-#ifdef CONFIG_CONTEXT_SWITCH_TRACER
+#ifdef CONFIG_TRACING
 extern void
 ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3);
 #else
-- 
cgit v1.2.3


From 57e6983cbde91b4569b4014b933f3a16e12b99fd Mon Sep 17 00:00:00 2001
From: Matt Carlson <mcarlson@broadcom.com>
Date: Sun, 25 May 2008 23:48:31 -0700
Subject: tg3: Add 5785 ASIC revision

This patch added the 5785 device ID and ASIC revision to the code.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tg3.c       | 58 ++++++++++++++++++++++++++++++++++++-------------
 drivers/net/tg3.h       |  1 +
 include/linux/pci_ids.h |  1 +
 3 files changed, 45 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index 028276edd3bc..69ded90cd96e 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -204,6 +204,7 @@ static struct pci_device_id tg3_pci_tbl[] = {
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5723)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5761)},
 	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5761E)},
+	{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5785)},
 	{PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, PCI_DEVICE_ID_SYSKONNECT_9DXX)},
 	{PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, PCI_DEVICE_ID_SYSKONNECT_9MXX)},
 	{PCI_DEVICE(PCI_VENDOR_ID_ALTIMA, PCI_DEVICE_ID_ALTIMA_AC1000)},
@@ -5710,7 +5711,8 @@ static int tg3_chip_reset(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		tw32(GRC_FASTBOOT_PC, 0);
 
 	/*
@@ -7004,7 +7006,8 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 		return err;
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5784 &&
-	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5761 &&
+	    GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5785) {
 		/* This value is determined during the probe time DMA
 		 * engine test, tg3_test_dma.
 		 */
@@ -7243,7 +7246,8 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 		      RDMAC_MODE_FIFOURUN_ENAB | RDMAC_MODE_FIFOOREAD_ENAB |
 		      RDMAC_MODE_LNGREAD_ENAB);
 
-	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784)
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		rdmac_mode |= RDMAC_MODE_BD_SBD_CRPT_ENAB |
 			      RDMAC_MODE_MBUF_RBD_CRPT_ENAB |
 			      RDMAC_MODE_MBUF_SBD_CRPT_ENAB;
@@ -7411,7 +7415,8 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 	if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755) ||
 	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787) ||
 	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784) ||
-	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761))
+	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) ||
+	    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785))
 		val |= WDMAC_MODE_STATUS_TAG_FIX;
 
 	tw32_f(WDMAC_MODE, val);
@@ -7473,7 +7478,9 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy)
 
 	tp->rx_mode = RX_MODE_ENABLE;
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		tp->rx_mode |= RX_MODE_IPV6_CSUM_ENABLE;
 
 	tw32_f(MAC_RX_MODE, tp->rx_mode);
@@ -9028,7 +9035,10 @@ static int tg3_set_tso(struct net_device *dev, u32 value)
 	    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906)) {
 		if (value) {
 			dev->features |= NETIF_F_TSO6;
-			if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+			if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+			    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 &&
+			     GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5784_AX) ||
+			    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 				dev->features |= NETIF_F_TSO_ECN;
 		} else
 			dev->features &= ~(NETIF_F_TSO6 | NETIF_F_TSO_ECN);
@@ -9286,7 +9296,8 @@ static int tg3_set_tx_csum(struct net_device *dev, u32 data)
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 		ethtool_op_set_tx_ipv6_csum(dev, data);
 	else
 		ethtool_op_set_tx_csum(dev, data);
@@ -9807,7 +9818,8 @@ static int tg3_test_memory(struct tg3 *tp)
 		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			mem_tbl = mem_tbl_5755;
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
 			mem_tbl = mem_tbl_5906;
@@ -10014,7 +10026,8 @@ static int tg3_test_loopback(struct tg3 *tp)
 		return TG3_LOOPBACK_FAILED;
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785) {
 		int i;
 		u32 status;
 
@@ -10042,7 +10055,8 @@ static int tg3_test_loopback(struct tg3 *tp)
 		err |= TG3_MAC_LOOPBACK_FAILED;
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785) {
 		tw32(TG3_CPMU_CTRL, cpmuctrl);
 
 		/* Release the mutex */
@@ -10690,7 +10704,8 @@ static void __devinit tg3_nvram_init(struct tg3 *tp)
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755)
 			tg3_get_5755_nvram_info(tp);
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
-			 GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784)
+			 GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
+			 GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			tg3_get_5787_nvram_info(tp);
 		else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
 			tg3_get_5761_nvram_info(tp);
@@ -11021,6 +11036,7 @@ static int tg3_nvram_write_block_buffered(struct tg3 *tp, u32 offset, u32 len,
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5787) &&
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5784) &&
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5761) &&
+		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5785) &&
 		    (tp->nvram_jedecnum == JEDEC_ST) &&
 		    (nvram_cmd & NVRAM_CMD_FIRST)) {
 
@@ -11924,6 +11940,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906 ||
 	    (tp->tg3_flags2 & TG3_FLG2_5780_CLASS))
 		tp->tg3_flags2 |= TG3_FLG2_5750_PLUS;
@@ -11945,6 +11962,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) {
 			tp->tg3_flags2 |= TG3_FLG2_HW_TSO_2;
 			tp->tg3_flags2 |= TG3_FLG2_1SHOT_MSI;
@@ -12147,7 +12165,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	}
 
 	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761) {
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785) {
 		tp->tg3_flags |= TG3_FLAG_CPMU_PRESENT;
 
 		if (tp->pci_chip_rev_id == CHIPREV_ID_5784_A0 ||
@@ -12231,7 +12250,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 				tp->tg3_flags2 |= TG3_FLG2_PHY_JITTER_BUG;
 			if (tp->pdev->device == PCI_DEVICE_ID_TIGON3_5755M)
 				tp->tg3_flags2 |= TG3_FLG2_PHY_ADJUST_TRIM;
-		} else if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906)
+		} else if (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906 &&
+			   GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5785)
 			tp->tg3_flags2 |= TG3_FLG2_PHY_BER_BUG;
 	}
 
@@ -12252,6 +12272,9 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5700_BX)
 		tp->coalesce_mode |= HOSTCC_MODE_32BYTE;
 
+	if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
+		tp->tg3_flags3 |= TG3_FLG3_USE_PHYLIB;
+
 	err = tg3_mdio_init(tp);
 	if (err)
 		return err;
@@ -12383,6 +12406,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp)
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785 ||
 	    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906)
 		tp->dev->hard_start_xmit = tg3_start_xmit;
 	else
@@ -13307,7 +13331,10 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 		if ((tp->tg3_flags2 & TG3_FLG2_HW_TSO_2) &&
 		    (GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906))
 			dev->features |= NETIF_F_TSO6;
-		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 &&
+		     GET_CHIP_REV(tp->pci_chip_rev_id) != CHIPREV_5784_AX) ||
+			GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			dev->features |= NETIF_F_TSO_ECN;
 	}
 
@@ -13373,7 +13400,8 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5755 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5787 ||
 		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5784 ||
-		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761)
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5761 ||
+		    GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
 			dev->features |= NETIF_F_IPV6_CSUM;
 
 		tp->tg3_flags |= TG3_FLAG_RX_CHECKSUMS;
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index 48f45c17f60d..95ac4c9590a1 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -128,6 +128,7 @@
 #define   ASIC_REV_USE_PROD_ID_REG	 0x0f
 #define   ASIC_REV_5784			 0x5784
 #define   ASIC_REV_5761			 0x5761
+#define   ASIC_REV_5785			 0x5785
 #define  GET_CHIP_REV(CHIP_REV_ID)	((CHIP_REV_ID) >> 8)
 #define   CHIPREV_5700_AX		 0x70
 #define   CHIPREV_5700_BX		 0x71
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index cf6dbd759395..72c038560e7d 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1979,6 +1979,7 @@
 #define PCI_DEVICE_ID_TIGON3_5787M	0x1693
 #define PCI_DEVICE_ID_TIGON3_5782	0x1696
 #define PCI_DEVICE_ID_TIGON3_5784	0x1698
+#define PCI_DEVICE_ID_TIGON3_5785	0x1699
 #define PCI_DEVICE_ID_TIGON3_5786	0x169a
 #define PCI_DEVICE_ID_TIGON3_5787	0x169b
 #define PCI_DEVICE_ID_TIGON3_5788	0x169c
-- 
cgit v1.2.3


From a9daf36746b1fb5c2db8d164ca70c30c63a0d7b2 Mon Sep 17 00:00:00 2001
From: Matt Carlson <mcarlson@broadcom.com>
Date: Sun, 25 May 2008 23:49:44 -0700
Subject: tg3: Add shmem options.

This patch adds some options obtained through shared memory.

Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tg3.c       | 95 +++++++++++++++++++++++++++++++++++++++++++++----
 drivers/net/tg3.h       | 30 +++++++++++++++-
 include/linux/brcmphy.h |  6 ++++
 3 files changed, 123 insertions(+), 8 deletions(-)
 create mode 100644 include/linux/brcmphy.h

(limited to 'include/linux')

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index 69ded90cd96e..365fcffc3144 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -33,6 +33,7 @@
 #include <linux/ethtool.h>
 #include <linux/mii.h>
 #include <linux/phy.h>
+#include <linux/brcmphy.h>
 #include <linux/if_vlan.h>
 #include <linux/ip.h>
 #include <linux/tcp.h>
@@ -869,6 +870,51 @@ static int tg3_mdio_reset(struct mii_bus *bp)
 	return 0;
 }
 
+static void tg3_mdio_config(struct tg3 *tp)
+{
+	u32 val;
+
+	if (tp->mdio_bus.phy_map[PHY_ADDR]->interface !=
+	    PHY_INTERFACE_MODE_RGMII)
+		return;
+
+	val = tr32(MAC_PHYCFG1) & ~(MAC_PHYCFG1_RGMII_EXT_RX_DEC |
+				    MAC_PHYCFG1_RGMII_SND_STAT_EN);
+	if (tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE) {
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_RX_EN)
+			val |= MAC_PHYCFG1_RGMII_EXT_RX_DEC;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_TX_EN)
+			val |= MAC_PHYCFG1_RGMII_SND_STAT_EN;
+	}
+	tw32(MAC_PHYCFG1, val | MAC_PHYCFG1_RGMII_INT | MAC_PHYCFG1_TXC_DRV);
+
+	val = tr32(MAC_PHYCFG2) & ~(MAC_PHYCFG2_INBAND_ENABLE);
+	if (!(tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE))
+		val |= MAC_PHYCFG2_INBAND_ENABLE;
+	tw32(MAC_PHYCFG2, val);
+
+	val = tr32(MAC_EXT_RGMII_MODE);
+	val &= ~(MAC_RGMII_MODE_RX_INT_B |
+		 MAC_RGMII_MODE_RX_QUALITY |
+		 MAC_RGMII_MODE_RX_ACTIVITY |
+		 MAC_RGMII_MODE_RX_ENG_DET |
+		 MAC_RGMII_MODE_TX_ENABLE |
+		 MAC_RGMII_MODE_TX_LOWPWR |
+		 MAC_RGMII_MODE_TX_RESET);
+	if (tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE) {
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_RX_EN)
+			val |= MAC_RGMII_MODE_RX_INT_B |
+			       MAC_RGMII_MODE_RX_QUALITY |
+			       MAC_RGMII_MODE_RX_ACTIVITY |
+			       MAC_RGMII_MODE_RX_ENG_DET;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_TX_EN)
+			val |= MAC_RGMII_MODE_TX_ENABLE |
+			       MAC_RGMII_MODE_TX_LOWPWR |
+			       MAC_RGMII_MODE_TX_RESET;
+	}
+	tw32(MAC_EXT_RGMII_MODE, val);
+}
+
 static void tg3_mdio_start(struct tg3 *tp)
 {
 	if (tp->tg3_flags3 & TG3_FLG3_MDIOBUS_INITED) {
@@ -880,6 +926,9 @@ static void tg3_mdio_start(struct tg3 *tp)
 	tp->mi_mode &= ~MAC_MI_MODE_AUTO_POLL;
 	tw32_f(MAC_MI_MODE, tp->mi_mode);
 	udelay(80);
+
+	if (tp->tg3_flags3 & TG3_FLG3_MDIOBUS_INITED)
+		tg3_mdio_config(tp);
 }
 
 static void tg3_mdio_stop(struct tg3 *tp)
@@ -895,6 +944,7 @@ static int tg3_mdio_init(struct tg3 *tp)
 {
 	int i;
 	u32 reg;
+	struct phy_device *phydev;
 	struct mii_bus *mdio_bus = &tp->mdio_bus;
 
 	tg3_mdio_start(tp);
@@ -928,13 +978,34 @@ static int tg3_mdio_init(struct tg3 *tp)
 		tg3_bmcr_reset(tp);
 
 	i = mdiobus_register(mdio_bus);
-	if (!i)
-		tp->tg3_flags3 |= TG3_FLG3_MDIOBUS_INITED;
-	else
+	if (i) {
 		printk(KERN_WARNING "%s: mdiobus_reg failed (0x%x)\n",
 			tp->dev->name, i);
+		return i;
+	}
 
-	return i;
+	tp->tg3_flags3 |= TG3_FLG3_MDIOBUS_INITED;
+
+	phydev = tp->mdio_bus.phy_map[PHY_ADDR];
+
+	switch (phydev->phy_id) {
+	case TG3_PHY_ID_BCM50610:
+		phydev->interface = PHY_INTERFACE_MODE_RGMII;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_STD_IBND_DISABLE)
+			phydev->dev_flags |= PHY_BRCM_STD_IBND_DISABLE;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_RX_EN)
+			phydev->dev_flags |= PHY_BRCM_EXT_IBND_RX_ENABLE;
+		if (tp->tg3_flags3 & TG3_FLG3_RGMII_EXT_IBND_TX_EN)
+			phydev->dev_flags |= PHY_BRCM_EXT_IBND_TX_ENABLE;
+		break;
+	case TG3_PHY_ID_BCMAC131:
+		phydev->interface = PHY_INTERFACE_MODE_MII;
+		break;
+	}
+
+	tg3_mdio_config(tp);
+
+	return 0;
 }
 
 static void tg3_mdio_fini(struct tg3 *tp)
@@ -1238,8 +1309,8 @@ static int tg3_phy_init(struct tg3 *tp)
 	phydev = tp->mdio_bus.phy_map[PHY_ADDR];
 
 	/* Attach the MAC to the PHY. */
-	phydev = phy_connect(tp->dev, phydev->dev.bus_id,
-			     tg3_adjust_link, 0, phydev->interface);
+	phydev = phy_connect(tp->dev, phydev->dev.bus_id, tg3_adjust_link,
+			     phydev->dev_flags, phydev->interface);
 	if (IS_ERR(phydev)) {
 		printk(KERN_ERR "%s: Could not attach to PHY\n", tp->dev->name);
 		return PTR_ERR(phydev);
@@ -11219,7 +11290,7 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp)
 	tg3_read_mem(tp, NIC_SRAM_DATA_SIG, &val);
 	if (val == NIC_SRAM_DATA_SIG_MAGIC) {
 		u32 nic_cfg, led_cfg;
-		u32 nic_phy_id, ver, cfg2 = 0, eeprom_phy_id;
+		u32 nic_phy_id, ver, cfg2 = 0, cfg4 = 0, eeprom_phy_id;
 		int eeprom_phy_serdes = 0;
 
 		tg3_read_mem(tp, NIC_SRAM_DATA_CFG, &nic_cfg);
@@ -11233,6 +11304,9 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp)
 		    (ver > 0) && (ver < 0x100))
 			tg3_read_mem(tp, NIC_SRAM_DATA_CFG_2, &cfg2);
 
+		if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5785)
+			tg3_read_mem(tp, NIC_SRAM_DATA_CFG_4, &cfg4);
+
 		if ((nic_cfg & NIC_SRAM_DATA_CFG_PHY_TYPE_MASK) ==
 		    NIC_SRAM_DATA_CFG_PHY_TYPE_FIBER)
 			eeprom_phy_serdes = 1;
@@ -11357,6 +11431,13 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp)
 			if (cfg3 & NIC_SRAM_ASPM_DEBOUNCE)
 				tp->tg3_flags |= TG3_FLAG_ASPM_WORKAROUND;
 		}
+
+		if (cfg4 & NIC_SRAM_RGMII_STD_IBND_DISABLE)
+			tp->tg3_flags3 |= TG3_FLG3_RGMII_STD_IBND_DISABLE;
+		if (cfg4 & NIC_SRAM_RGMII_EXT_IBND_RX_EN)
+			tp->tg3_flags3 |= TG3_FLG3_RGMII_EXT_IBND_RX_EN;
+		if (cfg4 & NIC_SRAM_RGMII_EXT_IBND_TX_EN)
+			tp->tg3_flags3 |= TG3_FLG3_RGMII_EXT_IBND_TX_EN;
 	}
 }
 
diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h
index 95ac4c9590a1..df07842172b7 100644
--- a/drivers/net/tg3.h
+++ b/drivers/net/tg3.h
@@ -529,7 +529,23 @@
 #define MAC_SERDES_CFG			0x00000590
 #define  MAC_SERDES_CFG_EDGE_SELECT	 0x00001000
 #define MAC_SERDES_STAT			0x00000594
-/* 0x598 --> 0x5b0 unused */
+/* 0x598 --> 0x5a0 unused */
+#define MAC_PHYCFG1			0x000005a0
+#define  MAC_PHYCFG1_RGMII_INT		 0x00000001
+#define  MAC_PHYCFG1_RGMII_EXT_RX_DEC	 0x02000000
+#define  MAC_PHYCFG1_RGMII_SND_STAT_EN	 0x04000000
+#define  MAC_PHYCFG1_TXC_DRV		 0x20000000
+#define MAC_PHYCFG2			0x000005a4
+#define  MAC_PHYCFG2_INBAND_ENABLE	 0x00000001
+#define MAC_EXT_RGMII_MODE		0x000005a8
+#define  MAC_RGMII_MODE_TX_ENABLE	 0x00000001
+#define  MAC_RGMII_MODE_TX_LOWPWR	 0x00000002
+#define  MAC_RGMII_MODE_TX_RESET	 0x00000004
+#define  MAC_RGMII_MODE_RX_INT_B	 0x00000100
+#define  MAC_RGMII_MODE_RX_QUALITY	 0x00000200
+#define  MAC_RGMII_MODE_RX_ACTIVITY	 0x00000400
+#define  MAC_RGMII_MODE_RX_ENG_DET	 0x00000800
+/* 0x5ac --> 0x5b0 unused */
 #define SERDES_RX_CTRL			0x000005b0	/* 5780/5714 only */
 #define  SERDES_RX_SIG_DETECT		 0x00000400
 #define SG_DIG_CTRL			0x000005b0
@@ -1715,6 +1731,12 @@
 #define NIC_SRAM_DATA_CFG_3		0x00000d3c
 #define  NIC_SRAM_ASPM_DEBOUNCE		 0x00000002
 
+#define NIC_SRAM_DATA_CFG_4		0x00000d60
+#define  NIC_SRAM_GMII_MODE		 0x00000002
+#define  NIC_SRAM_RGMII_STD_IBND_DISABLE 0x00000004
+#define  NIC_SRAM_RGMII_EXT_IBND_RX_EN	 0x00000008
+#define  NIC_SRAM_RGMII_EXT_IBND_TX_EN	 0x00000010
+
 #define NIC_SRAM_RX_MINI_BUFFER_DESC	0x00001000
 
 #define NIC_SRAM_DMA_DESC_POOL_BASE	0x00002000
@@ -2486,6 +2508,9 @@ struct tg3 {
 #define TG3_FLG3_MDIOBUS_INITED		0x00000020
 #define TG3_FLG3_MDIOBUS_PAUSED		0x00000040
 #define TG3_FLG3_PHY_CONNECTED		0x00000080
+#define TG3_FLG3_RGMII_STD_IBND_DISABLE	0x00000100
+#define TG3_FLG3_RGMII_EXT_IBND_RX_EN	0x00000200
+#define TG3_FLG3_RGMII_EXT_IBND_TX_EN	0x00000400
 
 	struct timer_list		timer;
 	u16				timer_counter;
@@ -2556,6 +2581,9 @@ struct tg3 {
 #define PHY_REV_BCM5401_B2		0x3
 #define PHY_REV_BCM5401_C0		0x6
 #define PHY_REV_BCM5411_X0		0x1 /* Found on Netgear GA302T */
+#define TG3_PHY_ID_BCM50610		0x143bd60
+#define TG3_PHY_ID_BCMAC131		0x143bc70
+
 
 	u32				led_ctrl;
 	u32				phy_otp;
diff --git a/include/linux/brcmphy.h b/include/linux/brcmphy.h
new file mode 100644
index 000000000000..9b64b6d67873
--- /dev/null
+++ b/include/linux/brcmphy.h
@@ -0,0 +1,6 @@
+#define PHY_BRCM_WIRESPEED_ENABLE	0x00000001
+#define PHY_BRCM_AUTO_PWRDWN_ENABLE	0x00000002
+#define PHY_BRCM_APD_CLK125_ENABLE	0x00000004
+#define PHY_BRCM_STD_IBND_DISABLE	0x00000008
+#define PHY_BRCM_EXT_IBND_RX_ENABLE	0x00000010
+#define PHY_BRCM_EXT_IBND_TX_ENABLE	0x00000020
-- 
cgit v1.2.3


From 75a9cd524c6bb441c61c85bae7020ce5e8b2e807 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 29 May 2008 01:43:48 -0700
Subject: wanrouter: Fix ioctl handler declaration.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/wanrouter.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/wanrouter.h b/include/linux/wanrouter.h
index 3add87465b1f..e0aa39612eba 100644
--- a/include/linux/wanrouter.h
+++ b/include/linux/wanrouter.h
@@ -522,7 +522,7 @@ extern int wanrouter_proc_init(void);
 extern void wanrouter_proc_cleanup(void);
 extern int wanrouter_proc_add(struct wan_device *wandev);
 extern int wanrouter_proc_delete(struct wan_device *wandev);
-extern int wanrouter_ioctl( struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg);
+extern long wanrouter_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 
 /* Public Data */
 /* list of registered devices */
-- 
cgit v1.2.3


From b79eeeb9e48457579cb742cd02e162fcd673c4a3 Mon Sep 17 00:00:00 2001
From: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Date: Thu, 29 May 2008 03:25:23 -0700
Subject: tcp: Reorganize tcp_sock to fill 64-bit holes & improve locality
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I tried to group recovery related fields nearby (non-CA_Open related
variables, to be more accurate) so that one to three cachelines would
not be necessary in CA_Open. These are now contiguously deployed:

  struct sk_buff_head        out_of_order_queue;   /*  1968    80 */
  /* --- cacheline 32 boundary (2048 bytes) --- */
  struct tcp_sack_block      duplicate_sack[1];    /*  2048     8 */
  struct tcp_sack_block      selective_acks[4];    /*  2056    32 */
  struct tcp_sack_block      recv_sack_cache[4];   /*  2088    32 */
  /* --- cacheline 33 boundary (2112 bytes) was 8 bytes ago --- */
  struct sk_buff *           highest_sack;         /*  2120     8 */
  int                        lost_cnt_hint;        /*  2128     4 */
  int                        retransmit_cnt_hint;  /*  2132     4 */
  u32                        lost_retrans_low;     /*  2136     4 */
  u8                         reordering;           /*  2140     1 */
  u8                         keepalive_probes;     /*  2141     1 */

  /* XXX 2 bytes hole, try to pack */

  u32                        prior_ssthresh;       /*  2144     4 */
  u32                        high_seq;             /*  2148     4 */
  u32                        retrans_stamp;        /*  2152     4 */
  u32                        undo_marker;          /*  2156     4 */
  int                        undo_retrans;         /*  2160     4 */
  u32                        total_retrans;        /*  2164     4 */

...and they're then followed by URG slowpath & keepalive related
variables.

Head of the out_of_order_queue always needed for empty checks, if
that's empty (and TCP is in CA_Open), following ~200 bytes (in 64-bit)
shouldn't be necessary for anything. If only OFO queue exists but TCP
is in CA_Open, selective_acks (and possibly duplicate_sack) are
necessary besides the out_of_order_queue but the rest of the block
again shouldn't be (ie., the other direction had losses).

As the cacheline boundaries depend on many factors in the preceeding
stuff, trying to align considering them doesn't make too much sense.

Commented one ordering hazard.

There are number of low utilized u8/16s that could be combined get 2
bytes less in total so that the hole could be made to vanish (includes
at least ecn_flags, urg_data, urg_mode, frto_counter, nonagle).

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h | 50 +++++++++++++++++++++++++-------------------------
 1 file changed, 25 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 18e62e3d406f..9881295f3857 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -296,10 +296,9 @@ struct tcp_sock {
 	u32	rcv_ssthresh;	/* Current window clamp			*/
 
 	u32	frto_highmark;	/* snd_nxt when RTO occurred */
-	u8	reordering;	/* Packet reordering metric.		*/
+	u16	advmss;		/* Advertised MSS			*/
 	u8	frto_counter;	/* Number of new acks after RTO */
 	u8	nonagle;	/* Disable Nagle algorithm?             */
-	u8	keepalive_probes; /* num of allowed keep alive probes	*/
 
 /* RTT measurement */
 	u32	srtt;		/* smoothed round trip time << 3	*/
@@ -310,6 +309,10 @@ struct tcp_sock {
 
 	u32	packets_out;	/* Packets which are "in flight"	*/
 	u32	retrans_out;	/* Retransmitted packets out		*/
+
+	u16	urg_data;	/* Saved octet of OOB data and control flags */
+	u8	urg_mode;	/* In urgent mode		*/
+	u8	ecn_flags;	/* ECN status bits.			*/
 /*
  *      Options received (usually on last packet, some only on SYN packets).
  */
@@ -325,13 +328,24 @@ struct tcp_sock {
 	u32	snd_cwnd_used;
 	u32	snd_cwnd_stamp;
 
-	struct sk_buff_head	out_of_order_queue; /* Out of order segments go here */
-
  	u32	rcv_wnd;	/* Current receiver window		*/
 	u32	write_seq;	/* Tail(+1) of data held in tcp send buffer */
 	u32	pushed_seq;	/* Last pushed seq, required to talk to windows */
+	u32	lost_out;	/* Lost packets			*/
+	u32	sacked_out;	/* SACK'd packets			*/
+	u32	fackets_out;	/* FACK'd packets			*/
+	u32	tso_deferred;
+	u32	bytes_acked;	/* Appropriate Byte Counting - RFC3465 */
 
-/*	SACKs data	*/
+	/* from STCP, retrans queue hinting */
+	struct sk_buff* lost_skb_hint;
+	struct sk_buff *scoreboard_skb_hint;
+	struct sk_buff *retransmit_skb_hint;
+	struct sk_buff *forward_skb_hint;
+
+	struct sk_buff_head	out_of_order_queue; /* Out of order segments go here */
+
+	/* SACKs data, these 2 need to be together (see tcp_build_and_update_options) */
 	struct tcp_sack_block duplicate_sack[1]; /* D-SACK block */
 	struct tcp_sack_block selective_acks[4]; /* The SACKS themselves*/
 
@@ -342,23 +356,14 @@ struct tcp_sock {
 					 * sacked_out > 0)
 					 */
 
-	/* from STCP, retrans queue hinting */
-	struct sk_buff* lost_skb_hint;
-
-	struct sk_buff *scoreboard_skb_hint;
-	struct sk_buff *retransmit_skb_hint;
-	struct sk_buff *forward_skb_hint;
-
 	int     lost_cnt_hint;
 	int     retransmit_cnt_hint;
 
 	u32	lost_retrans_low;	/* Sent seq after any rxmit (lowest) */
 
-	u16	advmss;		/* Advertised MSS			*/
+	u8	reordering;	/* Packet reordering metric.		*/
+	u8	keepalive_probes; /* num of allowed keep alive probes	*/
 	u32	prior_ssthresh; /* ssthresh saved at recovery start	*/
-	u32	lost_out;	/* Lost packets			*/
-	u32	sacked_out;	/* SACK'd packets			*/
-	u32	fackets_out;	/* FACK'd packets			*/
 	u32	high_seq;	/* snd_nxt at onset of congestion	*/
 
 	u32	retrans_stamp;	/* Timestamp of the last retransmit,
@@ -366,25 +371,18 @@ struct tcp_sock {
 				 * the first SYN. */
 	u32	undo_marker;	/* tracking retrans started here. */
 	int	undo_retrans;	/* number of undoable retransmissions. */
+	u32	total_retrans;	/* Total retransmits for entire connection */
+
 	u32	urg_seq;	/* Seq of received urgent pointer */
-	u16	urg_data;	/* Saved octet of OOB data and control flags */
-	u8	urg_mode;	/* In urgent mode		*/
-	u8	ecn_flags;	/* ECN status bits.			*/
 	u32	snd_up;		/* Urgent pointer		*/
 
-	u32	total_retrans;	/* Total retransmits for entire connection */
-	u32	bytes_acked;	/* Appropriate Byte Counting - RFC3465 */
-
 	unsigned int		keepalive_time;	  /* time before keep alive takes place */
 	unsigned int		keepalive_intvl;  /* time interval between keep alive probes */
-	int			linger2;
 
 	struct tcp_deferred_accept_info defer_tcp_accept;
 
 	unsigned long last_synq_overflow; 
 
-	u32	tso_deferred;
-
 /* Receiver side RTT estimation */
 	struct {
 		u32	rtt;
@@ -412,6 +410,8 @@ struct tcp_sock {
 /* TCP MD5 Signagure Option information */
 	struct tcp_md5sig_info	*md5sig_info;
 #endif
+
+	int			linger2;
 };
 
 static inline struct tcp_sock *tcp_sk(const struct sock *sk)
-- 
cgit v1.2.3


From f6d65610df3bd4e7138da03aec391224219df135 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: Mon, 2 Jun 2008 00:39:45 -0400
Subject: Input: atkbd - use ushort instead of uchar keymap

Since some of the keycodes defined in input.h have values greater
than 255 we should use unsigned shorts in keymaps.

Tested-by: Carlos Corbacho <carlos@strangeworlds.co.uk>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/keyboard/atkbd.c | 20 ++++++++++----------
 include/linux/input.h          |  2 ++
 2 files changed, 12 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c
index 4a95adc4cc78..56857d1e56d5 100644
--- a/drivers/input/keyboard/atkbd.c
+++ b/drivers/input/keyboard/atkbd.c
@@ -68,7 +68,7 @@ MODULE_PARM_DESC(extra, "Enable extra LEDs and keys on IBM RapidAcces, EzKey and
  * are loadable via an userland utility.
  */
 
-static unsigned char atkbd_set2_keycode[512] = {
+static const unsigned short atkbd_set2_keycode[512] = {
 
 #ifdef CONFIG_KEYBOARD_ATKBD_HP_KEYCODES
 
@@ -99,7 +99,7 @@ static unsigned char atkbd_set2_keycode[512] = {
 #endif
 };
 
-static unsigned char atkbd_set3_keycode[512] = {
+static const unsigned short atkbd_set3_keycode[512] = {
 
 	  0,  0,  0,  0,  0,  0,  0, 59,  1,138,128,129,130, 15, 41, 60,
 	131, 29, 42, 86, 58, 16,  2, 61,133, 56, 44, 31, 30, 17,  3, 62,
@@ -115,7 +115,7 @@ static unsigned char atkbd_set3_keycode[512] = {
 	148,149,147,140
 };
 
-static unsigned char atkbd_unxlate_table[128] = {
+static const unsigned short atkbd_unxlate_table[128] = {
           0,118, 22, 30, 38, 37, 46, 54, 61, 62, 70, 69, 78, 85,102, 13,
          21, 29, 36, 45, 44, 53, 60, 67, 68, 77, 84, 91, 90, 20, 28, 27,
          35, 43, 52, 51, 59, 66, 75, 76, 82, 14, 18, 93, 26, 34, 33, 42,
@@ -161,7 +161,7 @@ static unsigned char atkbd_unxlate_table[128] = {
 #define ATKBD_SCR_LEFT		249
 #define ATKBD_SCR_RIGHT		248
 
-#define ATKBD_SPECIAL		248
+#define ATKBD_SPECIAL		ATKBD_SCR_RIGHT
 
 #define ATKBD_LED_EVENT_BIT	0
 #define ATKBD_REP_EVENT_BIT	1
@@ -173,7 +173,7 @@ static unsigned char atkbd_unxlate_table[128] = {
 #define ATKBD_XL_HANGEUL	0x10
 #define ATKBD_XL_HANJA		0x20
 
-static struct {
+static const struct {
 	unsigned char keycode;
 	unsigned char set2;
 } atkbd_scroll_keys[] = {
@@ -200,7 +200,7 @@ struct atkbd {
 	char phys[32];
 
 	unsigned short id;
-	unsigned char keycode[512];
+	unsigned short keycode[512];
 	DECLARE_BITMAP(force_release_mask, 512);
 	unsigned char set;
 	unsigned char translated;
@@ -357,7 +357,7 @@ static irqreturn_t atkbd_interrupt(struct serio *serio, unsigned char data,
 	unsigned int code = data;
 	int scroll = 0, hscroll = 0, click = -1;
 	int value;
-	unsigned char keycode;
+	unsigned short keycode;
 
 #ifdef ATKBD_DEBUG
 	printk(KERN_DEBUG "atkbd.c: Received %02x flags %02x\n", data, flags);
@@ -959,16 +959,16 @@ static void atkbd_set_device_attrs(struct atkbd *atkbd)
 		input_dev->evbit[0] |= BIT_MASK(EV_REL);
 		input_dev->relbit[0] = BIT_MASK(REL_WHEEL) |
 			BIT_MASK(REL_HWHEEL);
-		set_bit(BTN_MIDDLE, input_dev->keybit);
+		__set_bit(BTN_MIDDLE, input_dev->keybit);
 	}
 
 	input_dev->keycode = atkbd->keycode;
-	input_dev->keycodesize = sizeof(unsigned char);
+	input_dev->keycodesize = sizeof(unsigned short);
 	input_dev->keycodemax = ARRAY_SIZE(atkbd_set2_keycode);
 
 	for (i = 0; i < 512; i++)
 		if (atkbd->keycode[i] && atkbd->keycode[i] < ATKBD_SPECIAL)
-			set_bit(atkbd->keycode[i], input_dev->keybit);
+			__set_bit(atkbd->keycode[i], input_dev->keybit);
 }
 
 /*
diff --git a/include/linux/input.h b/include/linux/input.h
index eff711d8a459..af9c6314153b 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -373,6 +373,8 @@ struct input_absinfo {
 
 #define KEY_WIMAX		246
 
+/* Range 248 - 255 is reserved for special needs of AT keyboard driver */
+
 #define BTN_MISC		0x100
 #define BTN_0			0x100
 #define BTN_1			0x101
-- 
cgit v1.2.3


From ad90c0e3ce8d20d6873b57e36181ef6d7a0097fe Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Tue, 27 May 2008 20:48:37 -0400
Subject: ftrace: user update and disable dynamic ftrace daemon

In dynamic ftrace, the mcount function starts off pointing to a stub
function that just returns.

On start up, the call to the stub is modified to point to a "record_ip"
function. The job of the record_ip function is to add the function to
a pre-allocated hash list. If the function is already there, it simply is
ignored, otherwise it is added to the list.

Later, a ftraced daemon wakes up and calls kstop_machine if any functions
have been recorded, and changes the calls to the recorded functions to
a simple nop.  If no functions were recorded, the daemon goes back to sleep.

The daemon wakes up once a second to see if it needs to update any newly
recorded functions into nops.  Usually it does not, but if a lot of code
has been executed for the first time in the kernel, the ftraced daemon
will call kstop_machine to update those into nops.

The problem currently is that there's no way to stop the daemon from doing
this, and it can cause unneeded latencies (800us which for some is bothersome).

This patch adds a new file /debugfs/tracing/ftraced_enabled. If the daemon
is active, reading this will return "enabled\n" and "disabled\n" when the
daemon is not running. To disable the daemon, the user can echo "0" or
"disable" into this file, and "1" or "enable" to re-enable the daemon.

Since the daemon is used to convert the functions into nops to increase
the performance of the system, I also added that anytime something is
written into the ftraced_enabled file, kstop_machine will run if there
are new functions that have been detected that need to be converted.

This way the user can disable the daemon but still be able to control the
conversion of the mcount calls to nops by simply,

  "echo 0 > /debugfs/tracing/ftraced_enabled"

when they need to do more conversions.

To see the number of converted functions:

  "cat /debugfs/tracing/dyn_ftrace_total_info"

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/ftrace.h |   6 ++
 kernel/trace/ftrace.c  | 157 ++++++++++++++++++++++++++++++++++---------------
 2 files changed, 116 insertions(+), 47 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index b482fe88bc04..623819433ed5 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -72,9 +72,15 @@ extern int ftrace_update_ftrace_func(ftrace_func_t func);
 extern void ftrace_caller(void);
 extern void ftrace_call(void);
 extern void mcount_call(void);
+
+void ftrace_disable_daemon(void);
+void ftrace_enable_daemon(void);
+
 #else
 # define ftrace_force_update()			({ 0; })
 # define ftrace_set_filter(buf, len, reset)	do { } while (0)
+# define ftrace_disable_daemon()		do { } while (0)
+# define ftrace_enable_daemon()			do { } while (0)
 #endif
 
 /* totally disable ftrace - can not re-enable after this */
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 1843edc098a6..f762f5a2d331 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -151,8 +151,6 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops)
 #ifdef CONFIG_DYNAMIC_FTRACE
 
 static struct task_struct *ftraced_task;
-static DECLARE_WAIT_QUEUE_HEAD(ftraced_waiters);
-static unsigned long ftraced_iteration_counter;
 
 enum {
 	FTRACE_ENABLE_CALLS		= (1 << 0),
@@ -189,6 +187,7 @@ static struct ftrace_page	*ftrace_pages;
 
 static int ftraced_trigger;
 static int ftraced_suspend;
+static int ftraced_stop;
 
 static int ftrace_record_suspend;
 
@@ -474,14 +473,21 @@ ftrace_code_disable(struct dyn_ftrace *rec)
 	return 1;
 }
 
+static int __ftrace_update_code(void *ignore);
+
 static int __ftrace_modify_code(void *data)
 {
 	unsigned long addr;
 	int *command = data;
 
-	if (*command & FTRACE_ENABLE_CALLS)
+	if (*command & FTRACE_ENABLE_CALLS) {
+		/*
+		 * Update any recorded ips now that we have the
+		 * machine stopped
+		 */
+		__ftrace_update_code(NULL);
 		ftrace_replace_code(1);
-	else if (*command & FTRACE_DISABLE_CALLS)
+	} else if (*command & FTRACE_DISABLE_CALLS)
 		ftrace_replace_code(0);
 
 	if (*command & FTRACE_UPDATE_TRACE_FUNC)
@@ -503,6 +509,25 @@ static void ftrace_run_update_code(int command)
 	stop_machine_run(__ftrace_modify_code, &command, NR_CPUS);
 }
 
+void ftrace_disable_daemon(void)
+{
+	/* Stop the daemon from calling kstop_machine */
+	mutex_lock(&ftraced_lock);
+	ftraced_stop = 1;
+	mutex_unlock(&ftraced_lock);
+
+	ftrace_force_update();
+}
+
+void ftrace_enable_daemon(void)
+{
+	mutex_lock(&ftraced_lock);
+	ftraced_stop = 0;
+	mutex_unlock(&ftraced_lock);
+
+	ftrace_force_update();
+}
+
 static ftrace_func_t saved_ftrace_func;
 
 static void ftrace_startup(void)
@@ -603,6 +628,7 @@ static int __ftrace_update_code(void *ignore)
 	int i;
 
 	/* Don't be recording funcs now */
+	ftrace_record_suspend++;
 	save_ftrace_enabled = ftrace_enabled;
 	ftrace_enabled = 0;
 
@@ -628,18 +654,23 @@ static int __ftrace_update_code(void *ignore)
 	stop = ftrace_now(raw_smp_processor_id());
 	ftrace_update_time = stop - start;
 	ftrace_update_tot_cnt += ftrace_update_cnt;
+	ftraced_trigger = 0;
 
 	ftrace_enabled = save_ftrace_enabled;
+	ftrace_record_suspend--;
 
 	return 0;
 }
 
-static void ftrace_update_code(void)
+static int ftrace_update_code(void)
 {
-	if (unlikely(ftrace_disabled))
-		return;
+	if (unlikely(ftrace_disabled) ||
+	    !ftrace_enabled || !ftraced_trigger)
+		return 0;
 
 	stop_machine_run(__ftrace_update_code, NULL, NR_CPUS);
+
+	return 1;
 }
 
 static int ftraced(void *ignore)
@@ -658,14 +689,13 @@ static int ftraced(void *ignore)
 
 		mutex_lock(&ftrace_sysctl_lock);
 		mutex_lock(&ftraced_lock);
-		if (ftrace_enabled && ftraced_trigger && !ftraced_suspend) {
-			ftrace_record_suspend++;
-			ftrace_update_code();
+		if (!ftraced_suspend && !ftraced_stop &&
+		    ftrace_update_code()) {
 			usecs = nsecs_to_usecs(ftrace_update_time);
 			if (ftrace_update_tot_cnt > 100000) {
 				ftrace_update_tot_cnt = 0;
 				pr_info("hm, dftrace overflow: %lu change%s"
-					 " (%lu total) in %lu usec%s\n",
+					" (%lu total) in %lu usec%s\n",
 					ftrace_update_cnt,
 					ftrace_update_cnt != 1 ? "s" : "",
 					ftrace_update_tot_cnt,
@@ -673,15 +703,10 @@ static int ftraced(void *ignore)
 				ftrace_disabled = 1;
 				WARN_ON_ONCE(1);
 			}
-			ftraced_trigger = 0;
-			ftrace_record_suspend--;
 		}
-		ftraced_iteration_counter++;
 		mutex_unlock(&ftraced_lock);
 		mutex_unlock(&ftrace_sysctl_lock);
 
-		wake_up_interruptible(&ftraced_waiters);
-
 		ftrace_shutdown_replenish();
 	}
 	__set_current_state(TASK_RUNNING);
@@ -1219,6 +1244,55 @@ ftrace_notrace_release(struct inode *inode, struct file *file)
 	return ftrace_regex_release(inode, file, 0);
 }
 
+static ssize_t
+ftraced_read(struct file *filp, char __user *ubuf,
+		     size_t cnt, loff_t *ppos)
+{
+	/* don't worry about races */
+	char *buf = ftraced_stop ? "disabled\n" : "enabled\n";
+	int r = strlen(buf);
+
+	return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
+}
+
+static ssize_t
+ftraced_write(struct file *filp, const char __user *ubuf,
+		      size_t cnt, loff_t *ppos)
+{
+	char buf[64];
+	long val;
+	int ret;
+
+	if (cnt >= sizeof(buf))
+		return -EINVAL;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	if (strncmp(buf, "enable", 6) == 0)
+		val = 1;
+	else if (strncmp(buf, "disable", 7) == 0)
+		val = 0;
+	else {
+		buf[cnt] = 0;
+
+		ret = strict_strtoul(buf, 10, &val);
+		if (ret < 0)
+			return ret;
+
+		val = !!val;
+	}
+
+	if (val)
+		ftrace_enable_daemon();
+	else
+		ftrace_disable_daemon();
+
+	filp->f_pos += cnt;
+
+	return cnt;
+}
+
 static struct file_operations ftrace_avail_fops = {
 	.open = ftrace_avail_open,
 	.read = seq_read,
@@ -1242,51 +1316,34 @@ static struct file_operations ftrace_notrace_fops = {
 	.release = ftrace_notrace_release,
 };
 
+static struct file_operations ftraced_fops = {
+	.open = tracing_open_generic,
+	.read = ftraced_read,
+	.write = ftraced_write,
+};
+
 /**
  * ftrace_force_update - force an update to all recording ftrace functions
- *
- * The ftrace dynamic update daemon only wakes up once a second.
- * There may be cases where an update needs to be done immediately
- * for tests or internal kernel tracing to begin. This function
- * wakes the daemon to do an update and will not return until the
- * update is complete.
  */
 int ftrace_force_update(void)
 {
-	unsigned long last_counter;
-	DECLARE_WAITQUEUE(wait, current);
 	int ret = 0;
 
 	if (unlikely(ftrace_disabled))
 		return -ENODEV;
 
+	mutex_lock(&ftrace_sysctl_lock);
 	mutex_lock(&ftraced_lock);
-	last_counter = ftraced_iteration_counter;
-
-	set_current_state(TASK_INTERRUPTIBLE);
-	add_wait_queue(&ftraced_waiters, &wait);
 
-	if (unlikely(!ftraced_task)) {
-		ret = -ENODEV;
-		goto out;
-	}
-
-	do {
-		mutex_unlock(&ftraced_lock);
-		wake_up_process(ftraced_task);
-		schedule();
-		mutex_lock(&ftraced_lock);
-		if (signal_pending(current)) {
-			ret = -EINTR;
-			break;
-		}
-		set_current_state(TASK_INTERRUPTIBLE);
-	} while (last_counter == ftraced_iteration_counter);
+	/*
+	 * If ftraced_trigger is not set, then there is nothing
+	 * to update.
+	 */
+	if (ftraced_trigger && !ftrace_update_code())
+		ret = -EBUSY;
 
- out:
 	mutex_unlock(&ftraced_lock);
-	remove_wait_queue(&ftraced_waiters, &wait);
-	set_current_state(TASK_RUNNING);
+	mutex_unlock(&ftrace_sysctl_lock);
 
 	return ret;
 }
@@ -1331,6 +1388,12 @@ static __init int ftrace_init_debugfs(void)
 	if (!entry)
 		pr_warning("Could not create debugfs "
 			   "'set_ftrace_notrace' entry\n");
+
+	entry = debugfs_create_file("ftraced_enabled", 0644, d_tracer,
+				    NULL, &ftraced_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs "
+			   "'ftraced_enabled' entry\n");
 	return 0;
 }
 
-- 
cgit v1.2.3


From 9306102ea5696a3815f8d24ac0c0fbd1e19be7d3 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 29 May 2008 16:35:23 +0800
Subject: mac80211: allow disable FAT in specific configurations

This patch allows to disable FAT channel in specific configurations.

For example the configuration (8, +1), (primary channel 8, extension
channel 12) isn't permitted in U.S., but (8, -1), (primary channel 8,
extension channel 4) is. When FAT channel configuration is not
permitted, FAT channel should be reported as not supported in the
capabilities of the HT IE in association request. And sssociation is
performed on 20Mhz channel.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h  |  2 ++
 include/net/wireless.h     |  6 ++++++
 net/mac80211/ieee80211_i.h |  2 ++
 net/mac80211/mlme.c        | 46 ++++++++++++++++++++++++++++++++++++++++++++--
 4 files changed, 54 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 3c2ac0c37aa8..9300f37cd7e8 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -320,6 +320,8 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_CAP_MCS_TX_UEQM		0x10
 /* 802.11n HT IE masks */
 #define IEEE80211_HT_IE_CHA_SEC_OFFSET		0x03
+#define IEEE80211_HT_IE_CHA_SEC_ABOVE 		0x01
+#define IEEE80211_HT_IE_CHA_SEC_BELOW 		0x03
 #define IEEE80211_HT_IE_CHA_WIDTH		0x04
 #define IEEE80211_HT_IE_HT_PROTECTION		0x0003
 #define IEEE80211_HT_IE_NON_GF_STA_PRSNT	0x0004
diff --git a/include/net/wireless.h b/include/net/wireless.h
index 667b4080d30f..9324f8dd183e 100644
--- a/include/net/wireless.h
+++ b/include/net/wireless.h
@@ -39,12 +39,18 @@ enum ieee80211_band {
  *	on this channel.
  * @IEEE80211_CHAN_NO_IBSS: IBSS is not allowed on this channel.
  * @IEEE80211_CHAN_RADAR: Radar detection is required on this channel.
+ * @IEEE80211_CHAN_NO_FAT_ABOVE: extension channel above this channel
+ * 	is not permitted.
+ * @IEEE80211_CHAN_NO_FAT_BELOW: extension channel below this channel
+ * 	is not permitted.
  */
 enum ieee80211_channel_flags {
 	IEEE80211_CHAN_DISABLED		= 1<<0,
 	IEEE80211_CHAN_PASSIVE_SCAN	= 1<<1,
 	IEEE80211_CHAN_NO_IBSS		= 1<<2,
 	IEEE80211_CHAN_RADAR		= 1<<3,
+	IEEE80211_CHAN_NO_FAT_ABOVE	= 1<<4,
+	IEEE80211_CHAN_NO_FAT_BELOW	= 1<<5,
 };
 
 /**
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 3f8601cafffb..432011cd3647 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -92,6 +92,8 @@ struct ieee80211_sta_bss {
 	size_t wmm_ie_len;
 	u8 *ht_ie;
 	size_t ht_ie_len;
+	u8 *ht_add_ie;
+	size_t ht_add_ie_len;
 #ifdef CONFIG_MAC80211_MESH
 	u8 *mesh_id;
 	size_t mesh_id_len;
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 6faa7006681a..d30c11337b04 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -808,8 +808,29 @@ static void ieee80211_send_assoc(struct net_device *dev,
 
 	/* wmm support is a must to HT */
 	if (wmm && (ifsta->flags & IEEE80211_STA_WMM_ENABLED) &&
-	    sband->ht_info.ht_supported) {
-		__le16 tmp = cpu_to_le16(sband->ht_info.cap);
+	    sband->ht_info.ht_supported && bss->ht_add_ie) {
+		struct ieee80211_ht_addt_info *ht_add_info =
+			(struct ieee80211_ht_addt_info *)bss->ht_add_ie;
+		u16 cap = sband->ht_info.cap;
+		__le16 tmp;
+		u32 flags = local->hw.conf.channel->flags;
+
+		switch (ht_add_info->ht_param & IEEE80211_HT_IE_CHA_SEC_OFFSET) {
+		case IEEE80211_HT_IE_CHA_SEC_ABOVE:
+			if (flags & IEEE80211_CHAN_NO_FAT_ABOVE) {
+				cap &= ~IEEE80211_HT_CAP_SUP_WIDTH;
+				cap &= ~IEEE80211_HT_CAP_SGI_40;
+			}
+			break;
+		case IEEE80211_HT_IE_CHA_SEC_BELOW:
+			if (flags & IEEE80211_CHAN_NO_FAT_BELOW) {
+				cap &= ~IEEE80211_HT_CAP_SUP_WIDTH;
+				cap &= ~IEEE80211_HT_CAP_SGI_40;
+			}
+			break;
+		}
+
+		tmp = cpu_to_le16(cap);
 		pos = skb_put(skb, sizeof(struct ieee80211_ht_cap)+2);
 		*pos++ = WLAN_EID_HT_CAPABILITY;
 		*pos++ = sizeof(struct ieee80211_ht_cap);
@@ -2264,6 +2285,7 @@ static void ieee80211_rx_bss_free(struct ieee80211_sta_bss *bss)
 	kfree(bss->rsn_ie);
 	kfree(bss->wmm_ie);
 	kfree(bss->ht_ie);
+	kfree(bss->ht_add_ie);
 	kfree(bss_mesh_id(bss));
 	kfree(bss_mesh_cfg(bss));
 	kfree(bss);
@@ -2640,6 +2662,26 @@ static void ieee80211_rx_bss_info(struct net_device *dev,
 		bss->ht_ie_len = 0;
 	}
 
+	if (elems.ht_info_elem &&
+	     (!bss->ht_add_ie ||
+	     bss->ht_add_ie_len != elems.ht_info_elem_len ||
+	     memcmp(bss->ht_add_ie, elems.ht_info_elem,
+			elems.ht_info_elem_len))) {
+		kfree(bss->ht_add_ie);
+		bss->ht_add_ie =
+			kmalloc(elems.ht_info_elem_len + 2, GFP_ATOMIC);
+		if (bss->ht_add_ie) {
+			memcpy(bss->ht_add_ie, elems.ht_info_elem - 2,
+				elems.ht_info_elem_len + 2);
+			bss->ht_add_ie_len = elems.ht_info_elem_len + 2;
+		} else
+			bss->ht_add_ie_len = 0;
+	} else if (!elems.ht_info_elem && bss->ht_add_ie) {
+		kfree(bss->ht_add_ie);
+		bss->ht_add_ie = NULL;
+		bss->ht_add_ie_len = 0;
+	}
+
 	bss->beacon_int = le16_to_cpu(mgmt->u.beacon.beacon_int);
 	bss->capability = le16_to_cpu(mgmt->u.beacon.capab_info);
 
-- 
cgit v1.2.3


From ba569b4c68f11906da2996ee252bcff0df61cb90 Mon Sep 17 00:00:00 2001
From: Masakazu Mokuno <mokuno@sm.sony.co.jp>
Date: Fri, 30 May 2008 16:52:23 +0900
Subject: WEXT: Add support for passing PMK and capability flags to WEXT

This defines the flags for setting the PMK to the driver and the
capability flag for this so that the user space program can figure out
whether the target driver wants to do 4-way hand shake by itself and
pass the PMK which is needed before 4-way handshake to the driver.

Signed-off-by: Masakazu Mokuno <mokuno@sm.sony.co.jp>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/wireless.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 0a9b5b41ed67..4a95a0e5eeca 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -611,6 +611,7 @@
 #define IW_ENCODE_ALG_WEP	1
 #define IW_ENCODE_ALG_TKIP	2
 #define IW_ENCODE_ALG_CCMP	3
+#define IW_ENCODE_ALG_PMK	4
 /* struct iw_encode_ext ->ext_flags */
 #define IW_ENCODE_EXT_TX_SEQ_VALID	0x00000001
 #define IW_ENCODE_EXT_RX_SEQ_VALID	0x00000002
@@ -630,6 +631,7 @@
 #define IW_ENC_CAPA_WPA2	0x00000002
 #define IW_ENC_CAPA_CIPHER_TKIP	0x00000004
 #define IW_ENC_CAPA_CIPHER_CCMP	0x00000008
+#define IW_ENC_CAPA_4WAY_HANDSHAKE	0x00000010
 
 /* Event capability macros - in (struct iw_range *)->event_capa
  * Because we have more than 32 possible events, we use an array of
-- 
cgit v1.2.3


From 18404756765c713a0be4eb1082920c04822ce588 Mon Sep 17 00:00:00 2001
From: Max Krasnyansky <maxk@qualcomm.com>
Date: Thu, 29 May 2008 11:02:52 -0700
Subject: genirq: Expose default irq affinity mask (take 3)

Current IRQ affinity interface does not provide a way to set affinity
for the IRQs that will be allocated/activated in the future.
This patch creates /proc/irq/default_smp_affinity that lets users set
default affinity mask for the newly allocated IRQs. Changing the default
does not affect affinity masks for the currently active IRQs, they
have to be changed explicitly.

Updated based on Paul J's comments and added some more documentation.

Signed-off-by: Max Krasnyansky <maxk@qualcomm.com>
Cc: pj@sgi.com
Cc: a.p.zijlstra@chello.nl
Cc: tglx@linutronix.de
Cc: rdunlap@xenotime.net
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/IRQ-affinity.txt     | 37 ++++++++++++++++++------
 Documentation/filesystems/proc.txt | 29 ++++++++++++-------
 arch/alpha/kernel/irq.c            |  5 ++--
 include/linux/interrupt.h          |  5 ++++
 include/linux/irq.h                |  9 ------
 kernel/irq/manage.c                | 28 ++++++++++++++++--
 kernel/irq/proc.c                  | 59 +++++++++++++++++++++++++++++++++++---
 7 files changed, 134 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/IRQ-affinity.txt b/Documentation/IRQ-affinity.txt
index 938d7dd05490..b4a615b78403 100644
--- a/Documentation/IRQ-affinity.txt
+++ b/Documentation/IRQ-affinity.txt
@@ -1,17 +1,26 @@
+ChangeLog:
+	Started by Ingo Molnar <mingo@redhat.com>
+	Update by Max Krasnyansky <maxk@qualcomm.com>
 
-SMP IRQ affinity, started by Ingo Molnar <mingo@redhat.com>
-
+SMP IRQ affinity
 
 /proc/irq/IRQ#/smp_affinity specifies which target CPUs are permitted
 for a given IRQ source. It's a bitmask of allowed CPUs. It's not allowed
 to turn off all CPUs, and if an IRQ controller does not support IRQ
 affinity then the value will not change from the default 0xffffffff.
 
+/proc/irq/default_smp_affinity specifies default affinity mask that applies
+to all non-active IRQs. Once IRQ is allocated/activated its affinity bitmask
+will be set to the default mask. It can then be changed as described above.
+Default mask is 0xffffffff.
+
 Here is an example of restricting IRQ44 (eth1) to CPU0-3 then restricting
-the IRQ to CPU4-7 (this is an 8-CPU SMP box):
+it to CPU4-7 (this is an 8-CPU SMP box):
 
+[root@moon 44]# cd /proc/irq/44
 [root@moon 44]# cat smp_affinity
 ffffffff
+
 [root@moon 44]# echo 0f > smp_affinity
 [root@moon 44]# cat smp_affinity
 0000000f
@@ -21,17 +30,27 @@ PING hell (195.4.7.3): 56 data bytes
 --- hell ping statistics ---
 6029 packets transmitted, 6027 packets received, 0% packet loss
 round-trip min/avg/max = 0.1/0.1/0.4 ms
-[root@moon 44]# cat /proc/interrupts | grep 44:
- 44:          0       1785       1785       1783       1783          1
-1          0   IO-APIC-level  eth1
+[root@moon 44]# cat /proc/interrupts | grep 'CPU\|44:'
+           CPU0       CPU1       CPU2       CPU3      CPU4       CPU5        CPU6       CPU7
+ 44:       1068       1785       1785       1783         0          0           0         0    IO-APIC-level  eth1
+
+As can be seen from the line above IRQ44 was delivered only to the first four
+processors (0-3).
+Now lets restrict that IRQ to CPU(4-7).
+
 [root@moon 44]# echo f0 > smp_affinity
+[root@moon 44]# cat smp_affinity
+000000f0
 [root@moon 44]# ping -f h
 PING hell (195.4.7.3): 56 data bytes
 ..
 --- hell ping statistics ---
 2779 packets transmitted, 2777 packets received, 0% packet loss
 round-trip min/avg/max = 0.1/0.5/585.4 ms
-[root@moon 44]# cat /proc/interrupts | grep 44:
- 44:       1068       1785       1785       1784       1784       1069       1070       1069   IO-APIC-level  eth1
-[root@moon 44]#
+[root@moon 44]# cat /proc/interrupts |  'CPU\|44:'
+           CPU0       CPU1       CPU2       CPU3      CPU4       CPU5        CPU6       CPU7
+ 44:       1068       1785       1785       1783      1784       1069        1070       1069   IO-APIC-level  eth1
+
+This time around IRQ44 was delivered only to the last four processors.
+i.e counters for the CPU0-3 did not change.
 
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index dbc3c6a3650f..7f268f327d75 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -380,28 +380,35 @@ i386 and x86_64 platforms support the new IRQ vector displays.
 Of some interest is the introduction of the /proc/irq directory to 2.4.
 It could be used to set IRQ to CPU affinity, this means that you can "hook" an
 IRQ to only one CPU, or to exclude a CPU of handling IRQs. The contents of the
-irq subdir is one subdir for each IRQ, and one file; prof_cpu_mask
+irq subdir is one subdir for each IRQ, and two files; default_smp_affinity and
+prof_cpu_mask.
 
 For example 
   > ls /proc/irq/
   0  10  12  14  16  18  2  4  6  8  prof_cpu_mask
-  1  11  13  15  17  19  3  5  7  9
+  1  11  13  15  17  19  3  5  7  9  default_smp_affinity
   > ls /proc/irq/0/
   smp_affinity
 
-The contents of the prof_cpu_mask file and each smp_affinity file for each IRQ
-is the same by default:
+smp_affinity is a bitmask, in which you can specify which CPUs can handle the
+IRQ, you can set it by doing:
 
-  > cat /proc/irq/0/smp_affinity 
-  ffffffff
+  > echo 1 > /proc/irq/10/smp_affinity
+
+This means that only the first CPU will handle the IRQ, but you can also echo
+5 which means that only the first and fourth CPU can handle the IRQ.
 
-It's a bitmask, in which you can specify which CPUs can handle the IRQ, you can
-set it by doing:
+The contents of each smp_affinity file is the same by default:
+
+  > cat /proc/irq/0/smp_affinity
+  ffffffff
 
-  > echo 1 > /proc/irq/prof_cpu_mask
+The default_smp_affinity mask applies to all non-active IRQs, which are the
+IRQs which have not yet been allocated/activated, and hence which lack a
+/proc/irq/[0-9]* directory.
 
-This means that only the first CPU will handle the IRQ, but you can also echo 5
-which means that only the first and fourth CPU can handle the IRQ.
+prof_cpu_mask specifies which CPUs are to be profiled by the system wide
+profiler. Default value is ffffffff (all cpus).
 
 The way IRQs are routed is handled by the IO-APIC, and it's Round Robin
 between all the CPUs which are allowed to handle it. As usual the kernel has
diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c
index facf82a5499a..c626a821cdcb 100644
--- a/arch/alpha/kernel/irq.c
+++ b/arch/alpha/kernel/irq.c
@@ -42,8 +42,7 @@ void ack_bad_irq(unsigned int irq)
 #ifdef CONFIG_SMP 
 static char irq_user_affinity[NR_IRQS];
 
-int
-select_smp_affinity(unsigned int irq)
+int irq_select_affinity(unsigned int irq)
 {
 	static int last_cpu;
 	int cpu = last_cpu + 1;
@@ -51,7 +50,7 @@ select_smp_affinity(unsigned int irq)
 	if (!irq_desc[irq].chip->set_affinity || irq_user_affinity[irq])
 		return 1;
 
-	while (!cpu_possible(cpu))
+	while (!cpu_possible(cpu) || !cpu_isset(cpu, irq_default_affinity))
 		cpu = (cpu < (NR_CPUS-1) ? cpu + 1 : 0);
 	last_cpu = cpu;
 
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index f1fc7470d26c..043400f3d458 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -104,8 +104,11 @@ extern void enable_irq(unsigned int irq);
 
 #if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
 
+extern cpumask_t irq_default_affinity;
+
 extern int irq_set_affinity(unsigned int irq, cpumask_t cpumask);
 extern int irq_can_set_affinity(unsigned int irq);
+extern int irq_select_affinity(unsigned int irq);
 
 #else /* CONFIG_SMP */
 
@@ -119,6 +122,8 @@ static inline int irq_can_set_affinity(unsigned int irq)
 	return 0;
 }
 
+static inline int irq_select_affinity(unsigned int irq)  { return 0; }
+
 #endif /* CONFIG_SMP && CONFIG_GENERIC_HARDIRQS */
 
 #ifdef CONFIG_GENERIC_HARDIRQS
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 552e0ec269c9..8ccb462ea42c 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -244,15 +244,6 @@ static inline void set_balance_irq_affinity(unsigned int irq, cpumask_t mask)
 }
 #endif
 
-#ifdef CONFIG_AUTO_IRQ_AFFINITY
-extern int select_smp_affinity(unsigned int irq);
-#else
-static inline int select_smp_affinity(unsigned int irq)
-{
-	return 1;
-}
-#endif
-
 extern int no_irq_affinity;
 
 static inline int irq_balancing_disabled(unsigned int irq)
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 46d6611a33bb..469814e9b9ee 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -17,6 +17,8 @@
 
 #ifdef CONFIG_SMP
 
+cpumask_t irq_default_affinity = CPU_MASK_ALL;
+
 /**
  *	synchronize_irq - wait for pending IRQ handlers (on other CPUs)
  *	@irq: interrupt number to wait for
@@ -95,6 +97,27 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
 	return 0;
 }
 
+#ifndef CONFIG_AUTO_IRQ_AFFINITY
+/*
+ * Generic version of the affinity autoselector.
+ */
+int irq_select_affinity(unsigned int irq)
+{
+	cpumask_t mask;
+
+	if (!irq_can_set_affinity(irq))
+		return 0;
+
+	cpus_and(mask, cpu_online_map, irq_default_affinity);
+
+	irq_desc[irq].affinity = mask;
+	irq_desc[irq].chip->set_affinity(irq, mask);
+
+	set_balance_irq_affinity(irq, mask);
+	return 0;
+}
+#endif
+
 #endif
 
 /**
@@ -382,6 +405,9 @@ int setup_irq(unsigned int irq, struct irqaction *new)
 		} else
 			/* Undo nested disables: */
 			desc->depth = 1;
+
+		/* Set default affinity mask once everything is setup */
+		irq_select_affinity(irq);
 	}
 	/* Reset broken irq detection when installing new handler */
 	desc->irq_count = 0;
@@ -571,8 +597,6 @@ int request_irq(unsigned int irq, irq_handler_t handler,
 	action->next = NULL;
 	action->dev_id = dev_id;
 
-	select_smp_affinity(irq);
-
 #ifdef CONFIG_DEBUG_SHIRQ
 	if (irqflags & IRQF_SHARED) {
 		/*
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index c2f2ccb0549a..6c6d35d68ee9 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -44,7 +44,7 @@ static int irq_affinity_write_proc(struct file *file, const char __user *buffer,
 				   unsigned long count, void *data)
 {
 	unsigned int irq = (int)(long)data, full_count = count, err;
-	cpumask_t new_value, tmp;
+	cpumask_t new_value;
 
 	if (!irq_desc[irq].chip->set_affinity || no_irq_affinity ||
 	    irq_balancing_disabled(irq))
@@ -62,17 +62,51 @@ static int irq_affinity_write_proc(struct file *file, const char __user *buffer,
 	 * way to make the system unusable accidentally :-) At least
 	 * one online CPU still has to be targeted.
 	 */
-	cpus_and(tmp, new_value, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpus_intersects(new_value, cpu_online_map))
 		/* Special case for empty set - allow the architecture
 		   code to set default SMP affinity. */
-		return select_smp_affinity(irq) ? -EINVAL : full_count;
+		return irq_select_affinity(irq) ? -EINVAL : full_count;
 
 	irq_set_affinity(irq, new_value);
 
 	return full_count;
 }
 
+static int default_affinity_read(char *page, char **start, off_t off,
+				  int count, int *eof, void *data)
+{
+	int len = cpumask_scnprintf(page, count, irq_default_affinity);
+	if (count - len < 2)
+		return -EINVAL;
+	len += sprintf(page + len, "\n");
+	return len;
+}
+
+static int default_affinity_write(struct file *file, const char __user *buffer,
+				   unsigned long count, void *data)
+{
+	unsigned int full_count = count, err;
+	cpumask_t new_value;
+
+	err = cpumask_parse_user(buffer, count, new_value);
+	if (err)
+		return err;
+
+	if (!is_affinity_mask_valid(new_value))
+		return -EINVAL;
+
+	/*
+	 * Do not allow disabling IRQs completely - it's a too easy
+	 * way to make the system unusable accidentally :-) At least
+	 * one online CPU still has to be targeted.
+	 */
+	if (!cpus_intersects(new_value, cpu_online_map))
+		return -EINVAL;
+
+	irq_default_affinity = new_value;
+
+	return full_count;
+}
 #endif
 
 static int irq_spurious_read(char *page, char **start, off_t off,
@@ -171,6 +205,21 @@ void unregister_handler_proc(unsigned int irq, struct irqaction *action)
 		remove_proc_entry(action->dir->name, irq_desc[irq].dir);
 }
 
+void register_default_affinity_proc(void)
+{
+#ifdef CONFIG_SMP
+	struct proc_dir_entry *entry;
+
+	/* create /proc/irq/default_smp_affinity */
+	entry = create_proc_entry("default_smp_affinity", 0600, root_irq_dir);
+	if (entry) {
+		entry->data = NULL;
+		entry->read_proc  = default_affinity_read;
+		entry->write_proc = default_affinity_write;
+	}
+#endif
+}
+
 void init_irq_proc(void)
 {
 	int i;
@@ -180,6 +229,8 @@ void init_irq_proc(void)
 	if (!root_irq_dir)
 		return;
 
+	register_default_affinity_proc();
+
 	/*
 	 * Create entries for all existing IRQs.
 	 */
-- 
cgit v1.2.3


From 9457afee85e0dfc2b5075a391d6f34463b4c2b90 Mon Sep 17 00:00:00 2001
From: "Denis V. Lunev" <den@openvz.org>
Date: Thu, 5 Jun 2008 11:23:39 -0700
Subject: netlink: Remove nonblock parameter from netlink_attachskb

Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h  | 2 +-
 ipc/mqueue.c             | 2 +-
 net/netlink/af_netlink.c | 4 ++--
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index bec1062a25a1..9ff1b54908f3 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -193,7 +193,7 @@ extern int netlink_unregister_notifier(struct notifier_block *nb);
 
 /* finegrained unicast helpers: */
 struct sock *netlink_getsockbyfilp(struct file *filp);
-int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock,
+int netlink_attachskb(struct sock *sk, struct sk_buff *skb,
 		      long *timeo, struct sock *ssk);
 void netlink_detachskb(struct sock *sk, struct sk_buff *skb);
 int netlink_sendskb(struct sock *sk, struct sk_buff *skb);
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index b3b69fd51330..3e84b958186b 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1054,7 +1054,7 @@ retry:
 			}
 
 			timeo = MAX_SCHEDULE_TIMEOUT;
-			ret = netlink_attachskb(sock, nc, 0, &timeo, NULL);
+			ret = netlink_attachskb(sock, nc, &timeo, NULL);
 			if (ret == 1)
 		       		goto retry;
 			if (ret) {
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 9b97f8006c9c..6507c02dbe0d 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -759,7 +759,7 @@ struct sock *netlink_getsockbyfilp(struct file *filp)
  * 0: continue
  * 1: repeat lookup - reference dropped while waiting for socket memory.
  */
-int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock,
+int netlink_attachskb(struct sock *sk, struct sk_buff *skb,
 		      long *timeo, struct sock *ssk)
 {
 	struct netlink_sock *nlk;
@@ -892,7 +892,7 @@ retry:
 		return err;
 	}
 
-	err = netlink_attachskb(sk, skb, nonblock, &timeo, ssk);
+	err = netlink_attachskb(sk, skb, &timeo, ssk);
 	if (err == 1)
 		goto retry;
 	if (err)
-- 
cgit v1.2.3


From 554ec22f075d46e4363520a407d2b7eeb5dfdd43 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 12 May 2008 21:21:03 +0200
Subject: namespacecheck: more sched.c fixes

[ Stephen Rothwell <sfr@canb.auug.org.au>: build fix ]

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h | 18 ------------------
 1 file changed, 18 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index ae0be3c62375..dc36c3aea018 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -134,7 +134,6 @@ extern unsigned long nr_running(void);
 extern unsigned long nr_uninterruptible(void);
 extern unsigned long nr_active(void);
 extern unsigned long nr_iowait(void);
-extern unsigned long weighted_cpuload(const int cpu);
 
 struct seq_file;
 struct cfs_rq;
@@ -823,23 +822,6 @@ extern int arch_reinit_sched_domains(void);
 
 #endif	/* CONFIG_SMP */
 
-/*
- * A runqueue laden with a single nice 0 task scores a weighted_cpuload of
- * SCHED_LOAD_SCALE. This function returns 1 if any cpu is laden with a
- * task of nice 0 or enough lower priority tasks to bring up the
- * weighted_cpuload
- */
-static inline int above_background_load(void)
-{
-	unsigned long cpu;
-
-	for_each_online_cpu(cpu) {
-		if (weighted_cpuload(cpu) >= SCHED_LOAD_SCALE)
-			return 1;
-	}
-	return 0;
-}
-
 struct io_context;			/* See blkdev.h */
 #define NGROUPS_SMALL		32
 #define NGROUPS_PER_BLOCK	((unsigned int)(PAGE_SIZE / sizeof(gid_t)))
-- 
cgit v1.2.3


From c7aceaba042702538b23cf4e0de1b2891ad8e671 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Thu, 15 May 2008 12:09:15 +0100
Subject: sched: reorder task_struct to reduce padding on 64bit builds

This patch removes 24 bytes of padding and allows 1 extra object per
slab on my fedora based config.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index dc36c3aea018..ea2857b99596 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1021,6 +1021,7 @@ struct task_struct {
 #endif
 
 	int prio, static_prio, normal_prio;
+	unsigned int rt_priority;
 	const struct sched_class *sched_class;
 	struct sched_entity se;
 	struct sched_rt_entity rt;
@@ -1104,7 +1105,6 @@ struct task_struct {
 	int __user *set_child_tid;		/* CLONE_CHILD_SETTID */
 	int __user *clear_child_tid;		/* CLONE_CHILD_CLEARTID */
 
-	unsigned int rt_priority;
 	cputime_t utime, stime, utimescaled, stimescaled;
 	cputime_t gtime;
 	cputime_t prev_utime, prev_stime;
@@ -1123,12 +1123,12 @@ struct task_struct {
 	gid_t gid,egid,sgid,fsgid;
 	struct group_info *group_info;
 	kernel_cap_t   cap_effective, cap_inheritable, cap_permitted, cap_bset;
-	unsigned securebits;
 	struct user_struct *user;
+	unsigned securebits;
 #ifdef CONFIG_KEYS
+	unsigned char jit_keyring;	/* default keyring to attach requested keys to */
 	struct key *request_key_auth;	/* assumed request_key authority */
 	struct key *thread_keyring;	/* keyring private to this thread */
-	unsigned char jit_keyring;	/* default keyring to attach requested keys to */
 #endif
 	char comm[TASK_COMM_LEN]; /* executable name excluding path
 				     - access with [gs]et_task_comm (which lock
@@ -1215,8 +1215,8 @@ struct task_struct {
 # define MAX_LOCK_DEPTH 48UL
 	u64 curr_chain_key;
 	int lockdep_depth;
-	struct held_lock held_locks[MAX_LOCK_DEPTH];
 	unsigned int lockdep_recursion;
+	struct held_lock held_locks[MAX_LOCK_DEPTH];
 #endif
 
 /* journalling filesystem info */
@@ -1244,10 +1244,6 @@ struct task_struct {
 	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
 	cputime_t acct_stimexpd;/* stime since last update */
 #endif
-#ifdef CONFIG_NUMA
-  	struct mempolicy *mempolicy;
-	short il_next;
-#endif
 #ifdef CONFIG_CPUSETS
 	nodemask_t mems_allowed;
 	int cpuset_mems_generation;
@@ -1266,6 +1262,10 @@ struct task_struct {
 #endif
 	struct list_head pi_state_list;
 	struct futex_pi_state *pi_state_cache;
+#endif
+#ifdef CONFIG_NUMA
+	struct mempolicy *mempolicy;
+	short il_next;
 #endif
 	atomic_t fs_excl;	/* holding fs exclusive resources */
 	struct rcu_head rcu;
-- 
cgit v1.2.3


From 1f11eb6a8bc92536d9e93ead48fa3ffbd1478571 Mon Sep 17 00:00:00 2001
From: Gregory Haskins <ghaskins@novell.com>
Date: Wed, 4 Jun 2008 15:04:05 -0400
Subject: sched: fix cpupri hotplug support

The RT folks over at RedHat found an issue w.r.t. hotplug support which
was traced to problems with the cpupri infrastructure in the scheduler:

https://bugzilla.redhat.com/show_bug.cgi?id=449676

This bug affects 23-rt12+, 24-rtX, 25-rtX, and sched-devel.  This patch
applies to 25.4-rt4, though it should trivially apply to most cpupri enabled
kernels mentioned above.

It turned out that the issue was that offline cpus could get inadvertently
registered with cpupri so that they were erroneously selected during
migration decisions.  The end result would be an OOPS as the offline cpu
had tasks routed to it.

This patch generalizes the old join/leave domain interface into an
online/offline interface, and adjusts the root-domain/hotplug code to
utilize it.

I was able to easily reproduce the issue prior to this patch, and am no
longer able to reproduce it after this patch.  I can offline cpus
indefinately and everything seems to be in working order.

Thanks to Arnaldo (acme), Thomas, and Peter for doing the legwork to point
me in the right direction.  Also thank you to Peter for reviewing the
early iterations of this patch.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/sched.h |  4 ++--
 kernel/sched.c        | 54 ++++++++++++++++++++++++++++++++++++++-------------
 kernel/sched_rt.c     | 24 +++++++++++++++++------
 3 files changed, 60 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index ea2857b99596..d25acf600a32 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -903,8 +903,8 @@ struct sched_class {
 	void (*set_cpus_allowed)(struct task_struct *p,
 				 const cpumask_t *newmask);
 
-	void (*join_domain)(struct rq *rq);
-	void (*leave_domain)(struct rq *rq);
+	void (*rq_online)(struct rq *rq);
+	void (*rq_offline)(struct rq *rq);
 
 	void (*switched_from) (struct rq *this_rq, struct task_struct *task,
 			       int running);
diff --git a/kernel/sched.c b/kernel/sched.c
index dc0be113f41d..f0ed81b71282 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -529,6 +529,7 @@ struct rq {
 	int push_cpu;
 	/* cpu of this runqueue: */
 	int cpu;
+	int online;
 
 	struct task_struct *migration_thread;
 	struct list_head migration_queue;
@@ -1498,6 +1499,8 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares)
 #endif
 
 #define sched_class_highest (&rt_sched_class)
+#define for_each_class(class) \
+   for (class = sched_class_highest; class; class = class->next)
 
 static inline void inc_load(struct rq *rq, const struct task_struct *p)
 {
@@ -6065,6 +6068,36 @@ static void unregister_sched_domain_sysctl(void)
 }
 #endif
 
+static void set_rq_online(struct rq *rq)
+{
+	if (!rq->online) {
+		const struct sched_class *class;
+
+		cpu_set(rq->cpu, rq->rd->online);
+		rq->online = 1;
+
+		for_each_class(class) {
+			if (class->rq_online)
+				class->rq_online(rq);
+		}
+	}
+}
+
+static void set_rq_offline(struct rq *rq)
+{
+	if (rq->online) {
+		const struct sched_class *class;
+
+		for_each_class(class) {
+			if (class->rq_offline)
+				class->rq_offline(rq);
+		}
+
+		cpu_clear(rq->cpu, rq->rd->online);
+		rq->online = 0;
+	}
+}
+
 /*
  * migration_call - callback that gets triggered when a CPU is added.
  * Here we can start up the necessary migration thread for the new CPU.
@@ -6102,7 +6135,8 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		spin_lock_irqsave(&rq->lock, flags);
 		if (rq->rd) {
 			BUG_ON(!cpu_isset(cpu, rq->rd->span));
-			cpu_set(cpu, rq->rd->online);
+
+			set_rq_online(rq);
 		}
 		spin_unlock_irqrestore(&rq->lock, flags);
 		break;
@@ -6163,7 +6197,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		spin_lock_irqsave(&rq->lock, flags);
 		if (rq->rd) {
 			BUG_ON(!cpu_isset(cpu, rq->rd->span));
-			cpu_clear(cpu, rq->rd->online);
+			set_rq_offline(rq);
 		}
 		spin_unlock_irqrestore(&rq->lock, flags);
 		break;
@@ -6385,20 +6419,16 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 static void rq_attach_root(struct rq *rq, struct root_domain *rd)
 {
 	unsigned long flags;
-	const struct sched_class *class;
 
 	spin_lock_irqsave(&rq->lock, flags);
 
 	if (rq->rd) {
 		struct root_domain *old_rd = rq->rd;
 
-		for (class = sched_class_highest; class; class = class->next) {
-			if (class->leave_domain)
-				class->leave_domain(rq);
-		}
+		if (cpu_isset(rq->cpu, old_rd->online))
+			set_rq_offline(rq);
 
 		cpu_clear(rq->cpu, old_rd->span);
-		cpu_clear(rq->cpu, old_rd->online);
 
 		if (atomic_dec_and_test(&old_rd->refcount))
 			kfree(old_rd);
@@ -6409,12 +6439,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd)
 
 	cpu_set(rq->cpu, rd->span);
 	if (cpu_isset(rq->cpu, cpu_online_map))
-		cpu_set(rq->cpu, rd->online);
-
-	for (class = sched_class_highest; class; class = class->next) {
-		if (class->join_domain)
-			class->join_domain(rq);
-	}
+		set_rq_online(rq);
 
 	spin_unlock_irqrestore(&rq->lock, flags);
 }
@@ -7824,6 +7849,7 @@ void __init sched_init(void)
 		rq->next_balance = jiffies;
 		rq->push_cpu = 0;
 		rq->cpu = i;
+		rq->online = 0;
 		rq->migration_thread = NULL;
 		INIT_LIST_HEAD(&rq->migration_queue);
 		rq_attach_root(rq, &def_root_domain);
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 44b06d75416e..e4821593d4de 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -12,6 +12,9 @@ static inline int rt_overloaded(struct rq *rq)
 
 static inline void rt_set_overload(struct rq *rq)
 {
+	if (!rq->online)
+		return;
+
 	cpu_set(rq->cpu, rq->rd->rto_mask);
 	/*
 	 * Make sure the mask is visible before we set
@@ -26,6 +29,9 @@ static inline void rt_set_overload(struct rq *rq)
 
 static inline void rt_clear_overload(struct rq *rq)
 {
+	if (!rq->online)
+		return;
+
 	/* the order here really doesn't matter */
 	atomic_dec(&rq->rd->rto_count);
 	cpu_clear(rq->cpu, rq->rd->rto_mask);
@@ -394,7 +400,10 @@ void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 	if (rt_se_prio(rt_se) < rt_rq->highest_prio) {
 		struct rq *rq = rq_of_rt_rq(rt_rq);
 		rt_rq->highest_prio = rt_se_prio(rt_se);
-		cpupri_set(&rq->rd->cpupri, rq->cpu, rt_se_prio(rt_se));
+
+		if (rq->online)
+			cpupri_set(&rq->rd->cpupri, rq->cpu,
+				   rt_se_prio(rt_se));
 	}
 #endif
 #ifdef CONFIG_SMP
@@ -448,7 +457,10 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 
 	if (rt_rq->highest_prio != highest_prio) {
 		struct rq *rq = rq_of_rt_rq(rt_rq);
-		cpupri_set(&rq->rd->cpupri, rq->cpu, rt_rq->highest_prio);
+
+		if (rq->online)
+			cpupri_set(&rq->rd->cpupri, rq->cpu,
+				   rt_rq->highest_prio);
 	}
 
 	update_rt_migration(rq_of_rt_rq(rt_rq));
@@ -1154,7 +1166,7 @@ static void set_cpus_allowed_rt(struct task_struct *p,
 }
 
 /* Assumes rq->lock is held */
-static void join_domain_rt(struct rq *rq)
+static void rq_online_rt(struct rq *rq)
 {
 	if (rq->rt.overloaded)
 		rt_set_overload(rq);
@@ -1163,7 +1175,7 @@ static void join_domain_rt(struct rq *rq)
 }
 
 /* Assumes rq->lock is held */
-static void leave_domain_rt(struct rq *rq)
+static void rq_offline_rt(struct rq *rq)
 {
 	if (rq->rt.overloaded)
 		rt_clear_overload(rq);
@@ -1331,8 +1343,8 @@ static const struct sched_class rt_sched_class = {
 	.load_balance		= load_balance_rt,
 	.move_one_task		= move_one_task_rt,
 	.set_cpus_allowed       = set_cpus_allowed_rt,
-	.join_domain            = join_domain_rt,
-	.leave_domain           = leave_domain_rt,
+	.rq_online              = rq_online_rt,
+	.rq_offline             = rq_offline_rt,
 	.pre_schedule		= pre_schedule_rt,
 	.post_schedule		= post_schedule_rt,
 	.task_wake_up		= task_wake_up_rt,
-- 
cgit v1.2.3


From 93f65158723ceb7078ee9a0fd4830c0de00f4b9e Mon Sep 17 00:00:00 2001
From: Kuo-lang Tseng <kuo-lang.tseng@intel.com>
Date: Mon, 9 Jun 2008 15:55:45 -0700
Subject: netfilter: ebtables: add IPv6 support

It implements matching functions for IPv6 address & traffic class
(merged from the patch sent by Jan Engelhardt [jengelh@computergmbh.de]
http://marc.info/?l=netfilter-devel&m=120182168424052&w=2), protocol,
and layer-4 port id. Corresponding watcher logging function is also
added for IPv6.

Signed-off-by: Kuo-lang Tseng <kuo-lang.tseng@intel.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_bridge/ebt_ip6.h |  40 +++++++++
 include/linux/netfilter_bridge/ebt_log.h |   3 +-
 net/bridge/netfilter/Kconfig             |   9 ++
 net/bridge/netfilter/Makefile            |   1 +
 net/bridge/netfilter/ebt_ip6.c           | 144 +++++++++++++++++++++++++++++++
 net/bridge/netfilter/ebt_log.c           |  64 ++++++++++----
 6 files changed, 244 insertions(+), 17 deletions(-)
 create mode 100644 include/linux/netfilter_bridge/ebt_ip6.h
 create mode 100644 net/bridge/netfilter/ebt_ip6.c

(limited to 'include/linux')

diff --git a/include/linux/netfilter_bridge/ebt_ip6.h b/include/linux/netfilter_bridge/ebt_ip6.h
new file mode 100644
index 000000000000..2273c3ae33ca
--- /dev/null
+++ b/include/linux/netfilter_bridge/ebt_ip6.h
@@ -0,0 +1,40 @@
+/*
+ *  ebt_ip6
+ *
+ *	Authors:
+ * Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
+ * Manohar Castelino <manohar.r.castelino@intel.com>
+ *
+ *  Jan 11, 2008
+ *
+ */
+
+#ifndef __LINUX_BRIDGE_EBT_IP6_H
+#define __LINUX_BRIDGE_EBT_IP6_H
+
+#define EBT_IP6_SOURCE 0x01
+#define EBT_IP6_DEST 0x02
+#define EBT_IP6_TCLASS 0x04
+#define EBT_IP6_PROTO 0x08
+#define EBT_IP6_SPORT 0x10
+#define EBT_IP6_DPORT 0x20
+#define EBT_IP6_MASK (EBT_IP6_SOURCE | EBT_IP6_DEST | EBT_IP6_TCLASS |\
+		      EBT_IP6_PROTO | EBT_IP6_SPORT | EBT_IP6_DPORT)
+#define EBT_IP6_MATCH "ip6"
+
+/* the same values are used for the invflags */
+struct ebt_ip6_info
+{
+	struct in6_addr saddr;
+	struct in6_addr daddr;
+	struct in6_addr smsk;
+	struct in6_addr dmsk;
+	uint8_t  tclass;
+	uint8_t  protocol;
+	uint8_t  bitmask;
+	uint8_t  invflags;
+	uint16_t sport[2];
+	uint16_t dport[2];
+};
+
+#endif
diff --git a/include/linux/netfilter_bridge/ebt_log.h b/include/linux/netfilter_bridge/ebt_log.h
index 96e231ae7554..b76e653157e5 100644
--- a/include/linux/netfilter_bridge/ebt_log.h
+++ b/include/linux/netfilter_bridge/ebt_log.h
@@ -4,7 +4,8 @@
 #define EBT_LOG_IP 0x01 /* if the frame is made by ip, log the ip information */
 #define EBT_LOG_ARP 0x02
 #define EBT_LOG_NFLOG 0x04
-#define EBT_LOG_MASK (EBT_LOG_IP | EBT_LOG_ARP)
+#define EBT_LOG_IP6 0x08
+#define EBT_LOG_MASK (EBT_LOG_IP | EBT_LOG_ARP | EBT_LOG_IP6)
 #define EBT_LOG_PREFIX_SIZE 30
 #define EBT_LOG_WATCHER "log"
 
diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig
index 7beeefa0f9c0..fb684c2ff8b6 100644
--- a/net/bridge/netfilter/Kconfig
+++ b/net/bridge/netfilter/Kconfig
@@ -83,6 +83,15 @@ config BRIDGE_EBT_IP
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config BRIDGE_EBT_IP6
+	tristate "ebt: IP6 filter support"
+	depends on BRIDGE_NF_EBTABLES
+	help
+	  This option adds the IP6 match, which allows basic IPV6 header field
+	  filtering.
+
+	  To compile it as a module, choose M here.  If unsure, say N.
+
 config BRIDGE_EBT_LIMIT
 	tristate "ebt: limit match support"
 	depends on BRIDGE_NF_EBTABLES
diff --git a/net/bridge/netfilter/Makefile b/net/bridge/netfilter/Makefile
index 83715d73a503..dd960645b413 100644
--- a/net/bridge/netfilter/Makefile
+++ b/net/bridge/netfilter/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_BRIDGE_EBT_802_3) += ebt_802_3.o
 obj-$(CONFIG_BRIDGE_EBT_AMONG) += ebt_among.o
 obj-$(CONFIG_BRIDGE_EBT_ARP) += ebt_arp.o
 obj-$(CONFIG_BRIDGE_EBT_IP) += ebt_ip.o
+obj-$(CONFIG_BRIDGE_EBT_IP) += ebt_ip6.o
 obj-$(CONFIG_BRIDGE_EBT_LIMIT) += ebt_limit.o
 obj-$(CONFIG_BRIDGE_EBT_MARK) += ebt_mark_m.o
 obj-$(CONFIG_BRIDGE_EBT_PKTTYPE) += ebt_pkttype.o
diff --git a/net/bridge/netfilter/ebt_ip6.c b/net/bridge/netfilter/ebt_ip6.c
new file mode 100644
index 000000000000..36efb3a75249
--- /dev/null
+++ b/net/bridge/netfilter/ebt_ip6.c
@@ -0,0 +1,144 @@
+/*
+ *  ebt_ip6
+ *
+ *	Authors:
+ *	Manohar Castelino <manohar.r.castelino@intel.com>
+ *	Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
+ *	Jan Engelhardt <jengelh@computergmbh.de>
+ *
+ * Summary:
+ * This is just a modification of the IPv4 code written by
+ * Bart De Schuymer <bdschuym@pandora.be>
+ * with the changes required to support IPv6
+ *
+ *  Jan, 2008
+ */
+
+#include <linux/netfilter_bridge/ebtables.h>
+#include <linux/netfilter_bridge/ebt_ip6.h>
+#include <linux/ipv6.h>
+#include <net/ipv6.h>
+#include <linux/in.h>
+#include <linux/module.h>
+#include <net/dsfield.h>
+
+struct tcpudphdr {
+	__be16 src;
+	__be16 dst;
+};
+
+static int ebt_filter_ip6(const struct sk_buff *skb,
+   const struct net_device *in,
+   const struct net_device *out, const void *data,
+   unsigned int datalen)
+{
+	const struct ebt_ip6_info *info = (struct ebt_ip6_info *)data;
+	const struct ipv6hdr *ih6;
+	struct ipv6hdr _ip6h;
+	const struct tcpudphdr *pptr;
+	struct tcpudphdr _ports;
+	struct in6_addr tmp_addr;
+	int i;
+
+	ih6 = skb_header_pointer(skb, 0, sizeof(_ip6h), &_ip6h);
+	if (ih6 == NULL)
+		return EBT_NOMATCH;
+	if (info->bitmask & EBT_IP6_TCLASS &&
+	   FWINV(info->tclass != ipv6_get_dsfield(ih6), EBT_IP6_TCLASS))
+		return EBT_NOMATCH;
+	for (i = 0; i < 4; i++)
+		tmp_addr.in6_u.u6_addr32[i] = ih6->saddr.in6_u.u6_addr32[i] &
+			info->smsk.in6_u.u6_addr32[i];
+	if (info->bitmask & EBT_IP6_SOURCE &&
+		FWINV((ipv6_addr_cmp(&tmp_addr, &info->saddr) != 0),
+			EBT_IP6_SOURCE))
+		return EBT_NOMATCH;
+	for (i = 0; i < 4; i++)
+		tmp_addr.in6_u.u6_addr32[i] = ih6->daddr.in6_u.u6_addr32[i] &
+			info->dmsk.in6_u.u6_addr32[i];
+	if (info->bitmask & EBT_IP6_DEST &&
+	   FWINV((ipv6_addr_cmp(&tmp_addr, &info->daddr) != 0), EBT_IP6_DEST))
+		return EBT_NOMATCH;
+	if (info->bitmask & EBT_IP6_PROTO) {
+		uint8_t nexthdr = ih6->nexthdr;
+		int offset_ph;
+
+		offset_ph = ipv6_skip_exthdr(skb, sizeof(_ip6h), &nexthdr);
+		if (offset_ph == -1)
+			return EBT_NOMATCH;
+		if (FWINV(info->protocol != nexthdr, EBT_IP6_PROTO))
+			return EBT_NOMATCH;
+		if (!(info->bitmask & EBT_IP6_DPORT) &&
+		    !(info->bitmask & EBT_IP6_SPORT))
+			return EBT_MATCH;
+		pptr = skb_header_pointer(skb, offset_ph, sizeof(_ports),
+					  &_ports);
+		if (pptr == NULL)
+			return EBT_NOMATCH;
+		if (info->bitmask & EBT_IP6_DPORT) {
+			u32 dst = ntohs(pptr->dst);
+			if (FWINV(dst < info->dport[0] ||
+				  dst > info->dport[1], EBT_IP6_DPORT))
+				return EBT_NOMATCH;
+		}
+		if (info->bitmask & EBT_IP6_SPORT) {
+			u32 src = ntohs(pptr->src);
+			if (FWINV(src < info->sport[0] ||
+				  src > info->sport[1], EBT_IP6_SPORT))
+			return EBT_NOMATCH;
+		}
+		return EBT_MATCH;
+	}
+	return EBT_MATCH;
+}
+
+static int ebt_ip6_check(const char *tablename, unsigned int hookmask,
+   const struct ebt_entry *e, void *data, unsigned int datalen)
+{
+	struct ebt_ip6_info *info = (struct ebt_ip6_info *)data;
+
+	if (datalen != EBT_ALIGN(sizeof(struct ebt_ip6_info)))
+		return -EINVAL;
+	if (e->ethproto != htons(ETH_P_IPV6) || e->invflags & EBT_IPROTO)
+		return -EINVAL;
+	if (info->bitmask & ~EBT_IP6_MASK || info->invflags & ~EBT_IP6_MASK)
+		return -EINVAL;
+	if (info->bitmask & (EBT_IP6_DPORT | EBT_IP6_SPORT)) {
+		if (info->invflags & EBT_IP6_PROTO)
+			return -EINVAL;
+		if (info->protocol != IPPROTO_TCP &&
+		    info->protocol != IPPROTO_UDP &&
+		    info->protocol != IPPROTO_UDPLITE &&
+		    info->protocol != IPPROTO_SCTP &&
+		    info->protocol != IPPROTO_DCCP)
+			 return -EINVAL;
+	}
+	if (info->bitmask & EBT_IP6_DPORT && info->dport[0] > info->dport[1])
+		return -EINVAL;
+	if (info->bitmask & EBT_IP6_SPORT && info->sport[0] > info->sport[1])
+		return -EINVAL;
+	return 0;
+}
+
+static struct ebt_match filter_ip6 =
+{
+	.name		= EBT_IP6_MATCH,
+	.match		= ebt_filter_ip6,
+	.check		= ebt_ip6_check,
+	.me		= THIS_MODULE,
+};
+
+static int __init ebt_ip6_init(void)
+{
+	return ebt_register_match(&filter_ip6);
+}
+
+static void __exit ebt_ip6_fini(void)
+{
+	ebt_unregister_match(&filter_ip6);
+}
+
+module_init(ebt_ip6_init);
+module_exit(ebt_ip6_fini);
+MODULE_DESCRIPTION("Ebtables: IPv6 protocol packet match");
+MODULE_LICENSE("GPL");
diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c
index 0b209e4aad0a..c883ec8a28b4 100644
--- a/net/bridge/netfilter/ebt_log.c
+++ b/net/bridge/netfilter/ebt_log.c
@@ -18,6 +18,9 @@
 #include <linux/if_arp.h>
 #include <linux/spinlock.h>
 #include <net/netfilter/nf_log.h>
+#include <linux/ipv6.h>
+#include <net/ipv6.h>
+#include <linux/in6.h>
 
 static DEFINE_SPINLOCK(ebt_log_lock);
 
@@ -58,6 +61,27 @@ static void print_MAC(const unsigned char *p)
 		printk("%02x%c", *p, i == ETH_ALEN - 1 ? ' ':':');
 }
 
+static void
+print_ports(const struct sk_buff *skb, uint8_t protocol, int offset)
+{
+	if (protocol == IPPROTO_TCP ||
+	    protocol == IPPROTO_UDP ||
+	    protocol == IPPROTO_UDPLITE ||
+	    protocol == IPPROTO_SCTP ||
+	    protocol == IPPROTO_DCCP) {
+		const struct tcpudphdr *pptr;
+		struct tcpudphdr _ports;
+
+		pptr = skb_header_pointer(skb, offset,
+					  sizeof(_ports), &_ports);
+		if (pptr == NULL) {
+			printk(" INCOMPLETE TCP/UDP header");
+			return;
+		}
+		printk(" SPT=%u DPT=%u", ntohs(pptr->src), ntohs(pptr->dst));
+	}
+}
+
 #define myNIPQUAD(a) a[0], a[1], a[2], a[3]
 static void
 ebt_log_packet(unsigned int pf, unsigned int hooknum,
@@ -95,23 +119,31 @@ ebt_log_packet(unsigned int pf, unsigned int hooknum,
 		printk(" IP SRC=%u.%u.%u.%u IP DST=%u.%u.%u.%u, IP "
 		       "tos=0x%02X, IP proto=%d", NIPQUAD(ih->saddr),
 		       NIPQUAD(ih->daddr), ih->tos, ih->protocol);
-		if (ih->protocol == IPPROTO_TCP ||
-		    ih->protocol == IPPROTO_UDP ||
-		    ih->protocol == IPPROTO_UDPLITE ||
-		    ih->protocol == IPPROTO_SCTP ||
-		    ih->protocol == IPPROTO_DCCP) {
-			const struct tcpudphdr *pptr;
-			struct tcpudphdr _ports;
-
-			pptr = skb_header_pointer(skb, ih->ihl*4,
-						  sizeof(_ports), &_ports);
-			if (pptr == NULL) {
-				printk(" INCOMPLETE TCP/UDP header");
-				goto out;
-			}
-			printk(" SPT=%u DPT=%u", ntohs(pptr->src),
-			   ntohs(pptr->dst));
+		print_ports(skb, ih->protocol, ih->ihl*4);
+		goto out;
+	}
+
+	if ((bitmask & EBT_LOG_IP6) && eth_hdr(skb)->h_proto ==
+	   htons(ETH_P_IPV6)) {
+		const struct ipv6hdr *ih;
+		struct ipv6hdr _iph;
+		uint8_t nexthdr;
+		int offset_ph;
+
+		ih = skb_header_pointer(skb, 0, sizeof(_iph), &_iph);
+		if (ih == NULL) {
+			printk(" INCOMPLETE IPv6 header");
+			goto out;
 		}
+		printk(" IPv6 SRC=%x:%x:%x:%x:%x:%x:%x:%x "
+		       "IPv6 DST=%x:%x:%x:%x:%x:%x:%x:%x, IPv6 "
+		       "priority=0x%01X, Next Header=%d", NIP6(ih->saddr),
+		       NIP6(ih->daddr), ih->priority, ih->nexthdr);
+		nexthdr = ih->nexthdr;
+		offset_ph = ipv6_skip_exthdr(skb, sizeof(_iph), &nexthdr);
+		if (offset_ph == -1)
+			goto out;
+		print_ports(skb, nexthdr, offset_ph);
 		goto out;
 	}
 
-- 
cgit v1.2.3


From a258860e01b80e8f554a4ab1a6c95e6042eb8b73 Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 9 Jun 2008 15:56:39 -0700
Subject: netfilter: ctnetlink: add full support for SCTP to ctnetlink

This patch adds full support for SCTP to ctnetlink. This includes three
new attributes: state, original vtag and reply vtag.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nfnetlink_conntrack.h | 10 ++++
 net/netfilter/nf_conntrack_proto_sctp.c       | 80 +++++++++++++++++++++++++++
 2 files changed, 90 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index 0a383ac083cb..759bc043dc65 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -81,6 +81,7 @@ enum ctattr_protoinfo {
 	CTA_PROTOINFO_UNSPEC,
 	CTA_PROTOINFO_TCP,
 	CTA_PROTOINFO_DCCP,
+	CTA_PROTOINFO_SCTP,
 	__CTA_PROTOINFO_MAX
 };
 #define CTA_PROTOINFO_MAX (__CTA_PROTOINFO_MAX - 1)
@@ -103,6 +104,15 @@ enum ctattr_protoinfo_dccp {
 };
 #define CTA_PROTOINFO_DCCP_MAX (__CTA_PROTOINFO_DCCP_MAX - 1)
 
+enum ctattr_protoinfo_sctp {
+	CTA_PROTOINFO_SCTP_UNSPEC,
+	CTA_PROTOINFO_SCTP_STATE,
+	CTA_PROTOINFO_SCTP_VTAG_ORIGINAL,
+	CTA_PROTOINFO_SCTP_VTAG_REPLY,
+	__CTA_PROTOINFO_SCTP_MAX
+};
+#define CTA_PROTOINFO_SCTP_MAX (__CTA_PROTOINFO_SCTP_MAX - 1)
+
 enum ctattr_counters {
 	CTA_COUNTERS_UNSPEC,
 	CTA_COUNTERS_PACKETS,		/* old 64bit counters */
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index cbf2e27a22b2..41183a4d2d62 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -463,6 +463,82 @@ static bool sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 	return true;
 }
 
+#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+
+#include <linux/netfilter/nfnetlink.h>
+#include <linux/netfilter/nfnetlink_conntrack.h>
+
+static int sctp_to_nlattr(struct sk_buff *skb, struct nlattr *nla,
+			  const struct nf_conn *ct)
+{
+	struct nlattr *nest_parms;
+
+	read_lock_bh(&sctp_lock);
+	nest_parms = nla_nest_start(skb, CTA_PROTOINFO_SCTP | NLA_F_NESTED);
+	if (!nest_parms)
+		goto nla_put_failure;
+
+	NLA_PUT_U8(skb, CTA_PROTOINFO_SCTP_STATE, ct->proto.sctp.state);
+
+	NLA_PUT_BE32(skb,
+		     CTA_PROTOINFO_SCTP_VTAG_ORIGINAL,
+		     htonl(ct->proto.sctp.vtag[IP_CT_DIR_ORIGINAL]));
+
+	NLA_PUT_BE32(skb,
+		     CTA_PROTOINFO_SCTP_VTAG_REPLY,
+		     htonl(ct->proto.sctp.vtag[IP_CT_DIR_REPLY]));
+
+	read_unlock_bh(&sctp_lock);
+
+	nla_nest_end(skb, nest_parms);
+
+	return 0;
+
+nla_put_failure:
+	read_unlock_bh(&sctp_lock);
+	return -1;
+}
+
+static const struct nla_policy sctp_nla_policy[CTA_PROTOINFO_SCTP_MAX+1] = {
+	[CTA_PROTOINFO_SCTP_STATE]	    = { .type = NLA_U8 },
+	[CTA_PROTOINFO_SCTP_VTAG_ORIGINAL]  = { .type = NLA_U32 },
+	[CTA_PROTOINFO_SCTP_VTAG_REPLY]     = { .type = NLA_U32 },
+};
+
+static int nlattr_to_sctp(struct nlattr *cda[], struct nf_conn *ct)
+{
+	struct nlattr *attr = cda[CTA_PROTOINFO_SCTP];
+	struct nlattr *tb[CTA_PROTOINFO_SCTP_MAX+1];
+	int err;
+
+	/* updates may not contain the internal protocol info, skip parsing */
+	if (!attr)
+		return 0;
+
+	err = nla_parse_nested(tb,
+			       CTA_PROTOINFO_SCTP_MAX,
+			       attr,
+			       sctp_nla_policy);
+	if (err < 0)
+		return err;
+
+	if (!tb[CTA_PROTOINFO_SCTP_STATE] ||
+	    !tb[CTA_PROTOINFO_SCTP_VTAG_ORIGINAL] ||
+	    !tb[CTA_PROTOINFO_SCTP_VTAG_REPLY])
+		return -EINVAL;
+
+	write_lock_bh(&sctp_lock);
+	ct->proto.sctp.state = nla_get_u8(tb[CTA_PROTOINFO_SCTP_STATE]);
+	ct->proto.sctp.vtag[IP_CT_DIR_ORIGINAL] =
+		ntohl(nla_get_be32(tb[CTA_PROTOINFO_SCTP_VTAG_ORIGINAL]));
+	ct->proto.sctp.vtag[IP_CT_DIR_REPLY] =
+		ntohl(nla_get_be32(tb[CTA_PROTOINFO_SCTP_VTAG_REPLY]));
+	write_unlock_bh(&sctp_lock);
+
+	return 0;
+}
+#endif
+
 #ifdef CONFIG_SYSCTL
 static unsigned int sctp_sysctl_table_users;
 static struct ctl_table_header *sctp_sysctl_header;
@@ -591,6 +667,8 @@ static struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp4 __read_mostly = {
 	.new 			= sctp_new,
 	.me 			= THIS_MODULE,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+	.to_nlattr		= sctp_to_nlattr,
+	.from_nlattr		= nlattr_to_sctp,
 	.tuple_to_nlattr	= nf_ct_port_tuple_to_nlattr,
 	.nlattr_to_tuple	= nf_ct_port_nlattr_to_tuple,
 	.nla_policy		= nf_ct_port_nla_policy,
@@ -617,6 +695,8 @@ static struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp6 __read_mostly = {
 	.new 			= sctp_new,
 	.me 			= THIS_MODULE,
 #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
+	.to_nlattr		= sctp_to_nlattr,
+	.from_nlattr		= nlattr_to_sctp,
 	.tuple_to_nlattr	= nf_ct_port_tuple_to_nlattr,
 	.nlattr_to_tuple	= nf_ct_port_nlattr_to_tuple,
 	.nla_policy		= nf_ct_port_nla_policy,
-- 
cgit v1.2.3


From 560ee653b67074b805f1b661988a72a0e58811a5 Mon Sep 17 00:00:00 2001
From: James Morris <jmorris@namei.org>
Date: Mon, 9 Jun 2008 15:57:24 -0700
Subject: netfilter: ip_tables: add iptables security table for mandatory
 access control rules

The following patch implements a new "security" table for iptables, so
that MAC (SELinux etc.) networking rules can be managed separately to
standard DAC rules.

This is to help with distro integration of the new secmark-based
network controls, per various previous discussions.

The need for a separate table arises from the fact that existing tools
and usage of iptables will likely clash with centralized MAC policy
management.

The SECMARK and CONNSECMARK targets will still be valid in the mangle
table to prevent breakage of existing users.

Signed-off-by: James Morris <jmorris@namei.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4.h        |   1 +
 include/net/netns/ipv4.h              |   1 +
 net/ipv4/netfilter/Kconfig            |  12 +++
 net/ipv4/netfilter/Makefile           |   1 +
 net/ipv4/netfilter/iptable_security.c | 180 ++++++++++++++++++++++++++++++++++
 net/netfilter/xt_CONNSECMARK.c        |  10 +-
 net/netfilter/xt_SECMARK.c            |  10 +-
 7 files changed, 209 insertions(+), 6 deletions(-)
 create mode 100644 net/ipv4/netfilter/iptable_security.c

(limited to 'include/linux')

diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h
index 650318b0c405..29c7727ff0e8 100644
--- a/include/linux/netfilter_ipv4.h
+++ b/include/linux/netfilter_ipv4.h
@@ -60,6 +60,7 @@ enum nf_ip_hook_priorities {
 	NF_IP_PRI_MANGLE = -150,
 	NF_IP_PRI_NAT_DST = -100,
 	NF_IP_PRI_FILTER = 0,
+	NF_IP_PRI_SECURITY = 50,
 	NF_IP_PRI_NAT_SRC = 100,
 	NF_IP_PRI_SELINUX_LAST = 225,
 	NF_IP_PRI_CONNTRACK_CONFIRM = INT_MAX,
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 34ee348a2cf2..6ef90b5fafb3 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -36,6 +36,7 @@ struct netns_ipv4 {
 	struct xt_table		*iptable_mangle;
 	struct xt_table		*iptable_raw;
 	struct xt_table		*arptable_filter;
+	struct xt_table		*iptable_security;
 #endif
 
 	int sysctl_icmp_echo_ignore_all;
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 2767841a8cef..6e251402506e 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -365,6 +365,18 @@ config IP_NF_RAW
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
 
+# security table for MAC policy
+config IP_NF_SECURITY
+	tristate "Security table"
+	depends on IP_NF_IPTABLES
+	depends on SECURITY
+	default m if NETFILTER_ADVANCED=n
+	help
+	  This option adds a `security' table to iptables, for use
+	  with Mandatory Access Control (MAC) policy.
+	 
+	  If unsure, say N.
+
 # ARP tables
 config IP_NF_ARPTABLES
 	tristate "ARP tables support"
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index d9b92fbf5579..3f31291f37ce 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -42,6 +42,7 @@ obj-$(CONFIG_IP_NF_FILTER) += iptable_filter.o
 obj-$(CONFIG_IP_NF_MANGLE) += iptable_mangle.o
 obj-$(CONFIG_NF_NAT) += iptable_nat.o
 obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o
+obj-$(CONFIG_IP_NF_SECURITY) += iptable_security.o
 
 # matches
 obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o
diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c
new file mode 100644
index 000000000000..2b472ac2263a
--- /dev/null
+++ b/net/ipv4/netfilter/iptable_security.c
@@ -0,0 +1,180 @@
+/*
+ * "security" table
+ *
+ * This is for use by Mandatory Access Control (MAC) security models,
+ * which need to be able to manage security policy in separate context
+ * to DAC.
+ *
+ * Based on iptable_mangle.c
+ *
+ * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling
+ * Copyright (C) 2000-2004 Netfilter Core Team <coreteam <at> netfilter.org>
+ * Copyright (C) 2008 Red Hat, Inc., James Morris <jmorris <at> redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/netfilter_ipv4/ip_tables.h>
+#include <net/ip.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("James Morris <jmorris <at> redhat.com>");
+MODULE_DESCRIPTION("iptables security table, for MAC rules");
+
+#define SECURITY_VALID_HOOKS	(1 << NF_INET_LOCAL_IN) | \
+				(1 << NF_INET_FORWARD) | \
+				(1 << NF_INET_LOCAL_OUT)
+
+static struct
+{
+	struct ipt_replace repl;
+	struct ipt_standard entries[3];
+	struct ipt_error term;
+} initial_table __initdata = {
+	.repl = {
+		.name = "security",
+		.valid_hooks = SECURITY_VALID_HOOKS,
+		.num_entries = 4,
+		.size = sizeof(struct ipt_standard) * 3 + sizeof(struct ipt_error),
+		.hook_entry = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 2,
+		},
+		.underflow = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ipt_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ipt_standard) * 2,
+		},
+	},
+	.entries = {
+		IPT_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_IN */
+		IPT_STANDARD_INIT(NF_ACCEPT),	/* FORWARD */
+		IPT_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_OUT */
+	},
+	.term = IPT_ERROR_INIT,			/* ERROR */
+};
+
+static struct xt_table security_table = {
+	.name		= "security",
+	.valid_hooks	= SECURITY_VALID_HOOKS,
+	.lock		= __RW_LOCK_UNLOCKED(security_table.lock),
+	.me		= THIS_MODULE,
+	.af		= AF_INET,
+};
+
+static unsigned int
+ipt_local_in_hook(unsigned int hook,
+		  struct sk_buff *skb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_in_net(in, out)->ipv4.iptable_security);
+}
+
+static unsigned int
+ipt_forward_hook(unsigned int hook,
+		 struct sk_buff *skb,
+		 const struct net_device *in,
+		 const struct net_device *out,
+		 int (*okfn)(struct sk_buff *))
+{
+	return ipt_do_table(skb, hook, in, out,
+			    nf_forward_net(in, out)->ipv4.iptable_security);
+}
+
+static unsigned int
+ipt_local_out_hook(unsigned int hook,
+		   struct sk_buff *skb,
+		   const struct net_device *in,
+		   const struct net_device *out,
+		   int (*okfn)(struct sk_buff *))
+{
+	/* Somebody is playing with raw sockets. */
+	if (skb->len < sizeof(struct iphdr)
+	    || ip_hdrlen(skb) < sizeof(struct iphdr)) {
+		if (net_ratelimit())
+			printk(KERN_INFO "iptable_security: ignoring short "
+			       "SOCK_RAW packet.\n");
+		return NF_ACCEPT;
+	}
+	return ipt_do_table(skb, hook, in, out,
+			    nf_local_out_net(in, out)->ipv4.iptable_security);
+}
+
+static struct nf_hook_ops ipt_ops[] __read_mostly = {
+	{
+		.hook		= ipt_local_in_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET,
+		.hooknum	= NF_INET_LOCAL_IN,
+		.priority	= NF_IP_PRI_SECURITY,
+	},
+	{
+		.hook		= ipt_forward_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET,
+		.hooknum	= NF_INET_FORWARD,
+		.priority	= NF_IP_PRI_SECURITY,
+	},
+	{
+		.hook		= ipt_local_out_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET,
+		.hooknum	= NF_INET_LOCAL_OUT,
+		.priority	= NF_IP_PRI_SECURITY,
+	},
+};
+
+static int __net_init iptable_security_net_init(struct net *net)
+{
+	net->ipv4.iptable_security =
+		ipt_register_table(net, &security_table, &initial_table.repl);
+
+	if (IS_ERR(net->ipv4.iptable_security))
+		return PTR_ERR(net->ipv4.iptable_security);
+
+	return 0;
+}
+
+static void __net_exit iptable_security_net_exit(struct net *net)
+{
+	ipt_unregister_table(net->ipv4.iptable_security);
+}
+
+static struct pernet_operations iptable_security_net_ops = {
+	.init = iptable_security_net_init,
+	.exit = iptable_security_net_exit,
+};
+
+static int __init iptable_security_init(void)
+{
+	int ret;
+
+	ret = register_pernet_subsys(&iptable_security_net_ops);
+        if (ret < 0)
+		return ret;
+
+	ret = nf_register_hooks(ipt_ops, ARRAY_SIZE(ipt_ops));
+	if (ret < 0)
+		goto cleanup_table;
+
+	return ret;
+
+cleanup_table:
+	unregister_pernet_subsys(&iptable_security_net_ops);
+	return ret;
+}
+
+static void __exit iptable_security_fini(void)
+{
+	nf_unregister_hooks(ipt_ops, ARRAY_SIZE(ipt_ops));
+	unregister_pernet_subsys(&iptable_security_net_ops);
+}
+
+module_init(iptable_security_init);
+module_exit(iptable_security_fini);
diff --git a/net/netfilter/xt_CONNSECMARK.c b/net/netfilter/xt_CONNSECMARK.c
index 211189eb2b67..76ca1f2421eb 100644
--- a/net/netfilter/xt_CONNSECMARK.c
+++ b/net/netfilter/xt_CONNSECMARK.c
@@ -8,7 +8,7 @@
  *   Copyright (C) 2002,2004 MARA Systems AB <http://www.marasystems.com>
  *    by Henrik Nordstrom <hno@marasystems.com>
  *
- * (C) 2006 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ * (C) 2006,2008 Red Hat, Inc., James Morris <jmorris@redhat.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -94,6 +94,12 @@ connsecmark_tg_check(const char *tablename, const void *entry,
 {
 	const struct xt_connsecmark_target_info *info = targinfo;
 
+	if (strcmp(tablename, "mangle") && strcmp(tablename, "security")) {
+		printk(KERN_INFO PFX "target only valid in the \'mangle\' "
+		       "or \'security\' tables, not \'%s\'.\n", tablename);
+		return false;
+	}
+
 	switch (info->mode) {
 	case CONNSECMARK_SAVE:
 	case CONNSECMARK_RESTORE:
@@ -126,7 +132,6 @@ static struct xt_target connsecmark_tg_reg[] __read_mostly = {
 		.destroy	= connsecmark_tg_destroy,
 		.target		= connsecmark_tg,
 		.targetsize	= sizeof(struct xt_connsecmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 	{
@@ -136,7 +141,6 @@ static struct xt_target connsecmark_tg_reg[] __read_mostly = {
 		.destroy	= connsecmark_tg_destroy,
 		.target		= connsecmark_tg,
 		.targetsize	= sizeof(struct xt_connsecmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 };
diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c
index c0284856ccd4..94f87ee7552b 100644
--- a/net/netfilter/xt_SECMARK.c
+++ b/net/netfilter/xt_SECMARK.c
@@ -5,7 +5,7 @@
  * Based on the nfmark match by:
  * (C) 1999-2001 Marc Boucher <marc@mbsi.ca>
  *
- * (C) 2006 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ * (C) 2006,2008 Red Hat, Inc., James Morris <jmorris@redhat.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -89,6 +89,12 @@ secmark_tg_check(const char *tablename, const void *entry,
 {
 	struct xt_secmark_target_info *info = targinfo;
 
+	if (strcmp(tablename, "mangle") && strcmp(tablename, "security")) {
+		printk(KERN_INFO PFX "target only valid in the \'mangle\' "
+		       "or \'security\' tables, not \'%s\'.\n", tablename);
+		return false;
+	}
+
 	if (mode && mode != info->mode) {
 		printk(KERN_INFO PFX "mode already set to %hu cannot mix with "
 		       "rules for mode %hu\n", mode, info->mode);
@@ -127,7 +133,6 @@ static struct xt_target secmark_tg_reg[] __read_mostly = {
 		.destroy	= secmark_tg_destroy,
 		.target		= secmark_tg,
 		.targetsize	= sizeof(struct xt_secmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 	{
@@ -137,7 +142,6 @@ static struct xt_target secmark_tg_reg[] __read_mostly = {
 		.destroy	= secmark_tg_destroy,
 		.target		= secmark_tg,
 		.targetsize	= sizeof(struct xt_secmark_target_info),
-		.table		= "mangle",
 		.me		= THIS_MODULE,
 	},
 };
-- 
cgit v1.2.3


From 17e6e59f0a1d7188d783c15dc3ccebd95a0840cd Mon Sep 17 00:00:00 2001
From: James Morris <jmorris@namei.org>
Date: Mon, 9 Jun 2008 15:58:05 -0700
Subject: netfilter: ip6_tables: add ip6tables security table

This is a port of the IPv4 security table for IPv6.

Signed-off-by: James Morris <jmorris@namei.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv6.h         |   1 +
 include/net/netns/ipv6.h               |   1 +
 net/ipv6/netfilter/Kconfig             |  12 +++
 net/ipv6/netfilter/Makefile            |   1 +
 net/ipv6/netfilter/ip6table_security.c | 172 +++++++++++++++++++++++++++++++++
 5 files changed, 187 insertions(+)
 create mode 100644 net/ipv6/netfilter/ip6table_security.c

(limited to 'include/linux')

diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index 3475a65dae9b..fd50988b83ec 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -64,6 +64,7 @@ enum nf_ip6_hook_priorities {
 	NF_IP6_PRI_MANGLE = -150,
 	NF_IP6_PRI_NAT_DST = -100,
 	NF_IP6_PRI_FILTER = 0,
+	NF_IP6_PRI_SECURITY = 50,
 	NF_IP6_PRI_NAT_SRC = 100,
 	NF_IP6_PRI_SELINUX_LAST = 225,
 	NF_IP6_PRI_LAST = INT_MAX,
diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h
index ac053be6c256..5bacd838e88b 100644
--- a/include/net/netns/ipv6.h
+++ b/include/net/netns/ipv6.h
@@ -35,6 +35,7 @@ struct netns_ipv6 {
 	struct xt_table		*ip6table_filter;
 	struct xt_table		*ip6table_mangle;
 	struct xt_table		*ip6table_raw;
+	struct xt_table		*ip6table_security;
 #endif
 	struct rt6_info         *ip6_null_entry;
 	struct rt6_statistics   *rt6_stats;
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 6cae5475737e..689dec899c57 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -208,5 +208,17 @@ config IP6_NF_RAW
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
 
+# security table for MAC policy
+config IP6_NF_SECURITY
+       tristate "Security table"
+       depends on IP6_NF_IPTABLES
+       depends on SECURITY
+       default m if NETFILTER_ADVANCED=n
+       help
+         This option adds a `security' table to iptables, for use
+         with Mandatory Access Control (MAC) policy.
+        
+         If unsure, say N.
+
 endmenu
 
diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile
index fbf2c14ed887..3f17c948eefb 100644
--- a/net/ipv6/netfilter/Makefile
+++ b/net/ipv6/netfilter/Makefile
@@ -8,6 +8,7 @@ obj-$(CONFIG_IP6_NF_FILTER) += ip6table_filter.o
 obj-$(CONFIG_IP6_NF_MANGLE) += ip6table_mangle.o
 obj-$(CONFIG_IP6_NF_QUEUE) += ip6_queue.o
 obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o
+obj-$(CONFIG_IP6_NF_SECURITY) += ip6table_security.o
 
 # objects for l3 independent conntrack
 nf_conntrack_ipv6-objs  :=  nf_conntrack_l3proto_ipv6.o nf_conntrack_proto_icmpv6.o nf_conntrack_reasm.o
diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c
new file mode 100644
index 000000000000..063a3d9c3c67
--- /dev/null
+++ b/net/ipv6/netfilter/ip6table_security.c
@@ -0,0 +1,172 @@
+/*
+ * "security" table for IPv6
+ *
+ * This is for use by Mandatory Access Control (MAC) security models,
+ * which need to be able to manage security policy in separate context
+ * to DAC.
+ *
+ * Based on iptable_mangle.c
+ *
+ * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling
+ * Copyright (C) 2000-2004 Netfilter Core Team <coreteam <at> netfilter.org>
+ * Copyright (C) 2008 Red Hat, Inc., James Morris <jmorris <at> redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/netfilter_ipv6/ip6_tables.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("James Morris <jmorris <at> redhat.com>");
+MODULE_DESCRIPTION("ip6tables security table, for MAC rules");
+
+#define SECURITY_VALID_HOOKS	(1 << NF_INET_LOCAL_IN) | \
+				(1 << NF_INET_FORWARD) | \
+				(1 << NF_INET_LOCAL_OUT)
+
+static struct
+{
+	struct ip6t_replace repl;
+	struct ip6t_standard entries[3];
+	struct ip6t_error term;
+} initial_table __initdata = {
+	.repl = {
+		.name = "security",
+		.valid_hooks = SECURITY_VALID_HOOKS,
+		.num_entries = 4,
+		.size = sizeof(struct ip6t_standard) * 3 + sizeof(struct ip6t_error),
+		.hook_entry = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ip6t_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 2,
+		},
+		.underflow = {
+			[NF_INET_LOCAL_IN] 	= 0,
+			[NF_INET_FORWARD] 	= sizeof(struct ip6t_standard),
+			[NF_INET_LOCAL_OUT] 	= sizeof(struct ip6t_standard) * 2,
+		},
+	},
+	.entries = {
+		IP6T_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_IN */
+		IP6T_STANDARD_INIT(NF_ACCEPT),	/* FORWARD */
+		IP6T_STANDARD_INIT(NF_ACCEPT),	/* LOCAL_OUT */
+	},
+	.term = IP6T_ERROR_INIT,		/* ERROR */
+};
+
+static struct xt_table security_table = {
+	.name		= "security",
+	.valid_hooks	= SECURITY_VALID_HOOKS,
+	.lock		= __RW_LOCK_UNLOCKED(security_table.lock),
+	.me		= THIS_MODULE,
+	.af		= AF_INET6,
+};
+
+static unsigned int
+ip6t_local_in_hook(unsigned int hook,
+		   struct sk_buff *skb,
+		   const struct net_device *in,
+		   const struct net_device *out,
+		   int (*okfn)(struct sk_buff *))
+{
+	return ip6t_do_table(skb, hook, in, out,
+			     init_net.ipv6.ip6table_security);
+}
+
+static unsigned int
+ip6t_forward_hook(unsigned int hook,
+		  struct sk_buff *skb,
+		  const struct net_device *in,
+		  const struct net_device *out,
+		  int (*okfn)(struct sk_buff *))
+{
+	return ip6t_do_table(skb, hook, in, out,
+			     init_net.ipv6.ip6table_security);
+}
+
+static unsigned int
+ip6t_local_out_hook(unsigned int hook,
+		    struct sk_buff *skb,
+		    const struct net_device *in,
+		    const struct net_device *out,
+		    int (*okfn)(struct sk_buff *))
+{
+	/* TBD: handle short packets via raw socket */
+	return ip6t_do_table(skb, hook, in, out,
+			     init_net.ipv6.ip6table_security);
+}
+
+static struct nf_hook_ops ip6t_ops[] __read_mostly = {
+	{
+		.hook		= ip6t_local_in_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET6,
+		.hooknum	= NF_INET_LOCAL_IN,
+		.priority	= NF_IP6_PRI_SECURITY,
+	},
+	{
+		.hook		= ip6t_forward_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET6,
+		.hooknum	= NF_INET_FORWARD,
+		.priority	= NF_IP6_PRI_SECURITY,
+	},
+	{
+		.hook		= ip6t_local_out_hook,
+		.owner		= THIS_MODULE,
+		.pf		= PF_INET6,
+		.hooknum	= NF_INET_LOCAL_OUT,
+		.priority	= NF_IP6_PRI_SECURITY,
+	},
+};
+
+static int __net_init ip6table_security_net_init(struct net *net)
+{
+	net->ipv6.ip6table_security =
+		ip6t_register_table(net, &security_table, &initial_table.repl);
+
+	if (IS_ERR(net->ipv6.ip6table_security))
+		return PTR_ERR(net->ipv6.ip6table_security);
+
+	return 0;
+}
+
+static void __net_exit ip6table_security_net_exit(struct net *net)
+{
+	ip6t_unregister_table(net->ipv6.ip6table_security);
+}
+
+static struct pernet_operations ip6table_security_net_ops = {
+	.init = ip6table_security_net_init,
+	.exit = ip6table_security_net_exit,
+};
+
+static int __init ip6table_security_init(void)
+{
+	int ret;
+
+	ret = register_pernet_subsys(&ip6table_security_net_ops);
+	if (ret < 0)
+		return ret;
+
+	ret = nf_register_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops));
+	if (ret < 0)
+		goto cleanup_table;
+
+	return ret;
+
+cleanup_table:
+	unregister_pernet_subsys(&ip6table_security_net_ops);
+	return ret;
+}
+
+static void __exit ip6table_security_fini(void)
+{
+	nf_unregister_hooks(ip6t_ops, ARRAY_SIZE(ip6t_ops));
+	unregister_pernet_subsys(&ip6table_security_net_ops);
+}
+
+module_init(ip6table_security_init);
+module_exit(ip6table_security_fini);
-- 
cgit v1.2.3


From cc1a9d86ce989083703c4bdc11b75a87e1cc404a Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Sun, 8 Jun 2008 19:39:16 -0700
Subject: mm, x86: shrink_active_range() should check all

Now we are using register_e820_active_regions() instead of
add_active_range() directly. So end_pfn could be different between the
value in early_node_map to node_end_pfn.

So we need to make shrink_active_range() smarter.

shrink_active_range() is a generic MM function in mm/page_alloc.c but
it is only used on 32-bit x86. Should we move it back to some file in
arch/x86?

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/discontig_32.c |  2 +-
 include/linux/mm.h         |  3 +--
 mm/page_alloc.c            | 44 ++++++++++++++++++++++++++++++++++----------
 3 files changed, 36 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c
index a89ccf3d4c14..489605bab85a 100644
--- a/arch/x86/mm/discontig_32.c
+++ b/arch/x86/mm/discontig_32.c
@@ -282,7 +282,7 @@ static unsigned long calculate_numa_remap_pages(void)
 
 		node_end_pfn[nid] -= size;
 		node_remap_start_pfn[nid] = node_end_pfn[nid];
-		shrink_active_range(nid, old_end_pfn, node_end_pfn[nid]);
+		shrink_active_range(nid, node_end_pfn[nid]);
 	}
 	printk("Reserving total of %ld pages for numa KVA remap\n",
 			reserve_pages);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index c31a9cd2a30e..7cbd949f2516 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -997,8 +997,7 @@ extern void free_area_init_node(int nid, pg_data_t *pgdat,
 extern void free_area_init_nodes(unsigned long *max_zone_pfn);
 extern void add_active_range(unsigned int nid, unsigned long start_pfn,
 					unsigned long end_pfn);
-extern void shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
-						unsigned long new_end_pfn);
+extern void shrink_active_range(unsigned int nid, unsigned long new_end_pfn);
 extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
 					unsigned long end_pfn);
 extern void remove_all_active_ranges(void);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 502223c3c2c6..215408684076 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3579,25 +3579,49 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn,
 /**
  * shrink_active_range - Shrink an existing registered range of PFNs
  * @nid: The node id the range is on that should be shrunk
- * @old_end_pfn: The old end PFN of the range
  * @new_end_pfn: The new PFN of the range
  *
  * i386 with NUMA use alloc_remap() to store a node_mem_map on a local node.
- * The map is kept at the end physical page range that has already been
- * registered with add_active_range(). This function allows an arch to shrink
- * an existing registered range.
+ * The map is kept near the end physical page range that has already been
+ * registered. This function allows an arch to shrink an existing registered
+ * range.
  */
-void __init shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
-						unsigned long new_end_pfn)
+void __init shrink_active_range(unsigned int nid, unsigned long new_end_pfn)
 {
-	int i;
+	int i, j;
+	int removed = 0;
 
 	/* Find the old active region end and shrink */
-	for_each_active_range_index_in_nid(i, nid)
-		if (early_node_map[i].end_pfn == old_end_pfn) {
+	for_each_active_range_index_in_nid(i, nid) {
+		if (early_node_map[i].start_pfn >= new_end_pfn) {
+			/* clear it */
+			early_node_map[i].end_pfn = 0;
+			removed = 1;
+			continue;
+		}
+		if (early_node_map[i].end_pfn > new_end_pfn) {
 			early_node_map[i].end_pfn = new_end_pfn;
-			break;
+			continue;
 		}
+	}
+
+	if (!removed)
+		return;
+
+	/* remove the blank ones */
+	for (i = nr_nodemap_entries - 1; i > 0; i--) {
+		if (early_node_map[i].nid != nid)
+			continue;
+		if (early_node_map[i].end_pfn)
+			continue;
+		/* we found it, get rid of it */
+		for (j = i; j < nr_nodemap_entries - 1; j++)
+			memcpy(&early_node_map[j], &early_node_map[j+1],
+				sizeof(early_node_map[j]));
+		j = nr_nodemap_entries - 1;
+		memset(&early_node_map[j], 0, sizeof(early_node_map[j]));
+		nr_nodemap_entries--;
+	}
 }
 
 /**
-- 
cgit v1.2.3


From 0eb967012ea15e6e8cfab483d9fa37bc602d400c Mon Sep 17 00:00:00 2001
From: Abhishek Sagar <sagar.abhishek@gmail.com>
Date: Sun, 1 Jun 2008 21:47:30 +0530
Subject: ftrace: prevent freeing of all failed updates

Prevent freeing of records which cause problems and correspond to function from
core kernel text. A new flag, FTRACE_FL_CONVERTED is used to mark a record
as "converted". All other records are patched lazily to NOPs. Failed records
now also remain on frace_hash table. Each invocation of ftrace_record_ip now
checks whether the traced function has ever been recorded (including past
failures) and doesn't re-record it again.

Signed-off-by: Abhishek Sagar <sagar.abhishek@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/ftrace.h |  1 +
 kernel/trace/ftrace.c  | 76 ++++++++++++++++++++++++++++++--------------------
 2 files changed, 47 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 623819433ed5..20e14d0093c7 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -49,6 +49,7 @@ enum {
 	FTRACE_FL_FILTER	= (1 << 2),
 	FTRACE_FL_ENABLED	= (1 << 3),
 	FTRACE_FL_NOTRACE	= (1 << 4),
+	FTRACE_FL_CONVERTED	= (1 << 5),
 };
 
 struct dyn_ftrace {
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index f762f5a2d331..ec54cb7d69d6 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -216,6 +216,12 @@ ftrace_add_hash(struct dyn_ftrace *node, unsigned long key)
 	hlist_add_head_rcu(&node->node, &ftrace_hash[key]);
 }
 
+/* called from kstop_machine */
+static inline void ftrace_del_hash(struct dyn_ftrace *node)
+{
+	hlist_del(&node->node);
+}
+
 static void ftrace_free_rec(struct dyn_ftrace *rec)
 {
 	/* no locking, only called from kstop_machine */
@@ -332,12 +338,11 @@ ftrace_record_ip(unsigned long ip)
 #define FTRACE_ADDR ((long)(ftrace_caller))
 #define MCOUNT_ADDR ((long)(mcount))
 
-static void
+static int
 __ftrace_replace_code(struct dyn_ftrace *rec,
 		      unsigned char *old, unsigned char *new, int enable)
 {
 	unsigned long ip, fl;
-	int failed;
 
 	ip = rec->ip;
 
@@ -364,7 +369,7 @@ __ftrace_replace_code(struct dyn_ftrace *rec,
 
 		if ((fl ==  (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) ||
 		    (fl == 0) || (rec->flags & FTRACE_FL_NOTRACE))
-			return;
+			return 0;
 
 		/*
 		 * If it is enabled disable it,
@@ -388,7 +393,7 @@ __ftrace_replace_code(struct dyn_ftrace *rec,
 			 */
 			fl = rec->flags & (FTRACE_FL_NOTRACE | FTRACE_FL_ENABLED);
 			if (fl == FTRACE_FL_NOTRACE)
-				return;
+				return 0;
 
 			new = ftrace_call_replace(ip, FTRACE_ADDR);
 		} else
@@ -396,34 +401,24 @@ __ftrace_replace_code(struct dyn_ftrace *rec,
 
 		if (enable) {
 			if (rec->flags & FTRACE_FL_ENABLED)
-				return;
+				return 0;
 			rec->flags |= FTRACE_FL_ENABLED;
 		} else {
 			if (!(rec->flags & FTRACE_FL_ENABLED))
-				return;
+				return 0;
 			rec->flags &= ~FTRACE_FL_ENABLED;
 		}
 	}
 
-	failed = ftrace_modify_code(ip, old, new);
-	if (failed) {
-		unsigned long key;
-		/* It is possible that the function hasn't been converted yet */
-		key = hash_long(ip, FTRACE_HASHBITS);
-		if (!ftrace_ip_in_hash(ip, key)) {
-			rec->flags |= FTRACE_FL_FAILED;
-			ftrace_free_rec(rec);
-		}
-
-	}
+	return ftrace_modify_code(ip, old, new);
 }
 
 static void ftrace_replace_code(int enable)
 {
+	int i, failed;
 	unsigned char *new = NULL, *old = NULL;
 	struct dyn_ftrace *rec;
 	struct ftrace_page *pg;
-	int i;
 
 	if (enable)
 		old = ftrace_nop_replace();
@@ -438,7 +433,15 @@ static void ftrace_replace_code(int enable)
 			if (rec->flags & FTRACE_FL_FAILED)
 				continue;
 
-			__ftrace_replace_code(rec, old, new, enable);
+			failed = __ftrace_replace_code(rec, old, new, enable);
+			if (failed && (rec->flags & FTRACE_FL_CONVERTED)) {
+				rec->flags |= FTRACE_FL_FAILED;
+				if ((system_state == SYSTEM_BOOTING) ||
+				    !kernel_text_address(rec->ip)) {
+					ftrace_del_hash(rec);
+					ftrace_free_rec(rec);
+				}
+			}
 		}
 	}
 }
@@ -467,7 +470,6 @@ ftrace_code_disable(struct dyn_ftrace *rec)
 	failed = ftrace_modify_code(ip, call, nop);
 	if (failed) {
 		rec->flags |= FTRACE_FL_FAILED;
-		ftrace_free_rec(rec);
 		return 0;
 	}
 	return 1;
@@ -621,8 +623,7 @@ unsigned long		ftrace_update_tot_cnt;
 static int __ftrace_update_code(void *ignore)
 {
 	struct dyn_ftrace *p;
-	struct hlist_head head;
-	struct hlist_node *t;
+	struct hlist_node *t, *n;
 	int save_ftrace_enabled;
 	cycle_t start, stop;
 	int i;
@@ -637,18 +638,33 @@ static int __ftrace_update_code(void *ignore)
 
 	/* No locks needed, the machine is stopped! */
 	for (i = 0; i < FTRACE_HASHSIZE; i++) {
-		if (hlist_empty(&ftrace_hash[i]))
-			continue;
+		/* all CPUS are stopped, we are safe to modify code */
+		hlist_for_each_entry_safe(p, t, n, &ftrace_hash[i], node) {
+			/* Skip over failed records which have not been
+			 * freed. */
+			if (p->flags & FTRACE_FL_FAILED)
+				continue;
 
-		head = ftrace_hash[i];
-		INIT_HLIST_HEAD(&ftrace_hash[i]);
+			/* Unconverted records are always at the head of the
+			 * hash bucket. Once we encounter a converted record,
+			 * simply skip over to the next bucket. Saves ftraced
+			 * some processor cycles (ftrace does its bid for
+			 * global warming :-p ). */
+			if (p->flags & (FTRACE_FL_CONVERTED))
+				break;
 
-		/* all CPUS are stopped, we are safe to modify code */
-		hlist_for_each_entry(p, t, &head, node) {
-			if (ftrace_code_disable(p))
+			if (ftrace_code_disable(p)) {
+				p->flags |= FTRACE_FL_CONVERTED;
 				ftrace_update_cnt++;
-		}
+			} else {
+				if ((system_state == SYSTEM_BOOTING) ||
+				    !kernel_text_address(p->ip)) {
+					ftrace_del_hash(p);
+					ftrace_free_rec(p);
 
+				}
+			}
+		}
 	}
 
 	stop = ftrace_now(raw_smp_processor_id());
-- 
cgit v1.2.3


From 9985b0bab332289f14837eff3c6e0bcc658b58f7 Mon Sep 17 00:00:00 2001
From: David Rientjes <rientjes@google.com>
Date: Thu, 5 Jun 2008 12:57:11 -0700
Subject: sched: prevent bound kthreads from changing cpus_allowed

Kthreads that have called kthread_bind() are bound to specific cpus, so
other tasks should not be able to change their cpus_allowed from under
them.  Otherwise, it is possible to move kthreads, such as the migration
or software watchdog threads, so they are not allowed access to the cpu
they work on.

Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |  1 +
 kernel/cpuset.c       | 14 +++++++++++++-
 kernel/kthread.c      |  1 +
 kernel/sched.c        |  6 ++++++
 4 files changed, 21 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index d25acf600a32..2db1485f865d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1486,6 +1486,7 @@ static inline void put_task_struct(struct task_struct *t)
 #define PF_SWAPWRITE	0x00800000	/* Allowed to write to swap */
 #define PF_SPREAD_PAGE	0x01000000	/* Spread page cache over cpuset */
 #define PF_SPREAD_SLAB	0x02000000	/* Spread some slab caches over cpuset */
+#define PF_THREAD_BOUND	0x04000000	/* Thread bound to specific cpu */
 #define PF_MEMPOLICY	0x10000000	/* Non-default NUMA mempolicy */
 #define PF_MUTEX_TESTER	0x20000000	/* Thread belongs to the rt mutex tester */
 #define PF_FREEZER_SKIP	0x40000000	/* Freezer should not count it as freezeable */
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 6090d18b58a9..b84354f4de36 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -1190,6 +1190,15 @@ static int cpuset_can_attach(struct cgroup_subsys *ss,
 
 	if (cpus_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))
 		return -ENOSPC;
+	if (tsk->flags & PF_THREAD_BOUND) {
+		cpumask_t mask;
+
+		mutex_lock(&callback_mutex);
+		mask = cs->cpus_allowed;
+		mutex_unlock(&callback_mutex);
+		if (!cpus_equal(tsk->cpus_allowed, mask))
+			return -EINVAL;
+	}
 
 	return security_task_setscheduler(tsk, 0, NULL);
 }
@@ -1203,11 +1212,14 @@ static void cpuset_attach(struct cgroup_subsys *ss,
 	struct mm_struct *mm;
 	struct cpuset *cs = cgroup_cs(cont);
 	struct cpuset *oldcs = cgroup_cs(oldcont);
+	int err;
 
 	mutex_lock(&callback_mutex);
 	guarantee_online_cpus(cs, &cpus);
-	set_cpus_allowed_ptr(tsk, &cpus);
+	err = set_cpus_allowed_ptr(tsk, &cpus);
 	mutex_unlock(&callback_mutex);
+	if (err)
+		return;
 
 	from = oldcs->mems_allowed;
 	to = cs->mems_allowed;
diff --git a/kernel/kthread.c b/kernel/kthread.c
index bd1b9ea024e1..97747cdd37c9 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -180,6 +180,7 @@ void kthread_bind(struct task_struct *k, unsigned int cpu)
 	set_task_cpu(k, cpu);
 	k->cpus_allowed = cpumask_of_cpu(cpu);
 	k->rt.nr_cpus_allowed = 1;
+	k->flags |= PF_THREAD_BOUND;
 }
 EXPORT_SYMBOL(kthread_bind);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index e9c24a128655..164fe7fe0d89 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5563,6 +5563,12 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
 		goto out;
 	}
 
+	if (unlikely((p->flags & PF_THREAD_BOUND) && p != current &&
+		     !cpus_equal(p->cpus_allowed, *new_mask))) {
+		ret = -EINVAL;
+		goto out;
+	}
+
 	if (p->sched_class->set_cpus_allowed)
 		p->sched_class->set_cpus_allowed(p, new_mask);
 	else {
-- 
cgit v1.2.3


From 19792a0859f96e9fc8ce87d97b269bcb895389e5 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Mon, 5 May 2008 21:25:47 +0300
Subject: PCI: drivers/pci/pci.c: add prototypes

This patch adds prototypes for pcibios_disable_device() and
pcibios_set_pcie_reset_state() in include/linux/pci.h

While I was at it, I also removed the unneeded "extern" from the
prototype of pcibios_add_platform_entries().

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/pci.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/pci.h b/include/linux/pci.h
index 509159bcd4e7..aaa9f333fb44 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1059,7 +1059,10 @@ extern int pci_pci_problems;
 extern unsigned long pci_cardbus_io_size;
 extern unsigned long pci_cardbus_mem_size;
 
-extern int pcibios_add_platform_entries(struct pci_dev *dev);
+int pcibios_add_platform_entries(struct pci_dev *dev);
+void pcibios_disable_device(struct pci_dev *dev);
+int pcibios_set_pcie_reset_state(struct pci_dev *dev,
+				 enum pcie_reset_state state);
 
 #ifdef CONFIG_PCI_MMCONFIG
 extern void __init pci_mmcfg_early_init(void);
-- 
cgit v1.2.3


From e1a2a51e684bfe9d6165992d4a065439617a3107 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Thu, 15 May 2008 21:51:31 +0200
Subject: Suspend/Resume bug in PCI layer wrt quirks

Some quirks should be called with interrupt disabled, we can't directly
call them in .resume_early. Also the patch introduces
pci_fixup_resume_early and pci_fixup_suspend, which matches current
device core callbacks (.suspend/.resume_early).

TBD: Somebody knows why we need quirk resume should double check if a
quirk should be called in resume or resume_early. I changed some per my
understanding, but can't make sure I fixed all.

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci-driver.c          |   6 +-
 drivers/pci/quirks.c              | 118 ++++++++++++++++++++++++++------------
 include/asm-generic/vmlinux.lds.h |   6 ++
 include/linux/pci.h               |  10 +++-
 4 files changed, 101 insertions(+), 39 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 72cf61ed8f96..677fd9d6db12 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -292,6 +292,9 @@ static int pci_device_suspend(struct device * dev, pm_message_t state)
 		if (pci_dev->current_state == PCI_D0)
 			pci_dev->current_state = PCI_UNKNOWN;
 	}
+
+	pci_fixup_device(pci_fixup_suspend, pci_dev);
+
 	return i;
 }
 
@@ -337,6 +340,7 @@ static int pci_device_resume(struct device * dev)
 		error = drv->resume(pci_dev);
 	else
 		error = pci_default_resume(pci_dev);
+	pci_fixup_device(pci_fixup_resume, pci_dev);
 	return error;
 }
 
@@ -346,7 +350,7 @@ static int pci_device_resume_early(struct device * dev)
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	struct pci_driver * drv = pci_dev->driver;
 
-	pci_fixup_device(pci_fixup_resume, pci_dev);
+	pci_fixup_device(pci_fixup_resume_early, pci_dev);
 
 	if (drv && drv->resume_early)
 		error = drv->resume_early(pci_dev);
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index dabb563f51d9..44aabb1519a9 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -556,7 +556,7 @@ static void quirk_via_ioapic(struct pci_dev *dev)
 	pci_write_config_byte (dev, 0x58, tmp);
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_82C686,	quirk_via_ioapic);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_82C686,	quirk_via_ioapic);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_82C686,	quirk_via_ioapic);
 
 /*
  * VIA 8237: Some BIOSs don't set the 'Bypass APIC De-Assert Message' Bit.
@@ -576,7 +576,7 @@ static void quirk_via_vt8237_bypass_apic_deassert(struct pci_dev *dev)
 	}
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8237,		quirk_via_vt8237_bypass_apic_deassert);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8237,		quirk_via_vt8237_bypass_apic_deassert);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8237,		quirk_via_vt8237_bypass_apic_deassert);
 
 /*
  * The AMD io apic can hang the box when an apic irq is masked.
@@ -622,7 +622,7 @@ static void quirk_amd_8131_ioapic(struct pci_dev *dev)
         }
 } 
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_amd_8131_ioapic);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_amd_8131_ioapic);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_amd_8131_ioapic);
 #endif /* CONFIG_X86_IO_APIC */
 
 /*
@@ -774,7 +774,7 @@ static void quirk_cardbus_legacy(struct pci_dev *dev)
 	pci_write_config_dword(dev, PCI_CB_LEGACY_MODE_BASE, 0);
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_ANY_ID, PCI_ANY_ID, quirk_cardbus_legacy);
-DECLARE_PCI_FIXUP_RESUME(PCI_ANY_ID, PCI_ANY_ID, quirk_cardbus_legacy);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_ANY_ID, PCI_ANY_ID, quirk_cardbus_legacy);
 
 /*
  * Following the PCI ordering rules is optional on the AMD762. I'm not
@@ -797,7 +797,7 @@ static void quirk_amd_ordering(struct pci_dev *dev)
 	}
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_FE_GATE_700C, quirk_amd_ordering);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_FE_GATE_700C, quirk_amd_ordering);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_AMD,	PCI_DEVICE_ID_AMD_FE_GATE_700C, quirk_amd_ordering);
 
 /*
  *	DreamWorks provided workaround for Dunord I-3000 problem
@@ -865,7 +865,7 @@ static void quirk_disable_pxb(struct pci_dev *pdev)
 	}
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82454NX,	quirk_disable_pxb);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82454NX,	quirk_disable_pxb);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82454NX,	quirk_disable_pxb);
 
 static void __devinit quirk_amd_ide_mode(struct pci_dev *pdev)
 {
@@ -885,9 +885,9 @@ static void __devinit quirk_amd_ide_mode(struct pci_dev *pdev)
 	}
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP600_SATA, quirk_amd_ide_mode);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP600_SATA, quirk_amd_ide_mode);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP600_SATA, quirk_amd_ide_mode);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP700_SATA, quirk_amd_ide_mode);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP700_SATA, quirk_amd_ide_mode);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP700_SATA, quirk_amd_ide_mode);
 
 /*
  *	Serverworks CSB5 IDE does not fully support native mode
@@ -1093,31 +1093,61 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801CA_0,	asu
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801CA_12,	asus_hides_smbus_lpc);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801DB_12,	asus_hides_smbus_lpc);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801EB_0,	asus_hides_smbus_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801AA_0,	asus_hides_smbus_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801DB_0,	asus_hides_smbus_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801BA_0,	asus_hides_smbus_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801CA_0,	asus_hides_smbus_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801CA_12,	asus_hides_smbus_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801DB_12,	asus_hides_smbus_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801EB_0,	asus_hides_smbus_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801AA_0,	asus_hides_smbus_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801DB_0,	asus_hides_smbus_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801BA_0,	asus_hides_smbus_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801CA_0,	asus_hides_smbus_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801CA_12,	asus_hides_smbus_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801DB_12,	asus_hides_smbus_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_82801EB_0,	asus_hides_smbus_lpc);
 
-static void asus_hides_smbus_lpc_ich6(struct pci_dev *dev)
+/* It appears we just have one such device. If not, we have a warning */
+static void __iomem *asus_rcba_base;
+static void asus_hides_smbus_lpc_ich6_suspend(struct pci_dev *dev)
 {
-	u32 val, rcba;
-	void __iomem *base;
+	u32 rcba;
 
 	if (likely(!asus_hides_smbus))
 		return;
+	WARN_ON(asus_rcba_base);
+
 	pci_read_config_dword(dev, 0xF0, &rcba);
-	base = ioremap_nocache(rcba & 0xFFFFC000, 0x4000); /* use bits 31:14, 16 kB aligned */
-	if (base == NULL) return;
-	val=readl(base + 0x3418); /* read the Function Disable register, dword mode only */
-	writel(val & 0xFFFFFFF7, base + 0x3418); /* enable the SMBus device */
-	iounmap(base);
+	/* use bits 31:14, 16 kB aligned */
+	asus_rcba_base = ioremap_nocache(rcba & 0xFFFFC000, 0x4000);
+	if (asus_rcba_base == NULL)
+		return;
+}
+
+static void asus_hides_smbus_lpc_ich6_resume_early(struct pci_dev *dev)
+{
+	u32 val;
+
+	if (likely(!asus_hides_smbus || !asus_rcba_base))
+		return;
+	/* read the Function Disable register, dword mode only */
+	val = readl(asus_rcba_base + 0x3418);
+	writel(val & 0xFFFFFFF7, asus_rcba_base + 0x3418); /* enable the SMBus device */
+}
+
+static void asus_hides_smbus_lpc_ich6_resume(struct pci_dev *dev)
+{
+	if (likely(!asus_hides_smbus || !asus_rcba_base))
+		return;
+	iounmap(asus_rcba_base);
+	asus_rcba_base = NULL;
 	dev_info(&dev->dev, "Enabled ICH6/i801 SMBus device\n");
 }
+
+static void asus_hides_smbus_lpc_ich6(struct pci_dev *dev)
+{
+	asus_hides_smbus_lpc_ich6_suspend(dev);
+	asus_hides_smbus_lpc_ich6_resume_early(dev);
+	asus_hides_smbus_lpc_ich6_resume(dev);
+}
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_ICH6_1,	asus_hides_smbus_lpc_ich6);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_ICH6_1,	asus_hides_smbus_lpc_ich6);
+DECLARE_PCI_FIXUP_SUSPEND(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_ICH6_1,	asus_hides_smbus_lpc_ich6_suspend);
+DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_ICH6_1,	asus_hides_smbus_lpc_ich6_resume);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_ICH6_1,	asus_hides_smbus_lpc_ich6_resume_early);
 
 /*
  * SiS 96x south bridge: BIOS typically hides SMBus device...
@@ -1135,10 +1165,10 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_961,		quirk_sis_96x_
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_962,		quirk_sis_96x_smbus);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_963,		quirk_sis_96x_smbus);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_LPC,		quirk_sis_96x_smbus);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_961,		quirk_sis_96x_smbus);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_962,		quirk_sis_96x_smbus);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_963,		quirk_sis_96x_smbus);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_LPC,		quirk_sis_96x_smbus);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_961,		quirk_sis_96x_smbus);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_962,		quirk_sis_96x_smbus);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_963,		quirk_sis_96x_smbus);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_LPC,		quirk_sis_96x_smbus);
 
 /*
  * ... This is further complicated by the fact that some SiS96x south
@@ -1172,7 +1202,7 @@ static void quirk_sis_503(struct pci_dev *dev)
 	quirk_sis_96x_smbus(dev);
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_503,		quirk_sis_503);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_503,		quirk_sis_503);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_SI,	PCI_DEVICE_ID_SI_503,		quirk_sis_503);
 
 
 /*
@@ -1205,7 +1235,7 @@ static void asus_hides_ac97_lpc(struct pci_dev *dev)
 	}
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8237, asus_hides_ac97_lpc);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8237, asus_hides_ac97_lpc);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_VIA,	PCI_DEVICE_ID_VIA_8237, asus_hides_ac97_lpc);
 
 #if defined(CONFIG_ATA) || defined(CONFIG_ATA_MODULE)
 
@@ -1270,12 +1300,12 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363, qui
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365, quirk_jmicron_ata);
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366, quirk_jmicron_ata);
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368, quirk_jmicron_ata);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB360, quirk_jmicron_ata);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB361, quirk_jmicron_ata);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363, quirk_jmicron_ata);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365, quirk_jmicron_ata);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366, quirk_jmicron_ata);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368, quirk_jmicron_ata);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB360, quirk_jmicron_ata);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB361, quirk_jmicron_ata);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363, quirk_jmicron_ata);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365, quirk_jmicron_ata);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366, quirk_jmicron_ata);
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368, quirk_jmicron_ata);
 
 #endif
 
@@ -1521,6 +1551,10 @@ extern struct pci_fixup __start_pci_fixups_enable[];
 extern struct pci_fixup __end_pci_fixups_enable[];
 extern struct pci_fixup __start_pci_fixups_resume[];
 extern struct pci_fixup __end_pci_fixups_resume[];
+extern struct pci_fixup __start_pci_fixups_resume_early[];
+extern struct pci_fixup __end_pci_fixups_resume_early[];
+extern struct pci_fixup __start_pci_fixups_suspend[];
+extern struct pci_fixup __end_pci_fixups_suspend[];
 
 
 void pci_fixup_device(enum pci_fixup_pass pass, struct pci_dev *dev)
@@ -1553,6 +1587,16 @@ void pci_fixup_device(enum pci_fixup_pass pass, struct pci_dev *dev)
 		end = __end_pci_fixups_resume;
 		break;
 
+	case pci_fixup_resume_early:
+		start = __start_pci_fixups_resume_early;
+		end = __end_pci_fixups_resume_early;
+		break;
+
+	case pci_fixup_suspend:
+		start = __start_pci_fixups_suspend;
+		end = __end_pci_fixups_suspend;
+		break;
+
 	default:
 		/* stupid compiler warning, you would think with an enum... */
 		return;
@@ -1629,7 +1673,7 @@ static void quirk_nvidia_ck804_pcie_aer_ext_cap(struct pci_dev *dev)
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA,  PCI_DEVICE_ID_NVIDIA_CK804_PCIE,
 			quirk_nvidia_ck804_pcie_aer_ext_cap);
-DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_NVIDIA,  PCI_DEVICE_ID_NVIDIA_CK804_PCIE,
+DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_NVIDIA,  PCI_DEVICE_ID_NVIDIA_CK804_PCIE,
 			quirk_nvidia_ck804_pcie_aer_ext_cap);
 
 static void __devinit quirk_via_cx700_pci_parking_caching(struct pci_dev *dev)
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index f054778e916c..cf108a3c7f59 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -84,6 +84,12 @@
 		VMLINUX_SYMBOL(__start_pci_fixups_resume) = .;		\
 		*(.pci_fixup_resume)					\
 		VMLINUX_SYMBOL(__end_pci_fixups_resume) = .;		\
+		VMLINUX_SYMBOL(__start_pci_fixups_resume_early) = .;	\
+		*(.pci_fixup_resume_early)				\
+		VMLINUX_SYMBOL(__end_pci_fixups_resume_early) = .;	\
+		VMLINUX_SYMBOL(__start_pci_fixups_suspend) = .;		\
+		*(.pci_fixup_suspend)					\
+		VMLINUX_SYMBOL(__end_pci_fixups_suspend) = .;		\
 	}								\
 									\
 	/* RapidIO route ops */						\
diff --git a/include/linux/pci.h b/include/linux/pci.h
index aaa9f333fb44..700704ef70f3 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1013,7 +1013,9 @@ enum pci_fixup_pass {
 	pci_fixup_header,	/* After reading configuration header */
 	pci_fixup_final,	/* Final phase of device fixups */
 	pci_fixup_enable,	/* pci_enable_device() time */
-	pci_fixup_resume,	/* pci_enable_device() time */
+	pci_fixup_resume,	/* pci_device_resume() */
+	pci_fixup_suspend,	/* pci_device_suspend */
+	pci_fixup_resume_early, /* pci_device_resume_early() */
 };
 
 /* Anonymous variables would be nice... */
@@ -1035,6 +1037,12 @@ enum pci_fixup_pass {
 #define DECLARE_PCI_FIXUP_RESUME(vendor, device, hook)			\
 	DECLARE_PCI_FIXUP_SECTION(.pci_fixup_resume,			\
 			resume##vendor##device##hook, vendor, device, hook)
+#define DECLARE_PCI_FIXUP_RESUME_EARLY(vendor, device, hook)		\
+	DECLARE_PCI_FIXUP_SECTION(.pci_fixup_resume_early,		\
+			resume_early##vendor##device##hook, vendor, device, hook)
+#define DECLARE_PCI_FIXUP_SUSPEND(vendor, device, hook)			\
+	DECLARE_PCI_FIXUP_SECTION(.pci_fixup_suspend,			\
+			suspend##vendor##device##hook, vendor, device, hook)
 
 
 void pci_fixup_device(enum pci_fixup_pass pass, struct pci_dev *dev);
-- 
cgit v1.2.3


From 1eede070a59e1cc73da51e1aaa00d9ab86572cfc Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Tue, 20 May 2008 23:00:01 +0200
Subject: Introduce new top level suspend and hibernation callbacks

Introduce 'struct pm_ops' and 'struct pm_ext_ops' ('ext' meaning
'extended') representing suspend and hibernation operations for bus
types, device classes, device types and device drivers.

Modify the PM core to use 'struct pm_ops' and 'struct pm_ext_ops'
objects, if defined, instead of the ->suspend(), ->resume(),
->suspend_late(), and ->resume_early() callbacks (the old callbacks
will be considered as legacy and gradually phased out).

The main purpose of doing this is to separate suspend (aka S2RAM and
standby) callbacks from hibernation callbacks in such a way that the
new callbacks won't take arguments and the semantics of each of them
will be clearly specified.  This has been requested for multiple
times by many people, including Linus himself, and the reason is that
within the current scheme if ->resume() is called, for example, it's
difficult to say why it's been called (ie. is it a resume from RAM or
from hibernation or a suspend/hibernation failure etc.?).

The second purpose is to make the suspend/hibernation callbacks more
flexible so that device drivers can handle more than they can within
the current scheme.  For example, some drivers may need to prevent
new children of the device from being registered before their
->suspend() callbacks are executed or they may want to carry out some
operations requiring the availability of some other devices, not
directly bound via the parent-child relationship, in order to prepare
for the execution of ->suspend(), etc.

Ultimately, we'd like to stop using the freezing of tasks for suspend
and therefore the drivers' suspend/hibernation code will have to take
care of the handling of the user space during suspend/hibernation.
That, in turn, would be difficult within the current scheme, without
the new ->prepare() and ->complete() callbacks.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/kernel/apm_32.c   |   8 +-
 drivers/base/power/main.c  | 675 +++++++++++++++++++++++++++++++++++----------
 drivers/base/power/power.h |   2 +-
 drivers/base/power/trace.c |   4 +-
 include/linux/device.h     |   9 +
 include/linux/pm.h         | 314 +++++++++++++++++++--
 kernel/power/disk.c        |  22 +-
 kernel/power/main.c        |   6 +-
 8 files changed, 845 insertions(+), 195 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index bf9290e29013..c1735f61a2c0 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -1211,9 +1211,9 @@ static int suspend(int vetoable)
 	if (err != APM_SUCCESS)
 		apm_error("suspend", err);
 	err = (err == APM_SUCCESS) ? 0 : -EIO;
-	device_power_up();
+	device_power_up(PMSG_RESUME);
 	local_irq_enable();
-	device_resume();
+	device_resume(PMSG_RESUME);
 	queue_event(APM_NORMAL_RESUME, NULL);
 	spin_lock(&user_list_lock);
 	for (as = user_list; as != NULL; as = as->next) {
@@ -1238,7 +1238,7 @@ static void standby(void)
 		apm_error("standby", err);
 
 	local_irq_disable();
-	device_power_up();
+	device_power_up(PMSG_RESUME);
 	local_irq_enable();
 }
 
@@ -1324,7 +1324,7 @@ static void check_events(void)
 			ignore_bounce = 1;
 			if ((event != APM_NORMAL_RESUME)
 			    || (ignore_normal_resume == 0)) {
-				device_resume();
+				device_resume(PMSG_RESUME);
 				queue_event(event, NULL);
 			}
 			ignore_normal_resume = 0;
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 45cc3d9eacb8..d571204aaff7 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -12,11 +12,9 @@
  * and add it to the list of power-controlled devices. sysfs entries for
  * controlling device power management will also be added.
  *
- * A different set of lists than the global subsystem list are used to
- * keep track of power info because we use different lists to hold
- * devices based on what stage of the power management process they
- * are in. The power domain dependencies may also differ from the
- * ancestral dependencies that the subsystem list maintains.
+ * A separate list is used for keeping track of power info, because the power
+ * domain dependencies may differ from the ancestral dependencies that the
+ * subsystem list maintains.
  */
 
 #include <linux/device.h>
@@ -30,31 +28,40 @@
 #include "power.h"
 
 /*
- * The entries in the dpm_active list are in a depth first order, simply
+ * The entries in the dpm_list list are in a depth first order, simply
  * because children are guaranteed to be discovered after parents, and
  * are inserted at the back of the list on discovery.
  *
- * All the other lists are kept in the same order, for consistency.
- * However the lists aren't always traversed in the same order.
- * Semaphores must be acquired from the top (i.e., front) down
- * and released in the opposite order.  Devices must be suspended
- * from the bottom (i.e., end) up and resumed in the opposite order.
- * That way no parent will be suspended while it still has an active
- * child.
- *
  * Since device_pm_add() may be called with a device semaphore held,
  * we must never try to acquire a device semaphore while holding
  * dpm_list_mutex.
  */
 
-LIST_HEAD(dpm_active);
-static LIST_HEAD(dpm_off);
-static LIST_HEAD(dpm_off_irq);
+LIST_HEAD(dpm_list);
 
 static DEFINE_MUTEX(dpm_list_mtx);
 
-/* 'true' if all devices have been suspended, protected by dpm_list_mtx */
-static bool all_sleeping;
+/*
+ * Set once the preparation of devices for a PM transition has started, reset
+ * before starting to resume devices.  Protected by dpm_list_mtx.
+ */
+static bool transition_started;
+
+/**
+ *	device_pm_lock - lock the list of active devices used by the PM core
+ */
+void device_pm_lock(void)
+{
+	mutex_lock(&dpm_list_mtx);
+}
+
+/**
+ *	device_pm_unlock - unlock the list of active devices used by the PM core
+ */
+void device_pm_unlock(void)
+{
+	mutex_unlock(&dpm_list_mtx);
+}
 
 /**
  *	device_pm_add - add a device to the list of active devices
@@ -68,17 +75,25 @@ int device_pm_add(struct device *dev)
 		 dev->bus ? dev->bus->name : "No Bus",
 		 kobject_name(&dev->kobj));
 	mutex_lock(&dpm_list_mtx);
-	if ((dev->parent && dev->parent->power.sleeping) || all_sleeping) {
-		if (dev->parent->power.sleeping)
-			dev_warn(dev, "parent %s is sleeping\n",
+	if (dev->parent) {
+		if (dev->parent->power.status >= DPM_SUSPENDING) {
+			dev_warn(dev, "parent %s is sleeping, will not add\n",
 				dev->parent->bus_id);
-		else
-			dev_warn(dev, "all devices are sleeping\n");
+			WARN_ON(true);
+		}
+	} else if (transition_started) {
+		/*
+		 * We refuse to register parentless devices while a PM
+		 * transition is in progress in order to avoid leaving them
+		 * unhandled down the road
+		 */
 		WARN_ON(true);
 	}
 	error = dpm_sysfs_add(dev);
-	if (!error)
-		list_add_tail(&dev->power.entry, &dpm_active);
+	if (!error) {
+		dev->power.status = DPM_ON;
+		list_add_tail(&dev->power.entry, &dpm_list);
+	}
 	mutex_unlock(&dpm_list_mtx);
 	return error;
 }
@@ -100,73 +115,243 @@ void device_pm_remove(struct device *dev)
 	mutex_unlock(&dpm_list_mtx);
 }
 
+/**
+ *	pm_op - execute the PM operation appropiate for given PM event
+ *	@dev:	Device.
+ *	@ops:	PM operations to choose from.
+ *	@state:	PM transition of the system being carried out.
+ */
+static int pm_op(struct device *dev, struct pm_ops *ops, pm_message_t state)
+{
+	int error = 0;
+
+	switch (state.event) {
+#ifdef CONFIG_SUSPEND
+	case PM_EVENT_SUSPEND:
+		if (ops->suspend) {
+			error = ops->suspend(dev);
+			suspend_report_result(ops->suspend, error);
+		}
+		break;
+	case PM_EVENT_RESUME:
+		if (ops->resume) {
+			error = ops->resume(dev);
+			suspend_report_result(ops->resume, error);
+		}
+		break;
+#endif /* CONFIG_SUSPEND */
+#ifdef CONFIG_HIBERNATION
+	case PM_EVENT_FREEZE:
+	case PM_EVENT_QUIESCE:
+		if (ops->freeze) {
+			error = ops->freeze(dev);
+			suspend_report_result(ops->freeze, error);
+		}
+		break;
+	case PM_EVENT_HIBERNATE:
+		if (ops->poweroff) {
+			error = ops->poweroff(dev);
+			suspend_report_result(ops->poweroff, error);
+		}
+		break;
+	case PM_EVENT_THAW:
+	case PM_EVENT_RECOVER:
+		if (ops->thaw) {
+			error = ops->thaw(dev);
+			suspend_report_result(ops->thaw, error);
+		}
+		break;
+	case PM_EVENT_RESTORE:
+		if (ops->restore) {
+			error = ops->restore(dev);
+			suspend_report_result(ops->restore, error);
+		}
+		break;
+#endif /* CONFIG_HIBERNATION */
+	default:
+		error = -EINVAL;
+	}
+	return error;
+}
+
+/**
+ *	pm_noirq_op - execute the PM operation appropiate for given PM event
+ *	@dev:	Device.
+ *	@ops:	PM operations to choose from.
+ *	@state: PM transition of the system being carried out.
+ *
+ *	The operation is executed with interrupts disabled by the only remaining
+ *	functional CPU in the system.
+ */
+static int pm_noirq_op(struct device *dev, struct pm_ext_ops *ops,
+			pm_message_t state)
+{
+	int error = 0;
+
+	switch (state.event) {
+#ifdef CONFIG_SUSPEND
+	case PM_EVENT_SUSPEND:
+		if (ops->suspend_noirq) {
+			error = ops->suspend_noirq(dev);
+			suspend_report_result(ops->suspend_noirq, error);
+		}
+		break;
+	case PM_EVENT_RESUME:
+		if (ops->resume_noirq) {
+			error = ops->resume_noirq(dev);
+			suspend_report_result(ops->resume_noirq, error);
+		}
+		break;
+#endif /* CONFIG_SUSPEND */
+#ifdef CONFIG_HIBERNATION
+	case PM_EVENT_FREEZE:
+	case PM_EVENT_QUIESCE:
+		if (ops->freeze_noirq) {
+			error = ops->freeze_noirq(dev);
+			suspend_report_result(ops->freeze_noirq, error);
+		}
+		break;
+	case PM_EVENT_HIBERNATE:
+		if (ops->poweroff_noirq) {
+			error = ops->poweroff_noirq(dev);
+			suspend_report_result(ops->poweroff_noirq, error);
+		}
+		break;
+	case PM_EVENT_THAW:
+	case PM_EVENT_RECOVER:
+		if (ops->thaw_noirq) {
+			error = ops->thaw_noirq(dev);
+			suspend_report_result(ops->thaw_noirq, error);
+		}
+		break;
+	case PM_EVENT_RESTORE:
+		if (ops->restore_noirq) {
+			error = ops->restore_noirq(dev);
+			suspend_report_result(ops->restore_noirq, error);
+		}
+		break;
+#endif /* CONFIG_HIBERNATION */
+	default:
+		error = -EINVAL;
+	}
+	return error;
+}
+
+static char *pm_verb(int event)
+{
+	switch (event) {
+	case PM_EVENT_SUSPEND:
+		return "suspend";
+	case PM_EVENT_RESUME:
+		return "resume";
+	case PM_EVENT_FREEZE:
+		return "freeze";
+	case PM_EVENT_QUIESCE:
+		return "quiesce";
+	case PM_EVENT_HIBERNATE:
+		return "hibernate";
+	case PM_EVENT_THAW:
+		return "thaw";
+	case PM_EVENT_RESTORE:
+		return "restore";
+	case PM_EVENT_RECOVER:
+		return "recover";
+	default:
+		return "(unknown PM event)";
+	}
+}
+
+static void pm_dev_dbg(struct device *dev, pm_message_t state, char *info)
+{
+	dev_dbg(dev, "%s%s%s\n", info, pm_verb(state.event),
+		((state.event & PM_EVENT_SLEEP) && device_may_wakeup(dev)) ?
+		", may wakeup" : "");
+}
+
+static void pm_dev_err(struct device *dev, pm_message_t state, char *info,
+			int error)
+{
+	printk(KERN_ERR "PM: Device %s failed to %s%s: error %d\n",
+		kobject_name(&dev->kobj), pm_verb(state.event), info, error);
+}
+
 /*------------------------- Resume routines -------------------------*/
 
 /**
- *	resume_device_early - Power on one device (early resume).
+ *	resume_device_noirq - Power on one device (early resume).
  *	@dev:	Device.
+ *	@state: PM transition of the system being carried out.
  *
  *	Must be called with interrupts disabled.
  */
-static int resume_device_early(struct device *dev)
+static int resume_device_noirq(struct device *dev, pm_message_t state)
 {
 	int error = 0;
 
 	TRACE_DEVICE(dev);
 	TRACE_RESUME(0);
 
-	if (dev->bus && dev->bus->resume_early) {
-		dev_dbg(dev, "EARLY resume\n");
+	if (!dev->bus)
+		goto End;
+
+	if (dev->bus->pm) {
+		pm_dev_dbg(dev, state, "EARLY ");
+		error = pm_noirq_op(dev, dev->bus->pm, state);
+	} else if (dev->bus->resume_early) {
+		pm_dev_dbg(dev, state, "legacy EARLY ");
 		error = dev->bus->resume_early(dev);
 	}
-
+ End:
 	TRACE_RESUME(error);
 	return error;
 }
 
 /**
  *	dpm_power_up - Power on all regular (non-sysdev) devices.
+ *	@state: PM transition of the system being carried out.
  *
- *	Walk the dpm_off_irq list and power each device up. This
- *	is used for devices that required they be powered down with
- *	interrupts disabled. As devices are powered on, they are moved
- *	to the dpm_off list.
+ *	Execute the appropriate "noirq resume" callback for all devices marked
+ *	as DPM_OFF_IRQ.
  *
  *	Must be called with interrupts disabled and only one CPU running.
  */
-static void dpm_power_up(void)
+static void dpm_power_up(pm_message_t state)
 {
+	struct device *dev;
 
-	while (!list_empty(&dpm_off_irq)) {
-		struct list_head *entry = dpm_off_irq.next;
-		struct device *dev = to_device(entry);
+	list_for_each_entry(dev, &dpm_list, power.entry)
+		if (dev->power.status > DPM_OFF) {
+			int error;
 
-		list_move_tail(entry, &dpm_off);
-		resume_device_early(dev);
-	}
+			dev->power.status = DPM_OFF;
+			error = resume_device_noirq(dev, state);
+			if (error)
+				pm_dev_err(dev, state, " early", error);
+		}
 }
 
 /**
  *	device_power_up - Turn on all devices that need special attention.
+ *	@state: PM transition of the system being carried out.
  *
  *	Power on system devices, then devices that required we shut them down
  *	with interrupts disabled.
  *
  *	Must be called with interrupts disabled.
  */
-void device_power_up(void)
+void device_power_up(pm_message_t state)
 {
 	sysdev_resume();
-	dpm_power_up();
+	dpm_power_up(state);
 }
 EXPORT_SYMBOL_GPL(device_power_up);
 
 /**
  *	resume_device - Restore state for one device.
  *	@dev:	Device.
- *
+ *	@state: PM transition of the system being carried out.
  */
-static int resume_device(struct device *dev)
+static int resume_device(struct device *dev, pm_message_t state)
 {
 	int error = 0;
 
@@ -175,21 +360,40 @@ static int resume_device(struct device *dev)
 
 	down(&dev->sem);
 
-	if (dev->bus && dev->bus->resume) {
-		dev_dbg(dev,"resuming\n");
-		error = dev->bus->resume(dev);
+	if (dev->bus) {
+		if (dev->bus->pm) {
+			pm_dev_dbg(dev, state, "");
+			error = pm_op(dev, &dev->bus->pm->base, state);
+		} else if (dev->bus->resume) {
+			pm_dev_dbg(dev, state, "legacy ");
+			error = dev->bus->resume(dev);
+		}
+		if (error)
+			goto End;
 	}
 
-	if (!error && dev->type && dev->type->resume) {
-		dev_dbg(dev,"resuming\n");
-		error = dev->type->resume(dev);
+	if (dev->type) {
+		if (dev->type->pm) {
+			pm_dev_dbg(dev, state, "type ");
+			error = pm_op(dev, dev->type->pm, state);
+		} else if (dev->type->resume) {
+			pm_dev_dbg(dev, state, "legacy type ");
+			error = dev->type->resume(dev);
+		}
+		if (error)
+			goto End;
 	}
 
-	if (!error && dev->class && dev->class->resume) {
-		dev_dbg(dev,"class resume\n");
-		error = dev->class->resume(dev);
+	if (dev->class) {
+		if (dev->class->pm) {
+			pm_dev_dbg(dev, state, "class ");
+			error = pm_op(dev, dev->class->pm, state);
+		} else if (dev->class->resume) {
+			pm_dev_dbg(dev, state, "legacy class ");
+			error = dev->class->resume(dev);
+		}
 	}
-
+ End:
 	up(&dev->sem);
 
 	TRACE_RESUME(error);
@@ -198,78 +402,161 @@ static int resume_device(struct device *dev)
 
 /**
  *	dpm_resume - Resume every device.
+ *	@state: PM transition of the system being carried out.
  *
- *	Resume the devices that have either not gone through
- *	the late suspend, or that did go through it but also
- *	went through the early resume.
+ *	Execute the appropriate "resume" callback for all devices the status of
+ *	which indicates that they are inactive.
+ */
+static void dpm_resume(pm_message_t state)
+{
+	struct list_head list;
+
+	INIT_LIST_HEAD(&list);
+	mutex_lock(&dpm_list_mtx);
+	transition_started = false;
+	while (!list_empty(&dpm_list)) {
+		struct device *dev = to_device(dpm_list.next);
+
+		get_device(dev);
+		if (dev->power.status >= DPM_OFF) {
+			int error;
+
+			dev->power.status = DPM_RESUMING;
+			mutex_unlock(&dpm_list_mtx);
+
+			error = resume_device(dev, state);
+
+			mutex_lock(&dpm_list_mtx);
+			if (error)
+				pm_dev_err(dev, state, "", error);
+		} else if (dev->power.status == DPM_SUSPENDING) {
+			/* Allow new children of the device to be registered */
+			dev->power.status = DPM_RESUMING;
+		}
+		if (!list_empty(&dev->power.entry))
+			list_move_tail(&dev->power.entry, &list);
+		put_device(dev);
+	}
+	list_splice(&list, &dpm_list);
+	mutex_unlock(&dpm_list_mtx);
+}
+
+/**
+ *	complete_device - Complete a PM transition for given device
+ *	@dev:	Device.
+ *	@state: PM transition of the system being carried out.
+ */
+static void complete_device(struct device *dev, pm_message_t state)
+{
+	down(&dev->sem);
+
+	if (dev->class && dev->class->pm && dev->class->pm->complete) {
+		pm_dev_dbg(dev, state, "completing class ");
+		dev->class->pm->complete(dev);
+	}
+
+	if (dev->type && dev->type->pm && dev->type->pm->complete) {
+		pm_dev_dbg(dev, state, "completing type ");
+		dev->type->pm->complete(dev);
+	}
+
+	if (dev->bus && dev->bus->pm && dev->bus->pm->base.complete) {
+		pm_dev_dbg(dev, state, "completing ");
+		dev->bus->pm->base.complete(dev);
+	}
+
+	up(&dev->sem);
+}
+
+/**
+ *	dpm_complete - Complete a PM transition for all devices.
+ *	@state: PM transition of the system being carried out.
  *
- *	Take devices from the dpm_off_list, resume them,
- *	and put them on the dpm_locked list.
+ *	Execute the ->complete() callbacks for all devices that are not marked
+ *	as DPM_ON.
  */
-static void dpm_resume(void)
+static void dpm_complete(pm_message_t state)
 {
+	struct list_head list;
+
+	INIT_LIST_HEAD(&list);
 	mutex_lock(&dpm_list_mtx);
-	all_sleeping = false;
-	while(!list_empty(&dpm_off)) {
-		struct list_head *entry = dpm_off.next;
-		struct device *dev = to_device(entry);
+	while (!list_empty(&dpm_list)) {
+		struct device *dev = to_device(dpm_list.prev);
 
-		list_move_tail(entry, &dpm_active);
-		dev->power.sleeping = false;
-		mutex_unlock(&dpm_list_mtx);
-		resume_device(dev);
-		mutex_lock(&dpm_list_mtx);
+		get_device(dev);
+		if (dev->power.status > DPM_ON) {
+			dev->power.status = DPM_ON;
+			mutex_unlock(&dpm_list_mtx);
+
+			complete_device(dev, state);
+
+			mutex_lock(&dpm_list_mtx);
+		}
+		if (!list_empty(&dev->power.entry))
+			list_move(&dev->power.entry, &list);
+		put_device(dev);
 	}
+	list_splice(&list, &dpm_list);
 	mutex_unlock(&dpm_list_mtx);
 }
 
 /**
  *	device_resume - Restore state of each device in system.
+ *	@state: PM transition of the system being carried out.
  *
  *	Resume all the devices, unlock them all, and allow new
  *	devices to be registered once again.
  */
-void device_resume(void)
+void device_resume(pm_message_t state)
 {
 	might_sleep();
-	dpm_resume();
+	dpm_resume(state);
+	dpm_complete(state);
 }
 EXPORT_SYMBOL_GPL(device_resume);
 
 
 /*------------------------- Suspend routines -------------------------*/
 
-static inline char *suspend_verb(u32 event)
+/**
+ *	resume_event - return a PM message representing the resume event
+ *	               corresponding to given sleep state.
+ *	@sleep_state: PM message representing a sleep state.
+ */
+static pm_message_t resume_event(pm_message_t sleep_state)
 {
-	switch (event) {
-	case PM_EVENT_SUSPEND:	return "suspend";
-	case PM_EVENT_FREEZE:	return "freeze";
-	case PM_EVENT_PRETHAW:	return "prethaw";
-	default:		return "(unknown suspend event)";
+	switch (sleep_state.event) {
+	case PM_EVENT_SUSPEND:
+		return PMSG_RESUME;
+	case PM_EVENT_FREEZE:
+	case PM_EVENT_QUIESCE:
+		return PMSG_RECOVER;
+	case PM_EVENT_HIBERNATE:
+		return PMSG_RESTORE;
 	}
-}
-
-static void
-suspend_device_dbg(struct device *dev, pm_message_t state, char *info)
-{
-	dev_dbg(dev, "%s%s%s\n", info, suspend_verb(state.event),
-		((state.event == PM_EVENT_SUSPEND) && device_may_wakeup(dev)) ?
-		", may wakeup" : "");
+	return PMSG_ON;
 }
 
 /**
- *	suspend_device_late - Shut down one device (late suspend).
+ *	suspend_device_noirq - Shut down one device (late suspend).
  *	@dev:	Device.
- *	@state:	Power state device is entering.
+ *	@state: PM transition of the system being carried out.
  *
  *	This is called with interrupts off and only a single CPU running.
  */
-static int suspend_device_late(struct device *dev, pm_message_t state)
+static int suspend_device_noirq(struct device *dev, pm_message_t state)
 {
 	int error = 0;
 
-	if (dev->bus && dev->bus->suspend_late) {
-		suspend_device_dbg(dev, state, "LATE ");
+	if (!dev->bus)
+		return 0;
+
+	if (dev->bus->pm) {
+		pm_dev_dbg(dev, state, "LATE ");
+		error = pm_noirq_op(dev, dev->bus->pm, state);
+	} else if (dev->bus->suspend_late) {
+		pm_dev_dbg(dev, state, "legacy LATE ");
 		error = dev->bus->suspend_late(dev, state);
 		suspend_report_result(dev->bus->suspend_late, error);
 	}
@@ -278,37 +565,30 @@ static int suspend_device_late(struct device *dev, pm_message_t state)
 
 /**
  *	device_power_down - Shut down special devices.
- *	@state:		Power state to enter.
+ *	@state: PM transition of the system being carried out.
  *
- *	Power down devices that require interrupts to be disabled
- *	and move them from the dpm_off list to the dpm_off_irq list.
+ *	Power down devices that require interrupts to be disabled.
  *	Then power down system devices.
  *
  *	Must be called with interrupts disabled and only one CPU running.
  */
 int device_power_down(pm_message_t state)
 {
+	struct device *dev;
 	int error = 0;
 
-	while (!list_empty(&dpm_off)) {
-		struct list_head *entry = dpm_off.prev;
-		struct device *dev = to_device(entry);
-
-		error = suspend_device_late(dev, state);
+	list_for_each_entry_reverse(dev, &dpm_list, power.entry) {
+		error = suspend_device_noirq(dev, state);
 		if (error) {
-			printk(KERN_ERR "Could not power down device %s: "
-					"error %d\n",
-					kobject_name(&dev->kobj), error);
+			pm_dev_err(dev, state, " late", error);
 			break;
 		}
-		if (!list_empty(&dev->power.entry))
-			list_move(&dev->power.entry, &dpm_off_irq);
+		dev->power.status = DPM_OFF_IRQ;
 	}
-
 	if (!error)
 		error = sysdev_suspend(state);
 	if (error)
-		dpm_power_up();
+		dpm_power_up(resume_event(state));
 	return error;
 }
 EXPORT_SYMBOL_GPL(device_power_down);
@@ -316,7 +596,7 @@ EXPORT_SYMBOL_GPL(device_power_down);
 /**
  *	suspend_device - Save state of one device.
  *	@dev:	Device.
- *	@state:	Power state device is entering.
+ *	@state: PM transition of the system being carried out.
  */
 static int suspend_device(struct device *dev, pm_message_t state)
 {
@@ -324,24 +604,43 @@ static int suspend_device(struct device *dev, pm_message_t state)
 
 	down(&dev->sem);
 
-	if (dev->class && dev->class->suspend) {
-		suspend_device_dbg(dev, state, "class ");
-		error = dev->class->suspend(dev, state);
-		suspend_report_result(dev->class->suspend, error);
+	if (dev->class) {
+		if (dev->class->pm) {
+			pm_dev_dbg(dev, state, "class ");
+			error = pm_op(dev, dev->class->pm, state);
+		} else if (dev->class->suspend) {
+			pm_dev_dbg(dev, state, "legacy class ");
+			error = dev->class->suspend(dev, state);
+			suspend_report_result(dev->class->suspend, error);
+		}
+		if (error)
+			goto End;
 	}
 
-	if (!error && dev->type && dev->type->suspend) {
-		suspend_device_dbg(dev, state, "type ");
-		error = dev->type->suspend(dev, state);
-		suspend_report_result(dev->type->suspend, error);
+	if (dev->type) {
+		if (dev->type->pm) {
+			pm_dev_dbg(dev, state, "type ");
+			error = pm_op(dev, dev->type->pm, state);
+		} else if (dev->type->suspend) {
+			pm_dev_dbg(dev, state, "legacy type ");
+			error = dev->type->suspend(dev, state);
+			suspend_report_result(dev->type->suspend, error);
+		}
+		if (error)
+			goto End;
 	}
 
-	if (!error && dev->bus && dev->bus->suspend) {
-		suspend_device_dbg(dev, state, "");
-		error = dev->bus->suspend(dev, state);
-		suspend_report_result(dev->bus->suspend, error);
+	if (dev->bus) {
+		if (dev->bus->pm) {
+			pm_dev_dbg(dev, state, "");
+			error = pm_op(dev, &dev->bus->pm->base, state);
+		} else if (dev->bus->suspend) {
+			pm_dev_dbg(dev, state, "legacy ");
+			error = dev->bus->suspend(dev, state);
+			suspend_report_result(dev->bus->suspend, error);
+		}
 	}
-
+ End:
 	up(&dev->sem);
 
 	return error;
@@ -349,67 +648,141 @@ static int suspend_device(struct device *dev, pm_message_t state)
 
 /**
  *	dpm_suspend - Suspend every device.
- *	@state:	Power state to put each device in.
+ *	@state: PM transition of the system being carried out.
  *
- *	Walk the dpm_locked list.  Suspend each device and move it
- *	to the dpm_off list.
- *
- *	(For historical reasons, if it returns -EAGAIN, that used to mean
- *	that the device would be called again with interrupts disabled.
- *	These days, we use the "suspend_late()" callback for that, so we
- *	print a warning and consider it an error).
+ *	Execute the appropriate "suspend" callbacks for all devices.
  */
 static int dpm_suspend(pm_message_t state)
 {
+	struct list_head list;
 	int error = 0;
 
+	INIT_LIST_HEAD(&list);
 	mutex_lock(&dpm_list_mtx);
-	while (!list_empty(&dpm_active)) {
-		struct list_head *entry = dpm_active.prev;
-		struct device *dev = to_device(entry);
-
-		WARN_ON(dev->parent && dev->parent->power.sleeping);
+	while (!list_empty(&dpm_list)) {
+		struct device *dev = to_device(dpm_list.prev);
 
-		dev->power.sleeping = true;
+		get_device(dev);
 		mutex_unlock(&dpm_list_mtx);
+
 		error = suspend_device(dev, state);
+
 		mutex_lock(&dpm_list_mtx);
 		if (error) {
-			printk(KERN_ERR "Could not suspend device %s: "
-					"error %d%s\n",
-					kobject_name(&dev->kobj),
-					error,
-					(error == -EAGAIN ?
-					" (please convert to suspend_late)" :
-					""));
-			dev->power.sleeping = false;
+			pm_dev_err(dev, state, "", error);
+			put_device(dev);
 			break;
 		}
+		dev->power.status = DPM_OFF;
 		if (!list_empty(&dev->power.entry))
-			list_move(&dev->power.entry, &dpm_off);
+			list_move(&dev->power.entry, &list);
+		put_device(dev);
 	}
-	if (!error)
-		all_sleeping = true;
+	list_splice(&list, dpm_list.prev);
 	mutex_unlock(&dpm_list_mtx);
+	return error;
+}
+
+/**
+ *	prepare_device - Execute the ->prepare() callback(s) for given device.
+ *	@dev:	Device.
+ *	@state: PM transition of the system being carried out.
+ */
+static int prepare_device(struct device *dev, pm_message_t state)
+{
+	int error = 0;
+
+	down(&dev->sem);
+
+	if (dev->bus && dev->bus->pm && dev->bus->pm->base.prepare) {
+		pm_dev_dbg(dev, state, "preparing ");
+		error = dev->bus->pm->base.prepare(dev);
+		suspend_report_result(dev->bus->pm->base.prepare, error);
+		if (error)
+			goto End;
+	}
+
+	if (dev->type && dev->type->pm && dev->type->pm->prepare) {
+		pm_dev_dbg(dev, state, "preparing type ");
+		error = dev->type->pm->prepare(dev);
+		suspend_report_result(dev->type->pm->prepare, error);
+		if (error)
+			goto End;
+	}
+
+	if (dev->class && dev->class->pm && dev->class->pm->prepare) {
+		pm_dev_dbg(dev, state, "preparing class ");
+		error = dev->class->pm->prepare(dev);
+		suspend_report_result(dev->class->pm->prepare, error);
+	}
+ End:
+	up(&dev->sem);
+
+	return error;
+}
 
+/**
+ *	dpm_prepare - Prepare all devices for a PM transition.
+ *	@state: PM transition of the system being carried out.
+ *
+ *	Execute the ->prepare() callback for all devices.
+ */
+static int dpm_prepare(pm_message_t state)
+{
+	struct list_head list;
+	int error = 0;
+
+	INIT_LIST_HEAD(&list);
+	mutex_lock(&dpm_list_mtx);
+	transition_started = true;
+	while (!list_empty(&dpm_list)) {
+		struct device *dev = to_device(dpm_list.next);
+
+		get_device(dev);
+		dev->power.status = DPM_PREPARING;
+		mutex_unlock(&dpm_list_mtx);
+
+		error = prepare_device(dev, state);
+
+		mutex_lock(&dpm_list_mtx);
+		if (error) {
+			dev->power.status = DPM_ON;
+			if (error == -EAGAIN) {
+				put_device(dev);
+				continue;
+			}
+			printk(KERN_ERR "PM: Failed to prepare device %s "
+				"for power transition: error %d\n",
+				kobject_name(&dev->kobj), error);
+			put_device(dev);
+			break;
+		}
+		dev->power.status = DPM_SUSPENDING;
+		if (!list_empty(&dev->power.entry))
+			list_move_tail(&dev->power.entry, &list);
+		put_device(dev);
+	}
+	list_splice(&list, &dpm_list);
+	mutex_unlock(&dpm_list_mtx);
 	return error;
 }
 
 /**
  *	device_suspend - Save state and stop all devices in system.
- *	@state: new power management state
+ *	@state: PM transition of the system being carried out.
  *
- *	Prevent new devices from being registered, then lock all devices
- *	and suspend them.
+ *	Prepare and suspend all devices.
  */
 int device_suspend(pm_message_t state)
 {
 	int error;
 
 	might_sleep();
-	error = dpm_suspend(state);
+	error = dpm_prepare(state);
+	if (!error)
+		error = dpm_suspend(state);
 	if (error)
-		device_resume();
+		device_resume(resume_event(state));
 	return error;
 }
 EXPORT_SYMBOL_GPL(device_suspend);
diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h
index a6894f2a4b99..a3252c0e2887 100644
--- a/drivers/base/power/power.h
+++ b/drivers/base/power/power.h
@@ -4,7 +4,7 @@
  * main.c
  */
 
-extern struct list_head dpm_active;	/* The active device list */
+extern struct list_head dpm_list;	/* The active device list */
 
 static inline struct device *to_device(struct list_head *entry)
 {
diff --git a/drivers/base/power/trace.c b/drivers/base/power/trace.c
index 2b4b392dcbc1..8c1e656b5f8b 100644
--- a/drivers/base/power/trace.c
+++ b/drivers/base/power/trace.c
@@ -188,9 +188,9 @@ static int show_file_hash(unsigned int value)
 static int show_dev_hash(unsigned int value)
 {
 	int match = 0;
-	struct list_head * entry = dpm_active.prev;
+	struct list_head *entry = dpm_list.prev;
 
-	while (entry != &dpm_active) {
+	while (entry != &dpm_list) {
 		struct device * dev = to_device(entry);
 		unsigned int hash = hash_string(DEVSEED, dev->bus_id, DEVHASH);
 		if (hash == value) {
diff --git a/include/linux/device.h b/include/linux/device.h
index 6a2d04c011bc..f71a78d123ae 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -68,6 +68,8 @@ struct bus_type {
 	int (*resume_early)(struct device *dev);
 	int (*resume)(struct device *dev);
 
+	struct pm_ext_ops *pm;
+
 	struct bus_type_private *p;
 };
 
@@ -131,6 +133,8 @@ struct device_driver {
 	int (*resume) (struct device *dev);
 	struct attribute_group **groups;
 
+	struct pm_ops *pm;
+
 	struct driver_private *p;
 };
 
@@ -197,6 +201,8 @@ struct class {
 
 	int (*suspend)(struct device *dev, pm_message_t state);
 	int (*resume)(struct device *dev);
+
+	struct pm_ops *pm;
 };
 
 extern int __must_check class_register(struct class *class);
@@ -248,8 +254,11 @@ struct device_type {
 	struct attribute_group **groups;
 	int (*uevent)(struct device *dev, struct kobj_uevent_env *env);
 	void (*release)(struct device *dev);
+
 	int (*suspend)(struct device *dev, pm_message_t state);
 	int (*resume)(struct device *dev);
+
+	struct pm_ops *pm;
 };
 
 /* interface for exporting device attributes */
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 39a7ee859b67..4ad9de94449a 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -112,7 +112,9 @@ typedef struct pm_message {
 	int event;
 } pm_message_t;
 
-/*
+/**
+ * struct pm_ops - device PM callbacks
+ *
  * Several driver power state transitions are externally visible, affecting
  * the state of pending I/O queues and (for drivers that touch hardware)
  * interrupts, wakeups, DMA, and other hardware state.  There may also be
@@ -120,6 +122,284 @@ typedef struct pm_message {
  * to the rest of the driver stack (such as a driver that's ON gating off
  * clocks which are not in active use).
  *
+ * The externally visible transitions are handled with the help of the following
+ * callbacks included in this structure:
+ *
+ * @prepare: Prepare the device for the upcoming transition, but do NOT change
+ *	its hardware state.  Prevent new children of the device from being
+ *	registered after @prepare() returns (the driver's subsystem and
+ *	generally the rest of the kernel is supposed to prevent new calls to the
+ *	probe method from being made too once @prepare() has succeeded).  If
+ *	@prepare() detects a situation it cannot handle (e.g. registration of a
+ *	child already in progress), it may return -EAGAIN, so that the PM core
+ *	can execute it once again (e.g. after the new child has been registered)
+ *	to recover from the race condition.  This method is executed for all
+ *	kinds of suspend transitions and is followed by one of the suspend
+ *	callbacks: @suspend(), @freeze(), or @poweroff().
+ *	The PM core executes @prepare() for all devices before starting to
+ *	execute suspend callbacks for any of them, so drivers may assume all of
+ *	the other devices to be present and functional while @prepare() is being
+ *	executed.  In particular, it is safe to make GFP_KERNEL memory
+ *	allocations from within @prepare().  However, drivers may NOT assume
+ *	anything about the availability of the user space at that time and it
+ *	is not correct to request firmware from within @prepare() (it's too
+ *	late to do that).  [To work around this limitation, drivers may
+ *	register suspend and hibernation notifiers that are executed before the
+ *	freezing of tasks.]
+ *
+ * @complete: Undo the changes made by @prepare().  This method is executed for
+ *	all kinds of resume transitions, following one of the resume callbacks:
+ *	@resume(), @thaw(), @restore().  Also called if the state transition
+ *	fails before the driver's suspend callback (@suspend(), @freeze(),
+ *	@poweroff()) can be executed (e.g. if the suspend callback fails for one
+ *	of the other devices that the PM core has unsuccessfully attempted to
+ *	suspend earlier).
+ *	The PM core executes @complete() after it has executed the appropriate
+ *	resume callback for all devices.
+ *
+ * @suspend: Executed before putting the system into a sleep state in which the
+ *	contents of main memory are preserved.  Quiesce the device, put it into
+ *	a low power state appropriate for the upcoming system state (such as
+ *	PCI_D3hot), and enable wakeup events as appropriate.
+ *
+ * @resume: Executed after waking the system up from a sleep state in which the
+ *	contents of main memory were preserved.  Put the device into the
+ *	appropriate state, according to the information saved in memory by the
+ *	preceding @suspend().  The driver starts working again, responding to
+ *	hardware events and software requests.  The hardware may have gone
+ *	through a power-off reset, or it may have maintained state from the
+ *	previous suspend() which the driver may rely on while resuming.  On most
+ *	platforms, there are no restrictions on availability of resources like
+ *	clocks during @resume().
+ *
+ * @freeze: Hibernation-specific, executed before creating a hibernation image.
+ *	Quiesce operations so that a consistent image can be created, but do NOT
+ *	otherwise put the device into a low power device state and do NOT emit
+ *	system wakeup events.  Save in main memory the device settings to be
+ *	used by @restore() during the subsequent resume from hibernation or by
+ *	the subsequent @thaw(), if the creation of the image or the restoration
+ *	of main memory contents from it fails.
+ *
+ * @thaw: Hibernation-specific, executed after creating a hibernation image OR
+ *	if the creation of the image fails.  Also executed after a failing
+ *	attempt to restore the contents of main memory from such an image.
+ *	Undo the changes made by the preceding @freeze(), so the device can be
+ *	operated in the same way as immediately before the call to @freeze().
+ *
+ * @poweroff: Hibernation-specific, executed after saving a hibernation image.
+ *	Quiesce the device, put it into a low power state appropriate for the
+ *	upcoming system state (such as PCI_D3hot), and enable wakeup events as
+ *	appropriate.
+ *
+ * @restore: Hibernation-specific, executed after restoring the contents of main
+ *	memory from a hibernation image.  Driver starts working again,
+ *	responding to hardware events and software requests.  Drivers may NOT
+ *	make ANY assumptions about the hardware state right prior to @restore().
+ *	On most platforms, there are no restrictions on availability of
+ *	resources like clocks during @restore().
+ *
+ * All of the above callbacks, except for @complete(), return error codes.
+ * However, the error codes returned by the resume operations, @resume(),
+ * @thaw(), and @restore(), do not cause the PM core to abort the resume
+ * transition during which they are returned.  The error codes returned in
+ * that cases are only printed by the PM core to the system logs for debugging
+ * purposes.  Still, it is recommended that drivers only return error codes
+ * from their resume methods in case of an unrecoverable failure (i.e. when the
+ * device being handled refuses to resume and becomes unusable) to allow us to
+ * modify the PM core in the future, so that it can avoid attempting to handle
+ * devices that failed to resume and their children.
+ *
+ * It is allowed to unregister devices while the above callbacks are being
+ * executed.  However, it is not allowed to unregister a device from within any
+ * of its own callbacks.
+ */
+
+struct pm_ops {
+	int (*prepare)(struct device *dev);
+	void (*complete)(struct device *dev);
+	int (*suspend)(struct device *dev);
+	int (*resume)(struct device *dev);
+	int (*freeze)(struct device *dev);
+	int (*thaw)(struct device *dev);
+	int (*poweroff)(struct device *dev);
+	int (*restore)(struct device *dev);
+};
+
+/**
+ * struct pm_ext_ops - extended device PM callbacks
+ *
+ * Some devices require certain operations related to suspend and hibernation
+ * to be carried out with interrupts disabled.  Thus, 'struct pm_ext_ops' below
+ * is defined, adding callbacks to be executed with interrupts disabled to
+ * 'struct pm_ops'.
+ *
+ * The following callbacks included in 'struct pm_ext_ops' are executed with
+ * the nonboot CPUs switched off and with interrupts disabled on the only
+ * functional CPU.  They also are executed with the PM core list of devices
+ * locked, so they must NOT unregister any devices.
+ *
+ * @suspend_noirq: Complete the operations of ->suspend() by carrying out any
+ *	actions required for suspending the device that need interrupts to be
+ *	disabled
+ *
+ * @resume_noirq: Prepare for the execution of ->resume() by carrying out any
+ *	actions required for resuming the device that need interrupts to be
+ *	disabled
+ *
+ * @freeze_noirq: Complete the operations of ->freeze() by carrying out any
+ *	actions required for freezing the device that need interrupts to be
+ *	disabled
+ *
+ * @thaw_noirq: Prepare for the execution of ->thaw() by carrying out any
+ *	actions required for thawing the device that need interrupts to be
+ *	disabled
+ *
+ * @poweroff_noirq: Complete the operations of ->poweroff() by carrying out any
+ *	actions required for handling the device that need interrupts to be
+ *	disabled
+ *
+ * @restore_noirq: Prepare for the execution of ->restore() by carrying out any
+ *	actions required for restoring the operations of the device that need
+ *	interrupts to be disabled
+ *
+ * All of the above callbacks return error codes, but the error codes returned
+ * by the resume operations, @resume_noirq(), @thaw_noirq(), and
+ * @restore_noirq(), do not cause the PM core to abort the resume transition
+ * during which they are returned.  The error codes returned in that cases are
+ * only printed by the PM core to the system logs for debugging purposes.
+ * Still, as stated above, it is recommended that drivers only return error
+ * codes from their resume methods if the device being handled fails to resume
+ * and is not usable any more.
+ */
+
+struct pm_ext_ops {
+	struct pm_ops base;
+	int (*suspend_noirq)(struct device *dev);
+	int (*resume_noirq)(struct device *dev);
+	int (*freeze_noirq)(struct device *dev);
+	int (*thaw_noirq)(struct device *dev);
+	int (*poweroff_noirq)(struct device *dev);
+	int (*restore_noirq)(struct device *dev);
+};
+
+/**
+ * PM_EVENT_ messages
+ *
+ * The following PM_EVENT_ messages are defined for the internal use of the PM
+ * core, in order to provide a mechanism allowing the high level suspend and
+ * hibernation code to convey the necessary information to the device PM core
+ * code:
+ *
+ * ON		No transition.
+ *
+ * FREEZE 	System is going to hibernate, call ->prepare() and ->freeze()
+ *		for all devices.
+ *
+ * SUSPEND	System is going to suspend, call ->prepare() and ->suspend()
+ *		for all devices.
+ *
+ * HIBERNATE	Hibernation image has been saved, call ->prepare() and
+ *		->poweroff() for all devices.
+ *
+ * QUIESCE	Contents of main memory are going to be restored from a (loaded)
+ *		hibernation image, call ->prepare() and ->freeze() for all
+ *		devices.
+ *
+ * RESUME	System is resuming, call ->resume() and ->complete() for all
+ *		devices.
+ *
+ * THAW		Hibernation image has been created, call ->thaw() and
+ *		->complete() for all devices.
+ *
+ * RESTORE	Contents of main memory have been restored from a hibernation
+ *		image, call ->restore() and ->complete() for all devices.
+ *
+ * RECOVER	Creation of a hibernation image or restoration of the main
+ *		memory contents from a hibernation image has failed, call
+ *		->thaw() and ->complete() for all devices.
+ */
+
+#define PM_EVENT_ON		0x0000
+#define PM_EVENT_FREEZE 	0x0001
+#define PM_EVENT_SUSPEND	0x0002
+#define PM_EVENT_HIBERNATE	0x0004
+#define PM_EVENT_QUIESCE	0x0008
+#define PM_EVENT_RESUME		0x0010
+#define PM_EVENT_THAW		0x0020
+#define PM_EVENT_RESTORE	0x0040
+#define PM_EVENT_RECOVER	0x0080
+
+#define PM_EVENT_SLEEP	(PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE)
+
+#define PMSG_FREEZE	((struct pm_message){ .event = PM_EVENT_FREEZE, })
+#define PMSG_QUIESCE	((struct pm_message){ .event = PM_EVENT_QUIESCE, })
+#define PMSG_SUSPEND	((struct pm_message){ .event = PM_EVENT_SUSPEND, })
+#define PMSG_HIBERNATE	((struct pm_message){ .event = PM_EVENT_HIBERNATE, })
+#define PMSG_RESUME	((struct pm_message){ .event = PM_EVENT_RESUME, })
+#define PMSG_THAW	((struct pm_message){ .event = PM_EVENT_THAW, })
+#define PMSG_RESTORE	((struct pm_message){ .event = PM_EVENT_RESTORE, })
+#define PMSG_RECOVER	((struct pm_message){ .event = PM_EVENT_RECOVER, })
+#define PMSG_ON		((struct pm_message){ .event = PM_EVENT_ON, })
+
+/**
+ * Device power management states
+ *
+ * These state labels are used internally by the PM core to indicate the current
+ * status of a device with respect to the PM core operations.
+ *
+ * DPM_ON		Device is regarded as operational.  Set this way
+ *			initially and when ->complete() is about to be called.
+ *			Also set when ->prepare() fails.
+ *
+ * DPM_PREPARING	Device is going to be prepared for a PM transition.  Set
+ *			when ->prepare() is about to be called.
+ *
+ * DPM_RESUMING		Device is going to be resumed.  Set when ->resume(),
+ *			->thaw(), or ->restore() is about to be called.
+ *
+ * DPM_SUSPENDING	Device has been prepared for a power transition.  Set
+ *			when ->prepare() has just succeeded.
+ *
+ * DPM_OFF		Device is regarded as inactive.  Set immediately after
+ *			->suspend(), ->freeze(), or ->poweroff() has succeeded.
+ *			Also set when ->resume()_noirq, ->thaw_noirq(), or
+ *			->restore_noirq() is about to be called.
+ *
+ * DPM_OFF_IRQ		Device is in a "deep sleep".  Set immediately after
+ *			->suspend_noirq(), ->freeze_noirq(), or
+ *			->poweroff_noirq() has just succeeded.
+ */
+
+enum dpm_state {
+	DPM_INVALID,
+	DPM_ON,
+	DPM_PREPARING,
+	DPM_RESUMING,
+	DPM_SUSPENDING,
+	DPM_OFF,
+	DPM_OFF_IRQ,
+};
+
+struct dev_pm_info {
+	pm_message_t		power_state;
+	unsigned		can_wakeup:1;
+	unsigned		should_wakeup:1;
+	enum dpm_state		status;		/* Owned by the PM core */
+#ifdef	CONFIG_PM_SLEEP
+	struct list_head	entry;
+#endif
+};
+
+/*
+ * The PM_EVENT_ messages are also used by drivers implementing the legacy
+ * suspend framework, based on the ->suspend() and ->resume() callbacks common
+ * for suspend and hibernation transitions, according to the rules below.
+ */
+
+/* Necessary, because several drivers use PM_EVENT_PRETHAW */
+#define PM_EVENT_PRETHAW PM_EVENT_QUIESCE
+
+/*
  * One transition is triggered by resume(), after a suspend() call; the
  * message is implicit:
  *
@@ -164,35 +444,13 @@ typedef struct pm_message {
  * or from system low-power states such as standby or suspend-to-RAM.
  */
 
-#define PM_EVENT_ON 0
-#define PM_EVENT_FREEZE 1
-#define PM_EVENT_SUSPEND 2
-#define PM_EVENT_HIBERNATE 4
-#define PM_EVENT_PRETHAW 8
-
-#define PM_EVENT_SLEEP	(PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE)
-
-#define PMSG_FREEZE	((struct pm_message){ .event = PM_EVENT_FREEZE, })
-#define PMSG_PRETHAW	((struct pm_message){ .event = PM_EVENT_PRETHAW, })
-#define PMSG_SUSPEND	((struct pm_message){ .event = PM_EVENT_SUSPEND, })
-#define PMSG_HIBERNATE	((struct pm_message){ .event = PM_EVENT_HIBERNATE, })
-#define PMSG_ON		((struct pm_message){ .event = PM_EVENT_ON, })
-
-struct dev_pm_info {
-	pm_message_t		power_state;
-	unsigned		can_wakeup:1;
-	unsigned		should_wakeup:1;
-	bool			sleeping:1;	/* Owned by the PM core */
-#ifdef	CONFIG_PM_SLEEP
-	struct list_head	entry;
-#endif
-};
+#ifdef CONFIG_PM_SLEEP
+extern void device_pm_lock(void);
+extern void device_power_up(pm_message_t state);
+extern void device_resume(pm_message_t state);
 
+extern void device_pm_unlock(void);
 extern int device_power_down(pm_message_t state);
-extern void device_power_up(void);
-extern void device_resume(void);
-
-#ifdef CONFIG_PM_SLEEP
 extern int device_suspend(pm_message_t state);
 extern int device_prepare_suspend(pm_message_t state);
 
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 14a656cdc652..d416be0efa8a 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -193,6 +193,7 @@ static int create_image(int platform_mode)
 	if (error)
 		return error;
 
+	device_pm_lock();
 	local_irq_disable();
 	/* At this point, device_suspend() has been called, but *not*
 	 * device_power_down(). We *must* call device_power_down() now.
@@ -224,9 +225,11 @@ static int create_image(int platform_mode)
 	/* NOTE:  device_power_up() is just a resume() for devices
 	 * that suspended with irqs off ... no overall powerup.
 	 */
-	device_power_up();
+	device_power_up(in_suspend ?
+		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
  Enable_irqs:
 	local_irq_enable();
+	device_pm_unlock();
 	return error;
 }
 
@@ -280,7 +283,8 @@ int hibernation_snapshot(int platform_mode)
  Finish:
 	platform_finish(platform_mode);
  Resume_devices:
-	device_resume();
+	device_resume(in_suspend ?
+		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
  Resume_console:
 	resume_console();
  Close:
@@ -300,8 +304,9 @@ static int resume_target_kernel(void)
 {
 	int error;
 
+	device_pm_lock();
 	local_irq_disable();
-	error = device_power_down(PMSG_PRETHAW);
+	error = device_power_down(PMSG_QUIESCE);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to power down, "
 			"aborting resume\n");
@@ -329,9 +334,10 @@ static int resume_target_kernel(void)
 	swsusp_free();
 	restore_processor_state();
 	touch_softlockup_watchdog();
-	device_power_up();
+	device_power_up(PMSG_RECOVER);
  Enable_irqs:
 	local_irq_enable();
+	device_pm_unlock();
 	return error;
 }
 
@@ -350,7 +356,7 @@ int hibernation_restore(int platform_mode)
 
 	pm_prepare_console();
 	suspend_console();
-	error = device_suspend(PMSG_PRETHAW);
+	error = device_suspend(PMSG_QUIESCE);
 	if (error)
 		goto Finish;
 
@@ -362,7 +368,7 @@ int hibernation_restore(int platform_mode)
 		enable_nonboot_cpus();
 	}
 	platform_restore_cleanup(platform_mode);
-	device_resume();
+	device_resume(PMSG_RECOVER);
  Finish:
 	resume_console();
 	pm_restore_console();
@@ -403,6 +409,7 @@ int hibernation_platform_enter(void)
 	if (error)
 		goto Finish;
 
+	device_pm_lock();
 	local_irq_disable();
 	error = device_power_down(PMSG_HIBERNATE);
 	if (!error) {
@@ -411,6 +418,7 @@ int hibernation_platform_enter(void)
 		while (1);
 	}
 	local_irq_enable();
+	device_pm_unlock();
 
 	/*
 	 * We don't need to reenable the nonboot CPUs or resume consoles, since
@@ -419,7 +427,7 @@ int hibernation_platform_enter(void)
  Finish:
 	hibernation_ops->finish();
  Resume_devices:
-	device_resume();
+	device_resume(PMSG_RESTORE);
  Resume_console:
 	resume_console();
  Close:
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6a6d5eb3524e..d023b6b584e5 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -228,6 +228,7 @@ static int suspend_enter(suspend_state_t state)
 {
 	int error = 0;
 
+	device_pm_lock();
 	arch_suspend_disable_irqs();
 	BUG_ON(!irqs_disabled());
 
@@ -239,10 +240,11 @@ static int suspend_enter(suspend_state_t state)
 	if (!suspend_test(TEST_CORE))
 		error = suspend_ops->enter(state);
 
-	device_power_up();
+	device_power_up(PMSG_RESUME);
  Done:
 	arch_suspend_enable_irqs();
 	BUG_ON(irqs_disabled());
+	device_pm_unlock();
 	return error;
 }
 
@@ -291,7 +293,7 @@ int suspend_devices_and_enter(suspend_state_t state)
 	if (suspend_ops->finish)
 		suspend_ops->finish();
  Resume_devices:
-	device_resume();
+	device_resume(PMSG_RESUME);
  Resume_console:
 	resume_console();
  Close:
-- 
cgit v1.2.3


From bbb44d9f23d868a2837c6b22b8dfb123d8e7800c Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Tue, 20 May 2008 00:49:04 +0200
Subject: PCI: implement new suspend/resume callbacks

Implement new suspend and hibernation callbacks for the PCI bus type.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci-driver.c | 392 +++++++++++++++++++++++++++++++++++++++++------
 include/linux/pci.h      |   2 +-
 2 files changed, 347 insertions(+), 47 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 677fd9d6db12..8eb8a3091dc7 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -274,7 +274,57 @@ static int pci_device_remove(struct device * dev)
 	return 0;
 }
 
-static int pci_device_suspend(struct device * dev, pm_message_t state)
+static void pci_device_shutdown(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
+
+	if (drv && drv->shutdown)
+		drv->shutdown(pci_dev);
+	pci_msi_shutdown(pci_dev);
+	pci_msix_shutdown(pci_dev);
+}
+
+#ifdef CONFIG_PM_SLEEP
+
+/*
+ * Default "suspend" method for devices that have no driver provided suspend,
+ * or not even a driver at all.
+ */
+static void pci_default_pm_suspend(struct pci_dev *pci_dev)
+{
+	pci_save_state(pci_dev);
+	/*
+	 * mark its power state as "unknown", since we don't know if
+	 * e.g. the BIOS will change its device state when we suspend.
+	 */
+	if (pci_dev->current_state == PCI_D0)
+		pci_dev->current_state = PCI_UNKNOWN;
+}
+
+/*
+ * Default "resume" method for devices that have no driver provided resume,
+ * or not even a driver at all.
+ */
+static int pci_default_pm_resume(struct pci_dev *pci_dev)
+{
+	int retval = 0;
+
+	/* restore the PCI config space */
+	pci_restore_state(pci_dev);
+	/* if the device was enabled before suspend, reenable */
+	retval = pci_reenable_device(pci_dev);
+	/*
+	 * if the device was busmaster before the suspend, make it busmaster
+	 * again
+	 */
+	if (pci_dev->is_busmaster)
+		pci_set_master(pci_dev);
+
+	return retval;
+}
+
+static int pci_legacy_suspend(struct device *dev, pm_message_t state)
 {
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	struct pci_driver * drv = pci_dev->driver;
@@ -284,21 +334,12 @@ static int pci_device_suspend(struct device * dev, pm_message_t state)
 		i = drv->suspend(pci_dev, state);
 		suspend_report_result(drv->suspend, i);
 	} else {
-		pci_save_state(pci_dev);
-		/*
-		 * mark its power state as "unknown", since we don't know if
-		 * e.g. the BIOS will change its device state when we suspend.
-		 */
-		if (pci_dev->current_state == PCI_D0)
-			pci_dev->current_state = PCI_UNKNOWN;
+		pci_default_pm_suspend(pci_dev);
 	}
-
-	pci_fixup_device(pci_fixup_suspend, pci_dev);
-
 	return i;
 }
 
-static int pci_device_suspend_late(struct device * dev, pm_message_t state)
+static int pci_legacy_suspend_late(struct device *dev, pm_message_t state)
 {
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	struct pci_driver * drv = pci_dev->driver;
@@ -311,26 +352,7 @@ static int pci_device_suspend_late(struct device * dev, pm_message_t state)
 	return i;
 }
 
-/*
- * Default resume method for devices that have no driver provided resume,
- * or not even a driver at all.
- */
-static int pci_default_resume(struct pci_dev *pci_dev)
-{
-	int retval = 0;
-
-	/* restore the PCI config space */
-	pci_restore_state(pci_dev);
-	/* if the device was enabled before suspend, reenable */
-	retval = pci_reenable_device(pci_dev);
-	/* if the device was busmaster before the suspend, make it busmaster again */
-	if (pci_dev->is_busmaster)
-		pci_set_master(pci_dev);
-
-	return retval;
-}
-
-static int pci_device_resume(struct device * dev)
+static int pci_legacy_resume(struct device *dev)
 {
 	int error;
 	struct pci_dev * pci_dev = to_pci_dev(dev);
@@ -339,35 +361,313 @@ static int pci_device_resume(struct device * dev)
 	if (drv && drv->resume)
 		error = drv->resume(pci_dev);
 	else
-		error = pci_default_resume(pci_dev);
-	pci_fixup_device(pci_fixup_resume, pci_dev);
+		error = pci_default_pm_resume(pci_dev);
 	return error;
 }
 
-static int pci_device_resume_early(struct device * dev)
+static int pci_legacy_resume_early(struct device *dev)
 {
 	int error = 0;
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	struct pci_driver * drv = pci_dev->driver;
 
-	pci_fixup_device(pci_fixup_resume_early, pci_dev);
-
 	if (drv && drv->resume_early)
 		error = drv->resume_early(pci_dev);
 	return error;
 }
 
-static void pci_device_shutdown(struct device *dev)
+static int pci_pm_prepare(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int error = 0;
+
+	if (drv && drv->pm && drv->pm->prepare)
+		error = drv->pm->prepare(dev);
+
+	return error;
+}
+
+static void pci_pm_complete(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+
+	if (drv && drv->pm && drv->pm->complete)
+		drv->pm->complete(dev);
+}
+
+#ifdef CONFIG_SUSPEND
+
+static int pci_pm_suspend(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct device_driver *drv = dev->driver;
+	int error = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->suspend) {
+			error = drv->pm->suspend(dev);
+			suspend_report_result(drv->pm->suspend, error);
+		} else {
+			pci_default_pm_suspend(pci_dev);
+		}
+	} else {
+		error = pci_legacy_suspend(dev, PMSG_SUSPEND);
+	}
+	pci_fixup_device(pci_fixup_suspend, pci_dev);
+
+	return error;
+}
+
+static int pci_pm_suspend_noirq(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
 	struct pci_driver *drv = pci_dev->driver;
+	int error = 0;
 
-	if (drv && drv->shutdown)
-		drv->shutdown(pci_dev);
-	pci_msi_shutdown(pci_dev);
-	pci_msix_shutdown(pci_dev);
+	if (drv && drv->pm) {
+		if (drv->pm->suspend_noirq) {
+			error = drv->pm->suspend_noirq(dev);
+			suspend_report_result(drv->pm->suspend_noirq, error);
+		}
+	} else {
+		error = pci_legacy_suspend_late(dev, PMSG_SUSPEND);
+	}
+
+	return error;
+}
+
+static int pci_pm_resume(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct device_driver *drv = dev->driver;
+	int error;
+
+	pci_fixup_device(pci_fixup_resume, pci_dev);
+
+	if (drv && drv->pm) {
+		error = drv->pm->resume ? drv->pm->resume(dev) :
+			pci_default_pm_resume(pci_dev);
+	} else {
+		error = pci_legacy_resume(dev);
+	}
+
+	return error;
+}
+
+static int pci_pm_resume_noirq(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
+	int error = 0;
+
+	pci_fixup_device(pci_fixup_resume_early, pci_dev);
+
+	if (drv && drv->pm) {
+		if (drv->pm->resume_noirq)
+			error = drv->pm->resume_noirq(dev);
+	} else {
+		error = pci_legacy_resume_early(dev);
+	}
+
+	return error;
+}
+
+#else /* !CONFIG_SUSPEND */
+
+#define pci_pm_suspend		NULL
+#define pci_pm_suspend_noirq	NULL
+#define pci_pm_resume		NULL
+#define pci_pm_resume_noirq	NULL
+
+#endif /* !CONFIG_SUSPEND */
+
+#ifdef CONFIG_HIBERNATION
+
+static int pci_pm_freeze(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct device_driver *drv = dev->driver;
+	int error = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->freeze) {
+			error = drv->pm->freeze(dev);
+			suspend_report_result(drv->pm->freeze, error);
+		} else {
+			pci_default_pm_suspend(pci_dev);
+		}
+	} else {
+		error = pci_legacy_suspend(dev, PMSG_FREEZE);
+		pci_fixup_device(pci_fixup_suspend, pci_dev);
+	}
+
+	return error;
+}
+
+static int pci_pm_freeze_noirq(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
+	int error = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->freeze_noirq) {
+			error = drv->pm->freeze_noirq(dev);
+			suspend_report_result(drv->pm->freeze_noirq, error);
+		}
+	} else {
+		error = pci_legacy_suspend_late(dev, PMSG_FREEZE);
+	}
+
+	return error;
+}
+
+static int pci_pm_thaw(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int error = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->thaw)
+			error =  drv->pm->thaw(dev);
+	} else {
+		pci_fixup_device(pci_fixup_resume, to_pci_dev(dev));
+		error = pci_legacy_resume(dev);
+	}
+
+	return error;
+}
+
+static int pci_pm_thaw_noirq(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
+	int error = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->thaw_noirq)
+			error = drv->pm->thaw_noirq(dev);
+	} else {
+		pci_fixup_device(pci_fixup_resume_early, pci_dev);
+		error = pci_legacy_resume_early(dev);
+	}
+
+	return error;
+}
+
+static int pci_pm_poweroff(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int error = 0;
+
+	pci_fixup_device(pci_fixup_suspend, to_pci_dev(dev));
+
+	if (drv && drv->pm) {
+		if (drv->pm->poweroff) {
+			error = drv->pm->poweroff(dev);
+			suspend_report_result(drv->pm->poweroff, error);
+		}
+	} else {
+		error = pci_legacy_suspend(dev, PMSG_HIBERNATE);
+	}
+
+	return error;
+}
+
+static int pci_pm_poweroff_noirq(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
+	int error = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->poweroff_noirq) {
+			error = drv->pm->poweroff_noirq(dev);
+			suspend_report_result(drv->pm->poweroff_noirq, error);
+		}
+	} else {
+		error = pci_legacy_suspend_late(dev, PMSG_HIBERNATE);
+	}
+
+	return error;
+}
+
+static int pci_pm_restore(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct device_driver *drv = dev->driver;
+	int error;
+
+	if (drv && drv->pm) {
+		error = drv->pm->restore ? drv->pm->restore(dev) :
+			pci_default_pm_resume(pci_dev);
+	} else {
+		error = pci_legacy_resume(dev);
+	}
+	pci_fixup_device(pci_fixup_resume, pci_dev);
+
+	return error;
+}
+
+static int pci_pm_restore_noirq(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct pci_driver *drv = pci_dev->driver;
+	int error = 0;
+
+	pci_fixup_device(pci_fixup_resume, pci_dev);
+
+	if (drv && drv->pm) {
+		if (drv->pm->restore_noirq)
+			error = drv->pm->restore_noirq(dev);
+	} else {
+		error = pci_legacy_resume_early(dev);
+	}
+	pci_fixup_device(pci_fixup_resume_early, pci_dev);
+
+	return error;
 }
 
+#else /* !CONFIG_HIBERNATION */
+
+#define pci_pm_freeze		NULL
+#define pci_pm_freeze_noirq	NULL
+#define pci_pm_thaw		NULL
+#define pci_pm_thaw_noirq	NULL
+#define pci_pm_poweroff		NULL
+#define pci_pm_poweroff_noirq	NULL
+#define pci_pm_restore		NULL
+#define pci_pm_restore_noirq	NULL
+
+#endif /* !CONFIG_HIBERNATION */
+
+struct pm_ext_ops pci_pm_ops = {
+	.base = {
+		.prepare = pci_pm_prepare,
+		.complete = pci_pm_complete,
+		.suspend = pci_pm_suspend,
+		.resume = pci_pm_resume,
+		.freeze = pci_pm_freeze,
+		.thaw = pci_pm_thaw,
+		.poweroff = pci_pm_poweroff,
+		.restore = pci_pm_restore,
+	},
+	.suspend_noirq = pci_pm_suspend_noirq,
+	.resume_noirq = pci_pm_resume_noirq,
+	.freeze_noirq = pci_pm_freeze_noirq,
+	.thaw_noirq = pci_pm_thaw_noirq,
+	.poweroff_noirq = pci_pm_poweroff_noirq,
+	.restore_noirq = pci_pm_restore_noirq,
+};
+
+#define PCI_PM_OPS_PTR	&pci_pm_ops
+
+#else /* !CONFIG_PM_SLEEP */
+
+#define PCI_PM_OPS_PTR	NULL
+
+#endif /* !CONFIG_PM_SLEEP */
+
 /**
  * __pci_register_driver - register a new pci driver
  * @drv: the driver structure to register
@@ -390,6 +690,9 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner,
 	drv->driver.owner = owner;
 	drv->driver.mod_name = mod_name;
 
+	if (drv->pm)
+		drv->driver.pm = &drv->pm->base;
+
 	spin_lock_init(&drv->dynids.lock);
 	INIT_LIST_HEAD(&drv->dynids.list);
 
@@ -515,12 +818,9 @@ struct bus_type pci_bus_type = {
 	.uevent		= pci_uevent,
 	.probe		= pci_device_probe,
 	.remove		= pci_device_remove,
-	.suspend	= pci_device_suspend,
-	.suspend_late	= pci_device_suspend_late,
-	.resume_early	= pci_device_resume_early,
-	.resume		= pci_device_resume,
 	.shutdown	= pci_device_shutdown,
 	.dev_attrs	= pci_dev_attrs,
+	.pm		= PCI_PM_OPS_PTR,
 };
 
 static int __init pci_driver_init(void)
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 700704ef70f3..507ee52323cd 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -389,7 +389,7 @@ struct pci_driver {
 	int  (*resume_early) (struct pci_dev *dev);
 	int  (*resume) (struct pci_dev *dev);	                /* Device woken up */
 	void (*shutdown) (struct pci_dev *dev);
-
+	struct pm_ext_ops *pm;
 	struct pci_error_handlers *err_handler;
 	struct device_driver	driver;
 	struct pci_dynids dynids;
-- 
cgit v1.2.3


From 25e18499e08cb097cbbfeab5de25d094d5312ee5 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 21 May 2008 01:40:43 +0200
Subject: Implement new suspend and hibernation callbacks for platform busses

Implement new suspend and hibernation callbacks for the platform bus
type.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Greg KH <greg@kroah.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/base/platform.c         | 296 ++++++++++++++++++++++++++++++++++++++--
 include/linux/platform_device.h |   1 +
 2 files changed, 289 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index 911ec600fe71..3f940393d6c7 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -453,6 +453,8 @@ int platform_driver_register(struct platform_driver *drv)
 		drv->driver.suspend = platform_drv_suspend;
 	if (drv->resume)
 		drv->driver.resume = platform_drv_resume;
+	if (drv->pm)
+		drv->driver.pm = &drv->pm->base;
 	return driver_register(&drv->driver);
 }
 EXPORT_SYMBOL_GPL(platform_driver_register);
@@ -560,7 +562,9 @@ static int platform_match(struct device *dev, struct device_driver *drv)
 	return (strncmp(pdev->name, drv->name, BUS_ID_SIZE) == 0);
 }
 
-static int platform_suspend(struct device *dev, pm_message_t mesg)
+#ifdef CONFIG_PM_SLEEP
+
+static int platform_legacy_suspend(struct device *dev, pm_message_t mesg)
 {
 	int ret = 0;
 
@@ -570,7 +574,7 @@ static int platform_suspend(struct device *dev, pm_message_t mesg)
 	return ret;
 }
 
-static int platform_suspend_late(struct device *dev, pm_message_t mesg)
+static int platform_legacy_suspend_late(struct device *dev, pm_message_t mesg)
 {
 	struct platform_driver *drv = to_platform_driver(dev->driver);
 	struct platform_device *pdev;
@@ -583,7 +587,7 @@ static int platform_suspend_late(struct device *dev, pm_message_t mesg)
 	return ret;
 }
 
-static int platform_resume_early(struct device *dev)
+static int platform_legacy_resume_early(struct device *dev)
 {
 	struct platform_driver *drv = to_platform_driver(dev->driver);
 	struct platform_device *pdev;
@@ -596,7 +600,7 @@ static int platform_resume_early(struct device *dev)
 	return ret;
 }
 
-static int platform_resume(struct device *dev)
+static int platform_legacy_resume(struct device *dev)
 {
 	int ret = 0;
 
@@ -606,15 +610,291 @@ static int platform_resume(struct device *dev)
 	return ret;
 }
 
+static int platform_pm_prepare(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int ret = 0;
+
+	if (drv && drv->pm && drv->pm->prepare)
+		ret = drv->pm->prepare(dev);
+
+	return ret;
+}
+
+static void platform_pm_complete(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+
+	if (drv && drv->pm && drv->pm->complete)
+		drv->pm->complete(dev);
+}
+
+#ifdef CONFIG_SUSPEND
+
+static int platform_pm_suspend(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int ret = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->suspend)
+			ret = drv->pm->suspend(dev);
+	} else {
+		ret = platform_legacy_suspend(dev, PMSG_SUSPEND);
+	}
+
+	return ret;
+}
+
+static int platform_pm_suspend_noirq(struct device *dev)
+{
+	struct platform_driver *pdrv;
+	int ret = 0;
+
+	if (!dev->driver)
+		return 0;
+
+	pdrv = to_platform_driver(dev->driver);
+	if (pdrv->pm) {
+		if (pdrv->pm->suspend_noirq)
+			ret = pdrv->pm->suspend_noirq(dev);
+	} else {
+		ret = platform_legacy_suspend_late(dev, PMSG_SUSPEND);
+	}
+
+	return ret;
+}
+
+static int platform_pm_resume(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int ret = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->resume)
+			ret = drv->pm->resume(dev);
+	} else {
+		ret = platform_legacy_resume(dev);
+	}
+
+	return ret;
+}
+
+static int platform_pm_resume_noirq(struct device *dev)
+{
+	struct platform_driver *pdrv;
+	int ret = 0;
+
+	if (!dev->driver)
+		return 0;
+
+	pdrv = to_platform_driver(dev->driver);
+	if (pdrv->pm) {
+		if (pdrv->pm->resume_noirq)
+			ret = pdrv->pm->resume_noirq(dev);
+	} else {
+		ret = platform_legacy_resume_early(dev);
+	}
+
+	return ret;
+}
+
+#else /* !CONFIG_SUSPEND */
+
+#define platform_pm_suspend		NULL
+#define platform_pm_resume		NULL
+#define platform_pm_suspend_noirq	NULL
+#define platform_pm_resume_noirq	NULL
+
+#endif /* !CONFIG_SUSPEND */
+
+#ifdef CONFIG_HIBERNATION
+
+static int platform_pm_freeze(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int ret = 0;
+
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
+		if (drv->pm->freeze)
+			ret = drv->pm->freeze(dev);
+	} else {
+		ret = platform_legacy_suspend(dev, PMSG_FREEZE);
+	}
+
+	return ret;
+}
+
+static int platform_pm_freeze_noirq(struct device *dev)
+{
+	struct platform_driver *pdrv;
+	int ret = 0;
+
+	if (!dev->driver)
+		return 0;
+
+	pdrv = to_platform_driver(dev->driver);
+	if (pdrv->pm) {
+		if (pdrv->pm->freeze_noirq)
+			ret = pdrv->pm->freeze_noirq(dev);
+	} else {
+		ret = platform_legacy_suspend_late(dev, PMSG_FREEZE);
+	}
+
+	return ret;
+}
+
+static int platform_pm_thaw(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int ret = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->thaw)
+			ret = drv->pm->thaw(dev);
+	} else {
+		ret = platform_legacy_resume(dev);
+	}
+
+	return ret;
+}
+
+static int platform_pm_thaw_noirq(struct device *dev)
+{
+	struct platform_driver *pdrv;
+	int ret = 0;
+
+	if (!dev->driver)
+		return 0;
+
+	pdrv = to_platform_driver(dev->driver);
+	if (pdrv->pm) {
+		if (pdrv->pm->thaw_noirq)
+			ret = pdrv->pm->thaw_noirq(dev);
+	} else {
+		ret = platform_legacy_resume_early(dev);
+	}
+
+	return ret;
+}
+
+static int platform_pm_poweroff(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int ret = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->poweroff)
+			ret = drv->pm->poweroff(dev);
+	} else {
+		ret = platform_legacy_suspend(dev, PMSG_HIBERNATE);
+	}
+
+	return ret;
+}
+
+static int platform_pm_poweroff_noirq(struct device *dev)
+{
+	struct platform_driver *pdrv;
+	int ret = 0;
+
+	if (!dev->driver)
+		return 0;
+
+	pdrv = to_platform_driver(dev->driver);
+	if (pdrv->pm) {
+		if (pdrv->pm->poweroff_noirq)
+			ret = pdrv->pm->poweroff_noirq(dev);
+	} else {
+		ret = platform_legacy_suspend_late(dev, PMSG_HIBERNATE);
+	}
+
+	return ret;
+}
+
+static int platform_pm_restore(struct device *dev)
+{
+	struct device_driver *drv = dev->driver;
+	int ret = 0;
+
+	if (drv && drv->pm) {
+		if (drv->pm->restore)
+			ret = drv->pm->restore(dev);
+	} else {
+		ret = platform_legacy_resume(dev);
+	}
+
+	return ret;
+}
+
+static int platform_pm_restore_noirq(struct device *dev)
+{
+	struct platform_driver *pdrv;
+	int ret = 0;
+
+	if (!dev->driver)
+		return 0;
+
+	pdrv = to_platform_driver(dev->driver);
+	if (pdrv->pm) {
+		if (pdrv->pm->restore_noirq)
+			ret = pdrv->pm->restore_noirq(dev);
+	} else {
+		ret = platform_legacy_resume_early(dev);
+	}
+
+	return ret;
+}
+
+#else /* !CONFIG_HIBERNATION */
+
+#define platform_pm_freeze		NULL
+#define platform_pm_thaw		NULL
+#define platform_pm_poweroff		NULL
+#define platform_pm_restore		NULL
+#define platform_pm_freeze_noirq	NULL
+#define platform_pm_thaw_noirq		NULL
+#define platform_pm_poweroff_noirq	NULL
+#define platform_pm_restore_noirq	NULL
+
+#endif /* !CONFIG_HIBERNATION */
+
+struct pm_ext_ops platform_pm_ops = {
+	.base = {
+		.prepare = platform_pm_prepare,
+		.complete = platform_pm_complete,
+		.suspend = platform_pm_suspend,
+		.resume = platform_pm_resume,
+		.freeze = platform_pm_freeze,
+		.thaw = platform_pm_thaw,
+		.poweroff = platform_pm_poweroff,
+		.restore = platform_pm_restore,
+	},
+	.suspend_noirq = platform_pm_suspend_noirq,
+	.resume_noirq = platform_pm_resume_noirq,
+	.freeze_noirq = platform_pm_freeze_noirq,
+	.thaw_noirq = platform_pm_thaw_noirq,
+	.poweroff_noirq = platform_pm_poweroff_noirq,
+	.restore_noirq = platform_pm_restore_noirq,
+};
+
+#define PLATFORM_PM_OPS_PTR	&platform_pm_ops
+
+#else /* !CONFIG_PM_SLEEP */
+
+#define PLATFORM_PM_OPS_PTR	NULL
+
+#endif /* !CONFIG_PM_SLEEP */
+
 struct bus_type platform_bus_type = {
 	.name		= "platform",
 	.dev_attrs	= platform_dev_attrs,
 	.match		= platform_match,
 	.uevent		= platform_uevent,
-	.suspend	= platform_suspend,
-	.suspend_late	= platform_suspend_late,
-	.resume_early	= platform_resume_early,
-	.resume		= platform_resume,
+	.pm		= PLATFORM_PM_OPS_PTR,
 };
 EXPORT_SYMBOL_GPL(platform_bus_type);
 
diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h
index 3261681c82a4..95ac21ab3a09 100644
--- a/include/linux/platform_device.h
+++ b/include/linux/platform_device.h
@@ -53,6 +53,7 @@ struct platform_driver {
 	int (*suspend_late)(struct platform_device *, pm_message_t state);
 	int (*resume_early)(struct platform_device *);
 	int (*resume)(struct platform_device *);
+	struct pm_ext_ops *pm;
 	struct device_driver driver;
 };
 
-- 
cgit v1.2.3


From ac9c052d10d8d6f46a30cb46c0d6d753997c299f Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 28 May 2008 15:01:03 +0900
Subject: shpchp: check firmware before taking control

Fix the following problems of shpchp driver about getting hotplug
control from firmware.

  - The shpchp driver must not control the hotplug controller if it
    fails to get control from the firmware. But current shpchp
    controls the hotplug controller regardless the result, because it
    doesn't check the return value of get_hp_hw_control_from_firmware().

  - Current shpchp driver doesn't support _OSC.

The pciehp driver already have the code for evaluating _OSC and OSHP
and shpchp and pciehp can share it. So this patch move that code from
pciehp to acpi_pcihp.c.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/acpi_pcihp.c  | 85 +++++++++++++++++++++++++++++++++++++--
 drivers/pci/hotplug/pciehp.h      | 10 +++--
 drivers/pci/hotplug/pciehp_hpc.c  | 69 -------------------------------
 drivers/pci/hotplug/shpchp.h      | 14 ++++---
 drivers/pci/hotplug/shpchp_core.c | 15 +++----
 drivers/pci/hotplug/shpchp_hpc.c  |  1 -
 include/linux/pci_hotplug.h       |  2 +-
 7 files changed, 105 insertions(+), 91 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c
index f8c187a763bd..93e37f0666ab 100644
--- a/drivers/pci/hotplug/acpi_pcihp.c
+++ b/drivers/pci/hotplug/acpi_pcihp.c
@@ -30,6 +30,7 @@
 #include <linux/types.h>
 #include <linux/pci.h>
 #include <linux/pci_hotplug.h>
+#include <linux/pci-acpi.h>
 #include <acpi/acpi.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/actypes.h>
@@ -299,7 +300,7 @@ free_and_return:
  *
  * @handle - the handle of the hotplug controller.
  */
-acpi_status acpi_run_oshp(acpi_handle handle)
+static acpi_status acpi_run_oshp(acpi_handle handle)
 {
 	acpi_status		status;
 	struct acpi_buffer	string = { ACPI_ALLOCATE_BUFFER, NULL };
@@ -322,9 +323,6 @@ acpi_status acpi_run_oshp(acpi_handle handle)
 	kfree(string.pointer);
 	return status;
 }
-EXPORT_SYMBOL_GPL(acpi_run_oshp);
-
-
 
 /* acpi_get_hp_params_from_firmware
  *
@@ -374,6 +372,85 @@ acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus,
 }
 EXPORT_SYMBOL_GPL(acpi_get_hp_params_from_firmware);
 
+/**
+ * acpi_get_hp_hw_control_from_firmware
+ * @dev: the pci_dev of the bridge that has a hotplug controller
+ * @flags: requested control bits for _OSC
+ *
+ * Attempt to take hotplug control from firmware.
+ */
+int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags)
+{
+	acpi_status status;
+	acpi_handle chandle, handle = DEVICE_ACPI_HANDLE(&(dev->dev));
+	struct pci_dev *pdev = dev;
+	struct pci_bus *parent;
+	struct acpi_buffer string = { ACPI_ALLOCATE_BUFFER, NULL };
+
+	flags &= (OSC_PCI_EXPRESS_NATIVE_HP_CONTROL |
+		  OSC_SHPC_NATIVE_HP_CONTROL |
+		  OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL);
+	if (!flags) {
+		err("Invalid flags %u specified!\n", flags);
+		return -EINVAL;
+	}
+
+	/*
+	 * Per PCI firmware specification, we should run the ACPI _OSC
+	 * method to get control of hotplug hardware before using it. If
+	 * an _OSC is missing, we look for an OSHP to do the same thing.
+	 * To handle different BIOS behavior, we look for _OSC and OSHP
+	 * within the scope of the hotplug controller and its parents,
+	 * upto the host bridge under which this controller exists.
+	 */
+	while (!handle) {
+		/*
+		 * This hotplug controller was not listed in the ACPI name
+		 * space at all. Try to get acpi handle of parent pci bus.
+		 */
+		if (!pdev || !pdev->bus->parent)
+			break;
+		parent = pdev->bus->parent;
+		dbg("Could not find %s in acpi namespace, trying parent\n",
+		    pci_name(pdev));
+		if (!parent->self)
+			/* Parent must be a host bridge */
+			handle = acpi_get_pci_rootbridge_handle(
+					pci_domain_nr(parent),
+					parent->number);
+		else
+			handle = DEVICE_ACPI_HANDLE(&(parent->self->dev));
+		pdev = parent->self;
+	}
+
+	while (handle) {
+		acpi_get_name(handle, ACPI_FULL_PATHNAME, &string);
+		dbg("Trying to get hotplug control for %s \n",
+		    (char *)string.pointer);
+		status = pci_osc_control_set(handle, flags);
+		if (status == AE_NOT_FOUND)
+			status = acpi_run_oshp(handle);
+		if (ACPI_SUCCESS(status)) {
+			dbg("Gained control for hotplug HW for pci %s (%s)\n",
+			    pci_name(dev), (char *)string.pointer);
+			kfree(string.pointer);
+			return 0;
+		}
+		if (acpi_root_bridge(handle))
+			break;
+		chandle = handle;
+		status = acpi_get_parent(chandle, &handle);
+		if (ACPI_FAILURE(status))
+			break;
+	}
+
+	dbg("Cannot get control of hotplug hardware for pci %s\n",
+	    pci_name(dev));
+
+	kfree(string.pointer);
+	return -ENODEV;
+}
+EXPORT_SYMBOL(acpi_get_hp_hw_control_from_firmware);
 
 /* acpi_root_bridge - check to see if this acpi object is a root bridge
  *
diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h
index 084b73efacb3..8492fab800cc 100644
--- a/drivers/pci/hotplug/pciehp.h
+++ b/drivers/pci/hotplug/pciehp.h
@@ -202,9 +202,13 @@ struct hpc_ops {
 #include <acpi/actypes.h>
 #include <linux/pci-acpi.h>
 
-extern int pciehp_acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev);
-#define pciehp_get_hp_hw_control_from_firmware(dev)			\
-	pciehp_acpi_get_hp_hw_control_from_firmware(dev)
+static inline int pciehp_get_hp_hw_control_from_firmware(struct pci_dev *dev)
+{
+	u32 flags = (OSC_PCI_EXPRESS_NATIVE_HP_CONTROL |
+		     OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL);
+	return acpi_get_hp_hw_control_from_firmware(dev, flags);
+}
+
 static inline int pciehp_get_hp_params_from_firmware(struct pci_dev *dev,
 			struct hotplug_params *hpp)
 {
diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c
index 7e3a3d17c334..a48021d85f22 100644
--- a/drivers/pci/hotplug/pciehp_hpc.c
+++ b/drivers/pci/hotplug/pciehp_hpc.c
@@ -1009,75 +1009,6 @@ static struct hpc_ops pciehp_hpc_ops = {
 	.check_lnk_status		= hpc_check_lnk_status,
 };
 
-#ifdef CONFIG_ACPI
-int pciehp_acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev)
-{
-	acpi_status status;
-	acpi_handle chandle, handle = DEVICE_ACPI_HANDLE(&(dev->dev));
-	struct pci_dev *pdev = dev;
-	struct pci_bus *parent;
-	struct acpi_buffer string = { ACPI_ALLOCATE_BUFFER, NULL };
-
-	/*
-	 * Per PCI firmware specification, we should run the ACPI _OSC
-	 * method to get control of hotplug hardware before using it.
-	 * If an _OSC is missing, we look for an OSHP to do the same thing.
-	 * To handle different BIOS behavior, we look for _OSC and OSHP
-	 * within the scope of the hotplug controller and its parents, upto
-	 * the host bridge under which this controller exists.
-	 */
-	while (!handle) {
-		/*
-		 * This hotplug controller was not listed in the ACPI name
-		 * space at all. Try to get acpi handle of parent pci bus.
-		 */
-		if (!pdev || !pdev->bus->parent)
-			break;
-		parent = pdev->bus->parent;
-		dbg("Could not find %s in acpi namespace, trying parent\n",
-				pci_name(pdev));
-		if (!parent->self)
-			/* Parent must be a host bridge */
-			handle = acpi_get_pci_rootbridge_handle(
-					pci_domain_nr(parent),
-					parent->number);
-		else
-			handle = DEVICE_ACPI_HANDLE(
-					&(parent->self->dev));
-		pdev = parent->self;
-	}
-
-	while (handle) {
-		acpi_get_name(handle, ACPI_FULL_PATHNAME, &string);
-		dbg("Trying to get hotplug control for %s \n",
-			(char *)string.pointer);
-		status = pci_osc_control_set(handle,
-				OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL |
-				OSC_PCI_EXPRESS_NATIVE_HP_CONTROL);
-		if (status == AE_NOT_FOUND)
-			status = acpi_run_oshp(handle);
-		if (ACPI_SUCCESS(status)) {
-			dbg("Gained control for hotplug HW for pci %s (%s)\n",
-				pci_name(dev), (char *)string.pointer);
-			kfree(string.pointer);
-			return 0;
-		}
-		if (acpi_root_bridge(handle))
-			break;
-		chandle = handle;
-		status = acpi_get_parent(chandle, &handle);
-		if (ACPI_FAILURE(status))
-			break;
-	}
-
-	dbg("Cannot get control of hotplug hardware for pci %s\n",
-			pci_name(dev));
-
-	kfree(string.pointer);
-	return -1;
-}
-#endif
-
 static int pcie_init_hardware_part1(struct controller *ctrl,
 				    struct pcie_device *dev)
 {
diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h
index f66e8d6315ab..8a026f750deb 100644
--- a/drivers/pci/hotplug/shpchp.h
+++ b/drivers/pci/hotplug/shpchp.h
@@ -170,6 +170,7 @@ extern void shpchp_queue_pushbutton_work(struct work_struct *work);
 extern int shpc_init( struct controller *ctrl, struct pci_dev *pdev);
 
 #ifdef CONFIG_ACPI
+#include <linux/pci-acpi.h>
 static inline int get_hp_params_from_firmware(struct pci_dev *dev,
 					      struct hotplug_params *hpp)
 {
@@ -177,14 +178,15 @@ static inline int get_hp_params_from_firmware(struct pci_dev *dev,
 			return -ENODEV;
 	return 0;
 }
-#define get_hp_hw_control_from_firmware(pdev)				\
-	do {								\
-		if (DEVICE_ACPI_HANDLE(&(pdev->dev)))			\
-			acpi_run_oshp(DEVICE_ACPI_HANDLE(&(pdev->dev)));\
-	} while (0)
+
+static inline int get_hp_hw_control_from_firmware(struct pci_dev *dev)
+{
+	u32 flags = OSC_SHPC_NATIVE_HP_CONTROL;
+	return acpi_get_hp_hw_control_from_firmware(dev, flags);
+}
 #else
 #define get_hp_params_from_firmware(dev, hpp) (-ENODEV)
-#define get_hp_hw_control_from_firmware(dev) do { } while (0)
+#define get_hp_hw_control_from_firmware(dev) (0)
 #endif
 
 struct ctrl_reg {
diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c
index 0066b0be0928..df41ecc4e7fc 100644
--- a/drivers/pci/hotplug/shpchp_core.c
+++ b/drivers/pci/hotplug/shpchp_core.c
@@ -330,13 +330,14 @@ static int get_cur_bus_speed (struct hotplug_slot *hotplug_slot, enum pci_bus_sp
 
 static int is_shpc_capable(struct pci_dev *dev)
 {
-       if ((dev->vendor == PCI_VENDOR_ID_AMD) || (dev->device ==
-                               PCI_DEVICE_ID_AMD_GOLAM_7450))
-               return 1;
-       if (pci_find_capability(dev, PCI_CAP_ID_SHPC))
-               return 1;
-
-       return 0;
+	if ((dev->vendor == PCI_VENDOR_ID_AMD) || (dev->device ==
+						PCI_DEVICE_ID_AMD_GOLAM_7450))
+		return 1;
+	if (!pci_find_capability(dev, PCI_CAP_ID_SHPC))
+		return 0;
+	if (get_hp_hw_control_from_firmware(dev))
+		return 0;
+	return 1;
 }
 
 static int shpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c
index 7d770b2cd889..7a0bff364cd4 100644
--- a/drivers/pci/hotplug/shpchp_hpc.c
+++ b/drivers/pci/hotplug/shpchp_hpc.c
@@ -1084,7 +1084,6 @@ int shpc_init(struct controller *ctrl, struct pci_dev *pdev)
 	dbg("%s: HPC at b:d:f:irq=0x%x:%x:%x:%x\n", __func__,
 			pdev->bus->number, PCI_SLOT(pdev->devfn),
 			PCI_FUNC(pdev->devfn), pdev->irq);
-	get_hp_hw_control_from_firmware(pdev);
 
 	/*
 	 * If this is the first controller to be initialized,
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index 8f67e8f2a3cc..dbdcd1ad3c6a 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -227,9 +227,9 @@ struct hotplug_params {
 #include <acpi/acpi.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/actypes.h>
-extern acpi_status acpi_run_oshp(acpi_handle handle);
 extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus,
 				struct hotplug_params *hpp);
+int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags);
 int acpi_root_bridge(acpi_handle handle);
 #endif
 #endif
-- 
cgit v1.2.3


From f46753c5e354b857b20ab8e0fe7b2579831dc369 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 10 Jun 2008 15:28:50 -0600
Subject: PCI: introduce pci_slot

Currently, /sys/bus/pci/slots/ only exposes hotplug attributes when a
hotplug driver is loaded, but PCI slots have attributes such as address,
speed, width, etc.  that are not related to hotplug at all.

Introduce pci_slot as the primary data structure and kobject model.
Hotplug attributes described in hotplug_slot become a secondary
structure associated with the pci_slot.

This patch only creates the infrastructure that allows the separation of
PCI slot attributes and hotplug attributes.  In this patch, the PCI
hotplug core remains the only user of this infrastructure, and thus,
/sys/bus/pci/slots/ will still only become populated when a hotplug
driver is loaded.

A later patch in this series will add a second user of this new
infrastructure and demonstrate splitting the task of exposing pci_slot
attributes from hotplug_slot attributes.

  - Make pci_slot the primary sysfs entity. hotplug_slot becomes a
    subsidiary structure.
    o pci_create_slot() creates and registers a slot with the PCI core
    o pci_slot_add_hotplug() gives it hotplug capability

  - Change the prototype of pci_hp_register() to take the bus and
    slot number (on parent bus) as parameters.

  - Remove all the ->get_address methods since this functionality is
    now handled by pci_slot directly.

[achiang@hp.com: rpaphp-correctly-pci_hp_register-for-empty-pci-slots]
Tested-by: Badari Pulavarty <pbadari@us.ibm.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: make headers_check happy]
[akpm@linux-foundation.org: nuther build fix]
[akpm@linux-foundation.org: fix typo in #include]
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Matthew Wilcox <matthew@wil.cx>
Cc: Greg KH <greg@kroah.com>
Cc: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Cc: Len Brown <lenb@kernel.org>
Acked-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/Makefile                    |   2 +-
 drivers/pci/hotplug/acpiphp.h           |   1 -
 drivers/pci/hotplug/acpiphp_core.c      |  25 +---
 drivers/pci/hotplug/acpiphp_glue.c      |  23 +--
 drivers/pci/hotplug/acpiphp_ibm.c       |   6 +-
 drivers/pci/hotplug/cpci_hotplug_core.c |   2 +-
 drivers/pci/hotplug/cpqphp_core.c       |   4 +-
 drivers/pci/hotplug/fakephp.c           |   2 +-
 drivers/pci/hotplug/ibmphp_ebda.c       |   3 +-
 drivers/pci/hotplug/pci_hotplug_core.c  | 258 ++++++++++++--------------------
 drivers/pci/hotplug/pciehp_core.c       |  32 ++--
 drivers/pci/hotplug/rpadlpar_sysfs.c    |   5 +-
 drivers/pci/hotplug/rpaphp_slot.c       |  44 +-----
 drivers/pci/hotplug/sgi_hotplug.c       |  10 +-
 drivers/pci/hotplug/shpchp_core.c       |  20 +--
 drivers/pci/pci.h                       |  13 ++
 drivers/pci/probe.c                     |   1 +
 drivers/pci/slot.c                      | 233 ++++++++++++++++++++++++++++
 include/linux/pci.h                     |  19 ++-
 include/linux/pci_hotplug.h             |  12 +-
 20 files changed, 422 insertions(+), 293 deletions(-)
 create mode 100644 drivers/pci/slot.c

(limited to 'include/linux')

diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile
index 4d1ce2e7361e..7d63f8ced24b 100644
--- a/drivers/pci/Makefile
+++ b/drivers/pci/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the PCI bus specific drivers.
 #
 
-obj-y		+= access.o bus.o probe.o remove.o pci.o quirks.o \
+obj-y		+= access.o bus.o probe.o remove.o pci.o quirks.o slot.o \
 			pci-driver.o search.o pci-sysfs.o rom.o setup-res.o
 obj-$(CONFIG_PROC_FS) += proc.o
 
diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h
index 7a29164d4b32..eecf7cbf4139 100644
--- a/drivers/pci/hotplug/acpiphp.h
+++ b/drivers/pci/hotplug/acpiphp.h
@@ -215,7 +215,6 @@ extern u8 acpiphp_get_power_status (struct acpiphp_slot *slot);
 extern u8 acpiphp_get_attention_status (struct acpiphp_slot *slot);
 extern u8 acpiphp_get_latch_status (struct acpiphp_slot *slot);
 extern u8 acpiphp_get_adapter_status (struct acpiphp_slot *slot);
-extern u32 acpiphp_get_address (struct acpiphp_slot *slot);
 
 /* variables */
 extern int acpiphp_debug;
diff --git a/drivers/pci/hotplug/acpiphp_core.c b/drivers/pci/hotplug/acpiphp_core.c
index 7af68ba27903..0e496e866a84 100644
--- a/drivers/pci/hotplug/acpiphp_core.c
+++ b/drivers/pci/hotplug/acpiphp_core.c
@@ -70,7 +70,6 @@ static int disable_slot		(struct hotplug_slot *slot);
 static int set_attention_status (struct hotplug_slot *slot, u8 value);
 static int get_power_status	(struct hotplug_slot *slot, u8 *value);
 static int get_attention_status (struct hotplug_slot *slot, u8 *value);
-static int get_address		(struct hotplug_slot *slot, u32 *value);
 static int get_latch_status	(struct hotplug_slot *slot, u8 *value);
 static int get_adapter_status	(struct hotplug_slot *slot, u8 *value);
 
@@ -83,7 +82,6 @@ static struct hotplug_slot_ops acpi_hotplug_slot_ops = {
 	.get_attention_status	= get_attention_status,
 	.get_latch_status	= get_latch_status,
 	.get_adapter_status	= get_adapter_status,
-	.get_address		= get_address,
 };
 
 
@@ -274,23 +272,6 @@ static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value)
 	return 0;
 }
 
-
-/**
- * get_address - get pci address of a slot
- * @hotplug_slot: slot to get status
- * @value: pointer to struct pci_busdev (seg, bus, dev)
- */
-static int get_address(struct hotplug_slot *hotplug_slot, u32 *value)
-{
-	struct slot *slot = hotplug_slot->private;
-
-	dbg("%s - physical_slot = %s\n", __func__, hotplug_slot->name);
-
-	*value = acpiphp_get_address(slot->acpi_slot);
-
-	return 0;
-}
-
 static int __init init_acpi(void)
 {
 	int retval;
@@ -357,7 +338,11 @@ int acpiphp_register_hotplug_slot(struct acpiphp_slot *acpiphp_slot)
 	acpiphp_slot->slot = slot;
 	snprintf(slot->name, sizeof(slot->name), "%u", slot->acpi_slot->sun);
 
-	retval = pci_hp_register(slot->hotplug_slot);
+	retval = pci_hp_register(slot->hotplug_slot,
+					acpiphp_slot->bridge->pci_bus,
+					acpiphp_slot->device);
+	if (retval == -EBUSY)
+		goto error_hpslot;
 	if (retval) {
 		err("pci_hp_register failed with error %d\n", retval);
 		goto error_hpslot;
diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 648596d469f6..9342c848db29 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -258,7 +258,12 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv)
 				bridge->pci_bus->number, slot->device);
 		retval = acpiphp_register_hotplug_slot(slot);
 		if (retval) {
-			warn("acpiphp_register_hotplug_slot failed(err code = 0x%x)\n", retval);
+			if (retval == -EBUSY)
+				warn("Slot %d already registered by another "
+					"hotplug driver\n", slot->sun);
+			else
+				warn("acpiphp_register_hotplug_slot failed "
+					"(err code = 0x%x)\n", retval);
 			goto err_exit;
 		}
 	}
@@ -1867,19 +1872,3 @@ u8 acpiphp_get_adapter_status(struct acpiphp_slot *slot)
 
 	return (sta == 0) ? 0 : 1;
 }
-
-
-/*
- * pci address (seg/bus/dev)
- */
-u32 acpiphp_get_address(struct acpiphp_slot *slot)
-{
-	u32 address;
-	struct pci_bus *pci_bus = slot->bridge->pci_bus;
-
-	address = (pci_domain_nr(pci_bus) << 16) |
-		  (pci_bus->number << 8) |
-		  slot->device;
-
-	return address;
-}
diff --git a/drivers/pci/hotplug/acpiphp_ibm.c b/drivers/pci/hotplug/acpiphp_ibm.c
index ede9051fdb5d..2b7c45e39370 100644
--- a/drivers/pci/hotplug/acpiphp_ibm.c
+++ b/drivers/pci/hotplug/acpiphp_ibm.c
@@ -33,8 +33,10 @@
 #include <linux/kobject.h>
 #include <asm/uaccess.h>
 #include <linux/moduleparam.h>
+#include <linux/pci.h>
 
 #include "acpiphp.h"
+#include "../pci.h"
 
 #define DRIVER_VERSION	"1.0.1"
 #define DRIVER_AUTHOR	"Irene Zubarev <zubarev@us.ibm.com>, Vernon Mauery <vernux@us.ibm.com>"
@@ -430,7 +432,7 @@ static int __init ibm_acpiphp_init(void)
 	int retval = 0;
 	acpi_status status;
 	struct acpi_device *device;
-	struct kobject *sysdir = &pci_hotplug_slots_kset->kobj;
+	struct kobject *sysdir = &pci_slots_kset->kobj;
 
 	dbg("%s\n", __func__);
 
@@ -477,7 +479,7 @@ init_return:
 static void __exit ibm_acpiphp_exit(void)
 {
 	acpi_status status;
-	struct kobject *sysdir = &pci_hotplug_slots_kset->kobj;
+	struct kobject *sysdir = &pci_slots_kset->kobj;
 
 	dbg("%s\n", __func__);
 
diff --git a/drivers/pci/hotplug/cpci_hotplug_core.c b/drivers/pci/hotplug/cpci_hotplug_core.c
index d8a6b80ab42a..935947991dc9 100644
--- a/drivers/pci/hotplug/cpci_hotplug_core.c
+++ b/drivers/pci/hotplug/cpci_hotplug_core.c
@@ -285,7 +285,7 @@ cpci_hp_register_bus(struct pci_bus *bus, u8 first, u8 last)
 		info->attention_status = cpci_get_attention_status(slot);
 
 		dbg("registering slot %s", slot->hotplug_slot->name);
-		status = pci_hp_register(slot->hotplug_slot);
+		status = pci_hp_register(slot->hotplug_slot, bus, i);
 		if (status) {
 			err("pci_hp_register failed with error %d", status);
 			goto error_name;
diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 36b115b27b0b..54defec51d08 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -434,7 +434,9 @@ static int ctrl_slot_setup(struct controller *ctrl,
 				slot->bus, slot->device,
 				slot->number, ctrl->slot_device_offset,
 				slot_number);
-		result = pci_hp_register(hotplug_slot);
+		result = pci_hp_register(hotplug_slot,
+					 ctrl->pci_dev->subordinate,
+					 slot->device);
 		if (result) {
 			err("pci_hp_register failed with error %d\n", result);
 			goto error_name;
diff --git a/drivers/pci/hotplug/fakephp.c b/drivers/pci/hotplug/fakephp.c
index b57223f7e830..40337a06c18a 100644
--- a/drivers/pci/hotplug/fakephp.c
+++ b/drivers/pci/hotplug/fakephp.c
@@ -126,7 +126,7 @@ static int add_slot(struct pci_dev *dev)
 	slot->release = &dummy_release;
 	slot->private = dslot;
 
-	retval = pci_hp_register(slot);
+	retval = pci_hp_register(slot, dev->bus, PCI_SLOT(dev->devfn));
 	if (retval) {
 		err("pci_hp_register failed with error %d\n", retval);
 		goto error_dslot;
diff --git a/drivers/pci/hotplug/ibmphp_ebda.c b/drivers/pci/hotplug/ibmphp_ebda.c
index dca7efc14be2..8467d0287325 100644
--- a/drivers/pci/hotplug/ibmphp_ebda.c
+++ b/drivers/pci/hotplug/ibmphp_ebda.c
@@ -1001,7 +1001,8 @@ static int __init ebda_rsrc_controller (void)
 		tmp_slot = list_entry (list, struct slot, ibm_slot_list);
 
 		snprintf (tmp_slot->hotplug_slot->name, 30, "%s", create_file_name (tmp_slot));
-		pci_hp_register (tmp_slot->hotplug_slot);
+		pci_hp_register(tmp_slot->hotplug_slot,
+			pci_find_bus(0, tmp_slot->bus), tmp_slot->device);
 	}
 
 	print_ebda_hpc ();
diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index a11021e8ce37..4df31f375197 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -40,6 +40,7 @@
 #include <linux/pci.h>
 #include <linux/pci_hotplug.h>
 #include <asm/uaccess.h>
+#include "../pci.h"
 
 #define MY_NAME	"pci_hotplug"
 
@@ -60,41 +61,7 @@ static int debug;
 //////////////////////////////////////////////////////////////////
 
 static LIST_HEAD(pci_hotplug_slot_list);
-
-struct kset *pci_hotplug_slots_kset;
-
-static ssize_t hotplug_slot_attr_show(struct kobject *kobj,
-		struct attribute *attr, char *buf)
-{
-	struct hotplug_slot *slot = to_hotplug_slot(kobj);
-	struct hotplug_slot_attribute *attribute = to_hotplug_attr(attr);
-	return attribute->show ? attribute->show(slot, buf) : -EIO;
-}
-
-static ssize_t hotplug_slot_attr_store(struct kobject *kobj,
-		struct attribute *attr, const char *buf, size_t len)
-{
-	struct hotplug_slot *slot = to_hotplug_slot(kobj);
-	struct hotplug_slot_attribute *attribute = to_hotplug_attr(attr);
-	return attribute->store ? attribute->store(slot, buf, len) : -EIO;
-}
-
-static struct sysfs_ops hotplug_slot_sysfs_ops = {
-	.show = hotplug_slot_attr_show,
-	.store = hotplug_slot_attr_store,
-};
-
-static void hotplug_slot_release(struct kobject *kobj)
-{
-	struct hotplug_slot *slot = to_hotplug_slot(kobj);
-	if (slot->release)
-		slot->release(slot);
-}
-
-static struct kobj_type hotplug_slot_ktype = {
-	.sysfs_ops = &hotplug_slot_sysfs_ops,
-	.release = &hotplug_slot_release,
-};
+static DEFINE_SPINLOCK(pci_hotplug_slot_list_lock);
 
 /* these strings match up with the values in pci_bus_speed */
 static char *pci_bus_speed_strings[] = {
@@ -149,16 +116,15 @@ GET_STATUS(power_status, u8)
 GET_STATUS(attention_status, u8)
 GET_STATUS(latch_status, u8)
 GET_STATUS(adapter_status, u8)
-GET_STATUS(address, u32)
 GET_STATUS(max_bus_speed, enum pci_bus_speed)
 GET_STATUS(cur_bus_speed, enum pci_bus_speed)
 
-static ssize_t power_read_file (struct hotplug_slot *slot, char *buf)
+static ssize_t power_read_file(struct pci_slot *slot, char *buf)
 {
 	int retval;
 	u8 value;
 
-	retval = get_power_status (slot, &value);
+	retval = get_power_status(slot->hotplug, &value);
 	if (retval)
 		goto exit;
 	retval = sprintf (buf, "%d\n", value);
@@ -166,9 +132,10 @@ exit:
 	return retval;
 }
 
-static ssize_t power_write_file (struct hotplug_slot *slot, const char *buf,
+static ssize_t power_write_file(struct pci_slot *pci_slot, const char *buf,
 		size_t count)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	unsigned long lpower;
 	u8 power;
 	int retval = 0;
@@ -204,29 +171,30 @@ exit:
 	return count;
 }
 
-static struct hotplug_slot_attribute hotplug_slot_attr_power = {
+static struct pci_slot_attribute hotplug_slot_attr_power = {
 	.attr = {.name = "power", .mode = S_IFREG | S_IRUGO | S_IWUSR},
 	.show = power_read_file,
 	.store = power_write_file
 };
 
-static ssize_t attention_read_file (struct hotplug_slot *slot, char *buf)
+static ssize_t attention_read_file(struct pci_slot *slot, char *buf)
 {
 	int retval;
 	u8 value;
 
-	retval = get_attention_status (slot, &value);
+	retval = get_attention_status(slot->hotplug, &value);
 	if (retval)
 		goto exit;
-	retval = sprintf (buf, "%d\n", value);
+	retval = sprintf(buf, "%d\n", value);
 
 exit:
 	return retval;
 }
 
-static ssize_t attention_write_file (struct hotplug_slot *slot, const char *buf,
+static ssize_t attention_write_file(struct pci_slot *slot, const char *buf,
 		size_t count)
 {
+	struct hotplug_slot_ops *ops = slot->hotplug->ops;
 	unsigned long lattention;
 	u8 attention;
 	int retval = 0;
@@ -235,13 +203,13 @@ static ssize_t attention_write_file (struct hotplug_slot *slot, const char *buf,
 	attention = (u8)(lattention & 0xff);
 	dbg (" - attention = %d\n", attention);
 
-	if (!try_module_get(slot->ops->owner)) {
+	if (!try_module_get(ops->owner)) {
 		retval = -ENODEV;
 		goto exit;
 	}
-	if (slot->ops->set_attention_status)
-		retval = slot->ops->set_attention_status(slot, attention);
-	module_put(slot->ops->owner);
+	if (ops->set_attention_status)
+		retval = ops->set_attention_status(slot->hotplug, attention);
+	module_put(ops->owner);
 
 exit:	
 	if (retval)
@@ -249,18 +217,18 @@ exit:
 	return count;
 }
 
-static struct hotplug_slot_attribute hotplug_slot_attr_attention = {
+static struct pci_slot_attribute hotplug_slot_attr_attention = {
 	.attr = {.name = "attention", .mode = S_IFREG | S_IRUGO | S_IWUSR},
 	.show = attention_read_file,
 	.store = attention_write_file
 };
 
-static ssize_t latch_read_file (struct hotplug_slot *slot, char *buf)
+static ssize_t latch_read_file(struct pci_slot *slot, char *buf)
 {
 	int retval;
 	u8 value;
 
-	retval = get_latch_status (slot, &value);
+	retval = get_latch_status(slot->hotplug, &value);
 	if (retval)
 		goto exit;
 	retval = sprintf (buf, "%d\n", value);
@@ -269,17 +237,17 @@ exit:
 	return retval;
 }
 
-static struct hotplug_slot_attribute hotplug_slot_attr_latch = {
+static struct pci_slot_attribute hotplug_slot_attr_latch = {
 	.attr = {.name = "latch", .mode = S_IFREG | S_IRUGO},
 	.show = latch_read_file,
 };
 
-static ssize_t presence_read_file (struct hotplug_slot *slot, char *buf)
+static ssize_t presence_read_file(struct pci_slot *slot, char *buf)
 {
 	int retval;
 	u8 value;
 
-	retval = get_adapter_status (slot, &value);
+	retval = get_adapter_status(slot->hotplug, &value);
 	if (retval)
 		goto exit;
 	retval = sprintf (buf, "%d\n", value);
@@ -288,42 +256,20 @@ exit:
 	return retval;
 }
 
-static struct hotplug_slot_attribute hotplug_slot_attr_presence = {
+static struct pci_slot_attribute hotplug_slot_attr_presence = {
 	.attr = {.name = "adapter", .mode = S_IFREG | S_IRUGO},
 	.show = presence_read_file,
 };
 
-static ssize_t address_read_file (struct hotplug_slot *slot, char *buf)
-{
-	int retval;
-	u32 address;
-
-	retval = get_address (slot, &address);
-	if (retval)
-		goto exit;
-	retval = sprintf (buf, "%04x:%02x:%02x\n",
-			  (address >> 16) & 0xffff,
-			  (address >> 8) & 0xff,
-			  address & 0xff);
-
-exit:
-	return retval;
-}
-
-static struct hotplug_slot_attribute hotplug_slot_attr_address = {
-	.attr = {.name = "address", .mode = S_IFREG | S_IRUGO},
-	.show = address_read_file,
-};
-
 static char *unknown_speed = "Unknown bus speed";
 
-static ssize_t max_bus_speed_read_file (struct hotplug_slot *slot, char *buf)
+static ssize_t max_bus_speed_read_file(struct pci_slot *slot, char *buf)
 {
 	char *speed_string;
 	int retval;
 	enum pci_bus_speed value;
 	
-	retval = get_max_bus_speed (slot, &value);
+	retval = get_max_bus_speed(slot->hotplug, &value);
 	if (retval)
 		goto exit;
 
@@ -338,18 +284,18 @@ exit:
 	return retval;
 }
 
-static struct hotplug_slot_attribute hotplug_slot_attr_max_bus_speed = {
+static struct pci_slot_attribute hotplug_slot_attr_max_bus_speed = {
 	.attr = {.name = "max_bus_speed", .mode = S_IFREG | S_IRUGO},
 	.show = max_bus_speed_read_file,
 };
 
-static ssize_t cur_bus_speed_read_file (struct hotplug_slot *slot, char *buf)
+static ssize_t cur_bus_speed_read_file(struct pci_slot *slot, char *buf)
 {
 	char *speed_string;
 	int retval;
 	enum pci_bus_speed value;
 
-	retval = get_cur_bus_speed (slot, &value);
+	retval = get_cur_bus_speed(slot->hotplug, &value);
 	if (retval)
 		goto exit;
 
@@ -364,14 +310,15 @@ exit:
 	return retval;
 }
 
-static struct hotplug_slot_attribute hotplug_slot_attr_cur_bus_speed = {
+static struct pci_slot_attribute hotplug_slot_attr_cur_bus_speed = {
 	.attr = {.name = "cur_bus_speed", .mode = S_IFREG | S_IRUGO},
 	.show = cur_bus_speed_read_file,
 };
 
-static ssize_t test_write_file (struct hotplug_slot *slot, const char *buf,
+static ssize_t test_write_file(struct pci_slot *pci_slot, const char *buf,
 		size_t count)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	unsigned long ltest;
 	u32 test;
 	int retval = 0;
@@ -394,13 +341,14 @@ exit:
 	return count;
 }
 
-static struct hotplug_slot_attribute hotplug_slot_attr_test = {
+static struct pci_slot_attribute hotplug_slot_attr_test = {
 	.attr = {.name = "test", .mode = S_IFREG | S_IRUGO | S_IWUSR},
 	.store = test_write_file
 };
 
-static int has_power_file (struct hotplug_slot *slot)
+static int has_power_file(struct pci_slot *pci_slot)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
 		return -ENODEV;
 	if ((slot->ops->enable_slot) ||
@@ -410,8 +358,9 @@ static int has_power_file (struct hotplug_slot *slot)
 	return -ENOENT;
 }
 
-static int has_attention_file (struct hotplug_slot *slot)
+static int has_attention_file(struct pci_slot *pci_slot)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
 		return -ENODEV;
 	if ((slot->ops->set_attention_status) ||
@@ -420,8 +369,9 @@ static int has_attention_file (struct hotplug_slot *slot)
 	return -ENOENT;
 }
 
-static int has_latch_file (struct hotplug_slot *slot)
+static int has_latch_file(struct pci_slot *pci_slot)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
 		return -ENODEV;
 	if (slot->ops->get_latch_status)
@@ -429,8 +379,9 @@ static int has_latch_file (struct hotplug_slot *slot)
 	return -ENOENT;
 }
 
-static int has_adapter_file (struct hotplug_slot *slot)
+static int has_adapter_file(struct pci_slot *pci_slot)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
 		return -ENODEV;
 	if (slot->ops->get_adapter_status)
@@ -438,17 +389,9 @@ static int has_adapter_file (struct hotplug_slot *slot)
 	return -ENOENT;
 }
 
-static int has_address_file (struct hotplug_slot *slot)
-{
-	if ((!slot) || (!slot->ops))
-		return -ENODEV;
-	if (slot->ops->get_address)
-		return 0;
-	return -ENOENT;
-}
-
-static int has_max_bus_speed_file (struct hotplug_slot *slot)
+static int has_max_bus_speed_file(struct pci_slot *pci_slot)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
 		return -ENODEV;
 	if (slot->ops->get_max_bus_speed)
@@ -456,8 +399,9 @@ static int has_max_bus_speed_file (struct hotplug_slot *slot)
 	return -ENOENT;
 }
 
-static int has_cur_bus_speed_file (struct hotplug_slot *slot)
+static int has_cur_bus_speed_file(struct pci_slot *pci_slot)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
 		return -ENODEV;
 	if (slot->ops->get_cur_bus_speed)
@@ -465,8 +409,9 @@ static int has_cur_bus_speed_file (struct hotplug_slot *slot)
 	return -ENOENT;
 }
 
-static int has_test_file (struct hotplug_slot *slot)
+static int has_test_file(struct pci_slot *pci_slot)
 {
+	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
 		return -ENODEV;
 	if (slot->ops->hardware_test)
@@ -474,7 +419,7 @@ static int has_test_file (struct hotplug_slot *slot)
 	return -ENOENT;
 }
 
-static int fs_add_slot (struct hotplug_slot *slot)
+static int fs_add_slot(struct pci_slot *slot)
 {
 	int retval = 0;
 
@@ -505,13 +450,6 @@ static int fs_add_slot (struct hotplug_slot *slot)
 			goto exit_adapter;
 	}
 
-	if (has_address_file(slot) == 0) {
-		retval = sysfs_create_file(&slot->kobj,
-					   &hotplug_slot_attr_address.attr);
-		if (retval)
-			goto exit_address;
-	}
-
 	if (has_max_bus_speed_file(slot) == 0) {
 		retval = sysfs_create_file(&slot->kobj,
 					   &hotplug_slot_attr_max_bus_speed.attr);
@@ -544,10 +482,6 @@ exit_cur_speed:
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr);
 
 exit_max_speed:
-	if (has_address_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_address.attr);
-
-exit_address:
 	if (has_adapter_file(slot) == 0)
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_presence.attr);
 
@@ -567,7 +501,7 @@ exit:
 	return retval;
 }
 
-static void fs_remove_slot (struct hotplug_slot *slot)
+static void fs_remove_slot(struct pci_slot *slot)
 {
 	if (has_power_file(slot) == 0)
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_power.attr);
@@ -581,9 +515,6 @@ static void fs_remove_slot (struct hotplug_slot *slot)
 	if (has_adapter_file(slot) == 0)
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_presence.attr);
 
-	if (has_address_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_address.attr);
-
 	if (has_max_bus_speed_file(slot) == 0)
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr);
 
@@ -599,12 +530,16 @@ static struct hotplug_slot *get_slot_from_name (const char *name)
 	struct hotplug_slot *slot;
 	struct list_head *tmp;
 
+	spin_lock(&pci_hotplug_slot_list_lock);
 	list_for_each (tmp, &pci_hotplug_slot_list) {
 		slot = list_entry (tmp, struct hotplug_slot, slot_list);
 		if (strcmp(slot->name, name) == 0)
-			return slot;
+			goto out;
 	}
-	return NULL;
+	slot = NULL;
+out:
+	spin_unlock(&pci_hotplug_slot_list_lock);
+	return slot;
 }
 
 /**
@@ -616,9 +551,10 @@ static struct hotplug_slot *get_slot_from_name (const char *name)
  *
  * Returns 0 if successful, anything else for an error.
  */
-int pci_hp_register (struct hotplug_slot *slot)
+int pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus, int slot_nr)
 {
 	int result;
+	struct pci_slot *pci_slot;
 	struct hotplug_slot *tmp;
 
 	if (slot == NULL)
@@ -636,19 +572,28 @@ int pci_hp_register (struct hotplug_slot *slot)
 	if (tmp)
 		return -EEXIST;
 
-	slot->kobj.kset = pci_hotplug_slots_kset;
-	result = kobject_init_and_add(&slot->kobj, &hotplug_slot_ktype, NULL,
-				      "%s", slot->name);
-	if (result) {
-		err("Unable to register kobject '%s'", slot->name);
-		return -EINVAL;
+	pci_slot = pci_create_slot(bus, slot_nr, slot->name);
+	if (IS_ERR(pci_slot))
+		return PTR_ERR(pci_slot);
+
+	if (pci_slot->hotplug) {
+		dbg("%s: already claimed\n", __func__);
+		pci_destroy_slot(pci_slot);
+		return -EBUSY;
 	}
 
-	list_add (&slot->slot_list, &pci_hotplug_slot_list);
+	slot->pci_slot = pci_slot;
+	pci_slot->hotplug = slot;
+
+	spin_lock(&pci_hotplug_slot_list_lock);
+	list_add(&slot->slot_list, &pci_hotplug_slot_list);
+	spin_unlock(&pci_hotplug_slot_list_lock);
+
+	result = fs_add_slot(pci_slot);
+	kobject_uevent(&pci_slot->kobj, KOBJ_ADD);
+	dbg("Added slot %s to the list\n", slot->name);
+
 
-	result = fs_add_slot (slot);
-	kobject_uevent(&slot->kobj, KOBJ_ADD);
-	dbg ("Added slot %s to the list\n", slot->name);
 	return result;
 }
 
@@ -661,22 +606,30 @@ int pci_hp_register (struct hotplug_slot *slot)
  *
  * Returns 0 if successful, anything else for an error.
  */
-int pci_hp_deregister (struct hotplug_slot *slot)
+int pci_hp_deregister(struct hotplug_slot *hotplug)
 {
 	struct hotplug_slot *temp;
+	struct pci_slot *slot;
 
-	if (slot == NULL)
+	if (!hotplug)
 		return -ENODEV;
 
-	temp = get_slot_from_name (slot->name);
-	if (temp != slot) {
+	temp = get_slot_from_name(hotplug->name);
+	if (temp != hotplug)
 		return -ENODEV;
-	}
-	list_del (&slot->slot_list);
 
-	fs_remove_slot (slot);
-	dbg ("Removed slot %s from the list\n", slot->name);
-	kobject_put(&slot->kobj);
+	spin_lock(&pci_hotplug_slot_list_lock);
+	list_del(&hotplug->slot_list);
+	spin_unlock(&pci_hotplug_slot_list_lock);
+
+	slot = hotplug->pci_slot;
+	fs_remove_slot(slot);
+	dbg("Removed slot %s from the list\n", hotplug->name);
+
+	hotplug->release(hotplug);
+	slot->hotplug = NULL;
+	pci_destroy_slot(slot);
+
 	return 0;
 }
 
@@ -690,13 +643,15 @@ int pci_hp_deregister (struct hotplug_slot *slot)
  *
  * Returns 0 if successful, anything else for an error.
  */
-int __must_check pci_hp_change_slot_info(struct hotplug_slot *slot,
+int __must_check pci_hp_change_slot_info(struct hotplug_slot *hotplug,
 					 struct hotplug_slot_info *info)
 {
-	if ((slot == NULL) || (info == NULL))
+	struct pci_slot *slot;
+	if (!hotplug || !info)
 		return -ENODEV;
+	slot = hotplug->pci_slot;
 
-	memcpy (slot->info, info, sizeof (struct hotplug_slot_info));
+	memcpy(hotplug->info, info, sizeof(struct hotplug_slot_info));
 
 	return 0;
 }
@@ -704,36 +659,22 @@ int __must_check pci_hp_change_slot_info(struct hotplug_slot *slot,
 static int __init pci_hotplug_init (void)
 {
 	int result;
-	struct kset *pci_bus_kset;
 
-	pci_bus_kset = bus_get_kset(&pci_bus_type);
-
-	pci_hotplug_slots_kset = kset_create_and_add("slots", NULL,
-						     &pci_bus_kset->kobj);
-	if (!pci_hotplug_slots_kset) {
-		result = -ENOMEM;
-		err("Register subsys error\n");
-		goto exit;
-	}
 	result = cpci_hotplug_init(debug);
 	if (result) {
 		err ("cpci_hotplug_init with error %d\n", result);
-		goto err_subsys;
+		goto err_cpci;
 	}
 
 	info (DRIVER_DESC " version: " DRIVER_VERSION "\n");
-	goto exit;
 
-err_subsys:
-	kset_unregister(pci_hotplug_slots_kset);
-exit:
+err_cpci:
 	return result;
 }
 
 static void __exit pci_hotplug_exit (void)
 {
 	cpci_hotplug_exit();
-	kset_unregister(pci_hotplug_slots_kset);
 }
 
 module_init(pci_hotplug_init);
@@ -745,7 +686,6 @@ MODULE_LICENSE("GPL");
 module_param(debug, bool, 0644);
 MODULE_PARM_DESC(debug, "Debugging mode enabled or not");
 
-EXPORT_SYMBOL_GPL(pci_hotplug_slots_kset);
 EXPORT_SYMBOL_GPL(pci_hp_register);
 EXPORT_SYMBOL_GPL(pci_hp_deregister);
 EXPORT_SYMBOL_GPL(pci_hp_change_slot_info);
diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c
index 49414e9100de..7b21c86e4bff 100644
--- a/drivers/pci/hotplug/pciehp_core.c
+++ b/drivers/pci/hotplug/pciehp_core.c
@@ -72,7 +72,6 @@ static int get_power_status	(struct hotplug_slot *slot, u8 *value);
 static int get_attention_status	(struct hotplug_slot *slot, u8 *value);
 static int get_latch_status	(struct hotplug_slot *slot, u8 *value);
 static int get_adapter_status	(struct hotplug_slot *slot, u8 *value);
-static int get_address		(struct hotplug_slot *slot, u32 *value);
 static int get_max_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 static int get_cur_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 
@@ -85,7 +84,6 @@ static struct hotplug_slot_ops pciehp_hotplug_slot_ops = {
 	.get_attention_status =	get_attention_status,
 	.get_latch_status =	get_latch_status,
 	.get_adapter_status =	get_adapter_status,
-	.get_address =		get_address,
   	.get_max_bus_speed =	get_max_bus_speed,
   	.get_cur_bus_speed =	get_cur_bus_speed,
 };
@@ -252,7 +250,9 @@ static int init_slots(struct controller *ctrl)
 		dbg("Registering bus=%x dev=%x hp_slot=%x sun=%x "
 		    "slot_device_offset=%x\n", slot->bus, slot->device,
 		    slot->hp_slot, slot->number, ctrl->slot_device_offset);
-		retval = pci_hp_register(hotplug_slot);
+		retval = pci_hp_register(hotplug_slot,
+					 ctrl->pci_dev->subordinate,
+					 slot->device);
 		if (retval) {
 			err("pci_hp_register failed with error %d\n", retval);
 			if (retval == -EEXIST)
@@ -263,7 +263,7 @@ static int init_slots(struct controller *ctrl)
 		}
 		/* create additional sysfs entries */
 		if (EMI(ctrl)) {
-			retval = sysfs_create_file(&hotplug_slot->kobj,
+			retval = sysfs_create_file(&hotplug_slot->pci_slot->kobj,
 				&hotplug_slot_attr_lock.attr);
 			if (retval) {
 				pci_hp_deregister(hotplug_slot);
@@ -296,7 +296,7 @@ static void cleanup_slots(struct controller *ctrl)
 		slot = list_entry(tmp, struct slot, slot_list);
 		list_del(&slot->slot_list);
 		if (EMI(ctrl))
-			sysfs_remove_file(&slot->hotplug_slot->kobj,
+			sysfs_remove_file(&slot->hotplug_slot->pci_slot->kobj,
 				&hotplug_slot_attr_lock.attr);
 		cancel_delayed_work(&slot->work);
 		flush_scheduled_work();
@@ -398,19 +398,8 @@ static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value)
 	return 0;
 }
 
-static int get_address(struct hotplug_slot *hotplug_slot, u32 *value)
-{
-	struct slot *slot = hotplug_slot->private;
-	struct pci_bus *bus = slot->ctrl->pci_dev->subordinate;
-
-	dbg("%s - physical_slot = %s\n", __func__, hotplug_slot->name);
-
-	*value = (pci_domain_nr(bus) << 16) | (slot->bus << 8) | slot->device;
-
-	return 0;
-}
-
-static int get_max_bus_speed(struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value)
+static int get_max_bus_speed(struct hotplug_slot *hotplug_slot,
+				enum pci_bus_speed *value)
 {
 	struct slot *slot = hotplug_slot->private;
 	int retval;
@@ -474,7 +463,12 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_
 	/* Setup the slot information structures */
 	rc = init_slots(ctrl);
 	if (rc) {
-		err("%s: slot initialization failed\n", PCIE_MODULE_NAME);
+		if (rc == -EBUSY)
+			warn("%s: slot already registered by another "
+				"hotplug driver\n", PCIE_MODULE_NAME);
+		else
+			err("%s: slot initialization failed\n",
+				PCIE_MODULE_NAME);
 		goto err_out_release_ctlr;
 	}
 
diff --git a/drivers/pci/hotplug/rpadlpar_sysfs.c b/drivers/pci/hotplug/rpadlpar_sysfs.c
index 779c5db71be4..a796301ea03f 100644
--- a/drivers/pci/hotplug/rpadlpar_sysfs.c
+++ b/drivers/pci/hotplug/rpadlpar_sysfs.c
@@ -14,8 +14,10 @@
  */
 #include <linux/kobject.h>
 #include <linux/string.h>
+#include <linux/pci.h>
 #include <linux/pci_hotplug.h>
 #include "rpadlpar.h"
+#include "../pci.h"
 
 #define DLPAR_KOBJ_NAME       "control"
 
@@ -27,7 +29,6 @@
 
 #define MAX_DRC_NAME_LEN 64
 
-
 static ssize_t add_slot_store(struct kobject *kobj, struct kobj_attribute *attr,
 			      const char *buf, size_t nbytes)
 {
@@ -112,7 +113,7 @@ int dlpar_sysfs_init(void)
 	int error;
 
 	dlpar_kobj = kobject_create_and_add(DLPAR_KOBJ_NAME,
-					    &pci_hotplug_slots_kset->kobj);
+					    &pci_slots_kset->kobj);
 	if (!dlpar_kobj)
 		return -EINVAL;
 
diff --git a/drivers/pci/hotplug/rpaphp_slot.c b/drivers/pci/hotplug/rpaphp_slot.c
index 56197b600d36..9b714ea93d20 100644
--- a/drivers/pci/hotplug/rpaphp_slot.c
+++ b/drivers/pci/hotplug/rpaphp_slot.c
@@ -33,33 +33,6 @@
 #include <asm/rtas.h>
 #include "rpaphp.h"
 
-static ssize_t address_read_file (struct hotplug_slot *php_slot, char *buf)
-{
-	int retval;
-	struct slot *slot = (struct slot *)php_slot->private;
-	struct pci_bus *bus;
-
-	if (!slot)
-		return -ENOENT;
-
-	bus = slot->bus;
-	if (!bus)
-		return -ENOENT;
-
-	if (bus->self)
-		retval = sprintf(buf, pci_name(bus->self));
-	else
-		retval = sprintf(buf, "%04x:%02x:00.0",
-		        pci_domain_nr(bus), bus->number);
-
-	return retval;
-}
-
-static struct hotplug_slot_attribute php_attr_address = {
-	.attr = {.name = "address", .mode = S_IFREG | S_IRUGO},
-	.show = address_read_file,
-};
-
 /* free up the memory used by a slot */
 static void rpaphp_release_slot(struct hotplug_slot *hotplug_slot)
 {
@@ -135,9 +108,6 @@ int rpaphp_deregister_slot(struct slot *slot)
 
 	list_del(&slot->rpaphp_slot_list);
 	
-	/* remove "address" file */
-	sysfs_remove_file(&php_slot->kobj, &php_attr_address.attr);
-
 	retval = pci_hp_deregister(php_slot);
 	if (retval)
 		err("Problem unregistering a slot %s\n", slot->name);
@@ -151,6 +121,7 @@ int rpaphp_register_slot(struct slot *slot)
 {
 	struct hotplug_slot *php_slot = slot->hotplug_slot;
 	int retval;
+	int slotno;
 
 	dbg("%s registering slot:path[%s] index[%x], name[%s] pdomain[%x] type[%d]\n", 
 		__func__, slot->dn->full_name, slot->index, slot->name,
@@ -162,19 +133,16 @@ int rpaphp_register_slot(struct slot *slot)
 		return -EAGAIN;
 	}	
 
-	retval = pci_hp_register(php_slot);
+	if (slot->dn->child)
+		slotno = PCI_SLOT(PCI_DN(slot->dn->child)->devfn);
+	else
+		slotno = -1;
+	retval = pci_hp_register(php_slot, slot->bus, slotno);
 	if (retval) {
 		err("pci_hp_register failed with error %d\n", retval);
 		return retval;
 	}
 
-	/* create "address" file */
-	retval = sysfs_create_file(&php_slot->kobj, &php_attr_address.attr);
-	if (retval) {
-		err("sysfs_create_file failed with error %d\n", retval);
-		goto sysfs_fail;
-	}
-
 	/* add slot to our internal list */
 	list_add(&slot->rpaphp_slot_list, &rpaphp_slot_head);
 	info("Slot [%s] registered\n", slot->name);
diff --git a/drivers/pci/hotplug/sgi_hotplug.c b/drivers/pci/hotplug/sgi_hotplug.c
index 2fe37cd85b69..2036a43ff2fd 100644
--- a/drivers/pci/hotplug/sgi_hotplug.c
+++ b/drivers/pci/hotplug/sgi_hotplug.c
@@ -197,13 +197,15 @@ static int sn_hp_slot_private_alloc(struct hotplug_slot *bss_hotplug_slot,
 static struct hotplug_slot * sn_hp_destroy(void)
 {
 	struct slot *slot;
+	struct pci_slot *pci_slot;
 	struct hotplug_slot *bss_hotplug_slot = NULL;
 
 	list_for_each_entry(slot, &sn_hp_list, hp_list) {
 		bss_hotplug_slot = slot->hotplug_slot;
+		pci_slot = bss_hotplug_slot->pci_slot;
 		list_del(&((struct slot *)bss_hotplug_slot->private)->
 			 hp_list);
-		sysfs_remove_file(&bss_hotplug_slot->kobj,
+		sysfs_remove_file(&pci_slot->kobj,
 				  &sn_slot_path_attr.attr);
 		break;
 	}
@@ -614,6 +616,7 @@ static void sn_release_slot(struct hotplug_slot *bss_hotplug_slot)
 static int sn_hotplug_slot_register(struct pci_bus *pci_bus)
 {
 	int device;
+	struct pci_slot *pci_slot;
 	struct hotplug_slot *bss_hotplug_slot;
 	int rc = 0;
 
@@ -650,11 +653,12 @@ static int sn_hotplug_slot_register(struct pci_bus *pci_bus)
 		bss_hotplug_slot->ops = &sn_hotplug_slot_ops;
 		bss_hotplug_slot->release = &sn_release_slot;
 
-		rc = pci_hp_register(bss_hotplug_slot);
+		rc = pci_hp_register(bss_hotplug_slot, pci_bus, device);
 		if (rc)
 			goto register_err;
 
-		rc = sysfs_create_file(&bss_hotplug_slot->kobj,
+		pci_slot = bss_hotplug_slot->pci_slot;
+		rc = sysfs_create_file(&pci_slot->kobj,
 				       &sn_slot_path_attr.attr);
 		if (rc)
 			goto register_err;
diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c
index df41ecc4e7fc..a8cbd039b85b 100644
--- a/drivers/pci/hotplug/shpchp_core.c
+++ b/drivers/pci/hotplug/shpchp_core.c
@@ -68,7 +68,6 @@ static int get_power_status	(struct hotplug_slot *slot, u8 *value);
 static int get_attention_status	(struct hotplug_slot *slot, u8 *value);
 static int get_latch_status	(struct hotplug_slot *slot, u8 *value);
 static int get_adapter_status	(struct hotplug_slot *slot, u8 *value);
-static int get_address		(struct hotplug_slot *slot, u32 *value);
 static int get_max_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 static int get_cur_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 
@@ -81,7 +80,6 @@ static struct hotplug_slot_ops shpchp_hotplug_slot_ops = {
 	.get_attention_status =	get_attention_status,
 	.get_latch_status =	get_latch_status,
 	.get_adapter_status =	get_adapter_status,
-	.get_address =		get_address,
 	.get_max_bus_speed =	get_max_bus_speed,
 	.get_cur_bus_speed =	get_cur_bus_speed,
 };
@@ -159,7 +157,8 @@ static int init_slots(struct controller *ctrl)
 		dbg("Registering bus=%x dev=%x hp_slot=%x sun=%x "
 		    "slot_device_offset=%x\n", slot->bus, slot->device,
 		    slot->hp_slot, slot->number, ctrl->slot_device_offset);
-		retval = pci_hp_register(slot->hotplug_slot);
+		retval = pci_hp_register(slot->hotplug_slot,
+				ctrl->pci_dev->subordinate, slot->device);
 		if (retval) {
 			err("pci_hp_register failed with error %d\n", retval);
 			if (retval == -EEXIST)
@@ -288,19 +287,8 @@ static int get_adapter_status (struct hotplug_slot *hotplug_slot, u8 *value)
 	return 0;
 }
 
-static int get_address (struct hotplug_slot *hotplug_slot, u32 *value)
-{
-	struct slot *slot = get_slot(hotplug_slot);
-	struct pci_bus *bus = slot->ctrl->pci_dev->subordinate;
-
-	dbg("%s - physical_slot = %s\n", __func__, hotplug_slot->name);
-
-	*value = (pci_domain_nr(bus) << 16) | (slot->bus << 8) | slot->device;
-
-	return 0;
-}
-
-static int get_max_bus_speed (struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value)
+static int get_max_bus_speed(struct hotplug_slot *hotplug_slot,
+				enum pci_bus_speed *value)
 {
 	struct slot *slot = get_slot(hotplug_slot);
 	int retval;
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 0a497c1b4227..e1d7bbf079b4 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -106,3 +106,16 @@ pci_match_one_device(const struct pci_device_id *id, const struct pci_dev *dev)
 }
 
 struct pci_dev *pci_find_upstream_pcie_bridge(struct pci_dev *pdev);
+
+/* PCI slot sysfs helper code */
+#define to_pci_slot(s) container_of(s, struct pci_slot, kobj)
+
+extern struct kset *pci_slots_kset;
+
+struct pci_slot_attribute {
+	struct attribute attr;
+	ssize_t (*show)(struct pci_slot *, char *);
+	ssize_t (*store)(struct pci_slot *, const char *, size_t);
+};
+#define to_pci_slot_attr(s) container_of(s, struct pci_slot_attribute, attr)
+
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index aebab71abebf..4562827b7e22 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -414,6 +414,7 @@ static struct pci_bus * pci_alloc_bus(void)
 		INIT_LIST_HEAD(&b->node);
 		INIT_LIST_HEAD(&b->children);
 		INIT_LIST_HEAD(&b->devices);
+		INIT_LIST_HEAD(&b->slots);
 	}
 	return b;
 }
diff --git a/drivers/pci/slot.c b/drivers/pci/slot.c
new file mode 100644
index 000000000000..7e5b85cbd948
--- /dev/null
+++ b/drivers/pci/slot.c
@@ -0,0 +1,233 @@
+/*
+ * drivers/pci/slot.c
+ * Copyright (C) 2006 Matthew Wilcox <matthew@wil.cx>
+ * Copyright (C) 2006-2008 Hewlett-Packard Development Company, L.P.
+ * 	Alex Chiang <achiang@hp.com>
+ */
+
+#include <linux/kobject.h>
+#include <linux/pci.h>
+#include <linux/err.h>
+#include "pci.h"
+
+struct kset *pci_slots_kset;
+EXPORT_SYMBOL_GPL(pci_slots_kset);
+
+static ssize_t pci_slot_attr_show(struct kobject *kobj,
+					struct attribute *attr, char *buf)
+{
+	struct pci_slot *slot = to_pci_slot(kobj);
+	struct pci_slot_attribute *attribute = to_pci_slot_attr(attr);
+	return attribute->show ? attribute->show(slot, buf) : -EIO;
+}
+
+static ssize_t pci_slot_attr_store(struct kobject *kobj,
+			struct attribute *attr, const char *buf, size_t len)
+{
+	struct pci_slot *slot = to_pci_slot(kobj);
+	struct pci_slot_attribute *attribute = to_pci_slot_attr(attr);
+	return attribute->store ? attribute->store(slot, buf, len) : -EIO;
+}
+
+static struct sysfs_ops pci_slot_sysfs_ops = {
+	.show = pci_slot_attr_show,
+	.store = pci_slot_attr_store,
+};
+
+static ssize_t address_read_file(struct pci_slot *slot, char *buf)
+{
+	if (slot->number == 0xff)
+		return sprintf(buf, "%04x:%02x\n",
+				pci_domain_nr(slot->bus),
+				slot->bus->number);
+	else
+		return sprintf(buf, "%04x:%02x:%02x\n",
+				pci_domain_nr(slot->bus),
+				slot->bus->number,
+				slot->number);
+}
+
+static void pci_slot_release(struct kobject *kobj)
+{
+	struct pci_slot *slot = to_pci_slot(kobj);
+
+	pr_debug("%s: releasing pci_slot on %x:%d\n", __func__,
+		 slot->bus->number, slot->number);
+
+	list_del(&slot->list);
+
+	kfree(slot);
+}
+
+static struct pci_slot_attribute pci_slot_attr_address =
+	__ATTR(address, (S_IFREG | S_IRUGO), address_read_file, NULL);
+
+static struct attribute *pci_slot_default_attrs[] = {
+	&pci_slot_attr_address.attr,
+	NULL,
+};
+
+static struct kobj_type pci_slot_ktype = {
+	.sysfs_ops = &pci_slot_sysfs_ops,
+	.release = &pci_slot_release,
+	.default_attrs = pci_slot_default_attrs,
+};
+
+/**
+ * pci_create_slot - create or increment refcount for physical PCI slot
+ * @parent: struct pci_bus of parent bridge
+ * @slot_nr: PCI_SLOT(pci_dev->devfn) or -1 for placeholder
+ * @name: user visible string presented in /sys/bus/pci/slots/<name>
+ *
+ * PCI slots have first class attributes such as address, speed, width,
+ * and a &struct pci_slot is used to manage them. This interface will
+ * either return a new &struct pci_slot to the caller, or if the pci_slot
+ * already exists, its refcount will be incremented.
+ *
+ * Slots are uniquely identified by a @pci_bus, @slot_nr, @name tuple.
+ *
+ * Placeholder slots:
+ * In most cases, @pci_bus, @slot_nr will be sufficient to uniquely identify
+ * a slot. There is one notable exception - pSeries (rpaphp), where the
+ * @slot_nr cannot be determined until a device is actually inserted into
+ * the slot. In this scenario, the caller may pass -1 for @slot_nr.
+ *
+ * The following semantics are imposed when the caller passes @slot_nr ==
+ * -1. First, the check for existing %struct pci_slot is skipped, as the
+ * caller may know about several unpopulated slots on a given %struct
+ * pci_bus, and each slot would have a @slot_nr of -1.  Uniqueness for
+ * these slots is then determined by the @name parameter. We expect
+ * kobject_init_and_add() to warn us if the caller attempts to create
+ * multiple slots with the same name. The other change in semantics is
+ * user-visible, which is the 'address' parameter presented in sysfs will
+ * consist solely of a dddd:bb tuple, where dddd is the PCI domain of the
+ * %struct pci_bus and bb is the bus number. In other words, the devfn of
+ * the 'placeholder' slot will not be displayed.
+ */
+
+struct pci_slot *pci_create_slot(struct pci_bus *parent, int slot_nr,
+				 const char *name)
+{
+	struct pci_slot *slot;
+	int err;
+
+	down_write(&pci_bus_sem);
+
+	if (slot_nr == -1)
+		goto placeholder;
+
+	/* If we've already created this slot, bump refcount and return. */
+	list_for_each_entry(slot, &parent->slots, list) {
+		if (slot->number == slot_nr) {
+			kobject_get(&slot->kobj);
+			pr_debug("%s: inc refcount to %d on %04x:%02x:%02x\n",
+				 __func__,
+				 atomic_read(&slot->kobj.kref.refcount),
+				 pci_domain_nr(parent), parent->number,
+				 slot_nr);
+			goto out;
+		}
+	}
+
+placeholder:
+	slot = kzalloc(sizeof(*slot), GFP_KERNEL);
+	if (!slot) {
+		slot = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+
+	slot->bus = parent;
+	slot->number = slot_nr;
+
+	slot->kobj.kset = pci_slots_kset;
+	err = kobject_init_and_add(&slot->kobj, &pci_slot_ktype, NULL,
+				   "%s", name);
+	if (err) {
+		printk(KERN_ERR "Unable to register kobject %s\n", name);
+		goto err;
+	}
+
+	INIT_LIST_HEAD(&slot->list);
+	list_add(&slot->list, &parent->slots);
+
+	/* Don't care if debug printk has a -1 for slot_nr */
+	pr_debug("%s: created pci_slot on %04x:%02x:%02x\n",
+		 __func__, pci_domain_nr(parent), parent->number, slot_nr);
+
+ out:
+	up_write(&pci_bus_sem);
+	return slot;
+ err:
+	kfree(slot);
+	slot = ERR_PTR(err);
+	goto out;
+}
+EXPORT_SYMBOL_GPL(pci_create_slot);
+
+/**
+ * pci_update_slot_number - update %struct pci_slot -> number
+ * @slot - %struct pci_slot to update
+ * @slot_nr - new number for slot
+ *
+ * The primary purpose of this interface is to allow callers who earlier
+ * created a placeholder slot in pci_create_slot() by passing a -1 as
+ * slot_nr, to update their %struct pci_slot with the correct @slot_nr.
+ */
+
+void pci_update_slot_number(struct pci_slot *slot, int slot_nr)
+{
+	int name_count = 0;
+	struct pci_slot *tmp;
+
+	down_write(&pci_bus_sem);
+
+	list_for_each_entry(tmp, &slot->bus->slots, list) {
+		WARN_ON(tmp->number == slot_nr);
+		if (!strcmp(kobject_name(&tmp->kobj), kobject_name(&slot->kobj)))
+			name_count++;
+	}
+
+	if (name_count > 1)
+		printk(KERN_WARNING "pci_update_slot_number found %d slots with the same name: %s\n", name_count, kobject_name(&slot->kobj));
+
+	slot->number = slot_nr;
+	up_write(&pci_bus_sem);
+}
+EXPORT_SYMBOL_GPL(pci_update_slot_number);
+
+/**
+ * pci_destroy_slot - decrement refcount for physical PCI slot
+ * @slot: struct pci_slot to decrement
+ *
+ * %struct pci_slot is refcounted, so destroying them is really easy; we
+ * just call kobject_put on its kobj and let our release methods do the
+ * rest.
+ */
+
+void pci_destroy_slot(struct pci_slot *slot)
+{
+	pr_debug("%s: dec refcount to %d on %04x:%02x:%02x\n", __func__,
+		 atomic_read(&slot->kobj.kref.refcount) - 1,
+		 pci_domain_nr(slot->bus), slot->bus->number, slot->number);
+
+	down_write(&pci_bus_sem);
+	kobject_put(&slot->kobj);
+	up_write(&pci_bus_sem);
+}
+EXPORT_SYMBOL_GPL(pci_destroy_slot);
+
+static int pci_slot_init(void)
+{
+	struct kset *pci_bus_kset;
+
+	pci_bus_kset = bus_get_kset(&pci_bus_type);
+	pci_slots_kset = kset_create_and_add("slots", NULL,
+						&pci_bus_kset->kobj);
+	if (!pci_slots_kset) {
+		printk(KERN_ERR "PCI: Slot initialization failure\n");
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+subsys_initcall(pci_slot_init);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 507ee52323cd..f1f73f79a180 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -17,8 +17,7 @@
 #ifndef LINUX_PCI_H
 #define LINUX_PCI_H
 
-/* Include the pci register defines */
-#include <linux/pci_regs.h>
+#include <linux/pci_regs.h>	/* The pci register defines */
 
 /*
  * The PCI interface treats multi-function devices as independent
@@ -49,12 +48,22 @@
 #include <linux/list.h>
 #include <linux/compiler.h>
 #include <linux/errno.h>
+#include <linux/kobject.h>
 #include <asm/atomic.h>
 #include <linux/device.h>
 
 /* Include the ID list */
 #include <linux/pci_ids.h>
 
+/* pci_slot represents a physical slot */
+struct pci_slot {
+	struct pci_bus *bus;		/* The bus this slot is on */
+	struct list_head list;		/* node in list of slots on this bus */
+	struct hotplug_slot *hotplug;	/* Hotplug info (migrate over time) */
+	unsigned char number;		/* PCI_SLOT(pci_dev->devfn) */
+	struct kobject kobj;
+};
+
 /* File state for mmap()s on /proc/bus/pci/X/Y */
 enum pci_mmap_state {
 	pci_mmap_io,
@@ -142,6 +151,7 @@ struct pci_dev {
 
 	void		*sysdata;	/* hook for sys-specific extension */
 	struct proc_dir_entry *procent;	/* device entry in /proc/bus/pci */
+	struct pci_slot	*slot;		/* Physical slot this device is in */
 
 	unsigned int	devfn;		/* encoded device & function index */
 	unsigned short	vendor;
@@ -266,6 +276,7 @@ struct pci_bus {
 	struct list_head children;	/* list of child buses */
 	struct list_head devices;	/* list of devices on this bus */
 	struct pci_dev	*self;		/* bridge device as seen by parent */
+	struct list_head slots;		/* list of slots on this bus */
 	struct resource	*resource[PCI_BUS_NUM_RESOURCES];
 					/* address space routed to this bus */
 
@@ -488,6 +499,10 @@ struct pci_bus *pci_create_bus(struct device *parent, int bus,
 			       struct pci_ops *ops, void *sysdata);
 struct pci_bus *pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev,
 				int busnr);
+struct pci_slot *pci_create_slot(struct pci_bus *parent, int slot_nr,
+				 const char *name);
+void pci_destroy_slot(struct pci_slot *slot);
+void pci_update_slot_number(struct pci_slot *slot, int slot_nr);
 int pci_scan_slot(struct pci_bus *bus, int devfn);
 struct pci_dev *pci_scan_single_device(struct pci_bus *bus, int devfn);
 void pci_device_add(struct pci_dev *dev, struct pci_bus *bus);
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index dbdcd1ad3c6a..a08cd06b541a 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -95,9 +95,6 @@ struct hotplug_slot_attribute {
  * @get_adapter_status: Called to get see if an adapter is present in the slot or not.
  *	If this field is NULL, the value passed in the struct hotplug_slot_info
  *	will be used when this value is requested by a user.
- * @get_address: Called to get pci address of a slot.
- *	If this field is NULL, the value passed in the struct hotplug_slot_info
- *	will be used when this value is requested by a user.
  * @get_max_bus_speed: Called to get the max bus speed for a slot.
  *	If this field is NULL, the value passed in the struct hotplug_slot_info
  *	will be used when this value is requested by a user.
@@ -120,7 +117,6 @@ struct hotplug_slot_ops {
 	int (*get_attention_status)	(struct hotplug_slot *slot, u8 *value);
 	int (*get_latch_status)		(struct hotplug_slot *slot, u8 *value);
 	int (*get_adapter_status)	(struct hotplug_slot *slot, u8 *value);
-	int (*get_address)		(struct hotplug_slot *slot, u32 *value);
 	int (*get_max_bus_speed)	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 	int (*get_cur_bus_speed)	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 };
@@ -140,7 +136,6 @@ struct hotplug_slot_info {
 	u8	attention_status;
 	u8	latch_status;
 	u8	adapter_status;
-	u32	address;
 	enum pci_bus_speed	max_bus_speed;
 	enum pci_bus_speed	cur_bus_speed;
 };
@@ -166,15 +161,14 @@ struct hotplug_slot {
 
 	/* Variables below this are for use only by the hotplug pci core. */
 	struct list_head		slot_list;
-	struct kobject			kobj;
+	struct pci_slot			*pci_slot;
 };
 #define to_hotplug_slot(n) container_of(n, struct hotplug_slot, kobj)
 
-extern int pci_hp_register		(struct hotplug_slot *slot);
-extern int pci_hp_deregister		(struct hotplug_slot *slot);
+extern int pci_hp_register(struct hotplug_slot *, struct pci_bus *, int nr);
+extern int pci_hp_deregister(struct hotplug_slot *slot);
 extern int __must_check pci_hp_change_slot_info	(struct hotplug_slot *slot,
 						 struct hotplug_slot_info *info);
-extern struct kset *pci_hotplug_slots_kset;
 
 /* PCI Setting Record (Type 0) */
 struct hpp_type0 {
-- 
cgit v1.2.3


From 12c03f59c3909159010b87a926f5626d4380d441 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Mon, 9 Jun 2008 16:33:55 -0700
Subject: smc911x: introduce platform data flags

This patch adds a new header file for platform data information
together with code that adds run time bus width and irq flag support.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/smc911x.c   | 17 ++++++++++-
 drivers/net/smc911x.h   | 81 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/smc911x.h | 12 ++++++++
 3 files changed, 109 insertions(+), 1 deletion(-)
 create mode 100644 include/linux/smc911x.h

(limited to 'include/linux')

diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c
index 2ca4db85f938..fc605f276c00 100644
--- a/drivers/net/smc911x.c
+++ b/drivers/net/smc911x.c
@@ -1819,6 +1819,7 @@ static int __init smc911x_probe(struct net_device *dev)
 	int i, retval;
 	unsigned int val, chip_id, revision;
 	const char *version_string;
+	unsigned long irq_flags;
 
 	DBG(SMC_DEBUG_FUNC, "%s: --> %s\n", dev->name, __FUNCTION__);
 
@@ -1985,9 +1986,15 @@ static int __init smc911x_probe(struct net_device *dev)
 	lp->ctl_rfduplx = 1;
 	lp->ctl_rspeed = 100;
 
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+	irq_flags = lp->cfg.irq_flags;
+#else
+	irq_flags = IRQF_SHARED | SMC_IRQ_SENSE;
+#endif
+
 	/* Grab the IRQ */
 	retval = request_irq(dev->irq, &smc911x_interrupt,
-			IRQF_SHARED | SMC_IRQ_SENSE, dev->name, dev);
+			     irq_flags, dev->name, dev);
 	if (retval)
 		goto err_out;
 
@@ -2057,6 +2064,7 @@ err_out:
  */
 static int smc911x_drv_probe(struct platform_device *pdev)
 {
+	struct smc91x_platdata *pd = pdev->dev.platform_data;
 	struct net_device *ndev;
 	struct resource *res;
 	struct smc911x_local *lp;
@@ -2090,6 +2098,13 @@ static int smc911x_drv_probe(struct platform_device *pdev)
 	ndev->irq = platform_get_irq(pdev, 0);
 	lp = netdev_priv(ndev);
 	lp->netdev = ndev;
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+	if (!pd) {
+		ret = -EINVAL;
+		goto release_both;
+	}
+	memcpy(&lp->cfg, pd, sizeof(lp->cfg));
+#endif
 
 	addr = ioremap(res->start, SMC911X_IO_EXTENT);
 	if (!addr) {
diff --git a/drivers/net/smc911x.h b/drivers/net/smc911x.h
index 271a3e8cf683..266232271a72 100644
--- a/drivers/net/smc911x.h
+++ b/drivers/net/smc911x.h
@@ -29,6 +29,7 @@
 #ifndef _SMC911X_H_
 #define _SMC911X_H_
 
+#include <linux/smc911x.h>
 /*
  * Use the DMA feature on PXA chips
  */
@@ -42,6 +43,12 @@
   #define SMC_USE_16BIT		0
   #define SMC_USE_32BIT		1
   #define SMC_IRQ_SENSE		IRQF_TRIGGER_LOW
+#else
+/*
+ * Default configuration
+ */
+
+#define SMC_DYNAMIC_BUS_CONFIG
 #endif
 
 /* store this information for the driver.. */
@@ -92,12 +99,84 @@ struct smc911x_local {
 	struct device *dev;
 #endif
 	void __iomem *base;
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+	struct smc911x_platdata cfg;
+#endif
 };
 
 /*
  * Define the bus width specific IO macros
  */
 
+#ifdef SMC_DYNAMIC_BUS_CONFIG
+static inline unsigned int SMC_inl(struct smc911x_local *lp, int reg)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT)
+		return readl(ioaddr);
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT)
+		return readw(ioaddr) | (readw(ioaddr + 2) << 16);
+
+	BUG();
+}
+
+static inline void SMC_outl(unsigned int value, struct smc911x_local *lp,
+			    int reg)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT) {
+		writel(value, ioaddr);
+		return;
+	}
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT) {
+		writew(value & 0xffff, ioaddr);
+		writew(value >> 16, ioaddr + 2);
+		return;
+	}
+
+	BUG();
+}
+
+static inline void SMC_insl(struct smc911x_local *lp, int reg,
+			      void *addr, unsigned int count)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT) {
+		readsl(ioaddr, addr, count);
+		return;
+	}
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT) {
+		readsw(ioaddr, addr, count * 2);
+		return;
+	}
+
+	BUG();
+}
+
+static inline void SMC_outsl(struct smc911x_local *lp, int reg,
+			     void *addr, unsigned int count)
+{
+	void __iomem *ioaddr = lp->base + reg;
+
+	if (lp->cfg.flags & SMC911X_USE_32BIT) {
+		writesl(ioaddr, addr, count);
+		return;
+	}
+
+	if (lp->cfg.flags & SMC911X_USE_16BIT) {
+		writesw(ioaddr, addr, count * 2);
+		return;
+	}
+
+	BUG();
+}
+#else
 #if	SMC_USE_16BIT
 #define SMC_inl(lp, r)		 ((readw((lp)->base + (r)) & 0xFFFF) + (readw((lp)->base + (r) + 2) << 16))
 #define SMC_outl(v, lp, r) 			 \
@@ -115,6 +194,8 @@ struct smc911x_local {
 #define SMC_outsl(lp, r, p, l)	 writesl((int*)((lp)->base + (r)), p, l)
 
 #endif /* SMC_USE_16BIT */
+#endif /* SMC_DYNAMIC_BUS_CONFIG */
+
 
 #ifdef SMC_USE_PXA_DMA
 #define SMC_USE_DMA
diff --git a/include/linux/smc911x.h b/include/linux/smc911x.h
new file mode 100644
index 000000000000..b58f54c24183
--- /dev/null
+++ b/include/linux/smc911x.h
@@ -0,0 +1,12 @@
+#ifndef __SMC911X_H__
+#define __SMC911X_H__
+
+#define SMC911X_USE_16BIT (1 << 0)
+#define SMC911X_USE_32BIT (1 << 1)
+
+struct smc911x_platdata {
+	unsigned long flags;
+	unsigned long irq_flags; /* IRQF_... */
+};
+
+#endif /* __SMC911X_H__ */
-- 
cgit v1.2.3


From 0b040829952d84bf2a62526f0e24b624e0699447 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 10 Jun 2008 22:46:50 -0700
Subject: net: remove CVS keywords

This patch removes CVS keywords that weren't updated for a long time
from comments.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_bridge.h          | 2 --
 include/linux/if_ppp.h             | 2 --
 include/linux/if_tun.h             | 2 --
 include/linux/ip6_tunnel.h         | 4 ----
 include/linux/ppp-comp.h           | 2 --
 include/linux/ppp_defs.h           | 2 --
 include/linux/sunrpc/auth_gss.h    | 2 --
 include/linux/sunrpc/gss_api.h     | 2 --
 include/linux/sunrpc/svcauth_gss.h | 3 ---
 include/net/inetpeer.h             | 2 --
 include/net/ip6_tunnel.h           | 4 ----
 include/net/ipconfig.h             | 2 --
 include/net/ipv6.h                 | 2 --
 include/net/snmp.h                 | 2 --
 net/bluetooth/bnep/bnep.h          | 4 ----
 net/bluetooth/bnep/core.c          | 4 ----
 net/bluetooth/bnep/netdev.c        | 4 ----
 net/bluetooth/bnep/sock.c          | 4 ----
 net/bluetooth/rfcomm/core.c        | 2 --
 net/bluetooth/rfcomm/sock.c        | 2 --
 net/bluetooth/rfcomm/tty.c         | 2 --
 net/bridge/br.c                    | 2 --
 net/bridge/br_device.c             | 2 --
 net/bridge/br_fdb.c                | 2 --
 net/bridge/br_forward.c            | 2 --
 net/bridge/br_if.c                 | 2 --
 net/bridge/br_input.c              | 2 --
 net/bridge/br_ioctl.c              | 2 --
 net/bridge/br_notify.c             | 2 --
 net/bridge/br_private.h            | 2 --
 net/bridge/br_private_stp.h        | 2 --
 net/bridge/br_stp.c                | 2 --
 net/bridge/br_stp_bpdu.c           | 2 --
 net/bridge/br_stp_if.c             | 2 --
 net/bridge/br_stp_timer.c          | 2 --
 net/core/skbuff.c                  | 2 --
 net/core/sock.c                    | 2 --
 net/ipv4/af_inet.c                 | 2 --
 net/ipv4/arp.c                     | 2 --
 net/ipv4/devinet.c                 | 2 --
 net/ipv4/fib_frontend.c            | 2 --
 net/ipv4/fib_hash.c                | 2 --
 net/ipv4/fib_semantics.c           | 2 --
 net/ipv4/fib_trie.c                | 2 --
 net/ipv4/icmp.c                    | 2 --
 net/ipv4/igmp.c                    | 2 --
 net/ipv4/inet_diag.c               | 2 --
 net/ipv4/inetpeer.c                | 2 --
 net/ipv4/ip_forward.c              | 2 --
 net/ipv4/ip_fragment.c             | 2 --
 net/ipv4/ip_input.c                | 2 --
 net/ipv4/ip_options.c              | 2 --
 net/ipv4/ip_output.c               | 2 --
 net/ipv4/ip_sockglue.c             | 2 --
 net/ipv4/ipconfig.c                | 2 --
 net/ipv4/ipip.c                    | 2 --
 net/ipv4/ipmr.c                    | 2 --
 net/ipv4/ipvs/ip_vs_app.c          | 2 --
 net/ipv4/ipvs/ip_vs_conn.c         | 2 --
 net/ipv4/ipvs/ip_vs_core.c         | 2 --
 net/ipv4/ipvs/ip_vs_ctl.c          | 2 --
 net/ipv4/ipvs/ip_vs_dh.c           | 2 --
 net/ipv4/ipvs/ip_vs_est.c          | 2 --
 net/ipv4/ipvs/ip_vs_ftp.c          | 2 --
 net/ipv4/ipvs/ip_vs_lblc.c         | 2 --
 net/ipv4/ipvs/ip_vs_lblcr.c        | 2 --
 net/ipv4/ipvs/ip_vs_lc.c           | 2 --
 net/ipv4/ipvs/ip_vs_nq.c           | 2 --
 net/ipv4/ipvs/ip_vs_proto.c        | 2 --
 net/ipv4/ipvs/ip_vs_proto_ah.c     | 2 --
 net/ipv4/ipvs/ip_vs_proto_esp.c    | 2 --
 net/ipv4/ipvs/ip_vs_proto_tcp.c    | 2 --
 net/ipv4/ipvs/ip_vs_proto_udp.c    | 2 --
 net/ipv4/ipvs/ip_vs_rr.c           | 2 --
 net/ipv4/ipvs/ip_vs_sched.c        | 2 --
 net/ipv4/ipvs/ip_vs_sed.c          | 2 --
 net/ipv4/ipvs/ip_vs_sh.c           | 2 --
 net/ipv4/ipvs/ip_vs_sync.c         | 2 --
 net/ipv4/ipvs/ip_vs_wlc.c          | 2 --
 net/ipv4/ipvs/ip_vs_wrr.c          | 2 --
 net/ipv4/ipvs/ip_vs_xmit.c         | 2 --
 net/ipv4/proc.c                    | 2 --
 net/ipv4/protocol.c                | 2 --
 net/ipv4/raw.c                     | 2 --
 net/ipv4/route.c                   | 2 --
 net/ipv4/syncookies.c              | 2 --
 net/ipv4/sysctl_net_ipv4.c         | 2 --
 net/ipv4/tcp.c                     | 2 --
 net/ipv4/tcp_diag.c                | 2 --
 net/ipv4/tcp_input.c               | 2 --
 net/ipv4/tcp_ipv4.c                | 2 --
 net/ipv4/tcp_minisocks.c           | 2 --
 net/ipv4/tcp_output.c              | 2 --
 net/ipv4/tcp_timer.c               | 2 --
 net/ipv4/udp.c                     | 2 --
 net/ipv4/udplite.c                 | 2 --
 net/ipv6/addrconf.c                | 2 --
 net/ipv6/af_inet6.c                | 2 --
 net/ipv6/datagram.c                | 2 --
 net/ipv6/exthdrs.c                 | 2 --
 net/ipv6/icmp.c                    | 2 --
 net/ipv6/ip6_fib.c                 | 2 --
 net/ipv6/ip6_input.c               | 2 --
 net/ipv6/ip6_output.c              | 2 --
 net/ipv6/ip6_tunnel.c              | 2 --
 net/ipv6/ipv6_sockglue.c           | 2 --
 net/ipv6/mcast.c                   | 2 --
 net/ipv6/proc.c                    | 2 --
 net/ipv6/protocol.c                | 2 --
 net/ipv6/raw.c                     | 2 --
 net/ipv6/reassembly.c              | 2 --
 net/ipv6/route.c                   | 2 --
 net/ipv6/sit.c                     | 2 --
 net/ipv6/tcp_ipv6.c                | 2 --
 net/ipv6/udp.c                     | 2 --
 net/ipv6/udplite.c                 | 2 --
 net/packet/af_packet.c             | 2 --
 net/sched/sch_htb.c                | 2 --
 net/sunrpc/auth_gss/auth_gss.c     | 2 --
 net/sysctl_net.c                   | 1 -
 net/unix/af_unix.c                 | 2 --
 121 files changed, 254 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_bridge.h b/include/linux/if_bridge.h
index 950e13d09e06..6badb3e2c4e4 100644
--- a/include/linux/if_bridge.h
+++ b/include/linux/if_bridge.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: if_bridge.h,v 1.1 2000/02/18 16:47:01 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/include/linux/if_ppp.h b/include/linux/if_ppp.h
index 0f2f70d4e48c..c3b1f8562709 100644
--- a/include/linux/if_ppp.h
+++ b/include/linux/if_ppp.h
@@ -1,5 +1,3 @@
-/*	$Id: if_ppp.h,v 1.21 2000/03/27 06:03:36 paulus Exp $	*/
-
 /*
  * if_ppp.h - Point-to-Point Protocol definitions.
  *
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 8c71fe2fb1f5..18f31b6187a3 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -11,8 +11,6 @@
  *  but WITHOUT ANY WARRANTY; without even the implied warranty of
  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  *  GNU General Public License for more details.
- *
- *  $Id: if_tun.h,v 1.2 2001/06/01 18:39:47 davem Exp $
  */
 
 #ifndef __IF_TUN_H
diff --git a/include/linux/ip6_tunnel.h b/include/linux/ip6_tunnel.h
index af3f4a70f3df..1e7cc4af40de 100644
--- a/include/linux/ip6_tunnel.h
+++ b/include/linux/ip6_tunnel.h
@@ -1,7 +1,3 @@
-/*
- * $Id$
- */
-
 #ifndef _IP6_TUNNEL_H
 #define _IP6_TUNNEL_H
 
diff --git a/include/linux/ppp-comp.h b/include/linux/ppp-comp.h
index e86a7a5cf355..b8d4ddd22736 100644
--- a/include/linux/ppp-comp.h
+++ b/include/linux/ppp-comp.h
@@ -23,8 +23,6 @@
  * ON AN "AS IS" BASIS, AND THE AUSTRALIAN NATIONAL UNIVERSITY HAS NO
  * OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS,
  * OR MODIFICATIONS.
- *
- * $Id: ppp-comp.h,v 1.6 1997/11/27 06:04:44 paulus Exp $
  */
 
 /*
diff --git a/include/linux/ppp_defs.h b/include/linux/ppp_defs.h
index c6b13ff85028..6e8adc77522c 100644
--- a/include/linux/ppp_defs.h
+++ b/include/linux/ppp_defs.h
@@ -1,5 +1,3 @@
-/*	$Id: ppp_defs.h,v 1.2 1994/09/21 01:31:06 paulus Exp $	*/
-
 /*
  * ppp_defs.h - PPP definitions.
  *
diff --git a/include/linux/sunrpc/auth_gss.h b/include/linux/sunrpc/auth_gss.h
index fec6899bf355..d48d4e605f74 100644
--- a/include/linux/sunrpc/auth_gss.h
+++ b/include/linux/sunrpc/auth_gss.h
@@ -7,8 +7,6 @@
  * Andy Adamson <andros@umich.edu>
  * Bruce Fields <bfields@umich.edu>
  * Copyright (c) 2000 The Regents of the University of Michigan
- *
- * $Id$
  */
 
 #ifndef _LINUX_SUNRPC_AUTH_GSS_H
diff --git a/include/linux/sunrpc/gss_api.h b/include/linux/sunrpc/gss_api.h
index 459c5fc11d51..03f33330ece2 100644
--- a/include/linux/sunrpc/gss_api.h
+++ b/include/linux/sunrpc/gss_api.h
@@ -7,8 +7,6 @@
  * Andy Adamson <andros@umich.edu>
  * Bruce Fields <bfields@umich.edu>
  * Copyright (c) 2000 The Regents of the University of Michigan
- *
- * $Id$
  */
 
 #ifndef _LINUX_SUNRPC_GSS_API_H
diff --git a/include/linux/sunrpc/svcauth_gss.h b/include/linux/sunrpc/svcauth_gss.h
index 417a1def56db..c9165d9771a8 100644
--- a/include/linux/sunrpc/svcauth_gss.h
+++ b/include/linux/sunrpc/svcauth_gss.h
@@ -3,9 +3,6 @@
  *
  * Bruce Fields <bfields@umich.edu>
  * Copyright (c) 2002 The Regents of the Unviersity of Michigan
- *
- * $Id$
- *
  */
 
 #ifndef _LINUX_SUNRPC_SVCAUTH_GSS_H
diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h
index ad8404b56113..15e1f8fe4c1f 100644
--- a/include/net/inetpeer.h
+++ b/include/net/inetpeer.h
@@ -1,8 +1,6 @@
 /*
  *		INETPEER - A storage for permanent information about peers
  *
- *  Version:	$Id: inetpeer.h,v 1.2 2002/01/12 07:54:56 davem Exp $
- *
  *  Authors:	Andrey V. Savochkin <saw@msu.ru>
  */
 
diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h
index 3780592ebe8e..83b4e008b16d 100644
--- a/include/net/ip6_tunnel.h
+++ b/include/net/ip6_tunnel.h
@@ -1,7 +1,3 @@
-/*
- * $Id$
- */
-
 #ifndef _NET_IP6_TUNNEL_H
 #define _NET_IP6_TUNNEL_H
 
diff --git a/include/net/ipconfig.h b/include/net/ipconfig.h
index 3924d7d2cb11..c74cc1bd5a02 100644
--- a/include/net/ipconfig.h
+++ b/include/net/ipconfig.h
@@ -1,6 +1,4 @@
 /*
- *  $Id: ipconfig.h,v 1.4 2001/04/30 04:51:46 davem Exp $
- *
  *  Copyright (C) 1997 Martin Mares
  *
  *  Automatic IP Layer Configuration
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index e0a612bc9c4e..7f7db8d57934 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: ipv6.h,v 1.1 2002/05/20 15:13:07 jgrimm Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/include/net/snmp.h b/include/net/snmp.h
index ce2f48507510..57c93628695f 100644
--- a/include/net/snmp.h
+++ b/include/net/snmp.h
@@ -14,8 +14,6 @@
  *		as published by the Free Software Foundation; either version
  *		2 of the License, or (at your option) any later version.
  *
- *		$Id: snmp.h,v 1.19 2001/06/14 13:40:46 davem Exp $
- *
  */
  
 #ifndef _SNMP_H
diff --git a/net/bluetooth/bnep/bnep.h b/net/bluetooth/bnep/bnep.h
index e69244dd8de8..b69bf4e7c48b 100644
--- a/net/bluetooth/bnep/bnep.h
+++ b/net/bluetooth/bnep/bnep.h
@@ -16,10 +16,6 @@
   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
 */
 
-/*
- * $Id: bnep.h,v 1.5 2002/08/04 21:23:58 maxk Exp $
- */
-
 #ifndef _BNEP_H
 #define _BNEP_H
 
diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c
index f85d94643aaf..1d98a1b80da7 100644
--- a/net/bluetooth/bnep/core.c
+++ b/net/bluetooth/bnep/core.c
@@ -25,10 +25,6 @@
    SOFTWARE IS DISCLAIMED.
 */
 
-/*
- * $Id: core.c,v 1.20 2002/08/04 21:23:58 maxk Exp $
- */
-
 #include <linux/module.h>
 
 #include <linux/kernel.h>
diff --git a/net/bluetooth/bnep/netdev.c b/net/bluetooth/bnep/netdev.c
index 95e3837e4312..d9fa0ab2c87f 100644
--- a/net/bluetooth/bnep/netdev.c
+++ b/net/bluetooth/bnep/netdev.c
@@ -25,10 +25,6 @@
    SOFTWARE IS DISCLAIMED.
 */
 
-/*
- * $Id: netdev.c,v 1.8 2002/08/04 21:23:58 maxk Exp $
- */
-
 #include <linux/module.h>
 
 #include <linux/socket.h>
diff --git a/net/bluetooth/bnep/sock.c b/net/bluetooth/bnep/sock.c
index 201e5b1ce473..8ffb57f2303a 100644
--- a/net/bluetooth/bnep/sock.c
+++ b/net/bluetooth/bnep/sock.c
@@ -24,10 +24,6 @@
    SOFTWARE IS DISCLAIMED.
 */
 
-/*
- * $Id: sock.c,v 1.4 2002/08/04 21:23:58 maxk Exp $
- */
-
 #include <linux/module.h>
 
 #include <linux/types.h>
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index 0c2c93735e93..b4fb84e398e5 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -23,8 +23,6 @@
 
 /*
  * Bluetooth RFCOMM core.
- *
- * $Id: core.c,v 1.42 2002/10/01 23:26:25 maxk Exp $
  */
 
 #include <linux/module.h>
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index 5083adcbfae5..c9054487670a 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -23,8 +23,6 @@
 
 /*
  * RFCOMM sockets.
- *
- * $Id: sock.c,v 1.24 2002/10/03 01:00:34 maxk Exp $
  */
 
 #include <linux/module.h>
diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c
index c9191871c1e0..be84f4fc1477 100644
--- a/net/bluetooth/rfcomm/tty.c
+++ b/net/bluetooth/rfcomm/tty.c
@@ -23,8 +23,6 @@
 
 /*
  * RFCOMM TTY.
- *
- * $Id: tty.c,v 1.24 2002/10/03 01:54:38 holtmann Exp $
  */
 
 #include <linux/module.h>
diff --git a/net/bridge/br.c b/net/bridge/br.c
index 8f3c58e5f7a5..cede010f4ddd 100644
--- a/net/bridge/br.c
+++ b/net/bridge/br.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br.c,v 1.47 2001/12/24 00:56:41 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 626c7795ae30..a6ffc6c2a69f 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_device.c,v 1.6 2001/12/24 00:59:55 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
index 72c5976a5ce3..4de74cdd091d 100644
--- a/net/bridge/br_fdb.c
+++ b/net/bridge/br_fdb.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_fdb.c,v 1.6 2002/01/17 00:57:07 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index a4711674b3df..512645727f51 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_forward.c,v 1.4 2001/08/14 22:05:57 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index c2397f503b0f..143c954681b8 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_if.c,v 1.7 2001/12/24 00:59:55 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index fa0f5711a996..0145e9416714 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_input.c,v 1.10 2001/12/24 04:50:20 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_ioctl.c b/net/bridge/br_ioctl.c
index 0655a5f07f58..eeee218eed80 100644
--- a/net/bridge/br_ioctl.c
+++ b/net/bridge/br_ioctl.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_ioctl.c,v 1.4 2000/11/08 05:16:40 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_notify.c b/net/bridge/br_notify.c
index 00644a544e3c..88d8ec7b3142 100644
--- a/net/bridge/br_notify.c
+++ b/net/bridge/br_notify.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_notify.c,v 1.2 2000/02/21 15:51:34 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 0243cb489edd..83ff5861c2d2 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_private.h,v 1.7 2001/12/24 00:59:55 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_private_stp.h b/net/bridge/br_private_stp.h
index e29f01ac1adf..8b650f7fbfa0 100644
--- a/net/bridge/br_private_stp.h
+++ b/net/bridge/br_private_stp.h
@@ -4,8 +4,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_private_stp.h,v 1.3 2001/02/05 06:03:47 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c
index e38034aa56f5..284d1b2fa1ff 100644
--- a/net/bridge/br_stp.c
+++ b/net/bridge/br_stp.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp.c,v 1.4 2000/06/19 10:13:35 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c
index ddeb6e5d45d6..9dc2de656965 100644
--- a/net/bridge/br_stp_bpdu.c
+++ b/net/bridge/br_stp_bpdu.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp_bpdu.c,v 1.3 2001/11/10 02:35:25 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c
index 1a430eccec9b..1a4e5c37a0cf 100644
--- a/net/bridge/br_stp_if.c
+++ b/net/bridge/br_stp_if.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp_if.c,v 1.4 2001/04/14 21:14:39 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/bridge/br_stp_timer.c b/net/bridge/br_stp_timer.c
index 77f5255e6915..772a140bfdf0 100644
--- a/net/bridge/br_stp_timer.c
+++ b/net/bridge/br_stp_timer.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Lennert Buytenhek		<buytenh@gnu.org>
  *
- *	$Id: br_stp_timer.c,v 1.3 2000/05/05 02:17:17 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 1e556d312117..3e18f8525e82 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4,8 +4,6 @@
  *	Authors:	Alan Cox <iiitac@pyr.swan.ac.uk>
  *			Florian La Roche <rzsfl@rz.uni-sb.de>
  *
- *	Version:	$Id: skbuff.c,v 1.90 2001/11/07 05:56:19 davem Exp $
- *
  *	Fixes:
  *		Alan Cox	:	Fixed the worst of the load
  *					balancer bugs.
diff --git a/net/core/sock.c b/net/core/sock.c
index 88094cb09c06..3879bf65897e 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -7,8 +7,6 @@
  *		handler for protocols to use and generic option handler.
  *
  *
- * Version:	$Id: sock.c,v 1.117 2002/02/01 22:01:03 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Florian La Roche, <flla@stud.uni-sb.de>
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 24eca23c2db3..42bd24b64b57 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -5,8 +5,6 @@
  *
  *		PF_INET protocol family socket handler.
  *
- * Version:	$Id: af_inet.c,v 1.137 2002/02/01 22:01:03 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Florian La Roche, <flla@stud.uni-sb.de>
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 9b539fa9fe18..20c515a1be28 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -1,6 +1,4 @@
 /* linux/net/ipv4/arp.c
- *
- * Version:	$Id: arp.c,v 1.99 2001/08/30 22:55:42 davem Exp $
  *
  * Copyright (C) 1994 by Florian  La Roche
  *
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 61011e1d580e..f8c0b0aea93a 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1,8 +1,6 @@
 /*
  *	NET3	IP device support routines.
  *
- *	Version: $Id: devinet.c,v 1.44 2001/10/31 21:55:54 davem Exp $
- *
  *		This program is free software; you can redistribute it and/or
  *		modify it under the terms of the GNU General Public License
  *		as published by the Free Software Foundation; either version
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 0b2ac6a3d903..5ad01d63f83b 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -5,8 +5,6 @@
  *
  *		IPv4 Forwarding Information Base: FIB frontend.
  *
- * Version:	$Id: fib_frontend.c,v 1.26 2001/10/31 21:55:54 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c
index 2e2fc3376ac9..eeec4bf982b8 100644
--- a/net/ipv4/fib_hash.c
+++ b/net/ipv4/fib_hash.c
@@ -5,8 +5,6 @@
  *
  *		IPv4 FIB: lookup engine and maintenance routines.
  *
- * Version:	$Id: fib_hash.c,v 1.13 2001/10/31 21:55:54 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 3b83c34019fc..9335eba683c3 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -5,8 +5,6 @@
  *
  *		IPv4 Forwarding Information Base: semantics.
  *
- * Version:	$Id: fib_semantics.c,v 1.19 2002/01/12 07:54:56 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 4b02d14e7ab9..394db9c941a1 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -22,8 +22,6 @@
  * IP-address lookup using LC-tries. Stefan Nilsson and Gunnar Karlsson
  * IEEE Journal on Selected Areas in Communications, 17(6):1083-1092, June 1999
  *
- * Version:	$Id: fib_trie.c,v 1.3 2005/06/08 14:20:01 robert Exp $
- *
  *
  * Code from fib_hash has been reused which includes the following header:
  *
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 87397351ddac..aa7cf46853b7 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -3,8 +3,6 @@
  *
  *		Alan Cox, <alan@redhat.com>
  *
- *	Version: $Id: icmp.c,v 1.85 2002/02/01 22:01:03 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 2769dc4a4c84..68e84a933e90 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -8,8 +8,6 @@
  *	the older version didn't come out right using gcc 2.5.8, the newer one
  *	seems to fall out with gcc 2.6.2.
  *
- *	Version: $Id: igmp.c,v 1.47 2002/02/01 22:01:03 davem Exp $
- *
  *	Authors:
  *		Alan Cox <Alan.Cox@linux.org>
  *
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index da97695e7096..c10036e7a463 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -1,8 +1,6 @@
 /*
  * inet_diag.c	Module for monitoring INET transport protocols sockets.
  *
- * Version:	$Id: inet_diag.c,v 1.3 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c
index af995198f643..a456ceeac3f2 100644
--- a/net/ipv4/inetpeer.c
+++ b/net/ipv4/inetpeer.c
@@ -3,8 +3,6 @@
  *
  *  This source is covered by the GNU GPL, the same as all kernel sources.
  *
- *  Version:	$Id: inetpeer.c,v 1.7 2001/09/20 21:22:50 davem Exp $
- *
  *  Authors:	Andrey V. Savochkin <saw@msu.ru>
  */
 
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 4813c39b438b..37d36a3f33cd 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -5,8 +5,6 @@
  *
  *		The IP forwarding functionality.
  *
- * Version:	$Id: ip_forward.c,v 1.48 2000/12/13 18:31:48 davem Exp $
- *
  * Authors:	see ip.c
  *
  * Fixes:
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index be1cb89a8d5a..91e321407313 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -5,8 +5,6 @@
  *
  *		The IP fragmentation functionality.
  *
- * Version:	$Id: ip_fragment.c,v 1.59 2002/01/12 07:54:56 davem Exp $
- *
  * Authors:	Fred N. van Kempen <waltje@uWalt.NL.Mugnet.ORG>
  *		Alan Cox <Alan.Cox@linux.org>
  *
diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index ff77a4a7f9ec..7c26428ea67b 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -5,8 +5,6 @@
  *
  *		The Internet Protocol (IP) module.
  *
- * Version:	$Id: ip_input.c,v 1.55 2002/01/12 07:39:45 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Donald Becker, <becker@super.org>
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index 33126ad2cfdc..be3f18a7a40e 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -5,8 +5,6 @@
  *
  *		The options processing module for ip.c
  *
- * Version:	$Id: ip_options.c,v 1.21 2001/09/01 00:31:50 davem Exp $
- *
  * Authors:	A.N.Kuznetsov
  *
  */
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index e527628f56cf..f1278eecf56d 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -5,8 +5,6 @@
  *
  *		The Internet Protocol (IP) output module.
  *
- * Version:	$Id: ip_output.c,v 1.100 2002/02/01 22:01:03 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Donald Becker, <becker@super.org>
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index e0514e82308e..105d92a039b9 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -5,8 +5,6 @@
  *
  *		The IP to API glue.
  *
- * Version:	$Id: ip_sockglue.c,v 1.62 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	see ip.c
  *
  * Fixes:
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index ed45037ce9be..b88aa9afa42e 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -1,6 +1,4 @@
 /*
- *  $Id: ipconfig.c,v 1.46 2002/02/01 22:01:04 davem Exp $
- *
  *  Automatic Configuration of IP -- use DHCP, BOOTP, RARP, or
  *  user-supplied information to configure own IP address and routes.
  *
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 86d8836551b9..4c6d2caf9203 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -1,8 +1,6 @@
 /*
  *	Linux NET3:	IP/IP protocol decoder.
  *
- *	Version: $Id: ipip.c,v 1.50 2001/10/02 02:22:36 davem Exp $
- *
  *	Authors:
  *		Sam Lantinga (slouken@cs.ucdavis.edu)  02/01/95
  *
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index a34da4977c73..300ab0c2919e 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -9,8 +9,6 @@
  *	as published by the Free Software Foundation; either version
  *	2 of the License, or (at your option) any later version.
  *
- *	Version: $Id: ipmr.c,v 1.65 2001/10/31 21:55:54 davem Exp $
- *
  *	Fixes:
  *	Michael Chastain	:	Incorrect size of copying.
  *	Alan Cox		:	Added the cache manager code
diff --git a/net/ipv4/ipvs/ip_vs_app.c b/net/ipv4/ipvs/ip_vs_app.c
index 535abe0c45e7..1f1897a1a702 100644
--- a/net/ipv4/ipvs/ip_vs_app.c
+++ b/net/ipv4/ipvs/ip_vs_app.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_app.c: Application module support for IPVS
  *
- * Version:     $Id: ip_vs_app.c,v 1.17 2003/03/22 06:31:21 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c
index 65f1ba112752..f8bdae47a77f 100644
--- a/net/ipv4/ipvs/ip_vs_conn.c
+++ b/net/ipv4/ipvs/ip_vs_conn.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_conn.c,v 1.31 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *              Julian Anastasov <ja@ssi.bg>
diff --git a/net/ipv4/ipvs/ip_vs_core.c b/net/ipv4/ipvs/ip_vs_core.c
index 963981a9d501..bcf6276ba4b2 100644
--- a/net/ipv4/ipvs/ip_vs_core.c
+++ b/net/ipv4/ipvs/ip_vs_core.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_core.c,v 1.34 2003/05/10 03:05:23 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *              Julian Anastasov <ja@ssi.bg>
diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c
index 94c5767c8e01..9a5ace0b4dd6 100644
--- a/net/ipv4/ipvs/ip_vs_ctl.c
+++ b/net/ipv4/ipvs/ip_vs_ctl.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_ctl.c,v 1.36 2003/06/08 09:31:19 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *              Julian Anastasov <ja@ssi.bg>
diff --git a/net/ipv4/ipvs/ip_vs_dh.c b/net/ipv4/ipvs/ip_vs_dh.c
index dcf5d46aaa5e..8afc1503ed20 100644
--- a/net/ipv4/ipvs/ip_vs_dh.c
+++ b/net/ipv4/ipvs/ip_vs_dh.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Destination Hashing scheduling module
  *
- * Version:     $Id: ip_vs_dh.c,v 1.5 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              Inspired by the consistent hashing scheduler patch from
diff --git a/net/ipv4/ipvs/ip_vs_est.c b/net/ipv4/ipvs/ip_vs_est.c
index dfa0d713c801..bc04eedd6dbb 100644
--- a/net/ipv4/ipvs/ip_vs_est.c
+++ b/net/ipv4/ipvs/ip_vs_est.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_est.c: simple rate estimator for IPVS
  *
- * Version:     $Id: ip_vs_est.c,v 1.4 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_ftp.c b/net/ipv4/ipvs/ip_vs_ftp.c
index 59aa166b7678..c1c758e4f733 100644
--- a/net/ipv4/ipvs/ip_vs_ftp.c
+++ b/net/ipv4/ipvs/ip_vs_ftp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_ftp.c: IPVS ftp application module
  *
- * Version:	$Id: ip_vs_ftp.c,v 1.13 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:	Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  * Changes:
diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c
index 3888642706ad..0efa3db4b180 100644
--- a/net/ipv4/ipvs/ip_vs_lblc.c
+++ b/net/ipv4/ipvs/ip_vs_lblc.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Locality-Based Least-Connection scheduling module
  *
- * Version:     $Id: ip_vs_lblc.c,v 1.10 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c
index daa260eb21cf..8e3bbeb45138 100644
--- a/net/ipv4/ipvs/ip_vs_lblcr.c
+++ b/net/ipv4/ipvs/ip_vs_lblcr.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Locality-Based Least-Connection with Replication scheduler
  *
- * Version:     $Id: ip_vs_lblcr.c,v 1.11 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_lc.c b/net/ipv4/ipvs/ip_vs_lc.c
index d88fef90a641..ac9f08e065d5 100644
--- a/net/ipv4/ipvs/ip_vs_lc.c
+++ b/net/ipv4/ipvs/ip_vs_lc.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Least-Connection Scheduling module
  *
- * Version:     $Id: ip_vs_lc.c,v 1.10 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_nq.c b/net/ipv4/ipvs/ip_vs_nq.c
index bc2a9e5f2a7b..a46bf258d420 100644
--- a/net/ipv4/ipvs/ip_vs_nq.c
+++ b/net/ipv4/ipvs/ip_vs_nq.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Never Queue scheduling module
  *
- * Version:     $Id: ip_vs_nq.c,v 1.2 2003/06/08 09:31:19 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_proto.c b/net/ipv4/ipvs/ip_vs_proto.c
index 4b1c16cbb16b..876714f23d65 100644
--- a/net/ipv4/ipvs/ip_vs_proto.c
+++ b/net/ipv4/ipvs/ip_vs_proto.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto.c: transport protocol load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto.c,v 1.2 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_ah.c b/net/ipv4/ipvs/ip_vs_proto_ah.c
index 4bf835e1d86d..73e0ea87c1f5 100644
--- a/net/ipv4/ipvs/ip_vs_proto_ah.c
+++ b/net/ipv4/ipvs/ip_vs_proto_ah.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_ah.c:	AH IPSec load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_ah.c,v 1.1 2003/07/04 15:04:37 wensong Exp $
- *
  * Authors:	Julian Anastasov <ja@ssi.bg>, February 2002
  *		Wensong Zhang <wensong@linuxvirtualserver.org>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_esp.c b/net/ipv4/ipvs/ip_vs_proto_esp.c
index db6a6b7b1a0b..21d70c8ffa54 100644
--- a/net/ipv4/ipvs/ip_vs_proto_esp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_esp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_esp.c:	ESP IPSec load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_esp.c,v 1.1 2003/07/04 15:04:37 wensong Exp $
- *
  * Authors:	Julian Anastasov <ja@ssi.bg>, February 2002
  *		Wensong Zhang <wensong@linuxvirtualserver.org>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c
index b83dc14b0a4d..d0ea467986a0 100644
--- a/net/ipv4/ipvs/ip_vs_proto_tcp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_tcp.c:	TCP load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_tcp.c,v 1.3 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c
index 75771cb3cd6f..c6be5d56823f 100644
--- a/net/ipv4/ipvs/ip_vs_proto_udp.c
+++ b/net/ipv4/ipvs/ip_vs_proto_udp.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_proto_udp.c:	UDP load balancing support for IPVS
  *
- * Version:     $Id: ip_vs_proto_udp.c,v 1.3 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/ipvs/ip_vs_rr.c b/net/ipv4/ipvs/ip_vs_rr.c
index 433f8a947924..c8db12d39e61 100644
--- a/net/ipv4/ipvs/ip_vs_rr.c
+++ b/net/ipv4/ipvs/ip_vs_rr.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Round-Robin Scheduling module
  *
- * Version:     $Id: ip_vs_rr.c,v 1.9 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *
diff --git a/net/ipv4/ipvs/ip_vs_sched.c b/net/ipv4/ipvs/ip_vs_sched.c
index 121a32b1b756..b64767309855 100644
--- a/net/ipv4/ipvs/ip_vs_sched.c
+++ b/net/ipv4/ipvs/ip_vs_sched.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_sched.c,v 1.13 2003/05/10 03:05:23 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *
diff --git a/net/ipv4/ipvs/ip_vs_sed.c b/net/ipv4/ipvs/ip_vs_sed.c
index dd7c128f9db3..2a7d31358181 100644
--- a/net/ipv4/ipvs/ip_vs_sed.c
+++ b/net/ipv4/ipvs/ip_vs_sed.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Shortest Expected Delay scheduling module
  *
- * Version:     $Id: ip_vs_sed.c,v 1.1 2003/05/10 03:06:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_sh.c b/net/ipv4/ipvs/ip_vs_sh.c
index 1b25b00ef1e1..b8fdfac65001 100644
--- a/net/ipv4/ipvs/ip_vs_sh.c
+++ b/net/ipv4/ipvs/ip_vs_sh.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Source Hashing scheduling module
  *
- * Version:     $Id: ip_vs_sh.c,v 1.5 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@gnuchina.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c
index eff54efe0351..2d4a86f73325 100644
--- a/net/ipv4/ipvs/ip_vs_sync.c
+++ b/net/ipv4/ipvs/ip_vs_sync.c
@@ -5,8 +5,6 @@
  *              high-performance and highly available server based on a
  *              cluster of servers.
  *
- * Version:     $Id: ip_vs_sync.c,v 1.13 2003/06/08 09:31:19 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  * ip_vs_sync:  sync connection info from master load balancer to backups
diff --git a/net/ipv4/ipvs/ip_vs_wlc.c b/net/ipv4/ipvs/ip_vs_wlc.c
index 8a9d913261d8..772c3cb4eca1 100644
--- a/net/ipv4/ipvs/ip_vs_wlc.c
+++ b/net/ipv4/ipvs/ip_vs_wlc.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Weighted Least-Connection Scheduling module
  *
- * Version:     $Id: ip_vs_wlc.c,v 1.13 2003/04/18 09:03:16 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Peter Kese <peter.kese@ijs.si>
  *
diff --git a/net/ipv4/ipvs/ip_vs_wrr.c b/net/ipv4/ipvs/ip_vs_wrr.c
index 85c680add6df..1d6932d7dc97 100644
--- a/net/ipv4/ipvs/ip_vs_wrr.c
+++ b/net/ipv4/ipvs/ip_vs_wrr.c
@@ -1,8 +1,6 @@
 /*
  * IPVS:        Weighted Round-Robin Scheduling module
  *
- * Version:     $Id: ip_vs_wrr.c,v 1.12 2002/09/15 08:14:08 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *
  *              This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/ipvs/ip_vs_xmit.c b/net/ipv4/ipvs/ip_vs_xmit.c
index f63006caea03..9892d4aca42e 100644
--- a/net/ipv4/ipvs/ip_vs_xmit.c
+++ b/net/ipv4/ipvs/ip_vs_xmit.c
@@ -1,8 +1,6 @@
 /*
  * ip_vs_xmit.c: various packet transmitters for IPVS
  *
- * Version:     $Id: ip_vs_xmit.c,v 1.2 2002/11/30 01:50:35 wensong Exp $
- *
  * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
  *              Julian Anastasov <ja@ssi.bg>
  *
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 552169b41b16..eb5cee279c5f 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -7,8 +7,6 @@
  *		PROC file system.  It is mainly used for debugging and
  *		statistics.
  *
- * Version:	$Id: proc.c,v 1.45 2001/05/16 16:45:35 davem Exp $
- *
  * Authors:	Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Gerald J. Heim, <heim@peanuts.informatik.uni-tuebingen.de>
  *		Fred Baumgarten, <dc6iq@insu1.etec.uni-karlsruhe.de>
diff --git a/net/ipv4/protocol.c b/net/ipv4/protocol.c
index 971ab9356e51..ea50da0649fd 100644
--- a/net/ipv4/protocol.c
+++ b/net/ipv4/protocol.c
@@ -5,8 +5,6 @@
  *
  *		INET protocol dispatch tables.
  *
- * Version:	$Id: protocol.c,v 1.14 2001/05/18 02:25:49 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index e7e091d365ff..1d0c97c8712d 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -5,8 +5,6 @@
  *
  *		RAW - implementation of IP "raw" sockets.
  *
- * Version:	$Id: raw.c,v 1.64 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 96be336064fb..fe3a02237286 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -5,8 +5,6 @@
  *
  *		ROUTE - implementation of the IP router.
  *
- * Version:	$Id: route.c,v 1.103 2002/01/12 07:44:09 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Alan Cox, <gw4pts@gw4pts.ampr.org>
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 73ba98921d64..6317d3c8dc0d 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -8,8 +8,6 @@
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
  *      2 of the License, or (at your option) any later version.
- *
- *  $Id: syncookies.c,v 1.18 2002/02/01 22:01:04 davem Exp $
  */
 
 #include <linux/tcp.h>
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index c437f804ee38..901607003205 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -1,8 +1,6 @@
 /*
  * sysctl_net_ipv4.c: sysctl interface to net IPV4 subsystem.
  *
- * $Id: sysctl_net_ipv4.c,v 1.50 2001/10/20 00:00:11 davem Exp $
- *
  * Begun April 1, 1996, Mike Shaver.
  * Added /proc/sys/net/ipv4 directory entry (empty =) ). [MS]
  */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ab66683b8043..ad66b09e0bcd 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp.c,v 1.216 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c
index 2fbcc7d1b1a0..838d491dfda7 100644
--- a/net/ipv4/tcp_diag.c
+++ b/net/ipv4/tcp_diag.c
@@ -1,8 +1,6 @@
 /*
  * tcp_diag.c	Module for monitoring TCP transport protocols sockets.
  *
- * Version:	$Id: tcp_diag.c,v 1.3 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index eba873e9b560..b68c3c7d906b 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_input.c,v 1.243 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index cd601a866c2f..f2926ae1de57 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_ipv4.c,v 1.240 2002/02/01 22:01:04 davem Exp $
- *
  *		IPv4 specific functions
  *
  *
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 019c8c16e5cc..1276cab85e3e 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_minisocks.c,v 1.15 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index ad993ecb4810..b171ac65ccab 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_output.c,v 1.146 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 4de68cf5f2aa..e77e7ae0bf2c 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -5,8 +5,6 @@
  *
  *		Implementation of the Transmission Control Protocol(TCP).
  *
- * Version:	$Id: tcp_timer.c,v 1.88 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Mark Evans, <evansmp@uhura.aston.ac.uk>
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 56fcda3694ba..355e6d62d483 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -5,8 +5,6 @@
  *
  *		The User Datagram Protocol (UDP).
  *
- * Version:	$Id: udp.c,v 1.102 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Arnt Gulbrandsen, <agulbra@nvg.unit.no>
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 72ce26b6c4d3..4ad16b6d5138 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -1,8 +1,6 @@
 /*
  *  UDPLITE     An implementation of the UDP-Lite protocol (RFC 3828).
  *
- *  Version:    $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $
- *
  *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
  *
  *  Changes:
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 147588f4c7c0..deb38bf03376 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -6,8 +6,6 @@
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
  *
- *	$Id: addrconf.c,v 1.69 2001/10/31 21:55:54 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e84b3fd17fb4..350457c761e6 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -7,8 +7,6 @@
  *
  *	Adapted from linux/net/ipv4/af_inet.c
  *
- *	$Id: af_inet6.c,v 1.66 2002/02/01 22:01:04 davem Exp $
- *
  * 	Fixes:
  *	piggy, Karl Knutson	:	Socket protocol table
  * 	Hideaki YOSHIFUJI	:	sin6_scope_id support
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index b9c2de84a8a2..8cdb6b65ee96 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: datagram.c,v 1.24 2002/02/01 22:01:04 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 3cd1c993d52b..602ea826f0a5 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -7,8 +7,6 @@
  *	Andi Kleen		<ak@muc.de>
  *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
  *
- *	$Id: exthdrs.c,v 1.13 2001/06/19 15:58:56 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index d42dd16d3487..399d41f65437 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: icmp.c,v 1.38 2002/02/08 03:57:19 davem Exp $
- *
  *	Based on net/ipv4/icmp.c
  *
  *	RFC 1885
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 1ee4fa17c129..4de2b9efcacb 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: ip6_fib.c,v 1.25 2001/10/31 21:55:55 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 4e5c8615832c..f77a6011c302 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -6,8 +6,6 @@
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *	Ian P. Morris		<I.P.Morris@soton.ac.uk>
  *
- *	$Id: ip6_input.c,v 1.19 2000/12/13 18:31:50 davem Exp $
- *
  *	Based in linux/net/ipv4/ip_input.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 48cdce9c696c..40a2813a63d1 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: ip6_output.c,v 1.34 2002/02/01 22:01:04 davem Exp $
- *
  *	Based on linux/net/ipv4/ip_output.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 37814810ac49..17c7b098cdb0 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -6,8 +6,6 @@
  *	Ville Nuorvala		<vnuorval@tcs.hut.fi>
  *	Yasuyuki Kozakai	<kozakai@linux-ipv6.org>
  *
- *	$Id$
- *
  *      Based on:
  *      linux/net/ipv6/sit.c and linux/net/ipv4/ipip.c
  *
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 26b83e512a09..237ebbb9383b 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -7,8 +7,6 @@
  *
  *	Based on linux/net/ipv4/ip_sockglue.c
  *
- *	$Id: ipv6_sockglue.c,v 1.41 2002/02/01 22:01:04 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index fd632dd7f98d..fbb2d12c41bc 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: mcast.c,v 1.40 2002/02/08 03:57:19 davem Exp $
- *
  *	Based on linux/ipv4/igmp.c and linux/ipv4/ip_sockglue.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index df0736a4cafa..cbc7e514d3ec 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -7,8 +7,6 @@
  *		PROC file system.  This is very similar to the IPv4 version,
  *		except it reports the sockets in the INET6 address family.
  *
- * Version:	$Id: proc.c,v 1.17 2002/02/01 22:01:04 davem Exp $
- *
  * Authors:	David S. Miller (davem@caip.rutgers.edu)
  * 		YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
  *
diff --git a/net/ipv6/protocol.c b/net/ipv6/protocol.c
index f929f47b925e..9ab789159913 100644
--- a/net/ipv6/protocol.c
+++ b/net/ipv6/protocol.c
@@ -5,8 +5,6 @@
  *
  *		PF_INET6 protocol dispatch tables.
  *
- * Version:	$Id: protocol.c,v 1.10 2001/05/18 02:25:49 davem Exp $
- *
  * Authors:	Pedro Roque	<roque@di.fc.ul.pt>
  *
  *		This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 8fee9a15b2d3..e03c1898ab2e 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -7,8 +7,6 @@
  *
  *	Adapted from linux/net/ipv4/raw.c
  *
- *	$Id: raw.c,v 1.51 2002/02/01 22:01:04 davem Exp $
- *
  *	Fixes:
  *	Hideaki YOSHIFUJI	:	sin6_scope_id support
  *	YOSHIFUJI,H.@USAGI	:	raw checksum (RFC2292(bis) compliance)
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 9391a6949b96..13509f906d89 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: reassembly.c,v 1.26 2001/03/07 22:00:57 davem Exp $
- *
  *	Based on: net/ipv4/ip_fragment.c
  *
  *	This program is free software; you can redistribute it and/or
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 220cffe9e63b..edae81319b51 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: route.c,v 1.56 2001/10/31 21:55:55 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 6b8f0583b637..b0c5080420a8 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -6,8 +6,6 @@
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
  *
- *	$Id: sit.c,v 1.53 2001/09/25 05:09:53 davem Exp $
- *
  *	This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
  *      as published by the Free Software Foundation; either version
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 715965f0fac0..155499197fc5 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -5,8 +5,6 @@
  *	Authors:
  *	Pedro Roque		<roque@di.fc.ul.pt>
  *
- *	$Id: tcp_ipv6.c,v 1.144 2002/02/01 22:01:04 davem Exp $
- *
  *	Based on:
  *	linux/net/ipv4/tcp.c
  *	linux/net/ipv4/tcp_input.c
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index dd309626ae9a..e0693fffc9bd 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -7,8 +7,6 @@
  *
  *	Based on linux/ipv4/udp.c
  *
- *	$Id: udp.c,v 1.65 2002/02/01 22:01:04 davem Exp $
- *
  *	Fixes:
  *	Hideaki YOSHIFUJI	:	sin6_scope_id support
  *	YOSHIFUJI Hideaki @USAGI and:	Support IPV6_V6ONLY socket option, which
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 491efd00a866..f6cdcb348e05 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -2,8 +2,6 @@
  *  UDPLITEv6   An implementation of the UDP-Lite protocol over IPv6.
  *              See also net/ipv4/udplite.c
  *
- *  Version:    $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $
- *
  *  Authors:    Gerrit Renker       <gerrit@erg.abdn.ac.uk>
  *
  *  Changes:
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 2cee87da4441..beca6402f1cf 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -5,8 +5,6 @@
  *
  *		PACKET - implements raw packet sockets.
  *
- * Version:	$Id: af_packet.c,v 1.61 2002/02/08 03:57:19 davem Exp $
- *
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Alan Cox, <gw4pts@gw4pts.ampr.org>
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 5bc1ed490180..213071859030 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -24,8 +24,6 @@
  *		Jiri Fojtasek
  *			fixed requeue routine
  *		and many others. thanks.
- *
- * $Id: sch_htb.c,v 1.25 2003/12/07 11:08:25 devik Exp devik $
  */
 #include <linux/module.h>
 #include <linux/types.h>
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index cc12d5f5d5da..019d4b4478c9 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -33,8 +33,6 @@
  *  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
  *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
  *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * $Id$
  */
 
 
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index d8e79162724c..007c1a6708ee 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -4,7 +4,6 @@
  * Begun April 1, 1996, Mike Shaver.
  * Added /proc/sys/net directories for each protocol family. [MS]
  *
- * $Log: sysctl_net.c,v $
  * Revision 1.2  1996/05/08  20:24:40  shaver
  * Added bits for NET_BRIDGE and the NET_IPV4_ARP stuff and
  * NET_IPV4_IP_FORWARD.
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index e18cd3628db4..392e80e3268d 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -8,8 +8,6 @@
  *		as published by the Free Software Foundation; either version
  *		2 of the License, or (at your option) any later version.
  *
- * Version:	$Id: af_unix.c,v 1.133 2002/02/08 03:57:19 davem Exp $
- *
  * Fixes:
  *		Linus Torvalds	:	Assorted bug cures.
  *		Niibe Yutaka	:	async I/O support.
-- 
cgit v1.2.3


From fc32b0e28df6655a15b488aaddfc1339f82dc13a Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 2 Jun 2008 00:28:40 +0200
Subject: mv643xx_eth: general cleanup

General cleanup of the mv643xx_eth driver.  Mainly fixes coding
style / indentation issues, get rid of some useless 'volatile's,
kill some more superfluous comments, and such.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
---
 drivers/net/mv643xx_eth.c   | 949 ++++++++++++++++++++------------------------
 include/linux/mv643xx_eth.h |  59 ++-
 2 files changed, 479 insertions(+), 529 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index ff6460124307..cf18419f96ab 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -67,8 +67,6 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define MAX_DESCS_PER_SKB	1
 #endif
 
-#define ETH_HW_IP_ALIGN		2
-
 /*
  * Registers shared between all ports.
  */
@@ -158,12 +156,6 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define DEFAULT_RX_QUEUE_SIZE		400
 #define DEFAULT_TX_QUEUE_SIZE		800
 
-/* SMI reg */
-#define SMI_BUSY		0x10000000	/* 0 - Write, 1 - Read	*/
-#define SMI_READ_VALID		0x08000000	/* 0 - Write, 1 - Read	*/
-#define SMI_OPCODE_WRITE	0		/* Completion of Read	*/
-#define SMI_OPCODE_READ		0x04000000	/* Operation is in progress */
-
 
 /*
  * RX/TX descriptors.
@@ -231,13 +223,24 @@ struct tx_desc {
 
 /* global *******************************************************************/
 struct mv643xx_eth_shared_private {
+	/*
+	 * Ethernet controller base address.
+	 */
 	void __iomem *base;
 
-	/* used to protect SMI_REG, which is shared across ports */
+	/*
+	 * Protects access to SMI_REG, which is shared between ports.
+	 */
 	spinlock_t phy_lock;
 
+	/*
+	 * Per-port MBUS window access register value.
+	 */
 	u32 win_protect;
 
+	/*
+	 * Hardware-specific parameters.
+	 */
 	unsigned int t_clk;
 };
 
@@ -306,16 +309,17 @@ struct tx_queue {
 
 struct mv643xx_eth_private {
 	struct mv643xx_eth_shared_private *shared;
-	int port_num;			/* User Ethernet port number	*/
+	int port_num;
 
-	struct mv643xx_eth_shared_private *shared_smi;
+	struct net_device *dev;
 
-	struct work_struct tx_timeout_task;
+	struct mv643xx_eth_shared_private *shared_smi;
+	int phy_addr;
 
-	struct net_device *dev;
-	struct mib_counters mib_counters;
 	spinlock_t lock;
 
+	struct mib_counters mib_counters;
+	struct work_struct tx_timeout_task;
 	struct mii_if_info mii;
 
 	/*
@@ -450,7 +454,12 @@ static void rxq_refill(struct rx_queue *rxq)
 						RX_ENABLE_INTERRUPT;
 		wmb();
 
-		skb_reserve(skb, ETH_HW_IP_ALIGN);
+		/*
+		 * The hardware automatically prepends 2 bytes of
+		 * dummy data to each received packet, so that the
+		 * IP header ends up 16-byte aligned.
+		 */
+		skb_reserve(skb, 2);
 	}
 
 	if (rxq->rx_desc_count == 0) {
@@ -474,9 +483,9 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 
 	rx = 0;
 	while (rx < budget) {
-		struct sk_buff *skb;
-		volatile struct rx_desc *rx_desc;
+		struct rx_desc *rx_desc;
 		unsigned int cmd_sts;
+		struct sk_buff *skb;
 		unsigned long flags;
 
 		spin_lock_irqsave(&mp->lock, flags);
@@ -497,34 +506,40 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 
 		spin_unlock_irqrestore(&mp->lock, flags);
 
-		dma_unmap_single(NULL, rx_desc->buf_ptr + ETH_HW_IP_ALIGN,
-					mp->dev->mtu + 24, DMA_FROM_DEVICE);
+		dma_unmap_single(NULL, rx_desc->buf_ptr + 2,
+				 mp->dev->mtu + 24, DMA_FROM_DEVICE);
 		rxq->rx_desc_count--;
 		rx++;
 
 		/*
 		 * Update statistics.
-		 * Note byte count includes 4 byte CRC count
+		 *
+		 * Note that the descriptor byte count includes 2 dummy
+		 * bytes automatically inserted by the hardware at the
+		 * start of the packet (which we don't count), and a 4
+		 * byte CRC at the end of the packet (which we do count).
 		 */
 		stats->rx_packets++;
-		stats->rx_bytes += rx_desc->byte_cnt - ETH_HW_IP_ALIGN;
+		stats->rx_bytes += rx_desc->byte_cnt - 2;
 
 		/*
-		 * In case received a packet without first / last bits on OR
-		 * the error summary bit is on, the packets needs to be dropeed.
+		 * In case we received a packet without first / last bits
+		 * on, or the error summary bit is set, the packet needs
+		 * to be dropped.
 		 */
 		if (((cmd_sts & (RX_FIRST_DESC | RX_LAST_DESC)) !=
 					(RX_FIRST_DESC | RX_LAST_DESC))
 				|| (cmd_sts & ERROR_SUMMARY)) {
 			stats->rx_dropped++;
+
 			if ((cmd_sts & (RX_FIRST_DESC | RX_LAST_DESC)) !=
 				(RX_FIRST_DESC | RX_LAST_DESC)) {
 				if (net_ratelimit())
-					printk(KERN_ERR
-						"%s: Received packet spread "
-						"on multiple descriptors\n",
-						mp->dev->name);
+					dev_printk(KERN_ERR, &mp->dev->dev,
+						   "received packet spanning "
+						   "multiple descriptors\n");
 			}
+
 			if (cmd_sts & ERROR_SUMMARY)
 				stats->rx_errors++;
 
@@ -534,7 +549,7 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 			 * The -4 is for the CRC in the trailer of the
 			 * received packet
 			 */
-			skb_put(skb, rx_desc->byte_cnt - ETH_HW_IP_ALIGN - 4);
+			skb_put(skb, rx_desc->byte_cnt - 2 - 4);
 
 			if (cmd_sts & LAYER_4_CHECKSUM_OK) {
 				skb->ip_summed = CHECKSUM_UNNECESSARY;
@@ -548,8 +563,10 @@ static int rxq_process(struct rx_queue *rxq, int budget)
 			netif_rx(skb);
 #endif
 		}
+
 		mp->dev->last_rx = jiffies;
 	}
+
 	rxq_refill(rxq);
 
 	return rx;
@@ -716,7 +733,7 @@ static void txq_submit_skb(struct tx_queue *txq, struct sk_buff *skb)
 	txq->tx_desc_count += nr_frags + 1;
 }
 
-static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	struct net_device_stats *stats = &dev->stats;
@@ -727,8 +744,9 @@ static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (has_tiny_unaligned_frags(skb) && __skb_linearize(skb)) {
 		stats->tx_dropped++;
-		printk(KERN_DEBUG "%s: failed to linearize tiny "
-				"unaligned fragment\n", dev->name);
+		dev_printk(KERN_DEBUG, &dev->dev,
+			   "failed to linearize skb with tiny "
+			   "unaligned fragment\n");
 		return NETDEV_TX_BUSY;
 	}
 
@@ -758,13 +776,15 @@ static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 
 /* mii management interface *************************************************/
-static int phy_addr_get(struct mv643xx_eth_private *mp);
+#define SMI_BUSY		0x10000000
+#define SMI_READ_VALID		0x08000000
+#define SMI_OPCODE_READ		0x04000000
+#define SMI_OPCODE_WRITE	0x00000000
 
-static void read_smi_reg(struct mv643xx_eth_private *mp,
-				unsigned int phy_reg, unsigned int *value)
+static void smi_reg_read(struct mv643xx_eth_private *mp, unsigned int addr,
+			 unsigned int reg, unsigned int *value)
 {
 	void __iomem *smi_reg = mp->shared_smi->base + SMI_REG;
-	int phy_addr = phy_addr_get(mp);
 	unsigned long flags;
 	int i;
 
@@ -780,7 +800,7 @@ static void read_smi_reg(struct mv643xx_eth_private *mp,
 		udelay(10);
 	}
 
-	writel((phy_addr << 16) | (phy_reg << 21) | SMI_OPCODE_READ, smi_reg);
+	writel(SMI_OPCODE_READ | (reg << 21) | (addr << 16), smi_reg);
 
 	/* now wait for the data to be valid */
 	for (i = 0; !(readl(smi_reg) & SMI_READ_VALID); i++) {
@@ -796,11 +816,11 @@ out:
 	spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags);
 }
 
-static void write_smi_reg(struct mv643xx_eth_private *mp,
-				   unsigned int phy_reg, unsigned int value)
+static void smi_reg_write(struct mv643xx_eth_private *mp,
+			  unsigned int addr,
+			  unsigned int reg, unsigned int value)
 {
 	void __iomem *smi_reg = mp->shared_smi->base + SMI_REG;
-	int phy_addr = phy_addr_get(mp);
 	unsigned long flags;
 	int i;
 
@@ -816,65 +836,63 @@ static void write_smi_reg(struct mv643xx_eth_private *mp,
 		udelay(10);
 	}
 
-	writel((phy_addr << 16) | (phy_reg << 21) |
-		SMI_OPCODE_WRITE | (value & 0xffff), smi_reg);
+	writel(SMI_OPCODE_WRITE | (reg << 21) |
+		(addr << 16) | (value & 0xffff), smi_reg);
 out:
 	spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags);
 }
 
 
 /* mib counters *************************************************************/
-static void clear_mib_counters(struct mv643xx_eth_private *mp)
+static inline u32 mib_read(struct mv643xx_eth_private *mp, int offset)
 {
-	unsigned int port_num = mp->port_num;
-	int i;
-
-	/* Perform dummy reads from MIB counters */
-	for (i = 0; i < 0x80; i += 4)
-		rdl(mp, MIB_COUNTERS(port_num) + i);
+	return rdl(mp, MIB_COUNTERS(mp->port_num) + offset);
 }
 
-static inline u32 read_mib(struct mv643xx_eth_private *mp, int offset)
+static void mib_counters_clear(struct mv643xx_eth_private *mp)
 {
-	return rdl(mp, MIB_COUNTERS(mp->port_num) + offset);
+	int i;
+
+	for (i = 0; i < 0x80; i += 4)
+		mib_read(mp, i);
 }
 
-static void update_mib_counters(struct mv643xx_eth_private *mp)
+static void mib_counters_update(struct mv643xx_eth_private *mp)
 {
 	struct mib_counters *p = &mp->mib_counters;
 
-	p->good_octets_received += read_mib(mp, 0x00);
-	p->good_octets_received += (u64)read_mib(mp, 0x04) << 32;
-	p->bad_octets_received += read_mib(mp, 0x08);
-	p->internal_mac_transmit_err += read_mib(mp, 0x0c);
-	p->good_frames_received += read_mib(mp, 0x10);
-	p->bad_frames_received += read_mib(mp, 0x14);
-	p->broadcast_frames_received += read_mib(mp, 0x18);
-	p->multicast_frames_received += read_mib(mp, 0x1c);
-	p->frames_64_octets += read_mib(mp, 0x20);
-	p->frames_65_to_127_octets += read_mib(mp, 0x24);
-	p->frames_128_to_255_octets += read_mib(mp, 0x28);
-	p->frames_256_to_511_octets += read_mib(mp, 0x2c);
-	p->frames_512_to_1023_octets += read_mib(mp, 0x30);
-	p->frames_1024_to_max_octets += read_mib(mp, 0x34);
-	p->good_octets_sent += read_mib(mp, 0x38);
-	p->good_octets_sent += (u64)read_mib(mp, 0x3c) << 32;
-	p->good_frames_sent += read_mib(mp, 0x40);
-	p->excessive_collision += read_mib(mp, 0x44);
-	p->multicast_frames_sent += read_mib(mp, 0x48);
-	p->broadcast_frames_sent += read_mib(mp, 0x4c);
-	p->unrec_mac_control_received += read_mib(mp, 0x50);
-	p->fc_sent += read_mib(mp, 0x54);
-	p->good_fc_received += read_mib(mp, 0x58);
-	p->bad_fc_received += read_mib(mp, 0x5c);
-	p->undersize_received += read_mib(mp, 0x60);
-	p->fragments_received += read_mib(mp, 0x64);
-	p->oversize_received += read_mib(mp, 0x68);
-	p->jabber_received += read_mib(mp, 0x6c);
-	p->mac_receive_error += read_mib(mp, 0x70);
-	p->bad_crc_event += read_mib(mp, 0x74);
-	p->collision += read_mib(mp, 0x78);
-	p->late_collision += read_mib(mp, 0x7c);
+	p->good_octets_received += mib_read(mp, 0x00);
+	p->good_octets_received += (u64)mib_read(mp, 0x04) << 32;
+	p->bad_octets_received += mib_read(mp, 0x08);
+	p->internal_mac_transmit_err += mib_read(mp, 0x0c);
+	p->good_frames_received += mib_read(mp, 0x10);
+	p->bad_frames_received += mib_read(mp, 0x14);
+	p->broadcast_frames_received += mib_read(mp, 0x18);
+	p->multicast_frames_received += mib_read(mp, 0x1c);
+	p->frames_64_octets += mib_read(mp, 0x20);
+	p->frames_65_to_127_octets += mib_read(mp, 0x24);
+	p->frames_128_to_255_octets += mib_read(mp, 0x28);
+	p->frames_256_to_511_octets += mib_read(mp, 0x2c);
+	p->frames_512_to_1023_octets += mib_read(mp, 0x30);
+	p->frames_1024_to_max_octets += mib_read(mp, 0x34);
+	p->good_octets_sent += mib_read(mp, 0x38);
+	p->good_octets_sent += (u64)mib_read(mp, 0x3c) << 32;
+	p->good_frames_sent += mib_read(mp, 0x40);
+	p->excessive_collision += mib_read(mp, 0x44);
+	p->multicast_frames_sent += mib_read(mp, 0x48);
+	p->broadcast_frames_sent += mib_read(mp, 0x4c);
+	p->unrec_mac_control_received += mib_read(mp, 0x50);
+	p->fc_sent += mib_read(mp, 0x54);
+	p->good_fc_received += mib_read(mp, 0x58);
+	p->bad_fc_received += mib_read(mp, 0x5c);
+	p->undersize_received += mib_read(mp, 0x60);
+	p->fragments_received += mib_read(mp, 0x64);
+	p->oversize_received += mib_read(mp, 0x68);
+	p->jabber_received += mib_read(mp, 0x6c);
+	p->mac_receive_error += mib_read(mp, 0x70);
+	p->bad_crc_event += mib_read(mp, 0x74);
+	p->collision += mib_read(mp, 0x78);
+	p->late_collision += mib_read(mp, 0x7c);
 }
 
 
@@ -944,7 +962,9 @@ static int mv643xx_eth_get_settings(struct net_device *dev, struct ethtool_cmd *
 	err = mii_ethtool_gset(&mp->mii, cmd);
 	spin_unlock_irq(&mp->lock);
 
-	/* The PHY may support 1000baseT_Half, but the mv643xx does not */
+	/*
+	 * The MAC does not support 1000baseT_Half.
+	 */
 	cmd->supported &= ~SUPPORTED_1000baseT_Half;
 	cmd->advertising &= ~ADVERTISED_1000baseT_Half;
 
@@ -956,6 +976,11 @@ static int mv643xx_eth_set_settings(struct net_device *dev, struct ethtool_cmd *
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	int err;
 
+	/*
+	 * The MAC does not support 1000baseT_Half.
+	 */
+	cmd->advertising &= ~ADVERTISED_1000baseT_Half;
+
 	spin_lock_irq(&mp->lock);
 	err = mii_ethtool_sset(&mp->mii, cmd);
 	spin_unlock_irq(&mp->lock);
@@ -963,17 +988,17 @@ static int mv643xx_eth_set_settings(struct net_device *dev, struct ethtool_cmd *
 	return err;
 }
 
-static void mv643xx_eth_get_drvinfo(struct net_device *netdev,
-				struct ethtool_drvinfo *drvinfo)
+static void mv643xx_eth_get_drvinfo(struct net_device *dev,
+				    struct ethtool_drvinfo *drvinfo)
 {
 	strncpy(drvinfo->driver,  mv643xx_eth_driver_name, 32);
 	strncpy(drvinfo->version, mv643xx_eth_driver_version, 32);
 	strncpy(drvinfo->fw_version, "N/A", 32);
-	strncpy(drvinfo->bus_info, "mv643xx", 32);
+	strncpy(drvinfo->bus_info, "platform", 32);
 	drvinfo->n_stats = ARRAY_SIZE(mv643xx_eth_stats);
 }
 
-static int mv643xx_eth_nway_restart(struct net_device *dev)
+static int mv643xx_eth_nway_reset(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
@@ -987,29 +1012,28 @@ static u32 mv643xx_eth_get_link(struct net_device *dev)
 	return mii_link_ok(&mp->mii);
 }
 
-static void mv643xx_eth_get_strings(struct net_device *netdev, uint32_t stringset,
-				uint8_t *data)
+static void mv643xx_eth_get_strings(struct net_device *dev,
+				    uint32_t stringset, uint8_t *data)
 {
 	int i;
 
-	switch(stringset) {
-	case ETH_SS_STATS:
-		for (i=0; i < ARRAY_SIZE(mv643xx_eth_stats); i++) {
+	if (stringset == ETH_SS_STATS) {
+		for (i = 0; i < ARRAY_SIZE(mv643xx_eth_stats); i++) {
 			memcpy(data + i * ETH_GSTRING_LEN,
 				mv643xx_eth_stats[i].stat_string,
 				ETH_GSTRING_LEN);
 		}
-		break;
 	}
 }
 
-static void mv643xx_eth_get_ethtool_stats(struct net_device *netdev,
-				struct ethtool_stats *stats, uint64_t *data)
+static void mv643xx_eth_get_ethtool_stats(struct net_device *dev,
+					  struct ethtool_stats *stats,
+					  uint64_t *data)
 {
-	struct mv643xx_eth_private *mp = netdev->priv;
+	struct mv643xx_eth_private *mp = dev->priv;
 	int i;
 
-	update_mib_counters(mp);
+	mib_counters_update(mp);
 
 	for (i = 0; i < ARRAY_SIZE(mv643xx_eth_stats); i++) {
 		const struct mv643xx_eth_stats *stat;
@@ -1027,38 +1051,35 @@ static void mv643xx_eth_get_ethtool_stats(struct net_device *netdev,
 	}
 }
 
-static int mv643xx_eth_get_sset_count(struct net_device *netdev, int sset)
+static int mv643xx_eth_get_sset_count(struct net_device *dev, int sset)
 {
-	switch (sset) {
-	case ETH_SS_STATS:
+	if (sset == ETH_SS_STATS)
 		return ARRAY_SIZE(mv643xx_eth_stats);
-	default:
-		return -EOPNOTSUPP;
-	}
+
+	return -EOPNOTSUPP;
 }
 
 static const struct ethtool_ops mv643xx_eth_ethtool_ops = {
-	.get_settings           = mv643xx_eth_get_settings,
-	.set_settings           = mv643xx_eth_set_settings,
-	.get_drvinfo            = mv643xx_eth_get_drvinfo,
-	.get_link               = mv643xx_eth_get_link,
+	.get_settings		= mv643xx_eth_get_settings,
+	.set_settings		= mv643xx_eth_set_settings,
+	.get_drvinfo		= mv643xx_eth_get_drvinfo,
+	.nway_reset		= mv643xx_eth_nway_reset,
+	.get_link		= mv643xx_eth_get_link,
 	.set_sg			= ethtool_op_set_sg,
+	.get_strings		= mv643xx_eth_get_strings,
+	.get_ethtool_stats	= mv643xx_eth_get_ethtool_stats,
 	.get_sset_count		= mv643xx_eth_get_sset_count,
-	.get_ethtool_stats      = mv643xx_eth_get_ethtool_stats,
-	.get_strings            = mv643xx_eth_get_strings,
-	.nway_reset		= mv643xx_eth_nway_restart,
 };
 
 
 /* address handling *********************************************************/
 static void uc_addr_get(struct mv643xx_eth_private *mp, unsigned char *addr)
 {
-	unsigned int port_num = mp->port_num;
 	unsigned int mac_h;
 	unsigned int mac_l;
 
-	mac_h = rdl(mp, MAC_ADDR_HIGH(port_num));
-	mac_l = rdl(mp, MAC_ADDR_LOW(port_num));
+	mac_h = rdl(mp, MAC_ADDR_HIGH(mp->port_num));
+	mac_l = rdl(mp, MAC_ADDR_LOW(mp->port_num));
 
 	addr[0] = (mac_h >> 24) & 0xff;
 	addr[1] = (mac_h >> 16) & 0xff;
@@ -1070,72 +1091,54 @@ static void uc_addr_get(struct mv643xx_eth_private *mp, unsigned char *addr)
 
 static void init_mac_tables(struct mv643xx_eth_private *mp)
 {
-	unsigned int port_num = mp->port_num;
-	int table_index;
-
-	/* Clear DA filter unicast table (Ex_dFUT) */
-	for (table_index = 0; table_index <= 0xC; table_index += 4)
-		wrl(mp, UNICAST_TABLE(port_num) + table_index, 0);
+	int i;
 
-	for (table_index = 0; table_index <= 0xFC; table_index += 4) {
-		/* Clear DA filter special multicast table (Ex_dFSMT) */
-		wrl(mp, SPECIAL_MCAST_TABLE(port_num) + table_index, 0);
-		/* Clear DA filter other multicast table (Ex_dFOMT) */
-		wrl(mp, OTHER_MCAST_TABLE(port_num) + table_index, 0);
+	for (i = 0; i < 0x100; i += 4) {
+		wrl(mp, SPECIAL_MCAST_TABLE(mp->port_num) + i, 0);
+		wrl(mp, OTHER_MCAST_TABLE(mp->port_num) + i, 0);
 	}
+
+	for (i = 0; i < 0x10; i += 4)
+		wrl(mp, UNICAST_TABLE(mp->port_num) + i, 0);
 }
 
 static void set_filter_table_entry(struct mv643xx_eth_private *mp,
-					    int table, unsigned char entry)
+				   int table, unsigned char entry)
 {
 	unsigned int table_reg;
-	unsigned int tbl_offset;
-	unsigned int reg_offset;
-
-	tbl_offset = (entry / 4) * 4;	/* Register offset of DA table entry */
-	reg_offset = entry % 4;		/* Entry offset within the register */
 
 	/* Set "accepts frame bit" at specified table entry */
-	table_reg = rdl(mp, table + tbl_offset);
-	table_reg |= 0x01 << (8 * reg_offset);
-	wrl(mp, table + tbl_offset, table_reg);
+	table_reg = rdl(mp, table + (entry & 0xfc));
+	table_reg |= 0x01 << (8 * (entry & 3));
+	wrl(mp, table + (entry & 0xfc), table_reg);
 }
 
 static void uc_addr_set(struct mv643xx_eth_private *mp, unsigned char *addr)
 {
-	unsigned int port_num = mp->port_num;
 	unsigned int mac_h;
 	unsigned int mac_l;
 	int table;
 
-	mac_l = (addr[4] << 8) | (addr[5]);
-	mac_h = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) |
-							(addr[3] << 0);
+	mac_l = (addr[4] << 8) | addr[5];
+	mac_h = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) | addr[3];
 
-	wrl(mp, MAC_ADDR_LOW(port_num), mac_l);
-	wrl(mp, MAC_ADDR_HIGH(port_num), mac_h);
+	wrl(mp, MAC_ADDR_LOW(mp->port_num), mac_l);
+	wrl(mp, MAC_ADDR_HIGH(mp->port_num), mac_h);
 
-	/* Accept frames with this address */
-	table = UNICAST_TABLE(port_num);
+	table = UNICAST_TABLE(mp->port_num);
 	set_filter_table_entry(mp, table, addr[5] & 0x0f);
 }
 
-static void mv643xx_eth_update_mac_address(struct net_device *dev)
+static int mv643xx_eth_set_mac_address(struct net_device *dev, void *addr)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
+	/* +2 is for the offset of the HW addr type */
+	memcpy(dev->dev_addr, addr + 2, 6);
+
 	init_mac_tables(mp);
 	uc_addr_set(mp, dev->dev_addr);
-}
-
-static int mv643xx_eth_set_mac_address(struct net_device *dev, void *addr)
-{
-	int i;
 
-	for (i = 0; i < 6; i++)
-		/* +2 is for the offset of the HW addr type */
-		dev->dev_addr[i] = ((unsigned char *)addr)[i + 2];
-	mv643xx_eth_update_mac_address(dev);
 	return 0;
 }
 
@@ -1157,95 +1160,53 @@ static int addr_crc(unsigned char *addr)
 	return crc;
 }
 
-static void mc_addr(struct mv643xx_eth_private *mp, unsigned char *addr)
+static void mv643xx_eth_set_rx_mode(struct net_device *dev)
 {
-	unsigned int port_num = mp->port_num;
-	int table;
-	int crc;
-
-	if ((addr[0] == 0x01) && (addr[1] == 0x00) &&
-	    (addr[2] == 0x5E) && (addr[3] == 0x00) && (addr[4] == 0x00)) {
-		table = SPECIAL_MCAST_TABLE(port_num);
-		set_filter_table_entry(mp, table, addr[5]);
-		return;
-	}
-
-	crc = addr_crc(addr);
-
-	table = OTHER_MCAST_TABLE(port_num);
-	set_filter_table_entry(mp, table, crc);
-}
+	struct mv643xx_eth_private *mp = netdev_priv(dev);
+	u32 port_config;
+	struct dev_addr_list *addr;
+	int i;
 
-static void set_multicast_list(struct net_device *dev)
-{
+	port_config = rdl(mp, PORT_CONFIG(mp->port_num));
+	if (dev->flags & IFF_PROMISC)
+		port_config |= UNICAST_PROMISCUOUS_MODE;
+	else
+		port_config &= ~UNICAST_PROMISCUOUS_MODE;
+	wrl(mp, PORT_CONFIG(mp->port_num), port_config);
 
-	struct dev_mc_list	*mc_list;
-	int			i;
-	int			table_index;
-	struct mv643xx_eth_private	*mp = netdev_priv(dev);
-	unsigned int		port_num = mp->port_num;
+	if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) {
+		int port_num = mp->port_num;
+		u32 accept = 0x01010101;
 
-	/* If the device is in promiscuous mode or in all multicast mode,
-	 * we will fully populate both multicast tables with accept.
-	 * This is guaranteed to yield a match on all multicast addresses...
-	 */
-	if ((dev->flags & IFF_PROMISC) || (dev->flags & IFF_ALLMULTI)) {
-		for (table_index = 0; table_index <= 0xFC; table_index += 4) {
-			/* Set all entries in DA filter special multicast
-			 * table (Ex_dFSMT)
-			 * Set for ETH_Q0 for now
-			 * Bits
-			 * 0	  Accept=1, Drop=0
-			 * 3-1  Queue	 ETH_Q0=0
-			 * 7-4  Reserved = 0;
-			 */
-			wrl(mp, SPECIAL_MCAST_TABLE(port_num) + table_index, 0x01010101);
-
-			/* Set all entries in DA filter other multicast
-			 * table (Ex_dFOMT)
-			 * Set for ETH_Q0 for now
-			 * Bits
-			 * 0	  Accept=1, Drop=0
-			 * 3-1  Queue	 ETH_Q0=0
-			 * 7-4  Reserved = 0;
-			 */
-			wrl(mp, OTHER_MCAST_TABLE(port_num) + table_index, 0x01010101);
+		for (i = 0; i < 0x100; i += 4) {
+			wrl(mp, SPECIAL_MCAST_TABLE(port_num) + i, accept);
+			wrl(mp, OTHER_MCAST_TABLE(port_num) + i, accept);
 		}
 		return;
 	}
 
-	/* We will clear out multicast tables every time we get the list.
-	 * Then add the entire new list...
-	 */
-	for (table_index = 0; table_index <= 0xFC; table_index += 4) {
-		/* Clear DA filter special multicast table (Ex_dFSMT) */
-		wrl(mp, SPECIAL_MCAST_TABLE(port_num) + table_index, 0);
-
-		/* Clear DA filter other multicast table (Ex_dFOMT) */
-		wrl(mp, OTHER_MCAST_TABLE(port_num) + table_index, 0);
+	for (i = 0; i < 0x100; i += 4) {
+		wrl(mp, SPECIAL_MCAST_TABLE(mp->port_num) + i, 0);
+		wrl(mp, OTHER_MCAST_TABLE(mp->port_num) + i, 0);
 	}
 
-	/* Get pointer to net_device multicast list and add each one... */
-	for (i = 0, mc_list = dev->mc_list;
-			(i < 256) && (mc_list != NULL) && (i < dev->mc_count);
-			i++, mc_list = mc_list->next)
-		if (mc_list->dmi_addrlen == 6)
-			mc_addr(mp, mc_list->dmi_addr);
-}
+	for (addr = dev->mc_list; addr != NULL; addr = addr->next) {
+		u8 *a = addr->da_addr;
+		int table;
 
-static void mv643xx_eth_set_rx_mode(struct net_device *dev)
-{
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	u32 config_reg;
+		if (addr->da_addrlen != 6)
+			continue;
 
-	config_reg = rdl(mp, PORT_CONFIG(mp->port_num));
-	if (dev->flags & IFF_PROMISC)
-		config_reg |= UNICAST_PROMISCUOUS_MODE;
-	else
-		config_reg &= ~UNICAST_PROMISCUOUS_MODE;
-	wrl(mp, PORT_CONFIG(mp->port_num), config_reg);
+		if (memcmp(a, "\x01\x00\x5e\x00\x00", 5) == 0) {
+			table = SPECIAL_MCAST_TABLE(mp->port_num);
+			set_filter_table_entry(mp, table, a[5]);
+		} else {
+			int crc = addr_crc(a);
 
-	set_multicast_list(dev);
+			table = OTHER_MCAST_TABLE(mp->port_num);
+			set_filter_table_entry(mp, table, crc);
+		}
+	}
 }
 
 
@@ -1483,10 +1444,7 @@ static void txq_deinit(struct tx_queue *txq)
 
 
 /* netdev ops and related ***************************************************/
-static void port_reset(struct mv643xx_eth_private *mp);
-
-static void mv643xx_eth_update_pscr(struct mv643xx_eth_private *mp,
-				    struct ethtool_cmd *ecmd)
+static void update_pscr(struct mv643xx_eth_private *mp, int speed, int duplex)
 {
 	u32 pscr_o;
 	u32 pscr_n;
@@ -1499,15 +1457,15 @@ static void mv643xx_eth_update_pscr(struct mv643xx_eth_private *mp,
 			    SET_FULL_DUPLEX_MODE   |
 			    MAX_RX_PACKET_MASK);
 
-	if (ecmd->speed == SPEED_1000) {
+	if (speed == SPEED_1000) {
 		pscr_n |= SET_GMII_SPEED_TO_1000 | MAX_RX_PACKET_9700BYTE;
 	} else {
-		if (ecmd->speed == SPEED_100)
+		if (speed == SPEED_100)
 			pscr_n |= SET_MII_SPEED_TO_100;
 		pscr_n |= MAX_RX_PACKET_1522BYTE;
 	}
 
-	if (ecmd->duplex == DUPLEX_FULL)
+	if (duplex == DUPLEX_FULL)
 		pscr_n |= SET_FULL_DUPLEX_MODE;
 
 	if (pscr_n != pscr_o) {
@@ -1524,27 +1482,30 @@ static void mv643xx_eth_update_pscr(struct mv643xx_eth_private *mp,
 	}
 }
 
-static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id)
+static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 {
 	struct net_device *dev = (struct net_device *)dev_id;
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	u32 int_cause, int_cause_ext = 0;
+	u32 int_cause;
+	u32 int_cause_ext;
 
-	/* Read interrupt cause registers */
 	int_cause = rdl(mp, INT_CAUSE(mp->port_num)) & (INT_RX | INT_EXT);
+	if (int_cause == 0)
+		return IRQ_NONE;
+
+	int_cause_ext = 0;
 	if (int_cause & INT_EXT) {
 		int_cause_ext = rdl(mp, INT_CAUSE_EXT(mp->port_num))
 				& (INT_EXT_LINK | INT_EXT_PHY | INT_EXT_TX);
 		wrl(mp, INT_CAUSE_EXT(mp->port_num), ~int_cause_ext);
 	}
 
-	/* PHY status changed */
-	if (int_cause_ext & (INT_EXT_LINK | INT_EXT_PHY)) {
+	if (int_cause_ext & (INT_EXT_PHY | INT_EXT_LINK)) {
 		if (mii_link_ok(&mp->mii)) {
 			struct ethtool_cmd cmd;
 
 			mii_ethtool_gset(&mp->mii, &cmd);
-			mv643xx_eth_update_pscr(mp, &cmd);
+			update_pscr(mp, cmd.speed, cmd.duplex);
 			txq_enable(mp->txq);
 			if (!netif_carrier_ok(dev)) {
 				netif_carrier_on(dev);
@@ -1558,10 +1519,7 @@ static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id)
 
 #ifdef MV643XX_ETH_NAPI
 	if (int_cause & INT_RX) {
-		/* schedule the NAPI poll routine to maintain port */
 		wrl(mp, INT_MASK(mp->port_num), 0x00000000);
-
-		/* wait for previous write to complete */
 		rdl(mp, INT_MASK(mp->port_num));
 
 		netif_rx_schedule(dev, &mp->napi);
@@ -1570,40 +1528,31 @@ static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id)
 	if (int_cause & INT_RX)
 		rxq_process(mp->rxq, INT_MAX);
 #endif
+
 	if (int_cause_ext & INT_EXT_TX) {
 		txq_reclaim(mp->txq, 0);
 		__txq_maybe_wake(mp->txq);
 	}
 
-	/*
-	 * If no real interrupt occured, exit.
-	 * This can happen when using gigE interrupt coalescing mechanism.
-	 */
-	if ((int_cause == 0x0) && (int_cause_ext == 0x0))
-		return IRQ_NONE;
-
 	return IRQ_HANDLED;
 }
 
 static void phy_reset(struct mv643xx_eth_private *mp)
 {
-	unsigned int phy_reg_data;
+	unsigned int data;
 
-	/* Reset the PHY */
-	read_smi_reg(mp, 0, &phy_reg_data);
-	phy_reg_data |= 0x8000;	/* Set bit 15 to reset the PHY */
-	write_smi_reg(mp, 0, phy_reg_data);
+	smi_reg_read(mp, mp->phy_addr, 0, &data);
+	data |= 0x8000;
+	smi_reg_write(mp, mp->phy_addr, 0, data);
 
-	/* wait for PHY to come out of reset */
 	do {
 		udelay(1);
-		read_smi_reg(mp, 0, &phy_reg_data);
-	} while (phy_reg_data & 0x8000);
+		smi_reg_read(mp, mp->phy_addr, 0, &data);
+	} while (data & 0x8000);
 }
 
-static void port_start(struct net_device *dev)
+static void port_start(struct mv643xx_eth_private *mp)
 {
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	u32 pscr;
 	struct ethtool_cmd ethtool_cmd;
 	int i;
@@ -1625,9 +1574,9 @@ static void port_start(struct net_device *dev)
 
 	wrl(mp, SDMA_CONFIG(mp->port_num), PORT_SDMA_CONFIG_DEFAULT_VALUE);
 
-	mv643xx_eth_get_settings(dev, &ethtool_cmd);
+	mv643xx_eth_get_settings(mp->dev, &ethtool_cmd);
 	phy_reset(mp);
-	mv643xx_eth_set_settings(dev, &ethtool_cmd);
+	mv643xx_eth_set_settings(mp->dev, &ethtool_cmd);
 
 	/*
 	 * Configure TX path and queues.
@@ -1643,8 +1592,10 @@ static void port_start(struct net_device *dev)
 		wrl(mp, off, addr);
 	}
 
-	/* Add the assigned Ethernet address to the port's address table */
-	uc_addr_set(mp, dev->dev_addr);
+	/*
+	 * Add configured unicast address to address filter table.
+	 */
+	uc_addr_set(mp, mp->dev->dev_addr);
 
 	/*
 	 * Receive all unmatched unicast, TCP, UDP, BPDU and broadcast
@@ -1675,13 +1626,14 @@ static void port_start(struct net_device *dev)
 
 static void set_rx_coal(struct mv643xx_eth_private *mp, unsigned int delay)
 {
-	unsigned int port_num = mp->port_num;
 	unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64;
 
-	/* Set RX Coalescing mechanism */
-	wrl(mp, SDMA_CONFIG(port_num),
+	if (coal > 0x3fff)
+		coal = 0x3fff;
+
+	wrl(mp, SDMA_CONFIG(mp->port_num),
 		((coal & 0x3fff) << 8) |
-		(rdl(mp, SDMA_CONFIG(port_num))
+		(rdl(mp, SDMA_CONFIG(mp->port_num))
 			& 0xffc000ff));
 }
 
@@ -1689,68 +1641,59 @@ static void set_tx_coal(struct mv643xx_eth_private *mp, unsigned int delay)
 {
 	unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64;
 
-	/* Set TX Coalescing mechanism */
-	wrl(mp, TX_FIFO_URGENT_THRESHOLD(mp->port_num), coal << 4);
-}
-
-static void port_init(struct mv643xx_eth_private *mp)
-{
-	port_reset(mp);
-
-	init_mac_tables(mp);
+	if (coal > 0x3fff)
+		coal = 0x3fff;
+	wrl(mp, TX_FIFO_URGENT_THRESHOLD(mp->port_num), (coal & 0x3fff) << 4);
 }
 
 static int mv643xx_eth_open(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	unsigned int port_num = mp->port_num;
 	int err;
 
-	/* Clear any pending ethernet port interrupts */
-	wrl(mp, INT_CAUSE(port_num), 0);
-	wrl(mp, INT_CAUSE_EXT(port_num), 0);
-	/* wait for previous write to complete */
-	rdl(mp, INT_CAUSE_EXT(port_num));
+	wrl(mp, INT_CAUSE(mp->port_num), 0);
+	wrl(mp, INT_CAUSE_EXT(mp->port_num), 0);
+	rdl(mp, INT_CAUSE_EXT(mp->port_num));
 
-	err = request_irq(dev->irq, mv643xx_eth_int_handler,
-			IRQF_SHARED | IRQF_SAMPLE_RANDOM, dev->name, dev);
+	err = request_irq(dev->irq, mv643xx_eth_irq,
+			  IRQF_SHARED | IRQF_SAMPLE_RANDOM,
+			  dev->name, dev);
 	if (err) {
-		printk(KERN_ERR "%s: Can not assign IRQ\n", dev->name);
+		dev_printk(KERN_ERR, &dev->dev, "can't assign irq\n");
 		return -EAGAIN;
 	}
 
-	port_init(mp);
+	init_mac_tables(mp);
 
 	err = rxq_init(mp);
 	if (err)
-		goto out_free_irq;
+		goto out;
 	rxq_refill(mp->rxq);
 
 	err = txq_init(mp);
 	if (err)
-		goto out_free_rx_skb;
+		goto out_free;
 
 #ifdef MV643XX_ETH_NAPI
 	napi_enable(&mp->napi);
 #endif
 
-	port_start(dev);
+	port_start(mp);
 
 	set_rx_coal(mp, 0);
 	set_tx_coal(mp, 0);
 
-	/* Unmask phy and link status changes interrupts */
-	wrl(mp, INT_MASK_EXT(port_num), INT_EXT_LINK | INT_EXT_PHY | INT_EXT_TX);
+	wrl(mp, INT_MASK_EXT(mp->port_num),
+	    INT_EXT_LINK | INT_EXT_PHY | INT_EXT_TX);
 
-	/* Unmask RX buffer and TX end interrupt */
-	wrl(mp, INT_MASK(port_num), INT_RX | INT_EXT);
+	wrl(mp, INT_MASK(mp->port_num), INT_RX | INT_EXT);
 
 	return 0;
 
 
-out_free_rx_skb:
+out_free:
 	rxq_deinit(mp->rxq);
-out_free_irq:
+out:
 	free_irq(dev->irq, dev);
 
 	return err;
@@ -1758,34 +1701,27 @@ out_free_irq:
 
 static void port_reset(struct mv643xx_eth_private *mp)
 {
-	unsigned int port_num = mp->port_num;
-	unsigned int reg_data;
+	unsigned int data;
 
 	txq_disable(mp->txq);
 	rxq_disable(mp->rxq);
 	while (!(rdl(mp, PORT_STATUS(mp->port_num)) & TX_FIFO_EMPTY))
 		udelay(10);
 
-	/* Clear all MIB counters */
-	clear_mib_counters(mp);
-
 	/* Reset the Enable bit in the Configuration Register */
-	reg_data = rdl(mp, PORT_SERIAL_CONTROL(port_num));
-	reg_data &= ~(SERIAL_PORT_ENABLE		|
-			DO_NOT_FORCE_LINK_FAIL	|
-			FORCE_LINK_PASS);
-	wrl(mp, PORT_SERIAL_CONTROL(port_num), reg_data);
+	data = rdl(mp, PORT_SERIAL_CONTROL(mp->port_num));
+	data &= ~(SERIAL_PORT_ENABLE		|
+		  DO_NOT_FORCE_LINK_FAIL	|
+		  FORCE_LINK_PASS);
+	wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), data);
 }
 
 static int mv643xx_eth_stop(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	unsigned int port_num = mp->port_num;
 
-	/* Mask all interrupts on ethernet port */
-	wrl(mp, INT_MASK(port_num), 0x00000000);
-	/* wait for previous write to complete */
-	rdl(mp, INT_MASK(port_num));
+	wrl(mp, INT_MASK(mp->port_num), 0x00000000);
+	rdl(mp, INT_MASK(mp->port_num));
 
 #ifdef MV643XX_ETH_NAPI
 	napi_disable(&mp->napi);
@@ -1793,17 +1729,18 @@ static int mv643xx_eth_stop(struct net_device *dev)
 	netif_carrier_off(dev);
 	netif_stop_queue(dev);
 
+	free_irq(dev->irq, dev);
+
 	port_reset(mp);
+	mib_counters_update(mp);
 
 	txq_deinit(mp->txq);
 	rxq_deinit(mp->rxq);
 
-	free_irq(dev->irq, dev);
-
 	return 0;
 }
 
-static int mv643xx_eth_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+static int mv643xx_eth_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
@@ -1812,7 +1749,7 @@ static int mv643xx_eth_do_ioctl(struct net_device *dev, struct ifreq *ifr, int c
 
 static int mv643xx_eth_change_mtu(struct net_device *dev, int new_mtu)
 {
-	if ((new_mtu > 9500) || (new_mtu < 64))
+	if (new_mtu < 64 || new_mtu > 9500)
 		return -EINVAL;
 
 	dev->mtu = new_mtu;
@@ -1823,73 +1760,70 @@ static int mv643xx_eth_change_mtu(struct net_device *dev, int new_mtu)
 	 * Stop and then re-open the interface. This will allocate RX
 	 * skbs of the new MTU.
 	 * There is a possible danger that the open will not succeed,
-	 * due to memory being full, which might fail the open function.
+	 * due to memory being full.
 	 */
 	mv643xx_eth_stop(dev);
 	if (mv643xx_eth_open(dev)) {
-		printk(KERN_ERR "%s: Fatal error on opening device\n",
-			dev->name);
+		dev_printk(KERN_ERR, &dev->dev,
+			   "fatal error on re-opening device after "
+			   "MTU change\n");
 	}
 
 	return 0;
 }
 
-static void mv643xx_eth_tx_timeout_task(struct work_struct *ugly)
+static void tx_timeout_task(struct work_struct *ugly)
 {
-	struct mv643xx_eth_private *mp = container_of(ugly, struct mv643xx_eth_private,
-						  tx_timeout_task);
-	struct net_device *dev = mp->dev;
-
-	if (!netif_running(dev))
-		return;
+	struct mv643xx_eth_private *mp;
 
-	netif_stop_queue(dev);
+	mp = container_of(ugly, struct mv643xx_eth_private, tx_timeout_task);
+	if (netif_running(mp->dev)) {
+		netif_stop_queue(mp->dev);
 
-	port_reset(mp);
-	port_start(dev);
+		port_reset(mp);
+		port_start(mp);
 
-	__txq_maybe_wake(mp->txq);
+		__txq_maybe_wake(mp->txq);
+	}
 }
 
 static void mv643xx_eth_tx_timeout(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
-	printk(KERN_INFO "%s: TX timeout  ", dev->name);
+	dev_printk(KERN_INFO, &dev->dev, "tx timeout\n");
 
-	/* Do the reset outside of interrupt context */
 	schedule_work(&mp->tx_timeout_task);
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
-static void mv643xx_eth_netpoll(struct net_device *netdev)
+static void mv643xx_eth_netpoll(struct net_device *dev)
 {
-	struct mv643xx_eth_private *mp = netdev_priv(netdev);
-	int port_num = mp->port_num;
+	struct mv643xx_eth_private *mp = netdev_priv(dev);
 
-	wrl(mp, INT_MASK(port_num), 0x00000000);
-	/* wait for previous write to complete */
-	rdl(mp, INT_MASK(port_num));
+	wrl(mp, INT_MASK(mp->port_num), 0x00000000);
+	rdl(mp, INT_MASK(mp->port_num));
 
-	mv643xx_eth_int_handler(netdev->irq, netdev);
+	mv643xx_eth_irq(dev->irq, dev);
 
-	wrl(mp, INT_MASK(port_num), INT_RX | INT_CAUSE_EXT);
+	wrl(mp, INT_MASK(mp->port_num), INT_RX | INT_CAUSE_EXT);
 }
 #endif
 
-static int mv643xx_eth_mdio_read(struct net_device *dev, int phy_id, int location)
+static int mv643xx_eth_mdio_read(struct net_device *dev, int addr, int reg)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	int val;
 
-	read_smi_reg(mp, location, &val);
+	smi_reg_read(mp, addr, reg, &val);
+
 	return val;
 }
 
-static void mv643xx_eth_mdio_write(struct net_device *dev, int phy_id, int location, int val)
+static void mv643xx_eth_mdio_write(struct net_device *dev, int addr, int reg, int val)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	write_smi_reg(mp, location, val);
+	smi_reg_write(mp, addr, reg, val);
 }
 
 
@@ -1956,9 +1890,6 @@ static int mv643xx_eth_shared_probe(struct platform_device *pdev)
 		goto out_free;
 
 	spin_lock_init(&msp->phy_lock);
-	msp->t_clk = (pd != NULL && pd->t_clk != 0) ? pd->t_clk : 133000000;
-
-	platform_set_drvdata(pdev, msp);
 
 	/*
 	 * (Re-)program MBUS remapping windows if we are asked to.
@@ -1966,6 +1897,13 @@ static int mv643xx_eth_shared_probe(struct platform_device *pdev)
 	if (pd != NULL && pd->dram != NULL)
 		mv643xx_eth_conf_mbus_windows(msp, pd->dram);
 
+	/*
+	 * Detect hardware parameters.
+	 */
+	msp->t_clk = (pd != NULL && pd->t_clk != 0) ? pd->t_clk : 133000000;
+
+	platform_set_drvdata(pdev, msp);
+
 	return 0;
 
 out_free:
@@ -1985,104 +1923,158 @@ static int mv643xx_eth_shared_remove(struct platform_device *pdev)
 }
 
 static struct platform_driver mv643xx_eth_shared_driver = {
-	.probe = mv643xx_eth_shared_probe,
-	.remove = mv643xx_eth_shared_remove,
+	.probe		= mv643xx_eth_shared_probe,
+	.remove		= mv643xx_eth_shared_remove,
 	.driver = {
-		.name = MV643XX_ETH_SHARED_NAME,
+		.name	= MV643XX_ETH_SHARED_NAME,
 		.owner	= THIS_MODULE,
 	},
 };
 
 static void phy_addr_set(struct mv643xx_eth_private *mp, int phy_addr)
 {
-	u32 reg_data;
 	int addr_shift = 5 * mp->port_num;
+	u32 data;
 
-	reg_data = rdl(mp, PHY_ADDR);
-	reg_data &= ~(0x1f << addr_shift);
-	reg_data |= (phy_addr & 0x1f) << addr_shift;
-	wrl(mp, PHY_ADDR, reg_data);
+	data = rdl(mp, PHY_ADDR);
+	data &= ~(0x1f << addr_shift);
+	data |= (phy_addr & 0x1f) << addr_shift;
+	wrl(mp, PHY_ADDR, data);
 }
 
 static int phy_addr_get(struct mv643xx_eth_private *mp)
 {
-	unsigned int reg_data;
+	unsigned int data;
+
+	data = rdl(mp, PHY_ADDR);
+
+	return (data >> (5 * mp->port_num)) & 0x1f;
+}
+
+static void set_params(struct mv643xx_eth_private *mp,
+		       struct mv643xx_eth_platform_data *pd)
+{
+	struct net_device *dev = mp->dev;
+
+	if (is_valid_ether_addr(pd->mac_addr))
+		memcpy(dev->dev_addr, pd->mac_addr, 6);
+	else
+		uc_addr_get(mp, dev->dev_addr);
+
+	if (pd->phy_addr == -1) {
+		mp->shared_smi = NULL;
+		mp->phy_addr = -1;
+	} else {
+		mp->shared_smi = mp->shared;
+		if (pd->shared_smi != NULL)
+			mp->shared_smi = platform_get_drvdata(pd->shared_smi);
+
+		if (pd->force_phy_addr || pd->phy_addr) {
+			mp->phy_addr = pd->phy_addr & 0x3f;
+			phy_addr_set(mp, mp->phy_addr);
+		} else {
+			mp->phy_addr = phy_addr_get(mp);
+		}
+	}
 
-	reg_data = rdl(mp, PHY_ADDR);
+	mp->default_rx_ring_size = DEFAULT_RX_QUEUE_SIZE;
+	if (pd->rx_queue_size)
+		mp->default_rx_ring_size = pd->rx_queue_size;
+	mp->rx_desc_sram_addr = pd->rx_sram_addr;
+	mp->rx_desc_sram_size = pd->rx_sram_size;
 
-	return ((reg_data >> (5 * mp->port_num)) & 0x1f);
+	mp->default_tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
+	if (pd->tx_queue_size)
+		mp->default_tx_ring_size = pd->tx_queue_size;
+	mp->tx_desc_sram_addr = pd->tx_sram_addr;
+	mp->tx_desc_sram_size = pd->tx_sram_size;
 }
 
 static int phy_detect(struct mv643xx_eth_private *mp)
 {
-	unsigned int phy_reg_data0;
-	int auto_neg;
+	unsigned int data;
+	unsigned int data2;
+
+	smi_reg_read(mp, mp->phy_addr, 0, &data);
+	smi_reg_write(mp, mp->phy_addr, 0, data ^ 0x1000);
 
-	read_smi_reg(mp, 0, &phy_reg_data0);
-	auto_neg = phy_reg_data0 & 0x1000;
-	phy_reg_data0 ^= 0x1000;	/* invert auto_neg */
-	write_smi_reg(mp, 0, phy_reg_data0);
+	smi_reg_read(mp, mp->phy_addr, 0, &data2);
+	if (((data ^ data2) & 0x1000) == 0)
+		return -ENODEV;
 
-	read_smi_reg(mp, 0, &phy_reg_data0);
-	if ((phy_reg_data0 & 0x1000) == auto_neg)
-		return -ENODEV;				/* change didn't take */
+	smi_reg_write(mp, mp->phy_addr, 0, data);
 
-	phy_reg_data0 ^= 0x1000;
-	write_smi_reg(mp, 0, phy_reg_data0);
 	return 0;
 }
 
-static void mv643xx_init_ethtool_cmd(struct net_device *dev, int phy_address,
-				     int speed, int duplex,
-				     struct ethtool_cmd *cmd)
+static int phy_init(struct mv643xx_eth_private *mp,
+		    struct mv643xx_eth_platform_data *pd)
 {
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
+	struct ethtool_cmd cmd;
+	int err;
 
-	memset(cmd, 0, sizeof(*cmd));
+	err = phy_detect(mp);
+	if (err) {
+		dev_printk(KERN_INFO, &mp->dev->dev,
+			   "no PHY detected at addr %d\n", mp->phy_addr);
+		return err;
+	}
+	phy_reset(mp);
+
+	mp->mii.phy_id = mp->phy_addr;
+	mp->mii.phy_id_mask = 0x3f;
+	mp->mii.reg_num_mask = 0x1f;
+	mp->mii.dev = mp->dev;
+	mp->mii.mdio_read = mv643xx_eth_mdio_read;
+	mp->mii.mdio_write = mv643xx_eth_mdio_write;
 
-	cmd->port = PORT_MII;
-	cmd->transceiver = XCVR_INTERNAL;
-	cmd->phy_address = phy_address;
+	mp->mii.supports_gmii = mii_check_gmii_support(&mp->mii);
 
-	if (speed == 0) {
-		cmd->autoneg = AUTONEG_ENABLE;
-		/* mii lib checks, but doesn't use speed on AUTONEG_ENABLE */
-		cmd->speed = SPEED_100;
-		cmd->advertising = ADVERTISED_10baseT_Half  |
-				   ADVERTISED_10baseT_Full  |
-				   ADVERTISED_100baseT_Half |
-				   ADVERTISED_100baseT_Full;
+	memset(&cmd, 0, sizeof(cmd));
+
+	cmd.port = PORT_MII;
+	cmd.transceiver = XCVR_INTERNAL;
+	cmd.phy_address = mp->phy_addr;
+	if (pd->speed == 0) {
+		cmd.autoneg = AUTONEG_ENABLE;
+		cmd.speed = SPEED_100;
+		cmd.advertising = ADVERTISED_10baseT_Half  |
+				  ADVERTISED_10baseT_Full  |
+				  ADVERTISED_100baseT_Half |
+				  ADVERTISED_100baseT_Full;
 		if (mp->mii.supports_gmii)
-			cmd->advertising |= ADVERTISED_1000baseT_Full;
+			cmd.advertising |= ADVERTISED_1000baseT_Full;
 	} else {
-		cmd->autoneg = AUTONEG_DISABLE;
-		cmd->speed = speed;
-		cmd->duplex = duplex;
+		cmd.autoneg = AUTONEG_DISABLE;
+		cmd.speed = pd->speed;
+		cmd.duplex = pd->duplex;
 	}
+
+	update_pscr(mp, cmd.speed, cmd.duplex);
+	mv643xx_eth_set_settings(mp->dev, &cmd);
+
+	return 0;
 }
 
 static int mv643xx_eth_probe(struct platform_device *pdev)
 {
 	struct mv643xx_eth_platform_data *pd;
-	int port_num;
 	struct mv643xx_eth_private *mp;
 	struct net_device *dev;
-	u8 *p;
 	struct resource *res;
-	int err;
-	struct ethtool_cmd cmd;
-	int duplex = DUPLEX_HALF;
-	int speed = 0;			/* default to auto-negotiation */
 	DECLARE_MAC_BUF(mac);
+	int err;
 
 	pd = pdev->dev.platform_data;
 	if (pd == NULL) {
-		printk(KERN_ERR "No mv643xx_eth_platform_data\n");
+		dev_printk(KERN_ERR, &pdev->dev,
+			   "no mv643xx_eth_platform_data\n");
 		return -ENODEV;
 	}
 
 	if (pd->shared == NULL) {
-		printk(KERN_ERR "No mv643xx_eth_platform_data->shared\n");
+		dev_printk(KERN_ERR, &pdev->dev,
+			   "no mv643xx_eth_platform_data->shared\n");
 		return -ENODEV;
 	}
 
@@ -2090,145 +2082,80 @@ static int mv643xx_eth_probe(struct platform_device *pdev)
 	if (!dev)
 		return -ENOMEM;
 
-	platform_set_drvdata(pdev, dev);
-
 	mp = netdev_priv(dev);
+	platform_set_drvdata(pdev, mp);
+
+	mp->shared = platform_get_drvdata(pd->shared);
+	mp->port_num = pd->port_number;
+
 	mp->dev = dev;
 #ifdef MV643XX_ETH_NAPI
 	netif_napi_add(dev, &mp->napi, mv643xx_eth_poll, 64);
 #endif
 
+	set_params(mp, pd);
+
+	spin_lock_init(&mp->lock);
+
+	mib_counters_clear(mp);
+	INIT_WORK(&mp->tx_timeout_task, tx_timeout_task);
+
+	err = phy_init(mp, pd);
+	if (err)
+		goto out;
+	SET_ETHTOOL_OPS(dev, &mv643xx_eth_ethtool_ops);
+
+
 	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
 	BUG_ON(!res);
 	dev->irq = res->start;
 
+	dev->hard_start_xmit = mv643xx_eth_xmit;
 	dev->open = mv643xx_eth_open;
 	dev->stop = mv643xx_eth_stop;
-	dev->hard_start_xmit = mv643xx_eth_start_xmit;
-	dev->set_mac_address = mv643xx_eth_set_mac_address;
 	dev->set_multicast_list = mv643xx_eth_set_rx_mode;
-
-	/* No need to Tx Timeout */
+	dev->set_mac_address = mv643xx_eth_set_mac_address;
+	dev->do_ioctl = mv643xx_eth_ioctl;
+	dev->change_mtu = mv643xx_eth_change_mtu;
 	dev->tx_timeout = mv643xx_eth_tx_timeout;
-
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	dev->poll_controller = mv643xx_eth_netpoll;
 #endif
-
 	dev->watchdog_timeo = 2 * HZ;
 	dev->base_addr = 0;
-	dev->change_mtu = mv643xx_eth_change_mtu;
-	dev->do_ioctl = mv643xx_eth_do_ioctl;
-	SET_ETHTOOL_OPS(dev, &mv643xx_eth_ethtool_ops);
 
 #ifdef MV643XX_ETH_CHECKSUM_OFFLOAD_TX
-#ifdef MAX_SKB_FRAGS
 	/*
 	 * Zero copy can only work if we use Discovery II memory. Else, we will
 	 * have to map the buffers to ISA memory which is only 16 MB
 	 */
 	dev->features = NETIF_F_SG | NETIF_F_IP_CSUM;
-#endif
 #endif
 
-	/* Configure the timeout task */
-	INIT_WORK(&mp->tx_timeout_task, mv643xx_eth_tx_timeout_task);
-
-	spin_lock_init(&mp->lock);
-
-	mp->shared = platform_get_drvdata(pd->shared);
-	port_num = mp->port_num = pd->port_number;
+	SET_NETDEV_DEV(dev, &pdev->dev);
 
 	if (mp->shared->win_protect)
-		wrl(mp, WINDOW_PROTECT(port_num), mp->shared->win_protect);
-
-	mp->shared_smi = mp->shared;
-	if (pd->shared_smi != NULL)
-		mp->shared_smi = platform_get_drvdata(pd->shared_smi);
-
-	/* set default config values */
-	uc_addr_get(mp, dev->dev_addr);
-
-	if (is_valid_ether_addr(pd->mac_addr))
-		memcpy(dev->dev_addr, pd->mac_addr, 6);
-
-	if (pd->phy_addr || pd->force_phy_addr)
-		phy_addr_set(mp, pd->phy_addr);
-
-	mp->default_rx_ring_size = DEFAULT_RX_QUEUE_SIZE;
-	if (pd->rx_queue_size)
-		mp->default_rx_ring_size = pd->rx_queue_size;
-
-	mp->default_tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
-	if (pd->tx_queue_size)
-		mp->default_tx_ring_size = pd->tx_queue_size;
-
-	if (pd->tx_sram_size) {
-		mp->tx_desc_sram_size = pd->tx_sram_size;
-		mp->tx_desc_sram_addr = pd->tx_sram_addr;
-	}
-
-	if (pd->rx_sram_size) {
-		mp->rx_desc_sram_addr = pd->rx_sram_addr;
-		mp->rx_desc_sram_size = pd->rx_sram_size;
-	}
-
-	duplex = pd->duplex;
-	speed = pd->speed;
+		wrl(mp, WINDOW_PROTECT(mp->port_num), mp->shared->win_protect);
 
-	/* Hook up MII support for ethtool */
-	mp->mii.dev = dev;
-	mp->mii.mdio_read = mv643xx_eth_mdio_read;
-	mp->mii.mdio_write = mv643xx_eth_mdio_write;
-	mp->mii.phy_id = phy_addr_get(mp);
-	mp->mii.phy_id_mask = 0x3f;
-	mp->mii.reg_num_mask = 0x1f;
-
-	err = phy_detect(mp);
-	if (err) {
-		pr_debug("%s: No PHY detected at addr %d\n",
-				dev->name, phy_addr_get(mp));
-		goto out;
-	}
-
-	phy_reset(mp);
-	mp->mii.supports_gmii = mii_check_gmii_support(&mp->mii);
-	mv643xx_init_ethtool_cmd(dev, mp->mii.phy_id, speed, duplex, &cmd);
-	mv643xx_eth_update_pscr(mp, &cmd);
-	mv643xx_eth_set_settings(dev, &cmd);
-
-	SET_NETDEV_DEV(dev, &pdev->dev);
 	err = register_netdev(dev);
 	if (err)
 		goto out;
 
-	p = dev->dev_addr;
-	printk(KERN_NOTICE
-		"%s: port %d with MAC address %s\n",
-		dev->name, port_num, print_mac(mac, p));
+	dev_printk(KERN_NOTICE, &dev->dev, "port %d with MAC address %s\n",
+		   mp->port_num, print_mac(mac, dev->dev_addr));
 
 	if (dev->features & NETIF_F_SG)
-		printk(KERN_NOTICE "%s: Scatter Gather Enabled\n", dev->name);
+		dev_printk(KERN_NOTICE, &dev->dev, "scatter/gather enabled\n");
 
 	if (dev->features & NETIF_F_IP_CSUM)
-		printk(KERN_NOTICE "%s: TX TCP/IP Checksumming Supported\n",
-								dev->name);
-
-#ifdef MV643XX_ETH_CHECKSUM_OFFLOAD_TX
-	printk(KERN_NOTICE "%s: RX TCP/UDP Checksum Offload ON \n", dev->name);
-#endif
-
-#ifdef MV643XX_ETH_COAL
-	printk(KERN_NOTICE "%s: TX and RX Interrupt Coalescing ON \n",
-								dev->name);
-#endif
+		dev_printk(KERN_NOTICE, &dev->dev, "tx checksum offload\n");
 
 #ifdef MV643XX_ETH_NAPI
-	printk(KERN_NOTICE "%s: RX NAPI Enabled \n", dev->name);
+	dev_printk(KERN_NOTICE, &dev->dev, "napi enabled\n");
 #endif
 
 	if (mp->tx_desc_sram_size > 0)
-		printk(KERN_NOTICE "%s: Using SRAM\n", dev->name);
+		dev_printk(KERN_NOTICE, &dev->dev, "configured with sram\n");
 
 	return 0;
 
@@ -2240,35 +2167,35 @@ out:
 
 static int mv643xx_eth_remove(struct platform_device *pdev)
 {
-	struct net_device *dev = platform_get_drvdata(pdev);
+	struct mv643xx_eth_private *mp = platform_get_drvdata(pdev);
 
-	unregister_netdev(dev);
+	unregister_netdev(mp->dev);
 	flush_scheduled_work();
+	free_netdev(mp->dev);
 
-	free_netdev(dev);
 	platform_set_drvdata(pdev, NULL);
+
 	return 0;
 }
 
 static void mv643xx_eth_shutdown(struct platform_device *pdev)
 {
-	struct net_device *dev = platform_get_drvdata(pdev);
-	struct mv643xx_eth_private *mp = netdev_priv(dev);
-	unsigned int port_num = mp->port_num;
+	struct mv643xx_eth_private *mp = platform_get_drvdata(pdev);
 
 	/* Mask all interrupts on ethernet port */
-	wrl(mp, INT_MASK(port_num), 0);
-	rdl(mp, INT_MASK(port_num));
+	wrl(mp, INT_MASK(mp->port_num), 0);
+	rdl(mp, INT_MASK(mp->port_num));
 
-	port_reset(mp);
+	if (netif_running(mp->dev))
+		port_reset(mp);
 }
 
 static struct platform_driver mv643xx_eth_driver = {
-	.probe = mv643xx_eth_probe,
-	.remove = mv643xx_eth_remove,
-	.shutdown = mv643xx_eth_shutdown,
+	.probe		= mv643xx_eth_probe,
+	.remove		= mv643xx_eth_remove,
+	.shutdown	= mv643xx_eth_shutdown,
 	.driver = {
-		.name = MV643XX_ETH_NAME,
+		.name	= MV643XX_ETH_NAME,
 		.owner	= THIS_MODULE,
 	},
 };
@@ -2283,21 +2210,21 @@ static int __init mv643xx_eth_init_module(void)
 		if (rc)
 			platform_driver_unregister(&mv643xx_eth_shared_driver);
 	}
+
 	return rc;
 }
+module_init(mv643xx_eth_init_module);
 
 static void __exit mv643xx_eth_cleanup_module(void)
 {
 	platform_driver_unregister(&mv643xx_eth_driver);
 	platform_driver_unregister(&mv643xx_eth_shared_driver);
 }
-
-module_init(mv643xx_eth_init_module);
 module_exit(mv643xx_eth_cleanup_module);
 
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR(	"Rabeeh Khoury, Assaf Hoffman, Matthew Dharm, Manish Lachwani"
-		" and Dale Farnsworth");
+MODULE_AUTHOR("Rabeeh Khoury, Assaf Hoffman, Matthew Dharm, Manish Lachwani "
+	      "and Dale Farnsworth");
 MODULE_DESCRIPTION("Ethernet driver for Marvell MV643XX");
-MODULE_ALIAS("platform:" MV643XX_ETH_NAME);
+MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:" MV643XX_ETH_SHARED_NAME);
+MODULE_ALIAS("platform:" MV643XX_ETH_NAME);
diff --git a/include/linux/mv643xx_eth.h b/include/linux/mv643xx_eth.h
index a15cdd4a8e58..646177660495 100644
--- a/include/linux/mv643xx_eth.h
+++ b/include/linux/mv643xx_eth.h
@@ -17,30 +17,53 @@
 
 struct mv643xx_eth_shared_platform_data {
 	struct mbus_dram_target_info	*dram;
-	unsigned int	t_clk;
+	unsigned int		t_clk;
 };
 
 struct mv643xx_eth_platform_data {
+	/*
+	 * Pointer back to our parent instance, and our port number.
+	 */
 	struct platform_device	*shared;
-	int		port_number;
+	int			port_number;
 
+	/*
+	 * Whether a PHY is present, and if yes, at which address.
+	 */
 	struct platform_device	*shared_smi;
+	int			force_phy_addr;
+	int			phy_addr;
 
-	u16		force_phy_addr;	/* force override if phy_addr == 0 */
-	u16		phy_addr;
-
-	/* If speed is 0, then speed and duplex are autonegotiated. */
-	int		speed;		/* 0, SPEED_10, SPEED_100, SPEED_1000 */
-	int		duplex;		/* DUPLEX_HALF or DUPLEX_FULL */
-
-	/* non-zero values of the following fields override defaults */
-	u32		tx_queue_size;
-	u32		rx_queue_size;
-	u32		tx_sram_addr;
-	u32		tx_sram_size;
-	u32		rx_sram_addr;
-	u32		rx_sram_size;
-	u8		mac_addr[6];	/* mac address if non-zero*/
+	/*
+	 * Use this MAC address if it is valid, overriding the
+	 * address that is already in the hardware.
+	 */
+	u8			mac_addr[6];
+
+	/*
+	 * If speed is 0, autonegotiation is enabled.
+	 *   Valid values for speed: 0, SPEED_10, SPEED_100, SPEED_1000.
+	 *   Valid values for duplex: DUPLEX_HALF, DUPLEX_FULL.
+	 */
+	int			speed;
+	int			duplex;
+
+	/*
+	 * Override default RX/TX queue sizes if nonzero.
+	 */
+	int			rx_queue_size;
+	int			tx_queue_size;
+
+	/*
+	 * Use on-chip SRAM for RX/TX descriptors if size is nonzero
+	 * and sufficient to contain all descriptors for the requested
+	 * ring sizes.
+	 */
+	unsigned long		rx_sram_addr;
+	int			rx_sram_size;
+	unsigned long		tx_sram_addr;
+	int			tx_sram_size;
 };
 
-#endif /* __LINUX_MV643XX_ETH_H */
+
+#endif
-- 
cgit v1.2.3


From 64da80a29c7455321a7df7b47e27d639e3944c1a Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 2 Jun 2008 01:01:26 +0200
Subject: mv643xx_eth: allow multiple RX queues

Allow the platform code to specify that we are running on hardware
that is capable of supporting multiple RX queues.  If this option
is used, initialise all of the given RX queues instead of just RX
queue zero.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
---
 drivers/net/mv643xx_eth.c   | 99 +++++++++++++++++++++++++++++++++------------
 include/linux/mv643xx_eth.h |  5 +++
 2 files changed, 79 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index 9ce7be09e295..3c8591853999 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -96,7 +96,7 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define TX_BW_MTU(p)			(0x0458 + ((p) << 10))
 #define TX_BW_BURST(p)			(0x045c + ((p) << 10))
 #define INT_CAUSE(p)			(0x0460 + ((p) << 10))
-#define  INT_RX				0x00000804
+#define  INT_RX				0x0007fbfc
 #define  INT_EXT			0x00000002
 #define INT_CAUSE_EXT(p)		(0x0464 + ((p) << 10))
 #define  INT_EXT_LINK			0x00100000
@@ -107,7 +107,7 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define INT_MASK(p)			(0x0468 + ((p) << 10))
 #define INT_MASK_EXT(p)			(0x046c + ((p) << 10))
 #define TX_FIFO_URGENT_THRESHOLD(p)	(0x0474 + ((p) << 10))
-#define RXQ_CURRENT_DESC_PTR(p)		(0x060c + ((p) << 10))
+#define RXQ_CURRENT_DESC_PTR(p, q)	(0x060c + ((p) << 10) + ((q) << 4))
 #define RXQ_COMMAND(p)			(0x0680 + ((p) << 10))
 #define TXQ_CURRENT_DESC_PTR(p)		(0x06c0 + ((p) << 10))
 #define TXQ_BW_TOKENS(p)		(0x0700 + ((p) << 10))
@@ -286,6 +286,8 @@ struct mib_counters {
 };
 
 struct rx_queue {
+	int index;
+
 	int rx_ring_size;
 
 	int rx_desc_count;
@@ -334,8 +336,10 @@ struct mv643xx_eth_private {
 	int default_rx_ring_size;
 	unsigned long rx_desc_sram_addr;
 	int rx_desc_sram_size;
+	u8 rxq_mask;
+	int rxq_primary;
 	struct napi_struct napi;
-	struct rx_queue rxq[1];
+	struct rx_queue rxq[8];
 
 	/*
 	 * TX state.
@@ -365,7 +369,7 @@ static inline void wrl(struct mv643xx_eth_private *mp, int offset, u32 data)
 /* rxq/txq helper functions *************************************************/
 static struct mv643xx_eth_private *rxq_to_mp(struct rx_queue *rxq)
 {
-	return container_of(rxq, struct mv643xx_eth_private, rxq[0]);
+	return container_of(rxq, struct mv643xx_eth_private, rxq[rxq->index]);
 }
 
 static struct mv643xx_eth_private *txq_to_mp(struct tx_queue *txq)
@@ -376,13 +380,13 @@ static struct mv643xx_eth_private *txq_to_mp(struct tx_queue *txq)
 static void rxq_enable(struct rx_queue *rxq)
 {
 	struct mv643xx_eth_private *mp = rxq_to_mp(rxq);
-	wrl(mp, RXQ_COMMAND(mp->port_num), 1);
+	wrl(mp, RXQ_COMMAND(mp->port_num), 1 << rxq->index);
 }
 
 static void rxq_disable(struct rx_queue *rxq)
 {
 	struct mv643xx_eth_private *mp = rxq_to_mp(rxq);
-	u8 mask = 1;
+	u8 mask = 1 << rxq->index;
 
 	wrl(mp, RXQ_COMMAND(mp->port_num), mask << 8);
 	while (rdl(mp, RXQ_COMMAND(mp->port_num)) & mask)
@@ -583,6 +587,7 @@ static int mv643xx_eth_poll(struct napi_struct *napi, int budget)
 {
 	struct mv643xx_eth_private *mp;
 	int rx;
+	int i;
 
 	mp = container_of(napi, struct mv643xx_eth_private, napi);
 
@@ -593,7 +598,10 @@ static int mv643xx_eth_poll(struct napi_struct *napi, int budget)
 	}
 #endif
 
-	rx = rxq_process(mp->rxq, budget);
+	rx = 0;
+	for (i = 7; rx < budget && i >= 0; i--)
+		if (mp->rxq_mask & (1 << i))
+			rx += rxq_process(mp->rxq + i, budget - rx);
 
 	if (rx < budget) {
 		netif_rx_complete(mp->dev, napi);
@@ -1306,13 +1314,15 @@ static void mv643xx_eth_set_rx_mode(struct net_device *dev)
 
 
 /* rx/tx queue initialisation ***********************************************/
-static int rxq_init(struct mv643xx_eth_private *mp)
+static int rxq_init(struct mv643xx_eth_private *mp, int index)
 {
-	struct rx_queue *rxq = mp->rxq;
+	struct rx_queue *rxq = mp->rxq + index;
 	struct rx_desc *rx_desc;
 	int size;
 	int i;
 
+	rxq->index = index;
+
 	rxq->rx_ring_size = mp->default_rx_ring_size;
 
 	rxq->rx_desc_count = 0;
@@ -1321,7 +1331,7 @@ static int rxq_init(struct mv643xx_eth_private *mp)
 
 	size = rxq->rx_ring_size * sizeof(struct rx_desc);
 
-	if (size <= mp->rx_desc_sram_size) {
+	if (index == mp->rxq_primary && size <= mp->rx_desc_sram_size) {
 		rxq->rx_desc_area = ioremap(mp->rx_desc_sram_addr,
 						mp->rx_desc_sram_size);
 		rxq->rx_desc_dma = mp->rx_desc_sram_addr;
@@ -1362,7 +1372,7 @@ static int rxq_init(struct mv643xx_eth_private *mp)
 
 
 out_free:
-	if (size <= mp->rx_desc_sram_size)
+	if (index == mp->rxq_primary && size <= mp->rx_desc_sram_size)
 		iounmap(rxq->rx_desc_area);
 	else
 		dma_free_coherent(NULL, size,
@@ -1395,7 +1405,8 @@ static void rxq_deinit(struct rx_queue *rxq)
 			   rxq->rx_desc_count);
 	}
 
-	if (rxq->rx_desc_area_size <= mp->rx_desc_sram_size)
+	if (rxq->index == mp->rxq_primary &&
+	    rxq->rx_desc_area_size <= mp->rx_desc_sram_size)
 		iounmap(rxq->rx_desc_area);
 	else
 		dma_free_coherent(NULL, rxq->rx_desc_area_size,
@@ -1612,6 +1623,9 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 		}
 	}
 
+	/*
+	 * RxBuffer or RxError set for any of the 8 queues?
+	 */
 #ifdef MV643XX_ETH_NAPI
 	if (int_cause & INT_RX) {
 		wrl(mp, INT_MASK(mp->port_num), 0x00000000);
@@ -1620,8 +1634,13 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 		netif_rx_schedule(dev, &mp->napi);
 	}
 #else
-	if (int_cause & INT_RX)
-		rxq_process(mp->rxq, INT_MAX);
+	if (int_cause & INT_RX) {
+		int i;
+
+		for (i = 7; i >= 0; i--)
+			if (mp->rxq_mask & (1 << i))
+				rxq_process(mp->rxq + i, INT_MAX);
+	}
 #endif
 
 	if (int_cause_ext & INT_EXT_TX) {
@@ -1707,13 +1726,16 @@ static void port_start(struct mv643xx_eth_private *mp)
 	wrl(mp, PORT_CONFIG_EXT(mp->port_num), 0x00000000);
 
 	/*
-	 * Enable the receive queue.
+	 * Enable the receive queues.
 	 */
-	for (i = 0; i < 1; i++) {
-		struct rx_queue *rxq = mp->rxq;
-		int off = RXQ_CURRENT_DESC_PTR(mp->port_num);
+	for (i = 0; i < 8; i++) {
+		struct rx_queue *rxq = mp->rxq + i;
+		int off = RXQ_CURRENT_DESC_PTR(mp->port_num, i);
 		u32 addr;
 
+		if ((mp->rxq_mask & (1 << i)) == 0)
+			continue;
+
 		addr = (u32)rxq->rx_desc_dma;
 		addr += rxq->rx_curr_desc * sizeof(struct rx_desc);
 		wrl(mp, off, addr);
@@ -1748,6 +1770,7 @@ static int mv643xx_eth_open(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
 	int err;
+	int i;
 
 	wrl(mp, INT_CAUSE(mp->port_num), 0);
 	wrl(mp, INT_CAUSE_EXT(mp->port_num), 0);
@@ -1763,10 +1786,20 @@ static int mv643xx_eth_open(struct net_device *dev)
 
 	init_mac_tables(mp);
 
-	err = rxq_init(mp);
-	if (err)
-		goto out;
-	rxq_refill(mp->rxq);
+	for (i = 0; i < 8; i++) {
+		if ((mp->rxq_mask & (1 << i)) == 0)
+			continue;
+
+		err = rxq_init(mp, i);
+		if (err) {
+			while (--i >= 0)
+				if (mp->rxq_mask & (1 << i))
+					rxq_deinit(mp->rxq + i);
+			goto out;
+		}
+
+		rxq_refill(mp->rxq + i);
+	}
 
 	err = txq_init(mp);
 	if (err)
@@ -1790,7 +1823,9 @@ static int mv643xx_eth_open(struct net_device *dev)
 
 
 out_free:
-	rxq_deinit(mp->rxq);
+	for (i = 0; i < 8; i++)
+		if (mp->rxq_mask & (1 << i))
+			rxq_deinit(mp->rxq + i);
 out:
 	free_irq(dev->irq, dev);
 
@@ -1800,9 +1835,13 @@ out:
 static void port_reset(struct mv643xx_eth_private *mp)
 {
 	unsigned int data;
+	int i;
 
+	for (i = 0; i < 8; i++) {
+		if (mp->rxq_mask & (1 << i))
+			rxq_disable(mp->rxq + i);
+	}
 	txq_disable(mp->txq);
-	rxq_disable(mp->rxq);
 	while (!(rdl(mp, PORT_STATUS(mp->port_num)) & TX_FIFO_EMPTY))
 		udelay(10);
 
@@ -1817,6 +1856,7 @@ static void port_reset(struct mv643xx_eth_private *mp)
 static int mv643xx_eth_stop(struct net_device *dev)
 {
 	struct mv643xx_eth_private *mp = netdev_priv(dev);
+	int i;
 
 	wrl(mp, INT_MASK(mp->port_num), 0x00000000);
 	rdl(mp, INT_MASK(mp->port_num));
@@ -1832,8 +1872,11 @@ static int mv643xx_eth_stop(struct net_device *dev)
 	port_reset(mp);
 	mib_counters_update(mp);
 
+	for (i = 0; i < 8; i++) {
+		if (mp->rxq_mask & (1 << i))
+			rxq_deinit(mp->rxq + i);
+	}
 	txq_deinit(mp->txq);
-	rxq_deinit(mp->rxq);
 
 	return 0;
 }
@@ -2085,6 +2128,12 @@ static void set_params(struct mv643xx_eth_private *mp,
 	mp->rx_desc_sram_addr = pd->rx_sram_addr;
 	mp->rx_desc_sram_size = pd->rx_sram_size;
 
+	if (pd->rx_queue_mask)
+		mp->rxq_mask = pd->rx_queue_mask;
+	else
+		mp->rxq_mask = 0x01;
+	mp->rxq_primary = fls(mp->rxq_mask) - 1;
+
 	mp->default_tx_ring_size = DEFAULT_TX_QUEUE_SIZE;
 	if (pd->tx_queue_size)
 		mp->default_tx_ring_size = pd->tx_queue_size;
diff --git a/include/linux/mv643xx_eth.h b/include/linux/mv643xx_eth.h
index 646177660495..1afd7ba6d303 100644
--- a/include/linux/mv643xx_eth.h
+++ b/include/linux/mv643xx_eth.h
@@ -48,6 +48,11 @@ struct mv643xx_eth_platform_data {
 	int			speed;
 	int			duplex;
 
+	/*
+	 * Which RX queues to use.
+	 */
+	int			rx_queue_mask;
+
 	/*
 	 * Override default RX/TX queue sizes if nonzero.
 	 */
-- 
cgit v1.2.3


From 3d6b35bc5090cf8d8b7e62eca1f9c21ca56fc6c7 Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Mon, 2 Jun 2008 01:28:22 +0200
Subject: mv643xx_eth: allow multiple TX queues

As with the multiple RX queue support, allow the platform code to
specify that the hardware we are running on supports multiple TX
queues.  This patch only uses the highest-numbered enabled queue
to send packets to for now, this can be extended later to enable
QoS and such.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Dale Farnsworth <dale@farnsworth.org>
---
 drivers/net/mv643xx_eth.c   | 146 +++++++++++++++++++++++++++++++-------------
 include/linux/mv643xx_eth.h |   3 +-
 2 files changed, 105 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index 3c8591853999..287155ea9ce1 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -103,16 +103,16 @@ static char mv643xx_eth_driver_version[] = "1.0";
 #define  INT_EXT_PHY			0x00010000
 #define  INT_EXT_TX_ERROR_0		0x00000100
 #define  INT_EXT_TX_0			0x00000001
-#define  INT_EXT_TX			0x00000101
+#define  INT_EXT_TX			0x0000ffff
 #define INT_MASK(p)			(0x0468 + ((p) << 10))
 #define INT_MASK_EXT(p)			(0x046c + ((p) << 10))
 #define TX_FIFO_URGENT_THRESHOLD(p)	(0x0474 + ((p) << 10))
 #define RXQ_CURRENT_DESC_PTR(p, q)	(0x060c + ((p) << 10) + ((q) << 4))
 #define RXQ_COMMAND(p)			(0x0680 + ((p) << 10))
-#define TXQ_CURRENT_DESC_PTR(p)		(0x06c0 + ((p) << 10))
-#define TXQ_BW_TOKENS(p)		(0x0700 + ((p) << 10))
-#define TXQ_BW_CONF(p)			(0x0704 + ((p) << 10))
-#define TXQ_BW_WRR_CONF(p)		(0x0708 + ((p) << 10))
+#define TXQ_CURRENT_DESC_PTR(p, q)	(0x06c0 + ((p) << 10) + ((q) << 2))
+#define TXQ_BW_TOKENS(p, q)		(0x0700 + ((p) << 10) + ((q) << 4))
+#define TXQ_BW_CONF(p, q)		(0x0704 + ((p) << 10) + ((q) << 4))
+#define TXQ_BW_WRR_CONF(p, q)		(0x0708 + ((p) << 10) + ((q) << 4))
 #define MIB_COUNTERS(p)			(0x1000 + ((p) << 7))
 #define SPECIAL_MCAST_TABLE(p)		(0x1400 + ((p) << 10))
 #define OTHER_MCAST_TABLE(p)		(0x1500 + ((p) << 10))
@@ -303,6 +303,8 @@ struct rx_queue {
 };
 
 struct tx_queue {
+	int index;
+
 	int tx_ring_size;
 
 	int tx_desc_count;
@@ -347,7 +349,9 @@ struct mv643xx_eth_private {
 	int default_tx_ring_size;
 	unsigned long tx_desc_sram_addr;
 	int tx_desc_sram_size;
-	struct tx_queue txq[1];
+	u8 txq_mask;
+	int txq_primary;
+	struct tx_queue txq[8];
 #ifdef MV643XX_ETH_TX_FAST_REFILL
 	int tx_clean_threshold;
 #endif
@@ -374,7 +378,7 @@ static struct mv643xx_eth_private *rxq_to_mp(struct rx_queue *rxq)
 
 static struct mv643xx_eth_private *txq_to_mp(struct tx_queue *txq)
 {
-	return container_of(txq, struct mv643xx_eth_private, txq[0]);
+	return container_of(txq, struct mv643xx_eth_private, txq[txq->index]);
 }
 
 static void rxq_enable(struct rx_queue *rxq)
@@ -396,13 +400,13 @@ static void rxq_disable(struct rx_queue *rxq)
 static void txq_enable(struct tx_queue *txq)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
-	wrl(mp, TXQ_COMMAND(mp->port_num), 1);
+	wrl(mp, TXQ_COMMAND(mp->port_num), 1 << txq->index);
 }
 
 static void txq_disable(struct tx_queue *txq)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
-	u8 mask = 1;
+	u8 mask = 1 << txq->index;
 
 	wrl(mp, TXQ_COMMAND(mp->port_num), mask << 8);
 	while (rdl(mp, TXQ_COMMAND(mp->port_num)) & mask)
@@ -413,6 +417,12 @@ static void __txq_maybe_wake(struct tx_queue *txq)
 {
 	struct mv643xx_eth_private *mp = txq_to_mp(txq);
 
+	/*
+	 * netif_{stop,wake}_queue() flow control only applies to
+	 * the primary queue.
+	 */
+	BUG_ON(txq->index != mp->txq_primary);
+
 	if (txq->tx_ring_size - txq->tx_desc_count >= MAX_DESCS_PER_SKB)
 		netif_wake_queue(mp->dev);
 }
@@ -593,8 +603,10 @@ static int mv643xx_eth_poll(struct napi_struct *napi, int budget)
 
 #ifdef MV643XX_ETH_TX_FAST_REFILL
 	if (++mp->tx_clean_threshold > 5) {
-		txq_reclaim(mp->txq, 0);
 		mp->tx_clean_threshold = 0;
+		for (i = 0; i < 8; i++)
+			if (mp->txq_mask & (1 << i))
+				txq_reclaim(mp->txq + i, 0);
 	}
 #endif
 
@@ -754,8 +766,6 @@ static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct tx_queue *txq;
 	unsigned long flags;
 
-	BUG_ON(netif_queue_stopped(dev));
-
 	if (has_tiny_unaligned_frags(skb) && __skb_linearize(skb)) {
 		stats->tx_dropped++;
 		dev_printk(KERN_DEBUG, &dev->dev,
@@ -766,13 +776,15 @@ static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	spin_lock_irqsave(&mp->lock, flags);
 
-	txq = mp->txq;
+	txq = mp->txq + mp->txq_primary;
 
 	if (txq->tx_ring_size - txq->tx_desc_count < MAX_DESCS_PER_SKB) {
-		printk(KERN_ERR "%s: transmit with queue full\n", dev->name);
-		netif_stop_queue(dev);
 		spin_unlock_irqrestore(&mp->lock, flags);
-		return NETDEV_TX_BUSY;
+		if (txq->index == mp->txq_primary && net_ratelimit())
+			dev_printk(KERN_ERR, &dev->dev,
+				   "primary tx queue full?!\n");
+		kfree_skb(skb);
+		return NETDEV_TX_OK;
 	}
 
 	txq_submit_skb(txq, skb);
@@ -780,8 +792,13 @@ static int mv643xx_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 	stats->tx_packets++;
 	dev->trans_start = jiffies;
 
-	if (txq->tx_ring_size - txq->tx_desc_count < MAX_DESCS_PER_SKB)
-		netif_stop_queue(dev);
+	if (txq->index == mp->txq_primary) {
+		int entries_left;
+
+		entries_left = txq->tx_ring_size - txq->tx_desc_count;
+		if (entries_left < MAX_DESCS_PER_SKB)
+			netif_stop_queue(dev);
+	}
 
 	spin_unlock_irqrestore(&mp->lock, flags);
 
@@ -831,8 +848,8 @@ static void txq_set_rate(struct tx_queue *txq, int rate, int burst)
 	if (bucket_size > 65535)
 		bucket_size = 65535;
 
-	wrl(mp, TXQ_BW_TOKENS(mp->port_num), token_rate << 14);
-	wrl(mp, TXQ_BW_CONF(mp->port_num),
+	wrl(mp, TXQ_BW_TOKENS(mp->port_num, txq->index), token_rate << 14);
+	wrl(mp, TXQ_BW_CONF(mp->port_num, txq->index),
 			(bucket_size << 10) | token_rate);
 }
 
@@ -848,7 +865,7 @@ static void txq_set_fixed_prio_mode(struct tx_queue *txq)
 	off = TXQ_FIX_PRIO_CONF(mp->port_num);
 
 	val = rdl(mp, off);
-	val |= 1;
+	val |= 1 << txq->index;
 	wrl(mp, off, val);
 }
 
@@ -864,13 +881,13 @@ static void txq_set_wrr(struct tx_queue *txq, int weight)
 	off = TXQ_FIX_PRIO_CONF(mp->port_num);
 
 	val = rdl(mp, off);
-	val &= ~1;
+	val &= ~(1 << txq->index);
 	wrl(mp, off, val);
 
 	/*
 	 * Configure WRR weight for this queue.
 	 */
-	off = TXQ_BW_WRR_CONF(mp->port_num);
+	off = TXQ_BW_WRR_CONF(mp->port_num, txq->index);
 
 	val = rdl(mp, off);
 	val = (val & ~0xff) | (weight & 0xff);
@@ -1415,13 +1432,15 @@ static void rxq_deinit(struct rx_queue *rxq)
 	kfree(rxq->rx_skb);
 }
 
-static int txq_init(struct mv643xx_eth_private *mp)
+static int txq_init(struct mv643xx_eth_private *mp, int index)
 {
-	struct tx_queue *txq = mp->txq;
+	struct tx_queue *txq = mp->txq + index;
 	struct tx_desc *tx_desc;
 	int size;
 	int i;
 
+	txq->index = index;
+
 	txq->tx_ring_size = mp->default_tx_ring_size;
 
 	txq->tx_desc_count = 0;
@@ -1430,7 +1449,7 @@ static int txq_init(struct mv643xx_eth_private *mp)
 
 	size = txq->tx_ring_size * sizeof(struct tx_desc);
 
-	if (size <= mp->tx_desc_sram_size) {
+	if (index == mp->txq_primary && size <= mp->tx_desc_sram_size) {
 		txq->tx_desc_area = ioremap(mp->tx_desc_sram_addr,
 						mp->tx_desc_sram_size);
 		txq->tx_desc_dma = mp->tx_desc_sram_addr;
@@ -1467,7 +1486,7 @@ static int txq_init(struct mv643xx_eth_private *mp)
 
 
 out_free:
-	if (size <= mp->tx_desc_sram_size)
+	if (index == mp->txq_primary && size <= mp->tx_desc_sram_size)
 		iounmap(txq->tx_desc_area);
 	else
 		dma_free_coherent(NULL, size,
@@ -1539,7 +1558,8 @@ static void txq_deinit(struct tx_queue *txq)
 
 	BUG_ON(txq->tx_used_desc != txq->tx_curr_desc);
 
-	if (txq->tx_desc_area_size <= mp->tx_desc_sram_size)
+	if (txq->index == mp->txq_primary &&
+	    txq->tx_desc_area_size <= mp->tx_desc_sram_size)
 		iounmap(txq->tx_desc_area);
 	else
 		dma_free_coherent(NULL, txq->tx_desc_area_size,
@@ -1578,12 +1598,20 @@ static void update_pscr(struct mv643xx_eth_private *mp, int speed, int duplex)
 		if ((pscr_o & SERIAL_PORT_ENABLE) == 0)
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_n);
 		else {
-			txq_disable(mp->txq);
+			int i;
+
+			for (i = 0; i < 8; i++)
+				if (mp->txq_mask & (1 << i))
+					txq_disable(mp->txq + i);
+
 			pscr_o &= ~SERIAL_PORT_ENABLE;
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_o);
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_n);
 			wrl(mp, PORT_SERIAL_CONTROL(mp->port_num), pscr_n);
-			txq_enable(mp->txq);
+
+			for (i = 0; i < 8; i++)
+				if (mp->txq_mask & (1 << i))
+					txq_enable(mp->txq + i);
 		}
 	}
 }
@@ -1609,13 +1637,17 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 	if (int_cause_ext & (INT_EXT_PHY | INT_EXT_LINK)) {
 		if (mii_link_ok(&mp->mii)) {
 			struct ethtool_cmd cmd;
+			int i;
 
 			mii_ethtool_gset(&mp->mii, &cmd);
 			update_pscr(mp, cmd.speed, cmd.duplex);
-			txq_enable(mp->txq);
+			for (i = 0; i < 8; i++)
+				if (mp->txq_mask & (1 << i))
+					txq_enable(mp->txq + i);
+
 			if (!netif_carrier_ok(dev)) {
 				netif_carrier_on(dev);
-				__txq_maybe_wake(mp->txq);
+				__txq_maybe_wake(mp->txq + mp->txq_primary);
 			}
 		} else if (netif_carrier_ok(dev)) {
 			netif_stop_queue(dev);
@@ -1643,9 +1675,17 @@ static irqreturn_t mv643xx_eth_irq(int irq, void *dev_id)
 	}
 #endif
 
+	/*
+	 * TxBuffer or TxError set for any of the 8 queues?
+	 */
 	if (int_cause_ext & INT_EXT_TX) {
-		txq_reclaim(mp->txq, 0);
-		__txq_maybe_wake(mp->txq);
+		int i;
+
+		for (i = 0; i < 8; i++)
+			if (mp->txq_mask & (1 << i))
+				txq_reclaim(mp->txq + i, 0);
+
+		__txq_maybe_wake(mp->txq + mp->txq_primary);
 	}
 
 	return IRQ_HANDLED;
@@ -1696,11 +1736,14 @@ static void port_start(struct mv643xx_eth_private *mp)
 	 * Configure TX path and queues.
 	 */
 	tx_set_rate(mp, 1000000000, 16777216);
-	for (i = 0; i < 1; i++) {
-		struct tx_queue *txq = mp->txq;
-		int off = TXQ_CURRENT_DESC_PTR(mp->port_num);
+	for (i = 0; i < 8; i++) {
+		struct tx_queue *txq = mp->txq + i;
+		int off = TXQ_CURRENT_DESC_PTR(mp->port_num, i);
 		u32 addr;
 
+		if ((mp->txq_mask & (1 << i)) == 0)
+			continue;
+
 		addr = (u32)txq->tx_desc_dma;
 		addr += txq->tx_curr_desc * sizeof(struct tx_desc);
 		wrl(mp, off, addr);
@@ -1801,9 +1844,18 @@ static int mv643xx_eth_open(struct net_device *dev)
 		rxq_refill(mp->rxq + i);
 	}
 
-	err = txq_init(mp);
-	if (err)
-		goto out_free;
+	for (i = 0; i < 8; i++) {
+		if ((mp->txq_mask & (1 << i)) == 0)
+			continue;
+
+		err = txq_init(mp, i);
+		if (err) {
+			while (--i >= 0)
+				if (mp->txq_mask & (1 << i))
+					txq_deinit(mp->txq + i);
+			goto out_free;
+		}
+	}
 
 #ifdef MV643XX_ETH_NAPI
 	napi_enable(&mp->napi);
@@ -1840,8 +1892,9 @@ static void port_reset(struct mv643xx_eth_private *mp)
 	for (i = 0; i < 8; i++) {
 		if (mp->rxq_mask & (1 << i))
 			rxq_disable(mp->rxq + i);
+		if (mp->txq_mask & (1 << i))
+			txq_disable(mp->txq + i);
 	}
-	txq_disable(mp->txq);
 	while (!(rdl(mp, PORT_STATUS(mp->port_num)) & TX_FIFO_EMPTY))
 		udelay(10);
 
@@ -1875,8 +1928,9 @@ static int mv643xx_eth_stop(struct net_device *dev)
 	for (i = 0; i < 8; i++) {
 		if (mp->rxq_mask & (1 << i))
 			rxq_deinit(mp->rxq + i);
+		if (mp->txq_mask & (1 << i))
+			txq_deinit(mp->txq + i);
 	}
-	txq_deinit(mp->txq);
 
 	return 0;
 }
@@ -1928,7 +1982,7 @@ static void tx_timeout_task(struct work_struct *ugly)
 		port_reset(mp);
 		port_start(mp);
 
-		__txq_maybe_wake(mp->txq);
+		__txq_maybe_wake(mp->txq + mp->txq_primary);
 	}
 }
 
@@ -2139,6 +2193,12 @@ static void set_params(struct mv643xx_eth_private *mp,
 		mp->default_tx_ring_size = pd->tx_queue_size;
 	mp->tx_desc_sram_addr = pd->tx_sram_addr;
 	mp->tx_desc_sram_size = pd->tx_sram_size;
+
+	if (pd->tx_queue_mask)
+		mp->txq_mask = pd->tx_queue_mask;
+	else
+		mp->txq_mask = 0x01;
+	mp->txq_primary = fls(mp->txq_mask) - 1;
 }
 
 static int phy_detect(struct mv643xx_eth_private *mp)
diff --git a/include/linux/mv643xx_eth.h b/include/linux/mv643xx_eth.h
index 1afd7ba6d303..12078577aef6 100644
--- a/include/linux/mv643xx_eth.h
+++ b/include/linux/mv643xx_eth.h
@@ -49,9 +49,10 @@ struct mv643xx_eth_platform_data {
 	int			duplex;
 
 	/*
-	 * Which RX queues to use.
+	 * Which RX/TX queues to use.
 	 */
 	int			rx_queue_mask;
+	int			tx_queue_mask;
 
 	/*
 	 * Override default RX/TX queue sizes if nonzero.
-- 
cgit v1.2.3


From d8f3de0d2412bb91639cfefc5b3c79dbf3812212 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Thu, 12 Jun 2008 23:24:06 +0200
Subject: Suspend-related patches for 2.6.27

ACPI PM: Add possibility to change suspend sequence

There are some systems out there that don't work correctly with
our current suspend/hibernation code ordering.  Provide a workaround
for these systems allowing them to pass 'acpi_sleep=old_ordering' in
the kernel command line so that it will use the pre-ACPI 2.0 ("old")
suspend code ordering.

Unfortunately, this requires us to add a platform hook to the
resuming of devices for recovering the platform in case one of the
device drivers' .suspend() routines returns error code.  Namely,
ACPI 1.0 specifies that _PTS should be called before suspending
devices, but _WAK still should be called before resuming them in
order to undo the changes made by _PTS.  However, if there is an
error during suspending devices, they are automatically resumed
without returning control to the PM core, so the _WAK has to be
called from within device_resume() in that cases.

The patch also reorders and refactors the ACPI suspend/hibernation
code to avoid duplication as far as reasonably possible.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/kernel-parameters.txt |   6 +-
 arch/x86/kernel/acpi/sleep.c        |   2 +
 drivers/acpi/sleep/main.c           | 276 +++++++++++++++++++++---------------
 drivers/base/power/main.c           |   2 -
 include/linux/acpi.h                |   3 +
 include/linux/suspend.h             |  14 +-
 kernel/power/disk.c                 |  28 +++-
 kernel/power/main.c                 |  10 +-
 8 files changed, 215 insertions(+), 126 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 9cf7b34f2db0..18d793ea0dd3 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -147,10 +147,14 @@ and is between 256 and 4096 characters. It is defined in the file
 			default: 0
 
 	acpi_sleep=	[HW,ACPI] Sleep options
-			Format: { s3_bios, s3_mode, s3_beep }
+			Format: { s3_bios, s3_mode, s3_beep, old_ordering }
 			See Documentation/power/video.txt for s3_bios and s3_mode.
 			s3_beep is for debugging; it makes the PC's speaker beep
 			as soon as the kernel's real-mode entry point is called.
+			old_ordering causes the ACPI 1.0 ordering of the _PTS
+			control method, wrt putting devices into low power
+			states, to be enforced (the ACPI 2.0 ordering of _PTS is
+			used by default).
 
 	acpi_sci=	[HW,ACPI] ACPI System Control Interrupt trigger mode
 			Format: { level | edge | high | low }
diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index afc25ee9964b..882e970032d5 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -124,6 +124,8 @@ static int __init acpi_sleep_setup(char *str)
 			acpi_realmode_flags |= 2;
 		if (strncmp(str, "s3_beep", 7) == 0)
 			acpi_realmode_flags |= 4;
+		if (strncmp(str, "old_ordering", 12) == 0)
+			acpi_old_suspend_ordering();
 		str = strchr(str, ',');
 		if (str != NULL)
 			str += strspn(str, ", \t");
diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c
index 0f2caea2fc83..4addf8ad50ae 100644
--- a/drivers/acpi/sleep/main.c
+++ b/drivers/acpi/sleep/main.c
@@ -24,10 +24,6 @@
 
 u8 sleep_states[ACPI_S_STATE_COUNT];
 
-#ifdef CONFIG_PM_SLEEP
-static u32 acpi_target_sleep_state = ACPI_STATE_S0;
-#endif
-
 static int acpi_sleep_prepare(u32 acpi_state)
 {
 #ifdef CONFIG_ACPI_SLEEP
@@ -50,9 +46,96 @@ static int acpi_sleep_prepare(u32 acpi_state)
 	return 0;
 }
 
-#ifdef CONFIG_SUSPEND
-static struct platform_suspend_ops acpi_suspend_ops;
+#ifdef CONFIG_PM_SLEEP
+static u32 acpi_target_sleep_state = ACPI_STATE_S0;
+
+/*
+ * ACPI 1.0 wants us to execute _PTS before suspending devices, so we allow the
+ * user to request that behavior by using the 'acpi_old_suspend_ordering'
+ * kernel command line option that causes the following variable to be set.
+ */
+static bool old_suspend_ordering;
+
+void __init acpi_old_suspend_ordering(void)
+{
+	old_suspend_ordering = true;
+}
+
+/**
+ *	acpi_pm_disable_gpes - Disable the GPEs.
+ */
+static int acpi_pm_disable_gpes(void)
+{
+	acpi_hw_disable_all_gpes();
+	return 0;
+}
+
+/**
+ *	__acpi_pm_prepare - Prepare the platform to enter the target state.
+ *
+ *	If necessary, set the firmware waking vector and do arch-specific
+ *	nastiness to get the wakeup code to the waking vector.
+ */
+static int __acpi_pm_prepare(void)
+{
+	int error = acpi_sleep_prepare(acpi_target_sleep_state);
+
+	if (error)
+		acpi_target_sleep_state = ACPI_STATE_S0;
+	return error;
+}
+
+/**
+ *	acpi_pm_prepare - Prepare the platform to enter the target sleep
+ *		state and disable the GPEs.
+ */
+static int acpi_pm_prepare(void)
+{
+	int error = __acpi_pm_prepare();
+
+	if (!error)
+		acpi_hw_disable_all_gpes();
+	return error;
+}
+
+/**
+ *	acpi_pm_finish - Instruct the platform to leave a sleep state.
+ *
+ *	This is called after we wake back up (or if entering the sleep state
+ *	failed).
+ */
+static void acpi_pm_finish(void)
+{
+	u32 acpi_state = acpi_target_sleep_state;
+
+	if (acpi_state == ACPI_STATE_S0)
+		return;
 
+	printk(KERN_INFO PREFIX "Waking up from system sleep state S%d\n",
+		acpi_state);
+	acpi_disable_wakeup_device(acpi_state);
+	acpi_leave_sleep_state(acpi_state);
+
+	/* reset firmware waking vector */
+	acpi_set_firmware_waking_vector((acpi_physical_address) 0);
+
+	acpi_target_sleep_state = ACPI_STATE_S0;
+}
+
+/**
+ *	acpi_pm_end - Finish up suspend sequence.
+ */
+static void acpi_pm_end(void)
+{
+	/*
+	 * This is necessary in case acpi_pm_finish() is not called during a
+	 * failing transition to a sleep state.
+	 */
+	acpi_target_sleep_state = ACPI_STATE_S0;
+}
+#endif /* CONFIG_PM_SLEEP */
+
+#ifdef CONFIG_SUSPEND
 extern void do_suspend_lowlevel(void);
 
 static u32 acpi_suspend_states[] = {
@@ -66,7 +149,6 @@ static u32 acpi_suspend_states[] = {
  *	acpi_suspend_begin - Set the target system sleep state to the state
  *		associated with given @pm_state, if supported.
  */
-
 static int acpi_suspend_begin(suspend_state_t pm_state)
 {
 	u32 acpi_state = acpi_suspend_states[pm_state];
@@ -82,25 +164,6 @@ static int acpi_suspend_begin(suspend_state_t pm_state)
 	return error;
 }
 
-/**
- *	acpi_suspend_prepare - Do preliminary suspend work.
- *
- *	If necessary, set the firmware waking vector and do arch-specific
- *	nastiness to get the wakeup code to the waking vector.
- */
-
-static int acpi_suspend_prepare(void)
-{
-	int error = acpi_sleep_prepare(acpi_target_sleep_state);
-
-	if (error) {
-		acpi_target_sleep_state = ACPI_STATE_S0;
-		return error;
-	}
-
-	return ACPI_SUCCESS(acpi_hw_disable_all_gpes()) ? 0 : -EFAULT;
-}
-
 /**
  *	acpi_suspend_enter - Actually enter a sleep state.
  *	@pm_state: ignored
@@ -109,7 +172,6 @@ static int acpi_suspend_prepare(void)
  *	assembly, which in turn call acpi_enter_sleep_state().
  *	It's unfortunate, but it works. Please fix if you're feeling frisky.
  */
-
 static int acpi_suspend_enter(suspend_state_t pm_state)
 {
 	acpi_status status = AE_OK;
@@ -166,39 +228,6 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
 	return ACPI_SUCCESS(status) ? 0 : -EFAULT;
 }
 
-/**
- *	acpi_suspend_finish - Instruct the platform to leave a sleep state.
- *
- *	This is called after we wake back up (or if entering the sleep state
- *	failed). 
- */
-
-static void acpi_suspend_finish(void)
-{
-	u32 acpi_state = acpi_target_sleep_state;
-
-	acpi_disable_wakeup_device(acpi_state);
-	acpi_leave_sleep_state(acpi_state);
-
-	/* reset firmware waking vector */
-	acpi_set_firmware_waking_vector((acpi_physical_address) 0);
-
-	acpi_target_sleep_state = ACPI_STATE_S0;
-}
-
-/**
- *	acpi_suspend_end - Finish up suspend sequence.
- */
-
-static void acpi_suspend_end(void)
-{
-	/*
-	 * This is necessary in case acpi_suspend_finish() is not called during a
-	 * failing transition to a sleep state.
-	 */
-	acpi_target_sleep_state = ACPI_STATE_S0;
-}
-
 static int acpi_suspend_state_valid(suspend_state_t pm_state)
 {
 	u32 acpi_state;
@@ -218,10 +247,39 @@ static int acpi_suspend_state_valid(suspend_state_t pm_state)
 static struct platform_suspend_ops acpi_suspend_ops = {
 	.valid = acpi_suspend_state_valid,
 	.begin = acpi_suspend_begin,
-	.prepare = acpi_suspend_prepare,
+	.prepare = acpi_pm_prepare,
+	.enter = acpi_suspend_enter,
+	.finish = acpi_pm_finish,
+	.end = acpi_pm_end,
+};
+
+/**
+ *	acpi_suspend_begin_old - Set the target system sleep state to the
+ *		state associated with given @pm_state, if supported, and
+ *		execute the _PTS control method.  This function is used if the
+ *		pre-ACPI 2.0 suspend ordering has been requested.
+ */
+static int acpi_suspend_begin_old(suspend_state_t pm_state)
+{
+	int error = acpi_suspend_begin(pm_state);
+
+	if (!error)
+		error = __acpi_pm_prepare();
+	return error;
+}
+
+/*
+ * The following callbacks are used if the pre-ACPI 2.0 suspend ordering has
+ * been requested.
+ */
+static struct platform_suspend_ops acpi_suspend_ops_old = {
+	.valid = acpi_suspend_state_valid,
+	.begin = acpi_suspend_begin_old,
+	.prepare = acpi_pm_disable_gpes,
 	.enter = acpi_suspend_enter,
-	.finish = acpi_suspend_finish,
-	.end = acpi_suspend_end,
+	.finish = acpi_pm_finish,
+	.end = acpi_pm_end,
+	.recover = acpi_pm_finish,
 };
 #endif /* CONFIG_SUSPEND */
 
@@ -229,22 +287,9 @@ static struct platform_suspend_ops acpi_suspend_ops = {
 static int acpi_hibernation_begin(void)
 {
 	acpi_target_sleep_state = ACPI_STATE_S4;
-
 	return 0;
 }
 
-static int acpi_hibernation_prepare(void)
-{
-	int error = acpi_sleep_prepare(ACPI_STATE_S4);
-
-	if (error) {
-		acpi_target_sleep_state = ACPI_STATE_S0;
-		return error;
-	}
-
-	return ACPI_SUCCESS(acpi_hw_disable_all_gpes()) ? 0 : -EFAULT;
-}
-
 static int acpi_hibernation_enter(void)
 {
 	acpi_status status = AE_OK;
@@ -274,52 +319,55 @@ static void acpi_hibernation_leave(void)
 	acpi_leave_sleep_state_prep(ACPI_STATE_S4);
 }
 
-static void acpi_hibernation_finish(void)
+static void acpi_pm_enable_gpes(void)
 {
-	acpi_disable_wakeup_device(ACPI_STATE_S4);
-	acpi_leave_sleep_state(ACPI_STATE_S4);
-
-	/* reset firmware waking vector */
-	acpi_set_firmware_waking_vector((acpi_physical_address) 0);
-
-	acpi_target_sleep_state = ACPI_STATE_S0;
+	acpi_hw_enable_all_runtime_gpes();
 }
 
-static void acpi_hibernation_end(void)
-{
-	/*
-	 * This is necessary in case acpi_hibernation_finish() is not called
-	 * during a failing transition to the sleep state.
-	 */
-	acpi_target_sleep_state = ACPI_STATE_S0;
-}
+static struct platform_hibernation_ops acpi_hibernation_ops = {
+	.begin = acpi_hibernation_begin,
+	.end = acpi_pm_end,
+	.pre_snapshot = acpi_pm_prepare,
+	.finish = acpi_pm_finish,
+	.prepare = acpi_pm_prepare,
+	.enter = acpi_hibernation_enter,
+	.leave = acpi_hibernation_leave,
+	.pre_restore = acpi_pm_disable_gpes,
+	.restore_cleanup = acpi_pm_enable_gpes,
+};
 
-static int acpi_hibernation_pre_restore(void)
+/**
+ *	acpi_hibernation_begin_old - Set the target system sleep state to
+ *		ACPI_STATE_S4 and execute the _PTS control method.  This
+ *		function is used if the pre-ACPI 2.0 suspend ordering has been
+ *		requested.
+ */
+static int acpi_hibernation_begin_old(void)
 {
-	acpi_status status;
-
-	status = acpi_hw_disable_all_gpes();
-
-	return ACPI_SUCCESS(status) ? 0 : -EFAULT;
-}
+	int error = acpi_sleep_prepare(ACPI_STATE_S4);
 
-static void acpi_hibernation_restore_cleanup(void)
-{
-	acpi_hw_enable_all_runtime_gpes();
+	if (!error)
+		acpi_target_sleep_state = ACPI_STATE_S4;
+	return error;
 }
 
-static struct platform_hibernation_ops acpi_hibernation_ops = {
-	.begin = acpi_hibernation_begin,
-	.end = acpi_hibernation_end,
-	.pre_snapshot = acpi_hibernation_prepare,
-	.finish = acpi_hibernation_finish,
-	.prepare = acpi_hibernation_prepare,
+/*
+ * The following callbacks are used if the pre-ACPI 2.0 suspend ordering has
+ * been requested.
+ */
+static struct platform_hibernation_ops acpi_hibernation_ops_old = {
+	.begin = acpi_hibernation_begin_old,
+	.end = acpi_pm_end,
+	.pre_snapshot = acpi_pm_disable_gpes,
+	.finish = acpi_pm_finish,
+	.prepare = acpi_pm_disable_gpes,
 	.enter = acpi_hibernation_enter,
 	.leave = acpi_hibernation_leave,
-	.pre_restore = acpi_hibernation_pre_restore,
-	.restore_cleanup = acpi_hibernation_restore_cleanup,
+	.pre_restore = acpi_pm_disable_gpes,
+	.restore_cleanup = acpi_pm_enable_gpes,
+	.recover = acpi_pm_finish,
 };
-#endif				/* CONFIG_HIBERNATION */
+#endif /* CONFIG_HIBERNATION */
 
 int acpi_suspend(u32 acpi_state)
 {
@@ -461,13 +509,15 @@ int __init acpi_sleep_init(void)
 		}
 	}
 
-	suspend_set_ops(&acpi_suspend_ops);
+	suspend_set_ops(old_suspend_ordering ?
+		&acpi_suspend_ops_old : &acpi_suspend_ops);
 #endif
 
 #ifdef CONFIG_HIBERNATION
 	status = acpi_get_sleep_type_data(ACPI_STATE_S4, &type_a, &type_b);
 	if (ACPI_SUCCESS(status)) {
-		hibernation_set_ops(&acpi_hibernation_ops);
+		hibernation_set_ops(old_suspend_ordering ?
+			&acpi_hibernation_ops_old : &acpi_hibernation_ops);
 		sleep_states[ACPI_STATE_S4] = 1;
 		printk(" S4");
 	}
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index d571204aaff7..3250c5257b74 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -781,8 +781,6 @@ int device_suspend(pm_message_t state)
 	error = dpm_prepare(state);
 	if (!error)
 		error = dpm_suspend(state);
-	if (error)
-		device_resume(resume_event(state));
 	return error;
 }
 EXPORT_SYMBOL_GPL(device_suspend);
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 41f7ce7edd7a..33adcf91ef41 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -234,6 +234,9 @@ int acpi_check_region(resource_size_t start, resource_size_t n,
 int acpi_check_mem_region(resource_size_t start, resource_size_t n,
 		      const char *name);
 
+#ifdef CONFIG_PM_SLEEP
+void __init acpi_old_suspend_ordering(void);
+#endif /* CONFIG_PM_SLEEP */
 #else	/* CONFIG_ACPI */
 
 static inline int early_acpi_boot_init(void)
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index a6977423baf7..e8e69159af71 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -86,6 +86,11 @@ typedef int __bitwise suspend_state_t;
  *	that implement @begin(), but platforms implementing @begin() should
  *	also provide a @end() which cleans up transitions aborted before
  *	@enter().
+ *
+ * @recover: Recover the platform from a suspend failure.
+ *	Called by the PM core if the suspending of devices fails.
+ *	This callback is optional and should only be implemented by platforms
+ *	which require special recovery actions in that situation.
  */
 struct platform_suspend_ops {
 	int (*valid)(suspend_state_t state);
@@ -94,6 +99,7 @@ struct platform_suspend_ops {
 	int (*enter)(suspend_state_t state);
 	void (*finish)(void);
 	void (*end)(void);
+	void (*recover)(void);
 };
 
 #ifdef CONFIG_SUSPEND
@@ -149,7 +155,7 @@ extern void mark_free_pages(struct zone *zone);
  * The methods in this structure allow a platform to carry out special
  * operations required by it during a hibernation transition.
  *
- * All the methods below must be implemented.
+ * All the methods below, except for @recover(), must be implemented.
  *
  * @begin: Tell the platform driver that we're starting hibernation.
  *	Called right after shrinking memory and before freezing devices.
@@ -189,6 +195,11 @@ extern void mark_free_pages(struct zone *zone);
  * @restore_cleanup: Clean up after a failing image restoration.
  *	Called right after the nonboot CPUs have been enabled and before
  *	thawing devices (runs with IRQs on).
+ *
+ * @recover: Recover the platform from a failure to suspend devices.
+ *	Called by the PM core if the suspending of devices during hibernation
+ *	fails.  This callback is optional and should only be implemented by
+ *	platforms which require special recovery actions in that situation.
  */
 struct platform_hibernation_ops {
 	int (*begin)(void);
@@ -200,6 +211,7 @@ struct platform_hibernation_ops {
 	void (*leave)(void);
 	int (*pre_restore)(void);
 	void (*restore_cleanup)(void);
+	void (*recover)(void);
 };
 
 #ifdef CONFIG_HIBERNATION
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index d416be0efa8a..f011e0870b52 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -179,6 +179,17 @@ static void platform_restore_cleanup(int platform_mode)
 		hibernation_ops->restore_cleanup();
 }
 
+/**
+ *	platform_recover - recover the platform from a failure to suspend
+ *	devices.
+ */
+
+static void platform_recover(int platform_mode)
+{
+	if (platform_mode && hibernation_ops && hibernation_ops->recover)
+		hibernation_ops->recover();
+}
+
 /**
  *	create_image - freeze devices that need to be frozen with interrupts
  *	off, create the hibernation image and thaw those devices.  Control
@@ -258,10 +269,10 @@ int hibernation_snapshot(int platform_mode)
 	suspend_console();
 	error = device_suspend(PMSG_FREEZE);
 	if (error)
-		goto Resume_console;
+		goto Recover_platform;
 
 	if (hibernation_test(TEST_DEVICES))
-		goto Resume_devices;
+		goto Recover_platform;
 
 	error = platform_pre_snapshot(platform_mode);
 	if (error || hibernation_test(TEST_PLATFORM))
@@ -285,11 +296,14 @@ int hibernation_snapshot(int platform_mode)
  Resume_devices:
 	device_resume(in_suspend ?
 		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
- Resume_console:
 	resume_console();
  Close:
 	platform_end(platform_mode);
 	return error;
+
+ Recover_platform:
+	platform_recover(platform_mode);
+	goto Resume_devices;
 }
 
 /**
@@ -398,8 +412,11 @@ int hibernation_platform_enter(void)
 
 	suspend_console();
 	error = device_suspend(PMSG_HIBERNATE);
-	if (error)
-		goto Resume_console;
+	if (error) {
+		if (hibernation_ops->recover)
+			hibernation_ops->recover();
+		goto Resume_devices;
+	}
 
 	error = hibernation_ops->prepare();
 	if (error)
@@ -428,7 +445,6 @@ int hibernation_platform_enter(void)
 	hibernation_ops->finish();
  Resume_devices:
 	device_resume(PMSG_RESTORE);
- Resume_console:
 	resume_console();
  Close:
 	hibernation_ops->end();
diff --git a/kernel/power/main.c b/kernel/power/main.c
index d023b6b584e5..3398f4651aa1 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -269,11 +269,11 @@ int suspend_devices_and_enter(suspend_state_t state)
 	error = device_suspend(PMSG_SUSPEND);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to suspend\n");
-		goto Resume_console;
+		goto Recover_platform;
 	}
 
 	if (suspend_test(TEST_DEVICES))
-		goto Resume_devices;
+		goto Recover_platform;
 
 	if (suspend_ops->prepare) {
 		error = suspend_ops->prepare();
@@ -294,12 +294,16 @@ int suspend_devices_and_enter(suspend_state_t state)
 		suspend_ops->finish();
  Resume_devices:
 	device_resume(PMSG_RESUME);
- Resume_console:
 	resume_console();
  Close:
 	if (suspend_ops->end)
 		suspend_ops->end();
 	return error;
+
+ Recover_platform:
+	if (suspend_ops->recover)
+		suspend_ops->recover();
+	goto Resume_devices;
 }
 
 /**
-- 
cgit v1.2.3


From c50cbb05a05cf1f9ca3592272eff053c847727d8 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Wed, 4 Jun 2008 21:47:29 -0700
Subject: cpu topology: always define CPU topology information

This can result in an empty topology directory in sysfs, and requires
in-kernel users to protect all uses with #ifdef - see
<http://marc.info/?l=linux-netdev&m=120639033904472&w=2>.

The documentation of CPU topology specifies what the defaults should be if
only partial information is available from the hardware.  So we can
provide these defaults as a fallback.

This patch:

- Adds default definitions of the 4 topology macros to <linux/topology.h>
- Changes drivers/base/topology.c to use the topology macros unconditionally
  and to cope with definitions that aren't lvalues
- Updates documentation accordingly

[ From: Andrew Morton <akpm@linux-foundation.org>
  - fold now-duplicated code
  - fix layout
]

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Chandra Seetharaman <sekharan@us.ibm.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Mike Travis <travis@sgi.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: John Hawkes <hawkes@sgi.com>
Cc: Zhang, Yanmin <yanmin.zhang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 Documentation/cputopology.txt | 26 +++++++++-----------------
 drivers/base/topology.c       | 38 ++++++++++----------------------------
 include/linux/topology.h      | 13 +++++++++++++
 3 files changed, 32 insertions(+), 45 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/cputopology.txt b/Documentation/cputopology.txt
index b61cb9564023..bd699da24666 100644
--- a/Documentation/cputopology.txt
+++ b/Documentation/cputopology.txt
@@ -14,9 +14,8 @@ represent the thread siblings to cpu X in the same physical package;
 To implement it in an architecture-neutral way, a new source file,
 drivers/base/topology.c, is to export the 4 attributes.
 
-If one architecture wants to support this feature, it just needs to
-implement 4 defines, typically in file include/asm-XXX/topology.h.
-The 4 defines are:
+For an architecture to support this feature, it must define some of
+these macros in include/asm-XXX/topology.h:
 #define topology_physical_package_id(cpu)
 #define topology_core_id(cpu)
 #define topology_thread_siblings(cpu)
@@ -25,17 +24,10 @@ The 4 defines are:
 The type of **_id is int.
 The type of siblings is cpumask_t.
 
-To be consistent on all architectures, the 4 attributes should have
-default values if their values are unavailable. Below is the rule.
-1) physical_package_id: If cpu has no physical package id, -1 is the
-default value.
-2) core_id: If cpu doesn't support multi-core, its core id is 0.
-3) thread_siblings: Just include itself, if the cpu doesn't support
-HT/multi-thread.
-4) core_siblings: Just include itself, if the cpu doesn't support
-multi-core and HT/Multi-thread.
-
-So be careful when declaring the 4 defines in include/asm-XXX/topology.h.
-
-If an attribute isn't defined on an architecture, it won't be exported.
-
+To be consistent on all architectures, include/linux/topology.h
+provides default definitions for any of the above macros that are
+not defined by include/asm-XXX/topology.h:
+1) physical_package_id: -1
+2) core_id: 0
+3) thread_siblings: just the given CPU
+4) core_siblings: just the given CPU
diff --git a/drivers/base/topology.c b/drivers/base/topology.c
index fdf4044d2e74..24d29a9fc25b 100644
--- a/drivers/base/topology.c
+++ b/drivers/base/topology.c
@@ -59,60 +59,42 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf)
 static inline ssize_t show_##name(struct sys_device *dev, char *buf)	\
 {									\
 	unsigned int cpu = dev->id;					\
-	return show_cpumap(0, &(topology_##name(cpu)), buf);		\
+	cpumask_t siblings = topology_##name(cpu);			\
+	return show_cpumap(0, &siblings, buf);				\
 }
 
 #define define_siblings_show_list(name)					\
 static inline ssize_t show_##name##_list(struct sys_device *dev, char *buf) \
 {									\
 	unsigned int cpu = dev->id;					\
-	return show_cpumap(1, &(topology_##name(cpu)), buf);		\
+	cpumask_t siblings = topology_##name(cpu);			\
+	return show_cpumap(1, &siblings, buf);				\
 }
 
 #define define_siblings_show_func(name)		\
 	define_siblings_show_map(name); define_siblings_show_list(name)
 
-#ifdef	topology_physical_package_id
 define_id_show_func(physical_package_id);
 define_one_ro(physical_package_id);
-#define ref_physical_package_id_attr	&attr_physical_package_id.attr,
-#else
-#define ref_physical_package_id_attr
-#endif
 
-#ifdef topology_core_id
 define_id_show_func(core_id);
 define_one_ro(core_id);
-#define ref_core_id_attr		&attr_core_id.attr,
-#else
-#define ref_core_id_attr
-#endif
 
-#ifdef topology_thread_siblings
 define_siblings_show_func(thread_siblings);
 define_one_ro(thread_siblings);
 define_one_ro(thread_siblings_list);
-#define ref_thread_siblings_attr	\
-		&attr_thread_siblings.attr, &attr_thread_siblings_list.attr,
-#else
-#define ref_thread_siblings_attr
-#endif
 
-#ifdef topology_core_siblings
 define_siblings_show_func(core_siblings);
 define_one_ro(core_siblings);
 define_one_ro(core_siblings_list);
-#define ref_core_siblings_attr		\
-		&attr_core_siblings.attr, &attr_core_siblings_list.attr,
-#else
-#define ref_core_siblings_attr
-#endif
 
 static struct attribute *default_attrs[] = {
-	ref_physical_package_id_attr
-	ref_core_id_attr
-	ref_thread_siblings_attr
-	ref_core_siblings_attr
+	&attr_physical_package_id.attr,
+	&attr_core_id.attr,
+	&attr_thread_siblings.attr,
+	&attr_thread_siblings_list.attr,
+	&attr_core_siblings.attr,
+	&attr_core_siblings_list.attr,
 	NULL
 };
 
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 24f3d2282e11..2158fc0d5a56 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -179,4 +179,17 @@ void arch_update_cpu_topology(void);
 #endif
 #endif /* CONFIG_NUMA */
 
+#ifndef topology_physical_package_id
+#define topology_physical_package_id(cpu)	((void)(cpu), -1)
+#endif
+#ifndef topology_core_id
+#define topology_core_id(cpu)			((void)(cpu), 0)
+#endif
+#ifndef topology_thread_siblings
+#define topology_thread_siblings(cpu)		cpumask_of_cpu(cpu)
+#endif
+#ifndef topology_core_siblings
+#define topology_core_siblings(cpu)		cpumask_of_cpu(cpu)
+#endif
+
 #endif /* _LINUX_TOPOLOGY_H */
-- 
cgit v1.2.3


From 564d9bdd89df4ceece5d126ff3b7db506ae06548 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Tue, 10 Jun 2008 14:04:14 +0200
Subject: nl80211: Fix comment merge error

The comments ended up in the wrong place due to a merge error.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index ea6517e58b04..aa8411e2a160 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -122,13 +122,13 @@ enum nl80211_commands {
 	NL80211_CMD_NEW_STATION,
 	NL80211_CMD_DEL_STATION,
 
-	/* add commands here */
-
 	NL80211_CMD_GET_MPATH,
 	NL80211_CMD_SET_MPATH,
 	NL80211_CMD_NEW_MPATH,
 	NL80211_CMD_DEL_MPATH,
 
+	/* add commands here */
+
 	/* used to define NL80211_CMD_MAX below */
 	__NL80211_CMD_AFTER_LAST,
 	NL80211_CMD_MAX = __NL80211_CMD_AFTER_LAST - 1
@@ -230,13 +230,13 @@ enum nl80211_attrs {
 
 	NL80211_ATTR_MNTR_FLAGS,
 
-	/* add attributes here, update the policy in nl80211.c */
-
 	NL80211_ATTR_MESH_ID,
 	NL80211_ATTR_STA_PLINK_ACTION,
 	NL80211_ATTR_MPATH_NEXT_HOP,
 	NL80211_ATTR_MPATH_INFO,
 
+	/* add attributes here, update the policy in nl80211.c */
+
 	__NL80211_ATTR_AFTER_LAST,
 	NL80211_ATTR_MAX = __NL80211_ATTR_AFTER_LAST - 1
 };
-- 
cgit v1.2.3


From 963f55178b25cb673ab438edaae4127b1d014bc1 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Thu, 12 Jun 2008 09:47:00 +0800
Subject: iwlwifi: remove redundant flags regarding to FAT channel

This patch removes redundant flags regarding to FAT channel. Use
mac80211's flag instead.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-core.c     | 42 +++++++++--------------------
 drivers/net/wireless/iwlwifi/iwl-dev.h      | 14 ----------
 drivers/net/wireless/iwlwifi/iwl-eeprom.c   | 19 ++++++++-----
 drivers/net/wireless/iwlwifi/iwl4965-base.c |  6 +++--
 include/linux/ieee80211.h                   |  1 +
 5 files changed, 31 insertions(+), 51 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c
index 6c7617c1bb91..fad26f71d7ce 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.c
+++ b/drivers/net/wireless/iwlwifi/iwl-core.c
@@ -478,24 +478,7 @@ static int iwlcore_init_geos(struct iwl_priv *priv)
 			if (ch->flags & EEPROM_CHANNEL_RADAR)
 				geo_ch->flags |= IEEE80211_CHAN_RADAR;
 
-			switch (ch->fat_extension_channel) {
-			case HT_IE_EXT_CHANNEL_ABOVE:
-				/* only above is allowed, disable below */
-				geo_ch->flags |= IEEE80211_CHAN_NO_FAT_BELOW;
-				break;
-			case HT_IE_EXT_CHANNEL_BELOW:
-				/* only below is allowed, disable above */
-				geo_ch->flags |= IEEE80211_CHAN_NO_FAT_ABOVE;
-				break;
-			case HT_IE_EXT_CHANNEL_NONE:
-				/* fat not allowed: disable both*/
-				geo_ch->flags |= (IEEE80211_CHAN_NO_FAT_ABOVE |
-						  IEEE80211_CHAN_NO_FAT_BELOW);
-				break;
-			case HT_IE_EXT_CHANNEL_MAX:
-				/* both above and below are permitted */
-				break;
-			}
+			geo_ch->flags |= ch->fat_extension_channel;
 
 			if (ch->max_power_avg > priv->max_channel_txpower_limit)
 				priv->max_channel_txpower_limit =
@@ -507,7 +490,7 @@ static int iwlcore_init_geos(struct iwl_priv *priv)
 		/* Save flags for reg domain usage */
 		geo_ch->orig_flags = geo_ch->flags;
 
-		IWL_DEBUG_INFO("Channel %d Freq=%d[%sGHz] %s flag=0%X\n",
+		IWL_DEBUG_INFO("Channel %d Freq=%d[%sGHz] %s flag=0x%X\n",
 				ch->channel, geo_ch->center_freq,
 				is_channel_a_band(ch) ?  "5.2" : "2.4",
 				geo_ch->flags & IEEE80211_CHAN_DISABLED ?
@@ -552,6 +535,7 @@ static u8 is_single_rx_stream(struct iwl_priv *priv)
 		(priv->current_ht_config.supp_mcs_set[2] == 0)) ||
 	       priv->ps_mode == IWL_MIMO_PS_STATIC;
 }
+
 static u8 iwl_is_channel_extension(struct iwl_priv *priv,
 				   enum ieee80211_band band,
 				   u16 channel, u8 extension_chan_offset)
@@ -562,12 +546,12 @@ static u8 iwl_is_channel_extension(struct iwl_priv *priv,
 	if (!is_channel_valid(ch_info))
 		return 0;
 
-	if (extension_chan_offset == IWL_EXT_CHANNEL_OFFSET_NONE)
-		return 0;
-
-	if ((ch_info->fat_extension_channel == extension_chan_offset) ||
-	    (ch_info->fat_extension_channel == HT_IE_EXT_CHANNEL_MAX))
-		return 1;
+	if (extension_chan_offset == IEEE80211_HT_IE_CHA_SEC_ABOVE)
+		return !(ch_info->fat_extension_channel &
+					IEEE80211_CHAN_NO_FAT_ABOVE);
+	else if (extension_chan_offset == IEEE80211_HT_IE_CHA_SEC_BELOW)
+		return !(ch_info->fat_extension_channel &
+					IEEE80211_CHAN_NO_FAT_BELOW);
 
 	return 0;
 }
@@ -579,7 +563,7 @@ u8 iwl_is_fat_tx_allowed(struct iwl_priv *priv,
 
 	if ((!iwl_ht_conf->is_ht) ||
 	   (iwl_ht_conf->supported_chan_width != IWL_CHANNEL_WIDTH_40MHZ) ||
-	   (iwl_ht_conf->extension_chan_offset == IWL_EXT_CHANNEL_OFFSET_NONE))
+	   (iwl_ht_conf->extension_chan_offset == IEEE80211_HT_IE_CHA_SEC_NONE))
 		return 0;
 
 	if (sta_ht_inf) {
@@ -619,13 +603,13 @@ void iwl_set_rxon_ht(struct iwl_priv *priv, struct iwl_ht_info *ht_info)
 
 	/* Note: control channel is opposite of extension channel */
 	switch (ht_info->extension_chan_offset) {
-	case IWL_EXT_CHANNEL_OFFSET_ABOVE:
+	case IEEE80211_HT_IE_CHA_SEC_ABOVE:
 		rxon->flags &= ~(RXON_FLG_CTRL_CHANNEL_LOC_HI_MSK);
 		break;
-	case IWL_EXT_CHANNEL_OFFSET_BELOW:
+	case IEEE80211_HT_IE_CHA_SEC_BELOW:
 		rxon->flags |= RXON_FLG_CTRL_CHANNEL_LOC_HI_MSK;
 		break;
-	case IWL_EXT_CHANNEL_OFFSET_NONE:
+	case IEEE80211_HT_IE_CHA_SEC_NONE:
 	default:
 		rxon->flags &= ~RXON_FLG_CHANNEL_MODE_MIXED_MSK;
 		break;
diff --git a/drivers/net/wireless/iwlwifi/iwl-dev.h b/drivers/net/wireless/iwlwifi/iwl-dev.h
index fa35193e5a7a..daa14d9e5a80 100644
--- a/drivers/net/wireless/iwlwifi/iwl-dev.h
+++ b/drivers/net/wireless/iwlwifi/iwl-dev.h
@@ -179,15 +179,6 @@ struct iwl4965_scan_power_info {
 	s8 requested_power;	/* scan pwr (dBm) requested for chnl/rate */
 };
 
-/* For fat_extension_channel */
-enum {
-	HT_IE_EXT_CHANNEL_NONE = 0,
-	HT_IE_EXT_CHANNEL_ABOVE,
-	HT_IE_EXT_CHANNEL_INVALID,
-	HT_IE_EXT_CHANNEL_BELOW,
-	HT_IE_EXT_CHANNEL_MAX
-};
-
 /*
  * One for each channel, holds all channel setup data
  * Some of the fields (e.g. eeprom and flags/max_power_avg) are redundant
@@ -782,11 +773,6 @@ struct iwl_kw {
 #define IWL_OPERATION_MODE_MIXED    2
 #define IWL_OPERATION_MODE_20MHZ    3
 
-#define IWL_EXT_CHANNEL_OFFSET_NONE      0
-#define IWL_EXT_CHANNEL_OFFSET_ABOVE     1
-#define IWL_EXT_CHANNEL_OFFSET_RESERVE1  2
-#define IWL_EXT_CHANNEL_OFFSET_BELOW     3
-
 #define IWL_TX_CRC_SIZE 4
 #define IWL_TX_DELIMITER_SIZE 4
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-eeprom.c b/drivers/net/wireless/iwlwifi/iwl-eeprom.c
index 11f9d9557a0e..cbb812f9f620 100644
--- a/drivers/net/wireless/iwlwifi/iwl-eeprom.c
+++ b/drivers/net/wireless/iwlwifi/iwl-eeprom.c
@@ -470,6 +470,11 @@ int iwl_init_channel_map(struct iwl_priv *priv)
 			/* Copy the run-time flags so they are there even on
 			 * invalid channels */
 			ch_info->flags = eeprom_ch_info[ch].flags;
+			/* First write that fat is not enabled, and then enable
+			 * one by one */
+			ch_info->fat_extension_channel =
+				(IEEE80211_CHAN_NO_FAT_ABOVE |
+				 IEEE80211_CHAN_NO_FAT_BELOW);
 
 			if (!(is_channel_valid(ch_info))) {
 				IWL_DEBUG_INFO("Ch. %d Flags %x [%sGHz] - "
@@ -534,12 +539,14 @@ int iwl_init_channel_map(struct iwl_priv *priv)
 		for (ch = 0; ch < eeprom_ch_count; ch++) {
 
 			if ((band == 6) &&
-			    ((eeprom_ch_index[ch] == 5) ||
-			    (eeprom_ch_index[ch] == 6) ||
-			    (eeprom_ch_index[ch] == 7)))
-			       fat_extension_chan = HT_IE_EXT_CHANNEL_MAX;
+				((eeprom_ch_index[ch] == 5) ||
+				 (eeprom_ch_index[ch] == 6) ||
+				 (eeprom_ch_index[ch] == 7)))
+				/* both are allowed: above and below */
+				fat_extension_chan = 0;
 			else
-				fat_extension_chan = HT_IE_EXT_CHANNEL_ABOVE;
+				fat_extension_chan =
+					IEEE80211_CHAN_NO_FAT_BELOW;
 
 			/* Set up driver's info for lower half */
 			iwl_set_fat_chan_info(priv, ieeeband,
@@ -551,7 +558,7 @@ int iwl_init_channel_map(struct iwl_priv *priv)
 			iwl_set_fat_chan_info(priv, ieeeband,
 						(eeprom_ch_index[ch] + 4),
 						&(eeprom_ch_info[ch]),
-						HT_IE_EXT_CHANNEL_BELOW);
+						IEEE80211_CHAN_NO_FAT_ABOVE);
 		}
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c
index 0acd42bd43c5..de40e893f09e 100644
--- a/drivers/net/wireless/iwlwifi/iwl4965-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c
@@ -664,9 +664,11 @@ static void iwl4965_ht_conf(struct iwl_priv *priv,
 	iwl_conf->extension_chan_offset =
 		ht_bss_conf->bss_cap & IEEE80211_HT_IE_CHA_SEC_OFFSET;
 	/* If no above or below channel supplied disable FAT channel */
-	if (iwl_conf->extension_chan_offset != IWL_EXT_CHANNEL_OFFSET_ABOVE &&
-	    iwl_conf->extension_chan_offset != IWL_EXT_CHANNEL_OFFSET_BELOW)
+	if (iwl_conf->extension_chan_offset != IEEE80211_HT_IE_CHA_SEC_ABOVE &&
+	    iwl_conf->extension_chan_offset != IEEE80211_HT_IE_CHA_SEC_BELOW) {
+		iwl_conf->extension_chan_offset = IEEE80211_HT_IE_CHA_SEC_NONE;
 		iwl_conf->supported_chan_width = 0;
+	}
 
 	iwl_conf->tx_mimo_ps_mode =
 		(u8)((ht_conf->cap & IEEE80211_HT_CAP_MIMO_PS) >> 2);
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 9300f37cd7e8..8f2c20b4a15d 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -320,6 +320,7 @@ struct ieee80211_ht_addt_info {
 #define IEEE80211_HT_CAP_MCS_TX_UEQM		0x10
 /* 802.11n HT IE masks */
 #define IEEE80211_HT_IE_CHA_SEC_OFFSET		0x03
+#define IEEE80211_HT_IE_CHA_SEC_NONE	 	0x00
 #define IEEE80211_HT_IE_CHA_SEC_ABOVE 		0x01
 #define IEEE80211_HT_IE_CHA_SEC_BELOW 		0x03
 #define IEEE80211_HT_IE_CHA_WIDTH		0x04
-- 
cgit v1.2.3


From fd7c8a40b2a63863f749e4d17f0d94d2e5ab1331 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 11 Jun 2008 14:21:56 -0700
Subject: mac80211: add helpers for frame control testing

A few general categories:

1) ieee80211_has_* tests if particular fctl bits are set, the helpers are de
in the same order as the fctl defines:

A combined _has_a4 was also added to test when both FROMDS and TODS are set.

2) ieee80211_is_* is meant to test whether the frame control is of a certain
ftype - data, mgmt, ctl, and two special helpers _is_data_qos, _is_data_pres
which also test a subset of the stype space.

When testing for a particular stype applicable only to one ftype, functions
like ieee80211_is_ack have been added.  Note that the ftype is also being
checked in these helpers.  They have been added for all mgmt and ctl stypes
in the same order as the STYPE defines.

3) ieee80211_get_* is meant to take a struct ieee80211_hdr * and returns a
pointer to somewhere in the struct, see get_SA, get_DA, get_qos_ctl.

The intel wireless drivers had helpers that used this namespace, convert the
all to use the new helpers and remove the byteshifting as they were defined
in cpu-order rather than little-endian.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-3945.c     |  36 +--
 drivers/net/wireless/iwlwifi/iwl-4965-rs.c  |  17 +-
 drivers/net/wireless/iwlwifi/iwl-4965.c     |  28 +-
 drivers/net/wireless/iwlwifi/iwl-5000.c     |   6 +-
 drivers/net/wireless/iwlwifi/iwl-helpers.h  | 109 --------
 drivers/net/wireless/iwlwifi/iwl-tx.c       |  49 ++--
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  33 ++-
 include/linux/ieee80211.h                   | 392 ++++++++++++++++++++++++++--
 8 files changed, 457 insertions(+), 213 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index 0ba6889dfd41..63f20370032d 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -388,7 +388,7 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 	u32 print_dump = 0;	/* set to 1 to dump all frames' contents */
 	u32 hundred = 0;
 	u32 dataframe = 0;
-	u16 fc;
+	__le16 fc;
 	u16 seq_ctl;
 	u16 channel;
 	u16 phy_flags;
@@ -407,7 +407,7 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 	u8 *data = IWL_RX_DATA(pkt);
 
 	/* MAC header */
-	fc = le16_to_cpu(header->frame_control);
+	fc = header->frame_control;
 	seq_ctl = le16_to_cpu(header->seq_ctrl);
 
 	/* metadata */
@@ -431,8 +431,8 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 
 	/* if data frame is to us and all is good,
 	 *   (optionally) print summary for only 1 out of every 100 */
-	if (to_us && (fc & ~IEEE80211_FCTL_PROTECTED) ==
-	    (IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
+	if (to_us && (fc & ~cpu_to_le16(IEEE80211_FCTL_PROTECTED)) ==
+	    cpu_to_le16(IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
 		dataframe = 1;
 		if (!group100)
 			print_summary = 1;	/* print each frame */
@@ -455,13 +455,13 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 
 		if (hundred)
 			title = "100Frames";
-		else if (fc & IEEE80211_FCTL_RETRY)
+		else if (ieee80211_has_retry(fc))
 			title = "Retry";
-		else if (ieee80211_is_assoc_response(fc))
+		else if (ieee80211_is_assoc_resp(fc))
 			title = "AscRsp";
-		else if (ieee80211_is_reassoc_response(fc))
+		else if (ieee80211_is_reassoc_resp(fc))
 			title = "RasRsp";
-		else if (ieee80211_is_probe_response(fc)) {
+		else if (ieee80211_is_probe_resp(fc)) {
 			title = "PrbRsp";
 			print_dump = 1;	/* dump frame contents */
 		} else if (ieee80211_is_beacon(fc)) {
@@ -490,14 +490,14 @@ static void iwl3945_dbg_report_frame(struct iwl3945_priv *priv,
 		if (dataframe)
 			IWL_DEBUG_RX("%s: mhd=0x%04x, dst=0x%02x, "
 				     "len=%u, rssi=%d, chnl=%d, rate=%u, \n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     length, rssi, channel, rate);
 		else {
 			/* src/dst addresses assume managed mode */
 			IWL_DEBUG_RX("%s: 0x%04x, dst=0x%02x, "
 				     "src=0x%02x, rssi=%u, tim=%lu usec, "
 				     "phy=0x%02x, chnl=%d\n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     header->addr3[5], rssi,
 				     tsf_low - priv->scan_start_tsf,
 				     phy_flags, channel);
@@ -971,7 +971,7 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 	u8 rts_retry_limit;
 	u8 data_retry_limit;
 	__le32 tx_flags;
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 
 	rate = iwl3945_rates[rate_index].plcp;
 	tx_flags = cmd->cmd.tx.tx_flags;
@@ -996,7 +996,7 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 	else
 		rts_retry_limit = 7;
 
-	if (ieee80211_is_probe_response(fc)) {
+	if (ieee80211_is_probe_resp(fc)) {
 		data_retry_limit = 3;
 		if (data_retry_limit < rts_retry_limit)
 			rts_retry_limit = data_retry_limit;
@@ -1006,12 +1006,12 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv,
 	if (priv->data_retry_limit != -1)
 		data_retry_limit = priv->data_retry_limit;
 
-	if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) {
-		switch (fc & IEEE80211_FCTL_STYPE) {
-		case IEEE80211_STYPE_AUTH:
-		case IEEE80211_STYPE_DEAUTH:
-		case IEEE80211_STYPE_ASSOC_REQ:
-		case IEEE80211_STYPE_REASSOC_REQ:
+	if (ieee80211_is_mgmt(fc)) {
+		switch (fc & cpu_to_le16(IEEE80211_FCTL_STYPE)) {
+		case cpu_to_le16(IEEE80211_STYPE_AUTH):
+		case cpu_to_le16(IEEE80211_STYPE_DEAUTH):
+		case cpu_to_le16(IEEE80211_STYPE_ASSOC_REQ):
+		case cpu_to_le16(IEEE80211_STYPE_REASSOC_REQ):
 			if (tx_flags & TX_CMD_FLG_RTS_MSK) {
 				tx_flags &= ~TX_CMD_FLG_RTS_MSK;
 				tx_flags |= TX_CMD_FLG_CTS_MSK;
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
index d601d31e0de9..202303117285 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c
@@ -281,11 +281,11 @@ static u8 rs_tl_add_packet(struct iwl4965_lq_sta *lq_data,
 	u32 time_diff;
 	s32 index;
 	struct iwl4965_traffic_load *tl = NULL;
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	u8 tid;
 
-	if (ieee80211_is_qos_data(fc)) {
-		u8 *qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(fc)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 	} else
 		return MAX_TID_COUNT;
@@ -794,7 +794,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev,
 	struct iwl4965_scale_tbl_info tbl_type;
 	struct iwl4965_scale_tbl_info *curr_tbl, *search_tbl;
 	u8 active_index = 0;
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	s32 tpt = 0;
 
 	IWL_DEBUG_RATE_LIMIT("get frame ack response, update rate scale window\n");
@@ -1651,7 +1651,8 @@ static void rs_rate_scale_perform(struct iwl_priv *priv,
 	int high_tpt = IWL_INVALID_VALUE;
 	u32 fail_count;
 	s8 scale_action = 0;
-	u16 fc, rate_mask;
+	__le16 fc;
+	u16 rate_mask;
 	u8 update_lq = 0;
 	struct iwl4965_lq_sta *lq_sta;
 	struct iwl4965_scale_tbl_info *tbl, *tbl1;
@@ -1666,7 +1667,7 @@ static void rs_rate_scale_perform(struct iwl_priv *priv,
 
 	IWL_DEBUG_RATE("rate scale calculate new rate for skb\n");
 
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 	if (!ieee80211_is_data(fc) || is_multicast_ether_addr(hdr->addr1)) {
 		/* Send management frames and broadcast/multicast data using
 		 * lowest rate. */
@@ -2100,7 +2101,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 	struct ieee80211_conf *conf = &local->hw.conf;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
 	struct sta_info *sta;
-	u16 fc;
+	__le16 fc;
 	struct iwl_priv *priv = (struct iwl_priv *)priv_rate;
 	struct iwl4965_lq_sta *lq_sta;
 
@@ -2112,7 +2113,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev,
 
 	/* Send management frames and broadcast/multicast data using lowest
 	 * rate. */
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 	if (!ieee80211_is_data(fc) || is_multicast_ether_addr(hdr->addr1) ||
 	    !sta || !sta->rate_ctrl_priv) {
 		sel->rate_idx = rate_lowest_index(local, sband, sta);
diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c
index df345cb6fd7d..ab5027345a01 100644
--- a/drivers/net/wireless/iwlwifi/iwl-4965.c
+++ b/drivers/net/wireless/iwlwifi/iwl-4965.c
@@ -2388,7 +2388,7 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 	u32 print_dump = 0;	/* set to 1 to dump all frames' contents */
 	u32 hundred = 0;
 	u32 dataframe = 0;
-	u16 fc;
+	__le16 fc;
 	u16 seq_ctl;
 	u16 channel;
 	u16 phy_flags;
@@ -2411,7 +2411,7 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 		return;
 
 	/* MAC header */
-	fc = le16_to_cpu(header->frame_control);
+	fc = header->frame_control;
 	seq_ctl = le16_to_cpu(header->seq_ctrl);
 
 	/* metadata */
@@ -2436,8 +2436,8 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 
 	/* if data frame is to us and all is good,
 	 *   (optionally) print summary for only 1 out of every 100 */
-	if (to_us && (fc & ~IEEE80211_FCTL_PROTECTED) ==
-	    (IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
+	if (to_us && (fc & ~cpu_to_le16(IEEE80211_FCTL_PROTECTED)) ==
+	    cpu_to_le16(IEEE80211_FCTL_FROMDS | IEEE80211_FTYPE_DATA)) {
 		dataframe = 1;
 		if (!group100)
 			print_summary = 1;	/* print each frame */
@@ -2461,13 +2461,13 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 
 		if (hundred)
 			title = "100Frames";
-		else if (fc & IEEE80211_FCTL_RETRY)
+		else if (ieee80211_has_retry(fc))
 			title = "Retry";
-		else if (ieee80211_is_assoc_response(fc))
+		else if (ieee80211_is_assoc_resp(fc))
 			title = "AscRsp";
-		else if (ieee80211_is_reassoc_response(fc))
+		else if (ieee80211_is_reassoc_resp(fc))
 			title = "RasRsp";
-		else if (ieee80211_is_probe_response(fc)) {
+		else if (ieee80211_is_probe_resp(fc)) {
 			title = "PrbRsp";
 			print_dump = 1;	/* dump frame contents */
 		} else if (ieee80211_is_beacon(fc)) {
@@ -2496,14 +2496,14 @@ static void iwl4965_dbg_report_frame(struct iwl_priv *priv,
 		if (dataframe)
 			IWL_DEBUG_RX("%s: mhd=0x%04x, dst=0x%02x, "
 				     "len=%u, rssi=%d, chnl=%d, rate=%u, \n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     length, rssi, channel, bitrate);
 		else {
 			/* src/dst addresses assume managed mode */
 			IWL_DEBUG_RX("%s: 0x%04x, dst=0x%02x, "
 				     "src=0x%02x, rssi=%u, tim=%lu usec, "
 				     "phy=0x%02x, chnl=%d\n",
-				     title, fc, header->addr1[5],
+				     title, le16_to_cpu(fc), header->addr1[5],
 				     header->addr3[5], rssi,
 				     tsf_low - priv->scan_start_tsf,
 				     phy_flags, channel);
@@ -3219,7 +3219,7 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 	struct iwl4965_tx_resp *tx_resp = (void *)&pkt->u.raw[0];
 	u32  status = le32_to_cpu(tx_resp->u.status);
 	int tid = MAX_TID_COUNT, sta_id = IWL_INVALID_STATION;
-	u16 fc;
+	__le16 fc;
 	struct ieee80211_hdr *hdr;
 	u8 *qc = NULL;
 
@@ -3235,9 +3235,9 @@ static void iwl4965_rx_reply_tx(struct iwl_priv *priv,
 	memset(&info->status, 0, sizeof(info->status));
 
 	hdr = iwl_tx_queue_get_hdr(priv, txq_id, index);
-	fc = le16_to_cpu(hdr->frame_control);
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	fc = hdr->frame_control;
+	if (ieee80211_is_data_qos(fc)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-5000.c b/drivers/net/wireless/iwlwifi/iwl-5000.c
index 8c466b02bdff..438c3812c390 100644
--- a/drivers/net/wireless/iwlwifi/iwl-5000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-5000.c
@@ -1252,7 +1252,6 @@ static void iwl5000_rx_reply_tx(struct iwl_priv *priv,
 	struct iwl5000_tx_resp *tx_resp = (void *)&pkt->u.raw[0];
 	u32  status = le16_to_cpu(tx_resp->status.status);
 	int tid = MAX_TID_COUNT, sta_id = IWL_INVALID_STATION;
-	u16 fc;
 	struct ieee80211_hdr *hdr;
 	u8 *qc = NULL;
 
@@ -1268,9 +1267,8 @@ static void iwl5000_rx_reply_tx(struct iwl_priv *priv,
 	memset(&info->status, 0, sizeof(info->status));
 
 	hdr = iwl_tx_queue_get_hdr(priv, txq_id, index);
-	fc = le16_to_cpu(hdr->frame_control);
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(hdr->frame_control)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 	}
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-helpers.h b/drivers/net/wireless/iwlwifi/iwl-helpers.h
index eec423a98fe6..41eed6793328 100644
--- a/drivers/net/wireless/iwlwifi/iwl-helpers.h
+++ b/drivers/net/wireless/iwlwifi/iwl-helpers.h
@@ -145,115 +145,6 @@ static inline struct ieee80211_conf *ieee80211_get_hw_conf(
 	return &hw->conf;
 }
 
-#define QOS_CONTROL_LEN 2
-
-
-static inline int ieee80211_is_management(u16 fc)
-{
-	return (fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT;
-}
-
-static inline int ieee80211_is_control(u16 fc)
-{
-	return (fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL;
-}
-
-static inline int ieee80211_is_data(u16 fc)
-{
-	return (fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA;
-}
-
-static inline int ieee80211_is_back_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BACK_REQ);
-}
-
-static inline int ieee80211_is_probe_response(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PROBE_RESP);
-}
-
-static inline int ieee80211_is_probe_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_PROBE_REQ);
-}
-
-static inline int ieee80211_is_beacon(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BEACON);
-}
-
-static inline int ieee80211_is_atim(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ATIM);
-}
-
-static inline int ieee80211_is_assoc_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_assoc_response(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_RESP);
-}
-
-static inline int ieee80211_is_auth(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_deauth(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_disassoc(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ);
-}
-
-static inline int ieee80211_is_reassoc_request(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_REQ);
-}
-
-static inline int ieee80211_is_reassoc_response(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_RESP);
-}
-
-static inline int ieee80211_is_qos_data(u16 fc)
-{
-	return ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) &&
-	       ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_QOS_DATA);
-}
-/**
- * ieee80211_get_qos_ctrl - get pointer to the QoS control field
- *
- * This function returns the pointer to 802.11 header QoS field (2 bytes)
- * This function doesn't check whether hdr is a QoS hdr, use with care
- * @hdr: struct ieee80211_hdr *hdr
- * @hdr_len: header length
- */
-
-static inline u8 *ieee80211_get_qos_ctrl(struct ieee80211_hdr *hdr, int hdr_len)
-{
-	return  ((u8 *) hdr + hdr_len - QOS_CONTROL_LEN);
-}
-
 static inline int iwl_check_bits(unsigned long field, unsigned long mask)
 {
 	return ((field & mask) == mask) ? 1 : 0;
diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index e804bf8aea80..98c434c52a64 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -569,15 +569,15 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 				  struct ieee80211_hdr *hdr,
 				  int is_unicast, u8 std_id)
 {
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	__le32 tx_flags = tx_cmd->tx_flags;
 
 	tx_cmd->stop_time.life_time = TX_CMD_LIFE_TIME_INFINITE;
 	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		tx_flags |= TX_CMD_FLG_ACK_MSK;
-		if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)
+		if (ieee80211_is_mgmt(fc))
 			tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
-		if (ieee80211_is_probe_response(fc) &&
+		if (ieee80211_is_probe_resp(fc) &&
 		    !(le16_to_cpu(hdr->seq_ctrl) & 0xf))
 			tx_flags |= TX_CMD_FLG_TSF_MSK;
 	} else {
@@ -585,7 +585,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 		tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
 	}
 
-	if (ieee80211_is_back_request(fc))
+	if (ieee80211_is_back_req(fc))
 		tx_flags |= TX_CMD_FLG_ACK_MSK | TX_CMD_FLG_IMM_BA_RSP_MASK;
 
 
@@ -593,8 +593,8 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 	if (ieee80211_get_morefrag(hdr))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
-	if (ieee80211_is_qos_data(fc)) {
-		u8 *qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(fc)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		tx_cmd->tid_tspec = qc[0] & 0xf;
 		tx_flags &= ~TX_CMD_FLG_SEQ_CTL_MSK;
 	} else {
@@ -613,9 +613,8 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 		tx_flags |= TX_CMD_FLG_FULL_TXOP_PROT_MSK;
 
 	tx_flags &= ~(TX_CMD_FLG_ANT_SEL_MSK);
-	if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) {
-		if ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ ||
-		    (fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_REQ)
+	if (ieee80211_is_mgmt(fc)) {
+		if (ieee80211_is_assoc_req(fc) || ieee80211_is_reassoc_req(fc))
 			tx_cmd->timeout.pm_frame_timeout = cpu_to_le16(3);
 		else
 			tx_cmd->timeout.pm_frame_timeout = cpu_to_le16(2);
@@ -634,7 +633,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 			      struct iwl_tx_cmd *tx_cmd,
 			      struct ieee80211_tx_info *info,
-			      u16 fc, int sta_id,
+			      __le16 fc, int sta_id,
 			      int is_hcca)
 {
 	u8 rts_retry_limit = 0;
@@ -655,7 +654,7 @@ static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 		rate_flags |= RATE_MCS_CCK_MSK;
 
 
-	if (ieee80211_is_probe_response(fc)) {
+	if (ieee80211_is_probe_resp(fc)) {
 		data_retry_limit = 3;
 		if (data_retry_limit < rts_retry_limit)
 			rts_retry_limit = data_retry_limit;
@@ -670,11 +669,11 @@ static void iwl_tx_cmd_build_rate(struct iwl_priv *priv,
 		tx_cmd->initial_rate_index = 0;
 		tx_cmd->tx_flags |= TX_CMD_FLG_STA_RATE_MSK;
 	} else {
-		switch (fc & IEEE80211_FCTL_STYPE) {
-		case IEEE80211_STYPE_AUTH:
-		case IEEE80211_STYPE_DEAUTH:
-		case IEEE80211_STYPE_ASSOC_REQ:
-		case IEEE80211_STYPE_REASSOC_REQ:
+		switch (fc & cpu_to_le16(IEEE80211_FCTL_STYPE)) {
+		case cpu_to_le16(IEEE80211_STYPE_AUTH):
+		case cpu_to_le16(IEEE80211_STYPE_DEAUTH):
+		case cpu_to_le16(IEEE80211_STYPE_ASSOC_REQ):
+		case cpu_to_le16(IEEE80211_STYPE_REASSOC_REQ):
 			if (tx_cmd->tx_flags & TX_CMD_FLG_RTS_MSK) {
 				tx_cmd->tx_flags &= ~TX_CMD_FLG_RTS_MSK;
 				tx_cmd->tx_flags |= TX_CMD_FLG_CTS_MSK;
@@ -771,7 +770,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	u16 seq_number = 0;
 	u8 id, hdr_len, unicast;
 	u8 sta_id;
-	u16 fc;
+	__le16 fc;
 	u8 wait_write_ptr = 0;
 	u8 tid = 0;
 	u8 *qc = NULL;
@@ -798,19 +797,19 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	unicast = !is_multicast_ether_addr(hdr->addr1);
 	id = 0;
 
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 
 #ifdef CONFIG_IWLWIFI_DEBUG
 	if (ieee80211_is_auth(fc))
 		IWL_DEBUG_TX("Sending AUTH frame\n");
-	else if (ieee80211_is_assoc_request(fc))
+	else if (ieee80211_is_assoc_req(fc))
 		IWL_DEBUG_TX("Sending ASSOC frame\n");
-	else if (ieee80211_is_reassoc_request(fc))
+	else if (ieee80211_is_reassoc_req(fc))
 		IWL_DEBUG_TX("Sending REASSOC frame\n");
 #endif
 
 	/* drop all data frame if we are not associated */
-	if (((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) &&
+	if (ieee80211_is_data(fc) &&
 	   (!iwl_is_associated(priv) ||
 	    ((priv->iw_mode == IEEE80211_IF_TYPE_STA) && !priv->assoc_id) ||
 	    !priv->assoc_station_added)) {
@@ -820,7 +819,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	hdr_len = ieee80211_get_hdrlen(fc);
+	hdr_len = ieee80211_get_hdrlen(le16_to_cpu(fc));
 
 	/* Find (or create) index into station table for destination station */
 	sta_id = iwl_get_sta_id(priv, hdr);
@@ -834,8 +833,8 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 
 	IWL_DEBUG_TX("station Id %d\n", sta_id);
 
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, hdr_len);
+	if (ieee80211_is_data_qos(fc)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 		seq_number = priv->stations[sta_id].tid[tid].seq_number &
 				IEEE80211_SCTL_SEQ;
@@ -938,7 +937,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	/* set is_hcca to 0; it probably will never be implemented */
 	iwl_tx_cmd_build_rate(priv, tx_cmd, info, fc, sta_id, 0);
 
-	iwl_update_tx_stats(priv, fc, len);
+	iwl_update_tx_stats(priv, le16_to_cpu(fc), len);
 
 	scratch_phys = txcmd_phys + sizeof(struct iwl_cmd_header) +
 		offsetof(struct iwl_tx_cmd, scratch);
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index d0084bcb1eca..0a5bbe9ee938 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2431,15 +2431,15 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 				  struct ieee80211_hdr *hdr,
 				  int is_unicast, u8 std_id)
 {
-	u16 fc = le16_to_cpu(hdr->frame_control);
+	__le16 fc = hdr->frame_control;
 	__le32 tx_flags = cmd->cmd.tx.tx_flags;
 
 	cmd->cmd.tx.stop_time.life_time = TX_CMD_LIFE_TIME_INFINITE;
 	if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		tx_flags |= TX_CMD_FLG_ACK_MSK;
-		if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)
+		if (ieee80211_is_mgmt(fc))
 			tx_flags |= TX_CMD_FLG_SEQ_CTL_MSK;
-		if (ieee80211_is_probe_response(fc) &&
+		if (ieee80211_is_probe_resp(fc) &&
 		    !(le16_to_cpu(hdr->seq_ctrl) & 0xf))
 			tx_flags |= TX_CMD_FLG_TSF_MSK;
 	} else {
@@ -2451,8 +2451,8 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 	if (ieee80211_get_morefrag(hdr))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
-	if (ieee80211_is_qos_data(fc)) {
-		u8 *qc = ieee80211_get_qos_ctrl(hdr, ieee80211_get_hdrlen(fc));
+	if (ieee80211_is_data_qos(fc)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		cmd->cmd.tx.tid_tspec = qc[0] & 0xf;
 		tx_flags &= ~TX_CMD_FLG_SEQ_CTL_MSK;
 	} else {
@@ -2471,9 +2471,8 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 		tx_flags |= TX_CMD_FLG_FULL_TXOP_PROT_MSK;
 
 	tx_flags &= ~(TX_CMD_FLG_ANT_SEL_MSK);
-	if ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT) {
-		if ((fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_ASSOC_REQ ||
-		    (fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_REASSOC_REQ)
+	if (ieee80211_is_mgmt(fc)) {
+		if (ieee80211_is_assoc_req(fc) || ieee80211_is_reassoc_req(fc))
 			cmd->cmd.tx.timeout.pm_frame_timeout = cpu_to_le16(3);
 		else
 			cmd->cmd.tx.timeout.pm_frame_timeout = cpu_to_le16(2);
@@ -2564,7 +2563,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 	u8 sta_id;
 	u8 tid = 0;
 	u16 seq_number = 0;
-	u16 fc;
+	__le16 fc;
 	u8 wait_write_ptr = 0;
 	u8 *qc = NULL;
 	unsigned long flags;
@@ -2589,28 +2588,28 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 	unicast = !is_multicast_ether_addr(hdr->addr1);
 	id = 0;
 
-	fc = le16_to_cpu(hdr->frame_control);
+	fc = hdr->frame_control;
 
 #ifdef CONFIG_IWL3945_DEBUG
 	if (ieee80211_is_auth(fc))
 		IWL_DEBUG_TX("Sending AUTH frame\n");
-	else if (ieee80211_is_assoc_request(fc))
+	else if (ieee80211_is_assoc_req(fc))
 		IWL_DEBUG_TX("Sending ASSOC frame\n");
-	else if (ieee80211_is_reassoc_request(fc))
+	else if (ieee80211_is_reassoc_req(fc))
 		IWL_DEBUG_TX("Sending REASSOC frame\n");
 #endif
 
 	/* drop all data frame if we are not associated */
 	if ((!iwl3945_is_associated(priv) ||
 	     ((priv->iw_mode == IEEE80211_IF_TYPE_STA) && !priv->assoc_id)) &&
-	    ((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA)) {
+	    ieee80211_is_data(fc)) {
 		IWL_DEBUG_DROP("Dropping - !iwl3945_is_associated\n");
 		goto drop_unlock;
 	}
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	hdr_len = ieee80211_get_hdrlen(fc);
+	hdr_len = ieee80211_get_hdrlen(le16_to_cpu(fc));
 
 	/* Find (or create) index into station table for destination station */
 	sta_id = iwl3945_get_sta_id(priv, hdr);
@@ -2624,8 +2623,8 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 
 	IWL_DEBUG_RATE("station Id %d\n", sta_id);
 
-	if (ieee80211_is_qos_data(fc)) {
-		qc = ieee80211_get_qos_ctrl(hdr, hdr_len);
+	if (ieee80211_is_data_qos(fc)) {
+		qc = ieee80211_get_qos_ctl(hdr);
 		tid = qc[0] & 0xf;
 		seq_number = priv->stations[sta_id].tid[tid].seq_number &
 				IEEE80211_SCTL_SEQ;
@@ -2746,7 +2745,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 			   sizeof(out_cmd->cmd.tx));
 
 	iwl3945_print_hex_dump(IWL_DL_TX, (u8 *)out_cmd->cmd.tx.hdr,
-			   ieee80211_get_hdrlen(fc));
+			   ieee80211_get_hdrlen(le16_to_cpu(fc)));
 
 	/* Tell device the write index *just past* this latest filled TFD */
 	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr, q->n_bd);
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 8f2c20b4a15d..371237b0d8b9 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -98,6 +98,7 @@
 
 #define IEEE80211_MAX_SSID_LEN		32
 #define IEEE80211_MAX_MESH_ID_LEN	32
+#define IEEE80211_QOS_CTL_LEN		2
 
 struct ieee80211_hdr {
 	__le16 frame_control;
@@ -109,6 +110,355 @@ struct ieee80211_hdr {
 	u8 addr4[6];
 } __attribute__ ((packed));
 
+/**
+ * ieee80211_has_tods - check if IEEE80211_FCTL_TODS is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_tods(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_TODS)) != 0;
+}
+
+/**
+ * ieee80211_has_fromds - check if IEEE80211_FCTL_FROMDS is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_fromds(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FROMDS)) != 0;
+}
+
+/**
+ * ieee80211_has_a4 - check if IEEE80211_FCTL_TODS and IEEE80211_FCTL_FROMDS are set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_a4(__le16 fc)
+{
+	__le16 tmp = cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS);
+	return (fc & tmp) == tmp;
+}
+
+/**
+ * ieee80211_has_morefrags - check if IEEE80211_FCTL_MOREFRAGS is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_morefrags(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_MOREFRAGS)) != 0;
+}
+
+/**
+ * ieee80211_has_retry - check if IEEE80211_FCTL_RETRY is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_retry(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_RETRY)) != 0;
+}
+
+/**
+ * ieee80211_has_pm - check if IEEE80211_FCTL_PM is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_pm(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_PM)) != 0;
+}
+
+/**
+ * ieee80211_has_moredata - check if IEEE80211_FCTL_MOREDATA is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_moredata(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_MOREDATA)) != 0;
+}
+
+/**
+ * ieee80211_has_protected - check if IEEE80211_FCTL_PROTECTED is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_protected(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_PROTECTED)) != 0;
+}
+
+/**
+ * ieee80211_has_order - check if IEEE80211_FCTL_ORDER is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_has_order(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_ORDER)) != 0;
+}
+
+/**
+ * ieee80211_is_mgmt - check if type is IEEE80211_FTYPE_MGMT
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_mgmt(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT);
+}
+
+/**
+ * ieee80211_is_ctl - check if type is IEEE80211_FTYPE_CTL
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_ctl(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL);
+}
+
+/**
+ * ieee80211_is_data - check if type is IEEE80211_FTYPE_DATA
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_data(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA);
+}
+
+/**
+ * ieee80211_is_data_qos - check if type is IEEE80211_FTYPE_DATA and IEEE80211_STYPE_QOS_DATA is set
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_data_qos(__le16 fc)
+{
+	/*
+	 * mask with QOS_DATA rather than IEEE80211_FCTL_STYPE as we just need
+	 * to check the one bit
+	 */
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_STYPE_QOS_DATA)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA | IEEE80211_STYPE_QOS_DATA);
+}
+
+/**
+ * ieee80211_is_data_present - check if type is IEEE80211_FTYPE_DATA and has data
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_data_present(__le16 fc)
+{
+	/*
+	 * mask with 0x40 and test that that bit is clear to only return true
+	 * for the data-containing substypes.
+	 */
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | 0x40)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA);
+}
+
+/**
+ * ieee80211_is_assoc_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ASSOC_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_assoc_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ASSOC_REQ);
+}
+
+/**
+ * ieee80211_is_assoc_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ASSOC_RESP
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_assoc_resp(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ASSOC_RESP);
+}
+
+/**
+ * ieee80211_is_reassoc_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_REASSOC_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_reassoc_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_REASSOC_REQ);
+}
+
+/**
+ * ieee80211_is_reassoc_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_REASSOC_RESP
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_reassoc_resp(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_REASSOC_RESP);
+}
+
+/**
+ * ieee80211_is_probe_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_PROBE_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_probe_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_REQ);
+}
+
+/**
+ * ieee80211_is_probe_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_PROBE_RESP
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_probe_resp(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_PROBE_RESP);
+}
+
+/**
+ * ieee80211_is_beacon - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_BEACON
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_beacon(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_BEACON);
+}
+
+/**
+ * ieee80211_is_atim - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ATIM
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_atim(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ATIM);
+}
+
+/**
+ * ieee80211_is_disassoc - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_DISASSOC
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_disassoc(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_DISASSOC);
+}
+
+/**
+ * ieee80211_is_auth - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_AUTH
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_auth(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_AUTH);
+}
+
+/**
+ * ieee80211_is_deauth - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_DEAUTH
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_deauth(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_DEAUTH);
+}
+
+/**
+ * ieee80211_is_action - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ACTION
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_action(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ACTION);
+}
+
+/**
+ * ieee80211_is_back_req - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_BACK_REQ
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_back_req(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_BACK_REQ);
+}
+
+/**
+ * ieee80211_is_back - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_BACK
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_back(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_BACK);
+}
+
+/**
+ * ieee80211_is_pspoll - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_PSPOLL
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_pspoll(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_PSPOLL);
+}
+
+/**
+ * ieee80211_is_rts - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_RTS
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_rts(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_RTS);
+}
+
+/**
+ * ieee80211_is_cts - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CTS
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_cts(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_CTS);
+}
+
+/**
+ * ieee80211_is_ack - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_ACK
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_ack(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_ACK);
+}
+
+/**
+ * ieee80211_is_cfend - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CFEND
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_cfend(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_CFEND);
+}
+
+/**
+ * ieee80211_is_cfendack - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CFENDACK
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_cfendack(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_CTL | IEEE80211_STYPE_CFENDACK);
+}
+
+/**
+ * ieee80211_is_nullfunc - check if FTYPE=IEEE80211_FTYPE_DATA and STYPE=IEEE80211_STYPE_NULLFUNC
+ * @fc: frame control bytes in little-endian byteorder
+ */
+static inline int ieee80211_is_nullfunc(__le16 fc)
+{
+	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
+	       cpu_to_le16(IEEE80211_FTYPE_DATA | IEEE80211_STYPE_NULLFUNC);
+}
 
 struct ieee80211s_hdr {
 	u8 flags;
@@ -552,49 +902,55 @@ enum ieee80211_back_parties {
 
 #define WLAN_MAX_KEY_LEN		32
 
+/**
+ * ieee80211_get_qos_ctl - get pointer to qos control bytes
+ * @hdr: the frame
+ *
+ * The qos ctrl bytes come after the frame_control, duration, seq_num
+ * and 3 or 4 addresses of length ETH_ALEN.
+ * 3 addr: 2 + 2 + 2 + 3*6 = 24
+ * 4 addr: 2 + 2 + 2 + 4*6 = 30
+ */
+static inline u8 *ieee80211_get_qos_ctl(struct ieee80211_hdr *hdr)
+{
+	if (ieee80211_has_a4(hdr->frame_control))
+		return (u8 *)hdr + 30;
+	else
+		return (u8 *)hdr + 24;
+}
+
 /**
  * ieee80211_get_SA - get pointer to SA
+ * @hdr: the frame
  *
  * Given an 802.11 frame, this function returns the offset
  * to the source address (SA). It does not verify that the
  * header is long enough to contain the address, and the
  * header must be long enough to contain the frame control
  * field.
- *
- * @hdr: the frame
  */
 static inline u8 *ieee80211_get_SA(struct ieee80211_hdr *hdr)
 {
-	__le16 fc = hdr->frame_control;
-	fc &= cpu_to_le16(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS);
-
-	switch (fc) {
-	case __constant_cpu_to_le16(IEEE80211_FCTL_FROMDS):
-		return hdr->addr3;
-	case __constant_cpu_to_le16(IEEE80211_FCTL_TODS|IEEE80211_FCTL_FROMDS):
+	if (ieee80211_has_a4(hdr->frame_control))
 		return hdr->addr4;
-	default:
-		return hdr->addr2;
-	}
+	if (ieee80211_has_fromds(hdr->frame_control))
+		return hdr->addr3;
+	return hdr->addr2;
 }
 
 /**
  * ieee80211_get_DA - get pointer to DA
+ * @hdr: the frame
  *
  * Given an 802.11 frame, this function returns the offset
  * to the destination address (DA). It does not verify that
  * the header is long enough to contain the address, and the
  * header must be long enough to contain the frame control
  * field.
- *
- * @hdr: the frame
  */
 static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 {
-	__le16 fc = hdr->frame_control;
-	fc &= cpu_to_le16(IEEE80211_FCTL_TODS);
-
-	if (fc)
+	if (ieee80211_has_tods(hdr->frame_control))
 		return hdr->addr3;
 	else
 		return hdr->addr1;
-- 
cgit v1.2.3


From 8b7b1e05b0454f232b8ae1e6ee134b7f0b38abfb Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 11 Jun 2008 14:21:56 -0700
Subject: mac80211: remove ieee80211_get_morefrag

Replaced by the new helper ieee80211_has_morefrags which is
more consistent with the intent of the function.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwlwifi/iwl-tx.c       |  4 ++--
 drivers/net/wireless/iwlwifi/iwl3945-base.c |  4 ++--
 drivers/net/wireless/rt2x00/rt2x00queue.c   |  2 +-
 drivers/net/wireless/rtl8187_dev.c          |  2 +-
 include/linux/ieee80211.h                   | 14 --------------
 5 files changed, 6 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index 98c434c52a64..7296e2846ec3 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -590,7 +590,7 @@ static void iwl_tx_cmd_build_basic(struct iwl_priv *priv,
 
 
 	tx_cmd->sta_id = std_id;
-	if (ieee80211_get_morefrag(hdr))
+	if (ieee80211_has_morefrags(fc))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
 	if (ieee80211_is_data_qos(fc)) {
@@ -944,7 +944,7 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	tx_cmd->dram_lsb_ptr = cpu_to_le32(scratch_phys);
 	tx_cmd->dram_msb_ptr = iwl_get_dma_hi_address(scratch_phys);
 
-	if (!ieee80211_get_morefrag(hdr)) {
+	if (!ieee80211_has_morefrags(hdr->frame_control)) {
 		txq->need_update = 1;
 		if (qc)
 			priv->stations[sta_id].tid[tid].seq_number = seq_number;
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 0a5bbe9ee938..47cf4b997f50 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -2448,7 +2448,7 @@ static void iwl3945_build_tx_cmd_basic(struct iwl3945_priv *priv,
 	}
 
 	cmd->cmd.tx.sta_id = std_id;
-	if (ieee80211_get_morefrag(hdr))
+	if (ieee80211_has_morefrags(fc))
 		tx_flags |= TX_CMD_FLG_MORE_FRAG_MSK;
 
 	if (ieee80211_is_data_qos(fc)) {
@@ -2731,7 +2731,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, struct sk_buff *skb)
 	out_cmd->cmd.tx.tx_flags &= ~TX_CMD_FLG_ANT_A_MSK;
 	out_cmd->cmd.tx.tx_flags &= ~TX_CMD_FLG_ANT_B_MSK;
 
-	if (!ieee80211_get_morefrag(hdr)) {
+	if (!ieee80211_has_morefrags(hdr->frame_control)) {
 		txq->need_update = 1;
 		if (qc) {
 			priv->stations[sta_id].tid[tid].seq_number = seq_number;
diff --git a/drivers/net/wireless/rt2x00/rt2x00queue.c b/drivers/net/wireless/rt2x00/rt2x00queue.c
index 2f3dd1d91a12..7b52039b01a6 100644
--- a/drivers/net/wireless/rt2x00/rt2x00queue.c
+++ b/drivers/net/wireless/rt2x00/rt2x00queue.c
@@ -130,7 +130,7 @@ void rt2x00queue_create_tx_descriptor(struct queue_entry *entry,
 	/*
 	 * Check if more fragments are pending
 	 */
-	if (ieee80211_get_morefrag(hdr)) {
+	if (ieee80211_has_morefrags(hdr->frame_control)) {
 		__set_bit(ENTRY_TXD_BURST, &txdesc->flags);
 		__set_bit(ENTRY_TXD_MORE_FRAG, &txdesc->flags);
 	}
diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c
index 0078c7e9918c..bec96d762c6c 100644
--- a/drivers/net/wireless/rtl8187_dev.c
+++ b/drivers/net/wireless/rtl8187_dev.c
@@ -181,7 +181,7 @@ static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
 	flags |= RTL8187_TX_FLAG_NO_ENCRYPT;
 
 	flags |= ieee80211_get_tx_rate(dev, info)->hw_value << 24;
-	if (ieee80211_get_morefrag((struct ieee80211_hdr *)skb->data))
+	if (ieee80211_has_morefrags(((struct ieee80211_hdr *)skb->data)->frame_control))
 		flags |= RTL8187_TX_FLAG_MORE_FRAG;
 	if (info->flags & IEEE80211_TX_CTL_USE_RTS_CTS) {
 		flags |= RTL8187_TX_FLAG_RTS;
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 371237b0d8b9..2998e3b5f166 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -956,18 +956,4 @@ static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 		return hdr->addr1;
 }
 
-/**
- * ieee80211_get_morefrag - determine whether the MOREFRAGS bit is set
- *
- * This function determines whether the "more fragments" bit is set
- * in the frame.
- *
- * @hdr: the frame
- */
-static inline int ieee80211_get_morefrag(struct ieee80211_hdr *hdr)
-{
-	__le16 fc = hdr->frame_control;
-	return !!(fc & cpu_to_le16(IEEE80211_FCTL_MOREFRAGS));
-}
-
 #endif /* IEEE80211_H */
-- 
cgit v1.2.3


From d6266281f8175e3ad68c28b20a609b278b47ade5 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:11:50 -0700
Subject: udp: introduce a udp_hashfn function

Currently the chain to store a UDP socket is calculated with
simple (x & (UDP_HTABLE_SIZE - 1)). But taking net into account
would make this calculation a bit more complex, so moving it into
a function would help.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h |  5 +++++
 net/ipv4/udp.c      | 12 ++++++------
 net/ipv6/udp.c      |  4 ++--
 3 files changed, 13 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 581ca2c14c52..9c94312b2de5 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -46,6 +46,11 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 
 #define UDP_HTABLE_SIZE		128
 
+static inline int udp_hashfn(const unsigned num)
+{
+	return num & (UDP_HTABLE_SIZE - 1);
+}
+
 struct udp_sock {
 	/* inet_sock has to be the first member */
 	struct inet_sock inet;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index eba790dcd16b..d8f527d15701 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -134,7 +134,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
 	struct sock *sk;
 	struct hlist_node *node;
 
-	sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)])
+	sk_for_each(sk, node, &udptable[udp_hashfn(num)])
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == num)
 			return 1;
 	return 0;
@@ -174,7 +174,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 		for (i = 0; i < UDP_HTABLE_SIZE; i++) {
 			int size = 0;
 
-			head = &udptable[rover & (UDP_HTABLE_SIZE - 1)];
+			head = &udptable[udp_hashfn(rover)];
 			if (hlist_empty(head))
 				goto gotit;
 
@@ -211,7 +211,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 gotit:
 		snum = rover;
 	} else {
-		head = &udptable[snum & (UDP_HTABLE_SIZE - 1)];
+		head = &udptable[udp_hashfn(snum)];
 
 		sk_for_each(sk2, node, head)
 			if (sk2->sk_hash == snum                             &&
@@ -227,7 +227,7 @@ gotit:
 	inet_sk(sk)->num = snum;
 	sk->sk_hash = snum;
 	if (sk_unhashed(sk)) {
-		head = &udptable[snum & (UDP_HTABLE_SIZE - 1)];
+		head = &udptable[udp_hashfn(snum)];
 		sk_add_node(sk, head);
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	}
@@ -264,7 +264,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -1068,7 +1068,7 @@ static int __udp4_lib_mcast_deliver(struct sk_buff *skb,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
+	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
 	dif = skb->dev->ifindex;
 	sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (sk) {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 09687f7a8564..6e4a822ba651 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -65,7 +65,7 @@ static struct sock *__udp6_lib_lookup(struct net *net,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -361,7 +361,7 @@ static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]);
+	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
 	dif = inet6_iif(skb);
 	sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (!sk) {
-- 
cgit v1.2.3


From 19c7578fb22b0aef103222cae9b522f03ae489d6 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:12:29 -0700
Subject: udp: add struct net argument to udp_hashfn

Every caller already has this one. The new argument is currently
unused, but this will be fixed shortly.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h |  2 +-
 net/ipv4/udp.c      | 12 ++++++------
 net/ipv6/udp.c      |  4 ++--
 3 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 9c94312b2de5..3deccac2e815 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -46,7 +46,7 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 
 #define UDP_HTABLE_SIZE		128
 
-static inline int udp_hashfn(const unsigned num)
+static inline int udp_hashfn(struct net *net, const unsigned num)
 {
 	return num & (UDP_HTABLE_SIZE - 1);
 }
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 6b0acb438f34..11eabf136144 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -134,7 +134,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
 	struct sock *sk;
 	struct hlist_node *node;
 
-	sk_for_each(sk, node, &udptable[udp_hashfn(num)])
+	sk_for_each(sk, node, &udptable[udp_hashfn(net, num)])
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == num)
 			return 1;
 	return 0;
@@ -174,7 +174,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 		for (i = 0; i < UDP_HTABLE_SIZE; i++) {
 			int size = 0;
 
-			head = &udptable[udp_hashfn(rover)];
+			head = &udptable[udp_hashfn(net, rover)];
 			if (hlist_empty(head))
 				goto gotit;
 
@@ -211,7 +211,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 gotit:
 		snum = rover;
 	} else {
-		head = &udptable[udp_hashfn(snum)];
+		head = &udptable[udp_hashfn(net, snum)];
 
 		sk_for_each(sk2, node, head)
 			if (sk2->sk_hash == snum                             &&
@@ -227,7 +227,7 @@ gotit:
 	inet_sk(sk)->num = snum;
 	sk->sk_hash = snum;
 	if (sk_unhashed(sk)) {
-		head = &udptable[udp_hashfn(snum)];
+		head = &udptable[udp_hashfn(net, snum)];
 		sk_add_node(sk, head);
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	}
@@ -264,7 +264,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(net, hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -1068,7 +1068,7 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
+	sk = sk_head(&udptable[udp_hashfn(net, ntohs(uh->dest))]);
 	dif = skb->dev->ifindex;
 	sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (sk) {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 80fb72c48976..432edaa882f6 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -65,7 +65,7 @@ static struct sock *__udp6_lib_lookup(struct net *net,
 	int badness = -1;
 
 	read_lock(&udp_hash_lock);
-	sk_for_each(sk, node, &udptable[udp_hashfn(hnum)]) {
+	sk_for_each(sk, node, &udptable[udp_hashfn(net, hnum)]) {
 		struct inet_sock *inet = inet_sk(sk);
 
 		if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum &&
@@ -362,7 +362,7 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
 	int dif;
 
 	read_lock(&udp_hash_lock);
-	sk = sk_head(&udptable[udp_hashfn(ntohs(uh->dest))]);
+	sk = sk_head(&udptable[udp_hashfn(net, ntohs(uh->dest))]);
 	dif = inet6_iif(skb);
 	sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif);
 	if (!sk) {
-- 
cgit v1.2.3


From 0b4419162aa6c4204843f3a13b48d9ab821d3167 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Mon, 16 Jun 2008 17:14:11 -0700
Subject: netns: introduce the net_hash_mix "salt" for hashes

There are many possible ways to add this "salt", thus I made this
patch to be the last in the series to change it if required.

Currently I propose to use the struct net pointer itself as this
salt, but since this pointer is most often cache-line aligned, shift
this right to eliminate the bits, that are most often zeroed.

After this, simply add this mix to prepared hashfn-s.

For CONFIG_NET_NS=n case this salt is 0 and no changes in hashfn
appear.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/udp.h            |  3 ++-
 include/net/inet6_hashtables.h |  3 ++-
 include/net/inet_hashtables.h  |  5 +++--
 include/net/inet_sock.h        |  3 ++-
 include/net/netns/hash.h       | 21 +++++++++++++++++++++
 5 files changed, 30 insertions(+), 5 deletions(-)
 create mode 100644 include/net/netns/hash.h

(limited to 'include/linux')

diff --git a/include/linux/udp.h b/include/linux/udp.h
index 3deccac2e815..0cf5c4c0ec81 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -38,6 +38,7 @@ struct udphdr {
 #ifdef __KERNEL__
 #include <net/inet_sock.h>
 #include <linux/skbuff.h>
+#include <net/netns/hash.h>
 
 static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 {
@@ -48,7 +49,7 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb)
 
 static inline int udp_hashfn(struct net *net, const unsigned num)
 {
-	return num & (UDP_HTABLE_SIZE - 1);
+	return (num + net_hash_mix(net)) & (UDP_HTABLE_SIZE - 1);
 }
 
 struct udp_sock {
diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h
index 72f13a9928e4..e48989f04c24 100644
--- a/include/net/inet6_hashtables.h
+++ b/include/net/inet6_hashtables.h
@@ -24,6 +24,7 @@
 #include <net/inet_sock.h>
 
 #include <net/ipv6.h>
+#include <net/netns/hash.h>
 
 struct inet_hashinfo;
 
@@ -36,7 +37,7 @@ static inline unsigned int inet6_ehashfn(struct net *net,
 
 	return jhash_3words((__force u32)laddr->s6_addr32[3],
 			    (__force u32)faddr->s6_addr32[3],
-			    ports, inet_ehash_secret);
+			    ports, inet_ehash_secret + net_hash_mix(net));
 }
 
 static inline int inet6_sk_ehashfn(const struct sock *sk)
diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index 26336cdcdc11..bb619d80f2e2 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -29,6 +29,7 @@
 #include <net/inet_sock.h>
 #include <net/sock.h>
 #include <net/tcp_states.h>
+#include <net/netns/hash.h>
 
 #include <asm/atomic.h>
 #include <asm/byteorder.h>
@@ -204,7 +205,7 @@ extern void inet_bind_bucket_destroy(struct kmem_cache *cachep,
 static inline int inet_bhashfn(struct net *net,
 		const __u16 lport, const int bhash_size)
 {
-	return lport & (bhash_size - 1);
+	return (lport + net_hash_mix(net)) & (bhash_size - 1);
 }
 
 extern void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
@@ -213,7 +214,7 @@ extern void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb,
 /* These can have wildcards, don't try too hard. */
 static inline int inet_lhashfn(struct net *net, const unsigned short num)
 {
-	return num & (INET_LHTABLE_SIZE - 1);
+	return (num + net_hash_mix(net)) & (INET_LHTABLE_SIZE - 1);
 }
 
 static inline int inet_sk_listen_hashfn(const struct sock *sk)
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 85bb420c5d86..643e26be058e 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -25,6 +25,7 @@
 #include <net/sock.h>
 #include <net/request_sock.h>
 #include <net/route.h>
+#include <net/netns/hash.h>
 
 /** struct ip_options - IP Options
  *
@@ -178,7 +179,7 @@ static inline unsigned int inet_ehashfn(struct net *net,
 	return jhash_3words((__force __u32) laddr,
 			    (__force __u32) faddr,
 			    ((__u32) lport) << 16 | (__force __u32)fport,
-			    inet_ehash_secret);
+			    inet_ehash_secret + net_hash_mix(net));
 }
 
 static inline int inet_sk_ehashfn(const struct sock *sk)
diff --git a/include/net/netns/hash.h b/include/net/netns/hash.h
new file mode 100644
index 000000000000..548d78f2cc47
--- /dev/null
+++ b/include/net/netns/hash.h
@@ -0,0 +1,21 @@
+#ifndef __NET_NS_HASH_H__
+#define __NET_NS_HASH_H__
+
+#include <asm/cache.h>
+
+struct net;
+
+static inline unsigned net_hash_mix(struct net *net)
+{
+#ifdef CONFIG_NET_NS
+	/*
+	 * shift this right to eliminate bits, that are
+	 * always zeroed
+	 */
+
+	return (unsigned)(((unsigned long)net) >> L1_CACHE_SHIFT);
+#else
+	return 0;
+#endif
+}
+#endif
-- 
cgit v1.2.3


From 87de87d5e47f94b4ea647a5bd1bc8dc1f7930db4 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 3 Jun 2008 09:14:03 -0700
Subject: wext: Dispatch and handle compat ioctls entirely in
 net/wireless/wext.c

Next we can kill the hacks in fs/compat_ioctl.c and also
dispatch compat ioctls down into the driver and 80211 protocol
helper layers in order to handle iw_point objects embedded in
stream replies which need to be translated.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/compat_ioctl.c        |   6 ---
 include/linux/wireless.h |  13 ++++++
 include/net/wext.h       |   7 ++++
 net/socket.c             |  10 +++++
 net/wireless/wext.c      | 104 +++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 134 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 97dba0d92348..8ab850bf2eee 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1757,12 +1757,6 @@ static int do_i2c_smbus_ioctl(unsigned int fd, unsigned int cmd, unsigned long a
 	return sys_ioctl(fd, cmd, (unsigned long)tdata);
 }
 
-struct compat_iw_point {
-	compat_caddr_t pointer;
-	__u16 length;
-	__u16 flags;
-};
-
 static int do_wireless_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
 {
 	struct iwreq __user *iwr;
diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 4a95a0e5eeca..79d846875825 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -677,6 +677,19 @@ struct	iw_point
   __u16		flags;		/* Optional params */
 };
 
+#ifdef __KERNEL__
+#ifdef CONFIG_COMPAT
+
+#include <linux/compat.h>
+
+struct compat_iw_point {
+	compat_caddr_t pointer;
+	__u16 length;
+	__u16 flags;
+};
+#endif
+#endif
+
 /*
  *	A frequency
  *	For numbers lower than 10^9, we encode the number in 'm' and
diff --git a/include/net/wext.h b/include/net/wext.h
index 80b31d826b7a..6d76a39a9c5b 100644
--- a/include/net/wext.h
+++ b/include/net/wext.h
@@ -12,6 +12,8 @@ extern int wext_proc_init(struct net *net);
 extern void wext_proc_exit(struct net *net);
 extern int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned int cmd,
 			     void __user *arg);
+extern int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
+				    unsigned long arg);
 #else
 static inline int wext_proc_init(struct net *net)
 {
@@ -26,6 +28,11 @@ static inline int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned
 {
 	return -EINVAL;
 }
+static inline int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
+					   unsigned long arg)
+{
+	return -EINVAL;
+}
 #endif
 
 #endif /* __NET_WEXT_H */
diff --git a/net/socket.c b/net/socket.c
index 66c4a8cf6db9..81fe82513046 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -90,6 +90,7 @@
 #include <asm/unistd.h>
 
 #include <net/compat.h>
+#include <net/wext.h>
 
 #include <net/sock.h>
 #include <linux/netfilter.h>
@@ -2210,10 +2211,19 @@ static long compat_sock_ioctl(struct file *file, unsigned cmd,
 {
 	struct socket *sock = file->private_data;
 	int ret = -ENOIOCTLCMD;
+	struct sock *sk;
+	struct net *net;
+
+	sk = sock->sk;
+	net = sock_net(sk);
 
 	if (sock->ops->compat_ioctl)
 		ret = sock->ops->compat_ioctl(sock, cmd, arg);
 
+	if (ret == -ENOIOCTLCMD &&
+	    (cmd >= SIOCIWFIRST && cmd <= SIOCIWLAST))
+		ret = compat_wext_handle_ioctl(net, cmd, arg);
+
 	return ret;
 }
 #endif
diff --git a/net/wireless/wext.c b/net/wireless/wext.c
index 09022cbb58ba..1a4636a9fcde 100644
--- a/net/wireless/wext.c
+++ b/net/wireless/wext.c
@@ -1112,6 +1112,110 @@ int wext_handle_ioctl(struct net *net, struct ifreq *ifr, unsigned int cmd,
 	return ret;
 }
 
+#ifdef CONFIG_COMPAT
+static int compat_standard_call(struct net_device	*dev,
+				struct iwreq		*iwr,
+				unsigned int		cmd,
+				iw_handler		handler)
+{
+	const struct iw_ioctl_description *descr;
+	struct compat_iw_point *iwp_compat;
+	struct iw_request_info info;
+	struct iw_point iwp;
+	int err;
+
+	descr = standard_ioctl + (cmd - SIOCIWFIRST);
+
+	if (descr->header_type != IW_HEADER_TYPE_POINT)
+		return ioctl_standard_call(dev, iwr, cmd, handler);
+
+	iwp_compat = (struct compat_iw_point *) &iwr->u.data;
+	iwp.pointer = compat_ptr(iwp_compat->pointer);
+	iwp.length = iwp_compat->length;
+	iwp.flags = iwp_compat->flags;
+
+	info.cmd = cmd;
+	info.flags = 0;
+
+	err = ioctl_standard_iw_point(&iwp, cmd, descr, handler, dev, &info);
+
+	iwp_compat->pointer = ptr_to_compat(iwp.pointer);
+	iwp_compat->length = iwp.length;
+	iwp_compat->flags = iwp.flags;
+
+	return err;
+}
+
+static int compat_private_call(struct net_device *dev, struct iwreq *iwr,
+			       unsigned int cmd, iw_handler handler)
+{
+	const struct iw_priv_args *descr;
+	struct iw_request_info info;
+	int ret, extra_size;
+
+	extra_size = get_priv_descr_and_size(dev, cmd, &descr);
+
+	/* Prepare the call */
+	info.cmd = cmd;
+	info.flags = 0;
+
+	/* Check if we have a pointer to user space data or not. */
+	if (extra_size == 0) {
+		/* No extra arguments. Trivial to handle */
+		ret = handler(dev, &info, &(iwr->u), (char *) &(iwr->u));
+	} else {
+		struct compat_iw_point *iwp_compat;
+		struct iw_point iwp;
+
+		iwp_compat = (struct compat_iw_point *) &iwr->u.data;
+		iwp.pointer = compat_ptr(iwp_compat->pointer);
+		iwp.length = iwp_compat->length;
+		iwp.flags = iwp_compat->flags;
+
+		ret = ioctl_private_iw_point(&iwp, cmd, descr,
+					     handler, dev, &info, extra_size);
+
+		iwp_compat->pointer = ptr_to_compat(iwp.pointer);
+		iwp_compat->length = iwp.length;
+		iwp_compat->flags = iwp.flags;
+	}
+
+	/* Call commit handler if needed and defined */
+	if (ret == -EIWCOMMIT)
+		ret = call_commit_handler(dev);
+
+	return ret;
+}
+
+int compat_wext_handle_ioctl(struct net *net, unsigned int cmd,
+			     unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	struct iwreq iwr;
+	char *colon;
+	int ret;
+
+	if (copy_from_user(&iwr, argp, sizeof(struct iwreq)))
+		return -EFAULT;
+
+	iwr.ifr_name[IFNAMSIZ-1] = 0;
+	colon = strchr(iwr.ifr_name, ':');
+	if (colon)
+		*colon = 0;
+
+	ret = wext_ioctl_dispatch(net, (struct ifreq *) &iwr, cmd,
+				  compat_standard_call,
+				  compat_private_call);
+
+	if (ret >= 0 &&
+	    IW_IS_GET(cmd) &&
+	    copy_to_user(argp, &iwr, sizeof(struct iwreq)))
+		return -EFAULT;
+
+	return ret;
+}
+#endif
+
 /************************* EVENT PROCESSING *************************/
 /*
  * Process events generated by the wireless layer or the driver.
-- 
cgit v1.2.3


From ccc580571cf0799d0460a085a7632b77753f083e Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 16 Jun 2008 18:50:49 -0700
Subject: wext: Emit event stream entries correctly when compat.

Three major portions to this change:

1) Add IW_EV_COMPAT_LCP_LEN, IW_EV_COMPAT_POINT_OFF,
   and IW_EV_COMPAT_POINT_LEN helper defines.

2) Delete iw_stream_check_add_*(), they are unused.

3) Add iw_request_info argument to iwe_stream_add_*(), and use it to
   size the event and pointer lengths correctly depending upon whether
   IW_REQUEST_FLAG_COMPAT is set or not.

4) The mechanical transformations to the drivers and wireless stack
   bits to get the iw_request_info passed down into the routines
   modified in #3.  Also, explicit references to IW_EV_LCP_LEN are
   replaced with iwe_stream_lcp_len(info).

With a lot of help and bug fixes from Masakazu Mokuno.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ps3_gelic_wireless.c           |  30 +++---
 drivers/net/wireless/airo.c                |  43 ++++++---
 drivers/net/wireless/atmel.c               |  24 +++--
 drivers/net/wireless/hostap/hostap.h       |   3 +-
 drivers/net/wireless/hostap/hostap_ap.c    |  32 +++----
 drivers/net/wireless/hostap/hostap_ioctl.c |  58 +++++------
 drivers/net/wireless/libertas/scan.c       |  36 +++----
 drivers/net/wireless/orinoco.c             |  30 +++---
 drivers/net/wireless/prism54/isl_ioctl.c   |  49 +++++-----
 drivers/net/wireless/rndis_wlan.c          |  32 ++++---
 drivers/net/wireless/wl3501_cs.c           |  10 +-
 drivers/net/wireless/zd1201.c              |  21 ++--
 include/linux/wireless.h                   |  15 +++
 include/net/iw_handler.h                   | 149 ++++++++++-------------------
 net/ieee80211/ieee80211_wx.c               |  48 ++++++----
 net/mac80211/ieee80211_i.h                 |   5 +-
 net/mac80211/mlme.c                        |  66 +++++++------
 net/mac80211/wext.c                        |   2 +-
 18 files changed, 345 insertions(+), 308 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/ps3_gelic_wireless.c b/drivers/net/ps3_gelic_wireless.c
index aa963ac1e37b..6b2dee0cf3a9 100644
--- a/drivers/net/ps3_gelic_wireless.c
+++ b/drivers/net/ps3_gelic_wireless.c
@@ -571,6 +571,7 @@ static void gelic_wl_parse_ie(u8 *data, size_t len,
  * independent format
  */
 static char *gelic_wl_translate_scan(struct net_device *netdev,
+				     struct iw_request_info *info,
 				     char *ev,
 				     char *stop,
 				     struct gelic_wl_scan_info *network)
@@ -588,26 +589,26 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, &scan->bssid[2], ETH_ALEN);
-	ev = iwe_stream_add_event(ev, stop, &iwe, IW_EV_ADDR_LEN);
+	ev = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_ADDR_LEN);
 
 	/* ESSID */
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
 	iwe.u.data.length = strnlen(scan->essid, 32);
-	ev = iwe_stream_add_point(ev, stop, &iwe, scan->essid);
+	ev = iwe_stream_add_point(info, ev, stop, &iwe, scan->essid);
 
 	/* FREQUENCY */
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = be16_to_cpu(scan->channel);
 	iwe.u.freq.e = 0; /* table value in MHz */
 	iwe.u.freq.i = 0;
-	ev = iwe_stream_add_event(ev, stop, &iwe, IW_EV_FREQ_LEN);
+	ev = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_FREQ_LEN);
 
 	/* RATES */
 	iwe.cmd = SIOCGIWRATE;
 	iwe.u.bitrate.fixed = iwe.u.bitrate.disabled = 0;
 	/* to stuff multiple values in one event */
-	tmp = ev + IW_EV_LCP_LEN;
+	tmp = ev + iwe_stream_lcp_len(info);
 	/* put them in ascendant order (older is first) */
 	i = 0;
 	j = 0;
@@ -620,16 +621,16 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 		else
 		    rate = scan->rate[i++] & 0x7f;
 		iwe.u.bitrate.value = rate * 500000; /* 500kbps unit */
-		tmp = iwe_stream_add_value(ev, tmp, stop, &iwe,
+		tmp = iwe_stream_add_value(info, ev, tmp, stop, &iwe,
 					   IW_EV_PARAM_LEN);
 	}
 	while (j < network->rate_ext_len) {
 		iwe.u.bitrate.value = (scan->ext_rate[j++] & 0x7f) * 500000;
-		tmp = iwe_stream_add_value(ev, tmp, stop, &iwe,
+		tmp = iwe_stream_add_value(info, ev, tmp, stop, &iwe,
 					   IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any rate */
-	if (IW_EV_LCP_LEN < (tmp - ev))
+	if (iwe_stream_lcp_len(info) < (tmp - ev))
 		ev = tmp;
 
 	/* ENCODE */
@@ -639,7 +640,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	ev = iwe_stream_add_point(ev, stop, &iwe, scan->essid);
+	ev = iwe_stream_add_point(info, ev, stop, &iwe, scan->essid);
 
 	/* MODE */
 	iwe.cmd = SIOCGIWMODE;
@@ -649,7 +650,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		ev = iwe_stream_add_event(ev, stop, &iwe, IW_EV_UINT_LEN);
+		ev = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_UINT_LEN);
 	}
 
 	/* QUAL */
@@ -659,7 +660,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 	iwe.u.qual.level = be16_to_cpu(scan->rssi);
 	iwe.u.qual.qual = be16_to_cpu(scan->rssi);
 	iwe.u.qual.noise = 0;
-	ev  = iwe_stream_add_event(ev, stop, &iwe, IW_EV_QUAL_LEN);
+	ev  = iwe_stream_add_event(info, ev, stop, &iwe, IW_EV_QUAL_LEN);
 
 	/* RSN */
 	memset(&iwe, 0, sizeof(iwe));
@@ -669,7 +670,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 		if (len) {
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = len;
-			ev = iwe_stream_add_point(ev, stop, &iwe, buf);
+			ev = iwe_stream_add_point(info, ev, stop, &iwe, buf);
 		}
 	} else {
 		/* this scan info has IE data */
@@ -684,7 +685,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 			memcpy(buf, ie_info.wpa.data, ie_info.wpa.len);
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = ie_info.wpa.len;
-			ev = iwe_stream_add_point(ev, stop, &iwe, buf);
+			ev = iwe_stream_add_point(info, ev, stop, &iwe, buf);
 		}
 
 		if (ie_info.rsn.len && (ie_info.rsn.len <= sizeof(buf))) {
@@ -692,7 +693,7 @@ static char *gelic_wl_translate_scan(struct net_device *netdev,
 			memcpy(buf, ie_info.rsn.data, ie_info.rsn.len);
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = ie_info.rsn.len;
-			ev = iwe_stream_add_point(ev, stop, &iwe, buf);
+			ev = iwe_stream_add_point(info, ev, stop, &iwe, buf);
 		}
 	}
 
@@ -737,7 +738,8 @@ static int gelic_wl_get_scan(struct net_device *netdev,
 		if (wl->scan_age == 0 ||
 		    time_after(scan_info->last_scanned + wl->scan_age,
 			       this_time))
-			ev = gelic_wl_translate_scan(netdev, ev, stop,
+			ev = gelic_wl_translate_scan(netdev, info,
+						     ev, stop,
 						     scan_info);
 		else
 			pr_debug("%s:entry too old\n", __func__);
diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c
index e30f8b79ea89..73d66a80c4a3 100644
--- a/drivers/net/wireless/airo.c
+++ b/drivers/net/wireless/airo.c
@@ -7156,6 +7156,7 @@ out:
  * format that the Wireless Tools will understand - Jean II
  */
 static inline char *airo_translate_scan(struct net_device *dev,
+					struct iw_request_info *info,
 					char *current_ev,
 					char *end_buf,
 					BSSListRid *bss)
@@ -7172,7 +7173,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bss->bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_ADDR_LEN);
 
 	/* Other entries will be displayed in the order we give them */
 
@@ -7182,7 +7184,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 		iwe.u.data.length = 32;
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->ssid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->ssid);
 
 	/* Add mode */
 	iwe.cmd = SIOCGIWMODE;
@@ -7192,7 +7195,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	/* Add frequency */
@@ -7203,7 +7207,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 	 */
 	iwe.u.freq.m = frequency_list[iwe.u.freq.m - 1] * 100000;
 	iwe.u.freq.e = 1;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_FREQ_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_FREQ_LEN);
 
 	dBm = le16_to_cpu(bss->dBm);
 
@@ -7223,7 +7228,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 				| IW_QUAL_DBM;
 	}
 	iwe.u.qual.noise = ai->wstats.qual.noise;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_QUAL_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -7232,11 +7238,12 @@ static inline char *airo_translate_scan(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->ssid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->ssid);
 
 	/* Rate : stuffing multiple values in a single event require a bit
 	 * more of magic - Jean II */
-	current_val = current_ev + IW_EV_LCP_LEN;
+	current_val = current_ev + iwe_stream_lcp_len(info);
 
 	iwe.cmd = SIOCGIWRATE;
 	/* Those two flags are ignored... */
@@ -7249,10 +7256,12 @@ static inline char *airo_translate_scan(struct net_device *dev,
 		/* Bit rate given in 500 kb/s units (+ 0x80) */
 		iwe.u.bitrate.value = ((bss->rates[i] & 0x7f) * 500000);
 		/* Add new value to event */
-		current_val = iwe_stream_add_value(current_ev, current_val, end_buf, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, current_ev,
+						   current_val, end_buf,
+						   &iwe, IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any event */
-	if((current_val - current_ev) > IW_EV_LCP_LEN)
+	if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 		current_ev = current_val;
 
 	/* Beacon interval */
@@ -7261,7 +7270,8 @@ static inline char *airo_translate_scan(struct net_device *dev,
 		iwe.cmd = IWEVCUSTOM;
 		sprintf(buf, "bcn_int=%d", bss->beaconInterval);
 		iwe.u.data.length = strlen(buf);
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, buf);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, buf);
 		kfree(buf);
 	}
 
@@ -7295,8 +7305,10 @@ static inline char *airo_translate_scan(struct net_device *dev,
 					iwe.cmd = IWEVGENIE;
 					iwe.u.data.length = min(info_element->len + 2,
 								  MAX_WPA_IE_LEN);
-					current_ev = iwe_stream_add_point(current_ev, end_buf,
-							&iwe, (char *) info_element);
+					current_ev = iwe_stream_add_point(
+							info, current_ev,
+							end_buf, &iwe,
+							(char *) info_element);
 				}
 				break;
 
@@ -7304,8 +7316,9 @@ static inline char *airo_translate_scan(struct net_device *dev,
 				iwe.cmd = IWEVGENIE;
 				iwe.u.data.length = min(info_element->len + 2,
 							  MAX_WPA_IE_LEN);
-				current_ev = iwe_stream_add_point(current_ev, end_buf,
-						&iwe, (char *) info_element);
+				current_ev = iwe_stream_add_point(
+					info, current_ev, end_buf,
+					&iwe, (char *) info_element);
 				break;
 
 			default:
@@ -7344,7 +7357,7 @@ static int airo_get_scan(struct net_device *dev,
 
 	list_for_each_entry (net, &ai->network_list, list) {
 		/* Translate to WE format this entry */
-		current_ev = airo_translate_scan(dev, current_ev,
+		current_ev = airo_translate_scan(dev, info, current_ev,
 						 extra + dwrq->length,
 						 &net->bss);
 
diff --git a/drivers/net/wireless/atmel.c b/drivers/net/wireless/atmel.c
index 7bb2646ae0ef..28b6ff3eaa37 100644
--- a/drivers/net/wireless/atmel.c
+++ b/drivers/net/wireless/atmel.c
@@ -2310,30 +2310,40 @@ static int atmel_get_scan(struct net_device *dev,
 		iwe.cmd = SIOCGIWAP;
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, priv->BSSinfo[i].BSSID, 6);
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, IW_EV_ADDR_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_ADDR_LEN);
 
 		iwe.u.data.length =  priv->BSSinfo[i].SSIDsize;
 		if (iwe.u.data.length > 32)
 			iwe.u.data.length = 32;
 		iwe.cmd = SIOCGIWESSID;
 		iwe.u.data.flags = 1;
-		current_ev = iwe_stream_add_point(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, priv->BSSinfo[i].SSID);
+		current_ev = iwe_stream_add_point(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, priv->BSSinfo[i].SSID);
 
 		iwe.cmd = SIOCGIWMODE;
 		iwe.u.mode = priv->BSSinfo[i].BSStype;
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_UINT_LEN);
 
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = priv->BSSinfo[i].channel;
 		iwe.u.freq.e = 0;
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, IW_EV_FREQ_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_FREQ_LEN);
 
 		/* Add quality statistics */
 		iwe.cmd = IWEVQUAL;
 		iwe.u.qual.level = priv->BSSinfo[i].RSSI;
 		iwe.u.qual.qual  = iwe.u.qual.level;
 		/* iwe.u.qual.noise  = SOMETHING */
-		current_ev = iwe_stream_add_event(current_ev, extra + IW_SCAN_MAX_DATA , &iwe, IW_EV_QUAL_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, IW_EV_QUAL_LEN);
 
 
 		iwe.cmd = SIOCGIWENCODE;
@@ -2342,7 +2352,9 @@ static int atmel_get_scan(struct net_device *dev,
 		else
 			iwe.u.data.flags = IW_ENCODE_DISABLED;
 		iwe.u.data.length = 0;
-		current_ev = iwe_stream_add_point(current_ev, extra + IW_SCAN_MAX_DATA, &iwe, NULL);
+		current_ev = iwe_stream_add_point(info, current_ev,
+						  extra + IW_SCAN_MAX_DATA,
+						  &iwe, NULL);
 	}
 
 	/* Length of data */
diff --git a/drivers/net/wireless/hostap/hostap.h b/drivers/net/wireless/hostap/hostap.h
index 547ba84dc797..3a386a636cca 100644
--- a/drivers/net/wireless/hostap/hostap.h
+++ b/drivers/net/wireless/hostap/hostap.h
@@ -67,7 +67,8 @@ void * ap_crypt_get_ptrs(struct ap_data *ap, u8 *addr, int permanent,
 int prism2_ap_get_sta_qual(local_info_t *local, struct sockaddr addr[],
 			   struct iw_quality qual[], int buf_size,
 			   int aplist);
-int prism2_ap_translate_scan(struct net_device *dev, char *buffer);
+int prism2_ap_translate_scan(struct net_device *dev,
+			     struct iw_request_info *info, char *buffer);
 int prism2_hostapd(struct ap_data *ap, struct prism2_hostapd_param *param);
 
 
diff --git a/drivers/net/wireless/hostap/hostap_ap.c b/drivers/net/wireless/hostap/hostap_ap.c
index 0acd9589c48c..06b23df8f69b 100644
--- a/drivers/net/wireless/hostap/hostap_ap.c
+++ b/drivers/net/wireless/hostap/hostap_ap.c
@@ -2420,7 +2420,8 @@ int prism2_ap_get_sta_qual(local_info_t *local, struct sockaddr addr[],
 
 /* Translate our list of Access Points & Stations to a card independant
  * format that the Wireless Tools will understand - Jean II */
-int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
+int prism2_ap_translate_scan(struct net_device *dev,
+			     struct iw_request_info *info, char *buffer)
 {
 	struct hostap_interface *iface;
 	local_info_t *local;
@@ -2449,8 +2450,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, sta->addr, ETH_ALEN);
 		iwe.len = IW_EV_ADDR_LEN;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_ADDR_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_ADDR_LEN);
 
 		/* Use the mode to indicate if it's a station or
 		 * an Access Point */
@@ -2461,8 +2462,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 		else
 			iwe.u.mode = IW_MODE_INFRA;
 		iwe.len = IW_EV_UINT_LEN;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 
 		/* Some quality */
 		memset(&iwe, 0, sizeof(iwe));
@@ -2477,8 +2478,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 		iwe.u.qual.noise = HFA384X_LEVEL_TO_dBm(sta->last_rx_silence);
 		iwe.u.qual.updated = sta->last_rx_updated;
 		iwe.len = IW_EV_QUAL_LEN;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_QUAL_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_QUAL_LEN);
 
 #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT
 		if (sta->ap) {
@@ -2486,8 +2487,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 			iwe.cmd = SIOCGIWESSID;
 			iwe.u.data.length = sta->u.ap.ssid_len;
 			iwe.u.data.flags = 1;
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe,
 							  sta->u.ap.ssid);
 
 			memset(&iwe, 0, sizeof(iwe));
@@ -2497,10 +2498,9 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 					IW_ENCODE_ENABLED | IW_ENCODE_NOKEY;
 			else
 				iwe.u.data.flags = IW_ENCODE_DISABLED;
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe,
-							  sta->u.ap.ssid
-							  /* 0 byte memcpy */);
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe,
+							  sta->u.ap.ssid);
 
 			if (sta->u.ap.channel > 0 &&
 			    sta->u.ap.channel <= FREQ_COUNT) {
@@ -2510,7 +2510,7 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 					* 100000;
 				iwe.u.freq.e = 1;
 				current_ev = iwe_stream_add_event(
-					current_ev, end_buf, &iwe,
+					info, current_ev, end_buf, &iwe,
 					IW_EV_FREQ_LEN);
 			}
 
@@ -2519,8 +2519,8 @@ int prism2_ap_translate_scan(struct net_device *dev, char *buffer)
 			sprintf(buf, "beacon_interval=%d",
 				sta->listen_interval);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe, buf);
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe, buf);
 		}
 #endif /* PRISM2_NO_KERNEL_IEEE80211_MGMT */
 
diff --git a/drivers/net/wireless/hostap/hostap_ioctl.c b/drivers/net/wireless/hostap/hostap_ioctl.c
index 0ca0bfeb0ada..ed52d98317cd 100644
--- a/drivers/net/wireless/hostap/hostap_ioctl.c
+++ b/drivers/net/wireless/hostap/hostap_ioctl.c
@@ -1793,6 +1793,7 @@ static int prism2_ioctl_siwscan(struct net_device *dev,
 
 #ifndef PRISM2_NO_STATION_MODES
 static char * __prism2_translate_scan(local_info_t *local,
+				      struct iw_request_info *info,
 				      struct hfa384x_hostscan_result *scan,
 				      struct hostap_bss_info *bss,
 				      char *current_ev, char *end_buf)
@@ -1823,7 +1824,7 @@ static char * __prism2_translate_scan(local_info_t *local,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_ADDR_LEN);
 
 	/* Other entries will be displayed in the order we give them */
@@ -1832,7 +1833,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.length = ssid_len;
 	iwe.u.data.flags = 1;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, ssid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, ssid);
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWMODE;
@@ -1847,8 +1849,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -1864,8 +1866,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 	if (chan > 0) {
 		iwe.u.freq.m = freq_list[chan - 1] * 100000;
 		iwe.u.freq.e = 1;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_FREQ_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_FREQ_LEN);
 	}
 
 	if (scan) {
@@ -1884,8 +1886,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 			| IW_QUAL_NOISE_UPDATED
 			| IW_QUAL_QUAL_INVALID
 			| IW_QUAL_DBM;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_QUAL_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_QUAL_LEN);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -1895,13 +1897,13 @@ static char * __prism2_translate_scan(local_info_t *local,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, "");
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf, &iwe, "");
 
 	/* TODO: add SuppRates into BSS table */
 	if (scan) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = SIOCGIWRATE;
-		current_val = current_ev + IW_EV_LCP_LEN;
+		current_val = current_ev + iwe_stream_lcp_len(info);
 		pos = scan->sup_rates;
 		for (i = 0; i < sizeof(scan->sup_rates); i++) {
 			if (pos[i] == 0)
@@ -1909,11 +1911,11 @@ static char * __prism2_translate_scan(local_info_t *local,
 			/* Bit rate given in 500 kb/s units (+ 0x80) */
 			iwe.u.bitrate.value = ((pos[i] & 0x7f) * 500000);
 			current_val = iwe_stream_add_value(
-				current_ev, current_val, end_buf, &iwe,
+				info, current_ev, current_val, end_buf, &iwe,
 				IW_EV_PARAM_LEN);
 		}
 		/* Check if we added any event */
-		if ((current_val - current_ev) > IW_EV_LCP_LEN)
+		if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 			current_ev = current_val;
 	}
 
@@ -1924,15 +1926,15 @@ static char * __prism2_translate_scan(local_info_t *local,
 		iwe.cmd = IWEVCUSTOM;
 		sprintf(buf, "bcn_int=%d", le16_to_cpu(scan->beacon_interval));
 		iwe.u.data.length = strlen(buf);
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  buf);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, buf);
 
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVCUSTOM;
 		sprintf(buf, "resp_rate=%d", le16_to_cpu(scan->rate));
 		iwe.u.data.length = strlen(buf);
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  buf);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, buf);
 
 		if (local->last_scan_type == PRISM2_HOSTSCAN &&
 		    (capabilities & WLAN_CAPABILITY_IBSS)) {
@@ -1940,8 +1942,8 @@ static char * __prism2_translate_scan(local_info_t *local,
 			iwe.cmd = IWEVCUSTOM;
 			sprintf(buf, "atim=%d", le16_to_cpu(scan->atim));
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
-							  &iwe, buf);
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf, &iwe, buf);
 		}
 	}
 	kfree(buf);
@@ -1950,16 +1952,16 @@ static char * __prism2_translate_scan(local_info_t *local,
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->wpa_ie_len;
-		current_ev = iwe_stream_add_point(
-			current_ev, end_buf, &iwe, bss->wpa_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->wpa_ie);
 	}
 
 	if (bss && bss->rsn_ie_len > 0 && bss->rsn_ie_len <= MAX_WPA_IE_LEN) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->rsn_ie_len;
-		current_ev = iwe_stream_add_point(
-			current_ev, end_buf, &iwe, bss->rsn_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->rsn_ie);
 	}
 
 	return current_ev;
@@ -1969,6 +1971,7 @@ static char * __prism2_translate_scan(local_info_t *local,
 /* Translate scan data returned from the card to a card independant
  * format that the Wireless Tools will understand - Jean II */
 static inline int prism2_translate_scan(local_info_t *local,
+					struct iw_request_info *info,
 					char *buffer, int buflen)
 {
 	struct hfa384x_hostscan_result *scan;
@@ -1999,13 +2002,14 @@ static inline int prism2_translate_scan(local_info_t *local,
 			if (memcmp(bss->bssid, scan->bssid, ETH_ALEN) == 0) {
 				bss->included = 1;
 				current_ev = __prism2_translate_scan(
-					local, scan, bss, current_ev, end_buf);
+					local, info, scan, bss, current_ev,
+					end_buf);
 				found++;
 			}
 		}
 		if (!found) {
 			current_ev = __prism2_translate_scan(
-				local, scan, NULL, current_ev, end_buf);
+				local, info, scan, NULL, current_ev, end_buf);
 		}
 		/* Check if there is space for one more entry */
 		if ((end_buf - current_ev) <= IW_EV_ADDR_LEN) {
@@ -2023,7 +2027,7 @@ static inline int prism2_translate_scan(local_info_t *local,
 		bss = list_entry(ptr, struct hostap_bss_info, list);
 		if (bss->included)
 			continue;
-		current_ev = __prism2_translate_scan(local, NULL, bss,
+		current_ev = __prism2_translate_scan(local, info, NULL, bss,
 						     current_ev, end_buf);
 		/* Check if there is space for one more entry */
 		if ((end_buf - current_ev) <= IW_EV_ADDR_LEN) {
@@ -2070,7 +2074,7 @@ static inline int prism2_ioctl_giwscan_sta(struct net_device *dev,
 	}
 	local->scan_timestamp = 0;
 
-	res = prism2_translate_scan(local, extra, data->length);
+	res = prism2_translate_scan(local, info, extra, data->length);
 
 	if (res >= 0) {
 		data->length = res;
@@ -2103,7 +2107,7 @@ static int prism2_ioctl_giwscan(struct net_device *dev,
 		 * Jean II */
 
 		/* Translate to WE format */
-		res = prism2_ap_translate_scan(dev, extra);
+		res = prism2_ap_translate_scan(dev, info, extra);
 		if (res >= 0) {
 			printk(KERN_DEBUG "Scan result translation succeeded "
 			       "(length=%d)\n", res);
diff --git a/drivers/net/wireless/libertas/scan.c b/drivers/net/wireless/libertas/scan.c
index d448c9702a0f..343ed38f772d 100644
--- a/drivers/net/wireless/libertas/scan.c
+++ b/drivers/net/wireless/libertas/scan.c
@@ -776,8 +776,9 @@ out:
 #define MAX_CUSTOM_LEN 64
 
 static inline char *lbs_translate_scan(struct lbs_private *priv,
-				       char *start, char *stop,
-				       struct bss_descriptor *bss)
+					    struct iw_request_info *info,
+					    char *start, char *stop,
+					    struct bss_descriptor *bss)
 {
 	struct chan_freq_power *cfp;
 	char *current_val;	/* For rates */
@@ -801,24 +802,24 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, &bss->bssid, ETH_ALEN);
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_ADDR_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_ADDR_LEN);
 
 	/* SSID */
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
 	iwe.u.data.length = min((uint32_t) bss->ssid_len, (uint32_t) IW_ESSID_MAX_SIZE);
-	start = iwe_stream_add_point(start, stop, &iwe, bss->ssid);
+	start = iwe_stream_add_point(info, start, stop, &iwe, bss->ssid);
 
 	/* Mode */
 	iwe.cmd = SIOCGIWMODE;
 	iwe.u.mode = bss->mode;
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_UINT_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_UINT_LEN);
 
 	/* Frequency */
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = (long)cfp->freq * 100000;
 	iwe.u.freq.e = 1;
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_FREQ_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_FREQ_LEN);
 
 	/* Add quality statistics */
 	iwe.cmd = IWEVQUAL;
@@ -852,7 +853,7 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		nf = priv->NF[TYPE_RXPD][TYPE_AVG] / AVG_SCALE;
 		iwe.u.qual.level = CAL_RSSI(snr, nf);
 	}
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_QUAL_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_QUAL_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -862,9 +863,9 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	}
 	iwe.u.data.length = 0;
-	start = iwe_stream_add_point(start, stop, &iwe, bss->ssid);
+	start = iwe_stream_add_point(info, start, stop, &iwe, bss->ssid);
 
-	current_val = start + IW_EV_LCP_LEN;
+	current_val = start + iwe_stream_lcp_len(info);
 
 	iwe.cmd = SIOCGIWRATE;
 	iwe.u.bitrate.fixed = 0;
@@ -874,19 +875,19 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 	for (j = 0; bss->rates[j] && (j < sizeof(bss->rates)); j++) {
 		/* Bit rate given in 500 kb/s units */
 		iwe.u.bitrate.value = bss->rates[j] * 500000;
-		current_val = iwe_stream_add_value(start, current_val,
-					 stop, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, start, current_val,
+						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	if ((bss->mode == IW_MODE_ADHOC) && priv->adhoccreate
 	    && !lbs_ssid_cmp(priv->curbssparams.ssid,
 			     priv->curbssparams.ssid_len,
 			     bss->ssid, bss->ssid_len)) {
 		iwe.u.bitrate.value = 22 * 500000;
-		current_val = iwe_stream_add_value(start, current_val,
+		current_val = iwe_stream_add_value(info, start, current_val,
 						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any event */
-	if((current_val - start) > IW_EV_LCP_LEN)
+	if ((current_val - start) > iwe_stream_lcp_len(info))
 		start = current_val;
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -895,7 +896,7 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		memcpy(buf, bss->wpa_ie, bss->wpa_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->wpa_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -904,7 +905,7 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		memcpy(buf, bss->rsn_ie, bss->rsn_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->rsn_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	if (bss->mesh) {
@@ -915,7 +916,8 @@ static inline char *lbs_translate_scan(struct lbs_private *priv,
 		p += snprintf(p, MAX_CUSTOM_LEN, "mesh-type: olpc");
 		iwe.u.data.length = p - custom;
 		if (iwe.u.data.length)
-			start = iwe_stream_add_point(start, stop, &iwe, custom);
+			start = iwe_stream_add_point(info, start, stop,
+						     &iwe, custom);
 	}
 
 out:
@@ -1036,7 +1038,7 @@ int lbs_get_scan(struct net_device *dev, struct iw_request_info *info,
 		}
 
 		/* Translate to WE format this entry */
-		next_ev = lbs_translate_scan(priv, ev, stop, iter_bss);
+		next_ev = lbs_translate_scan(priv, info, ev, stop, iter_bss);
 		if (next_ev == NULL)
 			continue;
 		ev = next_ev;
diff --git a/drivers/net/wireless/orinoco.c b/drivers/net/wireless/orinoco.c
index 6d13a0d15a0c..b047306bf386 100644
--- a/drivers/net/wireless/orinoco.c
+++ b/drivers/net/wireless/orinoco.c
@@ -4046,6 +4046,7 @@ static int orinoco_ioctl_setscan(struct net_device *dev,
  * format that the Wireless Tools will understand - Jean II
  * Return message length or -errno for fatal errors */
 static inline char *orinoco_translate_scan(struct net_device *dev,
+					   struct iw_request_info *info,
 					   char *current_ev,
 					   char *end_buf,
 					   union hermes_scan_info *bss,
@@ -4062,7 +4063,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bss->a.bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_ADDR_LEN);
 
 	/* Other entries will be displayed in the order we give them */
 
@@ -4072,7 +4074,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		iwe.u.data.length = 32;
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.data.flags = 1;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->a.essid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->a.essid);
 
 	/* Add mode */
 	iwe.cmd = SIOCGIWMODE;
@@ -4082,7 +4085,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	channel = bss->s.channel;
@@ -4091,7 +4095,7 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = channel_frequency[channel-1] * 100000;
 		iwe.u.freq.e = 1;
-		current_ev = iwe_stream_add_event(current_ev, end_buf,
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
 						  &iwe, IW_EV_FREQ_LEN);
 	}
 
@@ -4106,7 +4110,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		iwe.u.qual.qual = iwe.u.qual.level - iwe.u.qual.noise;
 	else
 		iwe.u.qual.qual = 0;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_QUAL_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -4115,7 +4120,8 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, bss->a.essid);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, bss->a.essid);
 
 	/* Add EXTRA: Age to display seconds since last beacon/probe response
 	 * for given network. */
@@ -4126,11 +4132,12 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 		      jiffies_to_msecs(jiffies - last_scanned));
 	iwe.u.data.length = p - custom;
 	if (iwe.u.data.length)
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, custom);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, custom);
 
 	/* Bit rate is not available in Lucent/Agere firmwares */
 	if (priv->firmware_type != FIRMWARE_TYPE_AGERE) {
-		char *current_val = current_ev + IW_EV_LCP_LEN;
+		char *current_val = current_ev + iwe_stream_lcp_len(info);
 		int i;
 		int step;
 
@@ -4149,12 +4156,13 @@ static inline char *orinoco_translate_scan(struct net_device *dev,
 				break;
 			/* Bit rate given in 500 kb/s units (+ 0x80) */
 			iwe.u.bitrate.value = ((bss->p.rates[i] & 0x7f) * 500000);
-			current_val = iwe_stream_add_value(current_ev, current_val,
+			current_val = iwe_stream_add_value(info, current_ev,
+							   current_val,
 							   end_buf, &iwe,
 							   IW_EV_PARAM_LEN);
 		}
 		/* Check if we added any event */
-		if ((current_val - current_ev) > IW_EV_LCP_LEN)
+		if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 			current_ev = current_val;
 	}
 
@@ -4190,7 +4198,7 @@ static int orinoco_ioctl_getscan(struct net_device *dev,
 
 	list_for_each_entry(bss, &priv->bss_list, list) {
 		/* Translate to WE format this entry */
-		current_ev = orinoco_translate_scan(dev, current_ev,
+		current_ev = orinoco_translate_scan(dev, info, current_ev,
 						    extra + srq->length,
 						    &bss->bss,
 						    bss->last_scanned);
diff --git a/drivers/net/wireless/prism54/isl_ioctl.c b/drivers/net/wireless/prism54/isl_ioctl.c
index 5b375b289036..97fa14e0a479 100644
--- a/drivers/net/wireless/prism54/isl_ioctl.c
+++ b/drivers/net/wireless/prism54/isl_ioctl.c
@@ -571,8 +571,9 @@ prism54_set_scan(struct net_device *dev, struct iw_request_info *info,
  */
 
 static char *
-prism54_translate_bss(struct net_device *ndev, char *current_ev,
-		      char *end_buf, struct obj_bss *bss, char noise)
+prism54_translate_bss(struct net_device *ndev, struct iw_request_info *info,
+		      char *current_ev, char *end_buf, struct obj_bss *bss,
+		      char noise)
 {
 	struct iw_event iwe;	/* Temporary buffer */
 	short cap;
@@ -584,8 +585,8 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 	memcpy(iwe.u.ap_addr.sa_data, bss->address, 6);
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	iwe.cmd = SIOCGIWAP;
-	current_ev =
-	    iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_ADDR_LEN);
 
 	/* The following entries will be displayed in the same order we give them */
 
@@ -593,7 +594,7 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 	iwe.u.data.length = bss->ssid.length;
 	iwe.u.data.flags = 1;
 	iwe.cmd = SIOCGIWESSID;
-	current_ev = iwe_stream_add_point(current_ev, end_buf,
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
 					  &iwe, bss->ssid.octets);
 
 	/* Capabilities */
@@ -610,9 +611,8 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 		iwe.u.mode = IW_MODE_ADHOC;
 	iwe.cmd = SIOCGIWMODE;
 	if (iwe.u.mode)
-		current_ev =
-		    iwe_stream_add_event(current_ev, end_buf, &iwe,
-					 IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 
 	/* Encryption capability */
 	if (cap & CAP_CRYPT)
@@ -621,14 +621,15 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
 	iwe.cmd = SIOCGIWENCODE;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, NULL);
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, NULL);
 
 	/* Add frequency. (short) bss->channel is the frequency in MHz */
 	iwe.u.freq.m = bss->channel;
 	iwe.u.freq.e = 6;
 	iwe.cmd = SIOCGIWFREQ;
-	current_ev =
-	    iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_FREQ_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_FREQ_LEN);
 
 	/* Add quality statistics */
 	iwe.u.qual.level = bss->rssi;
@@ -636,20 +637,20 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 	/* do a simple SNR for quality */
 	iwe.u.qual.qual = bss->rssi - noise;
 	iwe.cmd = IWEVQUAL;
-	current_ev =
-	    iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+					  &iwe, IW_EV_QUAL_LEN);
 
 	/* Add WPA/RSN Information Element, if any */
 	wpa_ie_len = prism54_wpa_bss_ie_get(priv, bss->address, wpa_ie);
 	if (wpa_ie_len > 0) {
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = min(wpa_ie_len, (size_t)MAX_WPA_IE_LEN);
-		current_ev = iwe_stream_add_point(current_ev, end_buf,
-				&iwe, wpa_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, wpa_ie);
 	}
 	/* Do the bitrates */
 	{
-		char *	current_val = current_ev + IW_EV_LCP_LEN;
+		char *current_val = current_ev + iwe_stream_lcp_len(info);
 		int i;
 		int mask;
 
@@ -662,14 +663,14 @@ prism54_translate_bss(struct net_device *ndev, char *current_ev,
 		for(i = 0; i < sizeof(scan_rate_list); i++) {
 			if(bss->rates & mask) {
 				iwe.u.bitrate.value = (scan_rate_list[i] * 500000);
-				current_val = iwe_stream_add_value(current_ev, current_val,
-								   end_buf, &iwe,
-								   IW_EV_PARAM_LEN);
+				current_val = iwe_stream_add_value(
+					info, current_ev, current_val,
+					end_buf, &iwe, IW_EV_PARAM_LEN);
 			}
 			mask <<= 1;
 		}
 		/* Check if we added any event */
-		if ((current_val - current_ev) > IW_EV_LCP_LEN)
+		if ((current_val - current_ev) > iwe_stream_lcp_len(info))
 			current_ev = current_val;
 	}
 
@@ -710,7 +711,7 @@ prism54_get_scan(struct net_device *ndev, struct iw_request_info *info,
 
 	/* ok now, scan the list and translate its info */
 	for (i = 0; i < (int) bsslist->nr; i++) {
-		current_ev = prism54_translate_bss(ndev, current_ev,
+		current_ev = prism54_translate_bss(ndev, info, current_ev,
 						   extra + dwrq->length,
 						   &(bsslist->bsslist[i]),
 						   noise);
@@ -2704,6 +2705,7 @@ prism2_ioctl_scan_req(struct net_device *ndev,
                      struct prism2_hostapd_param *param)
 {
 	islpci_private *priv = netdev_priv(ndev);
+	struct iw_request_info info;
 	int i, rvalue;
 	struct obj_bsslist *bsslist;
 	u32 noise = 0;
@@ -2727,9 +2729,12 @@ prism2_ioctl_scan_req(struct net_device *ndev,
 	rvalue |= mgt_get_request(priv, DOT11_OID_BSSLIST, 0, NULL, &r);
 	bsslist = r.ptr;
 
+	info.cmd = PRISM54_HOSTAPD;
+	info.flags = 0;
+
 	/* ok now, scan the list and translate its info */
 	for (i = 0; i < min(IW_MAX_AP, (int) bsslist->nr); i++)
-		current_ev = prism54_translate_bss(ndev, current_ev,
+		current_ev = prism54_translate_bss(ndev, &info, current_ev,
 						   extra + IW_SCAN_MAX_DATA,
 						   &(bsslist->bsslist[i]),
 						   noise);
diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
index a36d2c85e26e..65c50025c88f 100644
--- a/drivers/net/wireless/rndis_wlan.c
+++ b/drivers/net/wireless/rndis_wlan.c
@@ -1648,7 +1648,9 @@ static int rndis_iw_set_scan(struct net_device *dev,
 
 
 static char *rndis_translate_scan(struct net_device *dev,
-    char *cev, char *end_buf, struct ndis_80211_bssid_ex *bssid)
+				  struct iw_request_info *info, char *cev,
+				  char *end_buf,
+				  struct ndis_80211_bssid_ex *bssid)
 {
 #ifdef DEBUG
 	struct usbnet *usbdev = dev->priv;
@@ -1667,14 +1669,14 @@ static char *rndis_translate_scan(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bssid->mac, ETH_ALEN);
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_ADDR_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_ADDR_LEN);
 
 	devdbg(usbdev, "SSID(%d) %s", le32_to_cpu(bssid->ssid.length),
 						bssid->ssid.essid);
 	iwe.cmd = SIOCGIWESSID;
 	iwe.u.essid.length = le32_to_cpu(bssid->ssid.length);
 	iwe.u.essid.flags = 1;
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, bssid->ssid.essid);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, bssid->ssid.essid);
 
 	devdbg(usbdev, "MODE %d", le32_to_cpu(bssid->net_infra));
 	iwe.cmd = SIOCGIWMODE;
@@ -1690,12 +1692,12 @@ static char *rndis_translate_scan(struct net_device *dev,
 		iwe.u.mode = IW_MODE_AUTO;
 		break;
 	}
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_UINT_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_UINT_LEN);
 
 	devdbg(usbdev, "FREQ %d kHz", le32_to_cpu(bssid->config.ds_config));
 	iwe.cmd = SIOCGIWFREQ;
 	dsconfig_to_freq(le32_to_cpu(bssid->config.ds_config), &iwe.u.freq);
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_FREQ_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_FREQ_LEN);
 
 	devdbg(usbdev, "QUAL %d", le32_to_cpu(bssid->rssi));
 	iwe.cmd = IWEVQUAL;
@@ -1704,7 +1706,7 @@ static char *rndis_translate_scan(struct net_device *dev,
 	iwe.u.qual.updated = IW_QUAL_QUAL_UPDATED
 			| IW_QUAL_LEVEL_UPDATED
 			| IW_QUAL_NOISE_INVALID;
-	cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_QUAL_LEN);
+	cev = iwe_stream_add_event(info, cev, end_buf, &iwe, IW_EV_QUAL_LEN);
 
 	devdbg(usbdev, "ENCODE %d", le32_to_cpu(bssid->privacy));
 	iwe.cmd = SIOCGIWENCODE;
@@ -1714,10 +1716,10 @@ static char *rndis_translate_scan(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_ENABLED | IW_ENCODE_NOKEY;
 
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, NULL);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, NULL);
 
 	devdbg(usbdev, "RATES:");
-	current_val = cev + IW_EV_LCP_LEN;
+	current_val = cev + iwe_stream_lcp_len(info);
 	iwe.cmd = SIOCGIWRATE;
 	for (i = 0; i < sizeof(bssid->rates); i++) {
 		if (bssid->rates[i] & 0x7f) {
@@ -1725,13 +1727,13 @@ static char *rndis_translate_scan(struct net_device *dev,
 				((bssid->rates[i] & 0x7f) *
 				500000);
 			devdbg(usbdev, " %d", iwe.u.bitrate.value);
-			current_val = iwe_stream_add_value(cev,
+			current_val = iwe_stream_add_value(info, cev,
 				current_val, end_buf, &iwe,
 				IW_EV_PARAM_LEN);
 		}
 	}
 
-	if ((current_val - cev) > IW_EV_LCP_LEN)
+	if ((current_val - cev) > iwe_stream_lcp_len(info))
 		cev = current_val;
 
 	beacon = le32_to_cpu(bssid->config.beacon_period);
@@ -1739,14 +1741,14 @@ static char *rndis_translate_scan(struct net_device *dev,
 	iwe.cmd = IWEVCUSTOM;
 	snprintf(sbuf, sizeof(sbuf), "bcn_int=%d", beacon);
 	iwe.u.data.length = strlen(sbuf);
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, sbuf);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, sbuf);
 
 	atim = le32_to_cpu(bssid->config.atim_window);
 	devdbg(usbdev, "ATIM %d", atim);
 	iwe.cmd = IWEVCUSTOM;
 	snprintf(sbuf, sizeof(sbuf), "atim=%u", atim);
 	iwe.u.data.length = strlen(sbuf);
-	cev = iwe_stream_add_point(cev, end_buf, &iwe, sbuf);
+	cev = iwe_stream_add_point(info, cev, end_buf, &iwe, sbuf);
 
 	ie = (void *)(bssid->ies + sizeof(struct ndis_80211_fixed_ies));
 	ie_len = min(bssid_len - (int)sizeof(*bssid),
@@ -1760,7 +1762,7 @@ static char *rndis_translate_scan(struct net_device *dev,
 					(ie->id == MFIE_TYPE_RSN) ? 2 : 1);
 			iwe.cmd = IWEVGENIE;
 			iwe.u.data.length = min(ie->len + 2, MAX_WPA_IE_LEN);
-			cev = iwe_stream_add_point(cev, end_buf, &iwe,
+			cev = iwe_stream_add_point(info, cev, end_buf, &iwe,
 								(u8 *)ie);
 		}
 
@@ -1803,8 +1805,8 @@ static int rndis_iw_get_scan(struct net_device *dev,
 	devdbg(usbdev, "SIOCGIWSCAN: %d BSSIDs found", count);
 
 	while (count && ((void *)bssid + bssid_len) <= (buf + len)) {
-		cev = rndis_translate_scan(dev, cev, extra + IW_SCAN_MAX_DATA,
-									bssid);
+		cev = rndis_translate_scan(dev, info, cev,
+					   extra + IW_SCAN_MAX_DATA, bssid);
 		bssid = (void *)bssid + bssid_len;
 		bssid_len = le32_to_cpu(bssid->length);
 		count--;
diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c
index 42a36b3f3ff7..377141995e36 100644
--- a/drivers/net/wireless/wl3501_cs.c
+++ b/drivers/net/wireless/wl3501_cs.c
@@ -1624,25 +1624,25 @@ static int wl3501_get_scan(struct net_device *dev, struct iw_request_info *info,
 		iwe.cmd			= SIOCGIWAP;
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, this->bss_set[i].bssid, ETH_ALEN);
-		current_ev = iwe_stream_add_event(current_ev,
+		current_ev = iwe_stream_add_event(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, IW_EV_ADDR_LEN);
 		iwe.cmd		  = SIOCGIWESSID;
 		iwe.u.data.flags  = 1;
 		iwe.u.data.length = this->bss_set[i].ssid.el.len;
-		current_ev = iwe_stream_add_point(current_ev,
+		current_ev = iwe_stream_add_point(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe,
 						  this->bss_set[i].ssid.essid);
 		iwe.cmd	   = SIOCGIWMODE;
 		iwe.u.mode = this->bss_set[i].bss_type;
-		current_ev = iwe_stream_add_event(current_ev,
+		current_ev = iwe_stream_add_event(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, IW_EV_UINT_LEN);
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = this->bss_set[i].ds_pset.chan;
 		iwe.u.freq.e = 0;
-		current_ev = iwe_stream_add_event(current_ev,
+		current_ev = iwe_stream_add_event(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, IW_EV_FREQ_LEN);
 		iwe.cmd = SIOCGIWENCODE;
@@ -1651,7 +1651,7 @@ static int wl3501_get_scan(struct net_device *dev, struct iw_request_info *info,
 		else
 			iwe.u.data.flags = IW_ENCODE_DISABLED;
 		iwe.u.data.length = 0;
-		current_ev = iwe_stream_add_point(current_ev,
+		current_ev = iwe_stream_add_point(info, current_ev,
 						  extra + IW_SCAN_MAX_DATA,
 						  &iwe, NULL);
 	}
diff --git a/drivers/net/wireless/zd1201.c b/drivers/net/wireless/zd1201.c
index d5c0c66188ca..07e4d1f73207 100644
--- a/drivers/net/wireless/zd1201.c
+++ b/drivers/net/wireless/zd1201.c
@@ -1152,32 +1152,36 @@ static int zd1201_get_scan(struct net_device *dev,
 		iwe.cmd = SIOCGIWAP;
 		iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 		memcpy(iwe.u.ap_addr.sa_data, zd->rxdata+i+6, 6);
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_ADDR_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_ADDR_LEN);
 
 		iwe.cmd = SIOCGIWESSID;
 		iwe.u.data.length = zd->rxdata[i+16];
 		iwe.u.data.flags = 1;
-		cev = iwe_stream_add_point(cev, end_buf, &iwe, zd->rxdata+i+18);
+		cev = iwe_stream_add_point(info, cev, end_buf,
+					   &iwe, zd->rxdata+i+18);
 
 		iwe.cmd = SIOCGIWMODE;
 		if (zd->rxdata[i+14]&0x01)
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_UINT_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_UINT_LEN);
 		
 		iwe.cmd = SIOCGIWFREQ;
 		iwe.u.freq.m = zd->rxdata[i+0];
 		iwe.u.freq.e = 0;
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_FREQ_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_FREQ_LEN);
 		
 		iwe.cmd = SIOCGIWRATE;
 		iwe.u.bitrate.fixed = 0;
 		iwe.u.bitrate.disabled = 0;
 		for (j=0; j<10; j++) if (zd->rxdata[i+50+j]) {
 			iwe.u.bitrate.value = (zd->rxdata[i+50+j]&0x7f)*500000;
-			cev=iwe_stream_add_event(cev, end_buf, &iwe,
-			    IW_EV_PARAM_LEN);
+			cev = iwe_stream_add_event(info, cev, end_buf,
+						   &iwe, IW_EV_PARAM_LEN);
 		}
 		
 		iwe.cmd = SIOCGIWENCODE;
@@ -1186,14 +1190,15 @@ static int zd1201_get_scan(struct net_device *dev,
 			iwe.u.data.flags = IW_ENCODE_ENABLED;
 		else
 			iwe.u.data.flags = IW_ENCODE_DISABLED;
-		cev = iwe_stream_add_point(cev, end_buf, &iwe, NULL);
+		cev = iwe_stream_add_point(info, cev, end_buf, &iwe, NULL);
 		
 		iwe.cmd = IWEVQUAL;
 		iwe.u.qual.qual = zd->rxdata[i+4];
 		iwe.u.qual.noise= zd->rxdata[i+2]/10-100;
 		iwe.u.qual.level = (256+zd->rxdata[i+4]*100)/255-100;
 		iwe.u.qual.updated = 7;
-		cev = iwe_stream_add_event(cev, end_buf, &iwe, IW_EV_QUAL_LEN);
+		cev = iwe_stream_add_event(info, cev, end_buf,
+					   &iwe, IW_EV_QUAL_LEN);
 	}
 
 	if (!enabled_save)
diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 79d846875825..d7958f9b52cb 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -1113,6 +1113,21 @@ struct iw_event
 #define IW_EV_POINT_LEN	(IW_EV_LCP_LEN + sizeof(struct iw_point) - \
 			 IW_EV_POINT_OFF)
 
+#ifdef __KERNEL__
+#ifdef CONFIG_COMPAT
+struct __compat_iw_event {
+	__u16		len;			/* Real length of this stuff */
+	__u16		cmd;			/* Wireless IOCTL */
+	compat_caddr_t	pointer;
+};
+#define IW_EV_COMPAT_LCP_LEN offsetof(struct __compat_iw_event, pointer)
+#define IW_EV_COMPAT_POINT_OFF offsetof(struct compat_iw_point, length)
+#define IW_EV_COMPAT_POINT_LEN	\
+	(IW_EV_COMPAT_LCP_LEN + sizeof(struct compat_iw_point) - \
+	 IW_EV_COMPAT_POINT_OFF)
+#endif
+#endif
+
 /* Size of the Event prefix when packed in stream */
 #define IW_EV_LCP_PK_LEN	(4)
 /* Size of the various events when packed in stream */
diff --git a/include/net/iw_handler.h b/include/net/iw_handler.h
index c99a8eec84e7..51b9a37de991 100644
--- a/include/net/iw_handler.h
+++ b/include/net/iw_handler.h
@@ -478,105 +478,58 @@ extern void wireless_spy_update(struct net_device *	dev,
  * Function that are so simple that it's more efficient inlining them
  */
 
-/*------------------------------------------------------------------*/
-/*
- * Wrapper to add an Wireless Event to a stream of events.
- */
-static inline char *
-iwe_stream_add_event(char *	stream,		/* Stream of events */
-		     char *	ends,		/* End of stream */
-		     struct iw_event *iwe,	/* Payload */
-		     int	event_len)	/* Real size of payload */
+static inline int iwe_stream_lcp_len(struct iw_request_info *info)
 {
-	/* Check if it's possible */
-	if(likely((stream + event_len) < ends)) {
-		iwe->len = event_len;
-		/* Beware of alignement issues on 64 bits */
-		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN,
-		       event_len - IW_EV_LCP_LEN);
-		stream += event_len;
-	}
-	return stream;
+#ifdef CONFIG_COMPAT
+	if (info->flags & IW_REQUEST_FLAG_COMPAT)
+		return IW_EV_COMPAT_LCP_LEN;
+#endif
+	return IW_EV_LCP_LEN;
 }
 
-/*------------------------------------------------------------------*/
-/*
- * Wrapper to add an short Wireless Event containing a pointer to a
- * stream of events.
- */
-static inline char *
-iwe_stream_add_point(char *	stream,		/* Stream of events */
-		     char *	ends,		/* End of stream */
-		     struct iw_event *iwe,	/* Payload length + flags */
-		     char *	extra)		/* More payload */
+static inline int iwe_stream_point_len(struct iw_request_info *info)
 {
-	int	event_len = IW_EV_POINT_LEN + iwe->u.data.length;
-	/* Check if it's possible */
-	if(likely((stream + event_len) < ends)) {
-		iwe->len = event_len;
-		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN + IW_EV_POINT_OFF,
-		       IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_POINT_LEN, extra, iwe->u.data.length);
-		stream += event_len;
-	}
-	return stream;
+#ifdef CONFIG_COMPAT
+	if (info->flags & IW_REQUEST_FLAG_COMPAT)
+		return IW_EV_COMPAT_POINT_LEN;
+#endif
+	return IW_EV_POINT_LEN;
 }
 
-/*------------------------------------------------------------------*/
-/*
- * Wrapper to add a value to a Wireless Event in a stream of events.
- * Be careful, this one is tricky to use properly :
- * At the first run, you need to have (value = event + IW_EV_LCP_LEN).
- */
-static inline char *
-iwe_stream_add_value(char *	event,		/* Event in the stream */
-		     char *	value,		/* Value in event */
-		     char *	ends,		/* End of stream */
-		     struct iw_event *iwe,	/* Payload */
-		     int	event_len)	/* Real size of payload */
+static inline int iwe_stream_event_len_adjust(struct iw_request_info *info,
+					      int event_len)
 {
-	/* Don't duplicate LCP */
-	event_len -= IW_EV_LCP_LEN;
-
-	/* Check if it's possible */
-	if(likely((value + event_len) < ends)) {
-		/* Add new value */
-		memcpy(value, (char *) iwe + IW_EV_LCP_LEN, event_len);
-		value += event_len;
-		/* Patch LCP */
-		iwe->len = value - event;
-		memcpy(event, (char *) iwe, IW_EV_LCP_LEN);
+#ifdef CONFIG_COMPAT
+	if (info->flags & IW_REQUEST_FLAG_COMPAT) {
+		event_len -= IW_EV_LCP_LEN;
+		event_len += IW_EV_COMPAT_LCP_LEN;
 	}
-	return value;
+#endif
+
+	return event_len;
 }
 
 /*------------------------------------------------------------------*/
 /*
  * Wrapper to add an Wireless Event to a stream of events.
- * Same as above, with explicit error check...
  */
 static inline char *
-iwe_stream_check_add_event(char *	stream,		/* Stream of events */
-			   char *	ends,		/* End of stream */
-			   struct iw_event *iwe,	/* Payload */
-			   int		event_len,	/* Size of payload */
-			   int *	perr)		/* Error report */
+iwe_stream_add_event(struct iw_request_info *info, char *stream, char *ends,
+		     struct iw_event *iwe, int event_len)
 {
-	/* Check if it's possible, set error if not */
+	int lcp_len = iwe_stream_lcp_len(info);
+
+	event_len = iwe_stream_event_len_adjust(info, event_len);
+
+	/* Check if it's possible */
 	if(likely((stream + event_len) < ends)) {
 		iwe->len = event_len;
 		/* Beware of alignement issues on 64 bits */
 		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN,
-		       event_len - IW_EV_LCP_LEN);
+		memcpy(stream + lcp_len, &iwe->u,
+		       event_len - lcp_len);
 		stream += event_len;
-	} else
-		*perr = -E2BIG;
+	}
 	return stream;
 }
 
@@ -584,27 +537,25 @@ iwe_stream_check_add_event(char *	stream,		/* Stream of events */
 /*
  * Wrapper to add an short Wireless Event containing a pointer to a
  * stream of events.
- * Same as above, with explicit error check...
  */
 static inline char *
-iwe_stream_check_add_point(char *	stream,		/* Stream of events */
-			   char *	ends,		/* End of stream */
-			   struct iw_event *iwe,	/* Payload length + flags */
-			   char *	extra,		/* More payload */
-			   int *	perr)		/* Error report */
+iwe_stream_add_point(struct iw_request_info *info, char *stream, char *ends,
+		     struct iw_event *iwe, char *extra)
 {
-	int	event_len = IW_EV_POINT_LEN + iwe->u.data.length;
+	int event_len = iwe_stream_point_len(info) + iwe->u.data.length;
+	int point_len = iwe_stream_point_len(info);
+	int lcp_len   = iwe_stream_lcp_len(info);
+
 	/* Check if it's possible */
 	if(likely((stream + event_len) < ends)) {
 		iwe->len = event_len;
 		memcpy(stream, (char *) iwe, IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_LCP_LEN,
-		       ((char *) iwe) + IW_EV_LCP_LEN + IW_EV_POINT_OFF,
+		memcpy(stream + lcp_len,
+		       ((char *) &iwe->u) + IW_EV_POINT_OFF,
 		       IW_EV_POINT_PK_LEN - IW_EV_LCP_PK_LEN);
-		memcpy(stream + IW_EV_POINT_LEN, extra, iwe->u.data.length);
+		memcpy(stream + point_len, extra, iwe->u.data.length);
 		stream += event_len;
-	} else
-		*perr = -E2BIG;
+	}
 	return stream;
 }
 
@@ -613,29 +564,25 @@ iwe_stream_check_add_point(char *	stream,		/* Stream of events */
  * Wrapper to add a value to a Wireless Event in a stream of events.
  * Be careful, this one is tricky to use properly :
  * At the first run, you need to have (value = event + IW_EV_LCP_LEN).
- * Same as above, with explicit error check...
  */
 static inline char *
-iwe_stream_check_add_value(char *	event,		/* Event in the stream */
-			   char *	value,		/* Value in event */
-			   char *	ends,		/* End of stream */
-			   struct iw_event *iwe,	/* Payload */
-			   int		event_len,	/* Size of payload */
-			   int *	perr)		/* Error report */
+iwe_stream_add_value(struct iw_request_info *info, char *event, char *value,
+		     char *ends, struct iw_event *iwe, int event_len)
 {
+	int lcp_len = iwe_stream_lcp_len(info);
+
 	/* Don't duplicate LCP */
 	event_len -= IW_EV_LCP_LEN;
 
 	/* Check if it's possible */
 	if(likely((value + event_len) < ends)) {
 		/* Add new value */
-		memcpy(value, (char *) iwe + IW_EV_LCP_LEN, event_len);
+		memcpy(value, &iwe->u, event_len);
 		value += event_len;
 		/* Patch LCP */
 		iwe->len = value - event;
-		memcpy(event, (char *) iwe, IW_EV_LCP_LEN);
-	} else
-		*perr = -E2BIG;
+		memcpy(event, (char *) iwe, lcp_len);
+	}
 	return value;
 }
 
diff --git a/net/ieee80211/ieee80211_wx.c b/net/ieee80211/ieee80211_wx.c
index 822606b615ca..973832dd7faf 100644
--- a/net/ieee80211/ieee80211_wx.c
+++ b/net/ieee80211/ieee80211_wx.c
@@ -43,8 +43,9 @@ static const char *ieee80211_modes[] = {
 
 #define MAX_CUSTOM_LEN 64
 static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
-					   char *start, char *stop,
-					   struct ieee80211_network *network)
+				      char *start, char *stop,
+				      struct ieee80211_network *network,
+				      struct iw_request_info *info)
 {
 	char custom[MAX_CUSTOM_LEN];
 	char *p;
@@ -57,7 +58,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, network->bssid, ETH_ALEN);
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_ADDR_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_ADDR_LEN);
 
 	/* Remaining entries will be displayed in the order we provide them */
 
@@ -66,17 +67,19 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	iwe.u.data.flags = 1;
 	if (network->flags & NETWORK_EMPTY_ESSID) {
 		iwe.u.data.length = sizeof("<hidden>");
-		start = iwe_stream_add_point(start, stop, &iwe, "<hidden>");
+		start = iwe_stream_add_point(info, start, stop,
+					     &iwe, "<hidden>");
 	} else {
 		iwe.u.data.length = min(network->ssid_len, (u8) 32);
-		start = iwe_stream_add_point(start, stop, &iwe, network->ssid);
+		start = iwe_stream_add_point(info, start, stop,
+					     &iwe, network->ssid);
 	}
 
 	/* Add the protocol name */
 	iwe.cmd = SIOCGIWNAME;
 	snprintf(iwe.u.name, IFNAMSIZ, "IEEE 802.11%s",
 		 ieee80211_modes[network->mode]);
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_CHAR_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_CHAR_LEN);
 
 	/* Add mode */
 	iwe.cmd = SIOCGIWMODE;
@@ -86,7 +89,8 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
 
-		start = iwe_stream_add_event(start, stop, &iwe, IW_EV_UINT_LEN);
+		start = iwe_stream_add_event(info, start, stop,
+					     &iwe, IW_EV_UINT_LEN);
 	}
 
 	/* Add channel and frequency */
@@ -95,7 +99,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	iwe.u.freq.m = ieee80211_channel_to_freq(ieee, network->channel);
 	iwe.u.freq.e = 6;
 	iwe.u.freq.i = 0;
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_FREQ_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_FREQ_LEN);
 
 	/* Add encryption capability */
 	iwe.cmd = SIOCGIWENCODE;
@@ -104,12 +108,13 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	start = iwe_stream_add_point(start, stop, &iwe, network->ssid);
+	start = iwe_stream_add_point(info, start, stop,
+				     &iwe, network->ssid);
 
 	/* Add basic and extended rates */
 	/* Rate : stuffing multiple values in a single event require a bit
 	 * more of magic - Jean II */
-	current_val = start + IW_EV_LCP_LEN;
+	current_val = start + iwe_stream_lcp_len(info);
 	iwe.cmd = SIOCGIWRATE;
 	/* Those two flags are ignored... */
 	iwe.u.bitrate.fixed = iwe.u.bitrate.disabled = 0;
@@ -124,17 +129,19 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		/* Bit rate given in 500 kb/s units (+ 0x80) */
 		iwe.u.bitrate.value = ((rate & 0x7f) * 500000);
 		/* Add new value to event */
-		current_val = iwe_stream_add_value(start, current_val, stop, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, start, current_val,
+						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	for (; j < network->rates_ex_len; j++) {
 		rate = network->rates_ex[j] & 0x7F;
 		/* Bit rate given in 500 kb/s units (+ 0x80) */
 		iwe.u.bitrate.value = ((rate & 0x7f) * 500000);
 		/* Add new value to event */
-		current_val = iwe_stream_add_value(start, current_val, stop, &iwe, IW_EV_PARAM_LEN);
+		current_val = iwe_stream_add_value(info, start, current_val,
+						   stop, &iwe, IW_EV_PARAM_LEN);
 	}
 	/* Check if we added any rate */
-	if((current_val - start) > IW_EV_LCP_LEN)
+	if ((current_val - start) > iwe_stream_lcp_len(info))
 		start = current_val;
 
 	/* Add quality statistics */
@@ -181,14 +188,14 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		iwe.u.qual.level = network->stats.signal;
 	}
 
-	start = iwe_stream_add_event(start, stop, &iwe, IW_EV_QUAL_LEN);
+	start = iwe_stream_add_event(info, start, stop, &iwe, IW_EV_QUAL_LEN);
 
 	iwe.cmd = IWEVCUSTOM;
 	p = custom;
 
 	iwe.u.data.length = p - custom;
 	if (iwe.u.data.length)
-		start = iwe_stream_add_point(start, stop, &iwe, custom);
+		start = iwe_stream_add_point(info, start, stop, &iwe, custom);
 
 	memset(&iwe, 0, sizeof(iwe));
 	if (network->wpa_ie_len) {
@@ -196,7 +203,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		memcpy(buf, network->wpa_ie, network->wpa_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = network->wpa_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -205,7 +212,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		memcpy(buf, network->rsn_ie, network->rsn_ie_len);
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = network->rsn_ie_len;
-		start = iwe_stream_add_point(start, stop, &iwe, buf);
+		start = iwe_stream_add_point(info, start, stop, &iwe, buf);
 	}
 
 	/* Add EXTRA: Age to display seconds since last beacon/probe response
@@ -217,7 +224,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 		      jiffies_to_msecs(jiffies - network->last_scanned));
 	iwe.u.data.length = p - custom;
 	if (iwe.u.data.length)
-		start = iwe_stream_add_point(start, stop, &iwe, custom);
+		start = iwe_stream_add_point(info, start, stop, &iwe, custom);
 
 	/* Add spectrum management information */
 	iwe.cmd = -1;
@@ -238,7 +245,7 @@ static char *ieee80211_translate_scan(struct ieee80211_device *ieee,
 
 	if (iwe.cmd == IWEVCUSTOM) {
 		iwe.u.data.length = p - custom;
-		start = iwe_stream_add_point(start, stop, &iwe, custom);
+		start = iwe_stream_add_point(info, start, stop, &iwe, custom);
 	}
 
 	return start;
@@ -272,7 +279,8 @@ int ieee80211_wx_get_scan(struct ieee80211_device *ieee,
 
 		if (ieee->scan_age == 0 ||
 		    time_after(network->last_scanned + ieee->scan_age, jiffies))
-			ev = ieee80211_translate_scan(ieee, ev, stop, network);
+			ev = ieee80211_translate_scan(ieee, ev, stop, network,
+						      info);
 		else
 			IEEE80211_DEBUG_SCAN("Not showing network '%s ("
 					     "%s)' due to age (%dms).\n",
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 14fccf16b80f..80a9e7c07b47 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -24,6 +24,7 @@
 #include <linux/spinlock.h>
 #include <linux/etherdevice.h>
 #include <net/wireless.h>
+#include <net/iw_handler.h>
 #include "key.h"
 #include "sta_info.h"
 
@@ -867,7 +868,9 @@ int ieee80211_sta_set_bssid(struct net_device *dev, u8 *bssid);
 int ieee80211_sta_req_scan(struct net_device *dev, u8 *ssid, size_t ssid_len);
 void ieee80211_sta_req_auth(struct net_device *dev,
 			    struct ieee80211_if_sta *ifsta);
-int ieee80211_sta_scan_results(struct net_device *dev, char *buf, size_t len);
+int ieee80211_sta_scan_results(struct net_device *dev,
+			       struct iw_request_info *info,
+			       char *buf, size_t len);
 ieee80211_rx_result ieee80211_sta_rx_scan(
 	struct net_device *dev, struct sk_buff *skb,
 	struct ieee80211_rx_status *rx_status);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 55659a730dc1..e06d6450f215 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -4087,6 +4087,7 @@ int ieee80211_sta_req_scan(struct net_device *dev, u8 *ssid, size_t ssid_len)
 
 static char *
 ieee80211_sta_scan_result(struct net_device *dev,
+			  struct iw_request_info *info,
 			  struct ieee80211_sta_bss *bss,
 			  char *current_ev, char *end_buf)
 {
@@ -4101,7 +4102,7 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	iwe.cmd = SIOCGIWAP;
 	iwe.u.ap_addr.sa_family = ARPHRD_ETHER;
 	memcpy(iwe.u.ap_addr.sa_data, bss->bssid, ETH_ALEN);
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_ADDR_LEN);
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -4109,13 +4110,13 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	if (bss_mesh_cfg(bss)) {
 		iwe.u.data.length = bss_mesh_id_len(bss);
 		iwe.u.data.flags = 1;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss_mesh_id(bss));
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss_mesh_id(bss));
 	} else {
 		iwe.u.data.length = bss->ssid_len;
 		iwe.u.data.flags = 1;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->ssid);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->ssid);
 	}
 
 	if (bss->capability & (WLAN_CAPABILITY_ESS | WLAN_CAPABILITY_IBSS)
@@ -4128,22 +4129,22 @@ ieee80211_sta_scan_result(struct net_device *dev,
 			iwe.u.mode = IW_MODE_MASTER;
 		else
 			iwe.u.mode = IW_MODE_ADHOC;
-		current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
-						  IW_EV_UINT_LEN);
+		current_ev = iwe_stream_add_event(info, current_ev, end_buf,
+						  &iwe, IW_EV_UINT_LEN);
 	}
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = ieee80211_frequency_to_channel(bss->freq);
 	iwe.u.freq.e = 0;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_FREQ_LEN);
 
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = SIOCGIWFREQ;
 	iwe.u.freq.m = bss->freq;
 	iwe.u.freq.e = 6;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_FREQ_LEN);
 	memset(&iwe, 0, sizeof(iwe));
 	iwe.cmd = IWEVQUAL;
@@ -4151,7 +4152,7 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	iwe.u.qual.level = bss->signal;
 	iwe.u.qual.noise = bss->noise;
 	iwe.u.qual.updated = local->wstats_flags;
-	current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe,
+	current_ev = iwe_stream_add_event(info, current_ev, end_buf, &iwe,
 					  IW_EV_QUAL_LEN);
 
 	memset(&iwe, 0, sizeof(iwe));
@@ -4161,35 +4162,36 @@ ieee80211_sta_scan_result(struct net_device *dev,
 	else
 		iwe.u.data.flags = IW_ENCODE_DISABLED;
 	iwe.u.data.length = 0;
-	current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, "");
+	current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+					  &iwe, "");
 
 	if (bss && bss->wpa_ie) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->wpa_ie_len;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->wpa_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->wpa_ie);
 	}
 
 	if (bss && bss->rsn_ie) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->rsn_ie_len;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->rsn_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->rsn_ie);
 	}
 
 	if (bss && bss->ht_ie) {
 		memset(&iwe, 0, sizeof(iwe));
 		iwe.cmd = IWEVGENIE;
 		iwe.u.data.length = bss->ht_ie_len;
-		current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe,
-						  bss->ht_ie);
+		current_ev = iwe_stream_add_point(info, current_ev, end_buf,
+						  &iwe, bss->ht_ie);
 	}
 
 	if (bss && bss->supp_rates_len > 0) {
 		/* display all supported rates in readable format */
-		char *p = current_ev + IW_EV_LCP_LEN;
+		char *p = current_ev + iwe_stream_lcp_len(info);
 		int i;
 
 		memset(&iwe, 0, sizeof(iwe));
@@ -4200,7 +4202,7 @@ ieee80211_sta_scan_result(struct net_device *dev,
 		for (i = 0; i < bss->supp_rates_len; i++) {
 			iwe.u.bitrate.value = ((bss->supp_rates[i] &
 							0x7f) * 500000);
-			p = iwe_stream_add_value(current_ev, p,
+			p = iwe_stream_add_value(info, current_ev, p,
 					end_buf, &iwe, IW_EV_PARAM_LEN);
 		}
 		current_ev = p;
@@ -4214,7 +4216,8 @@ ieee80211_sta_scan_result(struct net_device *dev,
 			iwe.cmd = IWEVCUSTOM;
 			sprintf(buf, "tsf=%016llx", (unsigned long long)(bss->timestamp));
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			kfree(buf);
 		}
@@ -4229,31 +4232,36 @@ ieee80211_sta_scan_result(struct net_device *dev,
 			iwe.cmd = IWEVCUSTOM;
 			sprintf(buf, "Mesh network (version %d)", cfg[0]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Path Selection Protocol ID: "
 				"0x%02X%02X%02X%02X", cfg[1], cfg[2], cfg[3],
 							cfg[4]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Path Selection Metric ID: "
 				"0x%02X%02X%02X%02X", cfg[5], cfg[6], cfg[7],
 							cfg[8]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Congestion Control Mode ID: "
 				"0x%02X%02X%02X%02X", cfg[9], cfg[10],
 							cfg[11], cfg[12]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			sprintf(buf, "Channel Precedence: "
 				"0x%02X%02X%02X%02X", cfg[13], cfg[14],
 							cfg[15], cfg[16]);
 			iwe.u.data.length = strlen(buf);
-			current_ev = iwe_stream_add_point(current_ev, end_buf,
+			current_ev = iwe_stream_add_point(info, current_ev,
+							  end_buf,
 							  &iwe, buf);
 			kfree(buf);
 		}
@@ -4263,7 +4271,9 @@ ieee80211_sta_scan_result(struct net_device *dev,
 }
 
 
-int ieee80211_sta_scan_results(struct net_device *dev, char *buf, size_t len)
+int ieee80211_sta_scan_results(struct net_device *dev,
+			       struct iw_request_info *info,
+			       char *buf, size_t len)
 {
 	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 	char *current_ev = buf;
@@ -4276,8 +4286,8 @@ int ieee80211_sta_scan_results(struct net_device *dev, char *buf, size_t len)
 			spin_unlock_bh(&local->sta_bss_lock);
 			return -E2BIG;
 		}
-		current_ev = ieee80211_sta_scan_result(dev, bss, current_ev,
-						       end_buf);
+		current_ev = ieee80211_sta_scan_result(dev, info, bss,
+						       current_ev, end_buf);
 	}
 	spin_unlock_bh(&local->sta_bss_lock);
 	return current_ev - buf;
diff --git a/net/mac80211/wext.c b/net/mac80211/wext.c
index 5af3862e7191..f47d13bdf7f7 100644
--- a/net/mac80211/wext.c
+++ b/net/mac80211/wext.c
@@ -567,7 +567,7 @@ static int ieee80211_ioctl_giwscan(struct net_device *dev,
 	if (local->sta_sw_scanning || local->sta_hw_scanning)
 		return -EAGAIN;
 
-	res = ieee80211_sta_scan_results(dev, extra, data->length);
+	res = ieee80211_sta_scan_results(dev, info, extra, data->length);
 	if (res >= 0) {
 		data->length = res;
 		return 0;
-- 
cgit v1.2.3


From df3bc8bd8f8fd17e9b22859d82af38fa702e75b7 Mon Sep 17 00:00:00 2001
From: Chas Williams <chas@cmf.nrl.navy.mil>
Date: Tue, 17 Jun 2008 16:19:24 -0700
Subject: atm: [suni] add support for setting loopback and framing modes

Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/atm/suni.c    | 121 +++++++++++++++++++++++++++++++++++++++++++-------
 drivers/atm/suni.h    |  31 +++++++++++--
 include/linux/sonet.h |   2 +-
 3 files changed, 133 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/atm/suni.c b/drivers/atm/suni.c
index 6097f6877a44..6dd3f5919968 100644
--- a/drivers/atm/suni.c
+++ b/drivers/atm/suni.c
@@ -1,8 +1,14 @@
-/* drivers/atm/suni.c - PMC PM5346 SUNI (PHY) driver */
+/*
+ * drivers/atm/suni.c - S/UNI PHY driver
+ *
+ * Supports the following:
+ * 	PMC PM5346 S/UNI LITE
+ * 	PMC PM5350 S/UNI 155 ULTRA
+ * 	PMC PM5355 S/UNI 622
+ */
  
 /* Written 1995-2000 by Werner Almesberger, EPFL LRC/ICA */
 
-
 #include <linux/module.h>
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
@@ -146,25 +152,105 @@ static int get_diag(struct atm_dev *dev,void __user *arg)
 static int set_loopback(struct atm_dev *dev,int mode)
 {
 	unsigned char control;
+	int reg, dle, lle;
+
+	if (PRIV(dev)->type == SUNI_MRI_TYPE_PM5355) {
+		reg = SUNI_MCM;
+		dle = SUNI_MCM_DLE;
+		lle = SUNI_MCM_LLE;
+	} else {
+		reg = SUNI_MCT;
+		dle = SUNI_MCT_DLE;
+		lle = SUNI_MCT_LLE;
+	}
 
-	control = GET(MCT) & ~(SUNI_MCT_DLE | SUNI_MCT_LLE);
+	control = dev->ops->phy_get(dev, reg) & ~(dle | lle);
 	switch (mode) {
 		case ATM_LM_NONE:
 			break;
 		case ATM_LM_LOC_PHY:
-			control |= SUNI_MCT_DLE;
+			control |= dle;
 			break;
 		case ATM_LM_RMT_PHY:
-			control |= SUNI_MCT_LLE;
+			control |= lle;
 			break;
 		default:
 			return -EINVAL;
 	}
-	PUT(control,MCT);
+	 dev->ops->phy_put(dev, control, reg);
 	PRIV(dev)->loop_mode = mode;
 	return 0;
 }
 
+/*
+ * SONET vs. SDH Configuration
+ *
+ * Z0INS (register 0x06): 0 for SONET, 1 for SDH
+ * ENSS (register 0x3D): 0 for SONET, 1 for SDH
+ * LEN16 (register 0x28): 0 for SONET, 1 for SDH (n/a for S/UNI 155 QUAD)
+ * LEN16 (register 0x50): 0 for SONET, 1 for SDH (n/a for S/UNI 155 QUAD)
+ * S[1:0] (register 0x46): 00 for SONET, 10 for SDH
+ */
+
+static int set_sonet(struct atm_dev *dev)
+{
+	if (PRIV(dev)->type == SUNI_MRI_TYPE_PM5355) {
+		PUT(GET(RPOP_RC) & ~SUNI_RPOP_RC_ENSS, RPOP_RC);
+		PUT(GET(SSTB_CTRL) & ~SUNI_SSTB_CTRL_LEN16, SSTB_CTRL);
+		PUT(GET(SPTB_CTRL) & ~SUNI_SPTB_CTRL_LEN16, SPTB_CTRL);
+	}
+
+	REG_CHANGE(SUNI_TPOP_APM_S, SUNI_TPOP_APM_S_SHIFT,
+		   SUNI_TPOP_S_SONET, TPOP_APM);
+
+	return 0;
+}
+
+static int set_sdh(struct atm_dev *dev)
+{
+	if (PRIV(dev)->type == SUNI_MRI_TYPE_PM5355) {
+		PUT(GET(RPOP_RC) | SUNI_RPOP_RC_ENSS, RPOP_RC);
+		PUT(GET(SSTB_CTRL) | SUNI_SSTB_CTRL_LEN16, SSTB_CTRL);
+		PUT(GET(SPTB_CTRL) | SUNI_SPTB_CTRL_LEN16, SPTB_CTRL);
+	}
+
+	REG_CHANGE(SUNI_TPOP_APM_S, SUNI_TPOP_APM_S_SHIFT,
+		   SUNI_TPOP_S_SDH, TPOP_APM);
+
+	return 0;
+}
+
+
+static int get_framing(struct atm_dev *dev, void __user *arg)
+{
+	int framing;
+	unsigned char s;
+
+
+	s = (GET(TPOP_APM) & SUNI_TPOP_APM_S) >> SUNI_TPOP_APM_S_SHIFT;
+	if (s == SUNI_TPOP_S_SONET)
+		framing = SONET_FRAME_SONET;
+	else
+		framing = SONET_FRAME_SDH;
+
+	return put_user(framing, (int __user *) arg) ? -EFAULT : 0;
+}
+
+static int set_framing(struct atm_dev *dev, void __user *arg)
+{
+	int mode;
+
+	if (get_user(mode, (int __user *) arg))
+		return -EFAULT;
+
+	if (mode == SONET_FRAME_SONET)
+		return set_sonet(dev);
+	else if (mode == SONET_FRAME_SDH)
+		return set_sdh(dev);
+
+	return -EINVAL;
+}
+
 
 static int suni_ioctl(struct atm_dev *dev,unsigned int cmd,void __user *arg)
 {
@@ -179,14 +265,16 @@ static int suni_ioctl(struct atm_dev *dev,unsigned int cmd,void __user *arg)
 		case SONET_GETDIAG:
 			return get_diag(dev,arg);
 		case SONET_SETFRAMING:
-			if ((int)(unsigned long)arg != SONET_FRAME_SONET) return -EINVAL;
-			return 0;
+			if (!capable(CAP_NET_ADMIN))
+				return -EPERM;
+			return set_framing(dev, arg);
 		case SONET_GETFRAMING:
-			return put_user(SONET_FRAME_SONET,(int __user *)arg) ?
-			    -EFAULT : 0;
+			return get_framing(dev, arg);
 		case SONET_GETFRSENSE:
 			return -EINVAL;
 		case ATM_SETLOOP:
+			if (!capable(CAP_NET_ADMIN))
+				return -EPERM;
 			return set_loopback(dev,(int)(unsigned long)arg);
 		case ATM_GETLOOP:
 			return put_user(PRIV(dev)->loop_mode,(int __user *)arg) ?
@@ -220,10 +308,6 @@ static int suni_start(struct atm_dev *dev)
 	unsigned long flags;
 	int first;
 
-	if (!(dev->phy_data = kmalloc(sizeof(struct suni_priv),GFP_KERNEL)))
-		return -ENOMEM;
-
-	PRIV(dev)->dev = dev;
 	spin_lock_irqsave(&sunis_lock,flags);
 	first = !sunis;
 	PRIV(dev)->next = sunis;
@@ -284,16 +368,21 @@ int suni_init(struct atm_dev *dev)
 {
 	unsigned char mri;
 
+	if (!(dev->phy_data = kmalloc(sizeof(struct suni_priv),GFP_KERNEL)))
+		return -ENOMEM;
+	PRIV(dev)->dev = dev;
+
 	mri = GET(MRI); /* reset SUNI */
+	PRIV(dev)->type = (mri & SUNI_MRI_TYPE) >> SUNI_MRI_TYPE_SHIFT;
 	PUT(mri | SUNI_MRI_RESET,MRI);
 	PUT(mri,MRI);
 	PUT((GET(MT) & SUNI_MT_DS27_53),MT); /* disable all tests */
-	REG_CHANGE(SUNI_TPOP_APM_S,SUNI_TPOP_APM_S_SHIFT,SUNI_TPOP_S_SONET,
-	    TPOP_APM); /* use SONET */
+        set_sonet(dev);
 	REG_CHANGE(SUNI_TACP_IUCHP_CLP,0,SUNI_TACP_IUCHP_CLP,
 	    TACP_IUCHP); /* idle cells */
 	PUT(SUNI_IDLE_PATTERN,TACP_IUCPOP);
 	dev->phy = &suni_ops;
+
 	return 0;
 }
 
diff --git a/drivers/atm/suni.h b/drivers/atm/suni.h
index efa79bfae75b..7e3e656b3993 100644
--- a/drivers/atm/suni.h
+++ b/drivers/atm/suni.h
@@ -1,7 +1,8 @@
-/* drivers/atm/suni.h - PMC PM5346 SUNI (PHY) declarations */
+/*
+ * drivers/atm/suni.h - S/UNI PHY driver
+ */
  
 /* Written 1995-2000 by Werner Almesberger, EPFL LRC/ICA */
- 
 
 #ifndef DRIVER_ATM_SUNI_H
 #define DRIVER_ATM_SUNI_H
@@ -39,7 +40,8 @@
 #define SUNI_RLOP_LFM		0x1F	/* RLOP Line FEBE MSB */
 #define SUNI_TLOP_CTRL		0x20	/* TLOP Control */
 #define SUNI_TLOP_DIAG		0x21	/* TLOP Diagnostic */
-			     /* 0x22-0x2F reserved */
+			     /* 0x22-0x27 reserved */
+#define SUNI_SSTB_CTRL		0x28
 #define SUNI_RPOP_SC		0x30	/* RPOP Status/Control */
 #define SUNI_RPOP_IS		0x31	/* RPOP Interrupt Status */
 			     /* 0x32 reserved */
@@ -52,6 +54,7 @@
 #define SUNI_RPOP_PFM		0x3B	/* RPOP Path FEBE MSB */
 			     /* 0x3C reserved */
 #define SUNI_RPOP_PBC		0x3D	/* RPOP Path BIP-8 Configuration */
+#define SUNI_RPOP_RC		0x3D	/* RPOP Ring Control (PM5355) */
 			     /* 0x3E-0x3F reserved */
 #define SUNI_TPOP_CD		0x40	/* TPOP Control/Diagnostic */
 #define SUNI_TPOP_PC		0x41	/* TPOP Pointer Control */
@@ -82,7 +85,8 @@
 #define SUNI_TACP_TCC		0x65	/* TACP Transmit Cell Counter */
 #define SUNI_TACP_TCCM		0x66	/* TACP Transmit Cell Counter MSB */
 #define SUNI_TACP_CFG		0x67	/* TACP Configuration */
-			     /* 0x68-0x7F reserved */
+#define SUNI_SPTB_CTRL		0x68	/* SPTB Control */
+			     /* 0x69-0x7F reserved */
 #define	SUNI_MT			0x80	/* Master Test */
 			     /* 0x81-0xFF reserved */
 
@@ -94,9 +98,18 @@
 #define SUNI_MRI_ID_SHIFT 	0
 #define SUNI_MRI_TYPE		0x70	/* R, SUNI type (lite is 011) */
 #define SUNI_MRI_TYPE_SHIFT 	4
+#define SUNI_MRI_TYPE_PM5346	0x3	/* S/UNI 155 LITE */
+#define SUNI_MRI_TYPE_PM5347	0x4	/* S/UNI 155 PLUS */
+#define SUNI_MRI_TYPE_PM5350	0x7	/* S/UNI 155 ULTRA */
+#define SUNI_MRI_TYPE_PM5355	0x1	/* S/UNI 622 */
 #define SUNI_MRI_RESET		0x80	/* RW, reset & power down chip
 					   0: normal operation
 					   1: reset & low power */
+
+/* MCM is reg 0x4 */
+#define SUNI_MCM_LLE		0x20	/* line loopback (PM5355) */
+#define SUNI_MCM_DLE		0x10	/* diagnostic loopback (PM5355) */
+
 /* MCT is reg 5 */
 #define SUNI_MCT_LOOPT		0x01	/* RW, timing source, 0: from
 					   TRCLK+/- */
@@ -144,6 +157,12 @@
 /* TLOP_DIAG is reg 0x21 */
 #define SUNI_TLOP_DIAG_DBIP	0x01	/* insert line BIP err (continuously) */
 
+/* SSTB_CTRL is reg 0x28 */
+#define SUNI_SSTB_CTRL_LEN16	0x01	/* path trace message length bit */
+
+/* RPOP_RC is reg 0x3D (PM5355) */
+#define SUNI_RPOP_RC_ENSS	0x40	/* enable size bit */
+
 /* TPOP_DIAG is reg 0x40 */
 #define SUNI_TPOP_DIAG_PAIS	0x01	/* insert STS path alarm ind (cont) */
 #define SUNI_TPOP_DIAG_DB3	0x02	/* insert path BIP err (continuously) */
@@ -191,6 +210,9 @@
 					   pattern */
 #define SUNI_TACP_IUCHP_GFC_SHIFT 4
 
+/* SPTB_CTRL is reg 0x68 */
+#define SUNI_SPTB_CTRL_LEN16	0x01	/* path trace message length */
+
 /* MT is reg 0x80 */
 #define SUNI_MT_HIZIO		0x01	/* RW, all but data bus & MP interface
 					   tri-state */
@@ -208,6 +230,7 @@
 struct suni_priv {
 	struct k_sonet_stats sonet_stats;	/* link diagnostics */
 	int loop_mode;				/* loopback mode */
+	int type;				/* phy type */
 	struct atm_dev *dev;			/* device back-pointer */
 	struct suni_priv *next;			/* next SUNI */
 };
diff --git a/include/linux/sonet.h b/include/linux/sonet.h
index 753680296e17..67ad11fcf88b 100644
--- a/include/linux/sonet.h
+++ b/include/linux/sonet.h
@@ -34,7 +34,7 @@ struct sonet_stats {
 					/* clear error insertion */
 #define SONET_GETDIAG	_IOR('a',ATMIOC_PHYTYP+4,int)
 					/* query error insertion */
-#define SONET_SETFRAMING _IO('a',ATMIOC_PHYTYP+5)
+#define SONET_SETFRAMING _IOW('a',ATMIOC_PHYTYP+5,int)
 					/* set framing mode (SONET/SDH) */
 #define SONET_GETFRAMING _IOR('a',ATMIOC_PHYTYP+6,int)
 					/* get framing mode */
-- 
cgit v1.2.3


From c1da4ac752b8b0411791d26c678fcf23d2eed242 Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@voltaire.com>
Date: Fri, 13 Jun 2008 18:12:00 -0700
Subject: net/core: add NETDEV_BONDING_FAILOVER event

Add NETDEV_BONDING_FAILOVER event to be used in a successive patch
by bonding to announce fail-over for the active-backup mode through the
netdev events notifier chain mechanism. Such an event can be of use for the
RDMA CM (communication manager) to let native RDMA ULPs (eg NFS-RDMA, iSER)
always be aligned with the IP stack, in the sense that they use the same
ports/links as the stack does. More usages can be done to allow monitoring
tools based on netlink events being aware to bonding fail-over.

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/netdevice.h | 1 +
 include/linux/notifier.h  | 1 +
 net/core/dev.c            | 6 ++++++
 3 files changed, 8 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f27fd2009334..e92fc839ab1d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1479,6 +1479,7 @@ extern void		__dev_addr_unsync(struct dev_addr_list **to, int *to_count, struct
 extern void		dev_set_promiscuity(struct net_device *dev, int inc);
 extern void		dev_set_allmulti(struct net_device *dev, int inc);
 extern void		netdev_state_change(struct net_device *dev);
+extern void		netdev_bonding_change(struct net_device *dev);
 extern void		netdev_features_change(struct net_device *dev);
 /* Load a device via the kmod */
 extern void		dev_load(struct net *net, const char *name);
diff --git a/include/linux/notifier.h b/include/linux/notifier.h
index 0ff6224d172a..bd3d72ddf333 100644
--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -197,6 +197,7 @@ static inline int notifier_to_errno(int ret)
 #define NETDEV_GOING_DOWN	0x0009
 #define NETDEV_CHANGENAME	0x000A
 #define NETDEV_FEAT_CHANGE	0x000B
+#define NETDEV_BONDING_FAILOVER 0x000C
 
 #define SYS_DOWN	0x0001	/* Notify of system down */
 #define SYS_RESTART	SYS_DOWN
diff --git a/net/core/dev.c b/net/core/dev.c
index 68d8df0992ab..0e45742e7158 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -961,6 +961,12 @@ void netdev_state_change(struct net_device *dev)
 	}
 }
 
+void netdev_bonding_change(struct net_device *dev)
+{
+	call_netdevice_notifiers(NETDEV_BONDING_FAILOVER, dev);
+}
+EXPORT_SYMBOL(netdev_bonding_change);
+
 /**
  *	dev_load 	- load a network module
  *	@net: the applicable net namespace
-- 
cgit v1.2.3


From b8a9787eddb0e4665f31dd1d64584732b2b5d051 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 13 Jun 2008 18:12:04 -0700
Subject: bonding: Allow setting max_bonds to zero

	Permit bonding to function rationally if max_bonds is set to
zero.  This will load the module, but create no master devices (which can
be created via sysfs).

	Requires some change to bond_create_sysfs; currently, the
netdev sysfs directory is determined from the first bonding device created,
but this is no longer possible.  Instead, an interface from net/core is
created to create and destroy files in net_class.

	Based on a patch submitted by Phil Oester <kernel@linuxaces.com>.
Modified by Jay Vosburgh to fix the sysfs issue mentioned above and to
update the documentation.

Signed-off-by: Phil Oester <kernel@linuxace.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 Documentation/networking/bonding.txt |  3 ++-
 drivers/net/bonding/bond_main.c      |  6 +++---
 drivers/net/bonding/bond_sysfs.c     | 22 +++-------------------
 include/linux/netdevice.h            |  3 +++
 net/core/net-sysfs.c                 | 13 +++++++++++++
 5 files changed, 24 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
index 370b7da73ab4..7fa7fe71d7a8 100644
--- a/Documentation/networking/bonding.txt
+++ b/Documentation/networking/bonding.txt
@@ -376,7 +376,8 @@ max_bonds
 	Specifies the number of bonding devices to create for this
 	instance of the bonding driver.  E.g., if max_bonds is 3, and
 	the bonding driver is not already loaded, then bond0, bond1
-	and bond2 will be created.  The default value is 1.
+	and bond2 will be created.  The default value is 1.  Specifying
+	a value of 0 will load bonding, but will not create any devices.
 
 miimon
 
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 3b6d66a8ab98..d57b65dc2c72 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -4750,11 +4750,11 @@ static int bond_check_params(struct bond_params *params)
 		}
 	}
 
-	if (max_bonds < 1 || max_bonds > INT_MAX) {
+	if (max_bonds < 0 || max_bonds > INT_MAX) {
 		printk(KERN_WARNING DRV_NAME
 		       ": Warning: max_bonds (%d) not in range %d-%d, so it "
 		       "was reset to BOND_DEFAULT_MAX_BONDS (%d)\n",
-		       max_bonds, 1, INT_MAX, BOND_DEFAULT_MAX_BONDS);
+		       max_bonds, 0, INT_MAX, BOND_DEFAULT_MAX_BONDS);
 		max_bonds = BOND_DEFAULT_MAX_BONDS;
 	}
 
@@ -4953,7 +4953,7 @@ static int bond_check_params(struct bond_params *params)
 
 		printk("\n");
 
-	} else {
+	} else if (max_bonds) {
 		/* miimon and arp_interval not set, we need one so things
 		 * work as expected, see bonding.txt for details
 		 */
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index dd265c69b0df..6caac0ffb2f2 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -53,7 +53,6 @@ extern struct bond_parm_tbl arp_validate_tbl[];
 extern struct bond_parm_tbl fail_over_mac_tbl[];
 
 static int expected_refcount = -1;
-static struct class *netdev_class;
 /*--------------------------- Data Structures -----------------------------*/
 
 /* Bonding sysfs lock.  Why can't we just use the subsystem lock?
@@ -1447,19 +1446,9 @@ static struct attribute_group bonding_group = {
  */
 int bond_create_sysfs(void)
 {
-	int ret = 0;
-	struct bonding *firstbond;
-
-	/* get the netdev class pointer */
-	firstbond = container_of(bond_dev_list.next, struct bonding, bond_list);
-	if (!firstbond)
-		return -ENODEV;
-
-	netdev_class = firstbond->dev->dev.class;
-	if (!netdev_class)
-		return -ENODEV;
+	int ret;
 
-	ret = class_create_file(netdev_class, &class_attr_bonding_masters);
+	ret = netdev_class_create_file(&class_attr_bonding_masters);
 	/*
 	 * Permit multiple loads of the module by ignoring failures to
 	 * create the bonding_masters sysfs file.  Bonding devices
@@ -1478,10 +1467,6 @@ int bond_create_sysfs(void)
 			printk(KERN_ERR
 			       "network device named %s already exists in sysfs",
 			       class_attr_bonding_masters.attr.name);
-		else {
-			netdev_class = NULL;
-			return 0;
-		}
 	}
 
 	return ret;
@@ -1493,8 +1478,7 @@ int bond_create_sysfs(void)
  */
 void bond_destroy_sysfs(void)
 {
-	if (netdev_class)
-		class_remove_file(netdev_class, &class_attr_bonding_masters);
+	netdev_class_remove_file(&class_attr_bonding_masters);
 }
 
 /*
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e92fc839ab1d..9ccbfac3fd95 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1506,6 +1506,9 @@ extern void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos);
 extern void dev_seq_stop(struct seq_file *seq, void *v);
 #endif
 
+extern int netdev_class_create_file(struct class_attribute *class_attr);
+extern void netdev_class_remove_file(struct class_attribute *class_attr);
+
 extern void linkwatch_run_queue(void);
 
 extern int netdev_compute_features(unsigned long all, unsigned long one);
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index dccd737ea2e3..3f7941319217 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -468,6 +468,19 @@ int netdev_register_kobject(struct net_device *net)
 	return device_add(dev);
 }
 
+int netdev_class_create_file(struct class_attribute *class_attr)
+{
+	return class_create_file(&net_class, class_attr);
+}
+
+void netdev_class_remove_file(struct class_attribute *class_attr)
+{
+	class_remove_file(&net_class, class_attr);
+}
+
+EXPORT_SYMBOL(netdev_class_create_file);
+EXPORT_SYMBOL(netdev_class_remove_file);
+
 void netdev_initialize_kobject(struct net_device *net)
 {
 	struct device *device = &(net->dev);
-- 
cgit v1.2.3


From 9d45abe1c2949183e5d9cb25721bf1c42c7b5e3b Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Tue, 17 Jun 2008 21:12:48 -0700
Subject: netdevice: change net_device->promiscuity/allmulti to unsigned int

The comments of dev_set_allmulti/promiscuity() is that "While the count in
the device remains above zero...". So negative count is useless.
Fix the type of the counter from "int" to "unsigned int".

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f27fd2009334..06d8ea5992df 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -595,8 +595,8 @@ struct net_device
 	int			uc_promisc;
 	struct dev_addr_list	*mc_list;	/* Multicast mac addresses	*/
 	int			mc_count;	/* Number of installed mcasts	*/
-	int			promiscuity;
-	int			allmulti;
+	unsigned int		promiscuity;
+	unsigned int		allmulti;
 
 
 	/* Protocol specific pointers */
-- 
cgit v1.2.3


From dad9b335c6940de2746a9788eb456d09cf102f81 Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Wed, 18 Jun 2008 01:48:28 -0700
Subject: netdevice: Fix promiscuity and allmulti overflow

Max of promiscuity and allmulti plus positive @inc can cause overflow.
Fox example: when allmulti=0xFFFFFFFF, any caller give dev_set_allmulti() a
positive @inc will cause allmulti be off.
This is not what we want, though it's rare case.
The fix is that only negative @inc will cause allmulti or promiscuity be off
and when any caller makes the counters touch the roof, we return error.

Change of v2:
Change void function dev_set_promiscuity/allmulti to return int.
So callers can get the overflow error.
Caller's fix will be done later.

Change of v3:
1. Since we return error to caller, we don't need to print KERN_ERROR,
KERN_WARNING is enough.
2. In dev_set_promiscuity(), if __dev_set_promiscuity() failed, we
return at once.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 ++--
 net/core/dev.c            | 55 ++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 47 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 4bf613cd9e2d..45dce2b58d4c 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1476,8 +1476,8 @@ extern int 		__dev_addr_delete(struct dev_addr_list **list, int *count, void *ad
 extern int		__dev_addr_add(struct dev_addr_list **list, int *count, void *addr, int alen, int newonly);
 extern int		__dev_addr_sync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count);
 extern void		__dev_addr_unsync(struct dev_addr_list **to, int *to_count, struct dev_addr_list **from, int *from_count);
-extern void		dev_set_promiscuity(struct net_device *dev, int inc);
-extern void		dev_set_allmulti(struct net_device *dev, int inc);
+extern int		dev_set_promiscuity(struct net_device *dev, int inc);
+extern int		dev_set_allmulti(struct net_device *dev, int inc);
 extern void		netdev_state_change(struct net_device *dev);
 extern void		netdev_bonding_change(struct net_device *dev);
 extern void		netdev_features_change(struct net_device *dev);
diff --git a/net/core/dev.c b/net/core/dev.c
index 0e45742e7158..a495f712d38c 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2771,16 +2771,29 @@ int netdev_set_master(struct net_device *slave, struct net_device *master)
 	return 0;
 }
 
-static void __dev_set_promiscuity(struct net_device *dev, int inc)
+static int __dev_set_promiscuity(struct net_device *dev, int inc)
 {
 	unsigned short old_flags = dev->flags;
 
 	ASSERT_RTNL();
 
-	if ((dev->promiscuity += inc) == 0)
-		dev->flags &= ~IFF_PROMISC;
-	else
-		dev->flags |= IFF_PROMISC;
+	dev->flags |= IFF_PROMISC;
+	dev->promiscuity += inc;
+	if (dev->promiscuity == 0) {
+		/*
+		 * Avoid overflow.
+		 * If inc causes overflow, untouch promisc and return error.
+		 */
+		if (inc < 0)
+			dev->flags &= ~IFF_PROMISC;
+		else {
+			dev->promiscuity -= inc;
+			printk(KERN_WARNING "%s: promiscuity touches roof, "
+				"set promiscuity failed, promiscuity feature "
+				"of device might be broken.\n", dev->name);
+			return -EOVERFLOW;
+		}
+	}
 	if (dev->flags != old_flags) {
 		printk(KERN_INFO "device %s %s promiscuous mode\n",
 		       dev->name, (dev->flags & IFF_PROMISC) ? "entered" :
@@ -2798,6 +2811,7 @@ static void __dev_set_promiscuity(struct net_device *dev, int inc)
 		if (dev->change_rx_flags)
 			dev->change_rx_flags(dev, IFF_PROMISC);
 	}
+	return 0;
 }
 
 /**
@@ -2809,14 +2823,19 @@ static void __dev_set_promiscuity(struct net_device *dev, int inc)
  *	remains above zero the interface remains promiscuous. Once it hits zero
  *	the device reverts back to normal filtering operation. A negative inc
  *	value is used to drop promiscuity on the device.
+ *	Return 0 if successful or a negative errno code on error.
  */
-void dev_set_promiscuity(struct net_device *dev, int inc)
+int dev_set_promiscuity(struct net_device *dev, int inc)
 {
 	unsigned short old_flags = dev->flags;
+	int err;
 
-	__dev_set_promiscuity(dev, inc);
+	err = __dev_set_promiscuity(dev, inc);
+	if (!err)
+		return err;
 	if (dev->flags != old_flags)
 		dev_set_rx_mode(dev);
+	return err;
 }
 
 /**
@@ -2829,22 +2848,38 @@ void dev_set_promiscuity(struct net_device *dev, int inc)
  *	to all interfaces. Once it hits zero the device reverts back to normal
  *	filtering operation. A negative @inc value is used to drop the counter
  *	when releasing a resource needing all multicasts.
+ *	Return 0 if successful or a negative errno code on error.
  */
 
-void dev_set_allmulti(struct net_device *dev, int inc)
+int dev_set_allmulti(struct net_device *dev, int inc)
 {
 	unsigned short old_flags = dev->flags;
 
 	ASSERT_RTNL();
 
 	dev->flags |= IFF_ALLMULTI;
-	if ((dev->allmulti += inc) == 0)
-		dev->flags &= ~IFF_ALLMULTI;
+	dev->allmulti += inc;
+	if (dev->allmulti == 0) {
+		/*
+		 * Avoid overflow.
+		 * If inc causes overflow, untouch allmulti and return error.
+		 */
+		if (inc < 0)
+			dev->flags &= ~IFF_ALLMULTI;
+		else {
+			dev->allmulti -= inc;
+			printk(KERN_WARNING "%s: allmulti touches roof, "
+				"set allmulti failed, allmulti feature of "
+				"device might be broken.\n", dev->name);
+			return -EOVERFLOW;
+		}
+	}
 	if (dev->flags ^ old_flags) {
 		if (dev->change_rx_flags)
 			dev->change_rx_flags(dev, IFF_ALLMULTI);
 		dev_set_rx_mode(dev);
 	}
+	return 0;
 }
 
 /*
-- 
cgit v1.2.3


From e17ba73b0ee6c0f24393c48b455e0d8db761782c Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Mon, 12 May 2008 15:44:40 +0200
Subject: x86, generic: mark early_printk as asmlinkage

It's not explicitly marked as asmlinkage, but invoked from x86_32
startup code with parameters on stack.

No other architectures define early_printk and none of them are affected
by this change, since defines asmlinkage as empty token.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/early_printk.c | 2 +-
 include/linux/kernel.h         | 2 +-
 kernel/printk.c                | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index 643fd861b724..ff9e7350da54 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -196,7 +196,7 @@ static struct console simnow_console = {
 static struct console *early_console = &early_vga_console;
 static int early_console_initialized;
 
-void early_printk(const char *fmt, ...)
+asmlinkage void early_printk(const char *fmt, ...)
 {
 	char buf[512];
 	int n;
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index f2a668c195bf..4cb8d3df414e 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -207,7 +207,7 @@ static inline bool printk_timed_ratelimit(unsigned long *caller_jiffies, \
 		{ return false; }
 #endif
 
-extern void __attribute__((format(printf, 1, 2)))
+extern void asmlinkage __attribute__((format(printf, 1, 2)))
 	early_printk(const char *fmt, ...);
 
 unsigned long int_sqrt(unsigned long);
diff --git a/kernel/printk.c b/kernel/printk.c
index 70cfa5ac75ce..de1a4f4470c3 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -38,7 +38,7 @@
 /*
  * Architectures can override it:
  */
-void __attribute__((weak)) early_printk(const char *fmt, ...)
+void asmlinkage __attribute__((weak)) early_printk(const char *fmt, ...)
 {
 }
 
-- 
cgit v1.2.3


From 0187bdfb05674147774ca79a79942537f3ad54bd Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 19 Jun 2008 16:15:47 -0700
Subject: net: Disable LRO on devices that are forwarding

Large Receive Offload (LRO) is only appropriate for packets that are
destined for the host, and should be disabled if received packets may be
forwarded.  It can also confuse the GSO on output.

Add dev_disable_lro() function which uses the appropriate ethtool ops to
disable LRO if enabled.

Add calls to dev_disable_lro() in br_add_if() and functions that enable
IPv4 and IPv6 forwarding.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 +
 net/bridge/br_if.c        |  1 +
 net/core/dev.c            | 24 ++++++++++++++++++++++++
 net/ipv4/devinet.c        | 21 ++++++++++++++++-----
 net/ipv6/addrconf.c       |  6 ++++++
 5 files changed, 48 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 45dce2b58d4c..1304ad2d7105 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -886,6 +886,7 @@ extern struct net_device	*__dev_get_by_name(struct net *net, const char *name);
 extern int		dev_alloc_name(struct net_device *dev, const char *name);
 extern int		dev_open(struct net_device *dev);
 extern int		dev_close(struct net_device *dev);
+extern void		dev_disable_lro(struct net_device *dev);
 extern int		dev_queue_xmit(struct sk_buff *skb);
 extern int		register_netdevice(struct net_device *dev);
 extern void		unregister_netdevice(struct net_device *dev);
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 143c954681b8..832a561500d9 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -387,6 +387,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
 		goto err2;
 
 	rcu_assign_pointer(dev->br_port, p);
+	dev_disable_lro(dev);
 	dev_set_promiscuity(dev, 1);
 
 	list_add_rcu(&p->list, &br->port_list);
diff --git a/net/core/dev.c b/net/core/dev.c
index a495f712d38c..f6944ecd5b2e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -90,6 +90,7 @@
 #include <linux/if_ether.h>
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
+#include <linux/ethtool.h>
 #include <linux/notifier.h>
 #include <linux/skbuff.h>
 #include <net/net_namespace.h>
@@ -1123,6 +1124,29 @@ int dev_close(struct net_device *dev)
 }
 
 
+/**
+ *	dev_disable_lro - disable Large Receive Offload on a device
+ *	@dev: device
+ *
+ *	Disable Large Receive Offload (LRO) on a net device.  Must be
+ *	called under RTNL.  This is needed if received packets may be
+ *	forwarded to another interface.
+ */
+void dev_disable_lro(struct net_device *dev)
+{
+	if (dev->ethtool_ops && dev->ethtool_ops->get_flags &&
+	    dev->ethtool_ops->set_flags) {
+		u32 flags = dev->ethtool_ops->get_flags(dev);
+		if (flags & ETH_FLAG_LRO) {
+			flags &= ~ETH_FLAG_LRO;
+			dev->ethtool_ops->set_flags(dev, flags);
+		}
+	}
+	WARN_ON(dev->features & NETIF_F_LRO);
+}
+EXPORT_SYMBOL(dev_disable_lro);
+
+
 static int dev_boot_phase = 1;
 
 /*
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index f8c0b0aea93a..9de2514946ca 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -168,6 +168,8 @@ static struct in_device *inetdev_init(struct net_device *dev)
 	in_dev->dev = dev;
 	if ((in_dev->arp_parms = neigh_parms_alloc(dev, &arp_tbl)) == NULL)
 		goto out_kfree;
+	if (IPV4_DEVCONF(in_dev->cnf, FORWARDING))
+		dev_disable_lro(dev);
 	/* Reference in_dev->dev */
 	dev_hold(dev);
 	/* Account for reference dev->ip_ptr (below) */
@@ -1241,6 +1243,8 @@ static void inet_forward_change(struct net *net)
 	read_lock(&dev_base_lock);
 	for_each_netdev(net, dev) {
 		struct in_device *in_dev;
+		if (on)
+			dev_disable_lro(dev);
 		rcu_read_lock();
 		in_dev = __in_dev_get_rcu(dev);
 		if (in_dev)
@@ -1248,8 +1252,6 @@ static void inet_forward_change(struct net *net)
 		rcu_read_unlock();
 	}
 	read_unlock(&dev_base_lock);
-
-	rt_cache_flush(0);
 }
 
 static int devinet_conf_proc(ctl_table *ctl, int write,
@@ -1335,10 +1337,19 @@ static int devinet_sysctl_forward(ctl_table *ctl, int write,
 	if (write && *valp != val) {
 		struct net *net = ctl->extra2;
 
-		if (valp == &IPV4_DEVCONF_ALL(net, FORWARDING))
-			inet_forward_change(net);
-		else if (valp != &IPV4_DEVCONF_DFLT(net, FORWARDING))
+		if (valp != &IPV4_DEVCONF_DFLT(net, FORWARDING)) {
+			rtnl_lock();
+			if (valp == &IPV4_DEVCONF_ALL(net, FORWARDING)) {
+				inet_forward_change(net);
+			} else if (*valp) {
+				struct ipv4_devconf *cnf = ctl->extra1;
+				struct in_device *idev =
+					container_of(cnf, struct in_device, cnf);
+				dev_disable_lro(idev->dev);
+			}
+			rtnl_unlock();
 			rt_cache_flush(0);
+		}
 	}
 
 	return ret;
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 9be6be3a7ff3..84127d854cfc 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -348,6 +348,8 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 		kfree(ndev);
 		return NULL;
 	}
+	if (ndev->cnf.forwarding)
+		dev_disable_lro(dev);
 	/* We refer to the device */
 	dev_hold(dev);
 
@@ -442,6 +444,8 @@ static void dev_forward_change(struct inet6_dev *idev)
 	if (!idev)
 		return;
 	dev = idev->dev;
+	if (idev->cnf.forwarding)
+		dev_disable_lro(dev);
 	if (dev && (dev->flags & IFF_MULTICAST)) {
 		if (idev->cnf.forwarding)
 			ipv6_dev_mc_inc(dev, &in6addr_linklocal_allrouters);
@@ -487,12 +491,14 @@ static void addrconf_fixup_forwarding(struct ctl_table *table, int *p, int old)
 	if (p == &net->ipv6.devconf_dflt->forwarding)
 		return;
 
+	rtnl_lock();
 	if (p == &net->ipv6.devconf_all->forwarding) {
 		__s32 newf = net->ipv6.devconf_all->forwarding;
 		net->ipv6.devconf_dflt->forwarding = newf;
 		addrconf_forward_change(net, newf);
 	} else if ((!*p) ^ (!old))
 		dev_forward_change((struct inet6_dev *)table->extra1);
+	rtnl_unlock();
 
 	if (*p)
 		rt6_purge_dflt_routers(net);
-- 
cgit v1.2.3


From 4497b0763cb1afae463f5e144c28b5d806e28b60 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 19 Jun 2008 16:22:28 -0700
Subject: net: Discard and warn about LRO'd skbs received for forwarding

Add skb_warn_if_lro() to test whether an skb was received with LRO and
warn if so.

Change br_forward(), ip_forward() and ip6_forward() to call it) and
discard the skb if it returns true.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h  | 14 ++++++++++++++
 net/bridge/br_forward.c |  2 +-
 net/core/skbuff.c       |  8 ++++++++
 net/ipv4/ip_forward.c   |  3 +++
 net/ipv6/ip6_output.c   |  3 +++
 5 files changed, 29 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 299ec4b31412..2220b9e2dab0 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1702,6 +1702,20 @@ static inline int skb_is_gso_v6(const struct sk_buff *skb)
 	return skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6;
 }
 
+extern void __skb_warn_lro_forwarding(const struct sk_buff *skb);
+
+static inline bool skb_warn_if_lro(const struct sk_buff *skb)
+{
+	/* LRO sets gso_size but not gso_type, whereas if GSO is really
+	 * wanted then gso_type will be set. */
+	struct skb_shared_info *shinfo = skb_shinfo(skb);
+	if (shinfo->gso_size != 0 && unlikely(shinfo->gso_type == 0)) {
+		__skb_warn_lro_forwarding(skb);
+		return true;
+	}
+	return false;
+}
+
 static inline void skb_forward_csum(struct sk_buff *skb)
 {
 	/* Unfortunately we don't support this one.  Any brave souls? */
diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 512645727f51..bdd9ccea17ce 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -89,7 +89,7 @@ void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb)
 /* called with rcu_read_lock */
 void br_forward(const struct net_bridge_port *to, struct sk_buff *skb)
 {
-	if (should_deliver(to, skb)) {
+	if (!skb_warn_if_lro(skb) && should_deliver(to, skb)) {
 		__br_forward(to, skb);
 		return;
 	}
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 3e18f8525e82..2df012be973d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -2583,6 +2583,13 @@ bool skb_partial_csum_set(struct sk_buff *skb, u16 start, u16 off)
 	return true;
 }
 
+void __skb_warn_lro_forwarding(const struct sk_buff *skb)
+{
+	if (net_ratelimit())
+		pr_warning("%s: received packets cannot be forwarded"
+			   " while LRO is enabled\n", skb->dev->name);
+}
+
 EXPORT_SYMBOL(___pskb_trim);
 EXPORT_SYMBOL(__kfree_skb);
 EXPORT_SYMBOL(kfree_skb);
@@ -2616,6 +2623,7 @@ EXPORT_SYMBOL(skb_seq_read);
 EXPORT_SYMBOL(skb_abort_seq_read);
 EXPORT_SYMBOL(skb_find_text);
 EXPORT_SYMBOL(skb_append_datato_frags);
+EXPORT_SYMBOL(__skb_warn_lro_forwarding);
 
 EXPORT_SYMBOL_GPL(skb_to_sgvec);
 EXPORT_SYMBOL_GPL(skb_cow_data);
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 37d36a3f33cd..da14725916d3 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -56,6 +56,9 @@ int ip_forward(struct sk_buff *skb)
 	struct rtable *rt;	/* Route we use */
 	struct ip_options * opt	= &(IPCB(skb)->opt);
 
+	if (skb_warn_if_lro(skb))
+		goto drop;
+
 	if (!xfrm4_policy_check(NULL, XFRM_POLICY_FWD, skb))
 		goto drop;
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 40a2813a63d1..fd7cd1bfe151 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -407,6 +407,9 @@ int ip6_forward(struct sk_buff *skb)
 	if (ipv6_devconf.forwarding == 0)
 		goto error;
 
+	if (skb_warn_if_lro(skb))
+		goto drop;
+
 	if (!xfrm6_policy_check(NULL, XFRM_POLICY_FWD, skb)) {
 		IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_INDISCARDS);
 		goto drop;
-- 
cgit v1.2.3


From 443cd507ce7f78c6f8742b72736585c031d5a921 Mon Sep 17 00:00:00 2001
From: "Huang, Ying" <ying.huang@intel.com>
Date: Fri, 20 Jun 2008 16:39:21 +0800
Subject: lockdep: add lock_class information to lock_chain and output it

This patch records array of lock_class into lock_chain, and export
lock_chain information via /proc/lockdep_chains.

It is based on x86/master branch of git-x86 tree, and has been tested
on x86_64 platform.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/lockdep.h    |  3 ++
 kernel/lockdep.c           | 38 +++++++++++++++++--
 kernel/lockdep_internals.h |  6 +++
 kernel/lockdep_proc.c      | 91 ++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 135 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 4c4d236ded18..b26fbc715a50 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -182,6 +182,9 @@ struct lock_list {
  * We record lock dependency chains, so that we can cache them:
  */
 struct lock_chain {
+	u8				irq_context;
+	u8				depth;
+	u16				base;
 	struct list_head		entry;
 	u64				chain_key;
 };
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 81a4e4a3f087..a796f1f38ac5 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -1458,7 +1458,14 @@ out_bug:
 }
 
 unsigned long nr_lock_chains;
-static struct lock_chain lock_chains[MAX_LOCKDEP_CHAINS];
+struct lock_chain lock_chains[MAX_LOCKDEP_CHAINS];
+atomic_t nr_chain_hlocks;
+static u16 chain_hlocks[MAX_LOCKDEP_CHAIN_HLOCKS];
+
+struct lock_class *lock_chain_get_class(struct lock_chain *chain, int i)
+{
+	return lock_classes + chain_hlocks[chain->base + i];
+}
 
 /*
  * Look up a dependency chain. If the key is not present yet then
@@ -1466,10 +1473,15 @@ static struct lock_chain lock_chains[MAX_LOCKDEP_CHAINS];
  * validated. If the key is already hashed, return 0.
  * (On return with 1 graph_lock is held.)
  */
-static inline int lookup_chain_cache(u64 chain_key, struct lock_class *class)
+static inline int lookup_chain_cache(struct task_struct *curr,
+				     struct held_lock *hlock,
+				     u64 chain_key)
 {
+	struct lock_class *class = hlock->class;
 	struct list_head *hash_head = chainhashentry(chain_key);
 	struct lock_chain *chain;
+	struct held_lock *hlock_curr, *hlock_next;
+	int i, j, n;
 
 	if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
 		return 0;
@@ -1517,6 +1529,26 @@ cache_hit:
 	}
 	chain = lock_chains + nr_lock_chains++;
 	chain->chain_key = chain_key;
+	chain->irq_context = hlock->irq_context;
+	/* Find the first held_lock of current chain */
+	hlock_next = hlock;
+	for (i = curr->lockdep_depth - 1; i >= 0; i--) {
+		hlock_curr = curr->held_locks + i;
+		if (hlock_curr->irq_context != hlock_next->irq_context)
+			break;
+		hlock_next = hlock;
+	}
+	i++;
+	chain->depth = curr->lockdep_depth + 1 - i;
+	n = atomic_add_return(chain->depth, &nr_chain_hlocks);
+	if (unlikely(n < MAX_LOCKDEP_CHAIN_HLOCKS)) {
+		chain->base = n - chain->depth;
+		for (j = 0; j < chain->depth - 1; j++, i++) {
+			int lock_id = curr->held_locks[i].class - lock_classes;
+			chain_hlocks[chain->base + j] = lock_id;
+		}
+		chain_hlocks[chain->base + j] = class - lock_classes;
+	}
 	list_add_tail_rcu(&chain->entry, hash_head);
 	debug_atomic_inc(&chain_lookup_misses);
 	inc_chains();
@@ -1538,7 +1570,7 @@ static int validate_chain(struct task_struct *curr, struct lockdep_map *lock,
 	 * graph_lock for us)
 	 */
 	if (!hlock->trylock && (hlock->check == 2) &&
-			lookup_chain_cache(chain_key, hlock->class)) {
+	    lookup_chain_cache(curr, hlock, chain_key)) {
 		/*
 		 * Check whether last held lock:
 		 *
diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h
index 8ce09bc4613d..db09b176dd34 100644
--- a/kernel/lockdep_internals.h
+++ b/kernel/lockdep_internals.h
@@ -23,6 +23,8 @@
 #define MAX_LOCKDEP_CHAINS_BITS	14
 #define MAX_LOCKDEP_CHAINS	(1UL << MAX_LOCKDEP_CHAINS_BITS)
 
+#define MAX_LOCKDEP_CHAIN_HLOCKS (MAX_LOCKDEP_CHAINS*5)
+
 /*
  * Stack-trace: tightly packed array of stack backtrace
  * addresses. Protected by the hash_lock.
@@ -30,15 +32,19 @@
 #define MAX_STACK_TRACE_ENTRIES	262144UL
 
 extern struct list_head all_lock_classes;
+extern struct lock_chain lock_chains[];
 
 extern void
 get_usage_chars(struct lock_class *class, char *c1, char *c2, char *c3, char *c4);
 
 extern const char * __get_key_name(struct lockdep_subclass_key *key, char *str);
 
+struct lock_class *lock_chain_get_class(struct lock_chain *chain, int i);
+
 extern unsigned long nr_lock_classes;
 extern unsigned long nr_list_entries;
 extern unsigned long nr_lock_chains;
+extern atomic_t nr_chain_hlocks;
 extern unsigned long nr_stack_trace_entries;
 
 extern unsigned int nr_hardirq_chains;
diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c
index 688c5f1940bd..14d052c8a835 100644
--- a/kernel/lockdep_proc.c
+++ b/kernel/lockdep_proc.c
@@ -178,6 +178,93 @@ static const struct file_operations proc_lockdep_operations = {
 	.release	= seq_release,
 };
 
+static void *lc_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	struct lock_chain *chain;
+
+	(*pos)++;
+
+	if (v == SEQ_START_TOKEN)
+		chain = m->private;
+	else {
+		chain = v;
+
+		if (*pos < nr_lock_chains)
+			chain = lock_chains + *pos;
+		else
+			chain = NULL;
+	}
+
+	return chain;
+}
+
+static void *lc_start(struct seq_file *m, loff_t *pos)
+{
+	if (*pos == 0)
+		return SEQ_START_TOKEN;
+
+	if (*pos < nr_lock_chains)
+		return lock_chains + *pos;
+
+	return NULL;
+}
+
+static void lc_stop(struct seq_file *m, void *v)
+{
+}
+
+static int lc_show(struct seq_file *m, void *v)
+{
+	struct lock_chain *chain = v;
+	struct lock_class *class;
+	int i;
+
+	if (v == SEQ_START_TOKEN) {
+		seq_printf(m, "all lock chains:\n");
+		return 0;
+	}
+
+	seq_printf(m, "irq_context: %d\n", chain->irq_context);
+
+	for (i = 0; i < chain->depth; i++) {
+		class = lock_chain_get_class(chain, i);
+		seq_printf(m, "[%p] ", class->key);
+		print_name(m, class);
+		seq_puts(m, "\n");
+	}
+	seq_puts(m, "\n");
+
+	return 0;
+}
+
+static const struct seq_operations lockdep_chains_ops = {
+	.start	= lc_start,
+	.next	= lc_next,
+	.stop	= lc_stop,
+	.show	= lc_show,
+};
+
+static int lockdep_chains_open(struct inode *inode, struct file *file)
+{
+	int res = seq_open(file, &lockdep_chains_ops);
+	if (!res) {
+		struct seq_file *m = file->private_data;
+
+		if (nr_lock_chains)
+			m->private = lock_chains;
+		else
+			m->private = NULL;
+	}
+	return res;
+}
+
+static const struct file_operations proc_lockdep_chains_operations = {
+	.open		= lockdep_chains_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
 static void lockdep_stats_debug_show(struct seq_file *m)
 {
 #ifdef CONFIG_DEBUG_LOCKDEP
@@ -294,6 +381,8 @@ static int lockdep_stats_show(struct seq_file *m, void *v)
 #ifdef CONFIG_PROVE_LOCKING
 	seq_printf(m, " dependency chains:             %11lu [max: %lu]\n",
 			nr_lock_chains, MAX_LOCKDEP_CHAINS);
+	seq_printf(m, " dependency chain hlocks:       %11d [max: %lu]\n",
+			atomic_read(&nr_chain_hlocks), MAX_LOCKDEP_CHAIN_HLOCKS);
 #endif
 
 #ifdef CONFIG_TRACE_IRQFLAGS
@@ -661,6 +750,8 @@ static const struct file_operations proc_lock_stat_operations = {
 static int __init lockdep_proc_init(void)
 {
 	proc_create("lockdep", S_IRUSR, NULL, &proc_lockdep_operations);
+	proc_create("lockdep_chains", S_IRUSR, NULL,
+		    &proc_lockdep_chains_operations);
 	proc_create("lockdep_stats", S_IRUSR, NULL,
 		    &proc_lockdep_stats_operations);
 
-- 
cgit v1.2.3


From 0b2806768899dba5967bcd4a3b93eaed9a1dc4f3 Mon Sep 17 00:00:00 2001
From: Jonathan Corbet <corbet@lwn.net>
Date: Sun, 18 May 2008 14:27:41 -0600
Subject: Add cycle_kernel_lock()

A number of driver functions are so obviously trivial that they do not need
the big kernel lock - at least not overtly.  It turns out that the
acquisition of the BKL in driver open() functions can perform a sort of
poor-hacker's serialization function, delaying the open operation until the
driver is certain to have completed its initialization.  Add a simple
cycle_kernel_lock() function for these cases to make it clear that there is
no need to *hold* the BKL, just to be sure that we can acquire it.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
---
 include/linux/smp_lock.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/smp_lock.h b/include/linux/smp_lock.h
index aab3a4cff4e1..813be59bf345 100644
--- a/include/linux/smp_lock.h
+++ b/include/linux/smp_lock.h
@@ -27,11 +27,24 @@ static inline int reacquire_kernel_lock(struct task_struct *task)
 extern void __lockfunc lock_kernel(void)	__acquires(kernel_lock);
 extern void __lockfunc unlock_kernel(void)	__releases(kernel_lock);
 
+/*
+ * Various legacy drivers don't really need the BKL in a specific
+ * function, but they *do* need to know that the BKL became available.
+ * This function just avoids wrapping a bunch of lock/unlock pairs
+ * around code which doesn't really need it.
+ */
+static inline void cycle_kernel_lock(void)
+{
+	lock_kernel();
+	unlock_kernel();
+}
+
 #else
 
 #define lock_kernel()				do { } while(0)
 #define unlock_kernel()				do { } while(0)
 #define release_kernel_lock(task)		do { } while(0)
+#define cycle_kernel_lock()			do { } while(0)
 #define reacquire_kernel_lock(task)		0
 #define kernel_locked()				1
 
-- 
cgit v1.2.3


From 7c11337d9d81cde0a08a0da63cbfb20653890fa1 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 4 Jun 2008 18:50:06 -0400
Subject: nfsd: remove three unused NFS4_ACE_* defines

These flag bits aren't used by either the protocol or our
implementation, so I don't know why they were here.

Thanks to Johann Dahm for running across these.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Cc: Johann Dahm <jdahm@umich.edu>
---
 include/linux/nfs4.h | 3 ---
 1 file changed, 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 8726491de154..ea0366769484 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -65,9 +65,6 @@
 #define NFS4_ACE_SUCCESSFUL_ACCESS_ACE_FLAG   0x00000010
 #define NFS4_ACE_FAILED_ACCESS_ACE_FLAG       0x00000020
 #define NFS4_ACE_IDENTIFIER_GROUP             0x00000040
-#define NFS4_ACE_OWNER                        0x00000080
-#define NFS4_ACE_GROUP                        0x00000100
-#define NFS4_ACE_EVERYONE                     0x00000200
 
 #define NFS4_ACE_READ_DATA                    0x00000001
 #define NFS4_ACE_LIST_DIRECTORY               0x00000001
-- 
cgit v1.2.3


From a5e561fee651eb03086ca21e9aa78e1fffa4581a Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Tue, 10 Jun 2008 12:59:07 +0300
Subject: nfsd: eliminate unused nfs4_callback.cb_program

The cb_program member of struct nfs4_callback unused
since commit ff7d9756 nfsd: use static memory for callback program and stats

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/nfsd/state.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index db348f749376..06e9686d9dea 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -98,7 +98,6 @@ struct nfs4_callback {
 	u32                     cb_ident;
 	/* RPC client info */
 	atomic_t		cb_set;     /* successful CB_NULL call */
-	struct rpc_program      cb_program;
 	struct rpc_stat         cb_stat;
 	struct rpc_clnt *       cb_client;
 };
-- 
cgit v1.2.3


From 0d169ca136357d51a65d686f3c84866a8ba20ae9 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Tue, 10 Jun 2008 13:39:43 +0300
Subject: nfsd: eliminate unused nfs4_callback.cb_stat

The cb_stat member of struct nfs4_callback is unused
since commit ff7d9756 nfsd: use static memory for callback program and stats

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/nfsd/state.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 06e9686d9dea..d0fe2e378452 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -98,7 +98,6 @@ struct nfs4_callback {
 	u32                     cb_ident;
 	/* RPC client info */
 	atomic_t		cb_set;     /* successful CB_NULL call */
-	struct rpc_stat         cb_stat;
 	struct rpc_clnt *       cb_client;
 };
 
-- 
cgit v1.2.3


From bedbdd8bada194a690d2901801bf8451965086b3 Mon Sep 17 00:00:00 2001
From: Neil Brown <neilb@suse.de>
Date: Tue, 10 Jun 2008 08:40:35 -0400
Subject: knfsd: Replace lock_kernel with a mutex for nfsd thread
 startup/shutdown locking.

This removes the BKL from the RPC service creation codepath. The BKL
really isn't adequate for this job since some of this info needs
protection across sleeps.

Also, add some comments to try and clarify how the locking should work
and to make it clear that the BKL isn't necessary as long as there is
adequate locking between tasks when touching the svc_serv fields.

Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c          | 37 +++++++++++++++++++++++--------------
 fs/nfsd/nfssvc.c          | 45 ++++++++++++++++++++++++++++++++-------------
 include/linux/nfsd/nfsd.h |  1 +
 net/sunrpc/svc.c          | 15 +++++++++------
 4 files changed, 65 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 5ac00c4fee91..049d2a9c7715 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -450,22 +450,26 @@ static ssize_t write_pool_threads(struct file *file, char *buf, size_t size)
 	int i;
 	int rv;
 	int len;
-    	int npools = nfsd_nrpools();
+	int npools;
 	int *nthreads;
 
+	mutex_lock(&nfsd_mutex);
+	npools = nfsd_nrpools();
 	if (npools == 0) {
 		/*
 		 * NFS is shut down.  The admin can start it by
 		 * writing to the threads file but NOT the pool_threads
 		 * file, sorry.  Report zero threads.
 		 */
+		mutex_unlock(&nfsd_mutex);
 		strcpy(buf, "0\n");
 		return strlen(buf);
 	}
 
 	nthreads = kcalloc(npools, sizeof(int), GFP_KERNEL);
+	rv = -ENOMEM;
 	if (nthreads == NULL)
-		return -ENOMEM;
+		goto out_free;
 
 	if (size > 0) {
 		for (i = 0; i < npools; i++) {
@@ -496,10 +500,12 @@ static ssize_t write_pool_threads(struct file *file, char *buf, size_t size)
 		mesg += len;
 	}
 
+	mutex_unlock(&nfsd_mutex);
 	return (mesg-buf);
 
 out_free:
 	kfree(nthreads);
+	mutex_unlock(&nfsd_mutex);
 	return rv;
 }
 
@@ -566,14 +572,13 @@ static ssize_t write_versions(struct file *file, char *buf, size_t size)
 	return len;
 }
 
-static ssize_t write_ports(struct file *file, char *buf, size_t size)
+static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 {
 	if (size == 0) {
 		int len = 0;
-		lock_kernel();
+
 		if (nfsd_serv)
 			len = svc_xprt_names(nfsd_serv, buf, 0);
-		unlock_kernel();
 		return len;
 	}
 	/* Either a single 'fd' number is written, in which
@@ -603,9 +608,7 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size)
 			/* Decrease the count, but don't shutdown the
 			 * the service
 			 */
-			lock_kernel();
 			nfsd_serv->sv_nrthreads--;
-			unlock_kernel();
 		}
 		return err < 0 ? err : 0;
 	}
@@ -614,10 +617,8 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size)
 		int len = 0;
 		if (!toclose)
 			return -ENOMEM;
-		lock_kernel();
 		if (nfsd_serv)
 			len = svc_sock_names(buf, nfsd_serv, toclose);
-		unlock_kernel();
 		if (len >= 0)
 			lockd_down();
 		kfree(toclose);
@@ -655,7 +656,6 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size)
 		if (sscanf(&buf[1], "%15s %4d", transport, &port) == 2) {
 			if (port == 0)
 				return -EINVAL;
-			lock_kernel();
 			if (nfsd_serv) {
 				xprt = svc_find_xprt(nfsd_serv, transport,
 						     AF_UNSPEC, port);
@@ -666,13 +666,22 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size)
 				} else
 					err = -ENOTCONN;
 			}
-			unlock_kernel();
 			return err < 0 ? err : 0;
 		}
 	}
 	return -EINVAL;
 }
 
+static ssize_t write_ports(struct file *file, char *buf, size_t size)
+{
+	ssize_t rv;
+	mutex_lock(&nfsd_mutex);
+	rv = __write_ports(file, buf, size);
+	mutex_unlock(&nfsd_mutex);
+	return rv;
+}
+
+
 int nfsd_max_blksize;
 
 static ssize_t write_maxblksize(struct file *file, char *buf, size_t size)
@@ -691,13 +700,13 @@ static ssize_t write_maxblksize(struct file *file, char *buf, size_t size)
 		if (bsize > NFSSVC_MAXBLKSIZE)
 			bsize = NFSSVC_MAXBLKSIZE;
 		bsize &= ~(1024-1);
-		lock_kernel();
+		mutex_lock(&nfsd_mutex);
 		if (nfsd_serv && nfsd_serv->sv_nrthreads) {
-			unlock_kernel();
+			mutex_unlock(&nfsd_mutex);
 			return -EBUSY;
 		}
 		nfsd_max_blksize = bsize;
-		unlock_kernel();
+		mutex_unlock(&nfsd_mutex);
 	}
 	return sprintf(buf, "%d\n", nfsd_max_blksize);
 }
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 941041f4b136..512bd04c6dda 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -53,11 +53,27 @@
 extern struct svc_program	nfsd_program;
 static void			nfsd(struct svc_rqst *rqstp);
 struct timeval			nfssvc_boot;
-       struct svc_serv 		*nfsd_serv;
 static atomic_t			nfsd_busy;
 static unsigned long		nfsd_last_call;
 static DEFINE_SPINLOCK(nfsd_call_lock);
 
+/*
+ * nfsd_mutex protects nfsd_serv -- both the pointer itself and the members
+ * of the svc_serv struct. In particular, ->sv_nrthreads but also to some
+ * extent ->sv_temp_socks and ->sv_permsocks. It also protects nfsdstats.th_cnt
+ *
+ * If (out side the lock) nfsd_serv is non-NULL, then it must point to a
+ * properly initialised 'struct svc_serv' with ->sv_nrthreads > 0. That number
+ * of nfsd threads must exist and each must listed in ->sp_all_threads in each
+ * entry of ->sv_pools[].
+ *
+ * Transitions of the thread count between zero and non-zero are of particular
+ * interest since the svc_serv needs to be created and initialized at that
+ * point, or freed.
+ */
+DEFINE_MUTEX(nfsd_mutex);
+struct svc_serv 		*nfsd_serv;
+
 #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
 static struct svc_stat	nfsd_acl_svcstats;
 static struct svc_version *	nfsd_acl_version[] = {
@@ -190,13 +206,14 @@ void nfsd_reset_versions(void)
 	}
 }
 
+
 int nfsd_create_serv(void)
 {
 	int err = 0;
-	lock_kernel();
+
+	WARN_ON(!mutex_is_locked(&nfsd_mutex));
 	if (nfsd_serv) {
 		svc_get(nfsd_serv);
-		unlock_kernel();
 		return 0;
 	}
 	if (nfsd_max_blksize == 0) {
@@ -223,7 +240,7 @@ int nfsd_create_serv(void)
 				      nfsd, SIG_NOCLEAN, THIS_MODULE);
 	if (nfsd_serv == NULL)
 		err = -ENOMEM;
-	unlock_kernel();
+
 	do_gettimeofday(&nfssvc_boot);		/* record boot time */
 	return err;
 }
@@ -282,6 +299,8 @@ int nfsd_set_nrthreads(int n, int *nthreads)
 	int tot = 0;
 	int err = 0;
 
+	WARN_ON(!mutex_is_locked(&nfsd_mutex));
+
 	if (nfsd_serv == NULL || n <= 0)
 		return 0;
 
@@ -316,7 +335,6 @@ int nfsd_set_nrthreads(int n, int *nthreads)
 		nthreads[0] = 1;
 
 	/* apply the new numbers */
-	lock_kernel();
 	svc_get(nfsd_serv);
 	for (i = 0; i < n; i++) {
 		err = svc_set_num_threads(nfsd_serv, &nfsd_serv->sv_pools[i],
@@ -325,7 +343,6 @@ int nfsd_set_nrthreads(int n, int *nthreads)
 			break;
 	}
 	svc_destroy(nfsd_serv);
-	unlock_kernel();
 
 	return err;
 }
@@ -334,8 +351,8 @@ int
 nfsd_svc(unsigned short port, int nrservs)
 {
 	int	error;
-	
-	lock_kernel();
+
+	mutex_lock(&nfsd_mutex);
 	dprintk("nfsd: creating service\n");
 	error = -EINVAL;
 	if (nrservs <= 0)
@@ -363,7 +380,7 @@ nfsd_svc(unsigned short port, int nrservs)
  failure:
 	svc_destroy(nfsd_serv);		/* Release server */
  out:
-	unlock_kernel();
+	mutex_unlock(&nfsd_mutex);
 	return error;
 }
 
@@ -399,7 +416,7 @@ nfsd(struct svc_rqst *rqstp)
 	sigset_t shutdown_mask, allowed_mask;
 
 	/* Lock module and set up kernel thread */
-	lock_kernel();
+	mutex_lock(&nfsd_mutex);
 	daemonize("nfsd");
 
 	/* After daemonize() this kernel thread shares current->fs
@@ -417,11 +434,13 @@ nfsd(struct svc_rqst *rqstp)
 	siginitsetinv(&shutdown_mask, SHUTDOWN_SIGS);
 	siginitsetinv(&allowed_mask, ALLOWED_SIGS);
 
+
 	nfsdstats.th_cnt++;
 
 	rqstp->rq_task = current;
 
-	unlock_kernel();
+	mutex_unlock(&nfsd_mutex);
+
 
 	/*
 	 * We want less throttling in balance_dirty_pages() so that nfs to
@@ -477,7 +496,7 @@ nfsd(struct svc_rqst *rqstp)
 	/* Clear signals before calling svc_exit_thread() */
 	flush_signals(current);
 
-	lock_kernel();
+	mutex_lock(&nfsd_mutex);
 
 	nfsdstats.th_cnt --;
 
@@ -486,7 +505,7 @@ out:
 	svc_exit_thread(rqstp);
 
 	/* Release module */
-	unlock_kernel();
+	mutex_unlock(&nfsd_mutex);
 	module_put_and_exit(0);
 }
 
diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index 41d30c9c9de6..88d85b964429 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -54,6 +54,7 @@ typedef int (*nfsd_dirop_t)(struct inode *, struct dentry *, int, int);
 extern struct svc_program	nfsd_program;
 extern struct svc_version	nfsd_version2, nfsd_version3,
 				nfsd_version4;
+extern struct mutex		nfsd_mutex;
 extern struct svc_serv		*nfsd_serv;
 
 extern struct seq_operations nfs_exports_op;
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 01c7e311b904..7bffaff2a3ab 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -461,7 +461,8 @@ svc_create_pooled(struct svc_program *prog, unsigned int bufsize,
 EXPORT_SYMBOL(svc_create_pooled);
 
 /*
- * Destroy an RPC service.  Should be called with the BKL held
+ * Destroy an RPC service. Should be called with appropriate locking to
+ * protect the sv_nrthreads, sv_permsocks and sv_tempsocks.
  */
 void
 svc_destroy(struct svc_serv *serv)
@@ -578,9 +579,10 @@ out_enomem:
 EXPORT_SYMBOL(svc_prepare_thread);
 
 /*
- * Create a thread in the given pool.  Caller must hold BKL.
- * On a NUMA or SMP machine, with a multi-pool serv, the thread
- * will be restricted to run on the cpus belonging to the pool.
+ * Create a thread in the given pool.  Caller must hold BKL or another lock to
+ * serialize access to the svc_serv struct. On a NUMA or SMP machine, with a
+ * multi-pool serv, the thread will be restricted to run on the cpus belonging
+ * to the pool.
  */
 static int
 __svc_create_thread(svc_thread_fn func, struct svc_serv *serv,
@@ -674,7 +676,7 @@ found_pool:
  * of threads the given number.  If `pool' is non-NULL, applies
  * only to threads in that pool, otherwise round-robins between
  * all pools.  Must be called with a svc_get() reference and
- * the BKL held.
+ * the BKL or another lock to protect access to svc_serv fields.
  *
  * Destroying threads relies on the service threads filling in
  * rqstp->rq_task, which only the nfs ones do.  Assumes the serv
@@ -722,7 +724,8 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 EXPORT_SYMBOL(svc_set_num_threads);
 
 /*
- * Called from a server thread as it's exiting.  Caller must hold BKL.
+ * Called from a server thread as it's exiting. Caller must hold the BKL or
+ * the "service mutex", whichever is appropriate for the service.
  */
 void
 svc_exit_thread(struct svc_rqst *rqstp)
-- 
cgit v1.2.3


From 9867d76ca16b3f455f9ca83861f4ce5c94a25928 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Tue, 10 Jun 2008 08:40:38 -0400
Subject: knfsd: convert knfsd to kthread API

This patch is rather large, but I couldn't figure out a way to break it
up that would remain bisectable. It does several things:

- change svc_thread_fn typedef to better match what kthread_create expects
- change svc_pool_map_set_cpumask to be more kthread friendly. Make it
  take a task arg and and get rid of the "oldmask"
- have svc_set_num_threads call kthread_create directly
- eliminate __svc_create_thread

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfssvc.c           |  45 ++++++++++++--------
 include/linux/sunrpc/svc.h |   2 +-
 net/sunrpc/svc.c           | 100 +++++++++++++++------------------------------
 3 files changed, 64 insertions(+), 83 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 6339cb70a08d..9e2156813710 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -21,6 +21,7 @@
 #include <linux/smp_lock.h>
 #include <linux/freezer.h>
 #include <linux/fs_struct.h>
+#include <linux/kthread.h>
 
 #include <linux/sunrpc/types.h>
 #include <linux/sunrpc/stats.h>
@@ -46,7 +47,7 @@
 #define SHUTDOWN_SIGS	(sigmask(SIGKILL) | sigmask(SIGHUP) | sigmask(SIGINT) | sigmask(SIGQUIT))
 
 extern struct svc_program	nfsd_program;
-static void			nfsd(struct svc_rqst *rqstp);
+static int			nfsd(void *vrqstp);
 struct timeval			nfssvc_boot;
 static atomic_t			nfsd_busy;
 static unsigned long		nfsd_last_call;
@@ -407,18 +408,19 @@ update_thread_usage(int busy_threads)
 /*
  * This is the NFS server kernel thread
  */
-static void
-nfsd(struct svc_rqst *rqstp)
+static int
+nfsd(void *vrqstp)
 {
+	struct svc_rqst *rqstp = (struct svc_rqst *) vrqstp;
 	struct fs_struct *fsp;
-	int		err;
 	sigset_t shutdown_mask, allowed_mask;
+	int err, preverr = 0;
+	unsigned int signo;
 
 	/* Lock module and set up kernel thread */
 	mutex_lock(&nfsd_mutex);
-	daemonize("nfsd");
 
-	/* After daemonize() this kernel thread shares current->fs
+	/* At this point, the thread shares current->fs
 	 * with the init process. We need to create files with a
 	 * umask of 0 instead of init's umask. */
 	fsp = copy_fs_struct(current->fs);
@@ -433,14 +435,18 @@ nfsd(struct svc_rqst *rqstp)
 	siginitsetinv(&shutdown_mask, SHUTDOWN_SIGS);
 	siginitsetinv(&allowed_mask, ALLOWED_SIGS);
 
+	/*
+	 * thread is spawned with all signals set to SIG_IGN, re-enable
+	 * the ones that matter
+	 */
+	for (signo = 1; signo <= _NSIG; signo++) {
+		if (!sigismember(&shutdown_mask, signo))
+			allow_signal(signo);
+	}
 
 	nfsdstats.th_cnt++;
-
-	rqstp->rq_task = current;
-
 	mutex_unlock(&nfsd_mutex);
 
-
 	/*
 	 * We want less throttling in balance_dirty_pages() so that nfs to
 	 * localhost doesn't cause nfsd to lock up due to all the client's
@@ -462,15 +468,25 @@ nfsd(struct svc_rqst *rqstp)
 		 */
 		while ((err = svc_recv(rqstp, 60*60*HZ)) == -EAGAIN)
 			;
-		if (err < 0)
+		if (err == -EINTR)
 			break;
+		else if (err < 0) {
+			if (err != preverr) {
+				printk(KERN_WARNING "%s: unexpected error "
+					"from svc_recv (%d)\n", __func__, -err);
+				preverr = err;
+			}
+			schedule_timeout_uninterruptible(HZ);
+			continue;
+		}
+
 		update_thread_usage(atomic_read(&nfsd_busy));
 		atomic_inc(&nfsd_busy);
 
 		/* Lock the export hash tables for reading. */
 		exp_readlock();
 
-		/* Process request with signals blocked.  */
+		/* Process request with signals blocked. */
 		sigprocmask(SIG_SETMASK, &allowed_mask, NULL);
 
 		svc_process(rqstp);
@@ -481,14 +497,10 @@ nfsd(struct svc_rqst *rqstp)
 		atomic_dec(&nfsd_busy);
 	}
 
-	if (err != -EINTR)
-		printk(KERN_WARNING "nfsd: terminating on error %d\n", -err);
-
 	/* Clear signals before calling svc_exit_thread() */
 	flush_signals(current);
 
 	mutex_lock(&nfsd_mutex);
-
 	nfsdstats.th_cnt --;
 
 out:
@@ -498,6 +510,7 @@ out:
 	/* Release module */
 	mutex_unlock(&nfsd_mutex);
 	module_put_and_exit(0);
+	return 0;
 }
 
 static __be32 map_new_errors(u32 vers, __be32 nfserr)
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 4b54c5fdcfd9..011d6d8100d8 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -22,7 +22,7 @@
 /*
  * This is the RPC server thread function prototype
  */
-typedef void		(*svc_thread_fn)(struct svc_rqst *);
+typedef int		(*svc_thread_fn)(void *);
 
 /*
  *
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 7bffaff2a3ab..03a9f1a9e75c 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -18,6 +18,7 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
+#include <linux/kthread.h>
 
 #include <linux/sunrpc/types.h>
 #include <linux/sunrpc/xdr.h>
@@ -291,15 +292,14 @@ svc_pool_map_put(void)
 
 
 /*
- * Set the current thread's cpus_allowed mask so that it
+ * Set the given thread's cpus_allowed mask so that it
  * will only run on cpus in the given pool.
- *
- * Returns 1 and fills in oldmask iff a cpumask was applied.
  */
-static inline int
-svc_pool_map_set_cpumask(unsigned int pidx, cpumask_t *oldmask)
+static inline void
+svc_pool_map_set_cpumask(struct task_struct *task, unsigned int pidx)
 {
 	struct svc_pool_map *m = &svc_pool_map;
+	unsigned int node = m->pool_to[pidx];
 
 	/*
 	 * The caller checks for sv_nrpools > 1, which
@@ -307,26 +307,17 @@ svc_pool_map_set_cpumask(unsigned int pidx, cpumask_t *oldmask)
 	 */
 	BUG_ON(m->count == 0);
 
-	switch (m->mode)
-	{
-	default:
-		return 0;
+	switch (m->mode) {
 	case SVC_POOL_PERCPU:
 	{
-		unsigned int cpu = m->pool_to[pidx];
-
-		*oldmask = current->cpus_allowed;
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-		return 1;
+		set_cpus_allowed_ptr(task, &cpumask_of_cpu(node));
+		break;
 	}
 	case SVC_POOL_PERNODE:
 	{
-		unsigned int node = m->pool_to[pidx];
 		node_to_cpumask_ptr(nodecpumask, node);
-
-		*oldmask = current->cpus_allowed;
-		set_cpus_allowed_ptr(current, nodecpumask);
-		return 1;
+		set_cpus_allowed_ptr(task, nodecpumask);
+		break;
 	}
 	}
 }
@@ -578,47 +569,6 @@ out_enomem:
 }
 EXPORT_SYMBOL(svc_prepare_thread);
 
-/*
- * Create a thread in the given pool.  Caller must hold BKL or another lock to
- * serialize access to the svc_serv struct. On a NUMA or SMP machine, with a
- * multi-pool serv, the thread will be restricted to run on the cpus belonging
- * to the pool.
- */
-static int
-__svc_create_thread(svc_thread_fn func, struct svc_serv *serv,
-		    struct svc_pool *pool)
-{
-	struct svc_rqst	*rqstp;
-	int		error = -ENOMEM;
-	int		have_oldmask = 0;
-	cpumask_t	uninitialized_var(oldmask);
-
-	rqstp = svc_prepare_thread(serv, pool);
-	if (IS_ERR(rqstp)) {
-		error = PTR_ERR(rqstp);
-		goto out;
-	}
-
-	if (serv->sv_nrpools > 1)
-		have_oldmask = svc_pool_map_set_cpumask(pool->sp_id, &oldmask);
-
-	error = kernel_thread((int (*)(void *)) func, rqstp, 0);
-
-	if (have_oldmask)
-		set_cpus_allowed(current, oldmask);
-
-	if (error < 0)
-		goto out_thread;
-	svc_sock_update_bufs(serv);
-	error = 0;
-out:
-	return error;
-
-out_thread:
-	svc_exit_thread(rqstp);
-	goto out;
-}
-
 /*
  * Choose a pool in which to create a new thread, for svc_set_num_threads
  */
@@ -688,7 +638,9 @@ found_pool:
 int
 svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 {
-	struct task_struct *victim;
+	struct svc_rqst	*rqstp;
+	struct task_struct *task;
+	struct svc_pool *chosen_pool;
 	int error = 0;
 	unsigned int state = serv->sv_nrthreads-1;
 
@@ -704,18 +656,34 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 	/* create new threads */
 	while (nrservs > 0) {
 		nrservs--;
+		chosen_pool = choose_pool(serv, pool, &state);
+
+		rqstp = svc_prepare_thread(serv, chosen_pool);
+		if (IS_ERR(rqstp)) {
+			error = PTR_ERR(rqstp);
+			break;
+		}
+
 		__module_get(serv->sv_module);
-		error = __svc_create_thread(serv->sv_function, serv,
-					    choose_pool(serv, pool, &state));
-		if (error < 0) {
+		task = kthread_create(serv->sv_function, rqstp, serv->sv_name);
+		if (IS_ERR(task)) {
+			error = PTR_ERR(task);
 			module_put(serv->sv_module);
+			svc_exit_thread(rqstp);
 			break;
 		}
+
+		rqstp->rq_task = task;
+		if (serv->sv_nrpools > 1)
+			svc_pool_map_set_cpumask(task, chosen_pool->sp_id);
+
+		svc_sock_update_bufs(serv);
+		wake_up_process(task);
 	}
 	/* destroy old threads */
 	while (nrservs < 0 &&
-	       (victim = choose_victim(serv, pool, &state)) != NULL) {
-		send_sig(serv->sv_kill_signal, victim, 1);
+	       (task = choose_victim(serv, pool, &state)) != NULL) {
+		send_sig(serv->sv_kill_signal, task, 1);
 		nrservs++;
 	}
 
-- 
cgit v1.2.3


From a75c5d01e4235a7dd785548ac756f248b1b40107 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Tue, 10 Jun 2008 08:40:39 -0400
Subject: sunrpc: remove sv_kill_signal field from svc_serv struct

Since we no longer make any distinction between shutdown signals with
nfsd, then it becomes easier to just standardize on a particular signal
to use to bring it down (SIGINT, in this case).

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfssvc.c           | 3 +--
 include/linux/sunrpc/svc.h | 5 ++---
 net/sunrpc/svc.c           | 5 ++---
 3 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 9e2156813710..26c81149d49a 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -236,8 +236,7 @@ int nfsd_create_serv(void)
 
 	atomic_set(&nfsd_busy, 0);
 	nfsd_serv = svc_create_pooled(&nfsd_program, nfsd_max_blksize,
-				      nfsd_last_thread, nfsd, SIGINT,
-				      THIS_MODULE);
+				      nfsd_last_thread, nfsd, THIS_MODULE);
 	if (nfsd_serv == NULL)
 		err = -ENOMEM;
 
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 011d6d8100d8..dc69068d94c7 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -80,7 +80,6 @@ struct svc_serv {
 	struct module *		sv_module;	/* optional module to count when
 						 * adding threads */
 	svc_thread_fn		sv_function;	/* main function for threads */
-	int			sv_kill_signal;	/* signal to kill threads */
 };
 
 /*
@@ -388,8 +387,8 @@ struct svc_rqst *svc_prepare_thread(struct svc_serv *serv,
 					struct svc_pool *pool);
 void		   svc_exit_thread(struct svc_rqst *);
 struct svc_serv *  svc_create_pooled(struct svc_program *, unsigned int,
-			void (*shutdown)(struct svc_serv*),
-			svc_thread_fn, int sig, struct module *);
+			void (*shutdown)(struct svc_serv*), svc_thread_fn,
+			struct module *);
 int		   svc_set_num_threads(struct svc_serv *, struct svc_pool *, int);
 void		   svc_destroy(struct svc_serv *);
 int		   svc_process(struct svc_rqst *);
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 03a9f1a9e75c..5a32cb7c4bb4 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -434,7 +434,7 @@ EXPORT_SYMBOL(svc_create);
 struct svc_serv *
 svc_create_pooled(struct svc_program *prog, unsigned int bufsize,
 		void (*shutdown)(struct svc_serv *serv),
-		  svc_thread_fn func, int sig, struct module *mod)
+		  svc_thread_fn func, struct module *mod)
 {
 	struct svc_serv *serv;
 	unsigned int npools = svc_pool_map_get();
@@ -443,7 +443,6 @@ svc_create_pooled(struct svc_program *prog, unsigned int bufsize,
 
 	if (serv != NULL) {
 		serv->sv_function = func;
-		serv->sv_kill_signal = sig;
 		serv->sv_module = mod;
 	}
 
@@ -683,7 +682,7 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs)
 	/* destroy old threads */
 	while (nrservs < 0 &&
 	       (task = choose_victim(serv, pool, &state)) != NULL) {
-		send_sig(serv->sv_kill_signal, task, 1);
+		send_sig(SIGINT, task, 1);
 		nrservs++;
 	}
 
-- 
cgit v1.2.3


From 8837abcab3d16608bd2c7fac051a839d48f2f30c Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Mon, 16 Jun 2008 13:20:29 +0200
Subject: nfsd: rename MAY_ flags

Rename nfsd_permission() specific MAY_* flags to NFSD_MAY_* to make it
clear, that these are not used outside nfsd, and to avoid name and
number space conflicts with the VFS.

[comment from hch: rename MAY_READ, MAY_WRITE and MAY_EXEC as well]

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/lockd.c           |   2 +-
 fs/nfsd/nfs2acl.c         |   7 +--
 fs/nfsd/nfs3acl.c         |   5 +-
 fs/nfsd/nfs3proc.c        |   8 ++--
 fs/nfsd/nfs4proc.c        |  23 ++++++----
 fs/nfsd/nfs4state.c       |   6 +--
 fs/nfsd/nfsfh.c           |   2 +-
 fs/nfsd/nfsproc.c         |   8 ++--
 fs/nfsd/vfs.c             | 115 ++++++++++++++++++++++++----------------------
 include/linux/nfsd/nfsd.h |  26 +++++------
 10 files changed, 105 insertions(+), 97 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c
index 9e4a568a5013..6b6225ac4926 100644
--- a/fs/nfsd/lockd.c
+++ b/fs/nfsd/lockd.c
@@ -35,7 +35,7 @@ nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp)
 	fh.fh_export = NULL;
 
 	exp_readlock();
-	nfserr = nfsd_open(rqstp, &fh, S_IFREG, MAY_LOCK, filp);
+	nfserr = nfsd_open(rqstp, &fh, S_IFREG, NFSD_MAY_LOCK, filp);
 	fh_put(&fh);
 	rqstp->rq_client = NULL;
 	exp_readunlock();
diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c
index 1c3b7654e966..4e3219e84116 100644
--- a/fs/nfsd/nfs2acl.c
+++ b/fs/nfsd/nfs2acl.c
@@ -40,7 +40,8 @@ static __be32 nfsacld_proc_getacl(struct svc_rqst * rqstp,
 	dprintk("nfsd: GETACL(2acl)   %s\n", SVCFH_fmt(&argp->fh));
 
 	fh = fh_copy(&resp->fh, &argp->fh);
-	if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP)))
+	nfserr = fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_NOP);
+	if (nfserr)
 		RETURN_STATUS(nfserr);
 
 	if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT))
@@ -107,7 +108,7 @@ static __be32 nfsacld_proc_setacl(struct svc_rqst * rqstp,
 	dprintk("nfsd: SETACL(2acl)   %s\n", SVCFH_fmt(&argp->fh));
 
 	fh = fh_copy(&resp->fh, &argp->fh);
-	nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_SATTR);
+	nfserr = fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_SATTR);
 
 	if (!nfserr) {
 		nfserr = nfserrno( nfsd_set_posix_acl(
@@ -134,7 +135,7 @@ static __be32 nfsacld_proc_getattr(struct svc_rqst * rqstp,
 	dprintk("nfsd: GETATTR  %s\n", SVCFH_fmt(&argp->fh));
 
 	fh_copy(&resp->fh, &argp->fh);
-	return fh_verify(rqstp, &resp->fh, 0, MAY_NOP);
+	return fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_NOP);
 }
 
 /*
diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c
index b647f2f872dc..9981dbb377a3 100644
--- a/fs/nfsd/nfs3acl.c
+++ b/fs/nfsd/nfs3acl.c
@@ -36,7 +36,8 @@ static __be32 nfsd3_proc_getacl(struct svc_rqst * rqstp,
 	__be32 nfserr = 0;
 
 	fh = fh_copy(&resp->fh, &argp->fh);
-	if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP)))
+	nfserr = fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_NOP);
+	if (nfserr)
 		RETURN_STATUS(nfserr);
 
 	if (argp->mask & ~(NFS_ACL|NFS_ACLCNT|NFS_DFACL|NFS_DFACLCNT))
@@ -101,7 +102,7 @@ static __be32 nfsd3_proc_setacl(struct svc_rqst * rqstp,
 	__be32 nfserr = 0;
 
 	fh = fh_copy(&resp->fh, &argp->fh);
-	nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_SATTR);
+	nfserr = fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_SATTR);
 
 	if (!nfserr) {
 		nfserr = nfserrno( nfsd_set_posix_acl(
diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index c721a1e6e9dd..4d617ea28cfc 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -63,7 +63,7 @@ nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle  *argp,
 		SVCFH_fmt(&argp->fh));
 
 	fh_copy(&resp->fh, &argp->fh);
-	nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP);
+	nfserr = fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_NOP);
 	if (nfserr)
 		RETURN_STATUS(nfserr);
 
@@ -242,7 +242,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp,
 	attr   = &argp->attrs;
 
 	/* Get the directory inode */
-	nfserr = fh_verify(rqstp, dirfhp, S_IFDIR, MAY_CREATE);
+	nfserr = fh_verify(rqstp, dirfhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (nfserr)
 		RETURN_STATUS(nfserr);
 
@@ -558,7 +558,7 @@ nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle    *argp,
 	resp->f_maxfilesize = ~(u32) 0;
 	resp->f_properties = NFS3_FSF_DEFAULT;
 
-	nfserr = fh_verify(rqstp, &argp->fh, 0, MAY_NOP);
+	nfserr = fh_verify(rqstp, &argp->fh, 0, NFSD_MAY_NOP);
 
 	/* Check special features of the file system. May request
 	 * different read/write sizes for file systems known to have
@@ -597,7 +597,7 @@ nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle      *argp,
 	resp->p_case_insensitive = 0;
 	resp->p_case_preserving = 1;
 
-	nfserr = fh_verify(rqstp, &argp->fh, 0, MAY_NOP);
+	nfserr = fh_verify(rqstp, &argp->fh, 0, NFSD_MAY_NOP);
 
 	if (nfserr == 0) {
 		struct super_block *sb = argp->fh.fh_dentry->d_inode->i_sb;
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 313484380a9b..5c3683cfd59e 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -71,11 +71,11 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs
 		return nfserr_inval;
 
 	if (open->op_share_access & NFS4_SHARE_ACCESS_READ)
-		accmode |= MAY_READ;
+		accmode |= NFSD_MAY_READ;
 	if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
-		accmode |= (MAY_WRITE | MAY_TRUNC);
+		accmode |= (NFSD_MAY_WRITE | NFSD_MAY_TRUNC);
 	if (open->op_share_deny & NFS4_SHARE_DENY_WRITE)
-		accmode |= MAY_WRITE;
+		accmode |= NFSD_MAY_WRITE;
 
 	status = fh_verify(rqstp, current_fh, S_IFREG, accmode);
 
@@ -126,7 +126,8 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o
 			&resfh.fh_handle.fh_base, resfh.fh_handle.fh_size);
 
 	if (!created)
-		status = do_open_permission(rqstp, current_fh, open, MAY_NOP);
+		status = do_open_permission(rqstp, current_fh, open,
+					    NFSD_MAY_NOP);
 
 out:
 	fh_put(&resfh);
@@ -157,7 +158,8 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_
 	open->op_truncate = (open->op_iattr.ia_valid & ATTR_SIZE) &&
 		(open->op_iattr.ia_size == 0);
 
-	status = do_open_permission(rqstp, current_fh, open, MAY_OWNER_OVERRIDE);
+	status = do_open_permission(rqstp, current_fh, open,
+				    NFSD_MAY_OWNER_OVERRIDE);
 
 	return status;
 }
@@ -186,7 +188,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		cstate->current_fh.fh_handle.fh_size = rp->rp_openfh_len;
 		memcpy(&cstate->current_fh.fh_handle.fh_base, rp->rp_openfh,
 				rp->rp_openfh_len);
-		status = fh_verify(rqstp, &cstate->current_fh, 0, MAY_NOP);
+		status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
 		if (status)
 			dprintk("nfsd4_open: replay failed"
 				" restoring previous filehandle\n");
@@ -285,7 +287,7 @@ nfsd4_putfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	cstate->current_fh.fh_handle.fh_size = putfh->pf_fhlen;
 	memcpy(&cstate->current_fh.fh_handle.fh_base, putfh->pf_fhval,
 	       putfh->pf_fhlen);
-	return fh_verify(rqstp, &cstate->current_fh, 0, MAY_NOP);
+	return fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
 }
 
 static __be32
@@ -363,7 +365,8 @@ nfsd4_create(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 
 	fh_init(&resfh, NFS4_FHSIZE);
 
-	status = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, MAY_CREATE);
+	status = fh_verify(rqstp, &cstate->current_fh, S_IFDIR,
+			   NFSD_MAY_CREATE);
 	if (status == nfserr_symlink)
 		status = nfserr_notdir;
 	if (status)
@@ -445,7 +448,7 @@ nfsd4_getattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 {
 	__be32 status;
 
-	status = fh_verify(rqstp, &cstate->current_fh, 0, MAY_NOP);
+	status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
 	if (status)
 		return status;
 
@@ -730,7 +733,7 @@ _nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	int count;
 	__be32 status;
 
-	status = fh_verify(rqstp, &cstate->current_fh, 0, MAY_NOP);
+	status = fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP);
 	if (status)
 		return status;
 
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index bf11d6879ab4..eca8aaa450f1 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1722,9 +1722,9 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 		/* Stateid was not found, this is a new OPEN */
 		int flags = 0;
 		if (open->op_share_access & NFS4_SHARE_ACCESS_READ)
-			flags |= MAY_READ;
+			flags |= NFSD_MAY_READ;
 		if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
-			flags |= MAY_WRITE;
+			flags |= NFSD_MAY_WRITE;
 		status = nfs4_new_open(rqstp, &stp, dp, current_fh, flags);
 		if (status)
 			goto out;
@@ -2610,7 +2610,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 		 return nfserr_inval;
 
 	if ((status = fh_verify(rqstp, &cstate->current_fh,
-				S_IFREG, MAY_LOCK))) {
+				S_IFREG, NFSD_MAY_LOCK))) {
 		dprintk("NFSD: nfsd4_lock: permission denied!\n");
 		return status;
 	}
diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index 100ae5641162..c7b0fdaeac96 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -279,7 +279,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access)
 	if (error)
 		goto out;
 
-	if (!(access & MAY_LOCK)) {
+	if (!(access & NFSD_MAY_LOCK)) {
 		/*
 		 * pseudoflavor restrictions are not enforced on NLM,
 		 * which clients virtually always use auth_sys for,
diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c
index b5a20c486712..0766f95d236a 100644
--- a/fs/nfsd/nfsproc.c
+++ b/fs/nfsd/nfsproc.c
@@ -65,7 +65,7 @@ nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle  *argp,
 	dprintk("nfsd: GETATTR  %s\n", SVCFH_fmt(&argp->fh));
 
 	fh_copy(&resp->fh, &argp->fh);
-	nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP);
+	nfserr = fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_NOP);
 	return nfsd_return_attrs(nfserr, resp);
 }
 
@@ -215,11 +215,11 @@ nfsd_proc_create(struct svc_rqst *rqstp, struct nfsd_createargs *argp,
 		SVCFH_fmt(dirfhp), argp->len, argp->name);
 
 	/* First verify the parent file handle */
-	nfserr = fh_verify(rqstp, dirfhp, S_IFDIR, MAY_EXEC);
+	nfserr = fh_verify(rqstp, dirfhp, S_IFDIR, NFSD_MAY_EXEC);
 	if (nfserr)
 		goto done; /* must fh_put dirfhp even on error */
 
-	/* Check for MAY_WRITE in nfsd_create if necessary */
+	/* Check for NFSD_MAY_WRITE in nfsd_create if necessary */
 
 	nfserr = nfserr_acces;
 	if (!argp->len)
@@ -281,7 +281,7 @@ nfsd_proc_create(struct svc_rqst *rqstp, struct nfsd_createargs *argp,
 					nfserr = nfsd_permission(rqstp,
 								 newfhp->fh_export,
 								 newfhp->fh_dentry,
-								 MAY_WRITE|MAY_LOCAL_ACCESS);
+								 NFSD_MAY_WRITE|NFSD_MAY_LOCAL_ACCESS);
 					if (nfserr && nfserr != nfserr_rofs)
 						goto out_unlock;
 				}
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index a3a291f771f4..5e05ddda4560 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -144,7 +144,7 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name);
 
 	/* Obtain dentry and export. */
-	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_EXEC);
+	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
 	if (err)
 		return err;
 
@@ -262,14 +262,14 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 {
 	struct dentry	*dentry;
 	struct inode	*inode;
-	int		accmode = MAY_SATTR;
+	int		accmode = NFSD_MAY_SATTR;
 	int		ftype = 0;
 	__be32		err;
 	int		host_err;
 	int		size_change = 0;
 
 	if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE))
-		accmode |= MAY_WRITE|MAY_OWNER_OVERRIDE;
+		accmode |= NFSD_MAY_WRITE|NFSD_MAY_OWNER_OVERRIDE;
 	if (iap->ia_valid & ATTR_SIZE)
 		ftype = S_IFREG;
 
@@ -331,7 +331,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 	 */
 	if (iap->ia_valid & ATTR_SIZE) {
 		if (iap->ia_size < inode->i_size) {
-			err = nfsd_permission(rqstp, fhp->fh_export, dentry, MAY_TRUNC|MAY_OWNER_OVERRIDE);
+			err = nfsd_permission(rqstp, fhp->fh_export, dentry,
+					NFSD_MAY_TRUNC|NFSD_MAY_OWNER_OVERRIDE);
 			if (err)
 				goto out;
 		}
@@ -462,7 +463,7 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	unsigned int flags = 0;
 
 	/* Get inode */
-	error = fh_verify(rqstp, fhp, 0 /* S_IFREG */, MAY_SATTR);
+	error = fh_verify(rqstp, fhp, 0 /* S_IFREG */, NFSD_MAY_SATTR);
 	if (error)
 		return error;
 
@@ -563,20 +564,20 @@ struct accessmap {
 	int		how;
 };
 static struct accessmap	nfs3_regaccess[] = {
-    {	NFS3_ACCESS_READ,	MAY_READ			},
-    {	NFS3_ACCESS_EXECUTE,	MAY_EXEC			},
-    {	NFS3_ACCESS_MODIFY,	MAY_WRITE|MAY_TRUNC		},
-    {	NFS3_ACCESS_EXTEND,	MAY_WRITE			},
+    {	NFS3_ACCESS_READ,	NFSD_MAY_READ			},
+    {	NFS3_ACCESS_EXECUTE,	NFSD_MAY_EXEC			},
+    {	NFS3_ACCESS_MODIFY,	NFSD_MAY_WRITE|NFSD_MAY_TRUNC	},
+    {	NFS3_ACCESS_EXTEND,	NFSD_MAY_WRITE			},
 
     {	0,			0				}
 };
 
 static struct accessmap	nfs3_diraccess[] = {
-    {	NFS3_ACCESS_READ,	MAY_READ			},
-    {	NFS3_ACCESS_LOOKUP,	MAY_EXEC			},
-    {	NFS3_ACCESS_MODIFY,	MAY_EXEC|MAY_WRITE|MAY_TRUNC	},
-    {	NFS3_ACCESS_EXTEND,	MAY_EXEC|MAY_WRITE		},
-    {	NFS3_ACCESS_DELETE,	MAY_REMOVE			},
+    {	NFS3_ACCESS_READ,	NFSD_MAY_READ			},
+    {	NFS3_ACCESS_LOOKUP,	NFSD_MAY_EXEC			},
+    {	NFS3_ACCESS_MODIFY,	NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC},
+    {	NFS3_ACCESS_EXTEND,	NFSD_MAY_EXEC|NFSD_MAY_WRITE	},
+    {	NFS3_ACCESS_DELETE,	NFSD_MAY_REMOVE			},
 
     {	0,			0				}
 };
@@ -589,10 +590,10 @@ static struct accessmap	nfs3_anyaccess[] = {
 	 * mainly at mode bits, and we make sure to ignore read-only
 	 * filesystem checks
 	 */
-    {	NFS3_ACCESS_READ,	MAY_READ			},
-    {	NFS3_ACCESS_EXECUTE,	MAY_EXEC			},
-    {	NFS3_ACCESS_MODIFY,	MAY_WRITE|MAY_LOCAL_ACCESS	},
-    {	NFS3_ACCESS_EXTEND,	MAY_WRITE|MAY_LOCAL_ACCESS	},
+    {	NFS3_ACCESS_READ,	NFSD_MAY_READ			},
+    {	NFS3_ACCESS_EXECUTE,	NFSD_MAY_EXEC			},
+    {	NFS3_ACCESS_MODIFY,	NFSD_MAY_WRITE|NFSD_MAY_LOCAL_ACCESS	},
+    {	NFS3_ACCESS_EXTEND,	NFSD_MAY_WRITE|NFSD_MAY_LOCAL_ACCESS	},
 
     {	0,			0				}
 };
@@ -606,7 +607,7 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor
 	u32			query, result = 0, sresult = 0;
 	__be32			error;
 
-	error = fh_verify(rqstp, fhp, 0, MAY_NOP);
+	error = fh_verify(rqstp, fhp, 0, NFSD_MAY_NOP);
 	if (error)
 		goto out;
 
@@ -678,7 +679,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	 * and (hopefully) checked permission - so allow OWNER_OVERRIDE
 	 * in case a chmod has now revoked permission.
 	 */
-	err = fh_verify(rqstp, fhp, type, access | MAY_OWNER_OVERRIDE);
+	err = fh_verify(rqstp, fhp, type, access | NFSD_MAY_OWNER_OVERRIDE);
 	if (err)
 		goto out;
 
@@ -689,7 +690,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	 * or any access when mandatory locking enabled
 	 */
 	err = nfserr_perm;
-	if (IS_APPEND(inode) && (access & MAY_WRITE))
+	if (IS_APPEND(inode) && (access & NFSD_MAY_WRITE))
 		goto out;
 	/*
 	 * We must ignore files (but only files) which might have mandatory
@@ -706,14 +707,14 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	 * Check to see if there are any leases on this file.
 	 * This may block while leases are broken.
 	 */
-	host_err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0));
+	host_err = break_lease(inode, O_NONBLOCK | ((access & NFSD_MAY_WRITE) ? FMODE_WRITE : 0));
 	if (host_err == -EWOULDBLOCK)
 		host_err = -ETIMEDOUT;
 	if (host_err) /* NOMEM or WOULDBLOCK */
 		goto out_nfserr;
 
-	if (access & MAY_WRITE) {
-		if (access & MAY_READ)
+	if (access & NFSD_MAY_WRITE) {
+		if (access & NFSD_MAY_READ)
 			flags = O_RDWR|O_LARGEFILE;
 		else
 			flags = O_WRONLY|O_LARGEFILE;
@@ -1069,12 +1070,12 @@ nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 
 	if (file) {
 		err = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry,
-				MAY_READ|MAY_OWNER_OVERRIDE);
+				NFSD_MAY_READ|NFSD_MAY_OWNER_OVERRIDE);
 		if (err)
 			goto out;
 		err = nfsd_vfs_read(rqstp, fhp, file, offset, vec, vlen, count);
 	} else {
-		err = nfsd_open(rqstp, fhp, S_IFREG, MAY_READ, &file);
+		err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_READ, &file);
 		if (err)
 			goto out;
 		err = nfsd_vfs_read(rqstp, fhp, file, offset, vec, vlen, count);
@@ -1098,13 +1099,13 @@ nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 
 	if (file) {
 		err = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry,
-				MAY_WRITE|MAY_OWNER_OVERRIDE);
+				NFSD_MAY_WRITE|NFSD_MAY_OWNER_OVERRIDE);
 		if (err)
 			goto out;
 		err = nfsd_vfs_write(rqstp, fhp, file, offset, vec, vlen, cnt,
 				stablep);
 	} else {
-		err = nfsd_open(rqstp, fhp, S_IFREG, MAY_WRITE, &file);
+		err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_WRITE, &file);
 		if (err)
 			goto out;
 
@@ -1136,7 +1137,8 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if ((u64)count > ~(u64)offset)
 		return nfserr_inval;
 
-	if ((err = nfsd_open(rqstp, fhp, S_IFREG, MAY_WRITE, &file)) != 0)
+	err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_WRITE, &file);
+	if (err)
 		return err;
 	if (EX_ISSYNC(fhp->fh_export)) {
 		if (file->f_op && file->f_op->fsync) {
@@ -1197,7 +1199,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (isdotent(fname, flen))
 		goto out;
 
-	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE);
+	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
 
@@ -1334,7 +1336,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		goto out;
 	if (!(iap->ia_valid & ATTR_MODE))
 		iap->ia_mode = 0;
-	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE);
+	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
 
@@ -1471,7 +1473,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp)
 	__be32		err;
 	int		host_err;
 
-	err = fh_verify(rqstp, fhp, S_IFLNK, MAY_NOP);
+	err = fh_verify(rqstp, fhp, S_IFLNK, NFSD_MAY_NOP);
 	if (err)
 		goto out;
 
@@ -1526,7 +1528,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (isdotent(fname, flen))
 		goto out;
 
-	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE);
+	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
 	fh_lock(fhp);
@@ -1591,10 +1593,10 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
 	__be32		err;
 	int		host_err;
 
-	err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_CREATE);
+	err = fh_verify(rqstp, ffhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
-	err = fh_verify(rqstp, tfhp, -S_IFDIR, MAY_NOP);
+	err = fh_verify(rqstp, tfhp, -S_IFDIR, NFSD_MAY_NOP);
 	if (err)
 		goto out;
 
@@ -1661,10 +1663,10 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
 	__be32		err;
 	int		host_err;
 
-	err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_REMOVE);
+	err = fh_verify(rqstp, ffhp, S_IFDIR, NFSD_MAY_REMOVE);
 	if (err)
 		goto out;
-	err = fh_verify(rqstp, tfhp, S_IFDIR, MAY_CREATE);
+	err = fh_verify(rqstp, tfhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
 
@@ -1768,7 +1770,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 	err = nfserr_acces;
 	if (!flen || isdotent(fname, flen))
 		goto out;
-	err = fh_verify(rqstp, fhp, S_IFDIR, MAY_REMOVE);
+	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_REMOVE);
 	if (err)
 		goto out;
 
@@ -1834,7 +1836,7 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp,
 	struct file	*file;
 	loff_t		offset = *offsetp;
 
-	err = nfsd_open(rqstp, fhp, S_IFDIR, MAY_READ, &file);
+	err = nfsd_open(rqstp, fhp, S_IFDIR, NFSD_MAY_READ, &file);
 	if (err)
 		goto out;
 
@@ -1875,7 +1877,7 @@ out:
 __be32
 nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat)
 {
-	__be32 err = fh_verify(rqstp, fhp, 0, MAY_NOP);
+	__be32 err = fh_verify(rqstp, fhp, 0, NFSD_MAY_NOP);
 	if (!err && vfs_statfs(fhp->fh_dentry,stat))
 		err = nfserr_io;
 	return err;
@@ -1896,18 +1898,18 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 	struct inode	*inode = dentry->d_inode;
 	int		err;
 
-	if (acc == MAY_NOP)
+	if (acc == NFSD_MAY_NOP)
 		return 0;
 #if 0
 	dprintk("nfsd: permission 0x%x%s%s%s%s%s%s%s mode 0%o%s%s%s\n",
 		acc,
-		(acc & MAY_READ)?	" read"  : "",
-		(acc & MAY_WRITE)?	" write" : "",
-		(acc & MAY_EXEC)?	" exec"  : "",
-		(acc & MAY_SATTR)?	" sattr" : "",
-		(acc & MAY_TRUNC)?	" trunc" : "",
-		(acc & MAY_LOCK)?	" lock"  : "",
-		(acc & MAY_OWNER_OVERRIDE)? " owneroverride" : "",
+		(acc & NFSD_MAY_READ)?	" read"  : "",
+		(acc & NFSD_MAY_WRITE)?	" write" : "",
+		(acc & NFSD_MAY_EXEC)?	" exec"  : "",
+		(acc & NFSD_MAY_SATTR)?	" sattr" : "",
+		(acc & NFSD_MAY_TRUNC)?	" trunc" : "",
+		(acc & NFSD_MAY_LOCK)?	" lock"  : "",
+		(acc & NFSD_MAY_OWNER_OVERRIDE)? " owneroverride" : "",
 		inode->i_mode,
 		IS_IMMUTABLE(inode)?	" immut" : "",
 		IS_APPEND(inode)?	" append" : "",
@@ -1920,18 +1922,18 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 	 * system.  But if it is IRIX doing check on write-access for a 
 	 * device special file, we ignore rofs.
 	 */
-	if (!(acc & MAY_LOCAL_ACCESS))
-		if (acc & (MAY_WRITE | MAY_SATTR | MAY_TRUNC)) {
+	if (!(acc & NFSD_MAY_LOCAL_ACCESS))
+		if (acc & (NFSD_MAY_WRITE | NFSD_MAY_SATTR | NFSD_MAY_TRUNC)) {
 			if (exp_rdonly(rqstp, exp) ||
 			    __mnt_is_readonly(exp->ex_path.mnt))
 				return nfserr_rofs;
-			if (/* (acc & MAY_WRITE) && */ IS_IMMUTABLE(inode))
+			if (/* (acc & NFSD_MAY_WRITE) && */ IS_IMMUTABLE(inode))
 				return nfserr_perm;
 		}
-	if ((acc & MAY_TRUNC) && IS_APPEND(inode))
+	if ((acc & NFSD_MAY_TRUNC) && IS_APPEND(inode))
 		return nfserr_perm;
 
-	if (acc & MAY_LOCK) {
+	if (acc & NFSD_MAY_LOCK) {
 		/* If we cannot rely on authentication in NLM requests,
 		 * just allow locks, otherwise require read permission, or
 		 * ownership
@@ -1939,7 +1941,7 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 		if (exp->ex_flags & NFSEXP_NOAUTHNLM)
 			return 0;
 		else
-			acc = MAY_READ | MAY_OWNER_OVERRIDE;
+			acc = NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE;
 	}
 	/*
 	 * The file owner always gets access permission for accesses that
@@ -1955,15 +1957,16 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 	 * We must trust the client to do permission checking - using "ACCESS"
 	 * with NFSv3.
 	 */
-	if ((acc & MAY_OWNER_OVERRIDE) &&
+	if ((acc & NFSD_MAY_OWNER_OVERRIDE) &&
 	    inode->i_uid == current->fsuid)
 		return 0;
 
+	/* This assumes  NFSD_MAY_{READ,WRITE,EXEC} == MAY_{READ,WRITE,EXEC} */
 	err = permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC), NULL);
 
 	/* Allow read access to binaries even when mode 111 */
 	if (err == -EACCES && S_ISREG(inode->i_mode) &&
-	    acc == (MAY_READ | MAY_OWNER_OVERRIDE))
+	    acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE))
 		err = permission(inode, MAY_EXEC, NULL);
 
 	return err? nfserrno(err) : 0;
diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index 88d85b964429..a2861d95ecc3 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -28,20 +28,20 @@
 #define NFSD_SUPPORTED_MINOR_VERSION	0
 
 /*
- * Special flags for nfsd_permission. These must be different from MAY_READ,
- * MAY_WRITE, and MAY_EXEC.
+ * Flags for nfsd_permission
  */
-#define MAY_NOP			0
-#define MAY_SATTR		8
-#define MAY_TRUNC		16
-#define MAY_LOCK		32
-#define MAY_OWNER_OVERRIDE	64
-#define	MAY_LOCAL_ACCESS	128 /* IRIX doing local access check on device special file*/
-#if (MAY_SATTR | MAY_TRUNC | MAY_LOCK | MAY_OWNER_OVERRIDE | MAY_LOCAL_ACCESS) & (MAY_READ | MAY_WRITE | MAY_EXEC)
-# error "please use a different value for MAY_SATTR or MAY_TRUNC or MAY_LOCK or MAY_LOCAL_ACCESS or MAY_OWNER_OVERRIDE."
-#endif
-#define MAY_CREATE		(MAY_EXEC|MAY_WRITE)
-#define MAY_REMOVE		(MAY_EXEC|MAY_WRITE|MAY_TRUNC)
+#define NFSD_MAY_NOP		0
+#define NFSD_MAY_EXEC		1 /* == MAY_EXEC */
+#define NFSD_MAY_WRITE		2 /* == MAY_WRITE */
+#define NFSD_MAY_READ		4 /* == MAY_READ */
+#define NFSD_MAY_SATTR		8
+#define NFSD_MAY_TRUNC		16
+#define NFSD_MAY_LOCK		32
+#define NFSD_MAY_OWNER_OVERRIDE	64
+#define NFSD_MAY_LOCAL_ACCESS	128 /* IRIX doing local access check on device special file*/
+
+#define NFSD_MAY_CREATE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE)
+#define NFSD_MAY_REMOVE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC)
 
 /*
  * Callback function for readdir
-- 
cgit v1.2.3


From d00953a53e9a2edbe005c1e596f1e96a8a293401 Mon Sep 17 00:00:00 2001
From: Kevin Coffman <kwc@citi.umich.edu>
Date: Wed, 30 Apr 2008 12:45:53 -0400
Subject: gss_krb5: create a define for token header size and clean up ptr
 location

cleanup:
Document token header size with a #define instead of open-coding it.

Don't needlessly increment "ptr" past the beginning of the header
which makes the values passed to functions more understandable and
eliminates the need for extra "krb5_hdr" pointer.

Clean up some intersecting  white-space issues flagged by checkpatch.pl.

This leaves the checksum length hard-coded at 8 for DES.  A later patch
cleans that up.

Signed-off-by: Kevin Coffman <kwc@citi.umich.edu>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/sunrpc/gss_krb5.h       |  3 +++
 net/sunrpc/auth_gss/gss_krb5_seal.c   | 26 +++++++++---------
 net/sunrpc/auth_gss/gss_krb5_unseal.c | 16 +++++------
 net/sunrpc/auth_gss/gss_krb5_wrap.c   | 50 +++++++++++++++++------------------
 4 files changed, 49 insertions(+), 46 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h
index a10f1fb0bf7c..e7bbdba474d5 100644
--- a/include/linux/sunrpc/gss_krb5.h
+++ b/include/linux/sunrpc/gss_krb5.h
@@ -51,6 +51,9 @@ struct krb5_ctx {
 
 extern spinlock_t krb5_seq_lock;
 
+/* The length of the Kerberos GSS token header */
+#define GSS_KRB5_TOK_HDR_LEN	(16)
+
 #define KG_TOK_MIC_MSG    0x0101
 #define KG_TOK_WRAP_MSG   0x0201
 
diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c
index 5f1d36dfbcf7..b8f42ef7178e 100644
--- a/net/sunrpc/auth_gss/gss_krb5_seal.c
+++ b/net/sunrpc/auth_gss/gss_krb5_seal.c
@@ -78,7 +78,7 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text,
 	struct krb5_ctx		*ctx = gss_ctx->internal_ctx_id;
 	char			cksumdata[16];
 	struct xdr_netobj	md5cksum = {.len = 0, .data = cksumdata};
-	unsigned char		*ptr, *krb5_hdr, *msg_start;
+	unsigned char		*ptr, *msg_start;
 	s32			now;
 	u32			seq_send;
 
@@ -87,36 +87,36 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text,
 
 	now = get_seconds();
 
-	token->len = g_token_size(&ctx->mech_used, 24);
+	token->len = g_token_size(&ctx->mech_used, GSS_KRB5_TOK_HDR_LEN + 8);
 
 	ptr = token->data;
-	g_make_token_header(&ctx->mech_used, 24, &ptr);
+	g_make_token_header(&ctx->mech_used, GSS_KRB5_TOK_HDR_LEN + 8, &ptr);
 
-	*ptr++ = (unsigned char) ((KG_TOK_MIC_MSG>>8)&0xff);
-	*ptr++ = (unsigned char) (KG_TOK_MIC_MSG&0xff);
+	/* ptr now at header described in rfc 1964, section 1.2.1: */
+	ptr[0] = (unsigned char) ((KG_TOK_MIC_MSG >> 8) & 0xff);
+	ptr[1] = (unsigned char) (KG_TOK_MIC_MSG & 0xff);
 
-	/* ptr now at byte 2 of header described in rfc 1964, section 1.2.1: */
-	krb5_hdr = ptr - 2;
-	msg_start = krb5_hdr + 24;
+	msg_start = ptr + GSS_KRB5_TOK_HDR_LEN + 8;
 
-	*(__be16 *)(krb5_hdr + 2) = htons(SGN_ALG_DES_MAC_MD5);
-	memset(krb5_hdr + 4, 0xff, 4);
+	*(__be16 *)(ptr + 2) = htons(SGN_ALG_DES_MAC_MD5);
+	memset(ptr + 4, 0xff, 4);
 
-	if (make_checksum("md5", krb5_hdr, 8, text, 0, &md5cksum))
+	if (make_checksum("md5", ptr, 8, text, 0, &md5cksum))
 		return GSS_S_FAILURE;
 
 	if (krb5_encrypt(ctx->seq, NULL, md5cksum.data,
 			  md5cksum.data, md5cksum.len))
 		return GSS_S_FAILURE;
 
-	memcpy(krb5_hdr + 16, md5cksum.data + md5cksum.len - 8, 8);
+	memcpy(ptr + GSS_KRB5_TOK_HDR_LEN, md5cksum.data + md5cksum.len - 8, 8);
 
 	spin_lock(&krb5_seq_lock);
 	seq_send = ctx->seq_send++;
 	spin_unlock(&krb5_seq_lock);
 
 	if (krb5_make_seq_num(ctx->seq, ctx->initiate ? 0 : 0xff,
-			      seq_send, krb5_hdr + 16, krb5_hdr + 8))
+			      seq_send, ptr + GSS_KRB5_TOK_HDR_LEN,
+			      ptr + 8))
 		return GSS_S_FAILURE;
 
 	return (ctx->endtime < now) ? GSS_S_CONTEXT_EXPIRED : GSS_S_COMPLETE;
diff --git a/net/sunrpc/auth_gss/gss_krb5_unseal.c b/net/sunrpc/auth_gss/gss_krb5_unseal.c
index d91a5d004803..066ec73c84d6 100644
--- a/net/sunrpc/auth_gss/gss_krb5_unseal.c
+++ b/net/sunrpc/auth_gss/gss_krb5_unseal.c
@@ -92,30 +92,30 @@ gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
 					read_token->len))
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if ((*ptr++ != ((KG_TOK_MIC_MSG>>8)&0xff)) ||
-	    (*ptr++ != ( KG_TOK_MIC_MSG    &0xff))   )
+	if ((ptr[0] != ((KG_TOK_MIC_MSG >> 8) & 0xff)) ||
+	    (ptr[1] !=  (KG_TOK_MIC_MSG & 0xff)))
 		return GSS_S_DEFECTIVE_TOKEN;
 
 	/* XXX sanity-check bodysize?? */
 
-	signalg = ptr[0] + (ptr[1] << 8);
+	signalg = ptr[2] + (ptr[3] << 8);
 	if (signalg != SGN_ALG_DES_MAC_MD5)
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	sealalg = ptr[2] + (ptr[3] << 8);
+	sealalg = ptr[4] + (ptr[5] << 8);
 	if (sealalg != SEAL_ALG_NONE)
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if ((ptr[4] != 0xff) || (ptr[5] != 0xff))
+	if ((ptr[6] != 0xff) || (ptr[7] != 0xff))
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if (make_checksum("md5", ptr - 2, 8, message_buffer, 0, &md5cksum))
+	if (make_checksum("md5", ptr, 8, message_buffer, 0, &md5cksum))
 		return GSS_S_FAILURE;
 
 	if (krb5_encrypt(ctx->seq, NULL, md5cksum.data, md5cksum.data, 16))
 		return GSS_S_FAILURE;
 
-	if (memcmp(md5cksum.data + 8, ptr + 14, 8))
+	if (memcmp(md5cksum.data + 8, ptr + GSS_KRB5_TOK_HDR_LEN, 8))
 		return GSS_S_BAD_SIG;
 
 	/* it got through unscathed.  Make sure the context is unexpired */
@@ -127,7 +127,7 @@ gss_verify_mic_kerberos(struct gss_ctx *gss_ctx,
 
 	/* do sequencing checks */
 
-	if (krb5_get_seq_num(ctx->seq, ptr + 14, ptr + 6, &direction, &seqnum))
+	if (krb5_get_seq_num(ctx->seq, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8, &direction, &seqnum))
 		return GSS_S_FAILURE;
 
 	if ((ctx->initiate && direction != 0xff) ||
diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c
index b00b1b426301..283cb25c6237 100644
--- a/net/sunrpc/auth_gss/gss_krb5_wrap.c
+++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c
@@ -122,7 +122,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
 	char			cksumdata[16];
 	struct xdr_netobj	md5cksum = {.len = 0, .data = cksumdata};
 	int			blocksize = 0, plainlen;
-	unsigned char		*ptr, *krb5_hdr, *msg_start;
+	unsigned char		*ptr, *msg_start;
 	s32			now;
 	int			headlen;
 	struct page		**tmp_pages;
@@ -149,26 +149,26 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
 	buf->len += headlen;
 	BUG_ON((buf->len - offset - headlen) % blocksize);
 
-	g_make_token_header(&kctx->mech_used, 24 + plainlen, &ptr);
+	g_make_token_header(&kctx->mech_used,
+				GSS_KRB5_TOK_HDR_LEN + 8 + plainlen, &ptr);
 
 
-	*ptr++ = (unsigned char) ((KG_TOK_WRAP_MSG>>8)&0xff);
-	*ptr++ = (unsigned char) (KG_TOK_WRAP_MSG&0xff);
+	/* ptr now at header described in rfc 1964, section 1.2.1: */
+	ptr[0] = (unsigned char) ((KG_TOK_WRAP_MSG >> 8) & 0xff);
+	ptr[1] = (unsigned char) (KG_TOK_WRAP_MSG & 0xff);
 
-	/* ptr now at byte 2 of header described in rfc 1964, section 1.2.1: */
-	krb5_hdr = ptr - 2;
-	msg_start = krb5_hdr + 24;
+	msg_start = ptr + 24;
 
-	*(__be16 *)(krb5_hdr + 2) = htons(SGN_ALG_DES_MAC_MD5);
-	memset(krb5_hdr + 4, 0xff, 4);
-	*(__be16 *)(krb5_hdr + 4) = htons(SEAL_ALG_DES);
+	*(__be16 *)(ptr + 2) = htons(SGN_ALG_DES_MAC_MD5);
+	memset(ptr + 4, 0xff, 4);
+	*(__be16 *)(ptr + 4) = htons(SEAL_ALG_DES);
 
 	make_confounder(msg_start, blocksize);
 
 	/* XXXJBF: UGH!: */
 	tmp_pages = buf->pages;
 	buf->pages = pages;
-	if (make_checksum("md5", krb5_hdr, 8, buf,
+	if (make_checksum("md5", ptr, 8, buf,
 				offset + headlen - blocksize, &md5cksum))
 		return GSS_S_FAILURE;
 	buf->pages = tmp_pages;
@@ -176,7 +176,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
 	if (krb5_encrypt(kctx->seq, NULL, md5cksum.data,
 			  md5cksum.data, md5cksum.len))
 		return GSS_S_FAILURE;
-	memcpy(krb5_hdr + 16, md5cksum.data + md5cksum.len - 8, 8);
+	memcpy(ptr + GSS_KRB5_TOK_HDR_LEN, md5cksum.data + md5cksum.len - 8, 8);
 
 	spin_lock(&krb5_seq_lock);
 	seq_send = kctx->seq_send++;
@@ -185,7 +185,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset,
 	/* XXX would probably be more efficient to compute checksum
 	 * and encrypt at the same time: */
 	if ((krb5_make_seq_num(kctx->seq, kctx->initiate ? 0 : 0xff,
-			       seq_send, krb5_hdr + 16, krb5_hdr + 8)))
+			       seq_send, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8)))
 		return GSS_S_FAILURE;
 
 	if (gss_encrypt_xdr_buf(kctx->enc, buf, offset + headlen - blocksize,
@@ -219,38 +219,38 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 					buf->len - offset))
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if ((*ptr++ != ((KG_TOK_WRAP_MSG>>8)&0xff)) ||
-	    (*ptr++ !=  (KG_TOK_WRAP_MSG    &0xff))   )
+	if ((ptr[0] != ((KG_TOK_WRAP_MSG >> 8) & 0xff)) ||
+	    (ptr[1] !=  (KG_TOK_WRAP_MSG & 0xff)))
 		return GSS_S_DEFECTIVE_TOKEN;
 
 	/* XXX sanity-check bodysize?? */
 
 	/* get the sign and seal algorithms */
 
-	signalg = ptr[0] + (ptr[1] << 8);
+	signalg = ptr[2] + (ptr[3] << 8);
 	if (signalg != SGN_ALG_DES_MAC_MD5)
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	sealalg = ptr[2] + (ptr[3] << 8);
+	sealalg = ptr[4] + (ptr[5] << 8);
 	if (sealalg != SEAL_ALG_DES)
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if ((ptr[4] != 0xff) || (ptr[5] != 0xff))
+	if ((ptr[6] != 0xff) || (ptr[7] != 0xff))
 		return GSS_S_DEFECTIVE_TOKEN;
 
 	if (gss_decrypt_xdr_buf(kctx->enc, buf,
-			ptr + 22 - (unsigned char *)buf->head[0].iov_base))
+			ptr + GSS_KRB5_TOK_HDR_LEN + 8 - (unsigned char *)buf->head[0].iov_base))
 		return GSS_S_DEFECTIVE_TOKEN;
 
-	if (make_checksum("md5", ptr - 2, 8, buf,
-		 ptr + 22 - (unsigned char *)buf->head[0].iov_base, &md5cksum))
+	if (make_checksum("md5", ptr, 8, buf,
+		 ptr + GSS_KRB5_TOK_HDR_LEN + 8 - (unsigned char *)buf->head[0].iov_base, &md5cksum))
 		return GSS_S_FAILURE;
 
 	if (krb5_encrypt(kctx->seq, NULL, md5cksum.data,
 			   md5cksum.data, md5cksum.len))
 		return GSS_S_FAILURE;
 
-	if (memcmp(md5cksum.data + 8, ptr + 14, 8))
+	if (memcmp(md5cksum.data + 8, ptr + GSS_KRB5_TOK_HDR_LEN, 8))
 		return GSS_S_BAD_SIG;
 
 	/* it got through unscathed.  Make sure the context is unexpired */
@@ -262,8 +262,8 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 
 	/* do sequencing checks */
 
-	if (krb5_get_seq_num(kctx->seq, ptr + 14, ptr + 6, &direction,
-				    &seqnum))
+	if (krb5_get_seq_num(kctx->seq, ptr + GSS_KRB5_TOK_HDR_LEN, ptr + 8,
+				    &direction, &seqnum))
 		return GSS_S_BAD_SIG;
 
 	if ((kctx->initiate && direction != 0xff) ||
@@ -274,7 +274,7 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf)
 	 * better to copy and encrypt at the same time. */
 
 	blocksize = crypto_blkcipher_blocksize(kctx->enc);
-	data_start = ptr + 22 + blocksize;
+	data_start = ptr + GSS_KRB5_TOK_HDR_LEN + 8 + blocksize;
 	orig_start = buf->head[0].iov_base + offset;
 	data_len = (buf->head[0].iov_base + buf->head[0].iov_len) - data_start;
 	memmove(orig_start, data_start, data_len);
-- 
cgit v1.2.3


From 395a59d0f8e86bb39cd700c3d185d30c670bb958 Mon Sep 17 00:00:00 2001
From: Abhishek Sagar <sagar.abhishek@gmail.com>
Date: Sat, 21 Jun 2008 23:47:27 +0530
Subject: ftrace: store mcount address in rec->ip

Record the address of the mcount call-site. Currently all archs except sparc64
record the address of the instruction following the mcount call-site. Some
general cleanups are entailed. Storing mcount addresses in rec->ip enables
looking them up in the kprobe hash table later on to check if they're kprobe'd.

Signed-off-by: Abhishek Sagar <sagar.abhishek@gmail.com>
Cc: davem@davemloft.net
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/arm/kernel/armksyms.c          | 10 +++++-----
 arch/arm/kernel/entry-common.S      |  4 ++++
 arch/arm/kernel/ftrace.c            | 16 +++++++---------
 arch/powerpc/kernel/entry_32.S      |  4 ++++
 arch/powerpc/kernel/entry_64.S      |  5 ++++-
 arch/powerpc/kernel/ftrace.c        | 21 +++++++--------------
 arch/sparc64/kernel/ftrace.c        | 10 ++++++----
 arch/sparc64/kernel/sparc64_ksyms.c |  2 +-
 arch/x86/kernel/entry_32.S          |  4 ++++
 arch/x86/kernel/entry_64.S          |  4 ++++
 arch/x86/kernel/ftrace.c            | 26 +++++++++-----------------
 arch/x86/kernel/i386_ksyms_32.c     |  2 +-
 arch/x86/kernel/x8664_ksyms_64.c    |  2 +-
 include/asm-arm/ftrace.h            | 14 ++++++++++++++
 include/asm-powerpc/ftrace.h        |  8 ++++++++
 include/asm-sparc64/ftrace.h        | 14 ++++++++++++++
 include/asm-x86/ftrace.h            | 14 ++++++++++++++
 include/linux/ftrace.h              |  3 +--
 kernel/trace/ftrace.c               |  3 ++-
 19 files changed, 110 insertions(+), 56 deletions(-)
 create mode 100644 include/asm-arm/ftrace.h
 create mode 100644 include/asm-sparc64/ftrace.h
 create mode 100644 include/asm-x86/ftrace.h

(limited to 'include/linux')

diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c
index 3b132215cbf8..cc7b246e9652 100644
--- a/arch/arm/kernel/armksyms.c
+++ b/arch/arm/kernel/armksyms.c
@@ -18,6 +18,7 @@
 #include <asm/io.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
+#include <asm/ftrace.h>
 
 /*
  * libgcc functions - functions that are used internally by the
@@ -48,11 +49,6 @@ extern void __aeabi_ulcmp(void);
 extern void fpundefinstr(void);
 extern void fp_enter(void);
 
-#ifdef CONFIG_FTRACE
-extern void mcount(void);
-EXPORT_SYMBOL(mcount);
-#endif
-
 /*
  * This has a special calling convention; it doesn't
  * modify any of the usual registers, except for LR.
@@ -186,3 +182,7 @@ EXPORT_SYMBOL(_find_next_bit_be);
 #endif
 
 EXPORT_SYMBOL(copy_page);
+
+#ifdef CONFIG_FTRACE
+EXPORT_SYMBOL(mcount);
+#endif
diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S
index 8f79a4789ed4..84694e88b428 100644
--- a/arch/arm/kernel/entry-common.S
+++ b/arch/arm/kernel/entry-common.S
@@ -9,6 +9,7 @@
  */
 
 #include <asm/unistd.h>
+#include <asm/ftrace.h>
 #include <asm/arch/entry-macro.S>
 
 #include "entry-header.S"
@@ -104,6 +105,7 @@ ENTRY(ret_from_fork)
 ENTRY(mcount)
 	stmdb sp!, {r0-r3, lr}
 	mov r0, lr
+	sub r0, r0, #MCOUNT_INSN_SIZE
 
 	.globl mcount_call
 mcount_call:
@@ -114,6 +116,7 @@ ENTRY(ftrace_caller)
 	stmdb sp!, {r0-r3, lr}
 	ldr r1, [fp, #-4]
 	mov r0, lr
+	sub r0, r0, #MCOUNT_INSN_SIZE
 
 	.globl ftrace_call
 ftrace_call:
@@ -134,6 +137,7 @@ ENTRY(mcount)
 trace:
 	ldr r1, [fp, #-4]
 	mov r0, lr
+	sub r0, r0, #MCOUNT_INSN_SIZE
 	mov lr, pc
 	mov pc, r2
 	ldmia sp!, {r0-r3, pc}
diff --git a/arch/arm/kernel/ftrace.c b/arch/arm/kernel/ftrace.c
index 22f3d6e309f9..76d50e6091bc 100644
--- a/arch/arm/kernel/ftrace.c
+++ b/arch/arm/kernel/ftrace.c
@@ -12,9 +12,10 @@
  */
 
 #include <linux/ftrace.h>
+
 #include <asm/cacheflush.h>
+#include <asm/ftrace.h>
 
-#define INSN_SIZE      4
 #define PC_OFFSET      8
 #define BL_OPCODE      0xeb000000
 #define BL_OFFSET_MASK 0x00ffffff
@@ -32,10 +33,10 @@ unsigned char *ftrace_call_replace(unsigned long pc, unsigned long addr)
 {
 	long offset;
 
-	offset = (long)addr - (long)(pc - INSN_SIZE + PC_OFFSET);
+	offset = (long)addr - (long)(pc + PC_OFFSET);
 	if (unlikely(offset < -33554432 || offset > 33554428)) {
 		/* Can't generate branches that far (from ARM ARM). Ftrace
-		 * doesn't generate branches outside of core kernel text.
+		 * doesn't generate branches outside of kernel text.
 		 */
 		WARN_ON_ONCE(1);
 		return NULL;
@@ -52,7 +53,6 @@ int ftrace_modify_code(unsigned long pc, unsigned char *old_code,
 
 	old = *(unsigned long *)old_code;
 	new = *(unsigned long *)new_code;
-	pc -= INSN_SIZE;
 
 	__asm__ __volatile__ (
 		"1:  ldr    %1, [%2]  \n"
@@ -77,7 +77,7 @@ int ftrace_modify_code(unsigned long pc, unsigned char *old_code,
 		: "memory");
 
 	if (!err && (replaced == old))
-		flush_icache_range(pc, pc + INSN_SIZE);
+		flush_icache_range(pc, pc + MCOUNT_INSN_SIZE);
 
 	return err;
 }
@@ -89,8 +89,7 @@ int ftrace_update_ftrace_func(ftrace_func_t func)
 	unsigned char *new;
 
 	pc = (unsigned long)&ftrace_call;
-	pc += INSN_SIZE;
-	memcpy(&old, &ftrace_call, INSN_SIZE);
+	memcpy(&old, &ftrace_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(pc, (unsigned long)func);
 	ret = ftrace_modify_code(pc, (unsigned char *)&old, new);
 	return ret;
@@ -103,8 +102,7 @@ int ftrace_mcount_set(unsigned long *data)
 	unsigned char *new;
 
 	pc = (unsigned long)&mcount_call;
-	pc += INSN_SIZE;
-	memcpy(&old, &mcount_call, INSN_SIZE);
+	memcpy(&old, &mcount_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(pc, *addr);
 	*addr = ftrace_modify_code(pc, (unsigned char *)&old, new);
 	return 0;
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index 3b1dd29d9f91..7231a708af0d 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -30,6 +30,7 @@
 #include <asm/ppc_asm.h>
 #include <asm/asm-offsets.h>
 #include <asm/unistd.h>
+#include <asm/ftrace.h>
 
 #undef SHOW_SYSCALLS
 #undef SHOW_SYSCALLS_TASK
@@ -1053,6 +1054,7 @@ _GLOBAL(_mcount)
 	stw	r10,40(r1)
 	stw	r3, 44(r1)
 	stw	r5, 8(r1)
+	subi	r3, r3, MCOUNT_INSN_SIZE
 	.globl mcount_call
 mcount_call:
 	bl	ftrace_stub
@@ -1090,6 +1092,7 @@ _GLOBAL(ftrace_caller)
 	stw	r10,40(r1)
 	stw	r3, 44(r1)
 	stw	r5, 8(r1)
+	subi	r3, r3, MCOUNT_INSN_SIZE
 .globl ftrace_call
 ftrace_call:
 	bl	ftrace_stub
@@ -1128,6 +1131,7 @@ _GLOBAL(_mcount)
 	stw	r3, 44(r1)
 	stw	r5, 8(r1)
 
+	subi	r3, r3, MCOUNT_INSN_SIZE
 	LOAD_REG_ADDR(r5, ftrace_trace_function)
 	lwz	r5,0(r5)
 
diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
index 2c4d9e056ead..2f511a969d2c 100644
--- a/arch/powerpc/kernel/entry_64.S
+++ b/arch/powerpc/kernel/entry_64.S
@@ -31,6 +31,7 @@
 #include <asm/bug.h>
 #include <asm/ptrace.h>
 #include <asm/irqflags.h>
+#include <asm/ftrace.h>
 
 /*
  * System calls.
@@ -879,6 +880,7 @@ _GLOBAL(_mcount)
 	mflr	r3
 	stdu	r1, -112(r1)
 	std	r3, 128(r1)
+	subi	r3, r3, MCOUNT_INSN_SIZE
 	.globl mcount_call
 mcount_call:
 	bl	ftrace_stub
@@ -895,6 +897,7 @@ _GLOBAL(ftrace_caller)
 	stdu	r1, -112(r1)
 	std	r3, 128(r1)
 	ld	r4, 16(r11)
+	subi	r3, r3, MCOUNT_INSN_SIZE
 .globl ftrace_call
 ftrace_call:
 	bl	ftrace_stub
@@ -916,7 +919,7 @@ _GLOBAL(_mcount)
 	std	r3, 128(r1)
 	ld	r4, 16(r11)
 
-
+	subi	r3, r3, MCOUNT_INSN_SIZE
 	LOAD_REG_ADDR(r5,ftrace_trace_function)
 	ld	r5,0(r5)
 	ld	r5,0(r5)
diff --git a/arch/powerpc/kernel/ftrace.c b/arch/powerpc/kernel/ftrace.c
index e12c593ab9ca..3855ceb937b0 100644
--- a/arch/powerpc/kernel/ftrace.c
+++ b/arch/powerpc/kernel/ftrace.c
@@ -15,8 +15,8 @@
 #include <linux/list.h>
 
 #include <asm/cacheflush.h>
+#include <asm/ftrace.h>
 
-#define CALL_BACK		4
 
 static unsigned int ftrace_nop = 0x60000000;
 
@@ -27,9 +27,10 @@ static unsigned int ftrace_nop = 0x60000000;
 # define GET_ADDR(addr) *(unsigned long *)addr
 #endif
 
+
 static unsigned int notrace ftrace_calc_offset(long ip, long addr)
 {
-	return (int)((addr + CALL_BACK) - ip);
+	return (int)(addr - ip);
 }
 
 notrace unsigned char *ftrace_nop_replace(void)
@@ -76,9 +77,6 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 	unsigned new = *(unsigned *)new_code;
 	int faulted = 0;
 
-	/* move the IP back to the start of the call */
-	ip -= CALL_BACK;
-
 	/*
 	 * Note: Due to modules and __init, code can
 	 *  disappear and change, we need to protect against faulting
@@ -118,12 +116,10 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 notrace int ftrace_update_ftrace_func(ftrace_func_t func)
 {
 	unsigned long ip = (unsigned long)(&ftrace_call);
-	unsigned char old[4], *new;
+	unsigned char old[MCOUNT_INSN_SIZE], *new;
 	int ret;
 
-	ip += CALL_BACK;
-
-	memcpy(old, &ftrace_call, 4);
+	memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(ip, (unsigned long)func);
 	ret = ftrace_modify_code(ip, old, new);
 
@@ -134,16 +130,13 @@ notrace int ftrace_mcount_set(unsigned long *data)
 {
 	unsigned long ip = (long)(&mcount_call);
 	unsigned long *addr = data;
-	unsigned char old[4], *new;
-
-	/* ip is at the location, but modify code will subtact this */
-	ip += CALL_BACK;
+	unsigned char old[MCOUNT_INSN_SIZE], *new;
 
 	/*
 	 * Replace the mcount stub with a pointer to the
 	 * ip recorder function.
 	 */
-	memcpy(old, &mcount_call, 4);
+	memcpy(old, &mcount_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(ip, *addr);
 	*addr = ftrace_modify_code(ip, old, new);
 
diff --git a/arch/sparc64/kernel/ftrace.c b/arch/sparc64/kernel/ftrace.c
index c17373195b1e..4298d0aee713 100644
--- a/arch/sparc64/kernel/ftrace.c
+++ b/arch/sparc64/kernel/ftrace.c
@@ -5,6 +5,8 @@
 #include <linux/init.h>
 #include <linux/list.h>
 
+#include <asm/ftrace.h>
+
 static const u32 ftrace_nop = 0x01000000;
 
 notrace unsigned char *ftrace_nop_replace(void)
@@ -60,9 +62,9 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 notrace int ftrace_update_ftrace_func(ftrace_func_t func)
 {
 	unsigned long ip = (unsigned long)(&ftrace_call);
-	unsigned char old[4], *new;
+	unsigned char old[MCOUNT_INSN_SIZE], *new;
 
-	memcpy(old, &ftrace_call, 4);
+	memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(ip, (unsigned long)func);
 	return ftrace_modify_code(ip, old, new);
 }
@@ -71,13 +73,13 @@ notrace int ftrace_mcount_set(unsigned long *data)
 {
 	unsigned long ip = (long)(&mcount_call);
 	unsigned long *addr = data;
-	unsigned char old[4], *new;
+	unsigned char old[MCOUNT_INSN_SIZE], *new;
 
 	/*
 	 * Replace the mcount stub with a pointer to the
 	 * ip recorder function.
 	 */
-	memcpy(old, &mcount_call, 4);
+	memcpy(old, &mcount_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(ip, *addr);
 	*addr = ftrace_modify_code(ip, old, new);
 
diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c
index 8ac0b99f2c55..b80d982a29c6 100644
--- a/arch/sparc64/kernel/sparc64_ksyms.c
+++ b/arch/sparc64/kernel/sparc64_ksyms.c
@@ -53,6 +53,7 @@
 #include <asm/ns87303.h>
 #include <asm/timer.h>
 #include <asm/cpudata.h>
+#include <asm/ftrace.h>
 
 struct poll {
 	int fd;
@@ -112,7 +113,6 @@ EXPORT_SYMBOL(smp_call_function);
 #endif /* CONFIG_SMP */
 
 #if defined(CONFIG_MCOUNT)
-extern void _mcount(void);
 EXPORT_SYMBOL(_mcount);
 #endif
 
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 04ea83ccb979..95e6bbe3665e 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -51,6 +51,7 @@
 #include <asm/percpu.h>
 #include <asm/dwarf2.h>
 #include <asm/processor-flags.h>
+#include <asm/ftrace.h>
 #include "irq_vectors.h"
 
 /*
@@ -1118,6 +1119,7 @@ ENTRY(mcount)
 	pushl %ecx
 	pushl %edx
 	movl 0xc(%esp), %eax
+	subl $MCOUNT_INSN_SIZE, %eax
 
 .globl mcount_call
 mcount_call:
@@ -1136,6 +1138,7 @@ ENTRY(ftrace_caller)
 	pushl %edx
 	movl 0xc(%esp), %eax
 	movl 0x4(%ebp), %edx
+	subl $MCOUNT_INSN_SIZE, %eax
 
 .globl ftrace_call
 ftrace_call:
@@ -1166,6 +1169,7 @@ trace:
 	pushl %edx
 	movl 0xc(%esp), %eax
 	movl 0x4(%ebp), %edx
+	subl $MCOUNT_INSN_SIZE, %eax
 
 	call *ftrace_trace_function
 
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index fe25e5febca3..b0f7308f78a6 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -51,6 +51,7 @@
 #include <asm/page.h>
 #include <asm/irqflags.h>
 #include <asm/paravirt.h>
+#include <asm/ftrace.h>
 
 	.code64
 
@@ -68,6 +69,7 @@ ENTRY(mcount)
 	movq %r9, 48(%rsp)
 
 	movq 0x38(%rsp), %rdi
+	subq $MCOUNT_INSN_SIZE, %rdi
 
 .globl mcount_call
 mcount_call:
@@ -99,6 +101,7 @@ ENTRY(ftrace_caller)
 
 	movq 0x38(%rsp), %rdi
 	movq 8(%rbp), %rsi
+	subq $MCOUNT_INSN_SIZE, %rdi
 
 .globl ftrace_call
 ftrace_call:
@@ -139,6 +142,7 @@ trace:
 
 	movq 0x38(%rsp), %rdi
 	movq 8(%rbp), %rsi
+	subq $MCOUNT_INSN_SIZE, %rdi
 
 	call   *ftrace_trace_function
 
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 55828149e01e..ab115cd15fdf 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -17,20 +17,21 @@
 #include <linux/list.h>
 
 #include <asm/alternative.h>
+#include <asm/ftrace.h>
 
-#define CALL_BACK		5
 
 /* Long is fine, even if it is only 4 bytes ;-) */
 static long *ftrace_nop;
 
 union ftrace_code_union {
-	char code[5];
+	char code[MCOUNT_INSN_SIZE];
 	struct {
 		char e8;
 		int offset;
 	} __attribute__((packed));
 };
 
+
 static int notrace ftrace_calc_offset(long ip, long addr)
 {
 	return (int)(addr - ip);
@@ -46,7 +47,7 @@ notrace unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr)
 	static union ftrace_code_union calc;
 
 	calc.e8		= 0xe8;
-	calc.offset	= ftrace_calc_offset(ip, addr);
+	calc.offset	= ftrace_calc_offset(ip + MCOUNT_INSN_SIZE, addr);
 
 	/*
 	 * No locking needed, this must be called via kstop_machine
@@ -65,9 +66,6 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 	unsigned char newch = new_code[4];
 	int faulted = 0;
 
-	/* move the IP back to the start of the call */
-	ip -= CALL_BACK;
-
 	/*
 	 * Note: Due to modules and __init, code can
 	 *  disappear and change, we need to protect against faulting
@@ -102,12 +100,10 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 notrace int ftrace_update_ftrace_func(ftrace_func_t func)
 {
 	unsigned long ip = (unsigned long)(&ftrace_call);
-	unsigned char old[5], *new;
+	unsigned char old[MCOUNT_INSN_SIZE], *new;
 	int ret;
 
-	ip += CALL_BACK;
-
-	memcpy(old, &ftrace_call, 5);
+	memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(ip, (unsigned long)func);
 	ret = ftrace_modify_code(ip, old, new);
 
@@ -118,16 +114,13 @@ notrace int ftrace_mcount_set(unsigned long *data)
 {
 	unsigned long ip = (long)(&mcount_call);
 	unsigned long *addr = data;
-	unsigned char old[5], *new;
-
-	/* ip is at the location, but modify code will subtact this */
-	ip += CALL_BACK;
+	unsigned char old[MCOUNT_INSN_SIZE], *new;
 
 	/*
 	 * Replace the mcount stub with a pointer to the
 	 * ip recorder function.
 	 */
-	memcpy(old, &mcount_call, 5);
+	memcpy(old, &mcount_call, MCOUNT_INSN_SIZE);
 	new = ftrace_call_replace(ip, *addr);
 	*addr = ftrace_modify_code(ip, old, new);
 
@@ -142,8 +135,7 @@ int __init ftrace_dyn_arch_init(void *data)
 
 	ftrace_mcount_set(data);
 
-	ftrace_nop = (unsigned long *)noptable[CALL_BACK];
+	ftrace_nop = (unsigned long *)noptable[MCOUNT_INSN_SIZE];
 
 	return 0;
 }
-
diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c
index 29999dbb754c..dd7ebee446af 100644
--- a/arch/x86/kernel/i386_ksyms_32.c
+++ b/arch/x86/kernel/i386_ksyms_32.c
@@ -1,9 +1,9 @@
-#include <linux/ftrace.h>
 #include <linux/module.h>
 
 #include <asm/checksum.h>
 #include <asm/pgtable.h>
 #include <asm/desc.h>
+#include <asm/ftrace.h>
 
 #ifdef CONFIG_FTRACE
 /* mcount is defined in assembly */
diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c
index 122885bc5f3b..16ff4bf418d9 100644
--- a/arch/x86/kernel/x8664_ksyms_64.c
+++ b/arch/x86/kernel/x8664_ksyms_64.c
@@ -1,7 +1,6 @@
 /* Exports for assembly files.
    All C exports should go in the respective C files. */
 
-#include <linux/ftrace.h>
 #include <linux/module.h>
 #include <linux/smp.h>
 
@@ -11,6 +10,7 @@
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
 #include <asm/desc.h>
+#include <asm/ftrace.h>
 
 #ifdef CONFIG_FTRACE
 /* mcount is defined in assembly */
diff --git a/include/asm-arm/ftrace.h b/include/asm-arm/ftrace.h
new file mode 100644
index 000000000000..584ef9a8e5a5
--- /dev/null
+++ b/include/asm-arm/ftrace.h
@@ -0,0 +1,14 @@
+#ifndef _ASM_ARM_FTRACE
+#define _ASM_ARM_FTRACE
+
+#ifdef CONFIG_FTRACE
+#define MCOUNT_ADDR		((long)(mcount))
+#define MCOUNT_INSN_SIZE	4 /* sizeof mcount call */
+
+#ifndef __ASSEMBLY__
+extern void mcount(void);
+#endif
+
+#endif
+
+#endif /* _ASM_ARM_FTRACE */
diff --git a/include/asm-powerpc/ftrace.h b/include/asm-powerpc/ftrace.h
index b1bfa704b6e5..de921326cca8 100644
--- a/include/asm-powerpc/ftrace.h
+++ b/include/asm-powerpc/ftrace.h
@@ -1,6 +1,14 @@
 #ifndef _ASM_POWERPC_FTRACE
 #define _ASM_POWERPC_FTRACE
 
+#ifdef CONFIG_FTRACE
+#define MCOUNT_ADDR		((long)(_mcount))
+#define MCOUNT_INSN_SIZE	4 /* sizeof mcount call */
+
+#ifndef __ASSEMBLY__
 extern void _mcount(void);
+#endif
 
 #endif
+
+#endif /* _ASM_POWERPC_FTRACE */
diff --git a/include/asm-sparc64/ftrace.h b/include/asm-sparc64/ftrace.h
new file mode 100644
index 000000000000..f76a40a338bb
--- /dev/null
+++ b/include/asm-sparc64/ftrace.h
@@ -0,0 +1,14 @@
+#ifndef _ASM_SPARC64_FTRACE
+#define _ASM_SPARC64_FTRACE
+
+#ifdef CONFIG_FTRACE
+#define MCOUNT_ADDR		((long)(_mcount))
+#define MCOUNT_INSN_SIZE	4 /* sizeof mcount call */
+
+#ifndef __ASSEMBLY__
+extern void _mcount(void);
+#endif
+
+#endif
+
+#endif /* _ASM_SPARC64_FTRACE */
diff --git a/include/asm-x86/ftrace.h b/include/asm-x86/ftrace.h
new file mode 100644
index 000000000000..c184441133f2
--- /dev/null
+++ b/include/asm-x86/ftrace.h
@@ -0,0 +1,14 @@
+#ifndef _ASM_X86_FTRACE
+#define _ASM_SPARC64_FTRACE
+
+#ifdef CONFIG_FTRACE
+#define MCOUNT_ADDR		((long)(mcount))
+#define MCOUNT_INSN_SIZE	5 /* sizeof mcount call */
+
+#ifndef __ASSEMBLY__
+extern void mcount(void);
+#endif
+
+#endif /* CONFIG_FTRACE */
+
+#endif /* _ASM_X86_FTRACE */
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 20e14d0093c7..366098d591de 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -31,7 +31,6 @@ int unregister_ftrace_function(struct ftrace_ops *ops);
 void clear_ftrace_function(void);
 
 extern void ftrace_stub(unsigned long a0, unsigned long a1);
-extern void mcount(void);
 
 #else /* !CONFIG_FTRACE */
 # define register_ftrace_function(ops) do { } while (0)
@@ -54,7 +53,7 @@ enum {
 
 struct dyn_ftrace {
 	struct hlist_node node;
-	unsigned long	  ip;
+	unsigned long	  ip; /* address of mcount call-site */
 	unsigned long	  flags;
 };
 
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 0d5bcf69952d..f1e9e5c74e64 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -27,6 +27,8 @@
 #include <linux/hash.h>
 #include <linux/list.h>
 
+#include <asm/ftrace.h>
+
 #include "trace.h"
 
 /* ftrace_enabled is a method to turn ftrace on or off */
@@ -329,7 +331,6 @@ ftrace_record_ip(unsigned long ip)
 }
 
 #define FTRACE_ADDR ((long)(ftrace_caller))
-#define MCOUNT_ADDR ((long)(mcount))
 
 static int
 __ftrace_replace_code(struct dyn_ftrace *rec,
-- 
cgit v1.2.3


From 785656a41f9a9c0e843a23d1ae05d900b5158f8f Mon Sep 17 00:00:00 2001
From: Abhishek Sagar <sagar.abhishek@gmail.com>
Date: Sat, 21 Jun 2008 23:47:39 +0530
Subject: kprobes: enable clean usage of get_kprobe

Allow clean use of get_kprobe() outside of core kprobe code. Ftrace makes use
of get_kprobe to identify probes installed on mcount call-sites.

Signed-off-by: Abhishek Sagar <sagar.abhishek@gmail.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: jkenisto@us.ibm.com
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/kprobes.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 1036631ff4fa..04a3556bdea6 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -259,6 +259,10 @@ void recycle_rp_inst(struct kretprobe_instance *ri, struct hlist_head *head);
 struct jprobe;
 struct kretprobe;
 
+static inline struct kprobe *get_kprobe(void *addr)
+{
+	return NULL;
+}
 static inline struct kprobe *kprobe_running(void)
 {
 	return NULL;
-- 
cgit v1.2.3


From ecea656d1d5e912d2f3d332657ea4a6d8380f891 Mon Sep 17 00:00:00 2001
From: Abhishek Sagar <sagar.abhishek@gmail.com>
Date: Sat, 21 Jun 2008 23:47:53 +0530
Subject: ftrace: freeze kprobe'd records

Let records identified as being kprobe'd be marked as "frozen". The trouble
with records which have a kprobe installed on their mcount call-site is
that they don't get updated. So if such a function which is currently being
traced gets its tracing disabled due to a new filter rule (or because it
was added to the notrace list) then it won't be updated and continue being
traced. This patch allows scanning of all frozen records during tracing to
check if they should be traced.

Signed-off-by: Abhishek Sagar <sagar.abhishek@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/ftrace.h |  6 ++++-
 kernel/trace/ftrace.c  | 72 +++++++++++++++++++++++++++++++++++++++++++++++++-
 kernel/trace/trace.c   |  3 +++
 3 files changed, 79 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 366098d591de..3121b95443d9 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -49,6 +49,7 @@ enum {
 	FTRACE_FL_ENABLED	= (1 << 3),
 	FTRACE_FL_NOTRACE	= (1 << 4),
 	FTRACE_FL_CONVERTED	= (1 << 5),
+	FTRACE_FL_FROZEN	= (1 << 6),
 };
 
 struct dyn_ftrace {
@@ -73,15 +74,18 @@ extern void ftrace_caller(void);
 extern void ftrace_call(void);
 extern void mcount_call(void);
 
+extern int skip_trace(unsigned long ip);
+
 void ftrace_disable_daemon(void);
 void ftrace_enable_daemon(void);
 
 #else
+# define skip_trace(ip)				({ 0; })
 # define ftrace_force_update()			({ 0; })
 # define ftrace_set_filter(buf, len, reset)	do { } while (0)
 # define ftrace_disable_daemon()		do { } while (0)
 # define ftrace_enable_daemon()			do { } while (0)
-#endif
+#endif /* CONFIG_DYNAMIC_FTRACE */
 
 /* totally disable ftrace - can not re-enable after this */
 void ftrace_kill(void);
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index f1e9e5c74e64..d1238163155f 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -163,6 +163,8 @@ enum {
 };
 
 static int ftrace_filtered;
+static int tracing_on;
+static int frozen_record_count;
 
 static struct hlist_head ftrace_hash[FTRACE_HASHSIZE];
 
@@ -195,6 +197,71 @@ static int ftrace_record_suspend;
 
 static struct dyn_ftrace *ftrace_free_records;
 
+
+#ifdef CONFIG_KPROBES
+static inline void freeze_record(struct dyn_ftrace *rec)
+{
+	if (!(rec->flags & FTRACE_FL_FROZEN)) {
+		rec->flags |= FTRACE_FL_FROZEN;
+		frozen_record_count++;
+	}
+}
+
+static inline void unfreeze_record(struct dyn_ftrace *rec)
+{
+	if (rec->flags & FTRACE_FL_FROZEN) {
+		rec->flags &= ~FTRACE_FL_FROZEN;
+		frozen_record_count--;
+	}
+}
+
+static inline int record_frozen(struct dyn_ftrace *rec)
+{
+	return rec->flags & FTRACE_FL_FROZEN;
+}
+#else
+# define freeze_record(rec)			({ 0; })
+# define unfreeze_record(rec)			({ 0; })
+# define record_frozen(rec)			({ 0; })
+#endif /* CONFIG_KPROBES */
+
+int skip_trace(unsigned long ip)
+{
+	unsigned long fl;
+	struct dyn_ftrace *rec;
+	struct hlist_node *t;
+	struct hlist_head *head;
+
+	if (frozen_record_count == 0)
+		return 0;
+
+	head = &ftrace_hash[hash_long(ip, FTRACE_HASHBITS)];
+	hlist_for_each_entry_rcu(rec, t, head, node) {
+		if (rec->ip == ip) {
+			if (record_frozen(rec)) {
+				if (rec->flags & FTRACE_FL_FAILED)
+					return 1;
+
+				if (!(rec->flags & FTRACE_FL_CONVERTED))
+					return 1;
+
+				if (!tracing_on || !ftrace_enabled)
+					return 1;
+
+				if (ftrace_filtered) {
+					fl = rec->flags & (FTRACE_FL_FILTER |
+							   FTRACE_FL_NOTRACE);
+					if (!fl || (fl & FTRACE_FL_NOTRACE))
+						return 1;
+				}
+			}
+			break;
+		}
+	}
+
+	return 0;
+}
+
 static inline int
 ftrace_ip_in_hash(unsigned long ip, unsigned long key)
 {
@@ -489,8 +556,11 @@ static int __ftrace_modify_code(void *data)
 		 */
 		__ftrace_update_code(NULL);
 		ftrace_replace_code(1);
-	} else if (*command & FTRACE_DISABLE_CALLS)
+		tracing_on = 1;
+	} else if (*command & FTRACE_DISABLE_CALLS) {
 		ftrace_replace_code(0);
+		tracing_on = 0;
+	}
 
 	if (*command & FTRACE_UPDATE_TRACE_FUNC)
 		ftrace_update_ftrace_func(ftrace_trace_function);
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 6e9dae7eb418..9ade79369bfb 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -988,6 +988,9 @@ function_trace_call(unsigned long ip, unsigned long parent_ip)
 	if (unlikely(!tracer_enabled))
 		return;
 
+	if (skip_trace(ip))
+		return;
+
 	local_irq_save(flags);
 	cpu = raw_smp_processor_id();
 	data = tr->data[cpu];
-- 
cgit v1.2.3


From 3da757daf86e498872855f0b5e101f763ba79499 Mon Sep 17 00:00:00 2001
From: Alok Kataria <akataria@vmware.com>
Date: Fri, 20 Jun 2008 15:06:33 -0700
Subject: x86: use cpu_khz for loops_per_jiffy calculation

On the x86 platform we can use the value of tsc_khz computed during tsc
calibration to calculate the loops_per_jiffy value. Its very important
to keep the error in lpj values to minimum as any error in that may
result in kernel panic in check_timer. In virtualization environment, On
a highly overloaded host the guest delay calibration may sometimes
result in errors beyond the ~50% that timer_irq_works can handle,
resulting in the guest panicking.

Does some formating changes to lpj_setup code to now have a single
printk to print the bogomips value.

We do this only for the boot processor because the AP's can have
different base frequencies or the BIOS might boot a AP at a different
frequency.

Signed-off-by: Alok N Kataria <akataria@vmware.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Daniel Hecht <dhecht@vmware.com>
Cc: Tim Mann <mann@vmware.com>
Cc: Zach Amsden <zach@vmware.com>
Cc: Sahil Rihan <srihan@vmware.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/time_64.c |  2 ++
 arch/x86/kernel/tsc_32.c  |  5 +++++
 include/linux/delay.h     |  1 +
 init/calibrate.c          | 36 +++++++++++++++++++-----------------
 4 files changed, 27 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index c737849e2ef7..12b4a71bd074 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -123,6 +123,8 @@ void __init time_init(void)
 		(boot_cpu_data.x86_vendor == X86_VENDOR_AMD))
 		cpu_khz = calculate_cpu_khz();
 
+	lpj_tsc = ((unsigned long)tsc_khz * 1000)/HZ;
+
 	if (unsynchronized_tsc())
 		mark_tsc_unstable("TSCs unsynchronized");
 
diff --git a/arch/x86/kernel/tsc_32.c b/arch/x86/kernel/tsc_32.c
index 068759db63dd..be729035b30b 100644
--- a/arch/x86/kernel/tsc_32.c
+++ b/arch/x86/kernel/tsc_32.c
@@ -401,6 +401,7 @@ static inline void check_geode_tsc_reliable(void) { }
 void __init tsc_init(void)
 {
 	int cpu;
+	u64 lpj;
 
 	if (!cpu_has_tsc || tsc_disabled) {
 		/* Disable the TSC in case of !cpu_has_tsc */
@@ -421,6 +422,10 @@ void __init tsc_init(void)
 		return;
 	}
 
+	lpj = ((u64)tsc_khz * 1000);
+	do_div(lpj, HZ);
+	lpj_tsc = lpj;
+
 	printk("Detected %lu.%03lu MHz processor.\n",
 				(unsigned long)cpu_khz / 1000,
 				(unsigned long)cpu_khz % 1000);
diff --git a/include/linux/delay.h b/include/linux/delay.h
index 54552d21296e..01aec60590ab 100644
--- a/include/linux/delay.h
+++ b/include/linux/delay.h
@@ -41,6 +41,7 @@ static inline void ndelay(unsigned long x)
 #define ndelay(x) ndelay(x)
 #endif
 
+extern unsigned long lpj_tsc;
 void calibrate_delay(void);
 void msleep(unsigned int msecs);
 unsigned long msleep_interruptible(unsigned int msecs);
diff --git a/init/calibrate.c b/init/calibrate.c
index ecb3822d4f70..86286974dada 100644
--- a/init/calibrate.c
+++ b/init/calibrate.c
@@ -8,7 +8,9 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/timex.h>
+#include <linux/smp.h>
 
+unsigned long lpj_tsc;
 unsigned long preset_lpj;
 static int __init lpj_setup(char *str)
 {
@@ -108,6 +110,10 @@ static unsigned long __cpuinit calibrate_delay_direct(void) {return 0;}
  * This is the number of bits of precision for the loops_per_jiffy.  Each
  * bit takes on average 1.5/HZ seconds.  This (like the original) is a little
  * better than 1%
+ * For the boot cpu we can skip the delay calibration and assign it a value
+ * calculated based on the tsc frequency.
+ * For the rest of the CPUs we cannot assume that the tsc frequency is same as
+ * the cpu frequency, hence do the calibration for those.
  */
 #define LPS_PREC 8
 
@@ -118,20 +124,20 @@ void __cpuinit calibrate_delay(void)
 
 	if (preset_lpj) {
 		loops_per_jiffy = preset_lpj;
-		printk("Calibrating delay loop (skipped)... "
-			"%lu.%02lu BogoMIPS preset\n",
-			loops_per_jiffy/(500000/HZ),
-			(loops_per_jiffy/(5000/HZ)) % 100);
+		printk(KERN_INFO
+			"Calibrating delay loop (skipped) preset value.. ");
+	} else if ((smp_processor_id() == 0) && lpj_tsc) {
+		loops_per_jiffy = lpj_tsc;
+		printk(KERN_INFO
+			"Calibrating delay loop (skipped), "
+			"using tsc calculated value.. ");
 	} else if ((loops_per_jiffy = calibrate_delay_direct()) != 0) {
-		printk("Calibrating delay using timer specific routine.. ");
-		printk("%lu.%02lu BogoMIPS (lpj=%lu)\n",
-			loops_per_jiffy/(500000/HZ),
-			(loops_per_jiffy/(5000/HZ)) % 100,
-			loops_per_jiffy);
+		printk(KERN_INFO
+			"Calibrating delay using timer specific routine.. ");
 	} else {
 		loops_per_jiffy = (1<<12);
 
-		printk(KERN_DEBUG "Calibrating delay loop... ");
+		printk(KERN_INFO "Calibrating delay loop... ");
 		while ((loops_per_jiffy <<= 1) != 0) {
 			/* wait for "start of" clock tick */
 			ticks = jiffies;
@@ -161,12 +167,8 @@ void __cpuinit calibrate_delay(void)
 			if (jiffies != ticks)	/* longer than 1 tick */
 				loops_per_jiffy &= ~loopbit;
 		}
-
-		/* Round the value and print it */
-		printk("%lu.%02lu BogoMIPS (lpj=%lu)\n",
-			loops_per_jiffy/(500000/HZ),
-			(loops_per_jiffy/(5000/HZ)) % 100,
-			loops_per_jiffy);
 	}
-
+	printk(KERN_INFO "%lu.%02lu BogoMIPS (lpj=%lu)\n",
+			loops_per_jiffy/(500000/HZ),
+			(loops_per_jiffy/(5000/HZ)) % 100, loops_per_jiffy);
 }
-- 
cgit v1.2.3


From 961ccddd59d627b89bd3dc284b6517833bbdf25d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 23 Jun 2008 13:55:38 +1000
Subject: sched: add new API sched_setscheduler_nocheck: add a flag to control
 access checks

Hidehiro Kawai noticed that sched_setscheduler() can fail in
stop_machine: it calls sched_setscheduler() from insmod, which can
have CAP_SYS_MODULE without CAP_SYS_NICE.

Two cases could have failed, so are changed to sched_setscheduler_nocheck:
  kernel/softirq.c:cpu_callback()
	- CPU hotplug callback
  kernel/stop_machine.c:__stop_machine_run()
	- Called from various places, including modprobe()

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-mm@kvack.org
Cc: sugita <yumiko.sugita.yf@hitachi.com>
Cc: Satoshi OSHIMA <satoshi.oshima.fk@hitachi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |  2 ++
 kernel/sched.c        | 48 ++++++++++++++++++++++++++++++++++++------------
 kernel/softirq.c      |  2 +-
 kernel/stop_machine.c |  2 +-
 4 files changed, 40 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index c5d3f847ca8d..fe3b9b5d7390 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1655,6 +1655,8 @@ extern int can_nice(const struct task_struct *p, const int nice);
 extern int task_curr(const struct task_struct *p);
 extern int idle_cpu(int cpu);
 extern int sched_setscheduler(struct task_struct *, int, struct sched_param *);
+extern int sched_setscheduler_nocheck(struct task_struct *, int,
+				      struct sched_param *);
 extern struct task_struct *idle_task(int cpu);
 extern struct task_struct *curr_task(int cpu);
 extern void set_curr_task(int cpu, struct task_struct *p);
diff --git a/kernel/sched.c b/kernel/sched.c
index b048ad8a11af..8d7c246ab864 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4746,16 +4746,8 @@ __setscheduler(struct rq *rq, struct task_struct *p, int policy, int prio)
 	set_load_weight(p);
 }
 
-/**
- * sched_setscheduler - change the scheduling policy and/or RT priority of a thread.
- * @p: the task in question.
- * @policy: new policy.
- * @param: structure containing the new RT priority.
- *
- * NOTE that the task may be already dead.
- */
-int sched_setscheduler(struct task_struct *p, int policy,
-		       struct sched_param *param)
+static int __sched_setscheduler(struct task_struct *p, int policy,
+				struct sched_param *param, bool user)
 {
 	int retval, oldprio, oldpolicy = -1, on_rq, running;
 	unsigned long flags;
@@ -4787,7 +4779,7 @@ recheck:
 	/*
 	 * Allow unprivileged RT tasks to decrease priority:
 	 */
-	if (!capable(CAP_SYS_NICE)) {
+	if (user && !capable(CAP_SYS_NICE)) {
 		if (rt_policy(policy)) {
 			unsigned long rlim_rtprio;
 
@@ -4823,7 +4815,8 @@ recheck:
 	 * Do not allow realtime tasks into groups that have no runtime
 	 * assigned.
 	 */
-	if (rt_policy(policy) && task_group(p)->rt_bandwidth.rt_runtime == 0)
+	if (user
+	    && rt_policy(policy) && task_group(p)->rt_bandwidth.rt_runtime == 0)
 		return -EPERM;
 #endif
 
@@ -4872,8 +4865,39 @@ recheck:
 
 	return 0;
 }
+
+/**
+ * sched_setscheduler - change the scheduling policy and/or RT priority of a thread.
+ * @p: the task in question.
+ * @policy: new policy.
+ * @param: structure containing the new RT priority.
+ *
+ * NOTE that the task may be already dead.
+ */
+int sched_setscheduler(struct task_struct *p, int policy,
+		       struct sched_param *param)
+{
+	return __sched_setscheduler(p, policy, param, true);
+}
 EXPORT_SYMBOL_GPL(sched_setscheduler);
 
+/**
+ * sched_setscheduler_nocheck - change the scheduling policy and/or RT priority of a thread from kernelspace.
+ * @p: the task in question.
+ * @policy: new policy.
+ * @param: structure containing the new RT priority.
+ *
+ * Just like sched_setscheduler, only don't bother checking if the
+ * current context has permission.  For example, this is needed in
+ * stop_machine(): we create temporary high priority worker threads,
+ * but our caller might not have that capability.
+ */
+int sched_setscheduler_nocheck(struct task_struct *p, int policy,
+			       struct sched_param *param)
+{
+	return __sched_setscheduler(p, policy, param, false);
+}
+
 static int
 do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param)
 {
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 36e061740047..afd9120c2fc4 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -645,7 +645,7 @@ static int __cpuinit cpu_callback(struct notifier_block *nfb,
 
 		p = per_cpu(ksoftirqd, hotcpu);
 		per_cpu(ksoftirqd, hotcpu) = NULL;
-		sched_setscheduler(p, SCHED_FIFO, &param);
+		sched_setscheduler_nocheck(p, SCHED_FIFO, &param);
 		kthread_stop(p);
 		takeover_tasklets(hotcpu);
 		break;
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index b7350bbfb076..ba9b2054ecbd 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -187,7 +187,7 @@ struct task_struct *__stop_machine_run(int (*fn)(void *), void *data,
 		struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
 
 		/* One high-prio thread per cpu.  We'll do this one. */
-		sched_setscheduler(p, SCHED_FIFO, &param);
+		sched_setscheduler_nocheck(p, SCHED_FIFO, &param);
 		kthread_bind(p, cpu);
 		wake_up_process(p);
 		wait_for_completion(&smdata.done);
-- 
cgit v1.2.3


From 34f80b04f325078ff21123579343d99756ad8d0e Mon Sep 17 00:00:00 2001
From: Eilon Greenstein <eilong@broadcom.com>
Date: Mon, 23 Jun 2008 20:33:01 -0700
Subject: bnx2x: Add support for BCM57711 HW

Supporting the 57711 and 57711E - refers to in the code as E1H. The
57710 is referred to as E1.

To support the new members in the family, the bnx2x structure was
divided to 3 parts: common, port and function. These changes caused some
rearrangement in the bnx2x.h file.

A set of accessories macros were added to make access to the bnx2x
structure more readable

Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/Kconfig         |    1 +
 drivers/net/bnx2x.h         |  739 ++++----
 drivers/net/bnx2x_fw_defs.h |  483 ++++--
 drivers/net/bnx2x_hsi.h     |  708 +++++---
 drivers/net/bnx2x_init.h    |   20 +-
 drivers/net/bnx2x_link.c    |   11 +-
 drivers/net/bnx2x_main.c    | 3919 +++++++++++++++++++++++++++----------------
 drivers/net/bnx2x_reg.h     |   66 +-
 include/linux/pci_ids.h     |    2 +
 9 files changed, 3774 insertions(+), 2175 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 0e0f6696ccac..287d0873c60d 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2600,6 +2600,7 @@ config BNX2X
 	tristate "Broadcom NetXtremeII 10Gb support"
 	depends on PCI
 	select ZLIB_INFLATE
+	select LIBCRC32C
 	help
 	  This driver supports Broadcom NetXtremeII 10 gigabit Ethernet cards.
 	  To compile this driver as a module, choose M here: the module
diff --git a/drivers/net/bnx2x.h b/drivers/net/bnx2x.h
index 0979ca0ae408..e08b9439a933 100644
--- a/drivers/net/bnx2x.h
+++ b/drivers/net/bnx2x.h
@@ -14,39 +14,46 @@
 #ifndef BNX2X_H
 #define BNX2X_H
 
+/* compilation time flags */
+
+/* define this to make the driver freeze on error to allow getting debug info
+ * (you will need to reboot afterwards) */
+/* #define BNX2X_STOP_ON_ERROR */
+
 /* error/debug prints */
 
-#define DRV_MODULE_NAME 	"bnx2x"
-#define PFX DRV_MODULE_NAME     ": "
+#define DRV_MODULE_NAME		"bnx2x"
+#define PFX DRV_MODULE_NAME	": "
 
 /* for messages that are currently off */
-#define BNX2X_MSG_OFF   		0
-#define BNX2X_MSG_MCP   		0x10000 /* was: NETIF_MSG_HW */
-#define BNX2X_MSG_STATS 		0x20000 /* was: NETIF_MSG_TIMER */
-#define NETIF_MSG_NVM   		0x40000 /* was: NETIF_MSG_HW */
-#define NETIF_MSG_DMAE  		0x80000 /* was: NETIF_MSG_HW */
+#define BNX2X_MSG_OFF			0
+#define BNX2X_MSG_MCP			0x010000 /* was: NETIF_MSG_HW */
+#define BNX2X_MSG_STATS			0x020000 /* was: NETIF_MSG_TIMER */
+#define BNX2X_MSG_NVM			0x040000 /* was: NETIF_MSG_HW */
+#define BNX2X_MSG_DMAE			0x080000 /* was: NETIF_MSG_HW */
 #define BNX2X_MSG_SP			0x100000 /* was: NETIF_MSG_INTR */
 #define BNX2X_MSG_FP			0x200000 /* was: NETIF_MSG_INTR */
 
-#define DP_LEVEL			KERN_NOTICE     /* was: KERN_DEBUG */
+#define DP_LEVEL			KERN_NOTICE	/* was: KERN_DEBUG */
 
 /* regular debug print */
 #define DP(__mask, __fmt, __args...) do { \
 	if (bp->msglevel & (__mask)) \
-		printk(DP_LEVEL "[%s:%d(%s)]" __fmt, __FUNCTION__, \
-		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+		printk(DP_LEVEL "[%s:%d(%s)]" __fmt, __func__, __LINE__, \
+			bp->dev?(bp->dev->name):"?", ##__args); \
 	} while (0)
 
-/* for errors (never masked) */
-#define BNX2X_ERR(__fmt, __args...) do { \
-	printk(KERN_ERR "[%s:%d(%s)]" __fmt, __FUNCTION__, \
-		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+/* errors debug print */
+#define BNX2X_DBG_ERR(__fmt, __args...) do { \
+	if (bp->msglevel & NETIF_MSG_PROBE) \
+		printk(KERN_ERR "[%s:%d(%s)]" __fmt, __func__, __LINE__, \
+			bp->dev?(bp->dev->name):"?", ##__args); \
 	} while (0)
 
-/* for logging (never masked) */
-#define BNX2X_LOG(__fmt, __args...) do { \
-	printk(KERN_NOTICE "[%s:%d(%s)]" __fmt, __FUNCTION__, \
-		__LINE__, bp->dev?(bp->dev->name):"?", ##__args); \
+/* for errors (never masked) */
+#define BNX2X_ERR(__fmt, __args...) do { \
+	printk(KERN_ERR "[%s:%d(%s)]" __fmt, __func__, __LINE__, \
+		bp->dev?(bp->dev->name):"?", ##__args); \
 	} while (0)
 
 /* before we have a dev->name use dev_info() */
@@ -60,7 +67,7 @@
 #define bnx2x_panic() do { \
 		bp->panic = 1; \
 		BNX2X_ERR("driver assert\n"); \
-		bnx2x_disable_int(bp); \
+		bnx2x_int_disable(bp); \
 		bnx2x_panic_dump(bp); \
 	} while (0)
 #else
@@ -71,24 +78,29 @@
 #endif
 
 
-#define U64_LO(x)       		(((u64)x) & 0xffffffff)
-#define U64_HI(x)       		(((u64)x) >> 32)
-#define HILO_U64(hi, lo)		(((u64)hi << 32) + lo)
+#ifdef NETIF_F_HW_VLAN_TX
+#define BCM_VLAN			1
+#endif
+
 
+#define U64_LO(x)			(u32)(((u64)(x)) & 0xffffffff)
+#define U64_HI(x)			(u32)(((u64)(x)) >> 32)
+#define HILO_U64(hi, lo)		((((u64)(hi)) << 32) + (lo))
 
-#define REG_ADDR(bp, offset)    	(bp->regview + offset)
 
-#define REG_RD(bp, offset)      	readl(REG_ADDR(bp, offset))
-#define REG_RD8(bp, offset)     	readb(REG_ADDR(bp, offset))
-#define REG_RD64(bp, offset)    	readq(REG_ADDR(bp, offset))
+#define REG_ADDR(bp, offset)		(bp->regview + offset)
 
-#define REG_WR(bp, offset, val) 	writel((u32)val, REG_ADDR(bp, offset))
+#define REG_RD(bp, offset)		readl(REG_ADDR(bp, offset))
+#define REG_RD8(bp, offset)		readb(REG_ADDR(bp, offset))
+#define REG_RD64(bp, offset)		readq(REG_ADDR(bp, offset))
+
+#define REG_WR(bp, offset, val)		writel((u32)val, REG_ADDR(bp, offset))
 #define REG_WR8(bp, offset, val)	writeb((u8)val, REG_ADDR(bp, offset))
-#define REG_WR16(bp, offset, val)       writew((u16)val, REG_ADDR(bp, offset))
-#define REG_WR32(bp, offset, val)       REG_WR(bp, offset, val)
+#define REG_WR16(bp, offset, val)	writew((u16)val, REG_ADDR(bp, offset))
+#define REG_WR32(bp, offset, val)	REG_WR(bp, offset, val)
 
-#define REG_RD_IND(bp, offset)  	bnx2x_reg_rd_ind(bp, offset)
-#define REG_WR_IND(bp, offset, val)     bnx2x_reg_wr_ind(bp, offset, val)
+#define REG_RD_IND(bp, offset)		bnx2x_reg_rd_ind(bp, offset)
+#define REG_WR_IND(bp, offset, val)	bnx2x_reg_wr_ind(bp, offset, val)
 
 #define REG_RD_DMAE(bp, offset, valp, len32) \
 	do { \
@@ -96,28 +108,28 @@
 		memcpy(valp, bnx2x_sp(bp, wb_data[0]), len32 * 4); \
 	} while (0)
 
-#define REG_WR_DMAE(bp, offset, val, len32) \
+#define REG_WR_DMAE(bp, offset, valp, len32) \
 	do { \
-		memcpy(bnx2x_sp(bp, wb_data[0]), val, len32 * 4); \
+		memcpy(bnx2x_sp(bp, wb_data[0]), valp, len32 * 4); \
 		bnx2x_write_dmae(bp, bnx2x_sp_mapping(bp, wb_data), \
 				 offset, len32); \
 	} while (0)
 
-#define SHMEM_RD(bp, type) \
-	REG_RD(bp, bp->shmem_base + offsetof(struct shmem_region, type))
-#define SHMEM_WR(bp, type, val) \
-	REG_WR(bp, bp->shmem_base + offsetof(struct shmem_region, type), val)
+#define SHMEM_ADDR(bp, field)		(bp->common.shmem_base + \
+					 offsetof(struct shmem_region, field))
+#define SHMEM_RD(bp, field)		REG_RD(bp, SHMEM_ADDR(bp, field))
+#define SHMEM_WR(bp, field, val)	REG_WR(bp, SHMEM_ADDR(bp, field), val)
 
 #define NIG_WR(reg, val)	REG_WR(bp, reg, val)
-#define EMAC_WR(reg, val)       REG_WR(bp, emac_base + reg, val)
-#define BMAC_WR(reg, val)       REG_WR(bp, GRCBASE_NIG + bmac_addr + reg, val)
+#define EMAC_WR(reg, val)	REG_WR(bp, emac_base + reg, val)
+#define BMAC_WR(reg, val)	REG_WR(bp, GRCBASE_NIG + bmac_addr + reg, val)
 
 
-#define for_each_queue(bp, var) for (var = 0; var < bp->num_queues; var++)
+#define for_each_queue(bp, var)	for (var = 0; var < bp->num_queues; var++)
 
 #define for_each_nondefault_queue(bp, var) \
 				for (var = 1; var < bp->num_queues; var++)
-#define is_multi(bp)    	(bp->num_queues > 1)
+#define is_multi(bp)		(bp->num_queues > 1)
 
 
 struct regp {
@@ -358,210 +370,122 @@ struct bnx2x_eth_stats {
 	u32 number_of_bugs_found_in_stats_spec; /* just kidding */
 };
 
-#define MAC_STX_NA      		0xffffffff
-
-#ifdef BNX2X_MULTI
-#define MAX_CONTEXT     		16
-#else
-#define MAX_CONTEXT     		1
-#endif
-
-union cdu_context {
-	struct eth_context eth;
-	char pad[1024];
-};
-
-#define MAX_DMAE_C      		5
-
-/* DMA memory not used in fastpath */
-struct bnx2x_slowpath {
-	union cdu_context       	context[MAX_CONTEXT];
-	struct eth_stats_query  	fw_stats;
-	struct mac_configuration_cmd    mac_config;
-	struct mac_configuration_cmd    mcast_config;
-
-	/* used by dmae command executer */
-	struct dmae_command     	dmae[MAX_DMAE_C];
-
-	union mac_stats 		mac_stats;
-	struct nig_stats		nig;
-	struct bnx2x_eth_stats  	eth_stats;
-
-	u32     			wb_comp;
-#define BNX2X_WB_COMP_VAL       	0xe0d0d0ae
-	u32     			wb_data[4];
-};
-
-#define bnx2x_sp(bp, var)       	(&bp->slowpath->var)
 #define bnx2x_sp_check(bp, var) ((bp->slowpath) ? (&bp->slowpath->var) : NULL)
-#define bnx2x_sp_mapping(bp, var) \
-		(bp->slowpath_mapping + offsetof(struct bnx2x_slowpath, var))
-
-
 struct sw_rx_bd {
-	struct sk_buff  *skb;
+	struct sk_buff	*skb;
 	DECLARE_PCI_UNMAP_ADDR(mapping)
 };
 
 struct sw_tx_bd {
-	struct sk_buff  *skb;
-	u16     	first_bd;
+	struct sk_buff	*skb;
+	u16		first_bd;
 };
 
 struct bnx2x_fastpath {
 
-	struct napi_struct      napi;
+	struct napi_struct	napi;
 
 	struct host_status_block *status_blk;
-	dma_addr_t      	status_blk_mapping;
+	dma_addr_t		status_blk_mapping;
 
-	struct eth_tx_db_data   *hw_tx_prods;
-	dma_addr_t      	tx_prods_mapping;
+	struct eth_tx_db_data	*hw_tx_prods;
+	dma_addr_t		tx_prods_mapping;
 
-	struct sw_tx_bd 	*tx_buf_ring;
+	struct sw_tx_bd		*tx_buf_ring;
 
 	struct eth_tx_bd	*tx_desc_ring;
-	dma_addr_t      	tx_desc_mapping;
+	dma_addr_t		tx_desc_mapping;
 
 	struct sw_rx_bd 	*rx_buf_ring;
 
 	struct eth_rx_bd	*rx_desc_ring;
-	dma_addr_t      	rx_desc_mapping;
+	dma_addr_t		rx_desc_mapping;
 
 	union eth_rx_cqe	*rx_comp_ring;
-	dma_addr_t      	rx_comp_mapping;
-
-	int     		state;
-#define BNX2X_FP_STATE_CLOSED   	0
-#define BNX2X_FP_STATE_IRQ      	0x80000
-#define BNX2X_FP_STATE_OPENING  	0x90000
-#define BNX2X_FP_STATE_OPEN     	0xa0000
-#define BNX2X_FP_STATE_HALTING  	0xb0000
-#define BNX2X_FP_STATE_HALTED   	0xc0000
-
-	int     		index;
-
-	u16     		tx_pkt_prod;
-	u16     		tx_pkt_cons;
-	u16     		tx_bd_prod;
-	u16     		tx_bd_cons;
-	u16     		*tx_cons_sb;
-
-	u16     		fp_c_idx;
-	u16     		fp_u_idx;
-
-	u16     		rx_bd_prod;
-	u16     		rx_bd_cons;
-	u16     		rx_comp_prod;
-	u16     		rx_comp_cons;
-	u16     		*rx_cons_sb;
-
-	unsigned long   	tx_pkt,
+	dma_addr_t		rx_comp_mapping;
+
+	int			state;
+#define BNX2X_FP_STATE_CLOSED		0
+#define BNX2X_FP_STATE_IRQ		0x80000
+#define BNX2X_FP_STATE_OPENING		0x90000
+#define BNX2X_FP_STATE_OPEN		0xa0000
+#define BNX2X_FP_STATE_HALTING		0xb0000
+#define BNX2X_FP_STATE_HALTED		0xc0000
+
+	u8			index;	/* number in fp array */
+	u8			cl_id;	/* eth client id */
+	u8			sb_id;	/* status block number in HW */
+#define FP_IDX(fp)			(fp->index)
+#define FP_CL_ID(fp)			(fp->cl_id)
+#define BP_CL_ID(bp)			(bp->fp[0].cl_id)
+#define FP_SB_ID(fp)			(fp->sb_id)
+#define CNIC_SB_ID			0
+
+	u16			tx_pkt_prod;
+	u16			tx_pkt_cons;
+	u16			tx_bd_prod;
+	u16			tx_bd_cons;
+	u16			*tx_cons_sb;
+
+	u16			fp_c_idx;
+	u16			fp_u_idx;
+
+	u16			rx_bd_prod;
+	u16			rx_bd_cons;
+	u16			rx_comp_prod;
+	u16			rx_comp_cons;
+	u16			*rx_cons_sb;
+
+	unsigned long		tx_pkt,
 				rx_pkt,
 				rx_calls;
 
-	struct bnx2x    	*bp; /* parent */
-};
-
-#define bnx2x_fp(bp, nr, var)   	(bp->fp[nr].var)
-
-
-/* attn group wiring */
-#define MAX_DYNAMIC_ATTN_GRPS   	8
-
-struct attn_route {
-	u32     sig[4];
+	struct bnx2x		*bp; /* parent */
 };
 
-struct bnx2x {
-	/* Fields used in the tx and intr/napi performance paths
-	 * are grouped together in the beginning of the structure
-	 */
-	struct bnx2x_fastpath   *fp;
-	void __iomem    	*regview;
-	void __iomem    	*doorbells;
-
-	struct net_device       *dev;
-	struct pci_dev  	*pdev;
-
-	atomic_t		intr_sem;
-	struct msix_entry       msix_table[MAX_CONTEXT+1];
-
-	int     		tx_ring_size;
+#define bnx2x_fp(bp, nr, var)		(bp->fp[nr].var)
+/* This is needed for determening of last_max */
+#define SUB_S16(a, b)			(s16)((s16)(a) - (s16)(b))
 
-#ifdef BCM_VLAN
-	struct vlan_group       *vlgrp;
-#endif
-
-	u32     		rx_csum;
-	u32     		rx_offset;
-	u32     		rx_buf_use_size;	/* useable size */
-	u32     		rx_buf_size;    	/* with alignment */
-#define ETH_OVREHEAD    		(ETH_HLEN + 8)  /* 8 for CRC + VLAN */
-#define ETH_MIN_PACKET_SIZE     	60
-#define ETH_MAX_PACKET_SIZE     	1500
-#define ETH_MAX_JUMBO_PACKET_SIZE       9600
+/* stuff added to make the code fit 80Col */
 
-	struct host_def_status_block *def_status_blk;
-#define DEF_SB_ID       	16
-	u16     		def_c_idx;
-	u16     		def_u_idx;
-	u16     		def_t_idx;
-	u16     		def_x_idx;
-	u16     		def_att_idx;
-	u32     		attn_state;
-	struct attn_route       attn_group[MAX_DYNAMIC_ATTN_GRPS];
-	u32     		aeu_mask;
-	u32     		nig_mask;
+#define CQE_TYPE(cqe_fp_flags)	((cqe_fp_flags) & ETH_FAST_PATH_RX_CQE_TYPE)
 
-	/* slow path ring */
-	struct eth_spe  	*spq;
-	dma_addr_t      	spq_mapping;
-	u16     		spq_prod_idx;
-	struct eth_spe  	*spq_prod_bd;
-	struct eth_spe  	*spq_last_bd;
-	u16     		*dsb_sp_prod;
-	u16     		spq_left; /* serialize spq */
-	spinlock_t      	spq_lock;
-
-	/* Flag for marking that there is either
-	 * STAT_QUERY or CFC DELETE ramrod pending
-	 */
-	u8      		stat_pending;
+#define ETH_RX_ERROR_FALGS	(ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG | \
+				 ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG | \
+				 ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG)
 
-	/* End of fields used in the performance code paths */
 
-	int     		panic;
-	int     		msglevel;
+#define U_SB_ETH_RX_CQ_INDEX		HC_INDEX_U_ETH_RX_CQ_CONS
+#define U_SB_ETH_RX_BD_INDEX		HC_INDEX_U_ETH_RX_BD_CONS
+#define C_SB_ETH_TX_CQ_INDEX		HC_INDEX_C_ETH_TX_CQ_CONS
 
-	u32     		flags;
-#define PCIX_FLAG       		1
-#define PCI_32BIT_FLAG  		2
-#define ONE_TDMA_FLAG   		4       /* no longer used */
-#define NO_WOL_FLAG     		8
-#define USING_DAC_FLAG  		0x10
-#define USING_MSIX_FLAG 		0x20
-#define ASF_ENABLE_FLAG 		0x40
+#define BNX2X_RX_SB_INDEX \
+	(&fp->status_blk->u_status_block.index_values[U_SB_ETH_RX_CQ_INDEX])
 
-	int     		port;
+#define BNX2X_RX_SB_BD_INDEX \
+	(&fp->status_blk->u_status_block.index_values[U_SB_ETH_RX_BD_INDEX])
 
-	int     		pm_cap;
-	int     		pcie_cap;
+#define BNX2X_RX_SB_INDEX_NUM \
+		(((U_SB_ETH_RX_CQ_INDEX << \
+		   USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER_SHIFT) & \
+		  USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER) | \
+		 ((U_SB_ETH_RX_BD_INDEX << \
+		   USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER_SHIFT) & \
+		  USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER))
 
-	struct work_struct	sp_task;
-	struct work_struct	reset_task;
+#define BNX2X_TX_SB_INDEX \
+	(&fp->status_blk->c_status_block.index_values[C_SB_ETH_TX_CQ_INDEX])
 
-	struct timer_list       timer;
-	int     		timer_interval;
-	int     		current_interval;
+/* common */
 
-	u32     		shmem_base;
+struct bnx2x_common {
 
 	u32			chip_id;
 /* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
-#define CHIP_ID(bp)			(bp->chip_id & 0xfffffff0)
+#define CHIP_ID(bp)			(bp->common.chip_id & 0xfffffff0)
 
-#define CHIP_NUM(bp)			(bp->chip_id >> 16)
+#define CHIP_NUM(bp)			(bp->common.chip_id >> 16)
 #define CHIP_NUM_57710			0x164e
 #define CHIP_NUM_57711			0x164f
 #define CHIP_NUM_57711E			0x1650
@@ -572,7 +496,7 @@ struct bnx2x {
 					 CHIP_IS_57711E(bp))
 #define IS_E1H_OFFSET			CHIP_IS_E1H(bp)
 
-#define CHIP_REV(bp)			(bp->chip_id & 0x0000f000)
+#define CHIP_REV(bp)			(bp->common.chip_id & 0x0000f000)
 #define CHIP_REV_Ax			0x00000000
 /* assume maximum 5 revisions */
 #define CHIP_REV_IS_SLOW(bp)		(CHIP_REV(bp) > 0x00005000)
@@ -586,86 +510,250 @@ struct bnx2x {
 #define CHIP_TIME(bp)			((CHIP_REV_IS_EMUL(bp)) ? 2000 : \
 					((CHIP_REV_IS_FPGA(bp)) ? 200 : 1))
 
-#define CHIP_METAL(bp)			(bp->chip_id & 0x00000ff0)
-#define CHIP_BOND_ID(bp)		(bp->chip_id & 0x0000000f)
+#define CHIP_METAL(bp)			(bp->common.chip_id & 0x00000ff0)
+#define CHIP_BOND_ID(bp)		(bp->common.chip_id & 0x0000000f)
 
-	u16     		fw_seq;
-	u16     		fw_drv_pulse_wr_seq;
-	u32     		fw_mb;
+	int			flash_size;
+#define NVRAM_1MB_SIZE			0x20000	/* 1M bit in bytes */
+#define NVRAM_TIMEOUT_COUNT		30000
+#define NVRAM_PAGE_SIZE			256
 
-	u32     		hw_config;
+	u32			shmem_base;
+
+	u32			hw_config;
 	u32			board;
 
-	struct link_params	link_params;
+	u32			bc_ver;
+
+	char			*name;
+};
 
-	struct link_vars	link_vars;
+
+/* end of common */
+
+/* port */
+
+struct bnx2x_port {
+	u32			pmf;
 
 	u32			link_config;
 
-	u32     		supported;
+	u32			supported;
+/* link settings - missing defines */
+#define SUPPORTED_2500baseX_Full	(1 << 15)
+
+	u32			advertising;
 /* link settings - missing defines */
-#define SUPPORTED_2500baseT_Full	(1 << 15)
+#define ADVERTISED_2500baseX_Full	(1 << 15)
 
-	u32     		phy_addr;
+	u32			phy_addr;
 
 	/* used to synchronize phy accesses */
 	struct mutex		phy_mutex;
 
-	u32     		phy_id;
+	u32			port_stx;
 
+	struct nig_stats	old_nig_stats;
+};
 
-	u32     		advertising;
-/* link settings - missing defines */
-#define ADVERTISED_2500baseT_Full       (1 << 15)
+/* end of port */
+
+#define MAC_STX_NA      		0xffffffff
+
+#ifdef BNX2X_MULTI
+#define MAX_CONTEXT			16
+#else
+#define MAX_CONTEXT			1
+#endif
+
+union cdu_context {
+	struct eth_context eth;
+	char pad[1024];
+};
+
+#define MAX_DMAE_C			6
+
+/* DMA memory not used in fastpath */
+struct bnx2x_slowpath {
+	union cdu_context		context[MAX_CONTEXT];
+	struct eth_stats_query		fw_stats;
+	struct mac_configuration_cmd	mac_config;
+	struct mac_configuration_cmd	mcast_config;
+
+	/* used by dmae command executer */
+	struct dmae_command		dmae[MAX_DMAE_C];
+
+	union mac_stats 		mac_stats;
+	struct nig_stats		nig;
+	struct bnx2x_eth_stats  	eth_stats;
+
+	u32				wb_comp;
+#define BNX2X_WB_COMP_VAL       	0xe0d0d0ae
+	u32				wb_data[4];
+};
+
+#define bnx2x_sp(bp, var)		(&bp->slowpath->var)
+#define bnx2x_sp_mapping(bp, var) \
+		(bp->slowpath_mapping + offsetof(struct bnx2x_slowpath, var))
+
+
+/* attn group wiring */
+#define MAX_DYNAMIC_ATTN_GRPS		8
+
+struct attn_route {
+	u32	sig[4];
+};
+
+struct bnx2x {
+	/* Fields used in the tx and intr/napi performance paths
+	 * are grouped together in the beginning of the structure
+	 */
+	struct bnx2x_fastpath	fp[MAX_CONTEXT];
+	void __iomem		*regview;
+	void __iomem		*doorbells;
+#define BNX2X_DB_SIZE		(16*2048)
+
+	struct net_device	*dev;
+	struct pci_dev		*pdev;
+
+	atomic_t		intr_sem;
+	struct msix_entry       msix_table[MAX_CONTEXT+1];
+
+	int			tx_ring_size;
+
+#ifdef BCM_VLAN
+	struct vlan_group	*vlgrp;
+#endif
 
+	u32			rx_csum;
+	u32			rx_offset;
+	u32			rx_buf_use_size;	/* useable size */
+	u32			rx_buf_size;		/* with alignment */
+#define ETH_OVREHEAD			(ETH_HLEN + 8)	/* 8 for CRC + VLAN */
+#define ETH_MIN_PACKET_SIZE		60
+#define ETH_MAX_PACKET_SIZE		1500
+#define ETH_MAX_JUMBO_PACKET_SIZE	9600
 
-	u32     		bc_ver;
+	struct host_def_status_block *def_status_blk;
+#define DEF_SB_ID			16
+	u16			def_c_idx;
+	u16			def_u_idx;
+	u16			def_x_idx;
+	u16			def_t_idx;
+	u16			def_att_idx;
+	u32			attn_state;
+	struct attn_route	attn_group[MAX_DYNAMIC_ATTN_GRPS];
+	u32			aeu_mask;
+	u32			nig_mask;
+
+	/* slow path ring */
+	struct eth_spe		*spq;
+	dma_addr_t		spq_mapping;
+	u16			spq_prod_idx;
+	struct eth_spe		*spq_prod_bd;
+	struct eth_spe		*spq_last_bd;
+	u16			*dsb_sp_prod;
+	u16			spq_left; /* serialize spq */
+	/* used to synchronize spq accesses */
+	spinlock_t		spq_lock;
+
+	/* Flag for marking that there is either
+	 * STAT_QUERY or CFC DELETE ramrod pending
+	 */
+	u8      		stat_pending;
+
+	/* End of fileds used in the performance code paths */
+
+	int			panic;
+	int			msglevel;
+
+	u32			flags;
+#define PCIX_FLAG			1
+#define PCI_32BIT_FLAG			2
+#define ONE_TDMA_FLAG			4	/* no longer used */
+#define NO_WOL_FLAG			8
+#define USING_DAC_FLAG			0x10
+#define USING_MSIX_FLAG			0x20
+#define ASF_ENABLE_FLAG			0x40
+#define NO_MCP_FLAG			0x100
+#define BP_NOMCP(bp)			(bp->flags & NO_MCP_FLAG)
+
+	int			func;
+#define BP_PORT(bp)			(bp->func % PORT_MAX)
+#define BP_FUNC(bp)			(bp->func)
+#define BP_E1HVN(bp)			(bp->func >> 1)
+#define BP_L_ID(bp)			(BP_E1HVN(bp) << 2)
+/* assorted E1HVN */
+#define IS_E1HMF(bp)			(bp->e1hmf != 0)
+#define BP_MAX_QUEUES(bp)		(IS_E1HMF(bp) ? 4 : 16)
+
+	int			pm_cap;
+	int			pcie_cap;
+
+	struct work_struct	sp_task;
+	struct work_struct	reset_task;
+
+	struct timer_list	timer;
+	int			timer_interval;
+	int			current_interval;
+
+	u16			fw_seq;
+	u16			fw_drv_pulse_wr_seq;
+	u32			func_stx;
+
+	struct link_params	link_params;
+	struct link_vars	link_vars;
 
-	int     		flash_size;
-#define NVRAM_1MB_SIZE  		0x20000 /* 1M bit in bytes */
-#define NVRAM_TIMEOUT_COUNT     	30000
-#define NVRAM_PAGE_SIZE 		256
+	struct bnx2x_common	common;
+	struct bnx2x_port	port;
+
+	u32			mf_config;
+	u16			e1hov;
+	u8			e1hmf;
 
 	u8			wol;
 
-	int     		rx_ring_size;
+	int			rx_ring_size;
 
-	u16     		tx_quick_cons_trip_int;
-	u16     		tx_quick_cons_trip;
-	u16     		tx_ticks_int;
-	u16     		tx_ticks;
+	u16			tx_quick_cons_trip_int;
+	u16			tx_quick_cons_trip;
+	u16			tx_ticks_int;
+	u16			tx_ticks;
 
-	u16     		rx_quick_cons_trip_int;
-	u16     		rx_quick_cons_trip;
-	u16     		rx_ticks_int;
-	u16     		rx_ticks;
+	u16			rx_quick_cons_trip_int;
+	u16			rx_quick_cons_trip;
+	u16			rx_ticks_int;
+	u16			rx_ticks;
 
-	u32     		stats_ticks;
+	u32			stats_ticks;
+	u32			lin_cnt;
 
-	int     		state;
-#define BNX2X_STATE_CLOSED      	0x0
-#define BNX2X_STATE_OPENING_WAIT4_LOAD  0x1000
-#define BNX2X_STATE_OPENING_WAIT4_PORT  0x2000
+	int			state;
+#define BNX2X_STATE_CLOSED		0x0
+#define BNX2X_STATE_OPENING_WAIT4_LOAD	0x1000
+#define BNX2X_STATE_OPENING_WAIT4_PORT	0x2000
 #define BNX2X_STATE_OPEN		0x3000
-#define BNX2X_STATE_CLOSING_WAIT4_HALT  0x4000
+#define BNX2X_STATE_CLOSING_WAIT4_HALT	0x4000
 #define BNX2X_STATE_CLOSING_WAIT4_DELETE 0x5000
 #define BNX2X_STATE_CLOSING_WAIT4_UNLOAD 0x6000
-#define BNX2X_STATE_ERROR       	0xF000
+#define BNX2X_STATE_DISABLED		0xd000
+#define BNX2X_STATE_DIAG		0xe000
+#define BNX2X_STATE_ERROR		0xf000
 
-	int     		num_queues;
+	int			num_queues;
 
-	u32     		rx_mode;
-#define BNX2X_RX_MODE_NONE      	0
-#define BNX2X_RX_MODE_NORMAL    	1
-#define BNX2X_RX_MODE_ALLMULTI  	2
-#define BNX2X_RX_MODE_PROMISC   	3
-#define BNX2X_MAX_MULTICAST     	64
-#define BNX2X_MAX_EMUL_MULTI    	16
+	u32			rx_mode;
+#define BNX2X_RX_MODE_NONE		0
+#define BNX2X_RX_MODE_NORMAL		1
+#define BNX2X_RX_MODE_ALLMULTI		2
+#define BNX2X_RX_MODE_PROMISC		3
+#define BNX2X_MAX_MULTICAST		64
+#define BNX2X_MAX_EMUL_MULTI		16
 
-	dma_addr_t      	def_status_blk_mapping;
+	dma_addr_t		def_status_blk_mapping;
 
-	struct bnx2x_slowpath   *slowpath;
-	dma_addr_t      	slowpath_mapping;
+	struct bnx2x_slowpath	*slowpath;
+	dma_addr_t		slowpath_mapping;
 
 #ifdef BCM_ISCSI
 	void    		*t1;
@@ -742,8 +830,10 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 
 /* MC hsi */
 #define RX_COPY_THRESH  		92
-#define BCM_PAGE_BITS   		12
-#define BCM_PAGE_SIZE   		(1 << BCM_PAGE_BITS)
+#define BCM_PAGE_SHIFT			12
+#define BCM_PAGE_SIZE			(1 << BCM_PAGE_SHIFT)
+#define BCM_PAGE_MASK			(~(BCM_PAGE_SIZE - 1))
+#define BCM_PAGE_ALIGN(addr)	(((addr) + BCM_PAGE_SIZE - 1) & BCM_PAGE_MASK)
 
 #define NUM_TX_RINGS    		16
 #define TX_DESC_CNT     	(BCM_PAGE_SIZE / sizeof(struct eth_tx_bd))
@@ -795,26 +885,11 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 
 
 /* must be used on a CID before placing it on a HW ring */
-#define HW_CID(bp, x)   		(x | (bp->port << 23))
+#define HW_CID(bp, x)		((BP_PORT(bp) << 23) | (BP_E1HVN(bp) << 17) | x)
 
 #define SP_DESC_CNT     	(BCM_PAGE_SIZE / sizeof(struct eth_spe))
 #define MAX_SP_DESC_CNT 		(SP_DESC_CNT - 1)
 
-#define ATTN_NIG_FOR_FUNC       	(1L << 8)
-#define ATTN_SW_TIMER_4_FUNC    	(1L << 9)
-#define GPIO_2_FUNC     		(1L << 10)
-#define GPIO_3_FUNC     		(1L << 11)
-#define GPIO_4_FUNC     		(1L << 12)
-#define ATTN_GENERAL_ATTN_1     	(1L << 13)
-#define ATTN_GENERAL_ATTN_2     	(1L << 14)
-#define ATTN_GENERAL_ATTN_3     	(1L << 15)
-#define ATTN_GENERAL_ATTN_4     	(1L << 13)
-#define ATTN_GENERAL_ATTN_5     	(1L << 14)
-#define ATTN_GENERAL_ATTN_6     	(1L << 15)
-
-#define ATTN_HARD_WIRED_MASK    	0xff00
-#define ATTENTION_ID    		4
-
 
 #define BNX2X_BTR       		3
 #define MAX_SPQ_PENDING 		8
@@ -831,6 +906,31 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 		       DPM_TRIGER_TYPE); \
 	} while (0)
 
+static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms,
+			   int wait)
+{
+	u32 val;
+
+	do {
+		val = REG_RD(bp, reg);
+		if (val == expected)
+			break;
+		ms -= wait;
+		msleep(wait);
+
+	} while (ms > 0);
+
+	return val;
+}
+
+
+/* load/unload mode */
+#define LOAD_NORMAL			0
+#define LOAD_OPEN			1
+#define LOAD_DIAG			2
+#define UNLOAD_NORMAL			0
+#define UNLOAD_CLOSE			1
+
 /* DMAE command defines */
 #define DMAE_CMD_SRC_PCI		0
 #define DMAE_CMD_SRC_GRC		DMAE_COMMAND_SRC
@@ -877,23 +977,48 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 
 #define pbd_tcp_flags(skb)  	(ntohl(tcp_flag_word(tcp_hdr(skb)))>>16 & 0xff)
 
-/* stuff added to make the code fit 80Col */
+/* must be used on a CID before placing it on a HW ring */
 
-#define TPA_TYPE_START  		ETH_FAST_PATH_RX_CQE_START_FLG
-#define TPA_TYPE_END    		ETH_FAST_PATH_RX_CQE_END_FLG
-#define TPA_TYPE(cqe)   	(cqe->fast_path_cqe.error_type_flags & \
-				 (TPA_TYPE_START | TPA_TYPE_END))
 #define BNX2X_RX_SUM_OK(cqe) \
 			(!(cqe->fast_path_cqe.status_flags & \
 			 (ETH_FAST_PATH_RX_CQE_IP_XSUM_NO_VALIDATION_FLG | \
 			  ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG)))
 
-#define BNX2X_RX_SUM_FIX(cqe) \
-			((le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags) & \
-			  PARSING_FLAGS_OVER_ETHERNET_PROTOCOL) == \
-			 (1 << PARSING_FLAGS_OVER_ETHERNET_PROTOCOL_SHIFT))
+/* CMNG constants
+   derived from lab experiments, and not from system spec calculations !!! */
+#define DEF_MIN_RATE			100
+/* resolution of the rate shaping timer - 100 usec */
+#define RS_PERIODIC_TIMEOUT_USEC	100
+/* resolution of fairness algorithm in usecs -
+   coefficient for clauclating the actuall t fair */
+#define T_FAIR_COEF			10000000
+/* number of bytes in single QM arbitration cycle -
+   coeffiecnt for calculating the fairness timer */
+#define QM_ARB_BYTES			40000
+#define FAIR_MEM			2
+
+
+#define ATTN_NIG_FOR_FUNC		(1L << 8)
+#define ATTN_SW_TIMER_4_FUNC		(1L << 9)
+#define GPIO_2_FUNC			(1L << 10)
+#define GPIO_3_FUNC			(1L << 11)
+#define GPIO_4_FUNC			(1L << 12)
+#define ATTN_GENERAL_ATTN_1		(1L << 13)
+#define ATTN_GENERAL_ATTN_2		(1L << 14)
+#define ATTN_GENERAL_ATTN_3		(1L << 15)
+#define ATTN_GENERAL_ATTN_4		(1L << 13)
+#define ATTN_GENERAL_ATTN_5		(1L << 14)
+#define ATTN_GENERAL_ATTN_6		(1L << 15)
+
+#define ATTN_HARD_WIRED_MASK		0xff00
+#define ATTENTION_ID			4
 
 
+/* stuff added to make the code fit 80Col */
+
+#define BNX2X_PMF_LINK_ASSERT \
+	GENERAL_ATTEN_OFFSET(LINK_SYNC_ATTENTION_BIT_FUNC_0 + BP_FUNC(bp))
+
 #define BNX2X_MC_ASSERT_BITS \
 	(GENERAL_ATTEN_OFFSET(TSTORM_FATAL_ASSERT_ATTENTION_BIT) | \
 	 GENERAL_ATTEN_OFFSET(USTORM_FATAL_ASSERT_ATTENTION_BIT) | \
@@ -906,12 +1031,20 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 #define BNX2X_DOORQ_ASSERT \
 	AEU_INPUTS_ATTN_BITS_DOORBELLQ_HW_INTERRUPT
 
+#define BNX2X_GRC_TIMEOUT	GENERAL_ATTEN_OFFSET(LATCHED_ATTN_TIMEOUT_GRC)
+#define BNX2X_GRC_RSV		(GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCR) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCT) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCN) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCU) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RBCP) | \
+				 GENERAL_ATTEN_OFFSET(LATCHED_ATTN_RSVD_GRC))
+
 #define HW_INTERRUT_ASSERT_SET_0 \
 				(AEU_INPUTS_ATTN_BITS_TSDM_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_TCM_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_TSEMI_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_PBF_HW_INTERRUPT)
-#define HW_PRTY_ASSERT_SET_0    (AEU_INPUTS_ATTN_BITS_BRB_PARITY_ERROR | \
+#define HW_PRTY_ASSERT_SET_0	(AEU_INPUTS_ATTN_BITS_BRB_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_PARSER_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_TSDM_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_SEARCHER_PARITY_ERROR |\
@@ -928,7 +1061,7 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 				 AEU_INPUTS_ATTN_BITS_UPB_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_CSDM_HW_INTERRUPT | \
 				 AEU_INPUTS_ATTN_BITS_CCM_HW_INTERRUPT)
-#define HW_PRTY_ASSERT_SET_1    (AEU_INPUTS_ATTN_BITS_PBCLIENT_PARITY_ERROR |\
+#define HW_PRTY_ASSERT_SET_1	(AEU_INPUTS_ATTN_BITS_PBCLIENT_PARITY_ERROR |\
 				 AEU_INPUTS_ATTN_BITS_QM_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_XSDM_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_XSEMI_PARITY_ERROR | \
@@ -945,7 +1078,7 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 				 AEU_INPUTS_ATTN_BITS_DMAE_HW_INTERRUPT | \
 			AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_HW_INTERRUPT |\
 				 AEU_INPUTS_ATTN_BITS_MISC_HW_INTERRUPT)
-#define HW_PRTY_ASSERT_SET_2    (AEU_INPUTS_ATTN_BITS_CSEMI_PARITY_ERROR | \
+#define HW_PRTY_ASSERT_SET_2	(AEU_INPUTS_ATTN_BITS_CSEMI_PARITY_ERROR | \
 				 AEU_INPUTS_ATTN_BITS_PXP_PARITY_ERROR | \
 			AEU_INPUTS_ATTN_BITS_PXPPCICLOCKCLIENT_PARITY_ERROR |\
 				 AEU_INPUTS_ATTN_BITS_CFC_PARITY_ERROR | \
@@ -954,42 +1087,44 @@ int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode);
 				 AEU_INPUTS_ATTN_BITS_MISC_PARITY_ERROR)
 
 
-#define ETH_RX_ERROR_FALGS      (ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG | \
-				 ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG | \
-				 ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG)
-
-
 #define MULTI_FLAGS \
-	(TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY     | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY     | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY | \
-	 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE)
+		(TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY | \
+		 TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE)
 
-#define MULTI_MASK      0x7f
+#define MULTI_MASK			0x7f
 
 
-#define U_SB_ETH_RX_CQ_INDEX    	HC_INDEX_U_ETH_RX_CQ_CONS
-#define C_SB_ETH_TX_CQ_INDEX    	HC_INDEX_C_ETH_TX_CQ_CONS
-#define C_DEF_SB_SP_INDEX       	HC_INDEX_DEF_C_ETH_SLOW_PATH
-
-#define BNX2X_RX_SB_INDEX \
-	&fp->status_blk->u_status_block.index_values[U_SB_ETH_RX_CQ_INDEX]
+#define DEF_USB_FUNC_OFF		(2 + 2*HC_USTORM_DEF_SB_NUM_INDICES)
+#define DEF_CSB_FUNC_OFF		(2 + 2*HC_CSTORM_DEF_SB_NUM_INDICES)
+#define DEF_XSB_FUNC_OFF		(2 + 2*HC_XSTORM_DEF_SB_NUM_INDICES)
+#define DEF_TSB_FUNC_OFF		(2 + 2*HC_TSTORM_DEF_SB_NUM_INDICES)
 
-#define BNX2X_TX_SB_INDEX \
-	&fp->status_blk->c_status_block.index_values[C_SB_ETH_TX_CQ_INDEX]
+#define C_DEF_SB_SP_INDEX		HC_INDEX_DEF_C_ETH_SLOW_PATH
 
 #define BNX2X_SP_DSB_INDEX \
-&bp->def_status_blk->c_def_status_block.index_values[C_DEF_SB_SP_INDEX]
+(&bp->def_status_blk->c_def_status_block.index_values[C_DEF_SB_SP_INDEX])
 
 
 #define CAM_IS_INVALID(x) \
 (x.target_table_entry.flags == TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE)
 
 #define CAM_INVALIDATE(x) \
-x.target_table_entry.flags = TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE
+	(x.target_table_entry.flags = TSTORM_CAM_TARGET_TABLE_ENTRY_ACTION_TYPE)
+
+
+/* Number of u32 elements in MC hash array */
+#define MC_HASH_SIZE			8
+#define MC_HASH_OFFSET(bp, i)		(BAR_TSTRORM_INTMEM + \
+	TSTORM_APPROXIMATE_MATCH_MULTICAST_FILTERING_OFFSET(BP_FUNC(bp)) + i*4)
 
 
+#ifndef PXP2_REG_PXP2_INT_STS
+#define PXP2_REG_PXP2_INT_STS		PXP2_REG_PXP2_INT_STS_0
+#endif
+
 /* MISC_REG_RESET_REG - this is here for the hsi to work don't touch */
 
 #endif /* bnx2x.h */
diff --git a/drivers/net/bnx2x_fw_defs.h b/drivers/net/bnx2x_fw_defs.h
index 3b968904ca65..e3da7f69d27b 100644
--- a/drivers/net/bnx2x_fw_defs.h
+++ b/drivers/net/bnx2x_fw_defs.h
@@ -8,191 +8,390 @@
  */
 
 
-#define CSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x1922 + (port * 0x40) + (index * 0x4))
-#define CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1900 + (port * 0x40))
-#define CSTORM_HC_BTR_OFFSET(port)\
-	(0x1984 + (port * 0xc0))
-#define CSTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index)\
-	(0x141a + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define CSTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index)\
-	(0x1418 + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define CSTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id)\
-	(0x1400 + (port * 0x280) + (cpu_id * 0x28))
-#define CSTORM_STATS_FLAGS_OFFSET(port) 		(0x5108 + (port * 0x8))
-#define TSTORM_CLIENT_CONFIG_OFFSET(port, client_id)\
-	(0x1510 + (port * 0x240) + (client_id * 0x20))
-#define TSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x138a + (port * 0x28) + (index * 0x4))
-#define TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1370 + (port * 0x28))
-#define TSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port)\
-	(0x4b70 + (port * 0x8))
-#define TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(function)\
-	(0x1418 + (function * 0x30))
-#define TSTORM_HC_BTR_OFFSET(port)\
-	(0x13c4 + (port * 0x18))
-#define TSTORM_INDIRECTION_TABLE_OFFSET(port)\
-	(0x22c8 + (port * 0x80))
-#define TSTORM_INDIRECTION_TABLE_SIZE			0x80
-#define TSTORM_MAC_FILTER_CONFIG_OFFSET(port)\
-	(0x1420 + (port * 0x30))
-#define TSTORM_RCQ_PROD_OFFSET(port, client_id)\
-	(0x1508 + (port * 0x240) + (client_id * 0x20))
-#define TSTORM_STATS_FLAGS_OFFSET(port) 		(0x4b90 + (port * 0x8))
-#define USTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x191a + (port * 0x28) + (index * 0x4))
-#define USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1900 + (port * 0x28))
-#define USTORM_HC_BTR_OFFSET(port)\
-	(0x1954 + (port * 0xb8))
-#define USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port)\
-	(0x5408 + (port * 0x8))
-#define USTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index)\
-	(0x141a + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define USTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index)\
-	(0x1418 + (port * 0x280) + (cpu_id * 0x28) + (index * 0x4))
-#define USTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id)\
-	(0x1400 + (port * 0x280) + (cpu_id * 0x28))
-#define XSTORM_ASSERT_LIST_INDEX_OFFSET 		0x1000
-#define XSTORM_ASSERT_LIST_OFFSET(idx)			(0x1020 + (idx * 0x10))
-#define XSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index)\
-	(0x141a + (port * 0x28) + (index * 0x4))
-#define XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)\
-	(0x1400 + (port * 0x28))
-#define XSTORM_ETH_STATS_QUERY_ADDR_OFFSET(port)\
-	(0x5408 + (port * 0x8))
-#define XSTORM_HC_BTR_OFFSET(port)\
-	(0x1454 + (port * 0x18))
-#define XSTORM_SPQ_PAGE_BASE_OFFSET(port)\
-	(0x5328 + (port * 0x18))
-#define XSTORM_SPQ_PROD_OFFSET(port)\
-	(0x5330 + (port * 0x18))
-#define XSTORM_STATS_FLAGS_OFFSET(port) 		(0x53f8 + (port * 0x8))
+#define CSTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0x7000 : 0x1000)
+#define CSTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0x7020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define CSTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0x8522 + ((function>>1) * 0x40) + ((function&1) \
+	* 0x100) + (index * 0x4)) : (0x1922 + (function * 0x40) + (index \
+	* 0x4)))
+#define CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x8500 + ((function>>1) * 0x40) + ((function&1) \
+	* 0x100)) : (0x1900 + (function * 0x40)))
+#define CSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x8508 + ((function>>1) * 0x40) + ((function&1) \
+	* 0x100)) : (0x1908 + (function * 0x40)))
+#define CSTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x11e8 : 0xffffffff)
+#define CSTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0x8704 + (port * 0xf0)) : (0x1984 + (port * 0xc0)))
+#define CSTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x801a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x141a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define CSTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x8018 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x1418 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define CSTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x8000 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1400 + (port * 0x280) + (cpu_id * 0x28)))
+#define CSTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x8008 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1408 + (port * 0x280) + (cpu_id * 0x28)))
+#define CSTORM_STATS_FLAGS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x1108 + (function * 0x8)) : (0x5108 + \
+	(function * 0x8)))
+#define TSTORM_APPROXIMATE_MATCH_MULTICAST_FILTERING_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x31c0 + (function * 0x20)) : 0xffffffff)
+#define TSTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0xa000 : 0x1000)
+#define TSTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0xa020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define TSTORM_CLIENT_CONFIG_OFFSET(port, client_id) \
+	(IS_E1H_OFFSET? (0x3358 + (port * 0x3e8) + (client_id * 0x28)) : \
+	(0x9c8 + (port * 0x2f8) + (client_id * 0x28)))
+#define TSTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0xb01a + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0) + (index * 0x4)) : (0x141a + (function * 0x28) + (index * \
+	0x4)))
+#define TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xb000 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1400 + (function * 0x28)))
+#define TSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xb008 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1408 + (function * 0x28)))
+#define TSTORM_ETH_STATS_QUERY_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2b80 + (function * 0x8)) : (0x4b68 + \
+	(function * 0x8)))
+#define TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x3000 + (function * 0x38)) : (0x1500 + \
+	(function * 0x38)))
+#define TSTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x1ad0 : 0xffffffff)
+#define TSTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0xb144 + (port * 0x30)) : (0x1454 + (port * 0x18)))
+#define TSTORM_INDIRECTION_TABLE_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x12c8 + (function * 0x80)) : (0x22c8 + \
+	(function * 0x80)))
+#define TSTORM_INDIRECTION_TABLE_SIZE 0x80
+#define TSTORM_MAC_FILTER_CONFIG_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x3008 + (function * 0x38)) : (0x1508 + \
+	(function * 0x38)))
+#define TSTORM_RX_PRODS_OFFSET(port, client_id) \
+	(IS_E1H_OFFSET? (0x3350 + (port * 0x3e8) + (client_id * 0x28)) : \
+	(0x9c0 + (port * 0x2f8) + (client_id * 0x28)))
+#define TSTORM_STATS_FLAGS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2c00 + (function * 0x8)) : (0x4b88 + \
+	(function * 0x8)))
+#define TSTORM_TPA_EXIST_OFFSET (IS_E1H_OFFSET? 0x3b30 : 0x1c20)
+#define USTORM_AGG_DATA_OFFSET (IS_E1H_OFFSET? 0xa040 : 0x2c10)
+#define USTORM_AGG_DATA_SIZE (IS_E1H_OFFSET? 0x2440 : 0x1200)
+#define USTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0x8000 : 0x1000)
+#define USTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0x8020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define USTORM_CQE_PAGE_BASE_OFFSET(port, clientId) \
+	(IS_E1H_OFFSET? (0x3298 + (port * 0x258) + (clientId * 0x18)) : \
+	(0x5450 + (port * 0x1c8) + (clientId * 0x18)))
+#define USTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0x951a + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0) + (index * 0x4)) : (0x191a + (function * 0x28) + (index * \
+	0x4)))
+#define USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x9500 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1900 + (function * 0x28)))
+#define USTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x9508 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1908 + (function * 0x28)))
+#define USTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x2448 : 0xffffffff)
+#define USTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0x9644 + (port * 0xd0)) : (0x1954 + (port * 0xb8)))
+#define USTORM_MAX_AGG_SIZE_OFFSET(port, clientId) \
+	(IS_E1H_OFFSET? (0x3290 + (port * 0x258) + (clientId * 0x18)) : \
+	(0x5448 + (port * 0x1c8) + (clientId * 0x18)))
+#define USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2408 + (function * 0x8)) : (0x5408 + \
+	(function * 0x8)))
+#define USTORM_SB_HC_DISABLE_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x901a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x141a + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define USTORM_SB_HC_TIMEOUT_OFFSET(port, cpu_id, index) \
+	(IS_E1H_OFFSET? (0x9018 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)) : (0x1418 + (port * 0x280) + (cpu_id * 0x28) + \
+	(index * 0x4)))
+#define USTORM_SB_HOST_SB_ADDR_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x9000 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1400 + (port * 0x280) + (cpu_id * 0x28)))
+#define USTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, cpu_id) \
+	(IS_E1H_OFFSET? (0x9008 + (port * 0x280) + (cpu_id * 0x28)) : \
+	(0x1408 + (port * 0x280) + (cpu_id * 0x28)))
+#define XSTORM_ASSERT_LIST_INDEX_OFFSET \
+	(IS_E1H_OFFSET? 0x9000 : 0x1000)
+#define XSTORM_ASSERT_LIST_OFFSET(idx) \
+	(IS_E1H_OFFSET? (0x9020 + (idx * 0x10)) : (0x1020 + (idx * 0x10)))
+#define XSTORM_CMNG_PER_PORT_VARS_OFFSET(port) \
+	(IS_E1H_OFFSET? (0x24a8 + (port * 0x40)) : (0x3ba0 + (port * 0x40)))
+#define XSTORM_DEF_SB_HC_DISABLE_OFFSET(function, index) \
+	(IS_E1H_OFFSET? (0xa01a + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0) + (index * 0x4)) : (0x141a + (function * 0x28) + (index * \
+	0x4)))
+#define XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xa000 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1400 + (function * 0x28)))
+#define XSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(function) \
+	(IS_E1H_OFFSET? (0xa008 + ((function>>1) * 0x28) + ((function&1) \
+	* 0xa0)) : (0x1408 + (function * 0x28)))
+#define XSTORM_E1HOV_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2ab8 + (function * 0x2)) : 0xffffffff)
+#define XSTORM_ETH_STATS_QUERY_ADDR_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2418 + (function * 0x8)) : (0x3b70 + \
+	(function * 0x8)))
+#define XSTORM_FAIRNESS_PER_VN_VARS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2568 + (function * 0x70)) : (0x3c60 + \
+	(function * 0x70)))
+#define XSTORM_FUNCTION_MODE_OFFSET \
+	(IS_E1H_OFFSET? 0x2ac8 : 0xffffffff)
+#define XSTORM_HC_BTR_OFFSET(port) \
+	(IS_E1H_OFFSET? (0xa144 + (port * 0x30)) : (0x1454 + (port * 0x18)))
+#define XSTORM_RATE_SHAPING_PER_VN_VARS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2528 + (function * 0x70)) : (0x3c20 + \
+	(function * 0x70)))
+#define XSTORM_SPQ_PAGE_BASE_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2000 + (function * 0x10)) : (0x3328 + \
+	(function * 0x10)))
+#define XSTORM_SPQ_PROD_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x2008 + (function * 0x10)) : (0x3330 + \
+	(function * 0x10)))
+#define XSTORM_STATS_FLAGS_OFFSET(function) \
+	(IS_E1H_OFFSET? (0x23d8 + (function * 0x8)) : (0x3b60 + \
+	(function * 0x8)))
 #define COMMON_ASM_INVALID_ASSERT_OPCODE 0x0
 
 /**
 * This file defines HSI constatnts for the ETH flow
 */
-
-/* hash types */
-#define DEFAULT_HASH_TYPE			0
-#define IPV4_HASH_TYPE				1
-#define TCP_IPV4_HASH_TYPE			2
-#define IPV6_HASH_TYPE				3
-#define TCP_IPV6_HASH_TYPE			4
+#ifdef _EVEREST_MICROCODE
+#include "microcode_constants.h"
+#include "eth_rx_bd.h"
+#include "eth_tx_bd.h"
+#include "eth_rx_cqe.h"
+#include "eth_rx_sge.h"
+#include "eth_rx_cqe_next_page.h"
+#endif
+
+/* RSS hash types */
+#define DEFAULT_HASH_TYPE 0
+#define IPV4_HASH_TYPE 1
+#define TCP_IPV4_HASH_TYPE 2
+#define IPV6_HASH_TYPE 3
+#define TCP_IPV6_HASH_TYPE 4
+
+/* Ethernet Ring parmaters */
+#define X_ETH_LOCAL_RING_SIZE 13
+#define FIRST_BD_IN_PKT 0
+#define PARSE_BD_INDEX 1
+#define NUM_OF_ETH_BDS_IN_PAGE \
+	((PAGE_SIZE) / (STRUCT_SIZE(eth_tx_bd)/8))
+
+
+/* Rx ring params */
+#define U_ETH_LOCAL_BD_RING_SIZE (16)
+#define U_ETH_LOCAL_SGE_RING_SIZE (12)
+#define U_ETH_SGL_SIZE (8)
+
+
+#define U_ETH_BDS_PER_PAGE_MASK \
+	((PAGE_SIZE/(STRUCT_SIZE(eth_rx_bd)/8))-1)
+#define U_ETH_CQE_PER_PAGE_MASK \
+	((PAGE_SIZE/(STRUCT_SIZE(eth_rx_cqe)/8))-1)
+#define U_ETH_SGES_PER_PAGE_MASK \
+	((PAGE_SIZE/(STRUCT_SIZE(eth_rx_sge)/8))-1)
+
+#define U_ETH_SGES_PER_PAGE_INVERSE_MASK \
+	(0xFFFF - ((PAGE_SIZE/((STRUCT_SIZE(eth_rx_sge))/8))-1))
+
+
+#define TU_ETH_CQES_PER_PAGE \
+	(PAGE_SIZE/(STRUCT_SIZE(eth_rx_cqe_next_page)/8))
+#define U_ETH_BDS_PER_PAGE (PAGE_SIZE/(STRUCT_SIZE(eth_rx_bd)/8))
+#define U_ETH_SGES_PER_PAGE (PAGE_SIZE/(STRUCT_SIZE(eth_rx_sge)/8))
+
+#define U_ETH_UNDEFINED_Q 0xFF
 
 /* values of command IDs in the ramrod message */
-#define RAMROD_CMD_ID_ETH_PORT_SETUP			(80)
-#define RAMROD_CMD_ID_ETH_CLIENT_SETUP			(85)
-#define RAMROD_CMD_ID_ETH_STAT_QUERY			(90)
-#define RAMROD_CMD_ID_ETH_UPDATE			(100)
-#define RAMROD_CMD_ID_ETH_HALT				(105)
-#define RAMROD_CMD_ID_ETH_SET_MAC			(110)
-#define RAMROD_CMD_ID_ETH_CFC_DEL			(115)
-#define RAMROD_CMD_ID_ETH_PORT_DEL			(120)
-#define RAMROD_CMD_ID_ETH_FORWARD_SETUP 		(125)
+#define RAMROD_CMD_ID_ETH_PORT_SETUP (80)
+#define RAMROD_CMD_ID_ETH_CLIENT_SETUP (85)
+#define RAMROD_CMD_ID_ETH_STAT_QUERY (90)
+#define RAMROD_CMD_ID_ETH_UPDATE (100)
+#define RAMROD_CMD_ID_ETH_HALT (105)
+#define RAMROD_CMD_ID_ETH_SET_MAC (110)
+#define RAMROD_CMD_ID_ETH_CFC_DEL (115)
+#define RAMROD_CMD_ID_ETH_PORT_DEL (120)
+#define RAMROD_CMD_ID_ETH_FORWARD_SETUP (125)
 
 
 /* command values for set mac command */
-#define T_ETH_MAC_COMMAND_SET				0
-#define T_ETH_MAC_COMMAND_INVALIDATE			1
+#define T_ETH_MAC_COMMAND_SET 0
+#define T_ETH_MAC_COMMAND_INVALIDATE 1
+
+#define T_ETH_INDIRECTION_TABLE_SIZE 128
 
-#define T_ETH_INDIRECTION_TABLE_SIZE			128
+/*The CRC32 seed, that is used for the hash(reduction) multicast address */
+#define T_ETH_CRC32_HASH_SEED 0x00000000
 
 /* Maximal L2 clients supported */
-#define ETH_MAX_RX_CLIENTS				(18)
+#define ETH_MAX_RX_CLIENTS_E1 19
+#define ETH_MAX_RX_CLIENTS_E1H 25
+
+/* Maximal aggregation queues supported */
+#define ETH_MAX_AGGREGATION_QUEUES_E1 (32)
+#define ETH_MAX_AGGREGATION_QUEUES_E1H (64)
+
 
 /**
 * This file defines HSI constatnts common to all microcode flows
 */
 
 /* Connection types */
-#define ETH_CONNECTION_TYPE			0
+#define ETH_CONNECTION_TYPE 0
+#define TOE_CONNECTION_TYPE 1
+#define RDMA_CONNECTION_TYPE 2
+#define ISCSI_CONNECTION_TYPE 3
+#define FCOE_CONNECTION_TYPE 4
+#define RESERVED_CONNECTION_TYPE_0 5
+#define RESERVED_CONNECTION_TYPE_1 6
+#define RESERVED_CONNECTION_TYPE_2 7
+
 
-#define PROTOCOL_STATE_BIT_OFFSET		6
+#define PROTOCOL_STATE_BIT_OFFSET 6
 
-#define ETH_STATE	(ETH_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define ETH_STATE (ETH_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define TOE_STATE (TOE_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define RDMA_STATE (RDMA_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define ISCSI_STATE \
+	(ISCSI_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
+#define FCOE_STATE (FCOE_CONNECTION_TYPE << PROTOCOL_STATE_BIT_OFFSET)
 
 /* microcode fixed page page size 4K (chains and ring segments) */
-#define MC_PAGE_SIZE						(4096)
+#define MC_PAGE_SIZE (4096)
 
-/* Host coalescing constants */
 
-/* IGU constants */
-#define IGU_PORT_BASE				0x0400
-
-#define IGU_ADDR_MSIX				0x0000
-#define IGU_ADDR_INT_ACK			0x0200
-#define IGU_ADDR_PROD_UPD			0x0201
-#define IGU_ADDR_ATTN_BITS_UPD			0x0202
-#define IGU_ADDR_ATTN_BITS_SET			0x0203
-#define IGU_ADDR_ATTN_BITS_CLR			0x0204
-#define IGU_ADDR_COALESCE_NOW			0x0205
-#define IGU_ADDR_SIMD_MASK			0x0206
-#define IGU_ADDR_SIMD_NOMASK			0x0207
-#define IGU_ADDR_MSI_CTL			0x0210
-#define IGU_ADDR_MSI_ADDR_LO			0x0211
-#define IGU_ADDR_MSI_ADDR_HI			0x0212
-#define IGU_ADDR_MSI_DATA			0x0213
-
-#define IGU_INT_ENABLE				0
-#define IGU_INT_DISABLE 			1
-#define IGU_INT_NOP				2
-#define IGU_INT_NOP2				3
+/* Host coalescing constants */
 
 /* index numbers */
-#define HC_USTORM_DEF_SB_NUM_INDICES		4
-#define HC_CSTORM_DEF_SB_NUM_INDICES		8
-#define HC_XSTORM_DEF_SB_NUM_INDICES		4
-#define HC_TSTORM_DEF_SB_NUM_INDICES		4
-#define HC_USTORM_SB_NUM_INDICES		4
-#define HC_CSTORM_SB_NUM_INDICES		4
+#define HC_USTORM_DEF_SB_NUM_INDICES 4
+#define HC_CSTORM_DEF_SB_NUM_INDICES 8
+#define HC_XSTORM_DEF_SB_NUM_INDICES 4
+#define HC_TSTORM_DEF_SB_NUM_INDICES 4
+#define HC_USTORM_SB_NUM_INDICES 4
+#define HC_CSTORM_SB_NUM_INDICES 4
 
 /* index values - which counterto update */
 
-#define HC_INDEX_U_ETH_RX_CQ_CONS		1
+#define HC_INDEX_U_TOE_RX_CQ_CONS 0
+#define HC_INDEX_U_ETH_RX_CQ_CONS 1
+#define HC_INDEX_U_ETH_RX_BD_CONS 2
+#define HC_INDEX_U_FCOE_EQ_CONS 3
+
+#define HC_INDEX_C_TOE_TX_CQ_CONS 0
+#define HC_INDEX_C_ETH_TX_CQ_CONS 1
+#define HC_INDEX_C_ISCSI_EQ_CONS 2
+
+#define HC_INDEX_DEF_X_SPQ_CONS 0
 
-#define HC_INDEX_C_ETH_TX_CQ_CONS		1
+#define HC_INDEX_DEF_C_RDMA_EQ_CONS 0
+#define HC_INDEX_DEF_C_RDMA_NAL_PROD 1
+#define HC_INDEX_DEF_C_ETH_FW_TX_CQ_CONS 2
+#define HC_INDEX_DEF_C_ETH_SLOW_PATH 3
+#define HC_INDEX_DEF_C_ETH_RDMA_CQ_CONS 4
+#define HC_INDEX_DEF_C_ETH_ISCSI_CQ_CONS 5
 
-#define HC_INDEX_DEF_X_SPQ_CONS 		0
+#define HC_INDEX_DEF_U_ETH_RDMA_RX_CQ_CONS 0
+#define HC_INDEX_DEF_U_ETH_ISCSI_RX_CQ_CONS 1
+#define HC_INDEX_DEF_U_ETH_RDMA_RX_BD_CONS 2
+#define HC_INDEX_DEF_U_ETH_ISCSI_RX_BD_CONS 3
 
-#define HC_INDEX_DEF_C_ETH_FW_TX_CQ_CONS	2
-#define HC_INDEX_DEF_C_ETH_SLOW_PATH		3
 
 /* used by the driver to get the SB offset */
-#define USTORM_ID			0
-#define CSTORM_ID			1
-#define XSTORM_ID			2
-#define TSTORM_ID			3
-#define ATTENTION_ID			4
+#define USTORM_ID 0
+#define CSTORM_ID 1
+#define XSTORM_ID 2
+#define TSTORM_ID 3
+#define ATTENTION_ID 4
 
 /* max number of slow path commands per port */
-#define MAX_RAMRODS_PER_PORT		(8)
+#define MAX_RAMRODS_PER_PORT (8)
 
 /* values for RX ETH CQE type field */
-#define RX_ETH_CQE_TYPE_ETH_FASTPATH	(0)
-#define RX_ETH_CQE_TYPE_ETH_RAMROD		(1)
-
-/* MAC address list size */
-#define T_MAC_ADDRESS_LIST_SIZE 	(96)
-
+#define RX_ETH_CQE_TYPE_ETH_FASTPATH (0)
+#define RX_ETH_CQE_TYPE_ETH_RAMROD (1)
+
+
+/**** DEFINES FOR TIMERS/CLOCKS RESOLUTIONS ****/
+#define EMULATION_FREQUENCY_FACTOR (1600)
+#define FPGA_FREQUENCY_FACTOR (100)
+
+#define TIMERS_TICK_SIZE_CHIP (1e-3)
+#define TIMERS_TICK_SIZE_EMUL \
+ ((TIMERS_TICK_SIZE_CHIP)/((EMULATION_FREQUENCY_FACTOR)))
+#define TIMERS_TICK_SIZE_FPGA \
+ ((TIMERS_TICK_SIZE_CHIP)/((FPGA_FREQUENCY_FACTOR)))
+
+#define TSEMI_CLK1_RESUL_CHIP (1e-3)
+#define TSEMI_CLK1_RESUL_EMUL \
+ ((TSEMI_CLK1_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define TSEMI_CLK1_RESUL_FPGA \
+ ((TSEMI_CLK1_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+#define USEMI_CLK1_RESUL_CHIP \
+ (TIMERS_TICK_SIZE_CHIP)
+#define USEMI_CLK1_RESUL_EMUL \
+ (TIMERS_TICK_SIZE_EMUL)
+#define USEMI_CLK1_RESUL_FPGA \
+ (TIMERS_TICK_SIZE_FPGA)
+
+#define XSEMI_CLK1_RESUL_CHIP (1e-3)
+#define XSEMI_CLK1_RESUL_EMUL \
+ ((XSEMI_CLK1_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define XSEMI_CLK1_RESUL_FPGA \
+ ((XSEMI_CLK1_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+#define XSEMI_CLK2_RESUL_CHIP (1e-6)
+#define XSEMI_CLK2_RESUL_EMUL \
+ ((XSEMI_CLK2_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define XSEMI_CLK2_RESUL_FPGA \
+ ((XSEMI_CLK2_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+#define SDM_TIMER_TICK_RESUL_CHIP (4*(1e-6))
+#define SDM_TIMER_TICK_RESUL_EMUL \
+ ((SDM_TIMER_TICK_RESUL_CHIP)/(EMULATION_FREQUENCY_FACTOR))
+#define SDM_TIMER_TICK_RESUL_FPGA \
+ ((SDM_TIMER_TICK_RESUL_CHIP)/(FPGA_FREQUENCY_FACTOR))
+
+
+/**** END DEFINES FOR TIMERS/CLOCKS RESOLUTIONS ****/
 #define XSTORM_IP_ID_ROLL_HALF 0x8000
 #define XSTORM_IP_ID_ROLL_ALL 0
 
-#define FW_LOG_LIST_SIZE	(50)
+#define FW_LOG_LIST_SIZE (50)
+
+#define NUM_OF_PROTOCOLS 4
+#define MAX_COS_NUMBER 16
+#define MAX_T_STAT_COUNTER_ID 18
+#define MAX_X_STAT_COUNTER_ID 18
 
-#define NUM_OF_PROTOCOLS		4
-#define MAX_COS_NUMBER			16
-#define MAX_T_STAT_COUNTER_ID	18
+#define UNKNOWN_ADDRESS 0
+#define UNICAST_ADDRESS 1
+#define MULTICAST_ADDRESS 2
+#define BROADCAST_ADDRESS 3
 
-#define T_FAIR							1
-#define FAIR_MEM						2
-#define RS_PERIODIC_TIMEOUT_IN_SDM_TICS 25
+#define SINGLE_FUNCTION 0
+#define MULTI_FUNCTION 1
 
-#define UNKNOWN_ADDRESS 	0
-#define UNICAST_ADDRESS 	1
-#define MULTICAST_ADDRESS	2
-#define BROADCAST_ADDRESS	3
+#define IP_V4 0
+#define IP_V6 1
 
diff --git a/drivers/net/bnx2x_hsi.h b/drivers/net/bnx2x_hsi.h
index 96208ace1466..e515d68ea20f 100644
--- a/drivers/net/bnx2x_hsi.h
+++ b/drivers/net/bnx2x_hsi.h
@@ -132,6 +132,12 @@ struct shared_hw_cfg {					 /* NVRAM Offset */
 #define SHARED_HW_CFG_BOARD_TYPE_BCM957710T1003G    0x00000008
 #define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G    0x00000009
 #define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1021G    0x0000000a
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1023G    0x0000000b
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957710A1033G    0x0000000c
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711T1101     0x0000000d
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711ET1201    0x0000000e
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711A1133G    0x0000000f
+#define SHARED_HW_CFG_BOARD_TYPE_BCM957711EA1233G   0x00000010
 
 #define SHARED_HW_CFG_BOARD_VER_MASK		    0xffff0000
 #define SHARED_HW_CFG_BOARD_VER_SHIFT		    16
@@ -313,6 +319,7 @@ struct shared_feat_cfg {				 /* NVRAM Offset */
 
 	u32 config;						/* 0x450 */
 #define SHARED_FEATURE_BMC_ECHO_MODE_EN 	    0x00000001
+#define SHARED_FEATURE_MF_MODE_DISABLED 	    0x00000100
 
 };
 
@@ -502,20 +509,20 @@ struct port_feat_cfg {			    /* port 0: 0x454  port 1: 0x4c8 */
 };
 
 
-/*****************************************************************************
- * Device Information							     *
- *****************************************************************************/
-struct dev_info {						     /* size */
+/****************************************************************************
+ * Device Information							    *
+ ****************************************************************************/
+struct dev_info {						    /* size */
 
-	u32    bc_rev; /* 8 bits each: major, minor, build */	        /* 4 */
+	u32    bc_rev; /* 8 bits each: major, minor, build */	       /* 4 */
 
-	struct shared_hw_cfg	 shared_hw_config;		       /* 40 */
+	struct shared_hw_cfg	 shared_hw_config;		      /* 40 */
 
-	struct port_hw_cfg	 port_hw_config[PORT_MAX];      /* 400*2=800 */
+	struct port_hw_cfg	 port_hw_config[PORT_MAX];     /* 400*2=800 */
 
-	struct shared_feat_cfg	 shared_feature_config; 	        /* 4 */
+	struct shared_feat_cfg	 shared_feature_config; 	       /* 4 */
 
-	struct port_feat_cfg	 port_feature_config[PORT_MAX]; /* 116*2=232 */
+	struct port_feat_cfg	 port_feature_config[PORT_MAX];/* 116*2=232 */
 
 };
 
@@ -632,7 +639,9 @@ struct drv_port_mb {
 #define LINK_STATUS_LINK_PARTNER_15GXFD_CAPABLE 	0x08000000
 #define LINK_STATUS_LINK_PARTNER_16GXFD_CAPABLE 	0x10000000
 
-	u32 reserved[3];
+	u32 port_stx;
+
+	u32 reserved[2];
 
 };
 
@@ -655,6 +664,11 @@ struct drv_func_mb {
 #define DRV_MSG_CODE_GET_MANUF_KEY			0x82000000
 #define DRV_MSG_CODE_LOAD_L2B_PRAM			0x90000000
 
+#define BIOS_MSG_CODE_LIC_CHALLENGE			0xff010000
+#define BIOS_MSG_CODE_LIC_RESPONSE			0xff020000
+#define BIOS_MSG_CODE_VIRT_MAC_PRIM			0xff030000
+#define BIOS_MSG_CODE_VIRT_MAC_ISCSI			0xff040000
+
 #define DRV_MSG_SEQ_NUMBER_MASK 			0x0000ffff
 
 	u32 drv_mb_param;
@@ -684,6 +698,11 @@ struct drv_func_mb {
 #define FW_MSG_CODE_L2B_PRAM_X_LOAD_FAILURE		0x90230000
 #define FW_MSG_CODE_L2B_PRAM_U_LOAD_FAILURE		0x90240000
 
+#define FW_MSG_CODE_LIC_CHALLENGE			0xff010000
+#define FW_MSG_CODE_LIC_RESPONSE			0xff020000
+#define FW_MSG_CODE_VIRT_MAC_PRIM			0xff030000
+#define FW_MSG_CODE_VIRT_MAC_ISCSI			0xff040000
+
 #define FW_MSG_SEQ_NUMBER_MASK				0x0000ffff
 
 	u32 fw_mb_param;
@@ -709,7 +728,13 @@ struct drv_func_mb {
 	u32 iscsi_boot_signature;
 	u32 iscsi_boot_block_offset;
 
-	u32 reserved[3];
+	u32 drv_status;
+#define DRV_STATUS_PMF					0x00000001
+
+	u32 virt_mac_upper;
+#define VIRT_MAC_SIGN_MASK				0xffff0000
+#define VIRT_MAC_SIGNATURE				0x564d0000
+	u32 virt_mac_lower;
 
 };
 
@@ -725,6 +750,92 @@ struct mgmtfw_state {
 };
 
 
+/****************************************************************************
+ * Multi-Function configuration 					    *
+ ****************************************************************************/
+struct shared_mf_cfg {
+
+	u32 clp_mb;
+#define SHARED_MF_CLP_SET_DEFAULT		    0x00000000
+	/* set by CLP */
+#define SHARED_MF_CLP_EXIT			    0x00000001
+	/* set by MCP */
+#define SHARED_MF_CLP_EXIT_DONE 		    0x00010000
+
+};
+
+struct port_mf_cfg {
+
+	u32 dynamic_cfg;	/* device control channel */
+#define PORT_MF_CFG_OUTER_VLAN_TAG_MASK 	    0x0000ffff
+#define PORT_MF_CFG_OUTER_VLAN_TAG_SHIFT	    0
+#define PORT_MF_CFG_DYNAMIC_CFG_ENABLED 	    0x00010000
+#define PORT_MF_CFG_DYNAMIC_CFG_DEFAULT 	    0x00000000
+
+	u32 reserved[3];
+
+};
+
+struct func_mf_cfg {
+
+	u32 config;
+	/* E/R/I/D */
+	/* function 0 of each port cannot be hidden */
+#define FUNC_MF_CFG_FUNC_HIDE			    0x00000001
+
+#define FUNC_MF_CFG_PROTOCOL_MASK		    0x00000007
+#define FUNC_MF_CFG_PROTOCOL_ETHERNET		    0x00000002
+#define FUNC_MF_CFG_PROTOCOL_ETHERNET_WITH_RDMA     0x00000004
+#define FUNC_MF_CFG_PROTOCOL_ISCSI		    0x00000006
+#define FUNC_MF_CFG_PROTOCOL_DEFAULT\
+	FUNC_MF_CFG_PROTOCOL_ETHERNET_WITH_RDMA
+
+#define FUNC_MF_CFG_FUNC_DISABLED		    0x00000008
+
+	/* PRI */
+	/* 0 - low priority, 3 - high priority */
+#define FUNC_MF_CFG_TRANSMIT_PRIORITY_MASK	    0x00000300
+#define FUNC_MF_CFG_TRANSMIT_PRIORITY_SHIFT	    8
+#define FUNC_MF_CFG_TRANSMIT_PRIORITY_DEFAULT	    0x00000000
+
+	/* MINBW, MAXBW */
+	/* value range - 0..100, increments in 100Mbps */
+#define FUNC_MF_CFG_MIN_BW_MASK 		    0x00ff0000
+#define FUNC_MF_CFG_MIN_BW_SHIFT		    16
+#define FUNC_MF_CFG_MIN_BW_DEFAULT		    0x00000000
+#define FUNC_MF_CFG_MAX_BW_MASK 		    0xff000000
+#define FUNC_MF_CFG_MAX_BW_SHIFT		    24
+#define FUNC_MF_CFG_MAX_BW_DEFAULT		    0x64000000
+
+	u32 mac_upper;		/* MAC */
+#define FUNC_MF_CFG_UPPERMAC_MASK		    0x0000ffff
+#define FUNC_MF_CFG_UPPERMAC_SHIFT		    0
+#define FUNC_MF_CFG_UPPERMAC_DEFAULT		    FUNC_MF_CFG_UPPERMAC_MASK
+	u32 mac_lower;
+#define FUNC_MF_CFG_LOWERMAC_DEFAULT		    0xffffffff
+
+	u32 e1hov_tag;	/* VNI */
+#define FUNC_MF_CFG_E1HOV_TAG_MASK		    0x0000ffff
+#define FUNC_MF_CFG_E1HOV_TAG_SHIFT		    0
+#define FUNC_MF_CFG_E1HOV_TAG_DEFAULT		    FUNC_MF_CFG_E1HOV_TAG_MASK
+
+	u32 reserved[2];
+
+};
+
+struct mf_cfg {
+
+	struct shared_mf_cfg	shared_mf_config;
+	struct port_mf_cfg	port_mf_config[PORT_MAX];
+#if defined(b710)
+	struct func_mf_cfg	func_mf_config[E1_FUNC_MAX];
+#else
+	struct func_mf_cfg	func_mf_config[E1H_FUNC_MAX];
+#endif
+
+};
+
+
 /****************************************************************************
  * Shared Memory Region 						    *
  ****************************************************************************/
@@ -760,18 +871,18 @@ struct shmem_region {			       /*   SharedMem Offset (size) */
 	struct mgmtfw_state	mgmtfw_state;	       /* 0x4ac     (0x1b8) */
 
 	struct drv_port_mb	port_mb[PORT_MAX];     /* 0x664 (16*2=0x20) */
-#if defined(b710)
-	struct drv_func_mb	func_mb[E1_FUNC_MAX];  /* 0x684 (44*2=0x58) */
-#else
 	struct drv_func_mb	func_mb[E1H_FUNC_MAX];
-#endif
+
+	struct mf_cfg		mf_cfg;
 
 };						       /* 0x6dc */
 
 
+
+
 #define BCM_5710_FW_MAJOR_VERSION			4
-#define BCM_5710_FW_MINOR_VERSION			0
-#define BCM_5710_FW_REVISION_VERSION			14
+#define BCM_5710_FW_MINOR_VERSION			5
+#define BCM_5710_FW_REVISION_VERSION			1
 #define BCM_5710_FW_COMPILE_FLAGS			1
 
 
@@ -810,7 +921,7 @@ struct doorbell_hdr {
 };
 
 /*
- * doorbell message send to the chip
+ * doorbell message sent to the chip
  */
 struct doorbell {
 #if defined(__BIG_ENDIAN)
@@ -866,8 +977,10 @@ struct parsing_flags {
 	u16 flags;
 #define PARSING_FLAGS_ETHERNET_ADDRESS_TYPE (0x1<<0)
 #define PARSING_FLAGS_ETHERNET_ADDRESS_TYPE_SHIFT 0
-#define PARSING_FLAGS_NUMBER_OF_NESTED_VLANS (0x3<<1)
-#define PARSING_FLAGS_NUMBER_OF_NESTED_VLANS_SHIFT 1
+#define PARSING_FLAGS_VLAN (0x1<<1)
+#define PARSING_FLAGS_VLAN_SHIFT 1
+#define PARSING_FLAGS_EXTRA_VLAN (0x1<<2)
+#define PARSING_FLAGS_EXTRA_VLAN_SHIFT 2
 #define PARSING_FLAGS_OVER_ETHERNET_PROTOCOL (0x3<<3)
 #define PARSING_FLAGS_OVER_ETHERNET_PROTOCOL_SHIFT 3
 #define PARSING_FLAGS_IP_OPTIONS (0x1<<5)
@@ -891,6 +1004,12 @@ struct parsing_flags {
 };
 
 
+struct regpair {
+	u32 lo;
+	u32 hi;
+};
+
+
 /*
  * dmae command structure
  */
@@ -971,72 +1090,107 @@ struct double_regpair {
 
 
 /*
- * The eth Rx Buffer Descriptor
+ * The eth storm context of Ustorm (configuration part)
  */
-struct eth_rx_bd {
-	u32 addr_lo;
-	u32 addr_hi;
-};
-
-/*
- * The eth storm context of Ustorm
- */
-struct ustorm_eth_st_context {
+struct ustorm_eth_st_context_config {
 #if defined(__BIG_ENDIAN)
-	u8 sb_index_number;
+	u8 flags;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT (0x1<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC (0x1<<1)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC_SHIFT 1
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA (0x1<<2)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA_SHIFT 2
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING (0x1<<3)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING_SHIFT 3
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0 (0xF<<4)
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0_SHIFT 4
 	u8 status_block_id;
-	u8 __local_rx_bd_cons;
-	u8 __local_rx_bd_prod;
+	u8 clientId;
+	u8 sb_index_numbers;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER (0xF<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER (0xF<<4)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER_SHIFT 4
 #elif defined(__LITTLE_ENDIAN)
-	u8 __local_rx_bd_prod;
-	u8 __local_rx_bd_cons;
+	u8 sb_index_numbers;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER (0xF<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_CQE_SB_INDEX_NUMBER_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER (0xF<<4)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_BD_SB_INDEX_NUMBER_SHIFT 4
+	u8 clientId;
 	u8 status_block_id;
-	u8 sb_index_number;
+	u8 flags;
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT (0x1<<0)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT_SHIFT 0
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC (0x1<<1)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_DYNAMIC_HC_SHIFT 1
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA (0x1<<2)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_TPA_SHIFT 2
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING (0x1<<3)
+#define USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_SGE_RING_SHIFT 3
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0 (0xF<<4)
+#define __USTORM_ETH_ST_CONTEXT_CONFIG_RESERVED0_SHIFT 4
 #endif
 #if defined(__BIG_ENDIAN)
-	u16 rcq_cons;
-	u16 rx_bd_cons;
+	u16 bd_buff_size;
+	u16 mc_alignment_size;
 #elif defined(__LITTLE_ENDIAN)
-	u16 rx_bd_cons;
-	u16 rcq_cons;
+	u16 mc_alignment_size;
+	u16 bd_buff_size;
 #endif
-	u32 rx_bd_page_base_lo;
-	u32 rx_bd_page_base_hi;
-	u32 rcq_base_address_lo;
-	u32 rcq_base_address_hi;
 #if defined(__BIG_ENDIAN)
-	u16 __num_of_returned_cqes;
-	u8 num_rss;
-	u8 flags;
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT (0x1<<0)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT_SHIFT 0
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC (0x1<<1)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC_SHIFT 1
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA (0x1<<2)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA_SHIFT 2
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0 (0x1F<<3)
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0_SHIFT 3
+	u8 __local_sge_prod;
+	u8 __local_bd_prod;
+	u16 sge_buff_size;
 #elif defined(__LITTLE_ENDIAN)
-	u8 flags;
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT (0x1<<0)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT_SHIFT 0
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC (0x1<<1)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_DYNAMIC_HC_SHIFT 1
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA (0x1<<2)
-#define USTORM_ETH_ST_CONTEXT_ENABLE_TPA_SHIFT 2
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0 (0x1F<<3)
-#define __USTORM_ETH_ST_CONTEXT_RESERVED0_SHIFT 3
-	u8 num_rss;
-	u16 __num_of_returned_cqes;
+	u16 sge_buff_size;
+	u8 __local_bd_prod;
+	u8 __local_sge_prod;
 #endif
 #if defined(__BIG_ENDIAN)
-	u16 mc_alignment_size;
-	u16 agg_threshold;
+	u16 __bd_cons;
+	u16 __sge_cons;
 #elif defined(__LITTLE_ENDIAN)
-	u16 agg_threshold;
-	u16 mc_alignment_size;
+	u16 __sge_cons;
+	u16 __bd_cons;
 #endif
+	u32 bd_page_base_lo;
+	u32 bd_page_base_hi;
+	u32 sge_page_base_lo;
+	u32 sge_page_base_hi;
+};
+
+/*
+ * The eth Rx Buffer Descriptor
+ */
+struct eth_rx_bd {
+	u32 addr_lo;
+	u32 addr_hi;
+};
+
+/*
+ * The eth Rx SGE Descriptor
+ */
+struct eth_rx_sge {
+	u32 addr_lo;
+	u32 addr_hi;
+};
+
+/*
+ * Local BDs and SGEs rings (in ETH)
+ */
+struct eth_local_rx_rings {
 	struct eth_rx_bd __local_bd_ring[16];
+	struct eth_rx_sge __local_sge_ring[12];
+};
+
+/*
+ * The eth storm context of Ustorm
+ */
+struct ustorm_eth_st_context {
+	struct ustorm_eth_st_context_config common;
+	struct eth_local_rx_rings __rings;
 };
 
 /*
@@ -1107,9 +1261,9 @@ struct xstorm_eth_extra_ag_context_section {
 #if defined(__BIG_ENDIAN)
 	u16 __reserved3;
 	u8 __reserved2;
-	u8 __agg_misc7;
+	u8 __da_only_cnt;
 #elif defined(__LITTLE_ENDIAN)
-	u8 __agg_misc7;
+	u8 __da_only_cnt;
 	u8 __reserved2;
 	u16 __reserved3;
 #endif
@@ -1387,7 +1541,13 @@ struct timers_block_context {
 	u32 __reserved_0;
 	u32 __reserved_1;
 	u32 __reserved_2;
-	u32 __reserved_flags;
+	u32 flags;
+#define __TIMERS_BLOCK_CONTEXT_NUM_OF_ACTIVE_TIMERS (0x3<<0)
+#define __TIMERS_BLOCK_CONTEXT_NUM_OF_ACTIVE_TIMERS_SHIFT 0
+#define TIMERS_BLOCK_CONTEXT_CONN_VALID_FLG (0x1<<2)
+#define TIMERS_BLOCK_CONTEXT_CONN_VALID_FLG_SHIFT 2
+#define __TIMERS_BLOCK_CONTEXT_RESERVED0 (0x1FFFFFFF<<3)
+#define __TIMERS_BLOCK_CONTEXT_RESERVED0_SHIFT 3
 };
 
 /*
@@ -1497,11 +1657,19 @@ struct xstorm_eth_st_context {
 	u32 tx_bd_page_base_hi;
 #if defined(__BIG_ENDIAN)
 	u16 tx_bd_cons;
-	u8 __reserved0;
+	u8 statistics_data;
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID (0x7F<<0)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID_SHIFT 0
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE (0x1<<7)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE_SHIFT 7
 	u8 __local_tx_bd_prod;
 #elif defined(__LITTLE_ENDIAN)
 	u8 __local_tx_bd_prod;
-	u8 __reserved0;
+	u8 statistics_data;
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID (0x7F<<0)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_COUNTER_ID_SHIFT 0
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE (0x1<<7)
+#define XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE_SHIFT 7
 	u16 tx_bd_cons;
 #endif
 	u32 db_data_addr_lo;
@@ -1578,7 +1746,7 @@ struct eth_tx_doorbell {
 struct ustorm_def_status_block {
 	u16 index_values[HC_USTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1589,7 +1757,7 @@ struct ustorm_def_status_block {
 struct cstorm_def_status_block {
 	u16 index_values[HC_CSTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1600,7 +1768,7 @@ struct cstorm_def_status_block {
 struct xstorm_def_status_block {
 	u16 index_values[HC_XSTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1611,7 +1779,7 @@ struct xstorm_def_status_block {
 struct tstorm_def_status_block {
 	u16 index_values[HC_TSTORM_DEF_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1634,7 +1802,7 @@ struct host_def_status_block {
 struct ustorm_status_block {
 	u16 index_values[HC_USTORM_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1645,7 +1813,7 @@ struct ustorm_status_block {
 struct cstorm_status_block {
 	u16 index_values[HC_CSTORM_SB_NUM_INDICES];
 	u16 status_block_index;
-	u8 reserved0;
+	u8 func;
 	u8 status_block_id;
 	u32 __flags;
 };
@@ -1683,20 +1851,21 @@ struct eth_dynamic_hc_config {
  * regular eth FP CQE parameters struct
  */
 struct eth_fast_path_rx_cqe {
-	u8 type;
-	u8 error_type_flags;
-#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG (0x1<<0)
-#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG_SHIFT 0
-#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG (0x1<<1)
-#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG_SHIFT 1
-#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG (0x1<<2)
-#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG_SHIFT 2
-#define ETH_FAST_PATH_RX_CQE_START_FLG (0x1<<3)
-#define ETH_FAST_PATH_RX_CQE_START_FLG_SHIFT 3
-#define ETH_FAST_PATH_RX_CQE_END_FLG (0x1<<4)
-#define ETH_FAST_PATH_RX_CQE_END_FLG_SHIFT 4
-#define ETH_FAST_PATH_RX_CQE_RESERVED0 (0x7<<5)
-#define ETH_FAST_PATH_RX_CQE_RESERVED0_SHIFT 5
+	u8 type_error_flags;
+#define ETH_FAST_PATH_RX_CQE_TYPE (0x1<<0)
+#define ETH_FAST_PATH_RX_CQE_TYPE_SHIFT 0
+#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG (0x1<<1)
+#define ETH_FAST_PATH_RX_CQE_PHY_DECODE_ERR_FLG_SHIFT 1
+#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG (0x1<<2)
+#define ETH_FAST_PATH_RX_CQE_IP_BAD_XSUM_FLG_SHIFT 2
+#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG (0x1<<3)
+#define ETH_FAST_PATH_RX_CQE_L4_BAD_XSUM_FLG_SHIFT 3
+#define ETH_FAST_PATH_RX_CQE_START_FLG (0x1<<4)
+#define ETH_FAST_PATH_RX_CQE_START_FLG_SHIFT 4
+#define ETH_FAST_PATH_RX_CQE_END_FLG (0x1<<5)
+#define ETH_FAST_PATH_RX_CQE_END_FLG_SHIFT 5
+#define ETH_FAST_PATH_RX_CQE_RESERVED0 (0x3<<6)
+#define ETH_FAST_PATH_RX_CQE_RESERVED0_SHIFT 6
 	u8 status_flags;
 #define ETH_FAST_PATH_RX_CQE_RSS_HASH_TYPE (0x7<<0)
 #define ETH_FAST_PATH_RX_CQE_RSS_HASH_TYPE_SHIFT 0
@@ -1711,11 +1880,13 @@ struct eth_fast_path_rx_cqe {
 #define ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG (0x1<<7)
 #define ETH_FAST_PATH_RX_CQE_L4_XSUM_NO_VALIDATION_FLG_SHIFT 7
 	u8 placement_offset;
+	u8 queue_index;
 	u32 rss_hash_result;
 	u16 vlan_tag;
 	u16 pkt_len;
-	u16 queue_index;
+	u16 len_on_bd;
 	struct parsing_flags pars_flags;
+	u16 sgl[8];
 };
 
 
@@ -1728,6 +1899,23 @@ struct eth_halt_ramrod_data {
 };
 
 
+/*
+ * The data for statistics query ramrod
+ */
+struct eth_query_ramrod_data {
+#if defined(__BIG_ENDIAN)
+	u8 reserved0;
+	u8 collect_port_1b;
+	u16 drv_counter;
+#elif defined(__LITTLE_ENDIAN)
+	u16 drv_counter;
+	u8 collect_port_1b;
+	u8 reserved0;
+#endif
+	u32 ctr_id_vector;
+};
+
+
 /*
  * Place holder for ramrods protocol specific data
  */
@@ -1758,15 +1946,20 @@ struct eth_rx_bd_next_page {
  * Eth Rx Cqe structure- general structure for ramrods
  */
 struct common_ramrod_eth_rx_cqe {
-	u8 type;
+	u8 ramrod_type;
+#define COMMON_RAMROD_ETH_RX_CQE_TYPE (0x1<<0)
+#define COMMON_RAMROD_ETH_RX_CQE_TYPE_SHIFT 0
+#define COMMON_RAMROD_ETH_RX_CQE_RESERVED0 (0x7F<<1)
+#define COMMON_RAMROD_ETH_RX_CQE_RESERVED0_SHIFT 1
 	u8 conn_type_3b;
-	u16 reserved;
+	u16 reserved1;
 	u32 conn_and_cmd_data;
 #define COMMON_RAMROD_ETH_RX_CQE_CID (0xFFFFFF<<0)
 #define COMMON_RAMROD_ETH_RX_CQE_CID_SHIFT 0
 #define COMMON_RAMROD_ETH_RX_CQE_CMD_ID (0xFF<<24)
 #define COMMON_RAMROD_ETH_RX_CQE_CMD_ID_SHIFT 24
 	struct ramrod_data protocol_data;
+	u32 reserved2[4];
 };
 
 /*
@@ -1775,8 +1968,7 @@ struct common_ramrod_eth_rx_cqe {
 struct eth_rx_cqe_next_page {
 	u32 addr_lo;
 	u32 addr_hi;
-	u32 reserved0;
-	u32 reserved1;
+	u32 reserved[6];
 };
 
 /*
@@ -1806,11 +1998,6 @@ struct spe_hdr {
 	u16 reserved;
 };
 
-struct regpair {
-	u32 lo;
-	u32 hi;
-};
-
 /*
  * ethernet slow path element
  */
@@ -1821,6 +2008,7 @@ union eth_specific_data {
 	struct eth_halt_ramrod_data halt_ramrod_data;
 	struct regpair leading_cqe_addr;
 	struct regpair update_data_addr;
+	struct eth_query_ramrod_data query_ramrod_data;
 };
 
 /*
@@ -1843,10 +2031,13 @@ struct eth_tx_db_data {
 
 
 /*
- * Common configuration parameters per port in Tstorm
+ * Common configuration parameters per function in Tstorm
  */
 struct tstorm_eth_function_common_config {
-	u32 config_flags;
+#if defined(__BIG_ENDIAN)
+	u8 leading_client_id;
+	u8 rss_result_mask;
+	u16 config_flags;
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY (0x1<<0)
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY_SHIFT 0
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY (0x1<<1)
@@ -1859,17 +2050,32 @@ struct tstorm_eth_function_common_config {
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE_SHIFT 4
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE (0x1<<5)
 #define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE_SHIFT 5
-#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0 (0x3FFFFFF<<6)
-#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0_SHIFT 6
-#if defined(__BIG_ENDIAN)
-	u16 __secondary_vlan_id;
-	u8 leading_client_id;
-	u8 rss_result_mask;
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM (0x1<<6)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM_SHIFT 6
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0 (0x1FF<<7)
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0_SHIFT 7
 #elif defined(__LITTLE_ENDIAN)
+	u16 config_flags;
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY (0x1<<0)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY_SHIFT 0
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY (0x1<<1)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY_SHIFT 1
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY (0x1<<2)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY_SHIFT 2
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY (0x1<<3)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY_SHIFT 3
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE (0x1<<4)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_ENABLE_SHIFT 4
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE (0x1<<5)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_DEFAULT_ENABLE_SHIFT 5
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM (0x1<<6)
+#define TSTORM_ETH_FUNCTION_COMMON_CONFIG_VLAN_IN_CAM_SHIFT 6
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0 (0x1FF<<7)
+#define __TSTORM_ETH_FUNCTION_COMMON_CONFIG_RESERVED0_SHIFT 7
 	u8 rss_result_mask;
 	u8 leading_client_id;
-	u16 __secondary_vlan_id;
 #endif
+	u16 vlan_id[2];
 };
 
 /*
@@ -1887,7 +2093,7 @@ struct eth_update_ramrod_data {
 struct mac_configuration_hdr {
 	u8 length_6b;
 	u8 offset;
-	u16 reserved0;
+	u16 client_id;
 	u32 reserved1;
 };
 
@@ -1943,16 +2149,56 @@ struct mac_configuration_cmd {
 };
 
 
+/*
+ * MAC address in list for ramrod
+ */
+struct mac_configuration_entry_e1h {
+	u16 lsb_mac_addr;
+	u16 middle_mac_addr;
+	u16 msb_mac_addr;
+	u16 vlan_id;
+	u16 e1hov_id;
+	u8 client_id;
+	u8 flags;
+#define MAC_CONFIGURATION_ENTRY_E1H_PORT (0x1<<0)
+#define MAC_CONFIGURATION_ENTRY_E1H_PORT_SHIFT 0
+#define MAC_CONFIGURATION_ENTRY_E1H_ACTION_TYPE (0x1<<1)
+#define MAC_CONFIGURATION_ENTRY_E1H_ACTION_TYPE_SHIFT 1
+#define MAC_CONFIGURATION_ENTRY_E1H_RDMA_MAC (0x1<<2)
+#define MAC_CONFIGURATION_ENTRY_E1H_RDMA_MAC_SHIFT 2
+#define MAC_CONFIGURATION_ENTRY_E1H_RESERVED0 (0x1F<<3)
+#define MAC_CONFIGURATION_ENTRY_E1H_RESERVED0_SHIFT 3
+};
+
+/*
+ * MAC filtering configuration command
+ */
+struct mac_configuration_cmd_e1h {
+	struct mac_configuration_hdr hdr;
+	struct mac_configuration_entry_e1h config_table[32];
+};
+
+
+/*
+ * approximate-match multicast filtering for E1H per function in Tstorm
+ */
+struct tstorm_eth_approximate_match_multicast_filtering {
+	u32 mcast_add_hash_bit_array[8];
+};
+
+
 /*
  * Configuration parameters per client in Tstorm
  */
 struct tstorm_eth_client_config {
 #if defined(__BIG_ENDIAN)
-	u16 statistics_counter_id;
+	u8 max_sges_for_packet;
+	u8 statistics_counter_id;
 	u16 mtu;
 #elif defined(__LITTLE_ENDIAN)
 	u16 mtu;
-	u16 statistics_counter_id;
+	u8 statistics_counter_id;
+	u8 max_sges_for_packet;
 #endif
 #if defined(__BIG_ENDIAN)
 	u16 drop_flags;
@@ -1960,42 +2206,42 @@ struct tstorm_eth_client_config {
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR_SHIFT 1
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR (0x1<<2)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR_SHIFT 2
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<3)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 3
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<4)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 4
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0x7FF<<5)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 5
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<3)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 3
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0xFFF<<4)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 4
 	u16 config_flags;
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE (0x1<<0)
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE_SHIFT 1
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x3FFF<<2)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING_SHIFT 2
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x1FFF<<3)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 3
 #elif defined(__LITTLE_ENDIAN)
 	u16 config_flags;
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE (0x1<<0)
 #define TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE_SHIFT 1
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x3FFF<<2)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_ENABLE_SGE_RING_SHIFT 2
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0 (0x1FFF<<3)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED0_SHIFT 3
 	u16 drop_flags;
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR (0x1<<0)
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_IP_CS_ERR_SHIFT 0
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR (0x1<<1)
 #define TSTORM_ETH_CLIENT_CONFIG_DROP_TCP_CS_ERR_SHIFT 1
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR (0x1<<2)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR_SHIFT 2
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<3)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 3
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<4)
-#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 4
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0x7FF<<5)
-#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 5
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0 (0x1<<2)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_TTL0_SHIFT 2
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR (0x1<<3)
+#define TSTORM_ETH_CLIENT_CONFIG_DROP_UDP_CS_ERR_SHIFT 3
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1 (0xFFF<<4)
+#define __TSTORM_ETH_CLIENT_CONFIG_RESERVED1_SHIFT 4
 #endif
 };
 
@@ -2011,96 +2257,112 @@ struct tstorm_eth_mac_filter_config {
 	u32 bcast_drop_all;
 	u32 bcast_accept_all;
 	u32 strict_vlan;
-	u32 __secondary_vlan_clients;
+	u32 vlan_filter[2];
+	u32 reserved;
 };
 
 
-struct rate_shaping_per_protocol {
+/*
+ * Three RX producers for ETH
+ */
+struct tstorm_eth_rx_producers {
 #if defined(__BIG_ENDIAN)
-	u16 reserved0;
-	u16 protocol_rate;
+	u16 bd_prod;
+	u16 cqe_prod;
 #elif defined(__LITTLE_ENDIAN)
-	u16 protocol_rate;
-	u16 reserved0;
+	u16 cqe_prod;
+	u16 bd_prod;
 #endif
-	u32 protocol_quota;
-	s32 current_credit;
-	u32 reserved;
-};
-
-struct rate_shaping_vars {
-	struct rate_shaping_per_protocol protocol_vars[NUM_OF_PROTOCOLS];
-	u32 pause_mask;
-	u32 periodic_stop;
-	u32 rs_periodic_timeout;
-	u32 rs_threshold;
-	u32 last_periodic_time;
-	u32 reserved;
-};
-
-struct fairness_per_protocol {
-	u32 credit_delta;
-	s32 fair_credit;
 #if defined(__BIG_ENDIAN)
-	u16 reserved0;
-	u8 state;
-	u8 weight;
+	u16 reserved;
+	u16 sge_prod;
 #elif defined(__LITTLE_ENDIAN)
-	u8 weight;
-	u8 state;
-	u16 reserved0;
+	u16 sge_prod;
+	u16 reserved;
 #endif
-	u32 reserved1;
 };
 
-struct fairness_vars {
-	struct fairness_per_protocol protocol_vars[NUM_OF_PROTOCOLS];
-	u32 upper_bound;
-	u32 port_rate;
-	u32 pause_mask;
-	u32 fair_threshold;
-};
 
-struct safc_struct {
-	u32 cur_pause_mask;
-	u32 expire_time;
+/*
+ * common flag to indicate existance of TPA.
+ */
+struct tstorm_eth_tpa_exist {
 #if defined(__BIG_ENDIAN)
-	u16 reserved0;
-	u8 cur_cos_types;
-	u8 safc_timeout_usec;
+	u16 reserved1;
+	u8 reserved0;
+	u8 tpa_exist;
 #elif defined(__LITTLE_ENDIAN)
-	u8 safc_timeout_usec;
-	u8 cur_cos_types;
-	u16 reserved0;
+	u8 tpa_exist;
+	u8 reserved0;
+	u16 reserved1;
 #endif
-	u32 reserved1;
+	u32 reserved2;
 };
 
-struct demo_struct {
+
+/*
+ * per-port SAFC demo variables
+ */
+struct cmng_flags_per_port {
 	u8 con_number[NUM_OF_PROTOCOLS];
 #if defined(__BIG_ENDIAN)
-	u8 reserved1;
 	u8 fairness_enable;
 	u8 rate_shaping_enable;
-	u8 cmng_enable;
+	u8 cmng_protocol_enable;
+	u8 cmng_vn_enable;
 #elif defined(__LITTLE_ENDIAN)
-	u8 cmng_enable;
+	u8 cmng_vn_enable;
+	u8 cmng_protocol_enable;
 	u8 rate_shaping_enable;
 	u8 fairness_enable;
-	u8 reserved1;
 #endif
 };
 
-struct cmng_struct {
-	struct rate_shaping_vars rs_vars;
-	struct fairness_vars fair_vars;
-	struct safc_struct safc_vars;
-	struct demo_struct demo_vars;
+
+/*
+ * per-port rate shaping variables
+ */
+struct rate_shaping_vars_per_port {
+	u32 rs_periodic_timeout;
+	u32 rs_threshold;
+};
+
+
+/*
+ * per-port fairness variables
+ */
+struct fairness_vars_per_port {
+	u32 upper_bound;
+	u32 fair_threshold;
+	u32 fairness_timeout;
+};
+
+
+/*
+ * per-port SAFC variables
+ */
+struct safc_struct_per_port {
+#if defined(__BIG_ENDIAN)
+	u16 __reserved0;
+	u8 cur_cos_types;
+	u8 safc_timeout_usec;
+#elif defined(__LITTLE_ENDIAN)
+	u8 safc_timeout_usec;
+	u8 cur_cos_types;
+	u16 __reserved0;
+#endif
+	u8 cos_to_protocol[MAX_COS_NUMBER];
 };
 
 
-struct cos_to_protocol {
-	u8 mask[MAX_COS_NUMBER];
+/*
+ * Per-port congestion management variables
+ */
+struct cmng_struct_per_port {
+	struct rate_shaping_vars_per_port rs_vars;
+	struct fairness_vars_per_port fair_vars;
+	struct safc_struct_per_port safc_vars;
+	struct cmng_flags_per_port flags;
 };
 
 
@@ -2161,6 +2423,16 @@ struct eth_stats_query {
 };
 
 
+/*
+ * per-vnic fairness variables
+ */
+struct fairness_vars_per_vn {
+	u32 protocol_credit_delta[NUM_OF_PROTOCOLS];
+	u32 vn_credit_delta;
+	u32 __reserved0;
+};
+
+
 /*
  * FW version stored in the Xstorm RAM
  */
@@ -2179,8 +2451,10 @@ struct fw_version {
 #define FW_VERSION_OPTIMIZED_SHIFT 0
 #define FW_VERSION_BIG_ENDIEN (0x1<<1)
 #define FW_VERSION_BIG_ENDIEN_SHIFT 1
-#define __FW_VERSION_RESERVED (0x3FFFFFFF<<2)
-#define __FW_VERSION_RESERVED_SHIFT 2
+#define FW_VERSION_CHIP_VERSION (0x3<<2)
+#define FW_VERSION_CHIP_VERSION_SHIFT 2
+#define __FW_VERSION_RESERVED (0xFFFFFFF<<4)
+#define __FW_VERSION_RESERVED_SHIFT 4
 };
 
 
@@ -2188,15 +2462,9 @@ struct fw_version {
  * FW version stored in first line of pram
  */
 struct pram_fw_version {
-#if defined(__BIG_ENDIAN)
-	u16 patch;
-	u8 primary;
-	u8 client;
-#elif defined(__LITTLE_ENDIAN)
 	u8 client;
 	u8 primary;
 	u16 patch;
-#endif
 	u8 flags;
 #define PRAM_FW_VERSION_OPTIMIZED (0x1<<0)
 #define PRAM_FW_VERSION_OPTIMIZED_SHIFT 0
@@ -2204,8 +2472,34 @@ struct pram_fw_version {
 #define PRAM_FW_VERSION_STORM_ID_SHIFT 1
 #define PRAM_FW_VERSION_BIG_ENDIEN (0x1<<3)
 #define PRAM_FW_VERSION_BIG_ENDIEN_SHIFT 3
-#define __PRAM_FW_VERSION_RESERVED0 (0xF<<4)
-#define __PRAM_FW_VERSION_RESERVED0_SHIFT 4
+#define PRAM_FW_VERSION_CHIP_VERSION (0x3<<4)
+#define PRAM_FW_VERSION_CHIP_VERSION_SHIFT 4
+#define __PRAM_FW_VERSION_RESERVED0 (0x3<<6)
+#define __PRAM_FW_VERSION_RESERVED0_SHIFT 6
+};
+
+
+/*
+ * a single rate shaping counter. can be used as protocol or vnic counter
+ */
+struct rate_shaping_counter {
+	u32 quota;
+#if defined(__BIG_ENDIAN)
+	u16 __reserved0;
+	u16 rate;
+#elif defined(__LITTLE_ENDIAN)
+	u16 rate;
+	u16 __reserved0;
+#endif
+};
+
+
+/*
+ * per-vnic rate shaping variables
+ */
+struct rate_shaping_vars_per_vn {
+	struct rate_shaping_counter protocol_counters[NUM_OF_PROTOCOLS];
+	struct rate_shaping_counter vn_counter;
 };
 
 
diff --git a/drivers/net/bnx2x_init.h b/drivers/net/bnx2x_init.h
index 5a4e82b9e7bf..4c7750789b62 100644
--- a/drivers/net/bnx2x_init.h
+++ b/drivers/net/bnx2x_init.h
@@ -226,28 +226,28 @@ static const u32 *bnx2x_sel_blob(u32 addr, const u32 *data, int is_e1)
 			       tsem_int_table_data_e1h;
 	else
 		IF_IS_INT_TABLE_ADDR(CSEM_REG_INT_TABLE, addr)
-		data = is_e1 ? csem_int_table_data_e1 :
-			       csem_int_table_data_e1h;
+			data = is_e1 ? csem_int_table_data_e1 :
+				       csem_int_table_data_e1h;
 	else
 		IF_IS_INT_TABLE_ADDR(USEM_REG_INT_TABLE, addr)
-		data = is_e1 ? usem_int_table_data_e1 :
-			       usem_int_table_data_e1h;
+			data = is_e1 ? usem_int_table_data_e1 :
+				       usem_int_table_data_e1h;
 	else
 		IF_IS_INT_TABLE_ADDR(XSEM_REG_INT_TABLE, addr)
-		data = is_e1 ? xsem_int_table_data_e1 :
-			       xsem_int_table_data_e1h;
+			data = is_e1 ? xsem_int_table_data_e1 :
+				       xsem_int_table_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(TSEM_REG_PRAM, addr)
-		data = is_e1 ? tsem_pram_data_e1 : tsem_pram_data_e1h;
+			data = is_e1 ? tsem_pram_data_e1 : tsem_pram_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(CSEM_REG_PRAM, addr)
-		data = is_e1 ? csem_pram_data_e1 : csem_pram_data_e1h;
+			data = is_e1 ? csem_pram_data_e1 : csem_pram_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(USEM_REG_PRAM, addr)
-		data = is_e1 ? usem_pram_data_e1 : usem_pram_data_e1h;
+			data = is_e1 ? usem_pram_data_e1 : usem_pram_data_e1h;
 	else
 		IF_IS_PRAM_ADDR(XSEM_REG_PRAM, addr)
-		data = is_e1 ? xsem_pram_data_e1 : xsem_pram_data_e1h;
+			data = is_e1 ? xsem_pram_data_e1 : xsem_pram_data_e1h;
 
 	return data;
 }
diff --git a/drivers/net/bnx2x_link.c b/drivers/net/bnx2x_link.c
index 22586ebd7b1e..ff2743db10d9 100644
--- a/drivers/net/bnx2x_link.c
+++ b/drivers/net/bnx2x_link.c
@@ -3572,7 +3572,8 @@ u8 bnx2x_set_led(struct bnx2x *bp, u8 port, u8 mode, u32 speed,
 			   LED_BLINK_RATE_VAL);
 		REG_WR(bp, NIG_REG_LED_CONTROL_BLINK_RATE_ENA_P0 +
 			   port*4, 1);
-		if (((speed == SPEED_2500) ||
+		if (!CHIP_IS_E1H(bp) &&
+		    ((speed == SPEED_2500) ||
 		     (speed == SPEED_1000) ||
 		     (speed == SPEED_100) ||
 		     (speed == SPEED_10))) {
@@ -3753,6 +3754,14 @@ u8 bnx2x_phy_init(struct link_params *params, struct link_vars *vars)
 		vars->duplex = DUPLEX_FULL;
 		vars->flow_ctrl = FLOW_CTRL_NONE;
 		vars->link_status = (LINK_STATUS_LINK_UP | LINK_10GTFD);
+		/* enable on E1.5 FPGA */
+		if (CHIP_IS_E1H(bp)) {
+			vars->flow_ctrl |=
+				(FLOW_CTRL_TX | FLOW_CTRL_RX);
+			vars->link_status |=
+					(LINK_STATUS_TX_FLOW_CONTROL_ENABLED |
+					 LINK_STATUS_RX_FLOW_CONTROL_ENABLED);
+		}
 
 		bnx2x_emac_enable(params, vars, 0);
 		bnx2x_pbf_update(params, vars->flow_ctrl, vars->line_speed);
diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c
index efa942688f84..90b54e4c5c3b 100644
--- a/drivers/net/bnx2x_main.c
+++ b/drivers/net/bnx2x_main.c
@@ -1,4 +1,4 @@
-/* bnx2x.c: Broadcom Everest network driver.
+/* bnx2x_main.c: Broadcom Everest network driver.
  *
  * Copyright (c) 2007-2008 Broadcom Corporation
  *
@@ -15,12 +15,6 @@
  *
  */
 
-/* define this to make the driver freeze on error
- * to allow getting debug info
- * (you will need to reboot afterwards)
- */
-/*#define BNX2X_STOP_ON_ERROR*/
-
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/kernel.h>
@@ -46,16 +40,17 @@
 #include <linux/mii.h>
 #ifdef NETIF_F_HW_VLAN_TX
 	#include <linux/if_vlan.h>
-	#define BCM_VLAN 1
 #endif
 #include <net/ip.h>
 #include <net/tcp.h>
 #include <net/checksum.h>
+#include <linux/version.h>
+#include <net/ip6_checksum.h>
 #include <linux/workqueue.h>
 #include <linux/crc32.h>
+#include <linux/crc32c.h>
 #include <linux/prefetch.h>
 #include <linux/zlib.h>
-#include <linux/version.h>
 #include <linux/io.h>
 
 #include "bnx2x_reg.h"
@@ -67,13 +62,13 @@
 
 #define DRV_MODULE_VERSION      "1.42.4"
 #define DRV_MODULE_RELDATE      "2008/4/9"
-#define BNX2X_BC_VER    	0x040200
+#define BNX2X_BC_VER		0x040200
 
-/* Time in jiffies before concluding the transmitter is hung. */
-#define TX_TIMEOUT      	(5*HZ)
+/* Time in jiffies before concluding the transmitter is hung */
+#define TX_TIMEOUT		(5*HZ)
 
 static char version[] __devinitdata =
-	"Broadcom NetXtreme II 5771X 10Gigabit Ethernet Driver "
+	"Broadcom NetXtreme II 5771x 10Gigabit Ethernet Driver "
 	DRV_MODULE_NAME " " DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
 
 MODULE_AUTHOR("Eliezer Tamir");
@@ -83,20 +78,19 @@ MODULE_VERSION(DRV_MODULE_VERSION);
 
 static int use_inta;
 static int poll;
-static int onefunc;
-static int nomcp;
 static int debug;
+static int nomcp;
+static int load_count[3]; /* 0-common, 1-port0, 2-port1 */
 static int use_multi;
 
 module_param(use_inta, int, 0);
 module_param(poll, int, 0);
-module_param(onefunc, int, 0);
 module_param(debug, int, 0);
+module_param(nomcp, int, 0);
 MODULE_PARM_DESC(use_inta, "use INT#A instead of MSI-X");
 MODULE_PARM_DESC(poll, "use polling (for debug)");
-MODULE_PARM_DESC(onefunc, "enable only first function");
-MODULE_PARM_DESC(nomcp, "ignore management CPU (Implies onefunc)");
 MODULE_PARM_DESC(debug, "default debug msglevel");
+MODULE_PARM_DESC(nomcp, "ignore management CPU");
 
 #ifdef BNX2X_MULTI
 module_param(use_multi, int, 0);
@@ -105,18 +99,27 @@ MODULE_PARM_DESC(use_multi, "use per-CPU queues");
 
 enum bnx2x_board_type {
 	BCM57710 = 0,
+	BCM57711 = 1,
+	BCM57711E = 2,
 };
 
-/* indexed by board_t, above */
+/* indexed by board_type, above */
 static struct {
 	char *name;
 } board_info[] __devinitdata = {
-	{ "Broadcom NetXtreme II BCM57710 XGb" }
+	{ "Broadcom NetXtreme II BCM57710 XGb" },
+	{ "Broadcom NetXtreme II BCM57711 XGb" },
+	{ "Broadcom NetXtreme II BCM57711E XGb" }
 };
 
+
 static const struct pci_device_id bnx2x_pci_tbl[] = {
 	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_57710,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM57710 },
+	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_57711,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM57711 },
+	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_57711E,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM57711E },
 	{ 0 }
 };
 
@@ -201,7 +204,8 @@ void bnx2x_write_dmae(struct bnx2x *bp, dma_addr_t dma_addr, u32 dst_addr,
 #else
 			DMAE_CMD_ENDIANITY_DW_SWAP |
 #endif
-			(bp->port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+			(BP_PORT(bp) ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0) |
+			(BP_E1HVN(bp) << DMAE_CMD_E1HVN_SHIFT));
 	dmae->src_addr_lo = U64_LO(dma_addr);
 	dmae->src_addr_hi = U64_HI(dma_addr);
 	dmae->dst_addr_lo = dst_addr >> 2;
@@ -224,7 +228,7 @@ void bnx2x_write_dmae(struct bnx2x *bp, dma_addr_t dma_addr, u32 dst_addr,
 
 	*wb_comp = 0;
 
-	bnx2x_post_dmae(bp, dmae, (bp->port)*MAX_DMAE_C_PER_PORT);
+	bnx2x_post_dmae(bp, dmae, INIT_DMAE_C(bp));
 
 	udelay(5);
 
@@ -277,7 +281,8 @@ void bnx2x_read_dmae(struct bnx2x *bp, u32 src_addr, u32 len32)
 #else
 			DMAE_CMD_ENDIANITY_DW_SWAP |
 #endif
-			(bp->port ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0));
+			(BP_PORT(bp) ? DMAE_CMD_PORT_1 : DMAE_CMD_PORT_0) |
+			(BP_E1HVN(bp) << DMAE_CMD_E1HVN_SHIFT));
 	dmae->src_addr_lo = src_addr >> 2;
 	dmae->src_addr_hi = 0;
 	dmae->dst_addr_lo = U64_LO(bnx2x_sp_mapping(bp, wb_data));
@@ -297,7 +302,7 @@ void bnx2x_read_dmae(struct bnx2x *bp, u32 src_addr, u32 len32)
 
 	*wb_comp = 0;
 
-	bnx2x_post_dmae(bp, dmae, (bp->port)*MAX_DMAE_C_PER_PORT);
+	bnx2x_post_dmae(bp, dmae, INIT_DMAE_C(bp));
 
 	udelay(5);
 
@@ -345,47 +350,122 @@ static u64 bnx2x_wb_rd(struct bnx2x *bp, int reg)
 
 static int bnx2x_mc_assert(struct bnx2x *bp)
 {
-	int i, j, rc = 0;
 	char last_idx;
-	const char storm[] = {"XTCU"};
-	const u32 intmem_base[] = {
-		BAR_XSTRORM_INTMEM,
-		BAR_TSTRORM_INTMEM,
-		BAR_CSTRORM_INTMEM,
-		BAR_USTRORM_INTMEM
-	};
-
-	/* Go through all instances of all SEMIs */
-	for (i = 0; i < 4; i++) {
-		last_idx = REG_RD8(bp, XSTORM_ASSERT_LIST_INDEX_OFFSET +
-				   intmem_base[i]);
-		if (last_idx)
-			BNX2X_LOG("DATA %cSTORM_ASSERT_LIST_INDEX 0x%x\n",
-				  storm[i], last_idx);
-
-		/* print the asserts */
-		for (j = 0; j < STROM_ASSERT_ARRAY_SIZE; j++) {
-			u32 row0, row1, row2, row3;
-
-			row0 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) +
-				      intmem_base[i]);
-			row1 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 4 +
-				      intmem_base[i]);
-			row2 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 8 +
-				      intmem_base[i]);
-			row3 = REG_RD(bp, XSTORM_ASSERT_LIST_OFFSET(j) + 12 +
-				      intmem_base[i]);
-
-			if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
-				BNX2X_LOG("DATA %cSTORM_ASSERT_INDEX 0x%x ="
-					  " 0x%08x 0x%08x 0x%08x 0x%08x\n",
-					  storm[i], j, row3, row2, row1, row0);
-				rc++;
-			} else {
-				break;
-			}
+	int i, rc = 0;
+	u32 row0, row1, row2, row3;
+
+	/* XSTORM */
+	last_idx = REG_RD8(bp, BAR_XSTRORM_INTMEM +
+			   XSTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("XSTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_XSTRORM_INTMEM +
+			      XSTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("XSTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
+		}
+	}
+
+	/* TSTORM */
+	last_idx = REG_RD8(bp, BAR_TSTRORM_INTMEM +
+			   TSTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("TSTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_TSTRORM_INTMEM +
+			      TSTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("TSTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
+		}
+	}
+
+	/* CSTORM */
+	last_idx = REG_RD8(bp, BAR_CSTRORM_INTMEM +
+			   CSTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("CSTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_CSTRORM_INTMEM +
+			      CSTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("CSTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
+		}
+	}
+
+	/* USTORM */
+	last_idx = REG_RD8(bp, BAR_USTRORM_INTMEM +
+			   USTORM_ASSERT_LIST_INDEX_OFFSET);
+	if (last_idx)
+		BNX2X_ERR("USTORM_ASSERT_LIST_INDEX 0x%x\n", last_idx);
+
+	/* print the asserts */
+	for (i = 0; i < STROM_ASSERT_ARRAY_SIZE; i++) {
+
+		row0 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i));
+		row1 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i) + 4);
+		row2 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i) + 8);
+		row3 = REG_RD(bp, BAR_USTRORM_INTMEM +
+			      USTORM_ASSERT_LIST_OFFSET(i) + 12);
+
+		if (row0 != COMMON_ASM_INVALID_ASSERT_OPCODE) {
+			BNX2X_ERR("USTORM_ASSERT_INDEX 0x%x = 0x%08x"
+				  " 0x%08x 0x%08x 0x%08x\n",
+				  i, row3, row2, row1, row0);
+			rc++;
+		} else {
+			break;
 		}
 	}
+
 	return rc;
 }
 
@@ -428,14 +508,16 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 		struct eth_tx_db_data *hw_prods = fp->hw_tx_prods;
 
 		BNX2X_ERR("queue[%d]: tx_pkt_prod(%x)  tx_pkt_cons(%x)"
-			  "  tx_bd_prod(%x)  tx_bd_cons(%x)  *tx_cons_sb(%x)"
-			  "  *rx_cons_sb(%x)  rx_comp_prod(%x)"
-			  "  rx_comp_cons(%x)  fp_c_idx(%x)  fp_u_idx(%x)"
-			  "  bd data(%x,%x)\n",
+			  "  tx_bd_prod(%x)  tx_bd_cons(%x)  *tx_cons_sb(%x)\n",
 			  i, fp->tx_pkt_prod, fp->tx_pkt_cons, fp->tx_bd_prod,
-			  fp->tx_bd_cons, *fp->tx_cons_sb, *fp->rx_cons_sb,
-			  fp->rx_comp_prod, fp->rx_comp_cons, fp->fp_c_idx,
-			  fp->fp_u_idx, hw_prods->packets_prod,
+			  fp->tx_bd_cons, le16_to_cpu(*fp->tx_cons_sb));
+		BNX2X_ERR("          rx_comp_prod(%x)  rx_comp_cons(%x)"
+			  "  *rx_cons_sb(%x)\n",
+			  fp->rx_comp_prod, fp->rx_comp_cons,
+			  le16_to_cpu(*fp->rx_cons_sb));
+		BNX2X_ERR("          fp_c_idx(%x)  fp_u_idx(%x)"
+			  "  bd data(%x,%x)\n",
+			  fp->fp_c_idx, fp->fp_u_idx, hw_prods->packets_prod,
 			  hw_prods->bds_prod);
 
 		start = TX_BD(le16_to_cpu(*fp->tx_cons_sb) - 10);
@@ -463,7 +545,7 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 			struct sw_rx_bd *sw_bd = &fp->rx_buf_ring[j];
 
 			BNX2X_ERR("rx_bd[%x]=[%x:%x]  sw_bd=[%p]\n",
-				  j, rx_bd[0], rx_bd[1], sw_bd->skb);
+				  j, rx_bd[1], rx_bd[0], sw_bd->skb);
 		}
 
 		start = RCQ_BD(fp->rx_comp_cons - 10);
@@ -482,7 +564,7 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 		  bp->def_c_idx, bp->def_u_idx, bp->def_x_idx, bp->def_t_idx,
 		  bp->def_att_idx, bp->attn_state, bp->spq_prod_idx);
 
-
+	bnx2x_fw_dump(bp);
 	bnx2x_mc_assert(bp);
 	BNX2X_ERR("end crash dump -----------------\n");
 
@@ -492,7 +574,7 @@ static void bnx2x_panic_dump(struct bnx2x *bp)
 
 static void bnx2x_int_enable(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	u32 addr = port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0;
 	u32 val = REG_RD(bp, addr);
 	int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
@@ -507,7 +589,6 @@ static void bnx2x_int_enable(struct bnx2x *bp)
 			HC_CONFIG_0_REG_INT_LINE_EN_0 |
 			HC_CONFIG_0_REG_ATTN_BIT_EN_0);
 
-		/* Errata A0.158 workaround */
 		DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x)  MSI-X %d\n",
 		   val, port, addr, msix);
 
@@ -520,11 +601,25 @@ static void bnx2x_int_enable(struct bnx2x *bp)
 	   val, port, addr, msix);
 
 	REG_WR(bp, addr, val);
+
+	if (CHIP_IS_E1H(bp)) {
+		/* init leading/trailing edge */
+		if (IS_E1HMF(bp)) {
+			val = (0xfe0f | (1 << (BP_E1HVN(bp) + 4)));
+			if (bp->port.pmf)
+				/* enable nig attention */
+				val |= 0x0100;
+		} else
+			val = 0xffff;
+
+		REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, val);
+		REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, val);
+	}
 }
 
 static void bnx2x_int_disable(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	u32 addr = port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0;
 	u32 val = REG_RD(bp, addr);
 
@@ -543,10 +638,10 @@ static void bnx2x_int_disable(struct bnx2x *bp)
 
 static void bnx2x_int_disable_sync(struct bnx2x *bp)
 {
-
 	int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
 	int i;
 
+	/* disable interrupt handling */
 	atomic_inc(&bp->intr_sem);
 	/* prevent the HW from sending interrupts */
 	bnx2x_int_disable(bp);
@@ -563,30 +658,29 @@ static void bnx2x_int_disable_sync(struct bnx2x *bp)
 
 	/* make sure sp_task is not running */
 	cancel_work_sync(&bp->sp_task);
-
 }
 
-/* fast path code */
+/* fast path */
 
 /*
- * general service functions
+ * General service functions
  */
 
-static inline void bnx2x_ack_sb(struct bnx2x *bp, u8 id,
+static inline void bnx2x_ack_sb(struct bnx2x *bp, u8 sb_id,
 				u8 storm, u16 index, u8 op, u8 update)
 {
-	u32 igu_addr = (IGU_ADDR_INT_ACK + IGU_PORT_BASE * bp->port) * 8;
+	u32 igu_addr = (IGU_ADDR_INT_ACK + IGU_FUNC_BASE * BP_FUNC(bp)) * 8;
 	struct igu_ack_register igu_ack;
 
 	igu_ack.status_block_index = index;
 	igu_ack.sb_id_and_flags =
-			((id << IGU_ACK_REGISTER_STATUS_BLOCK_ID_SHIFT) |
+			((sb_id << IGU_ACK_REGISTER_STATUS_BLOCK_ID_SHIFT) |
 			 (storm << IGU_ACK_REGISTER_STORM_ID_SHIFT) |
 			 (update << IGU_ACK_REGISTER_UPDATE_INDEX_SHIFT) |
 			 (op << IGU_ACK_REGISTER_INTERRUPT_MODE_SHIFT));
 
-/*      DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
-	   (*(u32 *)&igu_ack), BAR_IGU_INTMEM + igu_addr); */
+	DP(BNX2X_MSG_OFF, "write 0x%08x to IGU addr 0x%x\n",
+	   (*(u32 *)&igu_ack), BAR_IGU_INTMEM + igu_addr);
 	REG_WR(bp, BAR_IGU_INTMEM + igu_addr, (*(u32 *)&igu_ack));
 }
 
@@ -614,8 +708,9 @@ static inline int bnx2x_has_work(struct bnx2x_fastpath *fp)
 	if ((rx_cons_sb & MAX_RCQ_DESC_CNT) == MAX_RCQ_DESC_CNT)
 		rx_cons_sb++;
 
-	if ((rx_cons_sb != fp->rx_comp_cons) ||
-	    (le16_to_cpu(*fp->tx_cons_sb) != fp->tx_pkt_cons))
+	if ((fp->rx_comp_cons != rx_cons_sb) ||
+	    (fp->tx_pkt_prod != le16_to_cpu(*fp->tx_cons_sb)) ||
+	    (fp->tx_pkt_prod != fp->tx_pkt_cons))
 		return 1;
 
 	return 0;
@@ -623,11 +718,11 @@ static inline int bnx2x_has_work(struct bnx2x_fastpath *fp)
 
 static u16 bnx2x_ack_int(struct bnx2x *bp)
 {
-	u32 igu_addr = (IGU_ADDR_SIMD_MASK + IGU_PORT_BASE * bp->port) * 8;
+	u32 igu_addr = (IGU_ADDR_SIMD_MASK + IGU_FUNC_BASE * BP_FUNC(bp)) * 8;
 	u32 result = REG_RD(bp, BAR_IGU_INTMEM + igu_addr);
 
-/*      DP(NETIF_MSG_INTR, "read 0x%08x from IGU addr 0x%x\n",
-	   result, BAR_IGU_INTMEM + igu_addr); */
+	DP(BNX2X_MSG_OFF, "read 0x%08x from IGU addr 0x%x\n",
+	   result, BAR_IGU_INTMEM + igu_addr);
 
 #ifdef IGU_DEBUG
 #warning IGU_DEBUG active
@@ -653,7 +748,7 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 	struct sw_tx_bd *tx_buf = &fp->tx_buf_ring[idx];
 	struct eth_tx_bd *tx_bd;
 	struct sk_buff *skb = tx_buf->skb;
-	u16 bd_idx = tx_buf->first_bd;
+	u16 bd_idx = TX_BD(tx_buf->first_bd), new_cons;
 	int nbd;
 
 	DP(BNX2X_MSG_OFF, "pkt_idx %d  buff @(%p)->skb %p\n",
@@ -666,9 +761,10 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 			 BD_UNMAP_LEN(tx_bd), PCI_DMA_TODEVICE);
 
 	nbd = le16_to_cpu(tx_bd->nbd) - 1;
+	new_cons = nbd + tx_buf->first_bd;
 #ifdef BNX2X_STOP_ON_ERROR
 	if (nbd > (MAX_SKB_FRAGS + 2)) {
-		BNX2X_ERR("bad nbd!\n");
+		BNX2X_ERR("BAD nbd!\n");
 		bnx2x_panic();
 	}
 #endif
@@ -708,32 +804,30 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 	tx_buf->first_bd = 0;
 	tx_buf->skb = NULL;
 
-	return bd_idx;
+	return new_cons;
 }
 
-static inline u32 bnx2x_tx_avail(struct bnx2x_fastpath *fp)
+static inline u16 bnx2x_tx_avail(struct bnx2x_fastpath *fp)
 {
-	u16 used;
-	u32 prod;
-	u32 cons;
+	s16 used;
+	u16 prod;
+	u16 cons;
 
-	/* Tell compiler that prod and cons can change */
-	barrier();
+	barrier(); /* Tell compiler that prod and cons can change */
 	prod = fp->tx_bd_prod;
 	cons = fp->tx_bd_cons;
 
-	used = (NUM_TX_BD - NUM_TX_RINGS + prod - cons +
-		(cons / TX_DESC_CNT) - (prod / TX_DESC_CNT));
-
-	if (prod >= cons) {
-		/* used = prod - cons - prod/size + cons/size */
-		used -= NUM_TX_BD - NUM_TX_RINGS;
-	}
+	/* NUM_TX_RINGS = number of "next-page" entries
+	   It will be used as a threshold */
+	used = SUB_S16(prod, cons) + (s16)NUM_TX_RINGS;
 
+#ifdef BNX2X_STOP_ON_ERROR
+	BUG_TRAP(used >= 0);
 	BUG_TRAP(used <= fp->bp->tx_ring_size);
 	BUG_TRAP((fp->bp->tx_ring_size - used) <= MAX_TX_AVAIL);
+#endif
 
-	return (fp->bp->tx_ring_size - used);
+	return (s16)(fp->bp->tx_ring_size) - used;
 }
 
 static void bnx2x_tx_int(struct bnx2x_fastpath *fp, int work)
@@ -757,10 +851,10 @@ static void bnx2x_tx_int(struct bnx2x_fastpath *fp, int work)
 
 		/* prefetch(bp->tx_buf_ring[pkt_cons].skb); */
 
-		DP(NETIF_MSG_TX_DONE, "hw_cons %u  sw_cons %u  pkt_cons %d\n",
+		DP(NETIF_MSG_TX_DONE, "hw_cons %u  sw_cons %u  pkt_cons %u\n",
 		   hw_cons, sw_cons, pkt_cons);
 
-/*      	if (NEXT_TX_IDX(sw_cons) != hw_cons) {
+/*		if (NEXT_TX_IDX(sw_cons) != hw_cons) {
 			rmb();
 			prefetch(fp->tx_buf_ring[NEXT_TX_IDX(sw_cons)].skb);
 		}
@@ -793,7 +887,6 @@ static void bnx2x_tx_int(struct bnx2x_fastpath *fp, int work)
 			netif_wake_queue(bp->dev);
 
 		netif_tx_unlock(bp->dev);
-
 	}
 }
 
@@ -804,13 +897,14 @@ static void bnx2x_sp_event(struct bnx2x_fastpath *fp,
 	int cid = SW_CID(rr_cqe->ramrod_cqe.conn_and_cmd_data);
 	int command = CQE_CMD(rr_cqe->ramrod_cqe.conn_and_cmd_data);
 
-	DP(NETIF_MSG_RX_STATUS,
+	DP(BNX2X_MSG_SP,
 	   "fp %d  cid %d  got ramrod #%d  state is %x  type is %d\n",
-	   fp->index, cid, command, bp->state, rr_cqe->ramrod_cqe.type);
+	   FP_IDX(fp), cid, command, bp->state,
+	   rr_cqe->ramrod_cqe.ramrod_type);
 
 	bp->spq_left++;
 
-	if (fp->index) {
+	if (FP_IDX(fp)) {
 		switch (command | fp->state) {
 		case (RAMROD_CMD_ID_ETH_CLIENT_SETUP |
 						BNX2X_FP_STATE_OPENING):
@@ -826,10 +920,11 @@ static void bnx2x_sp_event(struct bnx2x_fastpath *fp,
 			break;
 
 		default:
-			BNX2X_ERR("unexpected MC reply(%d)  state is %x\n",
-				  command, fp->state);
+			BNX2X_ERR("unexpected MC reply (%d)  "
+				  "fp->state is %x\n", command, fp->state);
+			break;
 		}
-		mb(); /* force bnx2x_wait_ramrod to see the change */
+		mb(); /* force bnx2x_wait_ramrod() to see the change */
 		return;
 	}
 
@@ -846,25 +941,25 @@ static void bnx2x_sp_event(struct bnx2x_fastpath *fp,
 		break;
 
 	case (RAMROD_CMD_ID_ETH_CFC_DEL | BNX2X_STATE_CLOSING_WAIT4_HALT):
-		DP(NETIF_MSG_IFDOWN, "got delete ramrod for MULTI[%d]\n",
-		   cid);
+		DP(NETIF_MSG_IFDOWN, "got delete ramrod for MULTI[%d]\n", cid);
 		bnx2x_fp(bp, cid, state) = BNX2X_FP_STATE_CLOSED;
 		break;
 
 	case (RAMROD_CMD_ID_ETH_SET_MAC | BNX2X_STATE_OPEN):
+	case (RAMROD_CMD_ID_ETH_SET_MAC | BNX2X_STATE_DIAG):
 		DP(NETIF_MSG_IFUP, "got set mac ramrod\n");
 		break;
 
 	case (RAMROD_CMD_ID_ETH_SET_MAC | BNX2X_STATE_CLOSING_WAIT4_HALT):
-		DP(NETIF_MSG_IFUP, "got (un)set mac ramrod\n");
+		DP(NETIF_MSG_IFDOWN, "got (un)set mac ramrod\n");
 		break;
 
 	default:
-		BNX2X_ERR("unexpected ramrod (%d)  state is %x\n",
+		BNX2X_ERR("unexpected MC reply (%d)  bp->state is %x\n",
 			  command, bp->state);
+		break;
 	}
-
-	mb(); /* force bnx2x_wait_ramrod to see the change */
+	mb(); /* force bnx2x_wait_ramrod() to see the change */
 }
 
 static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp,
@@ -882,7 +977,6 @@ static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp,
 	mapping = pci_map_single(bp->pdev, skb->data, bp->rx_buf_use_size,
 				 PCI_DMA_FROMDEVICE);
 	if (unlikely(dma_mapping_error(mapping))) {
-
 		dev_kfree_skb(skb);
 		return -ENOMEM;
 	}
@@ -924,7 +1018,7 @@ static void bnx2x_reuse_rx_skb(struct bnx2x_fastpath *fp,
 static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 {
 	struct bnx2x *bp = fp->bp;
-	u16 bd_cons, bd_prod, comp_ring_cons;
+	u16 bd_cons, bd_prod, bd_prod_fw, comp_ring_cons;
 	u16 hw_comp_cons, sw_comp_cons, sw_comp_prod;
 	int rx_pkt = 0;
 
@@ -933,12 +1027,15 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 		return 0;
 #endif
 
+	/* CQ "next element" is of the size of the regular element,
+	   that's why it's ok here */
 	hw_comp_cons = le16_to_cpu(*fp->rx_cons_sb);
 	if ((hw_comp_cons & MAX_RCQ_DESC_CNT) == MAX_RCQ_DESC_CNT)
 		hw_comp_cons++;
 
 	bd_cons = fp->rx_bd_cons;
 	bd_prod = fp->rx_bd_prod;
+	bd_prod_fw = bd_prod;
 	sw_comp_cons = fp->rx_comp_cons;
 	sw_comp_prod = fp->rx_comp_prod;
 
@@ -949,34 +1046,31 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 
 	DP(NETIF_MSG_RX_STATUS,
 	   "queue[%d]:  hw_comp_cons %u  sw_comp_cons %u\n",
-	   fp->index, hw_comp_cons, sw_comp_cons);
+	   FP_IDX(fp), hw_comp_cons, sw_comp_cons);
 
 	while (sw_comp_cons != hw_comp_cons) {
-		unsigned int len, pad;
-		struct sw_rx_bd *rx_buf;
+		struct sw_rx_bd *rx_buf = NULL;
 		struct sk_buff *skb;
 		union eth_rx_cqe *cqe;
+		u8 cqe_fp_flags;
+		u16 len, pad;
 
 		comp_ring_cons = RCQ_BD(sw_comp_cons);
 		bd_prod = RX_BD(bd_prod);
 		bd_cons = RX_BD(bd_cons);
 
 		cqe = &fp->rx_comp_ring[comp_ring_cons];
+		cqe_fp_flags = cqe->fast_path_cqe.type_error_flags;
 
-		DP(NETIF_MSG_RX_STATUS, "hw_comp_cons %u  sw_comp_cons %u"
-		   "  comp_ring (%u)  bd_ring (%u,%u)\n",
-		   hw_comp_cons, sw_comp_cons,
-		   comp_ring_cons, bd_prod, bd_cons);
 		DP(NETIF_MSG_RX_STATUS, "CQE type %x  err %x  status %x"
-		   "  queue %x  vlan %x  len %x\n",
-		   cqe->fast_path_cqe.type,
-		   cqe->fast_path_cqe.error_type_flags,
-		   cqe->fast_path_cqe.status_flags,
+		   "  queue %x  vlan %x  len %u\n", CQE_TYPE(cqe_fp_flags),
+		   cqe_fp_flags, cqe->fast_path_cqe.status_flags,
 		   cqe->fast_path_cqe.rss_hash_result,
-		   cqe->fast_path_cqe.vlan_tag, cqe->fast_path_cqe.pkt_len);
+		   le16_to_cpu(cqe->fast_path_cqe.vlan_tag),
+		   le16_to_cpu(cqe->fast_path_cqe.pkt_len));
 
 		/* is this a slowpath msg? */
-		if (unlikely(cqe->fast_path_cqe.type)) {
+		if (unlikely(CQE_TYPE(cqe_fp_flags))) {
 			bnx2x_sp_event(fp, cqe);
 			goto next_cqe;
 
@@ -984,7 +1078,6 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 		} else {
 			rx_buf = &fp->rx_buf_ring[bd_cons];
 			skb = rx_buf->skb;
-
 			len = le16_to_cpu(cqe->fast_path_cqe.pkt_len);
 			pad = cqe->fast_path_cqe.placement_offset;
 
@@ -996,13 +1089,11 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 			prefetch(((char *)(skb)) + 128);
 
 			/* is this an error packet? */
-			if (unlikely(cqe->fast_path_cqe.error_type_flags &
-							ETH_RX_ERROR_FALGS)) {
+			if (unlikely(cqe_fp_flags & ETH_RX_ERROR_FALGS)) {
 			/* do we sometimes forward error packets anyway? */
 				DP(NETIF_MSG_RX_ERR,
-				   "ERROR flags(%u) Rx packet(%u)\n",
-				   cqe->fast_path_cqe.error_type_flags,
-				   sw_comp_cons);
+				   "ERROR  flags %x  rx packet %u\n",
+				   cqe_fp_flags, sw_comp_cons);
 				/* TBD make sure MC counts this as a drop */
 				goto reuse_rx;
 			}
@@ -1018,7 +1109,7 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 							   len + pad);
 				if (new_skb == NULL) {
 					DP(NETIF_MSG_RX_ERR,
-					   "ERROR packet dropped "
+					   "ERROR  packet dropped "
 					   "because of alloc failure\n");
 					/* TBD count this as a drop? */
 					goto reuse_rx;
@@ -1044,7 +1135,7 @@ static int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
 
 			} else {
 				DP(NETIF_MSG_RX_ERR,
-				   "ERROR packet dropped because "
+				   "ERROR  packet dropped because "
 				   "of alloc failure\n");
 reuse_rx:
 				bnx2x_reuse_rx_skb(fp, skb, bd_cons, bd_prod);
@@ -1061,14 +1152,14 @@ reuse_rx:
 		}
 
 #ifdef BCM_VLAN
-		if ((le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags)
-				& PARSING_FLAGS_NUMBER_OF_NESTED_VLANS)
-		    && (bp->vlgrp != NULL))
+		if ((bp->vlgrp != NULL) &&
+		    (le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags) &
+		     PARSING_FLAGS_VLAN))
 			vlan_hwaccel_receive_skb(skb, bp->vlgrp,
 				le16_to_cpu(cqe->fast_path_cqe.vlan_tag));
 		else
 #endif
-		netif_receive_skb(skb);
+			netif_receive_skb(skb);
 
 		bp->dev->last_rx = jiffies;
 
@@ -1077,22 +1168,25 @@ next_rx:
 
 		bd_cons = NEXT_RX_IDX(bd_cons);
 		bd_prod = NEXT_RX_IDX(bd_prod);
+		bd_prod_fw = NEXT_RX_IDX(bd_prod_fw);
+		rx_pkt++;
 next_cqe:
 		sw_comp_prod = NEXT_RCQ_IDX(sw_comp_prod);
 		sw_comp_cons = NEXT_RCQ_IDX(sw_comp_cons);
-		rx_pkt++;
 
-		if ((rx_pkt == budget))
+		if (rx_pkt == budget)
 			break;
 	} /* while */
 
 	fp->rx_bd_cons = bd_cons;
-	fp->rx_bd_prod = bd_prod;
+	fp->rx_bd_prod = bd_prod_fw;
 	fp->rx_comp_cons = sw_comp_cons;
 	fp->rx_comp_prod = sw_comp_prod;
 
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       TSTORM_RCQ_PROD_OFFSET(bp->port, fp->index), sw_comp_prod);
+		TSTORM_RX_PRODS_OFFSET(BP_PORT(bp), FP_CL_ID(fp)),
+		sw_comp_prod);
+
 
 	mmiowb(); /* keep prod updates ordered */
 
@@ -1107,10 +1201,11 @@ static irqreturn_t bnx2x_msix_fp_int(int irq, void *fp_cookie)
 	struct bnx2x_fastpath *fp = fp_cookie;
 	struct bnx2x *bp = fp->bp;
 	struct net_device *dev = bp->dev;
-	int index = fp->index;
+	int index = FP_IDX(fp);
 
-	DP(NETIF_MSG_INTR, "got an msix interrupt on [%d]\n", index);
-	bnx2x_ack_sb(bp, index, USTORM_ID, 0, IGU_INT_DISABLE, 0);
+	DP(BNX2X_MSG_FP, "got an MSI-X interrupt on IDX:SB [%d:%d]\n",
+	   index, FP_SB_ID(fp));
+	bnx2x_ack_sb(bp, FP_SB_ID(fp), USTORM_ID, 0, IGU_INT_DISABLE, 0);
 
 #ifdef BNX2X_STOP_ON_ERROR
 	if (unlikely(bp->panic))
@@ -1123,6 +1218,7 @@ static irqreturn_t bnx2x_msix_fp_int(int irq, void *fp_cookie)
 	prefetch(&fp->status_blk->u_status_block.status_block_index);
 
 	netif_rx_schedule(dev, &bnx2x_fp(bp, index, napi));
+
 	return IRQ_HANDLED;
 }
 
@@ -1131,26 +1227,28 @@ static irqreturn_t bnx2x_interrupt(int irq, void *dev_instance)
 	struct net_device *dev = dev_instance;
 	struct bnx2x *bp = netdev_priv(dev);
 	u16 status = bnx2x_ack_int(bp);
+	u16 mask;
 
+	/* Return here if interrupt is shared and it's not for us */
 	if (unlikely(status == 0)) {
 		DP(NETIF_MSG_INTR, "not our interrupt!\n");
 		return IRQ_NONE;
 	}
-
-	DP(NETIF_MSG_INTR, "got an interrupt status is %u\n", status);
+	DP(NETIF_MSG_INTR, "got an interrupt  status %u\n", status);
 
 #ifdef BNX2X_STOP_ON_ERROR
 	if (unlikely(bp->panic))
 		return IRQ_HANDLED;
 #endif
 
-	/* Return here if interrupt is shared and is disabled */
+	/* Return here if interrupt is disabled */
 	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
 		DP(NETIF_MSG_INTR, "called but intr_sem not 0, returning\n");
 		return IRQ_HANDLED;
 	}
 
-	if (status & 0x2) {
+	mask = 0x2 << bp->fp[0].sb_id;
+	if (status & mask) {
 		struct bnx2x_fastpath *fp = &bp->fp[0];
 
 		prefetch(fp->rx_cons_sb);
@@ -1160,13 +1258,11 @@ static irqreturn_t bnx2x_interrupt(int irq, void *dev_instance)
 
 		netif_rx_schedule(dev, &bnx2x_fp(bp, 0, napi));
 
-		status &= ~0x2;
-		if (!status)
-			return IRQ_HANDLED;
+		status &= ~mask;
 	}
 
-	if (unlikely(status & 0x1)) {
 
+	if (unlikely(status & 0x1)) {
 		schedule_work(&bp->sp_task);
 
 		status &= ~0x1;
@@ -1174,8 +1270,9 @@ static irqreturn_t bnx2x_interrupt(int irq, void *dev_instance)
 			return IRQ_HANDLED;
 	}
 
-	DP(NETIF_MSG_INTR, "got an unknown interrupt! (status is %u)\n",
-	   status);
+	if (status)
+		DP(NETIF_MSG_INTR, "got an unknown interrupt! (status %u)\n",
+		   status);
 
 	return IRQ_HANDLED;
 }
@@ -1193,7 +1290,7 @@ static int bnx2x_hw_lock(struct bnx2x *bp, u32 resource)
 {
 	u32 lock_status;
 	u32 resource_bit = (1 << resource);
-	u8 port = bp->port;
+	u8 port = BP_PORT(bp);
 	int cnt;
 
 	/* Validating that the resource is within range */
@@ -1231,7 +1328,7 @@ static int bnx2x_hw_unlock(struct bnx2x *bp, u32 resource)
 {
 	u32 lock_status;
 	u32 resource_bit = (1 << resource);
-	u8 port = bp->port;
+	u8 port = BP_PORT(bp);
 
 	/* Validating that the resource is within range */
 	if (resource > HW_LOCK_MAX_RESOURCE_VALUE) {
@@ -1258,7 +1355,7 @@ static void bnx2x_phy_hw_lock(struct bnx2x *bp)
 {
 	u32 ext_phy_type = XGXS_EXT_PHY_TYPE(bp->link_params.ext_phy_config);
 
-	mutex_lock(&bp->phy_mutex);
+	mutex_lock(&bp->port.phy_mutex);
 
 	if ((ext_phy_type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8072) ||
 	    (ext_phy_type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073))
@@ -1273,14 +1370,14 @@ static void bnx2x_phy_hw_unlock(struct bnx2x *bp)
 	    (ext_phy_type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073))
 		bnx2x_hw_unlock(bp, HW_LOCK_RESOURCE_8072_MDIO);
 
-	mutex_unlock(&bp->phy_mutex);
+	mutex_unlock(&bp->port.phy_mutex);
 }
 
 int bnx2x_set_gpio(struct bnx2x *bp, int gpio_num, u32 mode)
 {
 	/* The GPIO should be swapped if swap register is set and active */
 	int gpio_port = (REG_RD(bp, NIG_REG_PORT_SWAP) &&
-			 REG_RD(bp, NIG_REG_STRAP_OVERRIDE)) ^ bp->port;
+			 REG_RD(bp, NIG_REG_STRAP_OVERRIDE)) ^ BP_PORT(bp);
 	int gpio_shift = gpio_num +
 			(gpio_port ? MISC_REGISTERS_GPIO_PORT_SHIFT : 0);
 	u32 gpio_mask = (1 << gpio_shift);
@@ -1379,18 +1476,18 @@ static void bnx2x_calc_fc_adv(struct bnx2x *bp)
 {
 	switch (bp->link_vars.ieee_fc) {
 	case MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_NONE:
-		bp->advertising &= ~(ADVERTISED_Asym_Pause |
+		bp->port.advertising &= ~(ADVERTISED_Asym_Pause |
 					  ADVERTISED_Pause);
 		break;
 	case MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_BOTH:
-		bp->advertising |= (ADVERTISED_Asym_Pause |
+		bp->port.advertising |= (ADVERTISED_Asym_Pause |
 					 ADVERTISED_Pause);
 		break;
 	case MDIO_COMBO_IEEE0_AUTO_NEG_ADV_PAUSE_ASYMMETRIC:
-		bp->advertising |= ADVERTISED_Asym_Pause;
+		bp->port.advertising |= ADVERTISED_Asym_Pause;
 		break;
 	default:
-		bp->advertising &= ~(ADVERTISED_Asym_Pause |
+		bp->port.advertising &= ~(ADVERTISED_Asym_Pause |
 					  ADVERTISED_Pause);
 		break;
 	}
@@ -1443,6 +1540,7 @@ static u8 bnx2x_initial_phy_init(struct bnx2x *bp)
 		bnx2x_link_report(bp);
 
 	bnx2x_calc_fc_adv(bp);
+
 	return rc;
 }
 
@@ -1473,15 +1571,261 @@ static u8 bnx2x_link_test(struct bnx2x *bp)
 	return rc;
 }
 
+/* Calculates the sum of vn_min_rates.
+   It's needed for further normalizing of the min_rates.
+
+   Returns:
+     sum of vn_min_rates
+       or
+     0 - if all the min_rates are 0.
+     In the later case fainess algorithm should be deactivated.
+     If not all min_rates are zero then those that are zeroes will
+     be set to 1.
+ */
+static u32 bnx2x_calc_vn_wsum(struct bnx2x *bp)
+{
+	int i, port = BP_PORT(bp);
+	u32 wsum = 0;
+	int all_zero = 1;
+
+	for (i = 0; i < E1HVN_MAX; i++) {
+		u32 vn_cfg =
+			SHMEM_RD(bp, mf_cfg.func_mf_config[2*i + port].config);
+		u32 vn_min_rate = ((vn_cfg & FUNC_MF_CFG_MIN_BW_MASK) >>
+				     FUNC_MF_CFG_MIN_BW_SHIFT) * 100;
+		if (!(vn_cfg & FUNC_MF_CFG_FUNC_HIDE)) {
+			/* If min rate is zero - set it to 1 */
+			if (!vn_min_rate)
+				vn_min_rate = DEF_MIN_RATE;
+			else
+				all_zero = 0;
+
+			wsum += vn_min_rate;
+		}
+	}
+
+	/* ... only if all min rates are zeros - disable FAIRNESS */
+	if (all_zero)
+		return 0;
+
+	return wsum;
+}
+
+static void bnx2x_init_port_minmax(struct bnx2x *bp,
+				   int en_fness,
+				   u16 port_rate,
+				   struct cmng_struct_per_port *m_cmng_port)
+{
+	u32 r_param = port_rate / 8;
+	int port = BP_PORT(bp);
+	int i;
+
+	memset(m_cmng_port, 0, sizeof(struct cmng_struct_per_port));
+
+	/* Enable minmax only if we are in e1hmf mode */
+	if (IS_E1HMF(bp)) {
+		u32 fair_periodic_timeout_usec;
+		u32 t_fair;
+
+		/* Enable rate shaping and fairness */
+		m_cmng_port->flags.cmng_vn_enable = 1;
+		m_cmng_port->flags.fairness_enable = en_fness ? 1 : 0;
+		m_cmng_port->flags.rate_shaping_enable = 1;
+
+		if (!en_fness)
+			DP(NETIF_MSG_IFUP, "All MIN values are zeroes"
+			   "  fairness will be disabled\n");
+
+		/* 100 usec in SDM ticks = 25 since each tick is 4 usec */
+		m_cmng_port->rs_vars.rs_periodic_timeout =
+						RS_PERIODIC_TIMEOUT_USEC / 4;
+
+		/* this is the threshold below which no timer arming will occur
+		   1.25 coefficient is for the threshold to be a little bigger
+		   than the real time, to compensate for timer in-accuracy */
+		m_cmng_port->rs_vars.rs_threshold =
+				(RS_PERIODIC_TIMEOUT_USEC * r_param * 5) / 4;
+
+		/* resolution of fairness timer */
+		fair_periodic_timeout_usec = QM_ARB_BYTES / r_param;
+		/* for 10G it is 1000usec. for 1G it is 10000usec. */
+		t_fair = T_FAIR_COEF / port_rate;
+
+		/* this is the threshold below which we won't arm
+		   the timer anymore */
+		m_cmng_port->fair_vars.fair_threshold = QM_ARB_BYTES;
+
+		/* we multiply by 1e3/8 to get bytes/msec.
+		   We don't want the credits to pass a credit
+		   of the T_FAIR*FAIR_MEM (algorithm resolution) */
+		m_cmng_port->fair_vars.upper_bound =
+						r_param * t_fair * FAIR_MEM;
+		/* since each tick is 4 usec */
+		m_cmng_port->fair_vars.fairness_timeout =
+						fair_periodic_timeout_usec / 4;
+
+	} else {
+		/* Disable rate shaping and fairness */
+		m_cmng_port->flags.cmng_vn_enable = 0;
+		m_cmng_port->flags.fairness_enable = 0;
+		m_cmng_port->flags.rate_shaping_enable = 0;
+
+		DP(NETIF_MSG_IFUP,
+		   "Single function mode  minmax will be disabled\n");
+	}
+
+	/* Store it to internal memory */
+	for (i = 0; i < sizeof(struct cmng_struct_per_port) / 4; i++)
+		REG_WR(bp, BAR_XSTRORM_INTMEM +
+		       XSTORM_CMNG_PER_PORT_VARS_OFFSET(port) + i * 4,
+		       ((u32 *)(m_cmng_port))[i]);
+}
+
+static void bnx2x_init_vn_minmax(struct bnx2x *bp, int func,
+				   u32 wsum, u16 port_rate,
+				 struct cmng_struct_per_port *m_cmng_port)
+{
+	struct rate_shaping_vars_per_vn m_rs_vn;
+	struct fairness_vars_per_vn m_fair_vn;
+	u32 vn_cfg = SHMEM_RD(bp, mf_cfg.func_mf_config[func].config);
+	u16 vn_min_rate, vn_max_rate;
+	int i;
+
+	/* If function is hidden - set min and max to zeroes */
+	if (vn_cfg & FUNC_MF_CFG_FUNC_HIDE) {
+		vn_min_rate = 0;
+		vn_max_rate = 0;
+
+	} else {
+		vn_min_rate = ((vn_cfg & FUNC_MF_CFG_MIN_BW_MASK) >>
+				FUNC_MF_CFG_MIN_BW_SHIFT) * 100;
+		/* If FAIRNESS is enabled (not all min rates are zeroes) and
+		   if current min rate is zero - set it to 1.
+		   This is a requirment of the algorithm. */
+		if ((vn_min_rate == 0) && wsum)
+			vn_min_rate = DEF_MIN_RATE;
+		vn_max_rate = ((vn_cfg & FUNC_MF_CFG_MAX_BW_MASK) >>
+				FUNC_MF_CFG_MAX_BW_SHIFT) * 100;
+	}
+
+	DP(NETIF_MSG_IFUP, "func %d: vn_min_rate=%d  vn_max_rate=%d  "
+	   "wsum=%d\n", func, vn_min_rate, vn_max_rate, wsum);
+
+	memset(&m_rs_vn, 0, sizeof(struct rate_shaping_vars_per_vn));
+	memset(&m_fair_vn, 0, sizeof(struct fairness_vars_per_vn));
+
+	/* global vn counter - maximal Mbps for this vn */
+	m_rs_vn.vn_counter.rate = vn_max_rate;
+
+	/* quota - number of bytes transmitted in this period */
+	m_rs_vn.vn_counter.quota =
+				(vn_max_rate * RS_PERIODIC_TIMEOUT_USEC) / 8;
+
+#ifdef BNX2X_PER_PROT_QOS
+	/* per protocol counter */
+	for (protocol = 0; protocol < NUM_OF_PROTOCOLS; protocol++) {
+		/* maximal Mbps for this protocol */
+		m_rs_vn.protocol_counters[protocol].rate =
+						protocol_max_rate[protocol];
+		/* the quota in each timer period -
+		   number of bytes transmitted in this period */
+		m_rs_vn.protocol_counters[protocol].quota =
+			(u32)(rs_periodic_timeout_usec *
+			  ((double)m_rs_vn.
+				   protocol_counters[protocol].rate/8));
+	}
+#endif
+
+	if (wsum) {
+		/* credit for each period of the fairness algorithm:
+		   number of bytes in T_FAIR (the vn share the port rate).
+		   wsum should not be larger than 10000, thus
+		   T_FAIR_COEF / (8 * wsum) will always be grater than zero */
+		m_fair_vn.vn_credit_delta =
+			max((u64)(vn_min_rate * (T_FAIR_COEF / (8 * wsum))),
+			    (u64)(m_cmng_port->fair_vars.fair_threshold * 2));
+		DP(NETIF_MSG_IFUP, "m_fair_vn.vn_credit_delta=%d\n",
+		   m_fair_vn.vn_credit_delta);
+	}
+
+#ifdef BNX2X_PER_PROT_QOS
+	do {
+		u32 protocolWeightSum = 0;
+
+		for (protocol = 0; protocol < NUM_OF_PROTOCOLS; protocol++)
+			protocolWeightSum +=
+					drvInit.protocol_min_rate[protocol];
+		/* per protocol counter -
+		   NOT NEEDED IF NO PER-PROTOCOL CONGESTION MANAGEMENT */
+		if (protocolWeightSum > 0) {
+			for (protocol = 0;
+			     protocol < NUM_OF_PROTOCOLS; protocol++)
+				/* credit for each period of the
+				   fairness algorithm - number of bytes in
+				   T_FAIR (the protocol share the vn rate) */
+				m_fair_vn.protocol_credit_delta[protocol] =
+					(u32)((vn_min_rate / 8) * t_fair *
+					protocol_min_rate / protocolWeightSum);
+		}
+	} while (0);
+#endif
+
+	/* Store it to internal memory */
+	for (i = 0; i < sizeof(struct rate_shaping_vars_per_vn)/4; i++)
+		REG_WR(bp, BAR_XSTRORM_INTMEM +
+		       XSTORM_RATE_SHAPING_PER_VN_VARS_OFFSET(func) + i * 4,
+		       ((u32 *)(&m_rs_vn))[i]);
+
+	for (i = 0; i < sizeof(struct fairness_vars_per_vn)/4; i++)
+		REG_WR(bp, BAR_XSTRORM_INTMEM +
+		       XSTORM_FAIRNESS_PER_VN_VARS_OFFSET(func) + i * 4,
+		       ((u32 *)(&m_fair_vn))[i]);
+}
+
 /* This function is called upon link interrupt */
 static void bnx2x_link_attn(struct bnx2x *bp)
 {
+	int vn;
+
 	bnx2x_phy_hw_lock(bp);
 	bnx2x_link_update(&bp->link_params, &bp->link_vars);
 	bnx2x_phy_hw_unlock(bp);
 
 	/* indicate link status */
 	bnx2x_link_report(bp);
+
+	if (IS_E1HMF(bp)) {
+		int func;
+
+		for (vn = VN_0; vn < E1HVN_MAX; vn++) {
+			if (vn == BP_E1HVN(bp))
+				continue;
+
+			func = ((vn << 1) | BP_PORT(bp));
+
+			/* Set the attention towards other drivers
+			   on the same port */
+			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_0 +
+			       (LINK_SYNC_ATTENTION_BIT_FUNC_0 + func)*4, 1);
+		}
+	}
+
+	if (CHIP_IS_E1H(bp) && (bp->link_vars.line_speed > 0)) {
+		struct cmng_struct_per_port m_cmng_port;
+		u32 wsum;
+		int port = BP_PORT(bp);
+
+		/* Init RATE SHAPING and FAIRNESS contexts */
+		wsum = bnx2x_calc_vn_wsum(bp);
+		bnx2x_init_port_minmax(bp, (int)wsum,
+					bp->link_vars.line_speed,
+					&m_cmng_port);
+		if (IS_E1HMF(bp))
+			for (vn = VN_0; vn < E1HVN_MAX; vn++)
+				bnx2x_init_vn_minmax(bp, 2*vn + port,
+					wsum, bp->link_vars.line_speed,
+						     &m_cmng_port);
+	}
 }
 
 static void bnx2x__link_status_update(struct bnx2x *bp)
@@ -1495,6 +1839,20 @@ static void bnx2x__link_status_update(struct bnx2x *bp)
 	bnx2x_link_report(bp);
 }
 
+static void bnx2x_pmf_update(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	u32 val;
+
+	bp->port.pmf = 1;
+	DP(NETIF_MSG_LINK, "pmf %d\n", bp->port.pmf);
+
+	/* enable nig attention */
+	val = (0xff0f | (1 << (BP_E1HVN(bp) + 4)));
+	REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, val);
+	REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, val);
+}
+
 /* end of Link */
 
 /* slow path */
@@ -1507,10 +1865,10 @@ static void bnx2x__link_status_update(struct bnx2x *bp)
 static int bnx2x_sp_post(struct bnx2x *bp, int command, int cid,
 			 u32 data_hi, u32 data_lo, int common)
 {
-	int port = bp->port;
+	int func = BP_FUNC(bp);
 
-	DP(NETIF_MSG_TIMER,
-	   "spe (%x:%x)  command %d  hw_cid %x  data (%x:%x)  left %x\n",
+	DP(BNX2X_MSG_SP/*NETIF_MSG_TIMER*/,
+	   "SPQE (%x:%x)  command %d  hw_cid %x  data (%x:%x)  left %x\n",
 	   (u32)U64_HI(bp->spq_mapping), (u32)(U64_LO(bp->spq_mapping) +
 	   (void *)bp->spq_prod_bd - (void *)bp->spq), command,
 	   HW_CID(bp, cid), data_hi, data_lo, bp->spq_left);
@@ -1520,11 +1878,11 @@ static int bnx2x_sp_post(struct bnx2x *bp, int command, int cid,
 		return -EIO;
 #endif
 
-	spin_lock(&bp->spq_lock);
+	spin_lock_bh(&bp->spq_lock);
 
 	if (!bp->spq_left) {
 		BNX2X_ERR("BUG! SPQ ring full!\n");
-		spin_unlock(&bp->spq_lock);
+		spin_unlock_bh(&bp->spq_lock);
 		bnx2x_panic();
 		return -EBUSY;
 	}
@@ -1553,18 +1911,18 @@ static int bnx2x_sp_post(struct bnx2x *bp, int command, int cid,
 		bp->spq_prod_idx++;
 	}
 
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PROD_OFFSET(port),
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PROD_OFFSET(func),
 	       bp->spq_prod_idx);
 
-	spin_unlock(&bp->spq_lock);
+	spin_unlock_bh(&bp->spq_lock);
 	return 0;
 }
 
 /* acquire split MCP access lock register */
 static int bnx2x_lock_alr(struct bnx2x *bp)
 {
-	int rc = 0;
 	u32 i, j, val;
+	int rc = 0;
 
 	might_sleep();
 	i = 100;
@@ -1577,10 +1935,8 @@ static int bnx2x_lock_alr(struct bnx2x *bp)
 
 		msleep(5);
 	}
-
 	if (!(val & (1L << 31))) {
 		BNX2X_ERR("Cannot acquire nvram interface\n");
-
 		rc = -EBUSY;
 	}
 
@@ -1631,8 +1987,9 @@ static inline u16 bnx2x_update_dsb_idx(struct bnx2x *bp)
 
 static void bnx2x_attn_int_asserted(struct bnx2x *bp, u32 asserted)
 {
-	int port = bp->port;
-	u32 igu_addr = (IGU_ADDR_ATTN_BITS_SET + IGU_PORT_BASE * port) * 8;
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	u32 igu_addr = (IGU_ADDR_ATTN_BITS_SET + IGU_FUNC_BASE * func) * 8;
 	u32 aeu_addr = port ? MISC_REG_AEU_MASK_ATTN_FUNC_1 :
 			      MISC_REG_AEU_MASK_ATTN_FUNC_0;
 	u32 nig_int_mask_addr = port ? NIG_REG_MASK_INTERRUPT_PORT1 :
@@ -1716,14 +2073,14 @@ static void bnx2x_attn_int_asserted(struct bnx2x *bp, u32 asserted)
 
 static inline void bnx2x_attn_int_deasserted0(struct bnx2x *bp, u32 attn)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int reg_offset;
 	u32 val;
 
-	if (attn & AEU_INPUTS_ATTN_BITS_SPIO5) {
+	reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0 :
+			     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
 
-		reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0 :
-				     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
+	if (attn & AEU_INPUTS_ATTN_BITS_SPIO5) {
 
 		val = REG_RD(bp, reg_offset);
 		val &= ~AEU_INPUTS_ATTN_BITS_SPIO5;
@@ -1731,7 +2088,7 @@ static inline void bnx2x_attn_int_deasserted0(struct bnx2x *bp, u32 attn)
 
 		BNX2X_ERR("SPIO5 hw attention\n");
 
-		switch (bp->board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
+		switch (bp->common.board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
 		case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
 			/* Fan failure attention */
 
@@ -1762,6 +2119,17 @@ static inline void bnx2x_attn_int_deasserted0(struct bnx2x *bp, u32 attn)
 			break;
 		}
 	}
+
+	if (attn & HW_INTERRUT_ASSERT_SET_0) {
+
+		val = REG_RD(bp, reg_offset);
+		val &= ~(attn & HW_INTERRUT_ASSERT_SET_0);
+		REG_WR(bp, reg_offset, val);
+
+		BNX2X_ERR("FATAL HW block attention set0 0x%x\n",
+			  (attn & HW_INTERRUT_ASSERT_SET_0));
+		bnx2x_panic();
+	}
 }
 
 static inline void bnx2x_attn_int_deasserted1(struct bnx2x *bp, u32 attn)
@@ -1776,6 +2144,23 @@ static inline void bnx2x_attn_int_deasserted1(struct bnx2x *bp, u32 attn)
 		if (val & 0x2)
 			BNX2X_ERR("FATAL error from DORQ\n");
 	}
+
+	if (attn & HW_INTERRUT_ASSERT_SET_1) {
+
+		int port = BP_PORT(bp);
+		int reg_offset;
+
+		reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_1 :
+				     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_1);
+
+		val = REG_RD(bp, reg_offset);
+		val &= ~(attn & HW_INTERRUT_ASSERT_SET_1);
+		REG_WR(bp, reg_offset, val);
+
+		BNX2X_ERR("FATAL HW block attention set1 0x%x\n",
+			  (attn & HW_INTERRUT_ASSERT_SET_1));
+		bnx2x_panic();
+	}
 }
 
 static inline void bnx2x_attn_int_deasserted2(struct bnx2x *bp, u32 attn)
@@ -1799,13 +2184,41 @@ static inline void bnx2x_attn_int_deasserted2(struct bnx2x *bp, u32 attn)
 		if (val & 0x18000)
 			BNX2X_ERR("FATAL error from PXP\n");
 	}
+
+	if (attn & HW_INTERRUT_ASSERT_SET_2) {
+
+		int port = BP_PORT(bp);
+		int reg_offset;
+
+		reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_2 :
+				     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_2);
+
+		val = REG_RD(bp, reg_offset);
+		val &= ~(attn & HW_INTERRUT_ASSERT_SET_2);
+		REG_WR(bp, reg_offset, val);
+
+		BNX2X_ERR("FATAL HW block attention set2 0x%x\n",
+			  (attn & HW_INTERRUT_ASSERT_SET_2));
+		bnx2x_panic();
+	}
 }
 
 static inline void bnx2x_attn_int_deasserted3(struct bnx2x *bp, u32 attn)
 {
+	u32 val;
+
 	if (attn & EVEREST_GEN_ATTN_IN_USE_MASK) {
 
-		if (attn & BNX2X_MC_ASSERT_BITS) {
+		if (attn & BNX2X_PMF_LINK_ASSERT) {
+			int func = BP_FUNC(bp);
+
+			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_12 + func*4, 0);
+			bnx2x__link_status_update(bp);
+			if (SHMEM_RD(bp, func_mb[func].drv_status) &
+							DRV_STATUS_PMF)
+				bnx2x_pmf_update(bp);
+
+		} else if (attn & BNX2X_MC_ASSERT_BITS) {
 
 			BNX2X_ERR("MC assert!\n");
 			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_10, 0);
@@ -1818,16 +2231,25 @@ static inline void bnx2x_attn_int_deasserted3(struct bnx2x *bp, u32 attn)
 
 			BNX2X_ERR("MCP assert!\n");
 			REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_11, 0);
-			bnx2x_mc_assert(bp);
+			bnx2x_fw_dump(bp);
 
 		} else
 			BNX2X_ERR("Unknown HW assert! (attn 0x%x)\n", attn);
 	}
 
 	if (attn & EVEREST_LATCHED_ATTN_IN_USE_MASK) {
-
+		BNX2X_ERR("LATCHED attention 0x%08x (masked)\n", attn);
+		if (attn & BNX2X_GRC_TIMEOUT) {
+			val = CHIP_IS_E1H(bp) ?
+				REG_RD(bp, MISC_REG_GRC_TIMEOUT_ATTN) : 0;
+			BNX2X_ERR("GRC time-out 0x%08x\n", val);
+		}
+		if (attn & BNX2X_GRC_RSV) {
+			val = CHIP_IS_E1H(bp) ?
+				REG_RD(bp, MISC_REG_GRC_RSV_ATTN) : 0;
+			BNX2X_ERR("GRC reserved 0x%08x\n", val);
+		}
 		REG_WR(bp, MISC_REG_AEU_CLR_LATCH_SIGNAL, 0x7ff);
-		BNX2X_ERR("LATCHED attention 0x%x (masked)\n", attn);
 	}
 }
 
@@ -1835,7 +2257,7 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 {
 	struct attn_route attn;
 	struct attn_route group_mask;
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int index;
 	u32 reg_addr;
 	u32 val;
@@ -1848,14 +2270,16 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 	attn.sig[1] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_2_FUNC_0 + port*4);
 	attn.sig[2] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_3_FUNC_0 + port*4);
 	attn.sig[3] = REG_RD(bp, MISC_REG_AEU_AFTER_INVERT_4_FUNC_0 + port*4);
-	DP(NETIF_MSG_HW, "attn %llx\n", (unsigned long long)attn.sig[0]);
+	DP(NETIF_MSG_HW, "attn: %08x %08x %08x %08x\n",
+	   attn.sig[0], attn.sig[1], attn.sig[2], attn.sig[3]);
 
 	for (index = 0; index < MAX_DYNAMIC_ATTN_GRPS; index++) {
 		if (deasserted & (1 << index)) {
 			group_mask = bp->attn_group[index];
 
-			DP(NETIF_MSG_HW, "group[%d]: %llx\n", index,
-			   (unsigned long long)group_mask.sig[0]);
+			DP(NETIF_MSG_HW, "group[%d]: %08x %08x %08x %08x\n",
+			   index, group_mask.sig[0], group_mask.sig[1],
+			   group_mask.sig[2], group_mask.sig[3]);
 
 			bnx2x_attn_int_deasserted3(bp,
 					attn.sig[3] & group_mask.sig[3]);
@@ -1866,22 +2290,6 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 			bnx2x_attn_int_deasserted0(bp,
 					attn.sig[0] & group_mask.sig[0]);
 
-			if ((attn.sig[0] & group_mask.sig[0] &
-						HW_INTERRUT_ASSERT_SET_0) ||
-			    (attn.sig[1] & group_mask.sig[1] &
-						HW_INTERRUT_ASSERT_SET_1) ||
-			    (attn.sig[2] & group_mask.sig[2] &
-						HW_INTERRUT_ASSERT_SET_2))
-				BNX2X_ERR("FATAL HW block attention"
-					  "  set0 0x%x  set1 0x%x"
-					  "  set2 0x%x\n",
-					  (attn.sig[0] & group_mask.sig[0] &
-					   HW_INTERRUT_ASSERT_SET_0),
-					  (attn.sig[1] & group_mask.sig[1] &
-					   HW_INTERRUT_ASSERT_SET_1),
-					  (attn.sig[2] & group_mask.sig[2] &
-					   HW_INTERRUT_ASSERT_SET_2));
-
 			if ((attn.sig[0] & group_mask.sig[0] &
 						HW_PRTY_ASSERT_SET_0) ||
 			    (attn.sig[1] & group_mask.sig[1] &
@@ -1894,17 +2302,17 @@ static void bnx2x_attn_int_deasserted(struct bnx2x *bp, u32 deasserted)
 
 	bnx2x_unlock_alr(bp);
 
-	reg_addr = (IGU_ADDR_ATTN_BITS_CLR + IGU_PORT_BASE * port) * 8;
+	reg_addr = (IGU_ADDR_ATTN_BITS_CLR + IGU_FUNC_BASE * BP_FUNC(bp)) * 8;
 
 	val = ~deasserted;
-/*      DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
+/*	DP(NETIF_MSG_INTR, "write 0x%08x to IGU addr 0x%x\n",
 	   val, BAR_IGU_INTMEM + reg_addr); */
 	REG_WR(bp, BAR_IGU_INTMEM + reg_addr, val);
 
 	if (bp->aeu_mask & (deasserted & 0xff))
-		BNX2X_ERR("IGU BUG\n");
+		BNX2X_ERR("IGU BUG!\n");
 	if (~bp->attn_state & deasserted)
-		BNX2X_ERR("IGU BUG\n");
+		BNX2X_ERR("IGU BUG!\n");
 
 	reg_addr = port ? MISC_REG_AEU_MASK_ATTN_FUNC_1 :
 			  MISC_REG_AEU_MASK_ATTN_FUNC_0;
@@ -1936,7 +2344,7 @@ static void bnx2x_attn_int(struct bnx2x *bp)
 	   attn_bits, attn_ack, asserted, deasserted);
 
 	if (~(attn_bits ^ attn_ack) & (attn_bits ^ attn_state))
-		BNX2X_ERR("bad attention state\n");
+		BNX2X_ERR("BAD attention state\n");
 
 	/* handle bits that were raised */
 	if (asserted)
@@ -1951,6 +2359,7 @@ static void bnx2x_sp_task(struct work_struct *work)
 	struct bnx2x *bp = container_of(work, struct bnx2x, sp_task);
 	u16 status;
 
+
 	/* Return here if interrupt is disabled */
 	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
 		DP(BNX2X_MSG_SP, "called but intr_sem not 0, returning\n");
@@ -1958,19 +2367,15 @@ static void bnx2x_sp_task(struct work_struct *work)
 	}
 
 	status = bnx2x_update_dsb_idx(bp);
-	if (status == 0)
-		BNX2X_ERR("spurious slowpath interrupt!\n");
+/*	if (status == 0)				     */
+/*		BNX2X_ERR("spurious slowpath interrupt!\n"); */
 
-	DP(NETIF_MSG_INTR, "got a slowpath interrupt (updated %x)\n", status);
+	DP(BNX2X_MSG_SP, "got a slowpath interrupt (updated %x)\n", status);
 
 	/* HW attentions */
 	if (status & 0x1)
 		bnx2x_attn_int(bp);
 
-	/* CStorm events: query_stats, port delete ramrod */
-	if (status & 0x2)
-		bp->stat_pending = 0;
-
 	bnx2x_ack_sb(bp, DEF_SB_ID, ATTENTION_ID, bp->def_att_idx,
 		     IGU_INT_NOP, 1);
 	bnx2x_ack_sb(bp, DEF_SB_ID, USTORM_ID, le16_to_cpu(bp->def_u_idx),
@@ -2109,13 +2514,13 @@ static inline long bnx2x_hilo(u32 *hiref)
 static void bnx2x_init_mac_stats(struct bnx2x *bp)
 {
 	struct dmae_command *dmae;
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int loader_idx = port * 8;
 	u32 opcode;
 	u32 mac_addr;
 
 	bp->executer_idx = 0;
-	if (bp->fw_mb) {
+	if (bp->func_stx) {
 		/* MCP */
 		opcode = (DMAE_CMD_SRC_PCI | DMAE_CMD_DST_GRC |
 			  DMAE_CMD_SRC_RESET | DMAE_CMD_DST_RESET |
@@ -2135,7 +2540,7 @@ static void bnx2x_init_mac_stats(struct bnx2x *bp)
 					   sizeof(u32));
 		dmae->src_addr_hi = U64_HI(bnx2x_sp_mapping(bp, eth_stats) +
 					   sizeof(u32));
-		dmae->dst_addr_lo = bp->fw_mb >> 2;
+		dmae->dst_addr_lo = bp->func_stx >> 2;
 		dmae->dst_addr_hi = 0;
 		dmae->len = (offsetof(struct bnx2x_eth_stats, mac_stx_end) -
 			     sizeof(u32)) >> 2;
@@ -2280,7 +2685,7 @@ static void bnx2x_init_mac_stats(struct bnx2x *bp)
 
 static void bnx2x_init_stats(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 
 	bp->stats_state = STATS_STATE_DISABLE;
 	bp->executer_idx = 0;
@@ -2641,8 +3046,6 @@ static void bnx2x_update_net_stats(struct bnx2x *bp)
 
 static void bnx2x_update_stats(struct bnx2x *bp)
 {
-	int i;
-
 	if (!bnx2x_update_storm_stats(bp)) {
 
 		if (bp->link_vars.mac_type == MAC_TYPE_BMAC) {
@@ -2662,6 +3065,7 @@ static void bnx2x_update_stats(struct bnx2x *bp)
 	if (bp->msglevel & NETIF_MSG_TIMER) {
 		struct bnx2x_eth_stats *estats = bnx2x_sp(bp, eth_stats);
 		struct net_device_stats *nstats = &bp->dev->stats;
+		int i;
 
 		printk(KERN_DEBUG "%s:\n", bp->dev->name);
 		printk(KERN_DEBUG "  tx avail (%4x)  tx hc idx (%x)"
@@ -2707,7 +3111,7 @@ static void bnx2x_update_stats(struct bnx2x *bp)
 	/* loader */
 	if (bp->executer_idx) {
 		struct dmae_command *dmae = &bp->dmae;
-		int port = bp->port;
+		int port = BP_PORT(bp);
 		int loader_idx = port * 8;
 
 		memset(dmae, 0, sizeof(struct dmae_command));
@@ -2766,8 +3170,8 @@ static void bnx2x_timer(unsigned long data)
 		rc = bnx2x_rx_int(fp, 1000);
 	}
 
-	if (!nomcp) {
-		int port = bp->port;
+	if (!BP_NOMCP(bp)) {
+		int func = BP_FUNC(bp);
 		u32 drv_pulse;
 		u32 mcp_pulse;
 
@@ -2775,9 +3179,9 @@ static void bnx2x_timer(unsigned long data)
 		bp->fw_drv_pulse_wr_seq &= DRV_PULSE_SEQ_MASK;
 		/* TBD - add SYSTEM_TIME */
 		drv_pulse = bp->fw_drv_pulse_wr_seq;
-		SHMEM_WR(bp, func_mb[port].drv_pulse_mb, drv_pulse);
+		SHMEM_WR(bp, func_mb[func].drv_pulse_mb, drv_pulse);
 
-		mcp_pulse = (SHMEM_RD(bp, func_mb[port].mcp_pulse_mb) &
+		mcp_pulse = (SHMEM_RD(bp, func_mb[func].mcp_pulse_mb) &
 			     MCP_PULSE_SEQ_MASK);
 		/* The delta between driver pulse and mcp response
 		 * should be 1 (before mcp response) or 0 (after mcp response)
@@ -2807,58 +3211,89 @@ timer_restart:
  * nic init service functions
  */
 
-static void bnx2x_init_sb(struct bnx2x *bp, struct host_status_block *sb,
-			  dma_addr_t mapping, int id)
+static void bnx2x_zero_sb(struct bnx2x *bp, int sb_id)
 {
-	int port = bp->port;
-	u64 section;
+	int port = BP_PORT(bp);
+
+	bnx2x_init_fill(bp, BAR_USTRORM_INTMEM +
+			USTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, sb_id), 0,
+			sizeof(struct ustorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_CSTRORM_INTMEM +
+			CSTORM_SB_HOST_STATUS_BLOCK_OFFSET(port, sb_id), 0,
+			sizeof(struct cstorm_def_status_block)/4);
+}
+
+static void bnx2x_init_sb(struct bnx2x *bp, int sb_id,
+			  struct host_status_block *sb,	dma_addr_t mapping)
+{
+	int port = BP_PORT(bp);
 	int index;
+	u64 section;
 
 	/* USTORM */
 	section = ((u64)mapping) + offsetof(struct host_status_block,
 					    u_status_block);
-	sb->u_status_block.status_block_id = id;
+	sb->u_status_block.status_block_id = sb_id;
 
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       USTORM_SB_HOST_SB_ADDR_OFFSET(port, id), U64_LO(section));
+	       USTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id), U64_LO(section));
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       ((USTORM_SB_HOST_SB_ADDR_OFFSET(port, id)) + 4),
+	       ((USTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id)) + 4),
 	       U64_HI(section));
 
 	for (index = 0; index < HC_USTORM_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_USTRORM_INTMEM +
-			 USTORM_SB_HC_DISABLE_OFFSET(port, id, index), 0x1);
+			 USTORM_SB_HC_DISABLE_OFFSET(port, sb_id, index), 1);
 
 	/* CSTORM */
 	section = ((u64)mapping) + offsetof(struct host_status_block,
 					    c_status_block);
-	sb->c_status_block.status_block_id = id;
+	sb->c_status_block.status_block_id = sb_id;
 
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       CSTORM_SB_HOST_SB_ADDR_OFFSET(port, id), U64_LO(section));
+	       CSTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id), U64_LO(section));
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       ((CSTORM_SB_HOST_SB_ADDR_OFFSET(port, id)) + 4),
+	       ((CSTORM_SB_HOST_SB_ADDR_OFFSET(port, sb_id)) + 4),
 	       U64_HI(section));
 
 	for (index = 0; index < HC_CSTORM_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_CSTRORM_INTMEM +
-			 CSTORM_SB_HC_DISABLE_OFFSET(port, id, index), 0x1);
+			 CSTORM_SB_HC_DISABLE_OFFSET(port, sb_id, index), 1);
+
+	bnx2x_ack_sb(bp, sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+}
+
+static void bnx2x_zero_def_sb(struct bnx2x *bp)
+{
+	int func = BP_FUNC(bp);
 
-	bnx2x_ack_sb(bp, id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_init_fill(bp, BAR_USTRORM_INTMEM +
+			USTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct ustorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_CSTRORM_INTMEM +
+			CSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct cstorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_XSTRORM_INTMEM +
+			XSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct xstorm_def_status_block)/4);
+	bnx2x_init_fill(bp, BAR_TSTRORM_INTMEM +
+			TSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), 0,
+			sizeof(struct tstorm_def_status_block)/4);
 }
 
 static void bnx2x_init_def_sb(struct bnx2x *bp,
 			      struct host_def_status_block *def_sb,
-			      dma_addr_t mapping, int id)
+			      dma_addr_t mapping, int sb_id)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
 	int index, val, reg_offset;
 	u64 section;
 
 	/* ATTN */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    atten_status_block);
-	def_sb->atten_status_block.status_block_id = id;
+	def_sb->atten_status_block.status_block_id = sb_id;
 
 	bp->def_att_idx = 0;
 	bp->attn_state = 0;
@@ -2866,7 +3301,7 @@ static void bnx2x_init_def_sb(struct bnx2x *bp,
 	reg_offset = (port ? MISC_REG_AEU_ENABLE1_FUNC_1_OUT_0 :
 			     MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
 
-	for (index = 0; index < 3; index++) {
+	for (index = 0; index < MAX_DYNAMIC_ATTN_GRPS; index++) {
 		bp->attn_group[index].sig[0] = REG_RD(bp,
 						     reg_offset + 0x10*index);
 		bp->attn_group[index].sig[1] = REG_RD(bp,
@@ -2889,116 +3324,123 @@ static void bnx2x_init_def_sb(struct bnx2x *bp,
 	reg_offset = (port ? HC_REG_ATTN_NUM_P1 : HC_REG_ATTN_NUM_P0);
 
 	val = REG_RD(bp, reg_offset);
-	val |= id;
+	val |= sb_id;
 	REG_WR(bp, reg_offset, val);
 
 	/* USTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    u_def_status_block);
-	def_sb->u_def_status_block.status_block_id = id;
+	def_sb->u_def_status_block.status_block_id = sb_id;
 
 	bp->def_u_idx = 0;
 
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_USTRORM_INTMEM +
-	       ((USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((USTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_USTRORM_INTMEM + USTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_USTRORM_INTMEM +  DEF_USB_FUNC_OFF +
+		USTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_USTRORM_INTMEM + USTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_USTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_USTRORM_INTMEM +
-			 USTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+			 USTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
 	/* CSTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    c_def_status_block);
-	def_sb->c_def_status_block.status_block_id = id;
+	def_sb->c_def_status_block.status_block_id = sb_id;
 
 	bp->def_c_idx = 0;
 
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_CSTRORM_INTMEM +
-	       ((CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((CSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_CSTRORM_INTMEM +  DEF_CSB_FUNC_OFF +
+		CSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_CSTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_CSTRORM_INTMEM +
-			 CSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+			 CSTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
 	/* TSTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    t_def_status_block);
-	def_sb->t_def_status_block.status_block_id = id;
+	def_sb->t_def_status_block.status_block_id = sb_id;
 
 	bp->def_t_idx = 0;
 
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       ((TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((TSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_TSTRORM_INTMEM +  DEF_TSB_FUNC_OFF +
+		TSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_TSTRORM_INTMEM + TSTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_TSTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_TSTRORM_INTMEM +
-			 TSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
+			 TSTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
 	/* XSTORM */
 	section = ((u64)mapping) + offsetof(struct host_def_status_block,
 					    x_def_status_block);
-	def_sb->x_def_status_block.status_block_id = id;
+	def_sb->x_def_status_block.status_block_id = sb_id;
 
 	bp->def_x_idx = 0;
 
 	REG_WR(bp, BAR_XSTRORM_INTMEM +
-	       XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port), U64_LO(section));
+	       XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func), U64_LO(section));
 	REG_WR(bp, BAR_XSTRORM_INTMEM +
-	       ((XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(port)) + 4),
+	       ((XSTORM_DEF_SB_HOST_SB_ADDR_OFFSET(func)) + 4),
 	       U64_HI(section));
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_HC_BTR_OFFSET(port),
+	REG_WR8(bp, BAR_XSTRORM_INTMEM +  DEF_XSB_FUNC_OFF +
+		XSTORM_DEF_SB_HOST_STATUS_BLOCK_OFFSET(func), func);
+	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_HC_BTR_OFFSET(func),
 	       BNX2X_BTR);
 
 	for (index = 0; index < HC_XSTORM_DEF_SB_NUM_INDICES; index++)
 		REG_WR16(bp, BAR_XSTRORM_INTMEM +
-			 XSTORM_DEF_SB_HC_DISABLE_OFFSET(port, index), 0x1);
-
-	bp->stat_pending = 0;
+			 XSTORM_DEF_SB_HC_DISABLE_OFFSET(func, index), 1);
 
-	bnx2x_ack_sb(bp, id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_ack_sb(bp, sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
 }
 
 static void bnx2x_update_coalesce(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int i;
 
 	for_each_queue(bp, i) {
+		int sb_id = bp->fp[i].sb_id;
 
 		/* HC_INDEX_U_ETH_RX_CQ_CONS */
 		REG_WR8(bp, BAR_USTRORM_INTMEM +
-			USTORM_SB_HC_TIMEOUT_OFFSET(port, i,
+			USTORM_SB_HC_TIMEOUT_OFFSET(port, sb_id,
 						   HC_INDEX_U_ETH_RX_CQ_CONS),
-			bp->rx_ticks_int/12);
+			bp->rx_ticks/12);
 		REG_WR16(bp, BAR_USTRORM_INTMEM +
-			 USTORM_SB_HC_DISABLE_OFFSET(port, i,
+			 USTORM_SB_HC_DISABLE_OFFSET(port, sb_id,
 						   HC_INDEX_U_ETH_RX_CQ_CONS),
-			 bp->rx_ticks_int ? 0 : 1);
+			 bp->rx_ticks ? 0 : 1);
 
 		/* HC_INDEX_C_ETH_TX_CQ_CONS */
 		REG_WR8(bp, BAR_CSTRORM_INTMEM +
-			CSTORM_SB_HC_TIMEOUT_OFFSET(port, i,
+			CSTORM_SB_HC_TIMEOUT_OFFSET(port, sb_id,
 						   HC_INDEX_C_ETH_TX_CQ_CONS),
-			bp->tx_ticks_int/12);
+			bp->tx_ticks/12);
 		REG_WR16(bp, BAR_CSTRORM_INTMEM +
-			 CSTORM_SB_HC_DISABLE_OFFSET(port, i,
+			 CSTORM_SB_HC_DISABLE_OFFSET(port, sb_id,
 						   HC_INDEX_C_ETH_TX_CQ_CONS),
-			 bp->tx_ticks_int ? 0 : 1);
+			 bp->tx_ticks ? 0 : 1);
 	}
 }
 
@@ -3006,7 +3448,6 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 {
 	u16 ring_prod;
 	int i, j;
-	int port = bp->port;
 
 	bp->rx_buf_use_size = bp->dev->mtu;
 
@@ -3025,13 +3466,13 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 			rx_bd = &fp->rx_desc_ring[RX_DESC_CNT * i - 2];
 			rx_bd->addr_hi =
 				cpu_to_le32(U64_HI(fp->rx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
+					    BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
 			rx_bd->addr_lo =
 				cpu_to_le32(U64_LO(fp->rx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
-
+					    BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
 		}
 
+		/* CQ ring */
 		for (i = 1; i <= NUM_RCQ_RINGS; i++) {
 			struct eth_rx_cqe_next_page *nextpg;
 
@@ -3039,10 +3480,10 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 				&fp->rx_comp_ring[RCQ_DESC_CNT * i - 1];
 			nextpg->addr_hi =
 				cpu_to_le32(U64_HI(fp->rx_comp_mapping +
-					  BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+					   BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
 			nextpg->addr_lo =
 				cpu_to_le32(U64_LO(fp->rx_comp_mapping +
-					  BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+					   BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
 		}
 
 		/* rx completion queue */
@@ -3064,15 +3505,16 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 		/* Warning! this will generate an interrupt (to the TSTORM) */
 		/* must only be done when chip is initialized */
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_RCQ_PROD_OFFSET(port, j), ring_prod);
+		       TSTORM_RX_PRODS_OFFSET(BP_PORT(bp), FP_CL_ID(fp)),
+			ring_prod);
 		if (j != 0)
 			continue;
 
 		REG_WR(bp, BAR_USTRORM_INTMEM +
-		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port),
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(BP_PORT(bp)),
 		       U64_LO(fp->rx_comp_mapping));
 		REG_WR(bp, BAR_USTRORM_INTMEM +
-		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(port) + 4,
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(BP_PORT(bp)) + 4,
 		       U64_HI(fp->rx_comp_mapping));
 	}
 }
@@ -3090,10 +3532,10 @@ static void bnx2x_init_tx_ring(struct bnx2x *bp)
 
 			tx_bd->addr_hi =
 				cpu_to_le32(U64_HI(fp->tx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
+					    BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
 			tx_bd->addr_lo =
 				cpu_to_le32(U64_LO(fp->tx_desc_mapping +
-					   BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
+					    BCM_PAGE_SIZE*(i % NUM_TX_RINGS)));
 		}
 
 		fp->tx_pkt_prod = 0;
@@ -3107,7 +3549,7 @@ static void bnx2x_init_tx_ring(struct bnx2x *bp)
 
 static void bnx2x_init_sp_ring(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int func = BP_FUNC(bp);
 
 	spin_lock_init(&bp->spq_lock);
 
@@ -3117,12 +3559,13 @@ static void bnx2x_init_sp_ring(struct bnx2x *bp)
 	bp->spq_prod_bd = bp->spq;
 	bp->spq_last_bd = bp->spq_prod_bd + MAX_SP_DESC_CNT;
 
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PAGE_BASE_OFFSET(port),
+	REG_WR(bp, XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PAGE_BASE_OFFSET(func),
 	       U64_LO(bp->spq_mapping));
-	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_SPQ_PAGE_BASE_OFFSET(port) + 4,
+	REG_WR(bp,
+	       XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PAGE_BASE_OFFSET(func) + 4,
 	       U64_HI(bp->spq_mapping));
 
-	REG_WR(bp, XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PROD_OFFSET(port),
+	REG_WR(bp, XSEM_REG_FAST_MEMORY + XSTORM_SPQ_PROD_OFFSET(func),
 	       bp->spq_prod_idx);
 }
 
@@ -3133,6 +3576,7 @@ static void bnx2x_init_context(struct bnx2x *bp)
 	for_each_queue(bp, i) {
 		struct eth_context *context = bnx2x_sp(bp, context[i].eth);
 		struct bnx2x_fastpath *fp = &bp->fp[i];
+		u8 sb_id = FP_SB_ID(fp);
 
 		context->xstorm_st_context.tx_bd_page_base_hi =
 						U64_HI(fp->tx_desc_mapping);
@@ -3142,26 +3586,25 @@ static void bnx2x_init_context(struct bnx2x *bp)
 						U64_HI(fp->tx_prods_mapping);
 		context->xstorm_st_context.db_data_addr_lo =
 						U64_LO(fp->tx_prods_mapping);
-
-		context->ustorm_st_context.rx_bd_page_base_hi =
+		context->xstorm_st_context.statistics_data = (BP_CL_ID(bp) |
+				XSTORM_ETH_ST_CONTEXT_STATISTICS_ENABLE);
+
+		context->ustorm_st_context.common.sb_index_numbers =
+						BNX2X_RX_SB_INDEX_NUM;
+		context->ustorm_st_context.common.clientId = FP_CL_ID(fp);
+		context->ustorm_st_context.common.status_block_id = sb_id;
+		context->ustorm_st_context.common.flags =
+			USTORM_ETH_ST_CONTEXT_CONFIG_ENABLE_MC_ALIGNMENT;
+		context->ustorm_st_context.common.mc_alignment_size = 64;
+		context->ustorm_st_context.common.bd_buff_size =
+						bp->rx_buf_use_size;
+		context->ustorm_st_context.common.bd_page_base_hi =
 						U64_HI(fp->rx_desc_mapping);
-		context->ustorm_st_context.rx_bd_page_base_lo =
+		context->ustorm_st_context.common.bd_page_base_lo =
 						U64_LO(fp->rx_desc_mapping);
-		context->ustorm_st_context.status_block_id = i;
-		context->ustorm_st_context.sb_index_number =
-						HC_INDEX_U_ETH_RX_CQ_CONS;
-		context->ustorm_st_context.rcq_base_address_hi =
-						U64_HI(fp->rx_comp_mapping);
-		context->ustorm_st_context.rcq_base_address_lo =
-						U64_LO(fp->rx_comp_mapping);
-		context->ustorm_st_context.flags =
-				USTORM_ETH_ST_CONTEXT_ENABLE_MC_ALIGNMENT;
-		context->ustorm_st_context.mc_alignment_size = 64;
-		context->ustorm_st_context.num_rss = bp->num_queues;
-
 		context->cstorm_st_context.sb_index_number =
 						HC_INDEX_C_ETH_TX_CQ_CONS;
-		context->cstorm_st_context.status_block_id = i;
+		context->cstorm_st_context.status_block_id = sb_id;
 
 		context->xstorm_ag_context.cdu_reserved =
 			CDU_RSRVD_VALUE_TYPE_A(HW_CID(bp, i),
@@ -3176,14 +3619,16 @@ static void bnx2x_init_context(struct bnx2x *bp)
 
 static void bnx2x_init_ind_table(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int i;
 
 	if (!is_multi(bp))
 		return;
 
+	DP(NETIF_MSG_IFUP, "Initializing indirection table\n");
 	for (i = 0; i < TSTORM_INDIRECTION_TABLE_SIZE; i++)
-		REG_WR8(bp, TSTORM_INDIRECTION_TABLE_OFFSET(port) + i,
+		REG_WR8(bp, BAR_TSTRORM_INTMEM +
+			TSTORM_INDIRECTION_TABLE_OFFSET(port) + i,
 			i % bp->num_queues);
 
 	REG_WR(bp, PRS_REG_A_PRSU_20, 0xf);
@@ -3191,77 +3636,74 @@ static void bnx2x_init_ind_table(struct bnx2x *bp)
 
 static void bnx2x_set_client_config(struct bnx2x *bp)
 {
-#ifdef BCM_VLAN
-	int mode = bp->rx_mode;
-#endif
-	int i, port = bp->port;
 	struct tstorm_eth_client_config tstorm_client = {0};
+	int port = BP_PORT(bp);
+	int i;
 
-	tstorm_client.mtu = bp->dev->mtu;
+	tstorm_client.mtu = bp->dev->mtu + ETH_OVREHEAD;
 	tstorm_client.statistics_counter_id = 0;
 	tstorm_client.config_flags =
 				TSTORM_ETH_CLIENT_CONFIG_STATSITICS_ENABLE;
 #ifdef BCM_VLAN
-	if (mode && bp->vlgrp) {
+	if (bp->rx_mode && bp->vlgrp) {
 		tstorm_client.config_flags |=
 				TSTORM_ETH_CLIENT_CONFIG_VLAN_REMOVAL_ENABLE;
 		DP(NETIF_MSG_IFUP, "vlan removal enabled\n");
 	}
 #endif
-	if (mode != BNX2X_RX_MODE_PROMISC)
-		tstorm_client.drop_flags =
-				TSTORM_ETH_CLIENT_CONFIG_DROP_MAC_ERR;
 
 	for_each_queue(bp, i) {
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_CLIENT_CONFIG_OFFSET(port, i),
+		       TSTORM_CLIENT_CONFIG_OFFSET(port, bp->fp[i].cl_id),
 		       ((u32 *)&tstorm_client)[0]);
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_CLIENT_CONFIG_OFFSET(port, i) + 4,
+		       TSTORM_CLIENT_CONFIG_OFFSET(port, bp->fp[i].cl_id) + 4,
 		       ((u32 *)&tstorm_client)[1]);
 	}
 
-/*	DP(NETIF_MSG_IFUP, "tstorm_client: 0x%08x 0x%08x\n",
-	   ((u32 *)&tstorm_client)[0], ((u32 *)&tstorm_client)[1]); */
+	DP(BNX2X_MSG_OFF, "tstorm_client: 0x%08x 0x%08x\n",
+	   ((u32 *)&tstorm_client)[0], ((u32 *)&tstorm_client)[1]);
 }
 
 static void bnx2x_set_storm_rx_mode(struct bnx2x *bp)
 {
-	int mode = bp->rx_mode;
-	int port = bp->port;
 	struct tstorm_eth_mac_filter_config tstorm_mac_filter = {0};
+	int mode = bp->rx_mode;
+	int mask = (1 << BP_L_ID(bp));
+	int func = BP_FUNC(bp);
 	int i;
 
 	DP(NETIF_MSG_RX_STATUS, "rx mode is %d\n", mode);
 
 	switch (mode) {
 	case BNX2X_RX_MODE_NONE: /* no Rx */
-		tstorm_mac_filter.ucast_drop_all = 1;
-		tstorm_mac_filter.mcast_drop_all = 1;
-		tstorm_mac_filter.bcast_drop_all = 1;
+		tstorm_mac_filter.ucast_drop_all = mask;
+		tstorm_mac_filter.mcast_drop_all = mask;
+		tstorm_mac_filter.bcast_drop_all = mask;
 		break;
 	case BNX2X_RX_MODE_NORMAL:
-		tstorm_mac_filter.bcast_accept_all = 1;
+		tstorm_mac_filter.bcast_accept_all = mask;
 		break;
 	case BNX2X_RX_MODE_ALLMULTI:
-		tstorm_mac_filter.mcast_accept_all = 1;
-		tstorm_mac_filter.bcast_accept_all = 1;
+		tstorm_mac_filter.mcast_accept_all = mask;
+		tstorm_mac_filter.bcast_accept_all = mask;
 		break;
 	case BNX2X_RX_MODE_PROMISC:
-		tstorm_mac_filter.ucast_accept_all = 1;
-		tstorm_mac_filter.mcast_accept_all = 1;
-		tstorm_mac_filter.bcast_accept_all = 1;
+		tstorm_mac_filter.ucast_accept_all = mask;
+		tstorm_mac_filter.mcast_accept_all = mask;
+		tstorm_mac_filter.bcast_accept_all = mask;
 		break;
 	default:
-		BNX2X_ERR("bad rx mode (%d)\n", mode);
+		BNX2X_ERR("BAD rx mode (%d)\n", mode);
+		break;
 	}
 
 	for (i = 0; i < sizeof(struct tstorm_eth_mac_filter_config)/4; i++) {
 		REG_WR(bp, BAR_TSTRORM_INTMEM +
-		       TSTORM_MAC_FILTER_CONFIG_OFFSET(port) + i * 4,
+		       TSTORM_MAC_FILTER_CONFIG_OFFSET(func) + i * 4,
 		       ((u32 *)&tstorm_mac_filter)[i]);
 
-/*      	DP(NETIF_MSG_IFUP, "tstorm_mac_filter[%d]: 0x%08x\n", i,
+/*		DP(NETIF_MSG_IFUP, "tstorm_mac_filter[%d]: 0x%08x\n", i,
 		   ((u32 *)&tstorm_mac_filter)[i]); */
 	}
 
@@ -3271,26 +3713,30 @@ static void bnx2x_set_storm_rx_mode(struct bnx2x *bp)
 
 static void bnx2x_init_internal(struct bnx2x *bp)
 {
-	int port = bp->port;
 	struct tstorm_eth_function_common_config tstorm_config = {0};
 	struct stats_indication_flags stats_flags = {0};
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	int i;
 
 	if (is_multi(bp)) {
 		tstorm_config.config_flags = MULTI_FLAGS;
 		tstorm_config.rss_result_mask = MULTI_MASK;
 	}
 
+	tstorm_config.leading_client_id = BP_L_ID(bp);
+
 	REG_WR(bp, BAR_TSTRORM_INTMEM +
-	       TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(port),
+	       TSTORM_FUNCTION_COMMON_CONFIG_OFFSET(func),
 	       (*(u32 *)&tstorm_config));
 
-/*      DP(NETIF_MSG_IFUP, "tstorm_config: 0x%08x\n",
+/*	DP(NETIF_MSG_IFUP, "tstorm_config: 0x%08x\n",
 	   (*(u32 *)&tstorm_config)); */
 
 	bp->rx_mode = BNX2X_RX_MODE_NONE; /* no rx until link is up */
 	bnx2x_set_storm_rx_mode(bp);
 
-	stats_flags.collect_eth = cpu_to_le32(1);
+	stats_flags.collect_eth = 1;
 
 	REG_WR(bp, BAR_XSTRORM_INTMEM + XSTORM_STATS_FLAGS_OFFSET(port),
 	       ((u32 *)&stats_flags)[0]);
@@ -3307,8 +3753,28 @@ static void bnx2x_init_internal(struct bnx2x *bp)
 	REG_WR(bp, BAR_CSTRORM_INTMEM + CSTORM_STATS_FLAGS_OFFSET(port) + 4,
 	       ((u32 *)&stats_flags)[1]);
 
-/*      DP(NETIF_MSG_IFUP, "stats_flags: 0x%08x 0x%08x\n",
+/*	DP(NETIF_MSG_IFUP, "stats_flags: 0x%08x 0x%08x\n",
 	   ((u32 *)&stats_flags)[0], ((u32 *)&stats_flags)[1]); */
+
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR8(bp, BAR_XSTRORM_INTMEM + XSTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+		REG_WR8(bp, BAR_TSTRORM_INTMEM + TSTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+		REG_WR8(bp, BAR_CSTRORM_INTMEM + CSTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+		REG_WR8(bp, BAR_USTRORM_INTMEM + USTORM_FUNCTION_MODE_OFFSET,
+			IS_E1HMF(bp));
+
+		REG_WR16(bp, BAR_XSTRORM_INTMEM +
+			 XSTORM_E1HOV_OFFSET(func), bp->e1hov);
+	}
+
+	/* Zero this manualy as its initialization is
+	   currently missing in the initTool */
+	for (i = 0; i < USTORM_AGG_DATA_SIZE >> 2; i++)
+		REG_WR(bp, BAR_USTRORM_INTMEM +
+		       USTORM_AGG_DATA_OFFSET + 4*i, 0);
 }
 
 static void bnx2x_nic_init(struct bnx2x *bp)
@@ -3318,15 +3784,20 @@ static void bnx2x_nic_init(struct bnx2x *bp)
 	for_each_queue(bp, i) {
 		struct bnx2x_fastpath *fp = &bp->fp[i];
 
+		fp->bp = bp;
 		fp->state = BNX2X_FP_STATE_CLOSED;
-		DP(NETIF_MSG_IFUP, "bnx2x_init_sb(%p,%p,%d);\n",
-		   bp, fp->status_blk, i);
 		fp->index = i;
-		bnx2x_init_sb(bp, fp->status_blk, fp->status_blk_mapping, i);
+		fp->cl_id = BP_L_ID(bp) + i;
+		fp->sb_id = fp->cl_id;
+		DP(NETIF_MSG_IFUP,
+		   "bnx2x_init_sb(%p,%p) index %d  cl_id %d  sb %d\n",
+		   bp, fp->status_blk, i, FP_CL_ID(fp), FP_SB_ID(fp));
+		bnx2x_init_sb(bp, FP_SB_ID(fp), fp->status_blk,
+			      fp->status_blk_mapping);
 	}
 
 	bnx2x_init_def_sb(bp, bp->def_status_blk,
-			  bp->def_status_blk_mapping, 0x10);
+			  bp->def_status_blk_mapping, DEF_SB_ID);
 	bnx2x_update_coalesce(bp);
 	bnx2x_init_rx_rings(bp);
 	bnx2x_init_tx_ring(bp);
@@ -3336,7 +3807,6 @@ static void bnx2x_nic_init(struct bnx2x *bp)
 	bnx2x_init_stats(bp);
 	bnx2x_init_ind_table(bp);
 	bnx2x_int_enable(bp);
-
 }
 
 /* end of nic init */
@@ -3374,7 +3844,7 @@ gunzip_nomem2:
 
 gunzip_nomem1:
 	printk(KERN_ERR PFX "%s: Cannot allocate firmware buffer for"
-	       " uncompression\n", bp->dev->name);
+	       " un-compression\n", bp->dev->name);
 	return -ENOMEM;
 }
 
@@ -3402,7 +3872,7 @@ static int bnx2x_gunzip(struct bnx2x *bp, u8 *zbuf, int len)
 
 	n = 10;
 
-#define FNAME   			0x8
+#define FNAME				0x8
 
 	if (zbuf[3] & FNAME)
 		while ((zbuf[n++] != 0) && (n < len));
@@ -3439,41 +3909,25 @@ static int bnx2x_gunzip(struct bnx2x *bp, u8 *zbuf, int len)
 /* nic load/unload */
 
 /*
- * general service functions
+ * General service functions
  */
 
 /* send a NIG loopback debug packet */
 static void bnx2x_lb_pckt(struct bnx2x *bp)
 {
-#ifdef USE_DMAE
 	u32 wb_write[3];
-#endif
 
 	/* Ethernet source and destination addresses */
-#ifdef USE_DMAE
 	wb_write[0] = 0x55555555;
 	wb_write[1] = 0x55555555;
-	wb_write[2] = 0x20;     	/* SOP */
+	wb_write[2] = 0x20;		/* SOP */
 	REG_WR_DMAE(bp, NIG_REG_DEBUG_PACKET_LB, wb_write, 3);
-#else
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB, 0x55555555);
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 4, 0x55555555);
-	/* SOP */
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 8, 0x20);
-#endif
 
 	/* NON-IP protocol */
-#ifdef USE_DMAE
 	wb_write[0] = 0x09000000;
 	wb_write[1] = 0x55555555;
-	wb_write[2] = 0x10;     	/* EOP, eop_bvalid = 0 */
+	wb_write[2] = 0x10;		/* EOP, eop_bvalid = 0 */
 	REG_WR_DMAE(bp, NIG_REG_DEBUG_PACKET_LB, wb_write, 3);
-#else
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB, 0x09000000);
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 4, 0x55555555);
-	/* EOP, eop_bvalid = 0 */
-	REG_WR_IND(bp, NIG_REG_DEBUG_PACKET_LB + 8, 0x10);
-#endif
 }
 
 /* some of the internal memories
@@ -3511,13 +3965,9 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	/* Wait until NIG register shows 1 packet of size 0x10 */
 	count = 1000 * factor;
 	while (count) {
-#ifdef BNX2X_DMAE_RD
+
 		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
 		val = *bnx2x_sp(bp, wb_data[0]);
-#else
-		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
-		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
-#endif
 		if (val == 0x10)
 			break;
 
@@ -3533,7 +3983,6 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	count = 1000 * factor;
 	while (count) {
 		val = REG_RD(bp, PRS_REG_NUM_OF_PACKETS);
-
 		if (val == 1)
 			break;
 
@@ -3546,9 +3995,9 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	}
 
 	/* Reset and init BRB, PRS */
-	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR, 0x3);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR, 0x03);
 	msleep(50);
-	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0x3);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0x03);
 	msleep(50);
 	bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
 	bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
@@ -3572,13 +4021,9 @@ static int bnx2x_int_mem_test(struct bnx2x *bp)
 	   packets of size 11*0x10 = 0xb0 */
 	count = 1000 * factor;
 	while (count) {
-#ifdef BNX2X_DMAE_RD
+
 		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
 		val = *bnx2x_sp(bp, wb_data[0]);
-#else
-		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
-		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
-#endif
 		if (val == 0xb0)
 			break;
 
@@ -3648,85 +4093,75 @@ static void enable_blocks_attention(struct bnx2x *bp)
 	REG_WR(bp, XSDM_REG_XSDM_INT_MASK_0, 0);
 	REG_WR(bp, XSDM_REG_XSDM_INT_MASK_1, 0);
 	REG_WR(bp, XCM_REG_XCM_INT_MASK, 0);
-/*      REG_WR(bp, XSEM_REG_XSEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, XSEM_REG_XSEM_INT_MASK_1, 0); */
+/*	REG_WR(bp, XSEM_REG_XSEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, XSEM_REG_XSEM_INT_MASK_1, 0); */
 	REG_WR(bp, USDM_REG_USDM_INT_MASK_0, 0);
 	REG_WR(bp, USDM_REG_USDM_INT_MASK_1, 0);
 	REG_WR(bp, UCM_REG_UCM_INT_MASK, 0);
-/*      REG_WR(bp, USEM_REG_USEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, USEM_REG_USEM_INT_MASK_1, 0); */
+/*	REG_WR(bp, USEM_REG_USEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, USEM_REG_USEM_INT_MASK_1, 0); */
 	REG_WR(bp, GRCBASE_UPB + PB_REG_PB_INT_MASK, 0);
 	REG_WR(bp, CSDM_REG_CSDM_INT_MASK_0, 0);
 	REG_WR(bp, CSDM_REG_CSDM_INT_MASK_1, 0);
 	REG_WR(bp, CCM_REG_CCM_INT_MASK, 0);
-/*      REG_WR(bp, CSEM_REG_CSEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, CSEM_REG_CSEM_INT_MASK_1, 0); */
-	REG_WR(bp, PXP2_REG_PXP2_INT_MASK, 0x480000);
+/*	REG_WR(bp, CSEM_REG_CSEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, CSEM_REG_CSEM_INT_MASK_1, 0); */
+	if (CHIP_REV_IS_FPGA(bp))
+		REG_WR(bp, PXP2_REG_PXP2_INT_MASK_0, 0x580000);
+	else
+		REG_WR(bp, PXP2_REG_PXP2_INT_MASK_0, 0x480000);
 	REG_WR(bp, TSDM_REG_TSDM_INT_MASK_0, 0);
 	REG_WR(bp, TSDM_REG_TSDM_INT_MASK_1, 0);
 	REG_WR(bp, TCM_REG_TCM_INT_MASK, 0);
-/*      REG_WR(bp, TSEM_REG_TSEM_INT_MASK_0, 0); */
-/*      REG_WR(bp, TSEM_REG_TSEM_INT_MASK_1, 0); */
+/*	REG_WR(bp, TSEM_REG_TSEM_INT_MASK_0, 0); */
+/*	REG_WR(bp, TSEM_REG_TSEM_INT_MASK_1, 0); */
 	REG_WR(bp, CDU_REG_CDU_INT_MASK, 0);
 	REG_WR(bp, DMAE_REG_DMAE_INT_MASK, 0);
-/*      REG_WR(bp, MISC_REG_MISC_INT_MASK, 0); */
-	REG_WR(bp, PBF_REG_PBF_INT_MASK, 0X18); 	/* bit 3,4 masked */
+/*	REG_WR(bp, MISC_REG_MISC_INT_MASK, 0); */
+	REG_WR(bp, PBF_REG_PBF_INT_MASK, 0X18);		/* bit 3,4 masked */
 }
 
-static int bnx2x_function_init(struct bnx2x *bp, int mode)
+
+static int bnx2x_init_common(struct bnx2x *bp)
 {
-	int func = bp->port;
-	int port = func ? PORT1 : PORT0;
 	u32 val, i;
-#ifdef USE_DMAE
-	u32 wb_write[2];
-#endif
-
-	DP(BNX2X_MSG_MCP, "function is %d  mode is %x\n", func, mode);
-	if ((func != 0) && (func != 1)) {
-		BNX2X_ERR("BAD function number (%d)\n", func);
-		return -ENODEV;
-	}
 
-	bnx2x_gunzip_init(bp);
+	DP(BNX2X_MSG_MCP, "starting common init  func %d\n", BP_FUNC(bp));
 
-	if (mode & 0x1) {       /* init common */
-		DP(BNX2X_MSG_MCP, "starting common init  func %d  mode %x\n",
-		   func, mode);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET,
-		       0xffffffff);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET,
-		       0xfffc);
-		bnx2x_init_block(bp, MISC_COMMON_START, MISC_COMMON_END);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET, 0xffffffff);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_SET, 0xfffc);
 
-		REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x100);
-		msleep(30);
-		REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x0);
+	bnx2x_init_block(bp, MISC_COMMON_START, MISC_COMMON_END);
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, MISC_REG_E1HMF_MODE, IS_E1HMF(bp));
 
-		bnx2x_init_block(bp, PXP_COMMON_START, PXP_COMMON_END);
-		bnx2x_init_block(bp, PXP2_COMMON_START, PXP2_COMMON_END);
+	REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x100);
+	msleep(30);
+	REG_WR(bp, MISC_REG_LCPLL_CTRL_REG_2, 0x0);
 
-		bnx2x_init_pxp(bp);
+	bnx2x_init_block(bp, PXP_COMMON_START, PXP_COMMON_END);
+	if (CHIP_IS_E1(bp)) {
+		/* enable HW interrupt from PXP on USDM overflow
+		   bit 16 on INT_MASK_0 */
+		REG_WR(bp, PXP_REG_PXP_INT_MASK_0, 0);
+	}
 
-		if (CHIP_REV(bp) == CHIP_REV_Ax) {
-			/* enable HW interrupt from PXP on USDM
-			   overflow bit 16 on INT_MASK_0 */
-			REG_WR(bp, PXP_REG_PXP_INT_MASK_0, 0);
-		}
+	bnx2x_init_block(bp, PXP2_COMMON_START, PXP2_COMMON_END);
+	bnx2x_init_pxp(bp);
 
 #ifdef __BIG_ENDIAN
-		REG_WR(bp, PXP2_REG_RQ_QM_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_TM_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_SRC_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_CDU_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_DBG_ENDIAN_M, 1);
-		REG_WR(bp, PXP2_REG_RQ_HC_ENDIAN_M, 1);
-
-/*      	REG_WR(bp, PXP2_REG_RD_PBF_SWAP_MODE, 1); */
-		REG_WR(bp, PXP2_REG_RD_QM_SWAP_MODE, 1);
-		REG_WR(bp, PXP2_REG_RD_TM_SWAP_MODE, 1);
-		REG_WR(bp, PXP2_REG_RD_SRC_SWAP_MODE, 1);
-		REG_WR(bp, PXP2_REG_RD_CDURD_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RQ_QM_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_TM_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_SRC_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_CDU_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_DBG_ENDIAN_M, 1);
+	REG_WR(bp, PXP2_REG_RQ_HC_ENDIAN_M, 1);
+
+/*	REG_WR(bp, PXP2_REG_RD_PBF_SWAP_MODE, 1); */
+	REG_WR(bp, PXP2_REG_RD_QM_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RD_TM_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RD_SRC_SWAP_MODE, 1);
+	REG_WR(bp, PXP2_REG_RD_CDURD_SWAP_MODE, 1);
 #endif
 
 #ifndef BCM_ISCSI
@@ -3734,92 +4169,105 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 		REG_WR(bp, PRS_REG_NIC_MODE, 1);
 #endif
 
-		REG_WR(bp, PXP2_REG_RQ_CDU_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_CDU_P_SIZE, 2);
 #ifdef BCM_ISCSI
-		REG_WR(bp, PXP2_REG_RQ_TM_P_SIZE, 5);
-		REG_WR(bp, PXP2_REG_RQ_QM_P_SIZE, 5);
-		REG_WR(bp, PXP2_REG_RQ_SRC_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_TM_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_QM_P_SIZE, 5);
+	REG_WR(bp, PXP2_REG_RQ_SRC_P_SIZE, 5);
 #endif
 
-		bnx2x_init_block(bp, DMAE_COMMON_START, DMAE_COMMON_END);
+	if (CHIP_REV_IS_FPGA(bp) && CHIP_IS_E1H(bp))
+		REG_WR(bp, PXP2_REG_PGL_TAGS_LIMIT, 0x1);
 
-		/* let the HW do it's magic ... */
-		msleep(100);
-		/* finish PXP init
-		   (can be moved up if we want to use the DMAE) */
-		val = REG_RD(bp, PXP2_REG_RQ_CFG_DONE);
-		if (val != 1) {
-			BNX2X_ERR("PXP2 CFG failed\n");
-			return -EBUSY;
-		}
+	/* let the HW do it's magic ... */
+	msleep(100);
+	/* finish PXP init */
+	val = REG_RD(bp, PXP2_REG_RQ_CFG_DONE);
+	if (val != 1) {
+		BNX2X_ERR("PXP2 CFG failed\n");
+		return -EBUSY;
+	}
+	val = REG_RD(bp, PXP2_REG_RD_INIT_DONE);
+	if (val != 1) {
+		BNX2X_ERR("PXP2 RD_INIT failed\n");
+		return -EBUSY;
+	}
 
-		val = REG_RD(bp, PXP2_REG_RD_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("PXP2 RD_INIT failed\n");
-			return -EBUSY;
-		}
+	REG_WR(bp, PXP2_REG_RQ_DISABLE_INPUTS, 0);
+	REG_WR(bp, PXP2_REG_RD_DISABLE_INPUTS, 0);
 
-		REG_WR(bp, PXP2_REG_RQ_DISABLE_INPUTS, 0);
-		REG_WR(bp, PXP2_REG_RD_DISABLE_INPUTS, 0);
+	bnx2x_init_block(bp, DMAE_COMMON_START, DMAE_COMMON_END);
 
-		bnx2x_init_fill(bp, TSEM_REG_PRAM, 0, 8);
+	/* clean the DMAE memory */
+	bp->dmae_ready = 1;
+	bnx2x_init_fill(bp, TSEM_REG_PRAM, 0, 8);
 
-		bnx2x_init_block(bp, TCM_COMMON_START, TCM_COMMON_END);
-		bnx2x_init_block(bp, UCM_COMMON_START, UCM_COMMON_END);
-		bnx2x_init_block(bp, CCM_COMMON_START, CCM_COMMON_END);
-		bnx2x_init_block(bp, XCM_COMMON_START, XCM_COMMON_END);
+	bnx2x_init_block(bp, TCM_COMMON_START, TCM_COMMON_END);
+	bnx2x_init_block(bp, UCM_COMMON_START, UCM_COMMON_END);
+	bnx2x_init_block(bp, CCM_COMMON_START, CCM_COMMON_END);
+	bnx2x_init_block(bp, XCM_COMMON_START, XCM_COMMON_END);
 
-#ifdef BNX2X_DMAE_RD
-		bnx2x_read_dmae(bp, XSEM_REG_PASSIVE_BUFFER, 3);
-		bnx2x_read_dmae(bp, CSEM_REG_PASSIVE_BUFFER, 3);
-		bnx2x_read_dmae(bp, TSEM_REG_PASSIVE_BUFFER, 3);
-		bnx2x_read_dmae(bp, USEM_REG_PASSIVE_BUFFER, 3);
-#else
-		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, XSEM_REG_PASSIVE_BUFFER + 8);
-		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, CSEM_REG_PASSIVE_BUFFER + 8);
-		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, TSEM_REG_PASSIVE_BUFFER + 8);
-		REG_RD(bp, USEM_REG_PASSIVE_BUFFER);
-		REG_RD(bp, USEM_REG_PASSIVE_BUFFER + 4);
-		REG_RD(bp, USEM_REG_PASSIVE_BUFFER + 8);
-#endif
-		bnx2x_init_block(bp, QM_COMMON_START, QM_COMMON_END);
-		/* soft reset pulse */
-		REG_WR(bp, QM_REG_SOFT_RESET, 1);
-		REG_WR(bp, QM_REG_SOFT_RESET, 0);
+	bnx2x_read_dmae(bp, XSEM_REG_PASSIVE_BUFFER, 3);
+	bnx2x_read_dmae(bp, CSEM_REG_PASSIVE_BUFFER, 3);
+	bnx2x_read_dmae(bp, TSEM_REG_PASSIVE_BUFFER, 3);
+	bnx2x_read_dmae(bp, USEM_REG_PASSIVE_BUFFER, 3);
+
+	bnx2x_init_block(bp, QM_COMMON_START, QM_COMMON_END);
+	/* soft reset pulse */
+	REG_WR(bp, QM_REG_SOFT_RESET, 1);
+	REG_WR(bp, QM_REG_SOFT_RESET, 0);
 
 #ifdef BCM_ISCSI
-		bnx2x_init_block(bp, TIMERS_COMMON_START, TIMERS_COMMON_END);
+	bnx2x_init_block(bp, TIMERS_COMMON_START, TIMERS_COMMON_END);
 #endif
-		bnx2x_init_block(bp, DQ_COMMON_START, DQ_COMMON_END);
-		REG_WR(bp, DORQ_REG_DPM_CID_OFST, BCM_PAGE_BITS);
-		if (CHIP_REV(bp) == CHIP_REV_Ax) {
-			/* enable hw interrupt from doorbell Q */
-			REG_WR(bp, DORQ_REG_DORQ_INT_MASK, 0);
-		}
 
-		bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
+	bnx2x_init_block(bp, DQ_COMMON_START, DQ_COMMON_END);
+	REG_WR(bp, DORQ_REG_DPM_CID_OFST, BCM_PAGE_SHIFT);
+	if (!CHIP_REV_IS_SLOW(bp)) {
+		/* enable hw interrupt from doorbell Q */
+		REG_WR(bp, DORQ_REG_DORQ_INT_MASK, 0);
+	}
 
-		if (CHIP_REV_IS_SLOW(bp)) {
-			/* fix for emulation and FPGA for no pause */
-			REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_0, 513);
-			REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_1, 513);
-			REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_0, 0);
-			REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_1, 0);
-		}
+	bnx2x_init_block(bp, BRB1_COMMON_START, BRB1_COMMON_END);
+	if (CHIP_REV_IS_SLOW(bp)) {
+		/* fix for emulation and FPGA for no pause */
+		REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_0, 513);
+		REG_WR(bp, BRB1_REG_PAUSE_HIGH_THRESHOLD_1, 513);
+		REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_0, 0);
+		REG_WR(bp, BRB1_REG_PAUSE_LOW_THRESHOLD_1, 0);
+	}
 
-		bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
+	bnx2x_init_block(bp, PRS_COMMON_START, PRS_COMMON_END);
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, PRS_REG_E1HOV_MODE, IS_E1HMF(bp));
 
-		bnx2x_init_block(bp, TSDM_COMMON_START, TSDM_COMMON_END);
-		bnx2x_init_block(bp, CSDM_COMMON_START, CSDM_COMMON_END);
-		bnx2x_init_block(bp, USDM_COMMON_START, USDM_COMMON_END);
-		bnx2x_init_block(bp, XSDM_COMMON_START, XSDM_COMMON_END);
+	bnx2x_init_block(bp, TSDM_COMMON_START, TSDM_COMMON_END);
+	bnx2x_init_block(bp, CSDM_COMMON_START, CSDM_COMMON_END);
+	bnx2x_init_block(bp, USDM_COMMON_START, USDM_COMMON_END);
+	bnx2x_init_block(bp, XSDM_COMMON_START, XSDM_COMMON_END);
 
+	if (CHIP_IS_E1H(bp)) {
+		bnx2x_init_fill(bp, TSTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				TSTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp, CSTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				CSTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp, XSTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				XSTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp, USTORM_INTMEM_ADDR, 0,
+				STORM_INTMEM_SIZE_E1H/2);
+		bnx2x_init_fill(bp,
+				USTORM_INTMEM_ADDR + STORM_INTMEM_SIZE_E1H/2,
+				0, STORM_INTMEM_SIZE_E1H/2);
+	} else { /* E1 */
 		bnx2x_init_fill(bp, TSTORM_INTMEM_ADDR, 0,
 				STORM_INTMEM_SIZE_E1);
 		bnx2x_init_fill(bp, CSTORM_INTMEM_ADDR, 0,
@@ -3828,157 +4276,141 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 				STORM_INTMEM_SIZE_E1);
 		bnx2x_init_fill(bp, USTORM_INTMEM_ADDR, 0,
 				STORM_INTMEM_SIZE_E1);
+	}
 
-		bnx2x_init_block(bp, TSEM_COMMON_START, TSEM_COMMON_END);
-		bnx2x_init_block(bp, USEM_COMMON_START, USEM_COMMON_END);
-		bnx2x_init_block(bp, CSEM_COMMON_START, CSEM_COMMON_END);
-		bnx2x_init_block(bp, XSEM_COMMON_START, XSEM_COMMON_END);
-
-		/* sync semi rtc */
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
-		       0x80000000);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET,
-		       0x80000000);
-
-		bnx2x_init_block(bp, UPB_COMMON_START, UPB_COMMON_END);
-		bnx2x_init_block(bp, XPB_COMMON_START, XPB_COMMON_END);
-		bnx2x_init_block(bp, PBF_COMMON_START, PBF_COMMON_END);
-
-		REG_WR(bp, SRC_REG_SOFT_RST, 1);
-		for (i = SRC_REG_KEYRSS0_0; i <= SRC_REG_KEYRSS1_9; i += 4) {
-			REG_WR(bp, i, 0xc0cac01a);
-			/* TODO: replace with something meaningful */
-		}
-		/* SRCH COMMON comes here */
-		REG_WR(bp, SRC_REG_SOFT_RST, 0);
-
-		if (sizeof(union cdu_context) != 1024) {
-			/* we currently assume that a context is 1024 bytes */
-			printk(KERN_ALERT PFX "please adjust the size of"
-			       " cdu_context(%ld)\n",
-			       (long)sizeof(union cdu_context));
-		}
-		val = (4 << 24) + (0 << 12) + 1024;
-		REG_WR(bp, CDU_REG_CDU_GLOBAL_PARAMS, val);
-		bnx2x_init_block(bp, CDU_COMMON_START, CDU_COMMON_END);
-
-		bnx2x_init_block(bp, CFC_COMMON_START, CFC_COMMON_END);
-		REG_WR(bp, CFC_REG_INIT_REG, 0x7FF);
-
-		bnx2x_init_block(bp, HC_COMMON_START, HC_COMMON_END);
-		bnx2x_init_block(bp, MISC_AEU_COMMON_START,
-				 MISC_AEU_COMMON_END);
-		/* RXPCS COMMON comes here */
-		/* EMAC0 COMMON comes here */
-		/* EMAC1 COMMON comes here */
-		/* DBU COMMON comes here */
-		/* DBG COMMON comes here */
-		bnx2x_init_block(bp, NIG_COMMON_START, NIG_COMMON_END);
+	bnx2x_init_block(bp, TSEM_COMMON_START, TSEM_COMMON_END);
+	bnx2x_init_block(bp, USEM_COMMON_START, USEM_COMMON_END);
+	bnx2x_init_block(bp, CSEM_COMMON_START, CSEM_COMMON_END);
+	bnx2x_init_block(bp, XSEM_COMMON_START, XSEM_COMMON_END);
 
-		if (CHIP_REV_IS_SLOW(bp))
-			msleep(200);
+	/* sync semi rtc */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+	       0x80000000);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_SET,
+	       0x80000000);
 
-		/* finish CFC init */
-		val = REG_RD(bp, CFC_REG_LL_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("CFC LL_INIT failed\n");
-			return -EBUSY;
-		}
+	bnx2x_init_block(bp, UPB_COMMON_START, UPB_COMMON_END);
+	bnx2x_init_block(bp, XPB_COMMON_START, XPB_COMMON_END);
+	bnx2x_init_block(bp, PBF_COMMON_START, PBF_COMMON_END);
 
-		val = REG_RD(bp, CFC_REG_AC_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("CFC AC_INIT failed\n");
-			return -EBUSY;
-		}
+	REG_WR(bp, SRC_REG_SOFT_RST, 1);
+	for (i = SRC_REG_KEYRSS0_0; i <= SRC_REG_KEYRSS1_9; i += 4) {
+		REG_WR(bp, i, 0xc0cac01a);
+		/* TODO: replace with something meaningful */
+	}
+	if (CHIP_IS_E1H(bp))
+		bnx2x_init_block(bp, SRCH_COMMON_START, SRCH_COMMON_END);
+	REG_WR(bp, SRC_REG_SOFT_RST, 0);
 
-		val = REG_RD(bp, CFC_REG_CAM_INIT_DONE);
-		if (val != 1) {
-			BNX2X_ERR("CFC CAM_INIT failed\n");
-			return -EBUSY;
-		}
+	if (sizeof(union cdu_context) != 1024)
+		/* we currently assume that a context is 1024 bytes */
+		printk(KERN_ALERT PFX "please adjust the size of"
+		       " cdu_context(%ld)\n", (long)sizeof(union cdu_context));
 
-		REG_WR(bp, CFC_REG_DEBUG0, 0);
+	bnx2x_init_block(bp, CDU_COMMON_START, CDU_COMMON_END);
+	val = (4 << 24) + (0 << 12) + 1024;
+	REG_WR(bp, CDU_REG_CDU_GLOBAL_PARAMS, val);
+	if (CHIP_IS_E1(bp)) {
+		/* !!! fix pxp client crdit until excel update */
+		REG_WR(bp, CDU_REG_CDU_DEBUG, 0x264);
+		REG_WR(bp, CDU_REG_CDU_DEBUG, 0);
+	}
 
-		/* read NIG statistic
-		   to see if this is our first up since powerup */
-#ifdef BNX2X_DMAE_RD
-		bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
-		val = *bnx2x_sp(bp, wb_data[0]);
-#else
-		val = REG_RD(bp, NIG_REG_STAT2_BRB_OCTET);
-		REG_RD(bp, NIG_REG_STAT2_BRB_OCTET + 4);
-#endif
-		/* do internal memory self test */
-		if ((val == 0) && bnx2x_int_mem_test(bp)) {
-			BNX2X_ERR("internal mem selftest failed\n");
-			return -EBUSY;
-		}
+	bnx2x_init_block(bp, CFC_COMMON_START, CFC_COMMON_END);
+	REG_WR(bp, CFC_REG_INIT_REG, 0x7FF);
 
-		/* clear PXP2 attentions */
-		REG_RD(bp, PXP2_REG_PXP2_INT_STS_CLR);
+	bnx2x_init_block(bp, HC_COMMON_START, HC_COMMON_END);
+	bnx2x_init_block(bp, MISC_AEU_COMMON_START, MISC_AEU_COMMON_END);
 
-		enable_blocks_attention(bp);
-		/* enable_blocks_parity(bp); */
+	/* PXPCS COMMON comes here */
+	/* Reset PCIE errors for debug */
+	REG_WR(bp, 0x2814, 0xffffffff);
+	REG_WR(bp, 0x3820, 0xffffffff);
 
-		switch (bp->board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
-		case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
-			/* Fan failure is indicated by SPIO 5 */
-			bnx2x_set_spio(bp, MISC_REGISTERS_SPIO_5,
-				       MISC_REGISTERS_SPIO_INPUT_HI_Z);
+	/* EMAC0 COMMON comes here */
+	/* EMAC1 COMMON comes here */
+	/* DBU COMMON comes here */
+	/* DBG COMMON comes here */
+
+	bnx2x_init_block(bp, NIG_COMMON_START, NIG_COMMON_END);
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR(bp, NIG_REG_LLH_MF_MODE, IS_E1HMF(bp));
+		REG_WR(bp, NIG_REG_LLH_E1HOV_MODE, IS_E1HMF(bp));
+	}
+
+	if (CHIP_REV_IS_SLOW(bp))
+		msleep(200);
+
+	/* finish CFC init */
+	val = reg_poll(bp, CFC_REG_LL_INIT_DONE, 1, 100, 10);
+	if (val != 1) {
+		BNX2X_ERR("CFC LL_INIT failed\n");
+		return -EBUSY;
+	}
+	val = reg_poll(bp, CFC_REG_AC_INIT_DONE, 1, 100, 10);
+	if (val != 1) {
+		BNX2X_ERR("CFC AC_INIT failed\n");
+		return -EBUSY;
+	}
+	val = reg_poll(bp, CFC_REG_CAM_INIT_DONE, 1, 100, 10);
+	if (val != 1) {
+		BNX2X_ERR("CFC CAM_INIT failed\n");
+		return -EBUSY;
+	}
+	REG_WR(bp, CFC_REG_DEBUG0, 0);
 
-			/* set to active low mode */
-			val = REG_RD(bp, MISC_REG_SPIO_INT);
-			val |= ((1 << MISC_REGISTERS_SPIO_5) <<
+	/* read NIG statistic
+	   to see if this is our first up since powerup */
+	bnx2x_read_dmae(bp, NIG_REG_STAT2_BRB_OCTET, 2);
+	val = *bnx2x_sp(bp, wb_data[0]);
+
+	/* do internal memory self test */
+	if ((CHIP_IS_E1(bp)) && (val == 0) && bnx2x_int_mem_test(bp)) {
+		BNX2X_ERR("internal mem self test failed\n");
+		return -EBUSY;
+	}
+
+	switch (bp->common.board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
+	case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
+		/* Fan failure is indicated by SPIO 5 */
+		bnx2x_set_spio(bp, MISC_REGISTERS_SPIO_5,
+			       MISC_REGISTERS_SPIO_INPUT_HI_Z);
+
+		/* set to active low mode */
+		val = REG_RD(bp, MISC_REG_SPIO_INT);
+		val |= ((1 << MISC_REGISTERS_SPIO_5) <<
 					MISC_REGISTERS_SPIO_INT_OLD_SET_POS);
-			REG_WR(bp, MISC_REG_SPIO_INT, val);
+		REG_WR(bp, MISC_REG_SPIO_INT, val);
 
-			/* enable interrupt to signal the IGU */
-			val = REG_RD(bp, MISC_REG_SPIO_EVENT_EN);
-			val |= (1 << MISC_REGISTERS_SPIO_5);
-			REG_WR(bp, MISC_REG_SPIO_EVENT_EN, val);
-			break;
+		/* enable interrupt to signal the IGU */
+		val = REG_RD(bp, MISC_REG_SPIO_EVENT_EN);
+		val |= (1 << MISC_REGISTERS_SPIO_5);
+		REG_WR(bp, MISC_REG_SPIO_EVENT_EN, val);
+		break;
 
-		default:
-			break;
-		}
+	default:
+		break;
+	}
 
-	} /* end of common init */
+	/* clear PXP2 attentions */
+	REG_RD(bp, PXP2_REG_PXP2_INT_STS_CLR_0);
 
-	/* per port init */
+	enable_blocks_attention(bp);
 
-	/* the phys address is shifted right 12 bits and has an added
-	   1=valid bit added to the 53rd bit
-	   then since this is a wide register(TM)
-	   we split it into two 32 bit writes
-	 */
-#define RQ_ONCHIP_AT_PORT_SIZE  384
-#define ONCHIP_ADDR1(x)   ((u32)(((u64)x >> 12) & 0xFFFFFFFF))
-#define ONCHIP_ADDR2(x)   ((u32)((1 << 20) | ((u64)x >> 44)))
-#define PXP_ONE_ILT(x)    ((x << 10) | x)
+	return 0;
+}
 
-	DP(BNX2X_MSG_MCP, "starting per-function init port is %x\n", func);
+static int bnx2x_init_port(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	u32 val;
 
-	REG_WR(bp, NIG_REG_MASK_INTERRUPT_PORT0 + func*4, 0);
+	DP(BNX2X_MSG_MCP, "starting port init  port %x\n", port);
+
+	REG_WR(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4, 0);
 
 	/* Port PXP comes here */
 	/* Port PXP2 comes here */
-
-	/* Offset is
-	 * Port0  0
-	 * Port1  384 */
-	i = func * RQ_ONCHIP_AT_PORT_SIZE;
-#ifdef USE_DMAE
-	wb_write[0] = ONCHIP_ADDR1(bnx2x_sp_mapping(bp, context));
-	wb_write[1] = ONCHIP_ADDR2(bnx2x_sp_mapping(bp, context));
-	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
-#else
-	REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + i*8,
-		   ONCHIP_ADDR1(bnx2x_sp_mapping(bp, context)));
-	REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + i*8 + 4,
-		   ONCHIP_ADDR2(bnx2x_sp_mapping(bp, context)));
-#endif
-	REG_WR(bp, PXP2_REG_PSWRQ_CDU0_L2P + func*4, PXP_ONE_ILT(i));
-
 #ifdef BCM_ISCSI
 	/* Port0  1
 	 * Port1  385 */
@@ -4004,30 +4436,9 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 	REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
 	REG_WR(bp, PXP2_REG_PSWRQ_SRC0_L2P + func*4, PXP_ONE_ILT(i));
 #endif
-
-	/* Port TCM comes here */
-	/* Port UCM comes here */
-	/* Port CCM comes here */
-	bnx2x_init_block(bp, func ? XCM_PORT1_START : XCM_PORT0_START,
-			     func ? XCM_PORT1_END : XCM_PORT0_END);
-
-#ifdef USE_DMAE
-	wb_write[0] = 0;
-	wb_write[1] = 0;
-#endif
-	for (i = 0; i < 32; i++) {
-		REG_WR(bp, QM_REG_BASEADDR + (func*32 + i)*4, 1024 * 4 * i);
-#ifdef USE_DMAE
-		REG_WR_DMAE(bp, QM_REG_PTRTBL + (func*32 + i)*8, wb_write, 2);
-#else
-		REG_WR_IND(bp, QM_REG_PTRTBL + (func*32 + i)*8, 0);
-		REG_WR_IND(bp, QM_REG_PTRTBL + (func*32 + i)*8 + 4, 0);
-#endif
-	}
-	REG_WR(bp, QM_REG_CONNNUM_0 + func*4, 1024/16 - 1);
+	/* Port CMs come here */
 
 	/* Port QM comes here */
-
 #ifdef BCM_ISCSI
 	REG_WR(bp, TM_REG_LIN0_SCAN_TIME + func*4, 1024/64*20);
 	REG_WR(bp, TM_REG_LIN0_MAX_ACTIVE_CID + func*4, 31);
@@ -4042,31 +4453,32 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 	/* Port CSDM comes here */
 	/* Port USDM comes here */
 	/* Port XSDM comes here */
-	bnx2x_init_block(bp, func ? TSEM_PORT1_START : TSEM_PORT0_START,
-			     func ? TSEM_PORT1_END : TSEM_PORT0_END);
-	bnx2x_init_block(bp, func ? USEM_PORT1_START : USEM_PORT0_START,
-			     func ? USEM_PORT1_END : USEM_PORT0_END);
-	bnx2x_init_block(bp, func ? CSEM_PORT1_START : CSEM_PORT0_START,
-			     func ? CSEM_PORT1_END : CSEM_PORT0_END);
-	bnx2x_init_block(bp, func ? XSEM_PORT1_START : XSEM_PORT0_START,
-			     func ? XSEM_PORT1_END : XSEM_PORT0_END);
+	bnx2x_init_block(bp, port ? TSEM_PORT1_START : TSEM_PORT0_START,
+			     port ? TSEM_PORT1_END : TSEM_PORT0_END);
+	bnx2x_init_block(bp, port ? USEM_PORT1_START : USEM_PORT0_START,
+			     port ? USEM_PORT1_END : USEM_PORT0_END);
+	bnx2x_init_block(bp, port ? CSEM_PORT1_START : CSEM_PORT0_START,
+			     port ? CSEM_PORT1_END : CSEM_PORT0_END);
+	bnx2x_init_block(bp, port ? XSEM_PORT1_START : XSEM_PORT0_START,
+			     port ? XSEM_PORT1_END : XSEM_PORT0_END);
 	/* Port UPB comes here */
-	/* Port XSDM comes here */
-	bnx2x_init_block(bp, func ? PBF_PORT1_START : PBF_PORT0_START,
-			     func ? PBF_PORT1_END : PBF_PORT0_END);
+	/* Port XPB comes here */
+
+	bnx2x_init_block(bp, port ? PBF_PORT1_START : PBF_PORT0_START,
+			     port ? PBF_PORT1_END : PBF_PORT0_END);
 
 	/* configure PBF to work without PAUSE mtu 9000 */
-	REG_WR(bp, PBF_REG_P0_PAUSE_ENABLE + func*4, 0);
+	REG_WR(bp, PBF_REG_P0_PAUSE_ENABLE + port*4, 0);
 
 	/* update threshold */
-	REG_WR(bp, PBF_REG_P0_ARB_THRSH + func*4, (9040/16));
+	REG_WR(bp, PBF_REG_P0_ARB_THRSH + port*4, (9040/16));
 	/* update init credit */
-	REG_WR(bp, PBF_REG_P0_INIT_CRD + func*4, (9040/16) + 553 - 22);
+	REG_WR(bp, PBF_REG_P0_INIT_CRD + port*4, (9040/16) + 553 - 22);
 
 	/* probe changes */
-	REG_WR(bp, PBF_REG_INIT_P0 + func*4, 1);
+	REG_WR(bp, PBF_REG_INIT_P0 + port*4, 1);
 	msleep(5);
-	REG_WR(bp, PBF_REG_INIT_P0 + func*4, 0);
+	REG_WR(bp, PBF_REG_INIT_P0 + port*4, 0);
 
 #ifdef BCM_ISCSI
 	/* tell the searcher where the T2 table is */
@@ -4084,23 +4496,57 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 #endif
 	/* Port CDU comes here */
 	/* Port CFC comes here */
-	bnx2x_init_block(bp, func ? HC_PORT1_START : HC_PORT0_START,
-			     func ? HC_PORT1_END : HC_PORT0_END);
-	bnx2x_init_block(bp, func ? MISC_AEU_PORT1_START :
+
+	if (CHIP_IS_E1(bp)) {
+		REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, 0);
+		REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, 0);
+	}
+	bnx2x_init_block(bp, port ? HC_PORT1_START : HC_PORT0_START,
+			     port ? HC_PORT1_END : HC_PORT0_END);
+
+	bnx2x_init_block(bp, port ? MISC_AEU_PORT1_START :
 				    MISC_AEU_PORT0_START,
-			     func ? MISC_AEU_PORT1_END : MISC_AEU_PORT0_END);
+			     port ? MISC_AEU_PORT1_END : MISC_AEU_PORT0_END);
+	/* init aeu_mask_attn_func_0/1:
+	 *  - SF mode: bits 3-7 are masked. only bits 0-2 are in use
+	 *  - MF mode: bit 3 is masked. bits 0-2 are in use as in SF
+	 *             bits 4-7 are used for "per vn group attention" */
+	REG_WR(bp, MISC_REG_AEU_MASK_ATTN_FUNC_0 + port*4,
+	       (IS_E1HMF(bp) ? 0xF7 : 0x7));
+
 	/* Port PXPCS comes here */
 	/* Port EMAC0 comes here */
 	/* Port EMAC1 comes here */
 	/* Port DBU comes here */
 	/* Port DBG comes here */
-	bnx2x_init_block(bp, func ? NIG_PORT1_START : NIG_PORT0_START,
-			     func ? NIG_PORT1_END : NIG_PORT0_END);
-	REG_WR(bp, NIG_REG_XGXS_SERDES0_MODE_SEL + func*4, 1);
+	bnx2x_init_block(bp, port ? NIG_PORT1_START : NIG_PORT0_START,
+			     port ? NIG_PORT1_END : NIG_PORT0_END);
+
+	REG_WR(bp, NIG_REG_XGXS_SERDES0_MODE_SEL + port*4, 1);
+
+	if (CHIP_IS_E1H(bp)) {
+		u32 wsum;
+		struct cmng_struct_per_port m_cmng_port;
+		int vn;
+
+		/* 0x2 disable e1hov, 0x1 enable */
+		REG_WR(bp, NIG_REG_LLH0_BRB1_DRV_MASK_MF + port*4,
+		       (IS_E1HMF(bp) ? 0x1 : 0x2));
+
+		/* Init RATE SHAPING and FAIRNESS contexts.
+		   Initialize as if there is 10G link. */
+		wsum = bnx2x_calc_vn_wsum(bp);
+		bnx2x_init_port_minmax(bp, (int)wsum, 10000, &m_cmng_port);
+		if (IS_E1HMF(bp))
+			for (vn = VN_0; vn < E1HVN_MAX; vn++)
+				bnx2x_init_vn_minmax(bp, 2*vn + port,
+					wsum, 10000, &m_cmng_port);
+	}
+
 	/* Port MCP comes here */
 	/* Port DMAE comes here */
 
-	switch (bp->board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
+	switch (bp->common.board & SHARED_HW_CFG_BOARD_TYPE_MASK) {
 	case SHARED_HW_CFG_BOARD_TYPE_BCM957710A1022G:
 		/* add SPIO 5 to group 0 */
 		val = REG_RD(bp, MISC_REG_AEU_ENABLE1_FUNC_0_OUT_0);
@@ -4114,48 +4560,150 @@ static int bnx2x_function_init(struct bnx2x *bp, int mode)
 
 	bnx2x__link_reset(bp);
 
+	return 0;
+}
+
+#define ILT_PER_FUNC		(768/2)
+#define FUNC_ILT_BASE(func)	(func * ILT_PER_FUNC)
+/* the phys address is shifted right 12 bits and has an added
+   1=valid bit added to the 53rd bit
+   then since this is a wide register(TM)
+   we split it into two 32 bit writes
+ */
+#define ONCHIP_ADDR1(x)		((u32)(((u64)x >> 12) & 0xFFFFFFFF))
+#define ONCHIP_ADDR2(x)		((u32)((1 << 20) | ((u64)x >> 44)))
+#define PXP_ONE_ILT(x)		(((x) << 10) | x)
+#define PXP_ILT_RANGE(f, l)	(((l) << 10) | f)
+
+#define CNIC_ILT_LINES		0
+
+static void bnx2x_ilt_wr(struct bnx2x *bp, u32 index, dma_addr_t addr)
+{
+	int reg;
+
+	if (CHIP_IS_E1H(bp))
+		reg = PXP2_REG_RQ_ONCHIP_AT_B0 + index*8;
+	else /* E1 */
+		reg = PXP2_REG_RQ_ONCHIP_AT + index*8;
+
+	bnx2x_wb_wr(bp, reg, ONCHIP_ADDR1(addr), ONCHIP_ADDR2(addr));
+}
+
+static int bnx2x_init_func(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	int i;
+
+	DP(BNX2X_MSG_MCP, "starting func init  func %x\n", func);
+
+	i = FUNC_ILT_BASE(func);
+
+	bnx2x_ilt_wr(bp, i, bnx2x_sp_mapping(bp, context));
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR(bp, PXP2_REG_RQ_CDU_FIRST_ILT, i);
+		REG_WR(bp, PXP2_REG_RQ_CDU_LAST_ILT, i + CNIC_ILT_LINES);
+	} else /* E1 */
+		REG_WR(bp, PXP2_REG_PSWRQ_CDU0_L2P + func*4,
+		       PXP_ILT_RANGE(i, i + CNIC_ILT_LINES));
+
+
+	if (CHIP_IS_E1H(bp)) {
+		for (i = 0; i < 9; i++)
+			bnx2x_init_block(bp,
+					 cm_start[func][i], cm_end[func][i]);
+
+		REG_WR(bp, NIG_REG_LLH0_FUNC_EN + port*8, 1);
+		REG_WR(bp, NIG_REG_LLH0_FUNC_VLAN_ID + port*8, bp->e1hov);
+	}
+
+	/* HC init per function */
+	if (CHIP_IS_E1H(bp)) {
+		REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_12 + func*4, 0);
+
+		REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, 0);
+		REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, 0);
+	}
+	bnx2x_init_block(bp, hc_limits[func][0], hc_limits[func][1]);
+
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, HC_REG_FUNC_NUM_P0 + port*4, func);
+
 	/* Reset PCIE errors for debug */
 	REG_WR(bp, 0x2114, 0xffffffff);
 	REG_WR(bp, 0x2120, 0xffffffff);
-	REG_WR(bp, 0x2814, 0xffffffff);
 
-	/* !!! move to init_values.h */
-	REG_WR(bp, XSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
-	REG_WR(bp, USDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
-	REG_WR(bp, CSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
-	REG_WR(bp, TSDM_REG_INIT_CREDIT_PXP_CTRL, 0x1);
+	return 0;
+}
+
+static int bnx2x_init_hw(struct bnx2x *bp, u32 load_code)
+{
+	int i, rc = 0;
 
-	REG_WR(bp, DBG_REG_PCI_REQ_CREDIT, 0x1);
-	REG_WR(bp, TM_REG_PCIARB_CRDCNT_VAL, 0x1);
-	REG_WR(bp, CDU_REG_CDU_DEBUG, 0x264);
-	REG_WR(bp, CDU_REG_CDU_DEBUG, 0x0);
+	DP(BNX2X_MSG_MCP, "function %d  load_code %x\n",
+	   BP_FUNC(bp), load_code);
 
-	bnx2x_gunzip_end(bp);
+	bp->dmae_ready = 0;
+	mutex_init(&bp->dmae_mutex);
+	bnx2x_gunzip_init(bp);
 
-	if (!nomcp) {
-		port = bp->port;
+	switch (load_code) {
+	case FW_MSG_CODE_DRV_LOAD_COMMON:
+		rc = bnx2x_init_common(bp);
+		if (rc)
+			goto init_hw_err;
+		/* no break */
+
+	case FW_MSG_CODE_DRV_LOAD_PORT:
+		bp->dmae_ready = 1;
+		rc = bnx2x_init_port(bp);
+		if (rc)
+			goto init_hw_err;
+		/* no break */
+
+	case FW_MSG_CODE_DRV_LOAD_FUNCTION:
+		bp->dmae_ready = 1;
+		rc = bnx2x_init_func(bp);
+		if (rc)
+			goto init_hw_err;
+		break;
+
+	default:
+		BNX2X_ERR("Unknown load_code (0x%x) from MCP\n", load_code);
+		break;
+	}
+
+	if (!BP_NOMCP(bp)) {
+		int func = BP_FUNC(bp);
 
 		bp->fw_drv_pulse_wr_seq =
-				(SHMEM_RD(bp, func_mb[port].drv_pulse_mb) &
+				(SHMEM_RD(bp, func_mb[func].drv_pulse_mb) &
 				 DRV_PULSE_SEQ_MASK);
-		bp->fw_mb = SHMEM_RD(bp, func_mb[port].fw_mb_param);
-		DP(BNX2X_MSG_MCP, "drv_pulse 0x%x  fw_mb 0x%x\n",
-		   bp->fw_drv_pulse_wr_seq, bp->fw_mb);
-	} else {
-		bp->fw_mb = 0;
-	}
+		bp->func_stx = SHMEM_RD(bp, func_mb[func].fw_mb_param);
+		DP(BNX2X_MSG_MCP, "drv_pulse 0x%x  func_stx 0x%x\n",
+		   bp->fw_drv_pulse_wr_seq, bp->func_stx);
+	} else
+		bp->func_stx = 0;
 
-	return 0;
+	/* this needs to be done before gunzip end */
+	bnx2x_zero_def_sb(bp);
+	for_each_queue(bp, i)
+		bnx2x_zero_sb(bp, BP_L_ID(bp) + i);
+
+init_hw_err:
+	bnx2x_gunzip_end(bp);
+
+	return rc;
 }
 
 /* send the MCP a request, block until there is a reply */
 static u32 bnx2x_fw_command(struct bnx2x *bp, u32 command)
 {
-	int port = bp->port;
+	int func = BP_FUNC(bp);
 	u32 seq = ++bp->fw_seq;
 	u32 rc = 0;
 
-	SHMEM_WR(bp, func_mb[port].drv_mb_header, (command | seq));
+	SHMEM_WR(bp, func_mb[func].drv_mb_header, (command | seq));
 	DP(BNX2X_MSG_MCP, "wrote command (%x) to FW MB\n", (command | seq));
 
 	/* let the FW do it's magic ... */
@@ -4164,7 +4712,7 @@ static u32 bnx2x_fw_command(struct bnx2x *bp, u32 command)
 	if (CHIP_REV_IS_SLOW(bp))
 		msleep(900);
 
-	rc = SHMEM_RD(bp, func_mb[port].fw_mb_header);
+	rc = SHMEM_RD(bp, func_mb[func].fw_mb_header);
 	DP(BNX2X_MSG_MCP, "read (%x) seq is (%x) from FW MB\n", rc, seq);
 
 	/* is this a reply to our command? */
@@ -4229,15 +4777,13 @@ static void bnx2x_free_mem(struct bnx2x *bp)
 			       NUM_RCQ_BD);
 	}
 
-	BNX2X_FREE(bp->fp);
-
 	/* end of fastpath */
 
 	BNX2X_PCI_FREE(bp->def_status_blk, bp->def_status_blk_mapping,
-		       (sizeof(struct host_def_status_block)));
+		       sizeof(struct host_def_status_block));
 
 	BNX2X_PCI_FREE(bp->slowpath, bp->slowpath_mapping,
-		       (sizeof(struct bnx2x_slowpath)));
+		       sizeof(struct bnx2x_slowpath));
 
 #ifdef BCM_ISCSI
 	BNX2X_PCI_FREE(bp->t1, bp->t1_mapping, 64*1024);
@@ -4273,8 +4819,6 @@ static int bnx2x_alloc_mem(struct bnx2x *bp)
 	int i;
 
 	/* fastpath */
-	BNX2X_ALLOC(bp->fp, sizeof(struct bnx2x_fastpath) * bp->num_queues);
-
 	for_each_queue(bp, i) {
 		bnx2x_fp(bp, i, bp) = bp;
 
@@ -4370,8 +4914,6 @@ static void bnx2x_free_tx_skbs(struct bnx2x *bp)
 		u16 sw_prod = fp->tx_pkt_prod;
 		u16 sw_cons = fp->tx_pkt_cons;
 
-		BUG_TRAP(fp->tx_buf_ring != NULL);
-
 		while (sw_cons != sw_prod) {
 			bd_cons = bnx2x_free_tx_pkt(bp, fp, TX_BD(sw_cons));
 			sw_cons++;
@@ -4386,8 +4928,6 @@ static void bnx2x_free_rx_skbs(struct bnx2x *bp)
 	for_each_queue(bp, j) {
 		struct bnx2x_fastpath *fp = &bp->fp[j];
 
-		BUG_TRAP(fp->rx_buf_ring != NULL);
-
 		for (i = 0; i < NUM_RX_BD; i++) {
 			struct sw_rx_bd *rx_buf = &fp->rx_buf_ring[i];
 			struct sk_buff *skb = rx_buf->skb;
@@ -4414,7 +4954,7 @@ static void bnx2x_free_skbs(struct bnx2x *bp)
 
 static void bnx2x_free_msix_irqs(struct bnx2x *bp)
 {
-	int i;
+	int i, offset = 1;
 
 	free_irq(bp->msix_table[0].vector, bp->dev);
 	DP(NETIF_MSG_IFDOWN, "released sp irq (%d)\n",
@@ -4422,26 +4962,22 @@ static void bnx2x_free_msix_irqs(struct bnx2x *bp)
 
 	for_each_queue(bp, i) {
 		DP(NETIF_MSG_IFDOWN, "about to release fp #%d->%d irq  "
-		   "state(%x)\n", i, bp->msix_table[i + 1].vector,
+		   "state %x\n", i, bp->msix_table[i + offset].vector,
 		   bnx2x_fp(bp, i, state));
 
 		if (bnx2x_fp(bp, i, state) != BNX2X_FP_STATE_CLOSED)
 			BNX2X_ERR("IRQ of fp #%d being freed while "
 				  "state != closed\n", i);
 
-		free_irq(bp->msix_table[i + 1].vector, &bp->fp[i]);
+		free_irq(bp->msix_table[i + offset].vector, &bp->fp[i]);
 	}
-
 }
 
 static void bnx2x_free_irq(struct bnx2x *bp)
 {
-
 	if (bp->flags & USING_MSIX_FLAG) {
-
 		bnx2x_free_msix_irqs(bp);
 		pci_disable_msix(bp->pdev);
-
 		bp->flags &= ~USING_MSIX_FLAG;
 
 	} else
@@ -4450,87 +4986,87 @@ static void bnx2x_free_irq(struct bnx2x *bp)
 
 static int bnx2x_enable_msix(struct bnx2x *bp)
 {
-
-	int i;
+	int i, rc, offset;
 
 	bp->msix_table[0].entry = 0;
-	for_each_queue(bp, i)
-		bp->msix_table[i + 1].entry = i + 1;
+	offset = 1;
+	DP(NETIF_MSG_IFUP, "msix_table[0].entry = 0 (slowpath)\n");
 
-	if (pci_enable_msix(bp->pdev, &bp->msix_table[0],
-				     bp->num_queues + 1)){
-		BNX2X_LOG("failed to enable MSI-X\n");
-		return -1;
+	for_each_queue(bp, i) {
+		int igu_vec = offset + i + BP_L_ID(bp);
 
+		bp->msix_table[i + offset].entry = igu_vec;
+		DP(NETIF_MSG_IFUP, "msix_table[%d].entry = %d "
+		   "(fastpath #%u)\n", i + offset, igu_vec, i);
 	}
 
+	rc = pci_enable_msix(bp->pdev, &bp->msix_table[0],
+			     bp->num_queues + offset);
+	if (rc) {
+		DP(NETIF_MSG_IFUP, "MSI-X is not attainable\n");
+		return -1;
+	}
 	bp->flags |= USING_MSIX_FLAG;
 
 	return 0;
-
 }
 
-
 static int bnx2x_req_msix_irqs(struct bnx2x *bp)
 {
-
-	int i, rc;
+	int i, rc, offset = 1;
 
 	rc = request_irq(bp->msix_table[0].vector, bnx2x_msix_sp_int, 0,
 			 bp->dev->name, bp->dev);
-
 	if (rc) {
 		BNX2X_ERR("request sp irq failed\n");
 		return -EBUSY;
 	}
 
 	for_each_queue(bp, i) {
-		rc = request_irq(bp->msix_table[i + 1].vector,
+		rc = request_irq(bp->msix_table[i + offset].vector,
 				 bnx2x_msix_fp_int, 0,
 				 bp->dev->name, &bp->fp[i]);
-
 		if (rc) {
-			BNX2X_ERR("request fp #%d irq failed  "
-				  "rc %d\n", i, rc);
+			BNX2X_ERR("request fp #%d irq failed  rc %d\n",
+				  i + offset, rc);
 			bnx2x_free_msix_irqs(bp);
 			return -EBUSY;
 		}
 
 		bnx2x_fp(bp, i, state) = BNX2X_FP_STATE_IRQ;
-
 	}
 
 	return 0;
-
 }
 
 static int bnx2x_req_irq(struct bnx2x *bp)
 {
+	int rc;
 
-	int rc = request_irq(bp->pdev->irq, bnx2x_interrupt,
-			     IRQF_SHARED, bp->dev->name, bp->dev);
+	rc = request_irq(bp->pdev->irq, bnx2x_interrupt, IRQF_SHARED,
+			 bp->dev->name, bp->dev);
 	if (!rc)
 		bnx2x_fp(bp, 0, state) = BNX2X_FP_STATE_IRQ;
 
 	return rc;
-
 }
 
 /*
  * Init service functions
  */
 
-static void bnx2x_set_mac_addr(struct bnx2x *bp)
+static void bnx2x_set_mac_addr_e1(struct bnx2x *bp)
 {
 	struct mac_configuration_cmd *config = bnx2x_sp(bp, mac_config);
+	int port = BP_PORT(bp);
 
 	/* CAM allocation
 	 * unicasts 0-31:port0 32-63:port1
 	 * multicast 64-127:port0 128-191:port1
 	 */
 	config->hdr.length_6b = 2;
-	config->hdr.offset = bp->port ? 31 : 0;
-	config->hdr.reserved0 = 0;
+	config->hdr.offset = port ? 31 : 0;
+	config->hdr.client_id = BP_CL_ID(bp);
 	config->hdr.reserved1 = 0;
 
 	/* primary MAC */
@@ -4540,7 +5076,7 @@ static void bnx2x_set_mac_addr(struct bnx2x *bp)
 					swab16(*(u16 *)&bp->dev->dev_addr[2]);
 	config->config_table[0].cam_entry.lsb_mac_addr =
 					swab16(*(u16 *)&bp->dev->dev_addr[4]);
-	config->config_table[0].cam_entry.flags = cpu_to_le16(bp->port);
+	config->config_table[0].cam_entry.flags = cpu_to_le16(port);
 	config->config_table[0].target_table_entry.flags = 0;
 	config->config_table[0].target_table_entry.client_id = 0;
 	config->config_table[0].target_table_entry.vlan_id = 0;
@@ -4554,7 +5090,7 @@ static void bnx2x_set_mac_addr(struct bnx2x *bp)
 	config->config_table[1].cam_entry.msb_mac_addr = 0xffff;
 	config->config_table[1].cam_entry.middle_mac_addr = 0xffff;
 	config->config_table[1].cam_entry.lsb_mac_addr = 0xffff;
-	config->config_table[1].cam_entry.flags = cpu_to_le16(bp->port);
+	config->config_table[1].cam_entry.flags = cpu_to_le16(port);
 	config->config_table[1].target_table_entry.flags =
 				TSTORM_CAM_TARGET_TABLE_ENTRY_BROADCAST;
 	config->config_table[1].target_table_entry.client_id = 0;
@@ -4565,64 +5101,105 @@ static void bnx2x_set_mac_addr(struct bnx2x *bp)
 		      U64_LO(bnx2x_sp_mapping(bp, mac_config)), 0);
 }
 
+static void bnx2x_set_mac_addr_e1h(struct bnx2x *bp)
+{
+	struct mac_configuration_cmd_e1h *config =
+		(struct mac_configuration_cmd_e1h *)bnx2x_sp(bp, mac_config);
+
+	if (bp->state != BNX2X_STATE_OPEN) {
+		DP(NETIF_MSG_IFUP, "state is %x, returning\n", bp->state);
+		return;
+	}
+
+	/* CAM allocation for E1H
+	 * unicasts: by func number
+	 * multicast: 20+FUNC*20, 20 each
+	 */
+	config->hdr.length_6b = 1;
+	config->hdr.offset = BP_FUNC(bp);
+	config->hdr.client_id = BP_CL_ID(bp);
+	config->hdr.reserved1 = 0;
+
+	/* primary MAC */
+	config->config_table[0].msb_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[0]);
+	config->config_table[0].middle_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[2]);
+	config->config_table[0].lsb_mac_addr =
+					swab16(*(u16 *)&bp->dev->dev_addr[4]);
+	config->config_table[0].client_id = BP_L_ID(bp);
+	config->config_table[0].vlan_id = 0;
+	config->config_table[0].e1hov_id = cpu_to_le16(bp->e1hov);
+	config->config_table[0].flags = BP_PORT(bp);
+
+	DP(NETIF_MSG_IFUP, "setting MAC (%04x:%04x:%04x)  E1HOV %d  CLID %d\n",
+	   config->config_table[0].msb_mac_addr,
+	   config->config_table[0].middle_mac_addr,
+	   config->config_table[0].lsb_mac_addr, bp->e1hov, BP_L_ID(bp));
+
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
+		      U64_HI(bnx2x_sp_mapping(bp, mac_config)),
+		      U64_LO(bnx2x_sp_mapping(bp, mac_config)), 0);
+}
+
 static int bnx2x_wait_ramrod(struct bnx2x *bp, int state, int idx,
 			     int *state_p, int poll)
 {
 	/* can take a while if any port is running */
-	int timeout = 500;
+	int cnt = 500;
 
 	DP(NETIF_MSG_IFUP, "%s for state to become %x on IDX [%d]\n",
 	   poll ? "polling" : "waiting", state, idx);
 
 	might_sleep();
-
-	while (timeout) {
-
+	while (cnt--) {
 		if (poll) {
 			bnx2x_rx_int(bp->fp, 10);
-			/* If index is different from 0
-			 * The reply for some commands will
+			/* if index is different from 0
+			 * the reply for some commands will
 			 * be on the none default queue
 			 */
 			if (idx)
 				bnx2x_rx_int(&bp->fp[idx], 10);
 		}
-
-		mb(); /* state is changed by bnx2x_sp_event()*/
+		mb(); /* state is changed by bnx2x_sp_event() */
 
 		if (*state_p == state)
 			return 0;
 
-		timeout--;
 		msleep(1);
-
 	}
 
 	/* timeout! */
 	BNX2X_ERR("timeout %s for state %x on IDX [%d]\n",
 		  poll ? "polling" : "waiting", state, idx);
+#ifdef BNX2X_STOP_ON_ERROR
+	bnx2x_panic();
+#endif
 
 	return -EBUSY;
 }
 
 static int bnx2x_setup_leading(struct bnx2x *bp)
 {
+	int rc;
 
 	/* reset IGU state */
-	bnx2x_ack_sb(bp, DEF_SB_ID, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_ack_sb(bp, bp->fp[0].sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
 
 	/* SETUP ramrod */
 	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_PORT_SETUP, 0, 0, 0, 0);
 
-	return bnx2x_wait_ramrod(bp, BNX2X_STATE_OPEN, 0, &(bp->state), 0);
+	/* Wait for completion */
+	rc = bnx2x_wait_ramrod(bp, BNX2X_STATE_OPEN, 0, &(bp->state), 0);
 
+	return rc;
 }
 
 static int bnx2x_setup_multi(struct bnx2x *bp, int index)
 {
-
 	/* reset IGU state */
-	bnx2x_ack_sb(bp, index, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
+	bnx2x_ack_sb(bp, bp->fp[index].sb_id, CSTORM_ID, 0, IGU_INT_ENABLE, 0);
 
 	/* SETUP ramrod */
 	bp->fp[index].state = BNX2X_FP_STATE_OPENING;
@@ -4631,82 +5208,107 @@ static int bnx2x_setup_multi(struct bnx2x *bp, int index)
 	/* Wait for completion */
 	return bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_OPEN, index,
 				 &(bp->fp[index].state), 0);
-
 }
 
-
 static int bnx2x_poll(struct napi_struct *napi, int budget);
 static void bnx2x_set_rx_mode(struct net_device *dev);
 
-static int bnx2x_nic_load(struct bnx2x *bp, int req_irq)
+/* must be called with rtnl_lock */
+static int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
 {
 	u32 load_code;
-	int i;
+	int i, rc;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return -EPERM;
+#endif
 
 	bp->state = BNX2X_STATE_OPENING_WAIT4_LOAD;
 
-	/* Send LOAD_REQUEST command to MCP.
-	   Returns the type of LOAD command: if it is the
-	   first port to be initialized common blocks should be
-	   initialized, otherwise - not.
+	/* Send LOAD_REQUEST command to MCP
+	   Returns the type of LOAD command:
+	   if it is the first port to be initialized
+	   common blocks should be initialized, otherwise - not
 	*/
-	if (!nomcp) {
+	if (!BP_NOMCP(bp)) {
 		load_code = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_REQ);
 		if (!load_code) {
 			BNX2X_ERR("MCP response failure, unloading\n");
 			return -EBUSY;
 		}
-		if (load_code == FW_MSG_CODE_DRV_LOAD_REFUSED) {
-			BNX2X_ERR("MCP refused load request, unloading\n");
+		if (load_code == FW_MSG_CODE_DRV_LOAD_REFUSED)
 			return -EBUSY; /* other port in diagnostic mode */
-		}
+
 	} else {
-		load_code = FW_MSG_CODE_DRV_LOAD_COMMON;
+		DP(NETIF_MSG_IFUP, "NO MCP load counts before us %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		load_count[0]++;
+		load_count[1 + BP_PORT(bp)]++;
+		DP(NETIF_MSG_IFUP, "NO MCP new load counts       %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		if (load_count[0] == 1)
+			load_code = FW_MSG_CODE_DRV_LOAD_COMMON;
+		else if (load_count[1 + BP_PORT(bp)] == 1)
+			load_code = FW_MSG_CODE_DRV_LOAD_PORT;
+		else
+			load_code = FW_MSG_CODE_DRV_LOAD_FUNCTION;
 	}
 
-	/* if we can't use msix we only need one fp,
-	 * so try to enable msix with the requested number of fp's
+	if ((load_code == FW_MSG_CODE_DRV_LOAD_COMMON) ||
+	    (load_code == FW_MSG_CODE_DRV_LOAD_PORT))
+		bp->port.pmf = 1;
+	else
+		bp->port.pmf = 0;
+	DP(NETIF_MSG_LINK, "pmf %d\n", bp->port.pmf);
+
+	/* if we can't use MSI-X we only need one fp,
+	 * so try to enable MSI-X with the requested number of fp's
 	 * and fallback to inta with one fp
 	 */
-	if (req_irq) {
-		if (use_inta) {
+	if (use_inta) {
+		bp->num_queues = 1;
+
+	} else {
+		if ((use_multi > 1) && (use_multi <= BP_MAX_QUEUES(bp)))
+			/* user requested number */
+			bp->num_queues = use_multi;
+
+		else if (use_multi)
+			bp->num_queues = min_t(u32, num_online_cpus(),
+					       BP_MAX_QUEUES(bp));
+		else
 			bp->num_queues = 1;
-		} else {
-			if ((use_multi > 1) && (use_multi <= 16))
-				/* user requested number */
-				bp->num_queues = use_multi;
-			else if (use_multi == 1)
-				bp->num_queues = num_online_cpus();
-			else
-				bp->num_queues = 1;
-
-			if (bnx2x_enable_msix(bp)) {
-				/* failed to enable msix */
-				bp->num_queues = 1;
-				if (use_multi)
-					BNX2X_ERR("Multi requested but failed"
-						  " to enable MSI-X\n");
-			}
+
+		if (bnx2x_enable_msix(bp)) {
+			/* failed to enable MSI-X */
+			bp->num_queues = 1;
+			if (use_multi)
+				BNX2X_ERR("Multi requested but failed"
+					  " to enable MSI-X\n");
 		}
 	}
-
-	DP(NETIF_MSG_IFUP, "set number of queues to %d\n", bp->num_queues);
+	DP(NETIF_MSG_IFUP,
+	   "set number of queues to %d\n", bp->num_queues);
 
 	if (bnx2x_alloc_mem(bp))
 		return -ENOMEM;
 
-	if (req_irq) {
-		if (bp->flags & USING_MSIX_FLAG) {
-			if (bnx2x_req_msix_irqs(bp)) {
-				pci_disable_msix(bp->pdev);
-				goto load_error;
-			}
+	/* Disable interrupt handling until HW is initialized */
+	atomic_set(&bp->intr_sem, 1);
 
-		} else {
-			if (bnx2x_req_irq(bp)) {
-				BNX2X_ERR("IRQ request failed, aborting\n");
-				goto load_error;
-			}
+	if (bp->flags & USING_MSIX_FLAG) {
+		rc = bnx2x_req_msix_irqs(bp);
+		if (rc) {
+			pci_disable_msix(bp->pdev);
+			goto load_error;
+		}
+	} else {
+		bnx2x_ack_int(bp);
+		rc = bnx2x_req_irq(bp);
+		if (rc) {
+			BNX2X_ERR("IRQ request failed, aborting\n");
+			goto load_error;
 		}
 	}
 
@@ -4714,26 +5316,25 @@ static int bnx2x_nic_load(struct bnx2x *bp, int req_irq)
 		netif_napi_add(bp->dev, &bnx2x_fp(bp, i, napi),
 			       bnx2x_poll, 128);
 
-
 	/* Initialize HW */
-	if (bnx2x_function_init(bp,
-				(load_code == FW_MSG_CODE_DRV_LOAD_COMMON))) {
+	rc = bnx2x_init_hw(bp, load_code);
+	if (rc) {
 		BNX2X_ERR("HW init failed, aborting\n");
 		goto load_error;
 	}
 
-
+	/* Enable interrupt handling */
 	atomic_set(&bp->intr_sem, 0);
 
-
 	/* Setup NIC internals and enable interrupts */
 	bnx2x_nic_init(bp);
 
 	/* Send LOAD_DONE command to MCP */
-	if (!nomcp) {
+	if (!BP_NOMCP(bp)) {
 		load_code = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE);
 		if (!load_code) {
 			BNX2X_ERR("MCP response failure, unloading\n");
+			rc = -EBUSY;
 			goto load_int_disable;
 		}
 	}
@@ -4745,33 +5346,68 @@ static int bnx2x_nic_load(struct bnx2x *bp, int req_irq)
 	for_each_queue(bp, i)
 		napi_enable(&bnx2x_fp(bp, i, napi));
 
-	if (bnx2x_setup_leading(bp))
+	rc = bnx2x_setup_leading(bp);
+	if (rc) {
+#ifdef BNX2X_STOP_ON_ERROR
+		bp->panic = 1;
+#endif
 		goto load_stop_netif;
+	}
 
-	for_each_nondefault_queue(bp, i)
-		if (bnx2x_setup_multi(bp, i))
-			goto load_stop_netif;
+	if (CHIP_IS_E1H(bp))
+		if (bp->mf_config & FUNC_MF_CFG_FUNC_DISABLED) {
+			BNX2X_ERR("!!!  mf_cfg function disabled\n");
+			bp->state = BNX2X_STATE_DISABLED;
+		}
 
-	bnx2x_set_mac_addr(bp);
+	if (bp->state == BNX2X_STATE_OPEN)
+		for_each_nondefault_queue(bp, i) {
+			rc = bnx2x_setup_multi(bp, i);
+			if (rc)
+				goto load_stop_netif;
+		}
 
-	bnx2x_initial_phy_init(bp);
+	if (CHIP_IS_E1(bp))
+		bnx2x_set_mac_addr_e1(bp);
+	else
+		bnx2x_set_mac_addr_e1h(bp);
+
+	if (bp->port.pmf)
+		bnx2x_initial_phy_init(bp);
 
 	/* Start fast path */
-	if (req_irq) { /* IRQ is only requested from bnx2x_open */
+	switch (load_mode) {
+	case LOAD_NORMAL:
+		/* Tx queue should be only reenabled */
+		netif_wake_queue(bp->dev);
+		bnx2x_set_rx_mode(bp->dev);
+		break;
+
+	case LOAD_OPEN:
+		/* IRQ is only requested from bnx2x_open */
 		netif_start_queue(bp->dev);
+		bnx2x_set_rx_mode(bp->dev);
 		if (bp->flags & USING_MSIX_FLAG)
 			printk(KERN_INFO PFX "%s: using MSI-X\n",
 			       bp->dev->name);
+		break;
 
-	/* Otherwise Tx queue should be only reenabled */
-	} else if (netif_running(bp->dev)) {
-		netif_wake_queue(bp->dev);
+	case LOAD_DIAG:
 		bnx2x_set_rx_mode(bp->dev);
+		bp->state = BNX2X_STATE_DIAG;
+		break;
+
+	default:
+		break;
 	}
 
+	if (!bp->port.pmf)
+		bnx2x__link_status_update(bp);
+
 	/* start the timer */
 	mod_timer(&bp->timer, jiffies + bp->current_interval);
 
+
 	return 0;
 
 load_stop_netif:
@@ -4781,7 +5417,7 @@ load_stop_netif:
 load_int_disable:
 	bnx2x_int_disable_sync(bp);
 
-	bnx2x_free_skbs(bp);
+	/* Release IRQs */
 	bnx2x_free_irq(bp);
 
 load_error:
@@ -4789,95 +5425,50 @@ load_error:
 
 	/* TBD we really need to reset the chip
 	   if we want to recover from this */
-	return -EBUSY;
-}
-
-
-static void bnx2x_reset_chip(struct bnx2x *bp, u32 reset_code)
-{
-	int port = bp->port;
-#ifdef USE_DMAE
-	u32 wb_write[2];
-#endif
-	int base, i;
-
-	DP(NETIF_MSG_IFDOWN, "reset called with code %x\n", reset_code);
-
-	/* Do not rcv packets to BRB */
-	REG_WR(bp, NIG_REG_LLH0_BRB1_DRV_MASK + port*4, 0x0);
-	/* Do not direct rcv packets that are not for MCP to the BRB */
-	REG_WR(bp, (port ? NIG_REG_LLH1_BRB1_NOT_MCP :
-			   NIG_REG_LLH0_BRB1_NOT_MCP), 0x0);
-
-	/* Configure IGU and AEU */
-	REG_WR(bp, HC_REG_CONFIG_0 + port*4, 0x1000);
-	REG_WR(bp, MISC_REG_AEU_MASK_ATTN_FUNC_0 + port*4, 0);
-
-	/* TODO: Close Doorbell port? */
-
-	/* Clear ILT */
-#ifdef USE_DMAE
-	wb_write[0] = 0;
-	wb_write[1] = 0;
-#endif
-	base = port * RQ_ONCHIP_AT_PORT_SIZE;
-	for (i = base; i < base + RQ_ONCHIP_AT_PORT_SIZE; i++) {
-#ifdef USE_DMAE
-		REG_WR_DMAE(bp, PXP2_REG_RQ_ONCHIP_AT + i*8, wb_write, 2);
-#else
-		REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT, 0);
-		REG_WR_IND(bp, PXP2_REG_RQ_ONCHIP_AT + 4, 0);
-#endif
-	}
-
-	if (reset_code == FW_MSG_CODE_DRV_UNLOAD_COMMON) {
-		/* reset_common */
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
-		       0xd3ffff7f);
-		REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
-		       0x1403);
-	}
+	return rc;
 }
 
 static int bnx2x_stop_multi(struct bnx2x *bp, int index)
 {
-
 	int rc;
 
 	/* halt the connection */
 	bp->fp[index].state = BNX2X_FP_STATE_HALTING;
 	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, index, 0, 0, 0);
 
-
+	/* Wait for completion */
 	rc = bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, index,
-				       &(bp->fp[index].state), 1);
+			       &(bp->fp[index].state), 1);
 	if (rc) /* timeout */
 		return rc;
 
 	/* delete cfc entry */
 	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_CFC_DEL, index, 0, 0, 1);
 
-	return bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_CLOSED, index,
-				 &(bp->fp[index].state), 1);
-
+	/* Wait for completion */
+	rc = bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_CLOSED, index,
+			       &(bp->fp[index].state), 1);
+	return rc;
 }
 
-
 static void bnx2x_stop_leading(struct bnx2x *bp)
 {
 	u16 dsb_sp_prod_idx;
 	/* if the other port is handling traffic,
 	   this can take a lot of time */
-	int timeout = 500;
+	int cnt = 500;
+	int rc;
 
 	might_sleep();
 
 	/* Send HALT ramrod */
 	bp->fp[0].state = BNX2X_FP_STATE_HALTING;
-	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, 0, 0, 0, 0);
+	bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_HALT, 0, 0, BP_CL_ID(bp), 0);
 
-	if (bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, 0,
-			       &(bp->fp[0].state), 1))
+	/* Wait for completion */
+	rc = bnx2x_wait_ramrod(bp, BNX2X_FP_STATE_HALTED, 0,
+			       &(bp->fp[0].state), 1);
+	if (rc) /* timeout */
 		return;
 
 	dsb_sp_prod_idx = *bp->dsb_sp_prod;
@@ -4889,29 +5480,110 @@ static void bnx2x_stop_leading(struct bnx2x *bp)
 	   we are going to reset the chip anyway
 	   so there is not much to do if this times out
 	 */
-	while ((dsb_sp_prod_idx == *bp->dsb_sp_prod) && timeout) {
-		timeout--;
+	while (dsb_sp_prod_idx == *bp->dsb_sp_prod) {
 		msleep(1);
-	}
-	if (!timeout) {
-		DP(NETIF_MSG_IFDOWN, "timeout polling for completion "
-		   "dsb_sp_prod 0x%x != dsb_sp_prod_idx 0x%x\n",
-		   *bp->dsb_sp_prod, dsb_sp_prod_idx);
+		if (!cnt) {
+			DP(NETIF_MSG_IFDOWN, "timeout waiting for port del "
+			   "dsb_sp_prod 0x%x != dsb_sp_prod_idx 0x%x\n",
+			   *bp->dsb_sp_prod, dsb_sp_prod_idx);
+#ifdef BNX2X_STOP_ON_ERROR
+			bnx2x_panic();
+#endif
+			break;
+		}
+		cnt--;
 	}
 	bp->state = BNX2X_STATE_CLOSING_WAIT4_UNLOAD;
 	bp->fp[0].state = BNX2X_FP_STATE_CLOSED;
 }
 
+static void bnx2x_reset_func(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	int func = BP_FUNC(bp);
+	int base, i;
+
+	/* Configure IGU */
+	REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, 0);
+	REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, 0);
+
+	REG_WR(bp, HC_REG_CONFIG_0 + port*4, 0x1000);
+
+	/* Clear ILT */
+	base = FUNC_ILT_BASE(func);
+	for (i = base; i < base + ILT_PER_FUNC; i++)
+		bnx2x_ilt_wr(bp, i, 0);
+}
+
+static void bnx2x_reset_port(struct bnx2x *bp)
+{
+	int port = BP_PORT(bp);
+	u32 val;
+
+	REG_WR(bp, NIG_REG_MASK_INTERRUPT_PORT0 + port*4, 0);
+
+	/* Do not rcv packets to BRB */
+	REG_WR(bp, NIG_REG_LLH0_BRB1_DRV_MASK + port*4, 0x0);
+	/* Do not direct rcv packets that are not for MCP to the BRB */
+	REG_WR(bp, (port ? NIG_REG_LLH1_BRB1_NOT_MCP :
+			   NIG_REG_LLH0_BRB1_NOT_MCP), 0x0);
+
+	/* Configure AEU */
+	REG_WR(bp, MISC_REG_AEU_MASK_ATTN_FUNC_0 + port*4, 0);
+
+	msleep(100);
+	/* Check for BRB port occupancy */
+	val = REG_RD(bp, BRB1_REG_PORT_NUM_OCC_BLOCKS_0 + port*4);
+	if (val)
+		DP(NETIF_MSG_IFDOWN,
+		   "BRB1 is not empty  %d blooks are occupied\n", val);
+
+	/* TODO: Close Doorbell port? */
+}
+
+static void bnx2x_reset_common(struct bnx2x *bp)
+{
+	/* reset_common */
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+	       0xd3ffff7f);
+	REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR, 0x1403);
+}
+
+static void bnx2x_reset_chip(struct bnx2x *bp, u32 reset_code)
+{
+	DP(BNX2X_MSG_MCP, "function %d  reset_code %x\n",
+	   BP_FUNC(bp), reset_code);
+
+	switch (reset_code) {
+	case FW_MSG_CODE_DRV_UNLOAD_COMMON:
+		bnx2x_reset_port(bp);
+		bnx2x_reset_func(bp);
+		bnx2x_reset_common(bp);
+		break;
+
+	case FW_MSG_CODE_DRV_UNLOAD_PORT:
+		bnx2x_reset_port(bp);
+		bnx2x_reset_func(bp);
+		break;
+
+	case FW_MSG_CODE_DRV_UNLOAD_FUNCTION:
+		bnx2x_reset_func(bp);
+		break;
 
-static int bnx2x_nic_unload(struct bnx2x *bp, int free_irq)
+	default:
+		BNX2X_ERR("Unknown reset_code (0x%x) from MCP\n", reset_code);
+		break;
+	}
+}
+
+/* msut be called with rtnl_lock */
+static int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode)
 {
 	u32 reset_code = 0;
-	int i, timeout;
+	int i, cnt;
 
 	bp->state = BNX2X_STATE_CLOSING_WAIT4_HALT;
 
-	del_timer_sync(&bp->timer);
-
 	bp->rx_mode = BNX2X_RX_MODE_NONE;
 	bnx2x_set_storm_rx_mode(bp);
 
@@ -4920,21 +5592,44 @@ static int bnx2x_nic_unload(struct bnx2x *bp, int free_irq)
 		bp->dev->trans_start = jiffies;	/* prevent tx timeout */
 	}
 
+	del_timer_sync(&bp->timer);
+	SHMEM_WR(bp, func_mb[BP_FUNC(bp)].drv_pulse_mb,
+		 (DRV_PULSE_ALWAYS_ALIVE | bp->fw_drv_pulse_wr_seq));
+
 	/* Wait until all fast path tasks complete */
 	for_each_queue(bp, i) {
 		struct bnx2x_fastpath *fp = &bp->fp[i];
 
-		timeout = 1000;
-		while (bnx2x_has_work(fp) && (timeout--))
+#ifdef BNX2X_STOP_ON_ERROR
+#ifdef __powerpc64__
+		DP(NETIF_MSG_RX_STATUS, "fp->tpa_queue_used = 0x%lx\n",
+#else
+		DP(NETIF_MSG_IFDOWN, "fp->tpa_queue_used = 0x%llx\n",
+#endif
+		   fp->tpa_queue_used);
+#endif
+		cnt = 1000;
+		smp_rmb();
+		while (bnx2x_has_work(fp)) {
 			msleep(1);
-		if (!timeout)
-			BNX2X_ERR("timeout waiting for queue[%d]\n", i);
+			if (!cnt) {
+				BNX2X_ERR("timeout waiting for queue[%d]\n",
+					  i);
+#ifdef BNX2X_STOP_ON_ERROR
+				bnx2x_panic();
+				return -EBUSY;
+#else
+				break;
+#endif
+			}
+			cnt--;
+			smp_rmb();
+		}
 	}
 
-	/* Wait until stat ramrod returns and all SP tasks complete */
-	timeout = 1000;
-	while ((bp->stat_pending || (bp->spq_left != MAX_SPQ_PENDING)) &&
-	       (timeout--))
+	/* Wait until all slow path tasks complete */
+	cnt = 1000;
+	while ((bp->spq_left != MAX_SPQ_PENDING) && cnt--)
 		msleep(1);
 
 	for_each_queue(bp, i)
@@ -4942,59 +5637,84 @@ static int bnx2x_nic_unload(struct bnx2x *bp, int free_irq)
 	/* Disable interrupts after Tx and Rx are disabled on stack level */
 	bnx2x_int_disable_sync(bp);
 
+	/* Release IRQs */
+	bnx2x_free_irq(bp);
+
 	if (bp->flags & NO_WOL_FLAG)
 		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_MCP;
 
 	else if (bp->wol) {
-		u32 emac_base = bp->port ? GRCBASE_EMAC0 : GRCBASE_EMAC1;
+		u32 emac_base = BP_PORT(bp) ? GRCBASE_EMAC1 : GRCBASE_EMAC0;
 		u8 *mac_addr = bp->dev->dev_addr;
-		u32 val = (EMAC_MODE_MPKT | EMAC_MODE_MPKT_RCVD |
-			   EMAC_MODE_ACPI_RCVD);
-
-		EMAC_WR(EMAC_REG_EMAC_MODE, val);
+		u32 val;
 
+		/* The mac address is written to entries 1-4 to
+		   preserve entry 0 which is used by the PMF */
 		val = (mac_addr[0] << 8) | mac_addr[1];
-		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH, val);
+		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + (BP_E1HVN(bp) + 1)*8, val);
 
 		val = (mac_addr[2] << 24) | (mac_addr[3] << 16) |
 		      (mac_addr[4] << 8) | mac_addr[5];
-		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + 4, val);
+		EMAC_WR(EMAC_REG_EMAC_MAC_MATCH + (BP_E1HVN(bp) + 1)*8 + 4,
+			val);
 
 		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_EN;
 
 	} else
 		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
 
-	/* Close multi and leading connections */
+	/* Close multi and leading connections
+	   Completions for ramrods are collected in a synchronous way */
 	for_each_nondefault_queue(bp, i)
 		if (bnx2x_stop_multi(bp, i))
 			goto unload_error;
 
-	bnx2x_stop_leading(bp);
+	if (CHIP_IS_E1H(bp))
+		REG_WR(bp, NIG_REG_LLH0_FUNC_EN + BP_PORT(bp)*8, 0);
+
+	bnx2x_stop_leading(bp);
+#ifdef BNX2X_STOP_ON_ERROR
+	/* If ramrod completion timed out - break here! */
+	if (bp->panic) {
+		BNX2X_ERR("Stop leading failed!\n");
+		return -EBUSY;
+	}
+#endif
+
 	if ((bp->state != BNX2X_STATE_CLOSING_WAIT4_UNLOAD) ||
 	    (bp->fp[0].state != BNX2X_FP_STATE_CLOSED)) {
-		DP(NETIF_MSG_IFDOWN, "failed to close leading properly!"
-		   "state 0x%x  fp[0].state 0x%x",
+		DP(NETIF_MSG_IFDOWN, "failed to close leading properly!  "
+		   "state 0x%x  fp[0].state 0x%x\n",
 		   bp->state, bp->fp[0].state);
 	}
 
 unload_error:
-	bnx2x__link_reset(bp);
-
-	if (!nomcp)
+	if (!BP_NOMCP(bp))
 		reset_code = bnx2x_fw_command(bp, reset_code);
-	else
-		reset_code = FW_MSG_CODE_DRV_UNLOAD_COMMON;
+	else {
+		DP(NETIF_MSG_IFDOWN, "NO MCP load counts      %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		load_count[0]--;
+		load_count[1 + BP_PORT(bp)]--;
+		DP(NETIF_MSG_IFDOWN, "NO MCP new load counts  %d, %d, %d\n",
+		   load_count[0], load_count[1], load_count[2]);
+		if (load_count[0] == 0)
+			reset_code = FW_MSG_CODE_DRV_UNLOAD_COMMON;
+		else if (load_count[1 + BP_PORT(bp)] == 0)
+			reset_code = FW_MSG_CODE_DRV_UNLOAD_PORT;
+		else
+			reset_code = FW_MSG_CODE_DRV_UNLOAD_FUNCTION;
+	}
 
-	/* Release IRQs */
-	if (free_irq)
-		bnx2x_free_irq(bp);
+	if ((reset_code == FW_MSG_CODE_DRV_UNLOAD_COMMON) ||
+	    (reset_code == FW_MSG_CODE_DRV_UNLOAD_PORT))
+		bnx2x__link_reset(bp);
 
 	/* Reset the chip */
 	bnx2x_reset_chip(bp, reset_code);
 
 	/* Report UNLOAD_DONE to MCP */
-	if (!nomcp)
+	if (!BP_NOMCP(bp))
 		bnx2x_fw_command(bp, DRV_MSG_CODE_UNLOAD_DONE);
 
 	/* Free SKBs and driver internals */
@@ -5008,6 +5728,29 @@ unload_error:
 	return 0;
 }
 
+static void bnx2x_reset_task(struct work_struct *work)
+{
+	struct bnx2x *bp = container_of(work, struct bnx2x, reset_task);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	BNX2X_ERR("reset task called but STOP_ON_ERROR defined"
+		  " so reset not done to allow debug dump,\n"
+	 KERN_ERR " you will need to reboot when done\n");
+	return;
+#endif
+
+	rtnl_lock();
+
+	if (!netif_running(bp->dev))
+		goto reset_task_exit;
+
+	bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+	bnx2x_nic_load(bp, LOAD_NORMAL);
+
+reset_task_exit:
+	rtnl_unlock();
+}
+
 /* end of nic load/unload */
 
 /* ethtool_ops */
@@ -5016,9 +5759,139 @@ unload_error:
  * Init service functions
  */
 
-static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
+static void __devinit bnx2x_undi_unload(struct bnx2x *bp)
+{
+	u32 val;
+
+	/* Check if there is any driver already loaded */
+	val = REG_RD(bp, MISC_REG_UNPREPARED);
+	if (val == 0x1) {
+		/* Check if it is the UNDI driver
+		 * UNDI driver initializes CID offset for normal bell to 0x7
+		 */
+		val = REG_RD(bp, DORQ_REG_NORM_CID_OFST);
+		if (val == 0x7) {
+			u32 reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
+			/* save our func and fw_seq */
+			int func = BP_FUNC(bp);
+			u16 fw_seq = bp->fw_seq;
+
+			BNX2X_DEV_INFO("UNDI is active! reset device\n");
+
+			/* try unload UNDI on port 0 */
+			bp->func = 0;
+			bp->fw_seq = (SHMEM_RD(bp,
+					     func_mb[bp->func].drv_mb_header) &
+				      DRV_MSG_SEQ_NUMBER_MASK);
+
+			reset_code = bnx2x_fw_command(bp, reset_code);
+			bnx2x_fw_command(bp, DRV_MSG_CODE_UNLOAD_DONE);
+
+			/* if UNDI is loaded on the other port */
+			if (reset_code != FW_MSG_CODE_DRV_UNLOAD_COMMON) {
+
+				bp->func = 1;
+				bp->fw_seq = (SHMEM_RD(bp,
+					     func_mb[bp->func].drv_mb_header) &
+					      DRV_MSG_SEQ_NUMBER_MASK);
+
+				bnx2x_fw_command(bp,
+					     DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS);
+				bnx2x_fw_command(bp,
+						 DRV_MSG_CODE_UNLOAD_DONE);
+
+				/* restore our func and fw_seq */
+				bp->func = func;
+				bp->fw_seq = fw_seq;
+			}
+
+			/* reset device */
+			REG_WR(bp,
+			       GRCBASE_MISC + MISC_REGISTERS_RESET_REG_1_CLEAR,
+			       0xd3ffff7f);
+			REG_WR(bp,
+			       GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR,
+			       0x1403);
+		}
+	}
+}
+
+static void __devinit bnx2x_get_common_hwinfo(struct bnx2x *bp)
+{
+	u32 val, val2, val3, val4, id;
+
+	/* Get the chip revision id and number. */
+	/* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
+	val = REG_RD(bp, MISC_REG_CHIP_NUM);
+	id = ((val & 0xffff) << 16);
+	val = REG_RD(bp, MISC_REG_CHIP_REV);
+	id |= ((val & 0xf) << 12);
+	val = REG_RD(bp, MISC_REG_CHIP_METAL);
+	id |= ((val & 0xff) << 4);
+	REG_RD(bp, MISC_REG_BOND_ID);
+	id |= (val & 0xf);
+	bp->common.chip_id = id;
+	bp->link_params.chip_id = bp->common.chip_id;
+	BNX2X_DEV_INFO("chip ID is 0x%x\n", id);
+
+	val = REG_RD(bp, MCP_REG_MCPR_NVM_CFG4);
+	bp->common.flash_size = (NVRAM_1MB_SIZE <<
+				 (val & MCPR_NVM_CFG4_FLASH_SIZE));
+	BNX2X_DEV_INFO("flash_size 0x%x (%d)\n",
+		       bp->common.flash_size, bp->common.flash_size);
+
+	bp->common.shmem_base = REG_RD(bp, MISC_REG_SHARED_MEM_ADDR);
+	bp->link_params.shmem_base = bp->common.shmem_base;
+	BNX2X_DEV_INFO("shmem offset is 0x%x\n", bp->common.shmem_base);
+
+	if (!bp->common.shmem_base ||
+	    (bp->common.shmem_base < 0xA0000) ||
+	    (bp->common.shmem_base >= 0xC0000)) {
+		BNX2X_DEV_INFO("MCP not active\n");
+		bp->flags |= NO_MCP_FLAG;
+		return;
+	}
+
+	val = SHMEM_RD(bp, validity_map[BP_PORT(bp)]);
+	if ((val & (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
+		!= (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
+		BNX2X_ERR("BAD MCP validity signature\n");
+
+	bp->common.hw_config = SHMEM_RD(bp, dev_info.shared_hw_config.config);
+	bp->common.board = SHMEM_RD(bp, dev_info.shared_hw_config.board);
+
+	BNX2X_DEV_INFO("hw_config 0x%08x  board 0x%08x\n",
+		       bp->common.hw_config, bp->common.board);
+
+	bp->link_params.hw_led_mode = ((bp->common.hw_config &
+					SHARED_HW_CFG_LED_MODE_MASK) >>
+				       SHARED_HW_CFG_LED_MODE_SHIFT);
+
+	val = SHMEM_RD(bp, dev_info.bc_rev) >> 8;
+	bp->common.bc_ver = val;
+	BNX2X_DEV_INFO("bc_ver %X\n", val);
+	if (val < BNX2X_BC_VER) {
+		/* for now only warn
+		 * later we might need to enforce this */
+		BNX2X_ERR("This driver needs bc_ver %X but found %X,"
+			  " please upgrade BC\n", BNX2X_BC_VER, val);
+	}
+	BNX2X_DEV_INFO("%sWoL Capable\n",
+		       (bp->flags & NO_WOL_FLAG)? "Not " : "");
+
+	val = SHMEM_RD(bp, dev_info.shared_hw_config.part_num);
+	val2 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[4]);
+	val3 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[8]);
+	val4 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[12]);
+
+	printk(KERN_INFO PFX "part number %X-%X-%X-%X\n",
+	       val, val2, val3, val4);
+}
+
+static void __devinit bnx2x_link_settings_supported(struct bnx2x *bp,
+						    u32 switch_cfg)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	u32 ext_phy_type;
 
 	switch (switch_cfg) {
@@ -5032,31 +5905,33 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (Direct)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10baseT_Half |
-					  SUPPORTED_10baseT_Full |
-					  SUPPORTED_100baseT_Half |
-					  SUPPORTED_100baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_2500baseX_Full |
-					  SUPPORTED_TP | SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10baseT_Half |
+					       SUPPORTED_10baseT_Full |
+					       SUPPORTED_100baseT_Half |
+					       SUPPORTED_100baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_2500baseX_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_SERDES_EXT_PHY_TYPE_BCM5482:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (5482)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10baseT_Half |
-					  SUPPORTED_10baseT_Full |
-					  SUPPORTED_100baseT_Half |
-					  SUPPORTED_100baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_TP | SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10baseT_Half |
+					       SUPPORTED_10baseT_Full |
+					       SUPPORTED_100baseT_Half |
+					       SUPPORTED_100baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		default:
@@ -5066,9 +5941,9 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			return;
 		}
 
-		bp->phy_addr = REG_RD(bp, NIG_REG_SERDES0_CTRL_PHY_ADDR +
-				      port*0x10);
-		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->phy_addr);
+		bp->port.phy_addr = REG_RD(bp, NIG_REG_SERDES0_CTRL_PHY_ADDR +
+					   port*0x10);
+		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->port.phy_addr);
 		break;
 
 	case SWITCH_CFG_10G:
@@ -5081,75 +5956,75 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (Direct)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10baseT_Half |
-					  SUPPORTED_10baseT_Full |
-					  SUPPORTED_100baseT_Half |
-					  SUPPORTED_100baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_2500baseX_Full |
-					  SUPPORTED_10000baseT_Full |
-					  SUPPORTED_TP | SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10baseT_Half |
+					       SUPPORTED_10baseT_Full |
+					       SUPPORTED_100baseT_Half |
+					       SUPPORTED_100baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_2500baseX_Full |
+					       SUPPORTED_10000baseT_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8705:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8705)\n",
-					ext_phy_type);
+				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8706)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8072:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8072)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8073:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (8073)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_2500baseX_Full |
-					  SUPPORTED_1000baseT_Full |
-					  SUPPORTED_FIBRE |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_2500baseX_Full |
+					       SUPPORTED_1000baseT_Full |
+					       SUPPORTED_FIBRE |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_SFX7101:
 			BNX2X_DEV_INFO("ext_phy_type 0x%x (SFX7101)\n",
 				       ext_phy_type);
 
-			bp->supported |= (SUPPORTED_10000baseT_Full |
-					  SUPPORTED_TP |
-					  SUPPORTED_Autoneg |
-					  SUPPORTED_Pause |
-					  SUPPORTED_Asym_Pause);
+			bp->port.supported |= (SUPPORTED_10000baseT_Full |
+					       SUPPORTED_TP |
+					       SUPPORTED_Autoneg |
+					       SUPPORTED_Pause |
+					       SUPPORTED_Asym_Pause);
 			break;
 
 		case PORT_HW_CFG_XGXS_EXT_PHY_TYPE_FAILURE:
@@ -5164,61 +6039,61 @@ static void bnx2x_link_settings_supported(struct bnx2x *bp, u32 switch_cfg)
 			return;
 		}
 
-		bp->phy_addr = REG_RD(bp, NIG_REG_XGXS0_CTRL_PHY_ADDR +
-				      port*0x18);
-		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->phy_addr);
+		bp->port.phy_addr = REG_RD(bp, NIG_REG_XGXS0_CTRL_PHY_ADDR +
+					   port*0x18);
+		BNX2X_DEV_INFO("phy_addr 0x%x\n", bp->port.phy_addr);
 
 		break;
 
 	default:
 		BNX2X_ERR("BAD switch_cfg link_config 0x%x\n",
-			  bp->link_config);
+			  bp->port.link_config);
 		return;
 	}
-	bp->link_params.phy_addr = bp->phy_addr;
+	bp->link_params.phy_addr = bp->port.phy_addr;
 
 	/* mask what we support according to speed_cap_mask */
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_HALF))
-		bp->supported &= ~SUPPORTED_10baseT_Half;
+		bp->port.supported &= ~SUPPORTED_10baseT_Half;
 
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_10M_FULL))
-		bp->supported &= ~SUPPORTED_10baseT_Full;
+		bp->port.supported &= ~SUPPORTED_10baseT_Full;
 
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_HALF))
-		bp->supported &= ~SUPPORTED_100baseT_Half;
+		bp->port.supported &= ~SUPPORTED_100baseT_Half;
 
 	if (!(bp->link_params.speed_cap_mask &
 				PORT_HW_CFG_SPEED_CAPABILITY_D0_100M_FULL))
-		bp->supported &= ~SUPPORTED_100baseT_Full;
+		bp->port.supported &= ~SUPPORTED_100baseT_Full;
 
 	if (!(bp->link_params.speed_cap_mask &
 					PORT_HW_CFG_SPEED_CAPABILITY_D0_1G))
-		bp->supported &= ~(SUPPORTED_1000baseT_Half |
-				   SUPPORTED_1000baseT_Full);
+		bp->port.supported &= ~(SUPPORTED_1000baseT_Half |
+					SUPPORTED_1000baseT_Full);
 
 	if (!(bp->link_params.speed_cap_mask &
 					PORT_HW_CFG_SPEED_CAPABILITY_D0_2_5G))
-		bp->supported &= ~SUPPORTED_2500baseX_Full;
+		bp->port.supported &= ~SUPPORTED_2500baseX_Full;
 
 	if (!(bp->link_params.speed_cap_mask &
 					PORT_HW_CFG_SPEED_CAPABILITY_D0_10G))
-		bp->supported &= ~SUPPORTED_10000baseT_Full;
+		bp->port.supported &= ~SUPPORTED_10000baseT_Full;
 
-	BNX2X_DEV_INFO("supported 0x%x\n", bp->supported);
+	BNX2X_DEV_INFO("supported 0x%x\n", bp->port.supported);
 }
 
-static void bnx2x_link_settings_requested(struct bnx2x *bp)
+static void __devinit bnx2x_link_settings_requested(struct bnx2x *bp)
 {
 	bp->link_params.req_duplex = DUPLEX_FULL;
 
-	switch (bp->link_config & PORT_FEATURE_LINK_SPEED_MASK) {
+	switch (bp->port.link_config & PORT_FEATURE_LINK_SPEED_MASK) {
 	case PORT_FEATURE_LINK_SPEED_AUTO:
-		if (bp->supported & SUPPORTED_Autoneg) {
+		if (bp->port.supported & SUPPORTED_Autoneg) {
 			bp->link_params.req_line_speed = SPEED_AUTO_NEG;
-			bp->advertising = bp->supported;
+			bp->port.advertising = bp->port.supported;
 		} else {
 			u32 ext_phy_type =
 			    XGXS_EXT_PHY_TYPE(bp->link_params.ext_phy_config);
@@ -5229,7 +6104,7 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 			     PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706)) {
 				/* force 10G, no AN */
 				bp->link_params.req_line_speed = SPEED_10000;
-				bp->advertising =
+				bp->port.advertising =
 						(ADVERTISED_10000baseT_Full |
 						 ADVERTISED_FIBRE);
 				break;
@@ -5237,98 +6112,98 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  Autoneg not supported\n",
-				  bp->link_config);
+				  bp->port.link_config);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_10M_FULL:
-		if (bp->supported & SUPPORTED_10baseT_Full) {
+		if (bp->port.supported & SUPPORTED_10baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_10;
-			bp->advertising = (ADVERTISED_10baseT_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_10baseT_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_10M_HALF:
-		if (bp->supported & SUPPORTED_10baseT_Half) {
+		if (bp->port.supported & SUPPORTED_10baseT_Half) {
 			bp->link_params.req_line_speed = SPEED_10;
 			bp->link_params.req_duplex = DUPLEX_HALF;
-			bp->advertising = (ADVERTISED_10baseT_Half |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_10baseT_Half |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_100M_FULL:
-		if (bp->supported & SUPPORTED_100baseT_Full) {
+		if (bp->port.supported & SUPPORTED_100baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_100;
-			bp->advertising = (ADVERTISED_100baseT_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_100baseT_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_100M_HALF:
-		if (bp->supported & SUPPORTED_100baseT_Half) {
+		if (bp->port.supported & SUPPORTED_100baseT_Half) {
 			bp->link_params.req_line_speed = SPEED_100;
 			bp->link_params.req_duplex = DUPLEX_HALF;
-			bp->advertising = (ADVERTISED_100baseT_Half |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_100baseT_Half |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_1G:
-		if (bp->supported & SUPPORTED_1000baseT_Full) {
+		if (bp->port.supported & SUPPORTED_1000baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_1000;
-			bp->advertising = (ADVERTISED_1000baseT_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_1000baseT_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
 		break;
 
 	case PORT_FEATURE_LINK_SPEED_2_5G:
-		if (bp->supported & SUPPORTED_2500baseX_Full) {
+		if (bp->port.supported & SUPPORTED_2500baseX_Full) {
 			bp->link_params.req_line_speed = SPEED_2500;
-			bp->advertising = (ADVERTISED_2500baseX_Full |
-					   ADVERTISED_TP);
+			bp->port.advertising = (ADVERTISED_2500baseX_Full |
+						ADVERTISED_TP);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
@@ -5337,15 +6212,15 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 	case PORT_FEATURE_LINK_SPEED_10G_CX4:
 	case PORT_FEATURE_LINK_SPEED_10G_KX4:
 	case PORT_FEATURE_LINK_SPEED_10G_KR:
-		if (bp->supported & SUPPORTED_10000baseT_Full) {
+		if (bp->port.supported & SUPPORTED_10000baseT_Full) {
 			bp->link_params.req_line_speed = SPEED_10000;
-			bp->advertising = (ADVERTISED_10000baseT_Full |
-					   ADVERTISED_FIBRE);
+			bp->port.advertising = (ADVERTISED_10000baseT_Full |
+						ADVERTISED_FIBRE);
 		} else {
 			BNX2X_ERR("NVRAM config error. "
 				  "Invalid link_config 0x%x"
 				  "  speed_cap_mask 0x%x\n",
-				  bp->link_config,
+				  bp->port.link_config,
 				  bp->link_params.speed_cap_mask);
 			return;
 		}
@@ -5354,64 +6229,33 @@ static void bnx2x_link_settings_requested(struct bnx2x *bp)
 	default:
 		BNX2X_ERR("NVRAM config error. "
 			  "BAD link speed link_config 0x%x\n",
-			  bp->link_config);
+			  bp->port.link_config);
 		bp->link_params.req_line_speed = SPEED_AUTO_NEG;
-		bp->advertising = bp->supported;
+		bp->port.advertising = bp->port.supported;
 		break;
 	}
 
-	bp->link_params.req_flow_ctrl = (bp->link_config &
-			     PORT_FEATURE_FLOW_CONTROL_MASK);
+	bp->link_params.req_flow_ctrl = (bp->port.link_config &
+					 PORT_FEATURE_FLOW_CONTROL_MASK);
 	if ((bp->link_params.req_flow_ctrl == FLOW_CTRL_AUTO) &&
-	    (!bp->supported & SUPPORTED_Autoneg))
+	    (!bp->port.supported & SUPPORTED_Autoneg))
 		bp->link_params.req_flow_ctrl = FLOW_CTRL_NONE;
 
 	BNX2X_DEV_INFO("req_line_speed %d  req_duplex %d  req_flow_ctrl 0x%x"
 		       "  advertising 0x%x\n",
 		       bp->link_params.req_line_speed,
 		       bp->link_params.req_duplex,
-		       bp->link_params.req_flow_ctrl, bp->advertising);
+		       bp->link_params.req_flow_ctrl, bp->port.advertising);
 }
 
-static void bnx2x_get_hwinfo(struct bnx2x *bp)
+static void __devinit bnx2x_get_port_hwinfo(struct bnx2x *bp)
 {
-	u32 val, val2, val3, val4, id;
-	int port = bp->port;
-
-	bp->shmem_base = REG_RD(bp, MISC_REG_SHARED_MEM_ADDR);
-	BNX2X_DEV_INFO("shmem offset is %x\n", bp->shmem_base);
-
-	/* Get the chip revision id and number. */
-	/* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */
-	val = REG_RD(bp, MISC_REG_CHIP_NUM);
-	id = ((val & 0xffff) << 16);
-	val = REG_RD(bp, MISC_REG_CHIP_REV);
-	id |= ((val & 0xf) << 12);
-	val = REG_RD(bp, MISC_REG_CHIP_METAL);
-	id |= ((val & 0xff) << 4);
-	REG_RD(bp, MISC_REG_BOND_ID);
-	id |= (val & 0xf);
-	bp->chip_id = id;
-	BNX2X_DEV_INFO("chip ID is %x\n", id);
+	int port = BP_PORT(bp);
+	u32 val, val2;
 
 	bp->link_params.bp = bp;
+	bp->link_params.port = port;
 
-	if (!bp->shmem_base || (bp->shmem_base != 0xAF900)) {
-		BNX2X_DEV_INFO("MCP not active\n");
-		nomcp = 1;
-		goto set_mac;
-	}
-
-	val = SHMEM_RD(bp, validity_map[port]);
-	if ((val & (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
-		!= (SHR_MEM_VALIDITY_DEV_INFO | SHR_MEM_VALIDITY_MB))
-		BNX2X_ERR("BAD MCP validity signature\n");
-
-	bp->fw_seq = (SHMEM_RD(bp, func_mb[port].drv_mb_header) &
-		      DRV_MSG_SEQ_NUMBER_MASK);
-
-	bp->hw_config = SHMEM_RD(bp, dev_info.shared_hw_config.config);
-	bp->board = SHMEM_RD(bp, dev_info.shared_hw_config.board);
 	bp->link_params.serdes_config =
 		SHMEM_RD(bp, dev_info.port_hw_config[port].serdes_config);
 	bp->link_params.lane_config =
@@ -5423,19 +6267,18 @@ static void bnx2x_get_hwinfo(struct bnx2x *bp)
 		SHMEM_RD(bp,
 			 dev_info.port_hw_config[port].speed_capability_mask);
 
-	bp->link_config =
+	bp->port.link_config =
 		SHMEM_RD(bp, dev_info.port_feature_config[port].link_config);
 
-	BNX2X_DEV_INFO("serdes_config (%08x)  lane_config (%08x)\n"
-	     KERN_INFO "  ext_phy_config (%08x)  speed_cap_mask (%08x)"
-		       "  link_config (%08x)\n",
+	BNX2X_DEV_INFO("serdes_config 0x%08x  lane_config 0x%08x\n"
+	     KERN_INFO "  ext_phy_config 0x%08x  speed_cap_mask 0x%08x"
+		       "  link_config 0x%08x\n",
 		       bp->link_params.serdes_config,
 		       bp->link_params.lane_config,
 		       bp->link_params.ext_phy_config,
-		       bp->link_params.speed_cap_mask,
-		       bp->link_config);
+		       bp->link_params.speed_cap_mask, bp->port.link_config);
 
-	bp->link_params.switch_cfg = (bp->link_config &
+	bp->link_params.switch_cfg = (bp->port.link_config &
 				      PORT_FEATURE_CONNECTED_SWITCH_MASK);
 	bnx2x_link_settings_supported(bp, bp->link_params.switch_cfg);
 
@@ -5451,43 +6294,126 @@ static void bnx2x_get_hwinfo(struct bnx2x *bp)
 	bp->dev->dev_addr[5] = (u8)(val & 0xff);
 	memcpy(bp->link_params.mac_addr, bp->dev->dev_addr, ETH_ALEN);
 	memcpy(bp->dev->perm_addr, bp->dev->dev_addr, ETH_ALEN);
+}
+
+static int __devinit bnx2x_get_hwinfo(struct bnx2x *bp)
+{
+	int func = BP_FUNC(bp);
+	u32 val, val2;
+	int rc = 0;
 
+	bnx2x_get_common_hwinfo(bp);
 
+	bp->e1hov = 0;
+	bp->e1hmf = 0;
+	if (CHIP_IS_E1H(bp)) {
+		bp->mf_config =
+			SHMEM_RD(bp, mf_cfg.func_mf_config[func].config);
 
-	val = SHMEM_RD(bp, dev_info.shared_hw_config.part_num);
-	val2 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[4]);
-	val3 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[8]);
-	val4 = SHMEM_RD(bp, dev_info.shared_hw_config.part_num[12]);
+		val =
+		   (SHMEM_RD(bp, mf_cfg.func_mf_config[func].e1hov_tag) &
+		    FUNC_MF_CFG_E1HOV_TAG_MASK);
+		if (val != FUNC_MF_CFG_E1HOV_TAG_DEFAULT) {
 
-	printk(KERN_INFO PFX "part number %X-%X-%X-%X\n",
-	       val, val2, val3, val4);
+			bp->e1hov = val;
+			bp->e1hmf = 1;
+			BNX2X_DEV_INFO("MF mode  E1HOV for func %d is %d "
+				       "(0x%04x)\n",
+				       func, bp->e1hov, bp->e1hov);
+		} else {
+			BNX2X_DEV_INFO("Single function mode\n");
+			if (BP_E1HVN(bp)) {
+				BNX2X_ERR("!!!  No valid E1HOV for func %d,"
+					  "  aborting\n", func);
+				rc = -EPERM;
+			}
+		}
+	}
 
-	/* bc ver */
-	if (!nomcp) {
-		bp->bc_ver = val = ((SHMEM_RD(bp, dev_info.bc_rev)) >> 8);
-		BNX2X_DEV_INFO("bc_ver %X\n", val);
-		if (val < BNX2X_BC_VER) {
-			/* for now only warn
-			 * later we might need to enforce this */
-			BNX2X_ERR("This driver needs bc_ver %X but found %X,"
-				  " please upgrade BC\n", BNX2X_BC_VER, val);
+	if (!BP_NOMCP(bp)) {
+		bnx2x_get_port_hwinfo(bp);
+
+		bp->fw_seq = (SHMEM_RD(bp, func_mb[func].drv_mb_header) &
+			      DRV_MSG_SEQ_NUMBER_MASK);
+		BNX2X_DEV_INFO("fw_seq 0x%08x\n", bp->fw_seq);
+	}
+
+	if (IS_E1HMF(bp)) {
+		val2 = SHMEM_RD(bp, mf_cfg.func_mf_config[func].mac_upper);
+		val = SHMEM_RD(bp,  mf_cfg.func_mf_config[func].mac_lower);
+		if ((val2 != FUNC_MF_CFG_UPPERMAC_DEFAULT) &&
+		    (val != FUNC_MF_CFG_LOWERMAC_DEFAULT)) {
+			bp->dev->dev_addr[0] = (u8)(val2 >> 8 & 0xff);
+			bp->dev->dev_addr[1] = (u8)(val2 & 0xff);
+			bp->dev->dev_addr[2] = (u8)(val >> 24 & 0xff);
+			bp->dev->dev_addr[3] = (u8)(val >> 16 & 0xff);
+			bp->dev->dev_addr[4] = (u8)(val >> 8  & 0xff);
+			bp->dev->dev_addr[5] = (u8)(val & 0xff);
+			memcpy(bp->link_params.mac_addr, bp->dev->dev_addr,
+			       ETH_ALEN);
+			memcpy(bp->dev->perm_addr, bp->dev->dev_addr,
+			       ETH_ALEN);
 		}
-	} else {
-		bp->bc_ver = 0;
+
+		return rc;
 	}
 
-	val = REG_RD(bp, MCP_REG_MCPR_NVM_CFG4);
-	bp->flash_size = (NVRAM_1MB_SIZE << (val & MCPR_NVM_CFG4_FLASH_SIZE));
-	BNX2X_DEV_INFO("flash_size 0x%x (%d)\n",
-		       bp->flash_size, bp->flash_size);
+	if (BP_NOMCP(bp)) {
+		/* only supposed to happen on emulation/FPGA */
+		BNX2X_ERR("warning rendom MAC workaround active\n");
+		random_ether_addr(bp->dev->dev_addr);
+		memcpy(bp->dev->perm_addr, bp->dev->dev_addr, ETH_ALEN);
+	}
 
-	return;
+	return rc;
+}
+
+static int __devinit bnx2x_init_bp(struct bnx2x *bp)
+{
+	int func = BP_FUNC(bp);
+	int rc;
+
+	if (nomcp)
+		bp->flags |= NO_MCP_FLAG;
 
-set_mac: /* only supposed to happen on emulation/FPGA */
-	BNX2X_ERR("warning rendom MAC workaround active\n");
-	random_ether_addr(bp->dev->dev_addr);
-	memcpy(bp->dev->perm_addr, bp->dev->dev_addr, 6);
+	mutex_init(&bp->port.phy_mutex);
 
+	INIT_WORK(&bp->sp_task, bnx2x_sp_task);
+	INIT_WORK(&bp->reset_task, bnx2x_reset_task);
+
+	rc = bnx2x_get_hwinfo(bp);
+
+	/* need to reset chip if undi was active */
+	if (!BP_NOMCP(bp))
+		bnx2x_undi_unload(bp);
+
+	if (CHIP_REV_IS_FPGA(bp))
+		printk(KERN_ERR PFX "FPGA detected\n");
+
+	if (BP_NOMCP(bp) && (func == 0))
+		printk(KERN_ERR PFX
+		       "MCP disabled, must load devices in order!\n");
+
+	bp->tx_ring_size = MAX_TX_AVAIL;
+	bp->rx_ring_size = MAX_RX_AVAIL;
+
+	bp->rx_csum = 1;
+	bp->rx_offset = 0;
+
+	bp->tx_ticks = 50;
+	bp->rx_ticks = 25;
+
+	bp->stats_ticks = 1000000 & 0xffff00;
+
+	bp->timer_interval = (CHIP_REV_IS_SLOW(bp) ? 5*HZ : HZ);
+	bp->current_interval = (poll ? poll : bp->timer_interval);
+
+	init_timer(&bp->timer);
+	bp->timer.expires = jiffies + bp->current_interval;
+	bp->timer.data = (unsigned long) bp;
+	bp->timer.function = bnx2x_timer;
+
+	return rc;
 }
 
 /*
@@ -5500,8 +6426,8 @@ static int bnx2x_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
-	cmd->supported = bp->supported;
-	cmd->advertising = bp->advertising;
+	cmd->supported = bp->port.supported;
+	cmd->advertising = bp->port.advertising;
 
 	if (netif_carrier_ok(dev)) {
 		cmd->speed = bp->link_vars.line_speed;
@@ -5510,6 +6436,14 @@ static int bnx2x_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		cmd->speed = bp->link_params.req_line_speed;
 		cmd->duplex = bp->link_params.req_duplex;
 	}
+	if (IS_E1HMF(bp)) {
+		u16 vn_max_rate;
+
+		vn_max_rate = ((bp->mf_config & FUNC_MF_CFG_MAX_BW_MASK) >>
+				FUNC_MF_CFG_MAX_BW_SHIFT) * 100;
+		if (vn_max_rate < cmd->speed)
+			cmd->speed = vn_max_rate;
+	}
 
 	if (bp->link_params.switch_cfg == SWITCH_CFG_10G) {
 		u32 ext_phy_type =
@@ -5541,7 +6475,7 @@ static int bnx2x_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	} else
 		cmd->port = PORT_TP;
 
-	cmd->phy_address = bp->phy_addr;
+	cmd->phy_address = bp->port.phy_addr;
 	cmd->transceiver = XCVR_INTERNAL;
 
 	if (bp->link_params.req_line_speed == SPEED_AUTO_NEG)
@@ -5568,6 +6502,9 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	struct bnx2x *bp = netdev_priv(dev);
 	u32 advertising;
 
+	if (IS_E1HMF(bp))
+		return 0;
+
 	DP(NETIF_MSG_LINK, "ethtool_cmd: cmd %d\n"
 	   DP_LEVEL "  supported 0x%x  advertising 0x%x  speed %d\n"
 	   DP_LEVEL "  duplex %d  port %d  phy_address %d  transceiver %d\n"
@@ -5577,24 +6514,25 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	   cmd->autoneg, cmd->maxtxpkt, cmd->maxrxpkt);
 
 	if (cmd->autoneg == AUTONEG_ENABLE) {
-		if (!(bp->supported & SUPPORTED_Autoneg)) {
-			DP(NETIF_MSG_LINK, "Aotoneg not supported\n");
+		if (!(bp->port.supported & SUPPORTED_Autoneg)) {
+			DP(NETIF_MSG_LINK, "Autoneg not supported\n");
 			return -EINVAL;
 		}
 
 		/* advertise the requested speed and duplex if supported */
-		cmd->advertising &= bp->supported;
+		cmd->advertising &= bp->port.supported;
 
 		bp->link_params.req_line_speed = SPEED_AUTO_NEG;
 		bp->link_params.req_duplex = DUPLEX_FULL;
-		bp->advertising |= (ADVERTISED_Autoneg | cmd->advertising);
+		bp->port.advertising |= (ADVERTISED_Autoneg |
+					 cmd->advertising);
 
 	} else { /* forced speed */
 		/* advertise the requested speed and duplex if supported */
 		switch (cmd->speed) {
 		case SPEED_10:
 			if (cmd->duplex == DUPLEX_FULL) {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 				      SUPPORTED_10baseT_Full)) {
 					DP(NETIF_MSG_LINK,
 					   "10M full not supported\n");
@@ -5604,7 +6542,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				advertising = (ADVERTISED_10baseT_Full |
 					       ADVERTISED_TP);
 			} else {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 				      SUPPORTED_10baseT_Half)) {
 					DP(NETIF_MSG_LINK,
 					   "10M half not supported\n");
@@ -5618,7 +6556,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 
 		case SPEED_100:
 			if (cmd->duplex == DUPLEX_FULL) {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 						SUPPORTED_100baseT_Full)) {
 					DP(NETIF_MSG_LINK,
 					   "100M full not supported\n");
@@ -5628,7 +6566,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				advertising = (ADVERTISED_100baseT_Full |
 					       ADVERTISED_TP);
 			} else {
-				if (!(bp->supported &
+				if (!(bp->port.supported &
 						SUPPORTED_100baseT_Half)) {
 					DP(NETIF_MSG_LINK,
 					   "100M half not supported\n");
@@ -5646,7 +6584,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				return -EINVAL;
 			}
 
-			if (!(bp->supported & SUPPORTED_1000baseT_Full)) {
+			if (!(bp->port.supported & SUPPORTED_1000baseT_Full)) {
 				DP(NETIF_MSG_LINK, "1G full not supported\n");
 				return -EINVAL;
 			}
@@ -5662,7 +6600,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				return -EINVAL;
 			}
 
-			if (!(bp->supported & SUPPORTED_2500baseX_Full)) {
+			if (!(bp->port.supported & SUPPORTED_2500baseX_Full)) {
 				DP(NETIF_MSG_LINK,
 				   "2.5G full not supported\n");
 				return -EINVAL;
@@ -5678,7 +6616,7 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 				return -EINVAL;
 			}
 
-			if (!(bp->supported & SUPPORTED_10000baseT_Full)) {
+			if (!(bp->port.supported & SUPPORTED_10000baseT_Full)) {
 				DP(NETIF_MSG_LINK, "10G full not supported\n");
 				return -EINVAL;
 			}
@@ -5694,16 +6632,18 @@ static int bnx2x_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 
 		bp->link_params.req_line_speed = cmd->speed;
 		bp->link_params.req_duplex = cmd->duplex;
-		bp->advertising = advertising;
+		bp->port.advertising = advertising;
 	}
 
 	DP(NETIF_MSG_LINK, "req_line_speed %d\n"
 	   DP_LEVEL "  req_duplex %d  advertising 0x%x\n",
 	   bp->link_params.req_line_speed, bp->link_params.req_duplex,
-	   bp->advertising);
+	   bp->port.advertising);
 
-	bnx2x_stop_stats(bp);
-	bnx2x_link_set(bp);
+	if (netif_running(dev)) {
+		bnx2x_stop_stats(bp);
+		bnx2x_link_set(bp);
+	}
 
 	return 0;
 }
@@ -5720,21 +6660,23 @@ static void bnx2x_get_drvinfo(struct net_device *dev,
 	strcpy(info->version, DRV_MODULE_VERSION);
 
 	phy_fw_ver[0] = '\0';
-	bnx2x_phy_hw_lock(bp);
-	bnx2x_get_ext_phy_fw_version(&bp->link_params,
-				     (bp->state != BNX2X_STATE_CLOSED),
-				     phy_fw_ver, PHY_FW_VER_LEN);
-	bnx2x_phy_hw_unlock(bp);
+	if (bp->port.pmf) {
+		bnx2x_phy_hw_lock(bp);
+		bnx2x_get_ext_phy_fw_version(&bp->link_params,
+					     (bp->state != BNX2X_STATE_CLOSED),
+					     phy_fw_ver, PHY_FW_VER_LEN);
+		bnx2x_phy_hw_unlock(bp);
+	}
 
 	snprintf(info->fw_version, 32, "%d.%d.%d:%d BC:%x%s%s",
 		 BCM_5710_FW_MAJOR_VERSION, BCM_5710_FW_MINOR_VERSION,
 		 BCM_5710_FW_REVISION_VERSION,
-		 BCM_5710_FW_COMPILE_FLAGS, bp->bc_ver,
+		 BCM_5710_FW_COMPILE_FLAGS, bp->common.bc_ver,
 		 ((phy_fw_ver[0] != '\0')? " PHY:":""), phy_fw_ver);
 	strcpy(info->bus_info, pci_name(bp->pdev));
 	info->n_stats = BNX2X_NUM_STATS;
 	info->testinfo_len = BNX2X_NUM_TESTS;
-	info->eedump_len = bp->flash_size;
+	info->eedump_len = bp->common.flash_size;
 	info->regdump_len = 0;
 }
 
@@ -5767,9 +6709,9 @@ static int bnx2x_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
 			return -EINVAL;
 
 		bp->wol = 1;
-	} else {
+	} else
 		bp->wol = 0;
-	}
+
 	return 0;
 }
 
@@ -5792,13 +6734,13 @@ static int bnx2x_nway_reset(struct net_device *dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
-	if (bp->state != BNX2X_STATE_OPEN) {
-		DP(NETIF_MSG_PROBE, "state is %x, returning\n", bp->state);
-		return -EAGAIN;
-	}
+	if (!bp->port.pmf)
+		return 0;
 
-	bnx2x_stop_stats(bp);
-	bnx2x_link_set(bp);
+	if (netif_running(dev)) {
+		bnx2x_stop_stats(bp);
+		bnx2x_link_set(bp);
+	}
 
 	return 0;
 }
@@ -5807,12 +6749,12 @@ static int bnx2x_get_eeprom_len(struct net_device *dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
-	return bp->flash_size;
+	return bp->common.flash_size;
 }
 
 static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int count, i;
 	u32 val = 0;
 
@@ -5834,7 +6776,7 @@ static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
 	}
 
 	if (!(val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port))) {
-		DP(NETIF_MSG_NVM, "cannot get access to nvram interface\n");
+		DP(BNX2X_MSG_NVM, "cannot get access to nvram interface\n");
 		return -EBUSY;
 	}
 
@@ -5843,7 +6785,7 @@ static int bnx2x_acquire_nvram_lock(struct bnx2x *bp)
 
 static int bnx2x_release_nvram_lock(struct bnx2x *bp)
 {
-	int port = bp->port;
+	int port = BP_PORT(bp);
 	int count, i;
 	u32 val = 0;
 
@@ -5865,7 +6807,7 @@ static int bnx2x_release_nvram_lock(struct bnx2x *bp)
 	}
 
 	if (val & (MCPR_NVM_SW_ARB_ARB_ARB1 << port)) {
-		DP(NETIF_MSG_NVM, "cannot free access to nvram interface\n");
+		DP(BNX2X_MSG_NVM, "cannot free access to nvram interface\n");
 		return -EBUSY;
 	}
 
@@ -5929,7 +6871,6 @@ static int bnx2x_nvram_read_dword(struct bnx2x *bp, u32 offset, u32 *ret_val,
 
 		if (val & MCPR_NVM_COMMAND_DONE) {
 			val = REG_RD(bp, MCP_REG_MCPR_NVM_READ);
-			DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
 			/* we read nvram data in cpu order
 			 * but ethtool sees it as an array of bytes
 			 * converting to big-endian will do the work */
@@ -5951,16 +6892,16 @@ static int bnx2x_nvram_read(struct bnx2x *bp, u32 offset, u8 *ret_buf,
 	u32 val;
 
 	if ((offset & 0x03) || (buf_size & 0x03) || (buf_size == 0)) {
-		DP(NETIF_MSG_NVM,
+		DP(BNX2X_MSG_NVM,
 		   "Invalid parameter: offset 0x%x  buf_size 0x%x\n",
 		   offset, buf_size);
 		return -EINVAL;
 	}
 
-	if (offset + buf_size > bp->flash_size) {
-		DP(NETIF_MSG_NVM, "Invalid parameter: offset (0x%x) +"
+	if (offset + buf_size > bp->common.flash_size) {
+		DP(BNX2X_MSG_NVM, "Invalid parameter: offset (0x%x) +"
 				  " buf_size (0x%x) > flash_size (0x%x)\n",
-		   offset, buf_size, bp->flash_size);
+		   offset, buf_size, bp->common.flash_size);
 		return -EINVAL;
 	}
 
@@ -6004,7 +6945,7 @@ static int bnx2x_get_eeprom(struct net_device *dev,
 	struct bnx2x *bp = netdev_priv(dev);
 	int rc;
 
-	DP(NETIF_MSG_NVM, "ethtool_eeprom: cmd %d\n"
+	DP(BNX2X_MSG_NVM, "ethtool_eeprom: cmd %d\n"
 	   DP_LEVEL "  magic 0x%x  offset 0x%x (%d)  len 0x%x (%d)\n",
 	   eeprom->cmd, eeprom->magic, eeprom->offset, eeprom->offset,
 	   eeprom->len, eeprom->len);
@@ -6066,10 +7007,10 @@ static int bnx2x_nvram_write1(struct bnx2x *bp, u32 offset, u8 *data_buf,
 	u32 align_offset;
 	u32 val;
 
-	if (offset + buf_size > bp->flash_size) {
-		DP(NETIF_MSG_NVM, "Invalid parameter: offset (0x%x) +"
+	if (offset + buf_size > bp->common.flash_size) {
+		DP(BNX2X_MSG_NVM, "Invalid parameter: offset (0x%x) +"
 				  " buf_size (0x%x) > flash_size (0x%x)\n",
-		   offset, buf_size, bp->flash_size);
+		   offset, buf_size, bp->common.flash_size);
 		return -EINVAL;
 	}
 
@@ -6093,8 +7034,6 @@ static int bnx2x_nvram_write1(struct bnx2x *bp, u32 offset, u8 *data_buf,
 		 * convert it back to cpu order */
 		val = be32_to_cpu(val);
 
-		DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
-
 		rc = bnx2x_nvram_write_dword(bp, align_offset, val,
 					     cmd_flags);
 	}
@@ -6114,21 +7053,20 @@ static int bnx2x_nvram_write(struct bnx2x *bp, u32 offset, u8 *data_buf,
 	u32 val;
 	u32 written_so_far;
 
-	if (buf_size == 1) {    /* ethtool */
+	if (buf_size == 1)	/* ethtool */
 		return bnx2x_nvram_write1(bp, offset, data_buf, buf_size);
-	}
 
 	if ((offset & 0x03) || (buf_size & 0x03) || (buf_size == 0)) {
-		DP(NETIF_MSG_NVM,
+		DP(BNX2X_MSG_NVM,
 		   "Invalid parameter: offset 0x%x  buf_size 0x%x\n",
 		   offset, buf_size);
 		return -EINVAL;
 	}
 
-	if (offset + buf_size > bp->flash_size) {
-		DP(NETIF_MSG_NVM, "Invalid parameter: offset (0x%x) +"
+	if (offset + buf_size > bp->common.flash_size) {
+		DP(BNX2X_MSG_NVM, "Invalid parameter: offset (0x%x) +"
 				  " buf_size (0x%x) > flash_size (0x%x)\n",
-		   offset, buf_size, bp->flash_size);
+		   offset, buf_size, bp->common.flash_size);
 		return -EINVAL;
 	}
 
@@ -6151,7 +7089,6 @@ static int bnx2x_nvram_write(struct bnx2x *bp, u32 offset, u8 *data_buf,
 			cmd_flags |= MCPR_NVM_COMMAND_FIRST;
 
 		memcpy(&val, data_buf, 4);
-		DP(NETIF_MSG_NVM, "val 0x%08x\n", val);
 
 		rc = bnx2x_nvram_write_dword(bp, offset, val, cmd_flags);
 
@@ -6175,7 +7112,7 @@ static int bnx2x_set_eeprom(struct net_device *dev,
 	struct bnx2x *bp = netdev_priv(dev);
 	int rc;
 
-	DP(NETIF_MSG_NVM, "ethtool_eeprom: cmd %d\n"
+	DP(BNX2X_MSG_NVM, "ethtool_eeprom: cmd %d\n"
 	   DP_LEVEL "  magic 0x%x  offset 0x%x (%d)  len 0x%x (%d)\n",
 	   eeprom->cmd, eeprom->magic, eeprom->offset, eeprom->offset,
 	   eeprom->len, eeprom->len);
@@ -6183,20 +7120,23 @@ static int bnx2x_set_eeprom(struct net_device *dev,
 	/* parameters already validated in ethtool_set_eeprom */
 
 	/* If the magic number is PHY (0x00504859) upgrade the PHY FW */
-	if (eeprom->magic == 0x00504859) {
-
-		bnx2x_phy_hw_lock(bp);
-		rc = bnx2x_flash_download(bp, bp->port,
-				     bp->link_params.ext_phy_config,
-				     (bp->state != BNX2X_STATE_CLOSED),
-				     eebuf, eeprom->len);
-		rc |= bnx2x_link_reset(&bp->link_params,
-				       &bp->link_vars);
-		rc |= bnx2x_phy_init(&bp->link_params,
-				     &bp->link_vars);
-		bnx2x_phy_hw_unlock(bp);
-
-	} else
+	if (eeprom->magic == 0x00504859)
+		if (bp->port.pmf) {
+
+			bnx2x_phy_hw_lock(bp);
+			rc = bnx2x_flash_download(bp, BP_PORT(bp),
+					     bp->link_params.ext_phy_config,
+					     (bp->state != BNX2X_STATE_CLOSED),
+					     eebuf, eeprom->len);
+				rc |= bnx2x_link_reset(&bp->link_params,
+						       &bp->link_vars);
+				rc |= bnx2x_phy_init(&bp->link_params,
+						     &bp->link_vars);
+			bnx2x_phy_hw_unlock(bp);
+
+		} else /* Only the PMF can access the PHY */
+			return -EINVAL;
+	else
 		rc = bnx2x_nvram_write(bp, eeprom->offset, eebuf, eeprom->len);
 
 	return rc;
@@ -6234,7 +7174,7 @@ static int bnx2x_set_coalesce(struct net_device *dev,
 		bp->stats_ticks = 0xffff00;
 	bp->stats_ticks &= 0xffff00;
 
-	if (netif_running(bp->dev))
+	if (netif_running(dev))
 		bnx2x_update_coalesce(bp);
 
 	return 0;
@@ -6261,6 +7201,7 @@ static int bnx2x_set_ringparam(struct net_device *dev,
 			       struct ethtool_ringparam *ering)
 {
 	struct bnx2x *bp = netdev_priv(dev);
+	int rc = 0;
 
 	if ((ering->rx_pending > MAX_RX_AVAIL) ||
 	    (ering->tx_pending > MAX_TX_AVAIL) ||
@@ -6270,12 +7211,12 @@ static int bnx2x_set_ringparam(struct net_device *dev,
 	bp->rx_ring_size = ering->rx_pending;
 	bp->tx_ring_size = ering->tx_pending;
 
-	if (netif_running(bp->dev)) {
-		bnx2x_nic_unload(bp, 0);
-		bnx2x_nic_load(bp, 0);
+	if (netif_running(dev)) {
+		bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+		rc = bnx2x_nic_load(bp, LOAD_NORMAL);
 	}
 
-	return 0;
+	return rc;
 }
 
 static void bnx2x_get_pauseparam(struct net_device *dev,
@@ -6301,6 +7242,9 @@ static int bnx2x_set_pauseparam(struct net_device *dev,
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
+	if (IS_E1HMF(bp))
+		return 0;
+
 	DP(NETIF_MSG_LINK, "ethtool_pauseparam: cmd %d\n"
 	   DP_LEVEL "  autoneg %d  rx_pause %d  tx_pause %d\n",
 	   epause->cmd, epause->autoneg, epause->rx_pause, epause->tx_pause);
@@ -6317,7 +7261,7 @@ static int bnx2x_set_pauseparam(struct net_device *dev,
 		bp->link_params.req_flow_ctrl = FLOW_CTRL_NONE;
 
 	if (epause->autoneg) {
-		if (!(bp->supported & SUPPORTED_Autoneg)) {
+		if (!(bp->port.supported & SUPPORTED_Autoneg)) {
 			DP(NETIF_MSG_LINK, "Autoneg not supported\n");
 			return -EINVAL;
 		}
@@ -6328,8 +7272,11 @@ static int bnx2x_set_pauseparam(struct net_device *dev,
 
 	DP(NETIF_MSG_LINK,
 	   "req_flow_ctrl 0x%x\n", bp->link_params.req_flow_ctrl);
-	bnx2x_stop_stats(bp);
-	bnx2x_link_set(bp);
+
+	if (netif_running(dev)) {
+		bnx2x_stop_stats(bp);
+		bnx2x_link_set(bp);
+	}
 
 	return 0;
 }
@@ -6531,18 +7478,25 @@ static void bnx2x_get_ethtool_stats(struct net_device *dev,
 static int bnx2x_phys_id(struct net_device *dev, u32 data)
 {
 	struct bnx2x *bp = netdev_priv(dev);
+	int port = BP_PORT(bp);
 	int i;
 
+	if (!netif_running(dev))
+		return 0;
+
+	if (!bp->port.pmf)
+		return 0;
+
 	if (data == 0)
 		data = 2;
 
 	for (i = 0; i < (data * 2); i++) {
 		if ((i % 2) == 0)
-			bnx2x_set_led(bp, bp->port, LED_MODE_OPER, SPEED_1000,
+			bnx2x_set_led(bp, port, LED_MODE_OPER, SPEED_1000,
 				      bp->link_params.hw_led_mode,
 				      bp->link_params.chip_id);
 		else
-			bnx2x_set_led(bp, bp->port, LED_MODE_OFF, 0,
+			bnx2x_set_led(bp, port, LED_MODE_OFF, 0,
 				      bp->link_params.hw_led_mode,
 				      bp->link_params.chip_id);
 
@@ -6552,7 +7506,7 @@ static int bnx2x_phys_id(struct net_device *dev, u32 data)
 	}
 
 	if (bp->link_vars.link_up)
-		bnx2x_set_led(bp, bp->port, LED_MODE_OPER,
+		bnx2x_set_led(bp, port, LED_MODE_OPER,
 			      bp->link_vars.line_speed,
 			      bp->link_params.hw_led_mode,
 			      bp->link_params.chip_id);
@@ -6609,117 +7563,40 @@ static int bnx2x_set_power_state(struct bnx2x *bp, pci_power_t state)
 
 	switch (state) {
 	case PCI_D0:
-		pci_write_config_word(bp->pdev,
-				      bp->pm_cap + PCI_PM_CTRL,
+		pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL,
 				      ((pmcsr & ~PCI_PM_CTRL_STATE_MASK) |
 				       PCI_PM_CTRL_PME_STATUS));
 
 		if (pmcsr & PCI_PM_CTRL_STATE_MASK)
 		/* delay required during transition out of D3hot */
 			msleep(20);
-		break;
-
-	case PCI_D3hot:
-		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
-		pmcsr |= 3;
-
-		if (bp->wol)
-			pmcsr |= PCI_PM_CTRL_PME_ENABLE;
-
-		pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL,
-				      pmcsr);
-
-		/* No more memory access after this point until
-		* device is brought back to D0.
-		*/
-		break;
-
-	default:
-		return -EINVAL;
-	}
-	return 0;
-}
-
-/*
- * net_device service functions
- */
-
-/* called with netif_tx_lock from set_multicast */
-static void bnx2x_set_rx_mode(struct net_device *dev)
-{
-	struct bnx2x *bp = netdev_priv(dev);
-	u32 rx_mode = BNX2X_RX_MODE_NORMAL;
-
-	DP(NETIF_MSG_IFUP, "called dev->flags = %x\n", dev->flags);
-
-	if (dev->flags & IFF_PROMISC)
-		rx_mode = BNX2X_RX_MODE_PROMISC;
-
-	else if ((dev->flags & IFF_ALLMULTI) ||
-		 (dev->mc_count > BNX2X_MAX_MULTICAST))
-		rx_mode = BNX2X_RX_MODE_ALLMULTI;
-
-	else { /* some multicasts */
-		int i, old, offset;
-		struct dev_mc_list *mclist;
-		struct mac_configuration_cmd *config =
-						bnx2x_sp(bp, mcast_config);
-
-		for (i = 0, mclist = dev->mc_list;
-		     mclist && (i < dev->mc_count);
-		     i++, mclist = mclist->next) {
-
-			config->config_table[i].cam_entry.msb_mac_addr =
-					swab16(*(u16 *)&mclist->dmi_addr[0]);
-			config->config_table[i].cam_entry.middle_mac_addr =
-					swab16(*(u16 *)&mclist->dmi_addr[2]);
-			config->config_table[i].cam_entry.lsb_mac_addr =
-					swab16(*(u16 *)&mclist->dmi_addr[4]);
-			config->config_table[i].cam_entry.flags =
-							cpu_to_le16(bp->port);
-			config->config_table[i].target_table_entry.flags = 0;
-			config->config_table[i].target_table_entry.
-								client_id = 0;
-			config->config_table[i].target_table_entry.
-								vlan_id = 0;
-
-			DP(NETIF_MSG_IFUP,
-			   "setting MCAST[%d] (%04x:%04x:%04x)\n",
-			   i, config->config_table[i].cam_entry.msb_mac_addr,
-			   config->config_table[i].cam_entry.middle_mac_addr,
-			   config->config_table[i].cam_entry.lsb_mac_addr);
-		}
-		old = config->hdr.length_6b;
-		if (old > i) {
-			for (; i < old; i++) {
-				if (CAM_IS_INVALID(config->config_table[i])) {
-					i--; /* already invalidated */
-					break;
-				}
-				/* invalidate */
-				CAM_INVALIDATE(config->config_table[i]);
-			}
-		}
+		break;
 
-		if (CHIP_REV_IS_SLOW(bp))
-			offset = BNX2X_MAX_EMUL_MULTI*(1 + bp->port);
-		else
-			offset = BNX2X_MAX_MULTICAST*(1 + bp->port);
+	case PCI_D3hot:
+		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
+		pmcsr |= 3;
 
-		config->hdr.length_6b = i;
-		config->hdr.offset = offset;
-		config->hdr.reserved0 = 0;
-		config->hdr.reserved1 = 0;
+		if (bp->wol)
+			pmcsr |= PCI_PM_CTRL_PME_ENABLE;
 
-		bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
-			      U64_HI(bnx2x_sp_mapping(bp, mcast_config)),
-			      U64_LO(bnx2x_sp_mapping(bp, mcast_config)), 0);
-	}
+		pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL,
+				      pmcsr);
 
-	bp->rx_mode = rx_mode;
-	bnx2x_set_storm_rx_mode(bp);
+		/* No more memory access after this point until
+		* device is brought back to D0.
+		*/
+		break;
+
+	default:
+		return -EINVAL;
+	}
+	return 0;
 }
 
+/*
+ * net_device service functions
+ */
+
 static int bnx2x_poll(struct napi_struct *napi, int budget)
 {
 	struct bnx2x_fastpath *fp = container_of(napi, struct bnx2x_fastpath,
@@ -6729,7 +7606,7 @@ static int bnx2x_poll(struct napi_struct *napi, int budget)
 
 #ifdef BNX2X_STOP_ON_ERROR
 	if (unlikely(bp->panic))
-		goto out_panic;
+		goto poll_panic;
 #endif
 
 	prefetch(fp->tx_buf_ring[TX_BD(fp->tx_pkt_cons)].skb);
@@ -6738,30 +7615,28 @@ static int bnx2x_poll(struct napi_struct *napi, int budget)
 
 	bnx2x_update_fpsb_idx(fp);
 
-	if (le16_to_cpu(*fp->tx_cons_sb) != fp->tx_pkt_cons)
+	if ((fp->tx_pkt_prod != le16_to_cpu(*fp->tx_cons_sb)) ||
+	    (fp->tx_pkt_prod != fp->tx_pkt_cons))
 		bnx2x_tx_int(fp, budget);
 
-
 	if (le16_to_cpu(*fp->rx_cons_sb) != fp->rx_comp_cons)
 		work_done = bnx2x_rx_int(fp, budget);
 
-
 	rmb(); /* bnx2x_has_work() reads the status block */
 
 	/* must not complete if we consumed full budget */
 	if ((work_done < budget) && !bnx2x_has_work(fp)) {
 
 #ifdef BNX2X_STOP_ON_ERROR
-out_panic:
+poll_panic:
 #endif
 		netif_rx_complete(bp->dev, napi);
 
-		bnx2x_ack_sb(bp, fp->index, USTORM_ID,
+		bnx2x_ack_sb(bp, FP_SB_ID(fp), USTORM_ID,
 			     le16_to_cpu(fp->fp_u_idx), IGU_INT_NOP, 1);
-		bnx2x_ack_sb(bp, fp->index, CSTORM_ID,
+		bnx2x_ack_sb(bp, FP_SB_ID(fp), CSTORM_ID,
 			     le16_to_cpu(fp->fp_c_idx), IGU_INT_ENABLE, 1);
 	}
-
 	return work_done;
 }
 
@@ -7055,18 +7930,145 @@ static int bnx2x_close(struct net_device *dev)
 	return 0;
 }
 
-/* Called with rtnl_lock */
+/* called with netif_tx_lock from set_multicast */
+static void bnx2x_set_rx_mode(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+	u32 rx_mode = BNX2X_RX_MODE_NORMAL;
+	int port = BP_PORT(bp);
+
+	if (bp->state != BNX2X_STATE_OPEN) {
+		DP(NETIF_MSG_IFUP, "state is %x, returning\n", bp->state);
+		return;
+	}
+
+	DP(NETIF_MSG_IFUP, "dev->flags = %x\n", dev->flags);
+
+	if (dev->flags & IFF_PROMISC)
+		rx_mode = BNX2X_RX_MODE_PROMISC;
+
+	else if ((dev->flags & IFF_ALLMULTI) ||
+		 ((dev->mc_count > BNX2X_MAX_MULTICAST) && CHIP_IS_E1(bp)))
+		rx_mode = BNX2X_RX_MODE_ALLMULTI;
+
+	else { /* some multicasts */
+		if (CHIP_IS_E1(bp)) {
+			int i, old, offset;
+			struct dev_mc_list *mclist;
+			struct mac_configuration_cmd *config =
+						bnx2x_sp(bp, mcast_config);
+
+			for (i = 0, mclist = dev->mc_list;
+			     mclist && (i < dev->mc_count);
+			     i++, mclist = mclist->next) {
+
+				config->config_table[i].
+					cam_entry.msb_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[0]);
+				config->config_table[i].
+					cam_entry.middle_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[2]);
+				config->config_table[i].
+					cam_entry.lsb_mac_addr =
+					swab16(*(u16 *)&mclist->dmi_addr[4]);
+				config->config_table[i].cam_entry.flags =
+							cpu_to_le16(port);
+				config->config_table[i].
+					target_table_entry.flags = 0;
+				config->config_table[i].
+					target_table_entry.client_id = 0;
+				config->config_table[i].
+					target_table_entry.vlan_id = 0;
+
+				DP(NETIF_MSG_IFUP,
+				   "setting MCAST[%d] (%04x:%04x:%04x)\n", i,
+				   config->config_table[i].
+						cam_entry.msb_mac_addr,
+				   config->config_table[i].
+						cam_entry.middle_mac_addr,
+				   config->config_table[i].
+						cam_entry.lsb_mac_addr);
+			}
+			old = config->hdr.length_6b;
+			if (old > i) {
+				for (; i < old; i++) {
+					if (CAM_IS_INVALID(config->
+							   config_table[i])) {
+						i--; /* already invalidated */
+						break;
+					}
+					/* invalidate */
+					CAM_INVALIDATE(config->
+						       config_table[i]);
+				}
+			}
+
+			if (CHIP_REV_IS_SLOW(bp))
+				offset = BNX2X_MAX_EMUL_MULTI*(1 + port);
+			else
+				offset = BNX2X_MAX_MULTICAST*(1 + port);
+
+			config->hdr.length_6b = i;
+			config->hdr.offset = offset;
+			config->hdr.client_id = BP_CL_ID(bp);
+			config->hdr.reserved1 = 0;
+
+			bnx2x_sp_post(bp, RAMROD_CMD_ID_ETH_SET_MAC, 0,
+				   U64_HI(bnx2x_sp_mapping(bp, mcast_config)),
+				   U64_LO(bnx2x_sp_mapping(bp, mcast_config)),
+				      0);
+		} else { /* E1H */
+			/* Accept one or more multicasts */
+			struct dev_mc_list *mclist;
+			u32 mc_filter[MC_HASH_SIZE];
+			u32 crc, bit, regidx;
+			int i;
+
+			memset(mc_filter, 0, 4 * MC_HASH_SIZE);
+
+			for (i = 0, mclist = dev->mc_list;
+			     mclist && (i < dev->mc_count);
+			     i++, mclist = mclist->next) {
+
+				DP(NETIF_MSG_IFUP, "Adding mcast MAC: "
+				   "%02x:%02x:%02x:%02x:%02x:%02x\n",
+				   mclist->dmi_addr[0], mclist->dmi_addr[1],
+				   mclist->dmi_addr[2], mclist->dmi_addr[3],
+				   mclist->dmi_addr[4], mclist->dmi_addr[5]);
+
+				crc = crc32c_le(0, mclist->dmi_addr, ETH_ALEN);
+				bit = (crc >> 24) & 0xff;
+				regidx = bit >> 5;
+				bit &= 0x1f;
+				mc_filter[regidx] |= (1 << bit);
+			}
+
+			for (i = 0; i < MC_HASH_SIZE; i++)
+				REG_WR(bp, MC_HASH_OFFSET(bp, i),
+				       mc_filter[i]);
+		}
+	}
+
+	bp->rx_mode = rx_mode;
+	bnx2x_set_storm_rx_mode(bp);
+}
+
+/* called with rtnl_lock */
 static int bnx2x_change_mac_addr(struct net_device *dev, void *p)
 {
 	struct sockaddr *addr = p;
 	struct bnx2x *bp = netdev_priv(dev);
 
-	if (!is_valid_ether_addr(addr->sa_data))
+	if (!is_valid_ether_addr((u8 *)(addr->sa_data)))
 		return -EINVAL;
 
 	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
-	if (netif_running(dev))
-		bnx2x_set_mac_addr(bp);
+	if (netif_running(dev)) {
+		if (CHIP_IS_E1(bp))
+			bnx2x_set_mac_addr_e1(bp);
+		else
+			bnx2x_set_mac_addr_e1h(bp);
+	}
 
 	return 0;
 }
@@ -7080,7 +8082,7 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 	switch (cmd) {
 	case SIOCGMIIPHY:
-		data->phy_id = bp->phy_addr;
+		data->phy_id = bp->port.phy_addr;
 
 		/* fallthrough */
 
@@ -7090,12 +8092,12 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		if (!netif_running(dev))
 			return -EAGAIN;
 
-		mutex_lock(&bp->phy_mutex);
-		err = bnx2x_cl45_read(bp, bp->port, 0, bp->phy_addr,
+		mutex_lock(&bp->port.phy_mutex);
+		err = bnx2x_cl45_read(bp, BP_PORT(bp), 0, bp->port.phy_addr,
 				      DEFAULT_PHY_DEV_ADDR,
 				      (data->reg_num & 0x1f), &mii_regval);
 		data->val_out = mii_regval;
-		mutex_unlock(&bp->phy_mutex);
+		mutex_unlock(&bp->port.phy_mutex);
 		return err;
 	}
 
@@ -7106,11 +8108,11 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		if (!netif_running(dev))
 			return -EAGAIN;
 
-		mutex_lock(&bp->phy_mutex);
-		err = bnx2x_cl45_write(bp, bp->port, 0, bp->phy_addr,
+		mutex_lock(&bp->port.phy_mutex);
+		err = bnx2x_cl45_write(bp, BP_PORT(bp), 0, bp->port.phy_addr,
 				       DEFAULT_PHY_DEV_ADDR,
 				       (data->reg_num & 0x1f), data->val_in);
-		mutex_unlock(&bp->phy_mutex);
+		mutex_unlock(&bp->port.phy_mutex);
 		return err;
 
 	default:
@@ -7121,10 +8123,11 @@ static int bnx2x_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return -EOPNOTSUPP;
 }
 
-/* Called with rtnl_lock */
+/* called with rtnl_lock */
 static int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
 {
 	struct bnx2x *bp = netdev_priv(dev);
+	int rc = 0;
 
 	if ((new_mtu > ETH_MAX_JUMBO_PACKET_SIZE) ||
 	    ((new_mtu + ETH_HLEN) < ETH_MIN_PACKET_SIZE))
@@ -7137,10 +8140,11 @@ static int bnx2x_change_mtu(struct net_device *dev, int new_mtu)
 	dev->mtu = new_mtu;
 
 	if (netif_running(dev)) {
-		bnx2x_nic_unload(bp, 0);
-		bnx2x_nic_load(bp, 0);
+		bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+		rc = bnx2x_nic_load(bp, LOAD_NORMAL);
 	}
-	return 0;
+
+	return rc;
 }
 
 static void bnx2x_tx_timeout(struct net_device *dev)
@@ -7156,7 +8160,7 @@ static void bnx2x_tx_timeout(struct net_device *dev)
 }
 
 #ifdef BCM_VLAN
-/* Called with rtnl_lock */
+/* called with rtnl_lock */
 static void bnx2x_vlan_rx_register(struct net_device *dev,
 				   struct vlan_group *vlgrp)
 {
@@ -7166,6 +8170,7 @@ static void bnx2x_vlan_rx_register(struct net_device *dev,
 	if (netif_running(dev))
 		bnx2x_set_client_config(bp);
 }
+
 #endif
 
 #if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
@@ -7179,36 +8184,8 @@ static void poll_bnx2x(struct net_device *dev)
 }
 #endif
 
-static void bnx2x_reset_task(struct work_struct *work)
-{
-	struct bnx2x *bp = container_of(work, struct bnx2x, reset_task);
-
-#ifdef BNX2X_STOP_ON_ERROR
-	BNX2X_ERR("reset task called but STOP_ON_ERROR defined"
-		  " so reset not done to allow debug dump,\n"
-	 KERN_ERR " you will need to reboot when done\n");
-	return;
-#endif
-
-	if (!netif_running(bp->dev))
-		return;
-
-	rtnl_lock();
-
-	if (bp->state != BNX2X_STATE_OPEN) {
-		DP(NETIF_MSG_TX_ERR, "state is %x, returning\n", bp->state);
-		goto reset_task_exit;
-	}
-
-	bnx2x_nic_unload(bp, 0);
-	bnx2x_nic_load(bp, 0);
-
-reset_task_exit:
-	rtnl_unlock();
-}
-
-static int __devinit bnx2x_init_board(struct pci_dev *pdev,
-				      struct net_device *dev)
+static int __devinit bnx2x_init_dev(struct pci_dev *pdev,
+				    struct net_device *dev)
 {
 	struct bnx2x *bp;
 	int rc;
@@ -7216,8 +8193,10 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 	SET_NETDEV_DEV(dev, &pdev->dev);
 	bp = netdev_priv(dev);
 
+	bp->dev = dev;
+	bp->pdev = pdev;
 	bp->flags = 0;
-	bp->port = PCI_FUNC(pdev->devfn);
+	bp->func = PCI_FUNC(pdev->devfn);
 
 	rc = pci_enable_device(pdev);
 	if (rc) {
@@ -7239,14 +8218,17 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 		goto err_out_disable;
 	}
 
-	rc = pci_request_regions(pdev, DRV_MODULE_NAME);
-	if (rc) {
-		printk(KERN_ERR PFX "Cannot obtain PCI resources,"
-		       " aborting\n");
-		goto err_out_disable;
-	}
+	if (atomic_read(&pdev->enable_cnt) == 1) {
+		rc = pci_request_regions(pdev, DRV_MODULE_NAME);
+		if (rc) {
+			printk(KERN_ERR PFX "Cannot obtain PCI resources,"
+			       " aborting\n");
+			goto err_out_disable;
+		}
 
-	pci_set_master(pdev);
+		pci_set_master(pdev);
+		pci_save_state(pdev);
+	}
 
 	bp->pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM);
 	if (bp->pm_cap == 0) {
@@ -7280,13 +8262,9 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 		goto err_out_release;
 	}
 
-	bp->dev = dev;
-	bp->pdev = pdev;
-
-	INIT_WORK(&bp->reset_task, bnx2x_reset_task);
-	INIT_WORK(&bp->sp_task, bnx2x_sp_task);
-
-	dev->base_addr = pci_resource_start(pdev, 0);
+	dev->mem_start = pci_resource_start(pdev, 0);
+	dev->base_addr = dev->mem_start;
+	dev->mem_end = pci_resource_end(pdev, 0);
 
 	dev->irq = pdev->irq;
 
@@ -7298,8 +8276,9 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 		goto err_out_release;
 	}
 
-	bp->doorbells = ioremap_nocache(pci_resource_start(pdev , 2),
-					pci_resource_len(pdev, 2));
+	bp->doorbells = ioremap_nocache(pci_resource_start(pdev, 2),
+					min_t(u64, BNX2X_DB_SIZE,
+					      pci_resource_len(pdev, 2)));
 	if (!bp->doorbells) {
 		printk(KERN_ERR PFX "Cannot map doorbell space, aborting\n");
 		rc = -ENOMEM;
@@ -7308,47 +8287,43 @@ static int __devinit bnx2x_init_board(struct pci_dev *pdev,
 
 	bnx2x_set_power_state(bp, PCI_D0);
 
-	bnx2x_get_hwinfo(bp);
-
-
-	if (nomcp) {
-		printk(KERN_ERR PFX "MCP disabled, will only"
-		       " init first device\n");
-		onefunc = 1;
-	}
-
-	if (onefunc && bp->port) {
-		printk(KERN_ERR PFX "Second device disabled, exiting\n");
-		rc = -ENODEV;
-		goto err_out_unmap;
-	}
-
-	bp->tx_ring_size = MAX_TX_AVAIL;
-	bp->rx_ring_size = MAX_RX_AVAIL;
-
-	bp->rx_csum = 1;
-
-	bp->rx_offset = 0;
-
-	bp->tx_quick_cons_trip_int = 0xff;
-	bp->tx_quick_cons_trip = 0xff;
-	bp->tx_ticks_int = 50;
-	bp->tx_ticks = 50;
+	/* clean indirect addresses */
+	pci_write_config_dword(bp->pdev, PCICFG_GRC_ADDRESS,
+			       PCICFG_VENDOR_ID_OFFSET);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_88_F0 + BP_PORT(bp)*16, 0);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F0 + BP_PORT(bp)*16, 0);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_90_F0 + BP_PORT(bp)*16, 0);
+	REG_WR(bp, PXP2_REG_PGL_ADDR_94_F0 + BP_PORT(bp)*16, 0);
 
-	bp->rx_quick_cons_trip_int = 0xff;
-	bp->rx_quick_cons_trip = 0xff;
-	bp->rx_ticks_int = 25;
-	bp->rx_ticks = 25;
+	dev->hard_start_xmit = bnx2x_start_xmit;
+	dev->watchdog_timeo = TX_TIMEOUT;
 
-	bp->stats_ticks = 1000000 & 0xffff00;
+	dev->ethtool_ops = &bnx2x_ethtool_ops;
+	dev->open = bnx2x_open;
+	dev->stop = bnx2x_close;
+	dev->set_multicast_list = bnx2x_set_rx_mode;
+	dev->set_mac_address = bnx2x_change_mac_addr;
+	dev->do_ioctl = bnx2x_ioctl;
+	dev->change_mtu = bnx2x_change_mtu;
+	dev->tx_timeout = bnx2x_tx_timeout;
+#ifdef BCM_VLAN
+	dev->vlan_rx_register = bnx2x_vlan_rx_register;
+#endif
+#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
+	dev->poll_controller = poll_bnx2x;
+#endif
+	dev->features |= NETIF_F_SG;
+	dev->features |= NETIF_F_HW_CSUM;
+	if (bp->flags & USING_DAC_FLAG)
+		dev->features |= NETIF_F_HIGHDMA;
+#ifdef BCM_VLAN
+	dev->features |= (NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX);
+#endif
+	dev->features |= (NETIF_F_TSO | NETIF_F_TSO_ECN);
 
 	bp->timer_interval = HZ;
 	bp->current_interval = (poll ? poll : HZ);
 
-	init_timer(&bp->timer);
-	bp->timer.expires = jiffies + bp->current_interval;
-	bp->timer.data = (unsigned long) bp;
-	bp->timer.function = bnx2x_timer;
 
 	return 0;
 
@@ -7357,14 +8332,14 @@ err_out_unmap:
 		iounmap(bp->regview);
 		bp->regview = NULL;
 	}
-
 	if (bp->doorbells) {
 		iounmap(bp->doorbells);
 		bp->doorbells = NULL;
 	}
 
 err_out_release:
-	pci_release_regions(pdev);
+	if (atomic_read(&pdev->enable_cnt) == 1)
+		pci_release_regions(pdev);
 
 err_out_disable:
 	pci_disable_device(pdev);
@@ -7398,7 +8373,6 @@ static int __devinit bnx2x_init_one(struct pci_dev *pdev,
 	struct net_device *dev = NULL;
 	struct bnx2x *bp;
 	int rc;
-	int port = PCI_FUNC(pdev->devfn);
 	DECLARE_MAC_BUF(mac);
 
 	if (version_printed++ == 0)
@@ -7406,78 +8380,62 @@ static int __devinit bnx2x_init_one(struct pci_dev *pdev,
 
 	/* dev zeroed in init_etherdev */
 	dev = alloc_etherdev(sizeof(*bp));
-	if (!dev)
+	if (!dev) {
+		printk(KERN_ERR PFX "Cannot allocate net device\n");
 		return -ENOMEM;
+	}
 
 	netif_carrier_off(dev);
 
 	bp = netdev_priv(dev);
 	bp->msglevel = debug;
 
-	if (port && onefunc) {
-		printk(KERN_ERR PFX "second function disabled. exiting\n");
-		free_netdev(dev);
-		return 0;
-	}
-
-	rc = bnx2x_init_board(pdev, dev);
+	rc = bnx2x_init_dev(pdev, dev);
 	if (rc < 0) {
 		free_netdev(dev);
 		return rc;
 	}
 
-	dev->hard_start_xmit = bnx2x_start_xmit;
-	dev->watchdog_timeo = TX_TIMEOUT;
-
-	dev->ethtool_ops = &bnx2x_ethtool_ops;
-	dev->open = bnx2x_open;
-	dev->stop = bnx2x_close;
-	dev->set_multicast_list = bnx2x_set_rx_mode;
-	dev->set_mac_address = bnx2x_change_mac_addr;
-	dev->do_ioctl = bnx2x_ioctl;
-	dev->change_mtu = bnx2x_change_mtu;
-	dev->tx_timeout = bnx2x_tx_timeout;
-#ifdef BCM_VLAN
-	dev->vlan_rx_register = bnx2x_vlan_rx_register;
-#endif
-#if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER)
-	dev->poll_controller = poll_bnx2x;
-#endif
-	dev->features |= NETIF_F_SG;
-	if (bp->flags & USING_DAC_FLAG)
-		dev->features |= NETIF_F_HIGHDMA;
-	dev->features |= NETIF_F_IP_CSUM;
-#ifdef BCM_VLAN
-	dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
-#endif
-	dev->features |= NETIF_F_TSO | NETIF_F_TSO_ECN;
-
 	rc = register_netdev(dev);
 	if (rc) {
 		dev_err(&pdev->dev, "Cannot register net device\n");
-		if (bp->regview)
-			iounmap(bp->regview);
-		if (bp->doorbells)
-			iounmap(bp->doorbells);
-		pci_release_regions(pdev);
-		pci_disable_device(pdev);
-		pci_set_drvdata(pdev, NULL);
-		free_netdev(dev);
-		return rc;
+		goto init_one_exit;
 	}
 
 	pci_set_drvdata(pdev, dev);
 
-	bp->name = board_info[ent->driver_data].name;
+	rc = bnx2x_init_bp(bp);
+	if (rc) {
+		unregister_netdev(dev);
+		goto init_one_exit;
+	}
+
+	bp->common.name = board_info[ent->driver_data].name;
 	printk(KERN_INFO "%s: %s (%c%d) PCI-E x%d %s found at mem %lx,"
-	       " IRQ %d, ", dev->name, bp->name,
-	       ((CHIP_ID(bp) & 0xf000) >> 12) + 'A',
-	       ((CHIP_ID(bp) & 0x0ff0) >> 4),
+	       " IRQ %d, ", dev->name, bp->common.name,
+	       (CHIP_REV(bp) >> 12) + 'A', (CHIP_METAL(bp) >> 4),
 	       bnx2x_get_pcie_width(bp),
 	       (bnx2x_get_pcie_speed(bp) == 2) ? "5GHz (Gen2)" : "2.5GHz",
 	       dev->base_addr, bp->pdev->irq);
 	printk(KERN_CONT "node addr %s\n", print_mac(mac, dev->dev_addr));
 	return 0;
+
+init_one_exit:
+	if (bp->regview)
+		iounmap(bp->regview);
+
+	if (bp->doorbells)
+		iounmap(bp->doorbells);
+
+	free_netdev(dev);
+
+	if (atomic_read(&pdev->enable_cnt) == 1)
+		pci_release_regions(pdev);
+
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+
+	return rc;
 }
 
 static void __devexit bnx2x_remove_one(struct pci_dev *pdev)
@@ -7486,11 +8444,9 @@ static void __devexit bnx2x_remove_one(struct pci_dev *pdev)
 	struct bnx2x *bp;
 
 	if (!dev) {
-		/* we get here if init_one() fails */
 		printk(KERN_ERR PFX "BAD net device from bnx2x_init_one\n");
 		return;
 	}
-
 	bp = netdev_priv(dev);
 
 	unregister_netdev(dev);
@@ -7502,7 +8458,10 @@ static void __devexit bnx2x_remove_one(struct pci_dev *pdev)
 		iounmap(bp->doorbells);
 
 	free_netdev(dev);
-	pci_release_regions(pdev);
+
+	if (atomic_read(&pdev->enable_cnt) == 1)
+		pci_release_regions(pdev);
+
 	pci_disable_device(pdev);
 	pci_set_drvdata(pdev, NULL);
 }
@@ -7512,21 +8471,29 @@ static int bnx2x_suspend(struct pci_dev *pdev, pm_message_t state)
 	struct net_device *dev = pci_get_drvdata(pdev);
 	struct bnx2x *bp;
 
-	if (!dev)
-		return 0;
+	if (!dev) {
+		printk(KERN_ERR PFX "BAD net device from bnx2x_init_one\n");
+		return -ENODEV;
+	}
+	bp = netdev_priv(dev);
 
-	if (!netif_running(dev))
-		return 0;
+	rtnl_lock();
 
-	bp = netdev_priv(dev);
+	pci_save_state(pdev);
 
-	bnx2x_nic_unload(bp, 0);
+	if (!netif_running(dev)) {
+		rtnl_unlock();
+		return 0;
+	}
 
 	netif_device_detach(dev);
 
-	pci_save_state(pdev);
+	bnx2x_nic_unload(bp, UNLOAD_NORMAL);
+
 	bnx2x_set_power_state(bp, pci_choose_state(pdev, state));
 
+	rtnl_unlock();
+
 	return 0;
 }
 
@@ -7540,21 +8507,25 @@ static int bnx2x_resume(struct pci_dev *pdev)
 		printk(KERN_ERR PFX "BAD net device from bnx2x_init_one\n");
 		return -ENODEV;
 	}
-
-	if (!netif_running(dev))
-		return 0;
-
 	bp = netdev_priv(dev);
 
+	rtnl_lock();
+
 	pci_restore_state(pdev);
+
+	if (!netif_running(dev)) {
+		rtnl_unlock();
+		return 0;
+	}
+
 	bnx2x_set_power_state(bp, PCI_D0);
 	netif_device_attach(dev);
 
-	rc = bnx2x_nic_load(bp, 0);
-	if (rc)
-		return rc;
+	rc = bnx2x_nic_load(bp, LOAD_NORMAL);
 
-	return 0;
+	rtnl_unlock();
+
+	return rc;
 }
 
 static struct pci_driver bnx2x_pci_driver = {
diff --git a/drivers/net/bnx2x_reg.h b/drivers/net/bnx2x_reg.h
index 8707c0d05d9a..15c9a9946724 100644
--- a/drivers/net/bnx2x_reg.h
+++ b/drivers/net/bnx2x_reg.h
@@ -38,21 +38,19 @@
    was asserted. */
 #define BRB1_REG_NUM_OF_FULL_CYCLES_0				 0x600c8
 #define BRB1_REG_NUM_OF_FULL_CYCLES_1				 0x600cc
-#define BRB1_REG_NUM_OF_FULL_CYCLES_2				 0x600d0
-#define BRB1_REG_NUM_OF_FULL_CYCLES_3				 0x600d4
 #define BRB1_REG_NUM_OF_FULL_CYCLES_4				 0x600d8
 /* [ST 32] The number of cycles that the pause signal towards MAC #0 was
    asserted. */
 #define BRB1_REG_NUM_OF_PAUSE_CYCLES_0				 0x600b8
 #define BRB1_REG_NUM_OF_PAUSE_CYCLES_1				 0x600bc
-#define BRB1_REG_NUM_OF_PAUSE_CYCLES_2				 0x600c0
-#define BRB1_REG_NUM_OF_PAUSE_CYCLES_3				 0x600c4
 /* [RW 10] Write client 0: De-assert pause threshold. */
 #define BRB1_REG_PAUSE_HIGH_THRESHOLD_0 			 0x60078
 #define BRB1_REG_PAUSE_HIGH_THRESHOLD_1 			 0x6007c
 /* [RW 10] Write client 0: Assert pause threshold. */
 #define BRB1_REG_PAUSE_LOW_THRESHOLD_0				 0x60068
 #define BRB1_REG_PAUSE_LOW_THRESHOLD_1				 0x6006c
+/* [R 24] The number of full blocks occpied by port. */
+#define BRB1_REG_PORT_NUM_OCC_BLOCKS_0				 0x60094
 /* [RW 1] Reset the design by software. */
 #define BRB1_REG_SOFT_RESET					 0x600dc
 /* [R 5] Used to read the value of the XX protection CAM occupancy counter. */
@@ -513,7 +511,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define CSEM_REG_INT_TABLE					 0x200400
-#define CSEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define CSEM_REG_MSG_NUM_FIC0					 0x200000
@@ -587,13 +584,10 @@
 #define DBG_REG_DBG_PRTY_MASK					 0xc0a8
 /* [R 1] Parity register #0 read */
 #define DBG_REG_DBG_PRTY_STS					 0xc09c
-/* [RW 2] debug only: These bits indicate the credit for PCI request type 4
-   interface; MUST be configured AFTER pci_ext_buffer_strt_addr_lsb/msb are
-   configured */
-#define DBG_REG_PCI_REQ_CREDIT					 0xc120
 /* [RW 32] Commands memory. The address to command X; row Y is to calculated
    as 14*X+Y. */
 #define DMAE_REG_CMD_MEM					 0x102400
+#define DMAE_REG_CMD_MEM_SIZE					 224
 /* [RW 1] If 0 - the CRC-16c initial value is all zeroes; if 1 - the CRC-16c
    initial value is all ones. */
 #define DMAE_REG_CRC16C_INIT					 0x10201c
@@ -1626,7 +1620,7 @@
    is reset to 0x080; giving a default blink period of approximately 8Hz. */
 #define NIG_REG_LED_CONTROL_BLINK_RATE_P0			 0x10310
 /* [RW 1] Port0: If set along with the
-   nig_registers_led_control_override_traffic_p0.led_control_override_traffic_p0
+ ~nig_registers_led_control_override_traffic_p0.led_control_override_traffic_p0
    bit and ~nig_registers_led_control_traffic_p0.led_control_traffic_p0 LED
    bit; the Traffic LED will blink with the blink rate specified in
    ~nig_registers_led_control_blink_rate_p0.led_control_blink_rate_p0 and
@@ -1733,9 +1727,21 @@
 /* [R 32] Rx statistics : In user packets discarded due to BRB backpressure
    for port0 */
 #define NIG_REG_STAT0_BRB_DISCARD				 0x105f0
+/* [WB_R 36] Tx statistics : Number of packets from emac0 or bmac0 that
+   between 1024 and 1522 bytes for port0 */
+#define NIG_REG_STAT0_EGRESS_MAC_PKT0				 0x10750
+/* [WB_R 36] Tx statistics : Number of packets from emac0 or bmac0 that
+   between 1523 bytes and above for port0 */
+#define NIG_REG_STAT0_EGRESS_MAC_PKT1				 0x10760
 /* [R 32] Rx statistics : In user packets discarded due to BRB backpressure
    for port1 */
 #define NIG_REG_STAT1_BRB_DISCARD				 0x10628
+/* [WB_R 36] Tx statistics : Number of packets from emac1 or bmac1 that
+   between 1024 and 1522 bytes for port1 */
+#define NIG_REG_STAT1_EGRESS_MAC_PKT0				 0x107a0
+/* [WB_R 36] Tx statistics : Number of packets from emac1 or bmac1 that
+   between 1523 bytes and above for port1 */
+#define NIG_REG_STAT1_EGRESS_MAC_PKT1				 0x107b0
 /* [WB_R 64] Rx statistics : User octets received for LP */
 #define NIG_REG_STAT2_BRB_OCTET 				 0x107e0
 #define NIG_REG_STATUS_INTERRUPT_PORT0				 0x10328
@@ -1849,7 +1855,6 @@
 #define PRS_REG_CFC_SEARCH_INITIAL_CREDIT			 0x4011c
 /* [RW 24] CID for port 0 if no match */
 #define PRS_REG_CID_PORT_0					 0x400fc
-#define PRS_REG_CID_PORT_1					 0x40100
 /* [RW 32] The CM header for flush message where 'load existed' bit in CFC
    load response is reset and packet type is 0. Used in packet start message
    to TCM. */
@@ -1957,6 +1962,10 @@
 #define PXP2_REG_HST_DATA_FIFO_STATUS				 0x12047c
 /* [R 7] Debug only: Number of used entries in the header FIFO */
 #define PXP2_REG_HST_HEADER_FIFO_STATUS 			 0x120478
+#define PXP2_REG_PGL_ADDR_88_F0 				 0x120534
+#define PXP2_REG_PGL_ADDR_8C_F0 				 0x120538
+#define PXP2_REG_PGL_ADDR_90_F0 				 0x12053c
+#define PXP2_REG_PGL_ADDR_94_F0 				 0x120540
 #define PXP2_REG_PGL_CONTROL0					 0x120490
 #define PXP2_REG_PGL_CONTROL1					 0x120514
 /* [RW 32] third dword data of expansion rom request. this register is
@@ -2060,12 +2069,13 @@
 #define PXP2_REG_PSWRQ_SRC0_L2P 				 0x120054
 #define PXP2_REG_PSWRQ_TM0_L2P					 0x12001c
 #define PXP2_REG_PSWRQ_TSDM0_L2P				 0x1200e0
-/* [RW 25] Interrupt mask register #0 read/write */
-#define PXP2_REG_PXP2_INT_MASK					 0x120578
-/* [R 25] Interrupt register #0 read */
-#define PXP2_REG_PXP2_INT_STS					 0x12056c
-/* [RC 25] Interrupt register #0 read clear */
-#define PXP2_REG_PXP2_INT_STS_CLR				 0x120570
+/* [RW 32] Interrupt mask register #0 read/write */
+#define PXP2_REG_PXP2_INT_MASK_0				 0x120578
+/* [R 32] Interrupt register #0 read */
+#define PXP2_REG_PXP2_INT_STS_0 				 0x12056c
+#define PXP2_REG_PXP2_INT_STS_1 				 0x120608
+/* [RC 32] Interrupt register #0 read clear */
+#define PXP2_REG_PXP2_INT_STS_CLR_0				 0x120570
 /* [RW 32] Parity mask register #0 read/write */
 #define PXP2_REG_PXP2_PRTY_MASK_0				 0x120588
 #define PXP2_REG_PXP2_PRTY_MASK_1				 0x120598
@@ -2811,22 +2821,6 @@
 #define QM_REG_QVOQIDX_97					 0x16e490
 #define QM_REG_QVOQIDX_98					 0x16e494
 #define QM_REG_QVOQIDX_99					 0x16e498
-/* [R 24] Remaining pause timeout for queues 15-0 */
-#define QM_REG_REMAINPAUSETM0					 0x168418
-/* [R 24] Remaining pause timeout for queues 31-16 */
-#define QM_REG_REMAINPAUSETM1					 0x16841c
-/* [R 24] Remaining pause timeout for queues 47-32 */
-#define QM_REG_REMAINPAUSETM2					 0x16e69c
-/* [R 24] Remaining pause timeout for queues 63-48 */
-#define QM_REG_REMAINPAUSETM3					 0x16e6a0
-/* [R 24] Remaining pause timeout for queues 79-64 */
-#define QM_REG_REMAINPAUSETM4					 0x16e6a4
-/* [R 24] Remaining pause timeout for queues 95-80 */
-#define QM_REG_REMAINPAUSETM5					 0x16e6a8
-/* [R 24] Remaining pause timeout for queues 111-96 */
-#define QM_REG_REMAINPAUSETM6					 0x16e6ac
-/* [R 24] Remaining pause timeout for queues 127-112 */
-#define QM_REG_REMAINPAUSETM7					 0x16e6b0
 /* [RW 1] Initialization bit command */
 #define QM_REG_SOFT_RESET					 0x168428
 /* [RW 8] The credit cost per every task in the QM. A value per each VOQ */
@@ -3826,7 +3820,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define TSEM_REG_INT_TABLE					 0x180400
-#define TSEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define TSEM_REG_MSG_NUM_FIC0					 0x180000
@@ -4283,7 +4276,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define USEM_REG_INT_TABLE					 0x300400
-#define USEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define USEM_REG_MSG_NUM_FIC0					 0x300000
@@ -4802,7 +4794,6 @@
 /* [RW 15] Interrupt table Read and write access to it is not possible in
    the middle of the work */
 #define XSEM_REG_INT_TABLE					 0x280400
-#define XSEM_REG_INT_TABLE_SIZE 				 256
 /* [ST 24] Statistics register. The number of messages that entered through
    FIC0 */
 #define XSEM_REG_MSG_NUM_FIC0					 0x280000
@@ -4930,10 +4921,7 @@
 #define EMAC_MDIO_MODE_CLOCK_CNT				 (0x3fL<<16)
 #define EMAC_MDIO_MODE_CLOCK_CNT_BITSHIFT			 16
 #define EMAC_MODE_25G_MODE					 (1L<<5)
-#define EMAC_MODE_ACPI_RCVD					 (1L<<20)
 #define EMAC_MODE_HALF_DUPLEX					 (1L<<1)
-#define EMAC_MODE_MPKT						 (1L<<18)
-#define EMAC_MODE_MPKT_RCVD					 (1L<<19)
 #define EMAC_MODE_PORT_GMII					 (2L<<2)
 #define EMAC_MODE_PORT_MII					 (1L<<2)
 #define EMAC_MODE_PORT_MII_10M					 (3L<<2)
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index caa000596b25..e74b14acf8e0 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1949,6 +1949,8 @@
 #define PCI_DEVICE_ID_NX2_5708		0x164c
 #define PCI_DEVICE_ID_TIGON3_5702FE	0x164d
 #define PCI_DEVICE_ID_NX2_57710		0x164e
+#define PCI_DEVICE_ID_NX2_57711		0x164f
+#define PCI_DEVICE_ID_NX2_57711E	0x1650
 #define PCI_DEVICE_ID_TIGON3_5705	0x1653
 #define PCI_DEVICE_ID_TIGON3_5705_2	0x1654
 #define PCI_DEVICE_ID_TIGON3_5720	0x1658
-- 
cgit v1.2.3


From a033c332e047397904ed74816946b2edd9b0d5cd Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Mon, 23 Jun 2008 10:52:42 +0800
Subject: lockdep: remove duplicate definition of STATIC_LOCKDEP_MAP_INIT

STATIC_LOCKDEP_MAP_INIT is defined twice in lockdep.h. I guess
it's a copy & paste.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/lockdep.h | 8 --------
 1 file changed, 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index b26fbc715a50..2486eb4edbf1 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -278,14 +278,6 @@ extern void lockdep_init_map(struct lockdep_map *lock, const char *name,
 		lockdep_init_map(&(lock)->dep_map, #lock, \
 				 (lock)->dep_map.key, sub)
 
-/*
- * To initialize a lockdep_map statically use this macro.
- * Note that _name must not be NULL.
- */
-#define STATIC_LOCKDEP_MAP_INIT(_name, _key) \
-	{ .name = (_name), .key = (void *)(_key), }
-
-
 /*
  * Acquire a lock.
  *
-- 
cgit v1.2.3


From f3f3149f35b9195ef4b761b1353fc0766b5f53be Mon Sep 17 00:00:00 2001
From: Alok Kataria <akataria@vmware.com>
Date: Mon, 23 Jun 2008 18:21:56 -0700
Subject: x86: use cpu_khz for loops_per_jiffy calculation, cleanup

As suggested by Ingo, remove all references to tsc from init/calibrate.c

TSC is x86 specific, and using tsc in variable names in a generic file should
be avoided. lpj_tsc is now called lpj_fine, since it is related to fine tuning
of lpj value. Also tsc_rate_*  is called timer_rate_*

Signed-off-by: Alok N Kataria <akataria@vmware.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Daniel Hecht <dhecht@vmware.com>
Cc: Tim Mann <mann@vmware.com>
Cc: Zach Amsden <zach@vmware.com>
Cc: Sahil Rihan <srihan@vmware.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/time_64.c |  2 +-
 arch/x86/kernel/tsc_32.c  |  2 +-
 include/linux/delay.h     |  2 +-
 init/calibrate.c          | 36 +++++++++++++++++++-----------------
 4 files changed, 22 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index 12b4a71bd074..39ae8511a137 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -123,7 +123,7 @@ void __init time_init(void)
 		(boot_cpu_data.x86_vendor == X86_VENDOR_AMD))
 		cpu_khz = calculate_cpu_khz();
 
-	lpj_tsc = ((unsigned long)tsc_khz * 1000)/HZ;
+	lpj_fine = ((unsigned long)tsc_khz * 1000)/HZ;
 
 	if (unsynchronized_tsc())
 		mark_tsc_unstable("TSCs unsynchronized");
diff --git a/arch/x86/kernel/tsc_32.c b/arch/x86/kernel/tsc_32.c
index 0af49fb533e1..048baab77268 100644
--- a/arch/x86/kernel/tsc_32.c
+++ b/arch/x86/kernel/tsc_32.c
@@ -425,7 +425,7 @@ void __init tsc_init(void)
 
 	lpj = ((u64)tsc_khz * 1000);
 	do_div(lpj, HZ);
-	lpj_tsc = lpj;
+	lpj_fine = lpj;
 
 	printk("Detected %lu.%03lu MHz processor.\n",
 				(unsigned long)cpu_khz / 1000,
diff --git a/include/linux/delay.h b/include/linux/delay.h
index 01aec60590ab..fd832c6d419e 100644
--- a/include/linux/delay.h
+++ b/include/linux/delay.h
@@ -41,7 +41,7 @@ static inline void ndelay(unsigned long x)
 #define ndelay(x) ndelay(x)
 #endif
 
-extern unsigned long lpj_tsc;
+extern unsigned long lpj_fine;
 void calibrate_delay(void);
 void msleep(unsigned int msecs);
 unsigned long msleep_interruptible(unsigned int msecs);
diff --git a/init/calibrate.c b/init/calibrate.c
index 86286974dada..7963e3fc51d9 100644
--- a/init/calibrate.c
+++ b/init/calibrate.c
@@ -10,7 +10,7 @@
 #include <linux/timex.h>
 #include <linux/smp.h>
 
-unsigned long lpj_tsc;
+unsigned long lpj_fine;
 unsigned long preset_lpj;
 static int __init lpj_setup(char *str)
 {
@@ -35,9 +35,9 @@ static unsigned long __cpuinit calibrate_delay_direct(void)
 	unsigned long pre_start, start, post_start;
 	unsigned long pre_end, end, post_end;
 	unsigned long start_jiffies;
-	unsigned long tsc_rate_min, tsc_rate_max;
-	unsigned long good_tsc_sum = 0;
-	unsigned long good_tsc_count = 0;
+	unsigned long timer_rate_min, timer_rate_max;
+	unsigned long good_timer_sum = 0;
+	unsigned long good_timer_count = 0;
 	int i;
 
 	if (read_current_timer(&pre_start) < 0 )
@@ -81,22 +81,24 @@ static unsigned long __cpuinit calibrate_delay_direct(void)
 		}
 		read_current_timer(&post_end);
 
-		tsc_rate_max = (post_end - pre_start) / DELAY_CALIBRATION_TICKS;
-		tsc_rate_min = (pre_end - post_start) / DELAY_CALIBRATION_TICKS;
+		timer_rate_max = (post_end - pre_start) /
+					DELAY_CALIBRATION_TICKS;
+		timer_rate_min = (pre_end - post_start) /
+					DELAY_CALIBRATION_TICKS;
 
 		/*
-	 	 * If the upper limit and lower limit of the tsc_rate is
+		 * If the upper limit and lower limit of the timer_rate is
 		 * >= 12.5% apart, redo calibration.
 		 */
 		if (pre_start != 0 && pre_end != 0 &&
-		    (tsc_rate_max - tsc_rate_min) < (tsc_rate_max >> 3)) {
-			good_tsc_count++;
-			good_tsc_sum += tsc_rate_max;
+		    (timer_rate_max - timer_rate_min) < (timer_rate_max >> 3)) {
+			good_timer_count++;
+			good_timer_sum += timer_rate_max;
 		}
 	}
 
-	if (good_tsc_count)
-		return (good_tsc_sum/good_tsc_count);
+	if (good_timer_count)
+		return (good_timer_sum/good_timer_count);
 
 	printk(KERN_WARNING "calibrate_delay_direct() failed to get a good "
 	       "estimate for loops_per_jiffy.\nProbably due to long platform interrupts. Consider using \"lpj=\" boot option.\n");
@@ -111,8 +113,8 @@ static unsigned long __cpuinit calibrate_delay_direct(void) {return 0;}
  * bit takes on average 1.5/HZ seconds.  This (like the original) is a little
  * better than 1%
  * For the boot cpu we can skip the delay calibration and assign it a value
- * calculated based on the tsc frequency.
- * For the rest of the CPUs we cannot assume that the tsc frequency is same as
+ * calculated based on the timer frequency.
+ * For the rest of the CPUs we cannot assume that the timer frequency is same as
  * the cpu frequency, hence do the calibration for those.
  */
 #define LPS_PREC 8
@@ -126,11 +128,11 @@ void __cpuinit calibrate_delay(void)
 		loops_per_jiffy = preset_lpj;
 		printk(KERN_INFO
 			"Calibrating delay loop (skipped) preset value.. ");
-	} else if ((smp_processor_id() == 0) && lpj_tsc) {
-		loops_per_jiffy = lpj_tsc;
+	} else if ((smp_processor_id() == 0) && lpj_fine) {
+		loops_per_jiffy = lpj_fine;
 		printk(KERN_INFO
 			"Calibrating delay loop (skipped), "
-			"using tsc calculated value.. ");
+			"value calculated using timer frequency.. ");
 	} else if ((loops_per_jiffy = calibrate_delay_direct()) != 0) {
 		printk(KERN_INFO
 			"Calibrating delay using timer specific routine.. ");
-- 
cgit v1.2.3


From f8dd0ecbb74d4b220b105d77c0633945ebb5453e Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 24 Jun 2008 22:16:04 +0100
Subject: DM9000: Allow the use of the NSR register to get link status.

The DM9000's internal PHY reports a copy of the link status
in the NSR register of the chip. Reading the status when
polling for link status is faster as it eliminates the need
to sleep, but does not print as much information.

Add an platform flag to force this behaviour, and a Kconfig
option to allow it to be forced to the faster method always.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/Kconfig    |  9 +++++++++
 drivers/net/dm9000.c   | 41 +++++++++++++++++++++++++++++++++++++++--
 include/linux/dm9000.h |  1 +
 3 files changed, 49 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 287d0873c60d..4d69474b6125 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -938,6 +938,15 @@ config DM9000
 	  To compile this driver as a module, choose M here.  The module
 	  will be called dm9000.
 
+config DM9000_FORCE_SIMPLE_PHY_POLL
+	bool "Force simple NSR based PHY polling"
+	depends on DM9000
+	---help---
+	  This configuration forces the DM9000 to use the NSR's LinkStatus
+	  bit to determine if the link is up or down instead of the more
+	  costly MII PHY reads. Note, this will not work if the chip is
+	  operating with an external PHY.
+
 config ENC28J60
 	tristate "ENC28J60 support"
 	depends on EXPERIMENTAL && SPI && NET_ETHERNET
diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c
index 7c38f6129b55..5ad2ec537684 100644
--- a/drivers/net/dm9000.c
+++ b/drivers/net/dm9000.c
@@ -552,15 +552,48 @@ static const struct ethtool_ops dm9000_ethtool_ops = {
  	.set_eeprom		= dm9000_set_eeprom,
 };
 
+static void dm9000_show_carrier(board_info_t *db,
+				unsigned carrier, unsigned nsr)
+{
+	struct net_device *ndev = db->ndev;
+	unsigned ncr = dm9000_read_locked(db, DM9000_NCR);
+
+	if (carrier)
+		dev_info(db->dev, "%s: link up, %dMbps, %s-duplex, no LPA\n",
+			 ndev->name, (nsr & NSR_SPEED) ? 10 : 100,
+			 (ncr & NCR_FDX) ? "full" : "half");
+	else
+		dev_info(db->dev, "%s: link down\n", ndev->name);
+}
+
 static void
 dm9000_poll_work(struct work_struct *w)
 {
 	struct delayed_work *dw = container_of(w, struct delayed_work, work);
 	board_info_t *db = container_of(dw, board_info_t, phy_poll);
+	struct net_device *ndev = db->ndev;
+
+	if (db->flags & DM9000_PLATF_SIMPLE_PHY &&
+	    !(db->flags & DM9000_PLATF_EXT_PHY)) {
+		unsigned nsr = dm9000_read_locked(db, DM9000_NSR);
+		unsigned old_carrier = netif_carrier_ok(ndev) ? 1 : 0;
+		unsigned new_carrier;
 
-	mii_check_media(&db->mii, netif_msg_link(db), 0);
+		new_carrier = (nsr & NSR_LINKST) ? 1 : 0;
+
+		if (old_carrier != new_carrier) {
+			if (netif_msg_link(db))
+				dm9000_show_carrier(db, new_carrier, nsr);
+
+			if (!new_carrier)
+				netif_carrier_off(ndev);
+			else
+				netif_carrier_on(ndev);
+		}
+	} else
+		mii_check_media(&db->mii, netif_msg_link(db), 0);
 	
-	if (netif_running(db->ndev))
+	if (netif_running(ndev))
 		dm9000_schedule_poll(db);
 }
 
@@ -1267,6 +1300,10 @@ dm9000_probe(struct platform_device *pdev)
 		db->flags = pdata->flags;
 	}
 
+#ifdef CONFIG_DM9000_FORCE_SIMPLE_PHY_POLL
+	db->flags |= DM9000_PLATF_SIMPLE_PHY;
+#endif
+
 	dm9000_reset(db);
 
 	/* try multiple times, DM9000 sometimes gets the read wrong */
diff --git a/include/linux/dm9000.h b/include/linux/dm9000.h
index a3750462f9e3..fc82446b6425 100644
--- a/include/linux/dm9000.h
+++ b/include/linux/dm9000.h
@@ -21,6 +21,7 @@
 #define DM9000_PLATF_32BITONLY	(0x0004)
 #define DM9000_PLATF_EXT_PHY	(0x0008)
 #define DM9000_PLATF_NO_EEPROM	(0x0010)
+#define DM9000_PLATF_SIMPLE_PHY (0x0020)  /* Use NSR to find LinkStatus */
 
 /* platfrom data for platfrom device structure's platfrom_data field */
 
-- 
cgit v1.2.3


From 3d4422332711ef48ef0f132f1fcbfcbd56c7f3d1 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 26 Jun 2008 11:21:34 +0200
Subject: Add generic helpers for arch IPI function calls

This adds kernel/smp.c which contains helpers for IPI function calls. In
addition to supporting the existing smp_call_function() in a more efficient
manner, it also adds a more scalable variant called smp_call_function_single()
for calling a given function on a single CPU only.

The core of this is based on the x86-64 patch from Nick Piggin, lots of
changes since then. "Alan D. Brunelle" <Alan.Brunelle@hp.com> has
contributed lots of fixes and suggestions as well. Also thanks to
Paul E. McKenney <paulmck@linux.vnet.ibm.com> for reviewing RCU usage
and getting rid of the data allocation fallback deadlock.

Acked-by: Ingo Molnar <mingo@elte.hu>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 arch/Kconfig              |   3 +
 arch/sparc64/kernel/smp.c |  11 +-
 include/linux/smp.h       |  35 ++++-
 init/main.c               |   2 +
 kernel/Makefile           |   1 +
 kernel/smp.c              | 383 ++++++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 428 insertions(+), 7 deletions(-)
 create mode 100644 kernel/smp.c

(limited to 'include/linux')

diff --git a/arch/Kconfig b/arch/Kconfig
index 3ea332b009e5..ad89a33d8c6e 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -39,3 +39,6 @@ config HAVE_KRETPROBES
 
 config HAVE_DMA_ATTRS
 	def_bool n
+
+config USE_GENERIC_SMP_HELPERS
+	def_bool n
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index fa63c68a1819..b82d017a1744 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -816,8 +816,9 @@ extern unsigned long xcall_call_function;
  * You must not call this function with disabled interrupts or from a
  * hardware interrupt handler or from a bottom half handler.
  */
-static int smp_call_function_mask(void (*func)(void *info), void *info,
-				  int nonatomic, int wait, cpumask_t mask)
+static int sparc64_smp_call_function_mask(void (*func)(void *info), void *info,
+					  int nonatomic, int wait,
+					  cpumask_t mask)
 {
 	struct call_data_struct data;
 	int cpus;
@@ -855,8 +856,8 @@ out_unlock:
 int smp_call_function(void (*func)(void *info), void *info,
 		      int nonatomic, int wait)
 {
-	return smp_call_function_mask(func, info, nonatomic, wait,
-				      cpu_online_map);
+	return sparc64_smp_call_function_mask(func, info, nonatomic, wait,
+						cpu_online_map);
 }
 
 void smp_call_function_client(int irq, struct pt_regs *regs)
@@ -893,7 +894,7 @@ static void tsb_sync(void *info)
 
 void smp_tsb_sync(struct mm_struct *mm)
 {
-	smp_call_function_mask(tsb_sync, mm, 0, 1, mm->cpu_vm_mask);
+	sparc64_smp_call_function_mask(tsb_sync, mm, 0, 1, mm->cpu_vm_mask);
 }
 
 extern unsigned long xcall_flush_tlb_mm;
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 55232ccf9cfd..eac3e062250f 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -7,9 +7,19 @@
  */
 
 #include <linux/errno.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/cpumask.h>
 
 extern void cpu_idle(void);
 
+struct call_single_data {
+	struct list_head list;
+	void (*func) (void *info);
+	void *info;
+	unsigned int flags;
+};
+
 #ifdef CONFIG_SMP
 
 #include <linux/preempt.h>
@@ -53,9 +63,28 @@ extern void smp_cpus_done(unsigned int max_cpus);
  * Call a function on all other processors
  */
 int smp_call_function(void(*func)(void *info), void *info, int retry, int wait);
-
+int smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
+				int wait);
 int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
 				int retry, int wait);
+void __smp_call_function_single(int cpuid, struct call_single_data *data);
+
+/*
+ * Generic and arch helpers
+ */
+#ifdef CONFIG_USE_GENERIC_SMP_HELPERS
+void generic_smp_call_function_single_interrupt(void);
+void generic_smp_call_function_interrupt(void);
+void init_call_single_data(void);
+void ipi_call_lock(void);
+void ipi_call_unlock(void);
+void ipi_call_lock_irq(void);
+void ipi_call_unlock_irq(void);
+#else
+static inline void init_call_single_data(void)
+{
+}
+#endif
 
 /*
  * Call a function on all processors
@@ -112,7 +141,9 @@ static inline void smp_send_reschedule(int cpu) { }
 })
 #define smp_call_function_mask(mask, func, info, wait) \
 			(up_smp_call_function(func, info))
-
+static inline void init_call_single_data(void)
+{
+}
 #endif /* !SMP */
 
 /*
diff --git a/init/main.c b/init/main.c
index f7fb20021d48..1efcccff1bdb 100644
--- a/init/main.c
+++ b/init/main.c
@@ -31,6 +31,7 @@
 #include <linux/kernel_stat.h>
 #include <linux/start_kernel.h>
 #include <linux/security.h>
+#include <linux/smp.h>
 #include <linux/workqueue.h>
 #include <linux/profile.h>
 #include <linux/rcupdate.h>
@@ -779,6 +780,7 @@ static void __init do_pre_smp_initcalls(void)
 {
 	extern int spawn_ksoftirqd(void);
 
+	init_call_single_data();
 	migration_init();
 	spawn_ksoftirqd();
 	if (!nosoftlockup)
diff --git a/kernel/Makefile b/kernel/Makefile
index 1c9938addb9d..9fa57976f252 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -28,6 +28,7 @@ obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o
 obj-$(CONFIG_RT_MUTEX_TESTER) += rtmutex-tester.o
 obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o
 obj-$(CONFIG_SMP) += cpu.o spinlock.o
+obj-$(CONFIG_USE_GENERIC_SMP_HELPERS) += smp.o
 obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o
 obj-$(CONFIG_PROVE_LOCKING) += spinlock.o
 obj-$(CONFIG_UID16) += uid16.o
diff --git a/kernel/smp.c b/kernel/smp.c
new file mode 100644
index 000000000000..f77b75c027ad
--- /dev/null
+++ b/kernel/smp.c
@@ -0,0 +1,383 @@
+/*
+ * Generic helpers for smp ipi calls
+ *
+ * (C) Jens Axboe <jens.axboe@oracle.com> 2008
+ *
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/percpu.h>
+#include <linux/rcupdate.h>
+#include <linux/smp.h>
+
+static DEFINE_PER_CPU(struct call_single_queue, call_single_queue);
+static LIST_HEAD(call_function_queue);
+__cacheline_aligned_in_smp DEFINE_SPINLOCK(call_function_lock);
+
+enum {
+	CSD_FLAG_WAIT		= 0x01,
+	CSD_FLAG_ALLOC		= 0x02,
+};
+
+struct call_function_data {
+	struct call_single_data csd;
+	spinlock_t lock;
+	unsigned int refs;
+	cpumask_t cpumask;
+	struct rcu_head rcu_head;
+};
+
+struct call_single_queue {
+	struct list_head list;
+	spinlock_t lock;
+};
+
+void __cpuinit init_call_single_data(void)
+{
+	int i;
+
+	for_each_possible_cpu(i) {
+		struct call_single_queue *q = &per_cpu(call_single_queue, i);
+
+		spin_lock_init(&q->lock);
+		INIT_LIST_HEAD(&q->list);
+	}
+}
+
+static void csd_flag_wait(struct call_single_data *data)
+{
+	/* Wait for response */
+	do {
+		/*
+		 * We need to see the flags store in the IPI handler
+		 */
+		smp_mb();
+		if (!(data->flags & CSD_FLAG_WAIT))
+			break;
+		cpu_relax();
+	} while (1);
+}
+
+/*
+ * Insert a previously allocated call_single_data element for execution
+ * on the given CPU. data must already have ->func, ->info, and ->flags set.
+ */
+static void generic_exec_single(int cpu, struct call_single_data *data)
+{
+	struct call_single_queue *dst = &per_cpu(call_single_queue, cpu);
+	int wait = data->flags & CSD_FLAG_WAIT, ipi;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dst->lock, flags);
+	ipi = list_empty(&dst->list);
+	list_add_tail(&data->list, &dst->list);
+	spin_unlock_irqrestore(&dst->lock, flags);
+
+	if (ipi)
+		arch_send_call_function_single_ipi(cpu);
+
+	if (wait)
+		csd_flag_wait(data);
+}
+
+static void rcu_free_call_data(struct rcu_head *head)
+{
+	struct call_function_data *data;
+
+	data = container_of(head, struct call_function_data, rcu_head);
+
+	kfree(data);
+}
+
+/*
+ * Invoked by arch to handle an IPI for call function. Must be called with
+ * interrupts disabled.
+ */
+void generic_smp_call_function_interrupt(void)
+{
+	struct call_function_data *data;
+	int cpu = get_cpu();
+
+	/*
+	 * It's ok to use list_for_each_rcu() here even though we may delete
+	 * 'pos', since list_del_rcu() doesn't clear ->next
+	 */
+	rcu_read_lock();
+	list_for_each_entry_rcu(data, &call_function_queue, csd.list) {
+		int refs;
+
+		if (!cpu_isset(cpu, data->cpumask))
+			continue;
+
+		data->csd.func(data->csd.info);
+
+		spin_lock(&data->lock);
+		cpu_clear(cpu, data->cpumask);
+		WARN_ON(data->refs == 0);
+		data->refs--;
+		refs = data->refs;
+		spin_unlock(&data->lock);
+
+		if (refs)
+			continue;
+
+		spin_lock(&call_function_lock);
+		list_del_rcu(&data->csd.list);
+		spin_unlock(&call_function_lock);
+
+		if (data->csd.flags & CSD_FLAG_WAIT) {
+			/*
+			 * serialize stores to data with the flag clear
+			 * and wakeup
+			 */
+			smp_wmb();
+			data->csd.flags &= ~CSD_FLAG_WAIT;
+		} else
+			call_rcu(&data->rcu_head, rcu_free_call_data);
+	}
+	rcu_read_unlock();
+
+	put_cpu();
+}
+
+/*
+ * Invoked by arch to handle an IPI for call function single. Must be called
+ * from the arch with interrupts disabled.
+ */
+void generic_smp_call_function_single_interrupt(void)
+{
+	struct call_single_queue *q = &__get_cpu_var(call_single_queue);
+	LIST_HEAD(list);
+
+	/*
+	 * Need to see other stores to list head for checking whether
+	 * list is empty without holding q->lock
+	 */
+	smp_mb();
+	while (!list_empty(&q->list)) {
+		unsigned int data_flags;
+
+		spin_lock(&q->lock);
+		list_replace_init(&q->list, &list);
+		spin_unlock(&q->lock);
+
+		while (!list_empty(&list)) {
+			struct call_single_data *data;
+
+			data = list_entry(list.next, struct call_single_data,
+						list);
+			list_del(&data->list);
+
+			/*
+			 * 'data' can be invalid after this call if
+			 * flags == 0 (when called through
+			 * generic_exec_single(), so save them away before
+			 * making the call.
+			 */
+			data_flags = data->flags;
+
+			data->func(data->info);
+
+			if (data_flags & CSD_FLAG_WAIT) {
+				smp_wmb();
+				data->flags &= ~CSD_FLAG_WAIT;
+			} else if (data_flags & CSD_FLAG_ALLOC)
+				kfree(data);
+		}
+		/*
+		 * See comment on outer loop
+		 */
+		smp_mb();
+	}
+}
+
+/*
+ * smp_call_function_single - Run a function on a specific CPU
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @retry: Unused
+ * @wait: If true, wait until function has completed on other CPUs.
+ *
+ * Returns 0 on success, else a negative status code. Note that @wait
+ * will be implicitly turned on in case of allocation failures, since
+ * we fall back to on-stack allocation.
+ */
+int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
+			     int retry, int wait)
+{
+	struct call_single_data d;
+	unsigned long flags;
+	/* prevent preemption and reschedule on another processor */
+	int me = get_cpu();
+
+	/* Can deadlock when called with interrupts disabled */
+	WARN_ON(irqs_disabled());
+
+	if (cpu == me) {
+		local_irq_save(flags);
+		func(info);
+		local_irq_restore(flags);
+	} else {
+		struct call_single_data *data = NULL;
+
+		if (!wait) {
+			data = kmalloc(sizeof(*data), GFP_ATOMIC);
+			if (data)
+				data->flags = CSD_FLAG_ALLOC;
+		}
+		if (!data) {
+			data = &d;
+			data->flags = CSD_FLAG_WAIT;
+		}
+
+		data->func = func;
+		data->info = info;
+		generic_exec_single(cpu, data);
+	}
+
+	put_cpu();
+	return 0;
+}
+EXPORT_SYMBOL(smp_call_function_single);
+
+/**
+ * __smp_call_function_single(): Run a function on another CPU
+ * @cpu: The CPU to run on.
+ * @data: Pre-allocated and setup data structure
+ *
+ * Like smp_call_function_single(), but allow caller to pass in a pre-allocated
+ * data structure. Useful for embedding @data inside other structures, for
+ * instance.
+ *
+ */
+void __smp_call_function_single(int cpu, struct call_single_data *data)
+{
+	/* Can deadlock when called with interrupts disabled */
+	WARN_ON((data->flags & CSD_FLAG_WAIT) && irqs_disabled());
+
+	generic_exec_single(cpu, data);
+}
+
+/**
+ * smp_call_function_mask(): Run a function on a set of other CPUs.
+ * @mask: The set of cpus to run on.
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @wait: If true, wait (atomically) until function has completed on other CPUs.
+ *
+ * Returns 0 on success, else a negative status code.
+ *
+ * If @wait is true, then returns once @func has returned. Note that @wait
+ * will be implicitly turned on in case of allocation failures, since
+ * we fall back to on-stack allocation.
+ *
+ * You must not call this function with disabled interrupts or from a
+ * hardware interrupt handler or from a bottom half handler. Preemption
+ * must be disabled when calling this function.
+ */
+int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
+			   int wait)
+{
+	struct call_function_data d;
+	struct call_function_data *data = NULL;
+	cpumask_t allbutself;
+	unsigned long flags;
+	int cpu, num_cpus;
+
+	/* Can deadlock when called with interrupts disabled */
+	WARN_ON(irqs_disabled());
+
+	cpu = smp_processor_id();
+	allbutself = cpu_online_map;
+	cpu_clear(cpu, allbutself);
+	cpus_and(mask, mask, allbutself);
+	num_cpus = cpus_weight(mask);
+
+	/*
+	 * If zero CPUs, return. If just a single CPU, turn this request
+	 * into a targetted single call instead since it's faster.
+	 */
+	if (!num_cpus)
+		return 0;
+	else if (num_cpus == 1) {
+		cpu = first_cpu(mask);
+		return smp_call_function_single(cpu, func, info, 0, wait);
+	}
+
+	if (!wait) {
+		data = kmalloc(sizeof(*data), GFP_ATOMIC);
+		if (data)
+			data->csd.flags = CSD_FLAG_ALLOC;
+	}
+	if (!data) {
+		data = &d;
+		data->csd.flags = CSD_FLAG_WAIT;
+	}
+
+	spin_lock_init(&data->lock);
+	data->csd.func = func;
+	data->csd.info = info;
+	data->refs = num_cpus;
+	data->cpumask = mask;
+
+	spin_lock_irqsave(&call_function_lock, flags);
+	list_add_tail_rcu(&data->csd.list, &call_function_queue);
+	spin_unlock_irqrestore(&call_function_lock, flags);
+
+	/* Send a message to all CPUs in the map */
+	arch_send_call_function_ipi(mask);
+
+	/* optionally wait for the CPUs to complete */
+	if (wait)
+		csd_flag_wait(&data->csd);
+
+	return 0;
+}
+EXPORT_SYMBOL(smp_call_function_mask);
+
+/**
+ * smp_call_function(): Run a function on all other CPUs.
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @natomic: Unused
+ * @wait: If true, wait (atomically) until function has completed on other CPUs.
+ *
+ * Returns 0 on success, else a negative status code.
+ *
+ * If @wait is true, then returns once @func has returned; otherwise
+ * it returns just before the target cpu calls @func. In case of allocation
+ * failure, @wait will be implicitly turned on.
+ *
+ * You must not call this function with disabled interrupts or from a
+ * hardware interrupt handler or from a bottom half handler.
+ */
+int smp_call_function(void (*func)(void *), void *info, int natomic, int wait)
+{
+	int ret;
+
+	preempt_disable();
+	ret = smp_call_function_mask(cpu_online_map, func, info, wait);
+	preempt_enable();
+	return ret;
+}
+EXPORT_SYMBOL(smp_call_function);
+
+void ipi_call_lock(void)
+{
+	spin_lock(&call_function_lock);
+}
+
+void ipi_call_unlock(void)
+{
+	spin_unlock(&call_function_lock);
+}
+
+void ipi_call_lock_irq(void)
+{
+	spin_lock_irq(&call_function_lock);
+}
+
+void ipi_call_unlock_irq(void)
+{
+	spin_unlock_irq(&call_function_lock);
+}
-- 
cgit v1.2.3


From 8691e5a8f691cc2a4fda0651e8d307aaba0e7d68 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 6 Jun 2008 11:18:06 +0200
Subject: smp_call_function: get rid of the unused nonatomic/retry argument

It's never used and the comments refer to nonatomic and retry
interchangably. So get rid of it.

Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 arch/alpha/kernel/core_marvel.c      |  2 +-
 arch/alpha/kernel/smp.c              |  6 +++---
 arch/alpha/oprofile/common.c         |  6 +++---
 arch/arm/oprofile/op_model_mpcore.c  |  2 +-
 arch/arm/vfp/vfpmodule.c             |  2 +-
 arch/cris/arch-v32/kernel/smp.c      |  5 ++---
 arch/ia64/kernel/mca.c               |  2 +-
 arch/ia64/kernel/palinfo.c           |  2 +-
 arch/ia64/kernel/perfmon.c           |  2 +-
 arch/ia64/kernel/process.c           |  2 +-
 arch/ia64/kernel/smpboot.c           |  2 +-
 arch/ia64/kernel/uncached.c          |  5 ++---
 arch/ia64/sn/kernel/sn2/sn_hwperf.c  |  2 +-
 arch/m32r/kernel/smp.c               |  4 ++--
 arch/mips/kernel/smp.c               |  4 ++--
 arch/mips/mm/c-r4k.c                 | 18 +++++++++---------
 arch/mips/pmc-sierra/yosemite/prom.c |  2 +-
 arch/mips/sibyte/cfe/setup.c         |  2 +-
 arch/mips/sibyte/sb1250/prom.c       |  2 +-
 arch/powerpc/kernel/smp.c            |  2 +-
 arch/s390/appldata/appldata_base.c   |  4 ++--
 arch/s390/kernel/smp.c               | 16 ++++++----------
 arch/s390/kernel/time.c              |  4 ++--
 arch/sh/kernel/smp.c                 | 10 +++++-----
 arch/sparc64/kernel/smp.c            | 12 ++++--------
 arch/um/kernel/smp.c                 |  3 +--
 arch/x86/kernel/cpu/mtrr/main.c      |  4 ++--
 arch/x86/kernel/cpuid.c              |  2 +-
 arch/x86/kernel/ldt.c                |  2 +-
 arch/x86/kernel/nmi_32.c             |  2 +-
 arch/x86/kernel/nmi_64.c             |  2 +-
 arch/x86/kernel/smp.c                |  2 +-
 arch/x86/kernel/vsyscall_64.c        |  2 +-
 arch/x86/kvm/vmx.c                   |  2 +-
 arch/x86/kvm/x86.c                   |  2 +-
 arch/x86/lib/msr-on-cpu.c            |  8 ++++----
 arch/x86/mach-voyager/voyager_smp.c  |  2 +-
 arch/x86/xen/smp.c                   |  2 +-
 drivers/acpi/processor_idle.c        |  2 +-
 drivers/cpuidle/cpuidle.c            |  2 +-
 include/asm-alpha/smp.h              |  2 +-
 include/asm-sparc/smp.h              |  2 +-
 include/linux/smp.h                  |  8 ++++----
 kernel/smp.c                         |  6 ++----
 kernel/softirq.c                     |  2 +-
 kernel/time/tick-broadcast.c         |  2 +-
 net/core/flow.c                      |  2 +-
 net/iucv/iucv.c                      | 14 +++++++-------
 virt/kvm/kvm_main.c                  |  6 +++---
 49 files changed, 95 insertions(+), 108 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/core_marvel.c b/arch/alpha/kernel/core_marvel.c
index ced4aae8b804..04dcc5e5d4c1 100644
--- a/arch/alpha/kernel/core_marvel.c
+++ b/arch/alpha/kernel/core_marvel.c
@@ -662,7 +662,7 @@ __marvel_rtc_io(u8 b, unsigned long addr, int write)
 		if (smp_processor_id() != boot_cpuid)
 			smp_call_function_single(boot_cpuid,
 						 __marvel_access_rtc,
-						 &rtc_access, 1, 1);
+						 &rtc_access, 1);
 		else
 			__marvel_access_rtc(&rtc_access);
 #else
diff --git a/arch/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c
index 95c905be9154..44114c8dbb2a 100644
--- a/arch/alpha/kernel/smp.c
+++ b/arch/alpha/kernel/smp.c
@@ -710,7 +710,7 @@ flush_tlb_mm(struct mm_struct *mm)
 		}
 	}
 
-	if (smp_call_function(ipi_flush_tlb_mm, mm, 1, 1)) {
+	if (smp_call_function(ipi_flush_tlb_mm, mm, 1)) {
 		printk(KERN_CRIT "flush_tlb_mm: timed out\n");
 	}
 
@@ -763,7 +763,7 @@ flush_tlb_page(struct vm_area_struct *vma, unsigned long addr)
 	data.mm = mm;
 	data.addr = addr;
 
-	if (smp_call_function(ipi_flush_tlb_page, &data, 1, 1)) {
+	if (smp_call_function(ipi_flush_tlb_page, &data, 1)) {
 		printk(KERN_CRIT "flush_tlb_page: timed out\n");
 	}
 
@@ -815,7 +815,7 @@ flush_icache_user_range(struct vm_area_struct *vma, struct page *page,
 		}
 	}
 
-	if (smp_call_function(ipi_flush_icache_page, mm, 1, 1)) {
+	if (smp_call_function(ipi_flush_icache_page, mm, 1)) {
 		printk(KERN_CRIT "flush_icache_page: timed out\n");
 	}
 
diff --git a/arch/alpha/oprofile/common.c b/arch/alpha/oprofile/common.c
index 9fc0eeb4f0ab..7c3d5ec6ec67 100644
--- a/arch/alpha/oprofile/common.c
+++ b/arch/alpha/oprofile/common.c
@@ -65,7 +65,7 @@ op_axp_setup(void)
 	model->reg_setup(&reg, ctr, &sys);
 
 	/* Configure the registers on all cpus.  */
-	(void)smp_call_function(model->cpu_setup, &reg, 0, 1);
+	(void)smp_call_function(model->cpu_setup, &reg, 1);
 	model->cpu_setup(&reg);
 	return 0;
 }
@@ -86,7 +86,7 @@ op_axp_cpu_start(void *dummy)
 static int
 op_axp_start(void)
 {
-	(void)smp_call_function(op_axp_cpu_start, NULL, 0, 1);
+	(void)smp_call_function(op_axp_cpu_start, NULL, 1);
 	op_axp_cpu_start(NULL);
 	return 0;
 }
@@ -101,7 +101,7 @@ op_axp_cpu_stop(void *dummy)
 static void
 op_axp_stop(void)
 {
-	(void)smp_call_function(op_axp_cpu_stop, NULL, 0, 1);
+	(void)smp_call_function(op_axp_cpu_stop, NULL, 1);
 	op_axp_cpu_stop(NULL);
 }
 
diff --git a/arch/arm/oprofile/op_model_mpcore.c b/arch/arm/oprofile/op_model_mpcore.c
index 74fae6045650..4458705021e0 100644
--- a/arch/arm/oprofile/op_model_mpcore.c
+++ b/arch/arm/oprofile/op_model_mpcore.c
@@ -201,7 +201,7 @@ static int em_call_function(int (*fn)(void))
 	data.ret = 0;
 
 	preempt_disable();
-	smp_call_function(em_func, &data, 1, 1);
+	smp_call_function(em_func, &data, 1);
 	em_func(&data);
 	preempt_enable();
 
diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index 32455c633f1c..c0d2c9bb952b 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -352,7 +352,7 @@ static int __init vfp_init(void)
 	else if (vfpsid & FPSID_NODOUBLE) {
 		printk("no double precision support\n");
 	} else {
-		smp_call_function(vfp_enable, NULL, 1, 1);
+		smp_call_function(vfp_enable, NULL, 1);
 
 		VFP_arch = (vfpsid & FPSID_ARCH_MASK) >> FPSID_ARCH_BIT;  /* Extract the architecture version */
 		printk("implementor %02x architecture %d part %02x variant %x rev %x\n",
diff --git a/arch/cris/arch-v32/kernel/smp.c b/arch/cris/arch-v32/kernel/smp.c
index a9c3334e46c9..952a24b2f5a9 100644
--- a/arch/cris/arch-v32/kernel/smp.c
+++ b/arch/cris/arch-v32/kernel/smp.c
@@ -194,7 +194,7 @@ void stop_this_cpu(void* dummy)
 /* Other calls */
 void smp_send_stop(void)
 {
-	smp_call_function(stop_this_cpu, NULL, 1, 0);
+	smp_call_function(stop_this_cpu, NULL, 0);
 }
 
 int setup_profiling_timer(unsigned int multiplier)
@@ -316,8 +316,7 @@ int send_ipi(int vector, int wait, cpumask_t cpu_mask)
  * You must not call this function with disabled interrupts or from a
  * hardware interrupt handler or from a bottom half handler.
  */
-int smp_call_function(void (*func)(void *info), void *info,
-		      int nonatomic, int wait)
+int smp_call_function(void (*func)(void *info), void *info, int wait)
 {
 	cpumask_t cpu_mask = CPU_MASK_ALL;
 	struct call_data_struct data;
diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index 705176b434b3..9cd818cc7008 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -1881,7 +1881,7 @@ static int __cpuinit mca_cpu_callback(struct notifier_block *nfb,
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
 		smp_call_function_single(hotcpu, ia64_mca_cmc_vector_adjust,
-					 NULL, 1, 0);
+					 NULL, 0);
 		break;
 	}
 	return NOTIFY_OK;
diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c
index 9dc00f7fe10e..e5c57f413ca2 100644
--- a/arch/ia64/kernel/palinfo.c
+++ b/arch/ia64/kernel/palinfo.c
@@ -921,7 +921,7 @@ int palinfo_handle_smp(pal_func_cpu_u_t *f, char *page)
 
 
 	/* will send IPI to other CPU and wait for completion of remote call */
-	if ((ret=smp_call_function_single(f->req_cpu, palinfo_smp_call, &ptr, 0, 1))) {
+	if ((ret=smp_call_function_single(f->req_cpu, palinfo_smp_call, &ptr, 1))) {
 		printk(KERN_ERR "palinfo: remote CPU call from %d to %d on function %d: "
 		       "error %d\n", smp_processor_id(), f->req_cpu, f->func_id, ret);
 		return 0;
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 7714a97b0104..9baa48255c12 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -1820,7 +1820,7 @@ pfm_syswide_cleanup_other_cpu(pfm_context_t *ctx)
 	int ret;
 
 	DPRINT(("calling CPU%d for cleanup\n", ctx->ctx_cpu));
-	ret = smp_call_function_single(ctx->ctx_cpu, pfm_syswide_force_stop, ctx, 0, 1);
+	ret = smp_call_function_single(ctx->ctx_cpu, pfm_syswide_force_stop, ctx, 1);
 	DPRINT(("called CPU%d for cleanup ret=%d\n", ctx->ctx_cpu, ret));
 }
 #endif /* CONFIG_SMP */
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index a3a34b4eb038..fabaf08d9a69 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -286,7 +286,7 @@ void cpu_idle_wait(void)
 {
 	smp_mb();
 	/* kick all the CPUs so that they exit out of pm_idle */
-	smp_call_function(do_nothing, NULL, 0, 1);
+	smp_call_function(do_nothing, NULL, 1);
 }
 EXPORT_SYMBOL_GPL(cpu_idle_wait);
 
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index eaa1b6795a13..9d1d429c6c59 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -317,7 +317,7 @@ ia64_sync_itc (unsigned int master)
 
 	go[MASTER] = 1;
 
-	if (smp_call_function_single(master, sync_master, NULL, 1, 0) < 0) {
+	if (smp_call_function_single(master, sync_master, NULL, 0) < 0) {
 		printk(KERN_ERR "sync_itc: failed to get attention of CPU %u!\n", master);
 		return;
 	}
diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
index e77995a6e3ed..8eff8c1d40a6 100644
--- a/arch/ia64/kernel/uncached.c
+++ b/arch/ia64/kernel/uncached.c
@@ -123,8 +123,7 @@ static int uncached_add_chunk(struct uncached_pool *uc_pool, int nid)
 	status = ia64_pal_prefetch_visibility(PAL_VISIBILITY_PHYSICAL);
 	if (status == PAL_VISIBILITY_OK_REMOTE_NEEDED) {
 		atomic_set(&uc_pool->status, 0);
-		status = smp_call_function(uncached_ipi_visibility, uc_pool,
-					   0, 1);
+		status = smp_call_function(uncached_ipi_visibility, uc_pool, 1);
 		if (status || atomic_read(&uc_pool->status))
 			goto failed;
 	} else if (status != PAL_VISIBILITY_OK)
@@ -146,7 +145,7 @@ static int uncached_add_chunk(struct uncached_pool *uc_pool, int nid)
 	if (status != PAL_STATUS_SUCCESS)
 		goto failed;
 	atomic_set(&uc_pool->status, 0);
-	status = smp_call_function(uncached_ipi_mc_drain, uc_pool, 0, 1);
+	status = smp_call_function(uncached_ipi_mc_drain, uc_pool, 1);
 	if (status || atomic_read(&uc_pool->status))
 		goto failed;
 
diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
index 8cc0c4753d89..636588e7e068 100644
--- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c
+++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
@@ -629,7 +629,7 @@ static int sn_hwperf_op_cpu(struct sn_hwperf_op_info *op_info)
 		if (use_ipi) {
 			/* use an interprocessor interrupt to call SAL */
 			smp_call_function_single(cpu, sn_hwperf_call_sal,
-				op_info, 1, 1);
+				op_info, 1);
 		}
 		else {
 			/* migrate the task before calling SAL */ 
diff --git a/arch/m32r/kernel/smp.c b/arch/m32r/kernel/smp.c
index 74eb7bcd5a40..7577f971ea4e 100644
--- a/arch/m32r/kernel/smp.c
+++ b/arch/m32r/kernel/smp.c
@@ -212,7 +212,7 @@ void smp_flush_tlb_all(void)
 	local_irq_save(flags);
 	__flush_tlb_all();
 	local_irq_restore(flags);
-	smp_call_function(flush_tlb_all_ipi, NULL, 1, 1);
+	smp_call_function(flush_tlb_all_ipi, NULL, 1);
 	preempt_enable();
 }
 
@@ -505,7 +505,7 @@ void smp_invalidate_interrupt(void)
  *==========================================================================*/
 void smp_send_stop(void)
 {
-	smp_call_function(stop_this_cpu, NULL, 1, 0);
+	smp_call_function(stop_this_cpu, NULL, 0);
 }
 
 /*==========================================================================*
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index c75b26cb61df..7a9ae830be86 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -167,7 +167,7 @@ static void stop_this_cpu(void *dummy)
 
 void smp_send_stop(void)
 {
-	smp_call_function(stop_this_cpu, NULL, 1, 0);
+	smp_call_function(stop_this_cpu, NULL, 0);
 }
 
 void __init smp_cpus_done(unsigned int max_cpus)
@@ -266,7 +266,7 @@ static void flush_tlb_mm_ipi(void *mm)
 static inline void smp_on_other_tlbs(void (*func) (void *info), void *info)
 {
 #ifndef CONFIG_MIPS_MT_SMTC
-	smp_call_function(func, info, 1, 1);
+	smp_call_function(func, info, 1);
 #endif
 }
 
diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 27096751ddce..71df3390c07b 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -43,12 +43,12 @@
  *    primary cache.
  */
 static inline void r4k_on_each_cpu(void (*func) (void *info), void *info,
-                                   int retry, int wait)
+                                   int wait)
 {
 	preempt_disable();
 
 #if !defined(CONFIG_MIPS_MT_SMP) && !defined(CONFIG_MIPS_MT_SMTC)
-	smp_call_function(func, info, retry, wait);
+	smp_call_function(func, info, wait);
 #endif
 	func(info);
 	preempt_enable();
@@ -350,7 +350,7 @@ static inline void local_r4k___flush_cache_all(void * args)
 
 static void r4k___flush_cache_all(void)
 {
-	r4k_on_each_cpu(local_r4k___flush_cache_all, NULL, 1, 1);
+	r4k_on_each_cpu(local_r4k___flush_cache_all, NULL, 1);
 }
 
 static inline int has_valid_asid(const struct mm_struct *mm)
@@ -397,7 +397,7 @@ static void r4k_flush_cache_range(struct vm_area_struct *vma,
 	int exec = vma->vm_flags & VM_EXEC;
 
 	if (cpu_has_dc_aliases || (exec && !cpu_has_ic_fills_f_dc))
-		r4k_on_each_cpu(local_r4k_flush_cache_range, vma, 1, 1);
+		r4k_on_each_cpu(local_r4k_flush_cache_range, vma, 1);
 }
 
 static inline void local_r4k_flush_cache_mm(void * args)
@@ -429,7 +429,7 @@ static void r4k_flush_cache_mm(struct mm_struct *mm)
 	if (!cpu_has_dc_aliases)
 		return;
 
-	r4k_on_each_cpu(local_r4k_flush_cache_mm, mm, 1, 1);
+	r4k_on_each_cpu(local_r4k_flush_cache_mm, mm, 1);
 }
 
 struct flush_cache_page_args {
@@ -521,7 +521,7 @@ static void r4k_flush_cache_page(struct vm_area_struct *vma,
 	args.addr = addr;
 	args.pfn = pfn;
 
-	r4k_on_each_cpu(local_r4k_flush_cache_page, &args, 1, 1);
+	r4k_on_each_cpu(local_r4k_flush_cache_page, &args, 1);
 }
 
 static inline void local_r4k_flush_data_cache_page(void * addr)
@@ -535,7 +535,7 @@ static void r4k_flush_data_cache_page(unsigned long addr)
 		local_r4k_flush_data_cache_page((void *)addr);
 	else
 		r4k_on_each_cpu(local_r4k_flush_data_cache_page, (void *) addr,
-			        1, 1);
+			        1);
 }
 
 struct flush_icache_range_args {
@@ -571,7 +571,7 @@ static void r4k_flush_icache_range(unsigned long start, unsigned long end)
 	args.start = start;
 	args.end = end;
 
-	r4k_on_each_cpu(local_r4k_flush_icache_range, &args, 1, 1);
+	r4k_on_each_cpu(local_r4k_flush_icache_range, &args, 1);
 	instruction_hazard();
 }
 
@@ -672,7 +672,7 @@ static void local_r4k_flush_cache_sigtramp(void * arg)
 
 static void r4k_flush_cache_sigtramp(unsigned long addr)
 {
-	r4k_on_each_cpu(local_r4k_flush_cache_sigtramp, (void *) addr, 1, 1);
+	r4k_on_each_cpu(local_r4k_flush_cache_sigtramp, (void *) addr, 1);
 }
 
 static void r4k_flush_icache_all(void)
diff --git a/arch/mips/pmc-sierra/yosemite/prom.c b/arch/mips/pmc-sierra/yosemite/prom.c
index 35dc435846a6..cf4c868715ac 100644
--- a/arch/mips/pmc-sierra/yosemite/prom.c
+++ b/arch/mips/pmc-sierra/yosemite/prom.c
@@ -64,7 +64,7 @@ static void prom_exit(void)
 #ifdef CONFIG_SMP
 	if (smp_processor_id())
 		/* CPU 1 */
-		smp_call_function(prom_cpu0_exit, NULL, 1, 1);
+		smp_call_function(prom_cpu0_exit, NULL, 1);
 #endif
 	prom_cpu0_exit(NULL);
 }
diff --git a/arch/mips/sibyte/cfe/setup.c b/arch/mips/sibyte/cfe/setup.c
index 33fce826f8bf..fd9604d5555a 100644
--- a/arch/mips/sibyte/cfe/setup.c
+++ b/arch/mips/sibyte/cfe/setup.c
@@ -74,7 +74,7 @@ static void __noreturn cfe_linux_exit(void *arg)
 		if (!reboot_smp) {
 			/* Get CPU 0 to do the cfe_exit */
 			reboot_smp = 1;
-			smp_call_function(cfe_linux_exit, arg, 1, 0);
+			smp_call_function(cfe_linux_exit, arg, 0);
 		}
 	} else {
 		printk("Passing control back to CFE...\n");
diff --git a/arch/mips/sibyte/sb1250/prom.c b/arch/mips/sibyte/sb1250/prom.c
index cf8f6b3de86c..65b1af66b674 100644
--- a/arch/mips/sibyte/sb1250/prom.c
+++ b/arch/mips/sibyte/sb1250/prom.c
@@ -66,7 +66,7 @@ static void prom_linux_exit(void)
 {
 #ifdef CONFIG_SMP
 	if (smp_processor_id()) {
-		smp_call_function(prom_cpu0_exit, NULL, 1, 1);
+		smp_call_function(prom_cpu0_exit, NULL, 1);
 	}
 #endif
 	while(1);
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 37a5ab410dcc..5191b46a611e 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -168,7 +168,7 @@ static void stop_this_cpu(void *dummy)
 
 void smp_send_stop(void)
 {
-	smp_call_function(stop_this_cpu, NULL, 0, 0);
+	smp_call_function(stop_this_cpu, NULL, 0);
 }
 
 extern struct gettimeofday_struct do_gtod;
diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c
index ad40729bec3d..837a3b3e7759 100644
--- a/arch/s390/appldata/appldata_base.c
+++ b/arch/s390/appldata/appldata_base.c
@@ -209,7 +209,7 @@ __appldata_vtimer_setup(int cmd)
 			per_cpu(appldata_timer, i).expires = per_cpu_interval;
 			smp_call_function_single(i, add_virt_timer_periodic,
 						 &per_cpu(appldata_timer, i),
-						 0, 1);
+						 1);
 		}
 		appldata_timer_active = 1;
 		P_INFO("Monitoring timer started.\n");
@@ -236,7 +236,7 @@ __appldata_vtimer_setup(int cmd)
 			args.timer = &per_cpu(appldata_timer, i);
 			args.expires = per_cpu_interval;
 			smp_call_function_single(i, __appldata_mod_vtimer_wrap,
-						 &args, 0, 1);
+						 &args, 1);
 		}
 	}
 }
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 5d4fa4b1c74c..276b105fb2a4 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -109,7 +109,7 @@ static void do_call_function(void)
 }
 
 static void __smp_call_function_map(void (*func) (void *info), void *info,
-				    int nonatomic, int wait, cpumask_t map)
+				    int wait, cpumask_t map)
 {
 	struct call_data_struct data;
 	int cpu, local = 0;
@@ -162,7 +162,6 @@ out:
  * smp_call_function:
  * @func: the function to run; this must be fast and non-blocking
  * @info: an arbitrary pointer to pass to the function
- * @nonatomic: unused
  * @wait: if true, wait (atomically) until function has completed on other CPUs
  *
  * Run a function on all other CPUs.
@@ -170,15 +169,14 @@ out:
  * You must not call this function with disabled interrupts, from a
  * hardware interrupt handler or from a bottom half.
  */
-int smp_call_function(void (*func) (void *info), void *info, int nonatomic,
-		      int wait)
+int smp_call_function(void (*func) (void *info), void *info, int wait)
 {
 	cpumask_t map;
 
 	spin_lock(&call_lock);
 	map = cpu_online_map;
 	cpu_clear(smp_processor_id(), map);
-	__smp_call_function_map(func, info, nonatomic, wait, map);
+	__smp_call_function_map(func, info, wait, map);
 	spin_unlock(&call_lock);
 	return 0;
 }
@@ -189,7 +187,6 @@ EXPORT_SYMBOL(smp_call_function);
  * @cpu: the CPU where func should run
  * @func: the function to run; this must be fast and non-blocking
  * @info: an arbitrary pointer to pass to the function
- * @nonatomic: unused
  * @wait: if true, wait (atomically) until function has completed on other CPUs
  *
  * Run a function on one processor.
@@ -198,11 +195,10 @@ EXPORT_SYMBOL(smp_call_function);
  * hardware interrupt handler or from a bottom half.
  */
 int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
-			     int nonatomic, int wait)
+			     int wait)
 {
 	spin_lock(&call_lock);
-	__smp_call_function_map(func, info, nonatomic, wait,
-				cpumask_of_cpu(cpu));
+	__smp_call_function_map(func, info, wait, cpumask_of_cpu(cpu));
 	spin_unlock(&call_lock);
 	return 0;
 }
@@ -228,7 +224,7 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
 {
 	spin_lock(&call_lock);
 	cpu_clear(smp_processor_id(), mask);
-	__smp_call_function_map(func, info, 0, wait, mask);
+	__smp_call_function_map(func, info, wait, mask);
 	spin_unlock(&call_lock);
 	return 0;
 }
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 7aec676fefd5..bf7bf2c2236a 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -690,7 +690,7 @@ static int etr_sync_clock(struct etr_aib *aib, int port)
 	 */
 	memset(&etr_sync, 0, sizeof(etr_sync));
 	preempt_disable();
-	smp_call_function(etr_sync_cpu_start, NULL, 0, 0);
+	smp_call_function(etr_sync_cpu_start, NULL, 0);
 	local_irq_disable();
 	etr_enable_sync_clock();
 
@@ -729,7 +729,7 @@ static int etr_sync_clock(struct etr_aib *aib, int port)
 		rc = -EAGAIN;
 	}
 	local_irq_enable();
-	smp_call_function(etr_sync_cpu_end,NULL,0,0);
+	smp_call_function(etr_sync_cpu_end,NULL,0);
 	preempt_enable();
 	return rc;
 }
diff --git a/arch/sh/kernel/smp.c b/arch/sh/kernel/smp.c
index 2ed8dceb297b..71781ba2675b 100644
--- a/arch/sh/kernel/smp.c
+++ b/arch/sh/kernel/smp.c
@@ -168,7 +168,7 @@ static void stop_this_cpu(void *unused)
 
 void smp_send_stop(void)
 {
-	smp_call_function(stop_this_cpu, 0, 1, 0);
+	smp_call_function(stop_this_cpu, 0, 0);
 }
 
 void arch_send_call_function_ipi(cpumask_t mask)
@@ -223,7 +223,7 @@ void flush_tlb_mm(struct mm_struct *mm)
 	preempt_disable();
 
 	if ((atomic_read(&mm->mm_users) != 1) || (current->mm != mm)) {
-		smp_call_function(flush_tlb_mm_ipi, (void *)mm, 1, 1);
+		smp_call_function(flush_tlb_mm_ipi, (void *)mm, 1);
 	} else {
 		int i;
 		for (i = 0; i < num_online_cpus(); i++)
@@ -260,7 +260,7 @@ void flush_tlb_range(struct vm_area_struct *vma,
 		fd.vma = vma;
 		fd.addr1 = start;
 		fd.addr2 = end;
-		smp_call_function(flush_tlb_range_ipi, (void *)&fd, 1, 1);
+		smp_call_function(flush_tlb_range_ipi, (void *)&fd, 1);
 	} else {
 		int i;
 		for (i = 0; i < num_online_cpus(); i++)
@@ -303,7 +303,7 @@ void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
 
 		fd.vma = vma;
 		fd.addr1 = page;
-		smp_call_function(flush_tlb_page_ipi, (void *)&fd, 1, 1);
+		smp_call_function(flush_tlb_page_ipi, (void *)&fd, 1);
 	} else {
 		int i;
 		for (i = 0; i < num_online_cpus(); i++)
@@ -327,6 +327,6 @@ void flush_tlb_one(unsigned long asid, unsigned long vaddr)
 	fd.addr1 = asid;
 	fd.addr2 = vaddr;
 
-	smp_call_function(flush_tlb_one_ipi, (void *)&fd, 1, 1);
+	smp_call_function(flush_tlb_one_ipi, (void *)&fd, 1);
 	local_flush_tlb_one(asid, vaddr);
 }
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index b82d017a1744..c099d96f1239 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -807,7 +807,6 @@ extern unsigned long xcall_call_function;
  * smp_call_function(): Run a function on all other CPUs.
  * @func: The function to run. This must be fast and non-blocking.
  * @info: An arbitrary pointer to pass to the function.
- * @nonatomic: currently unused.
  * @wait: If true, wait (atomically) until function has completed on other CPUs.
  *
  * Returns 0 on success, else a negative status code. Does not return until
@@ -817,8 +816,7 @@ extern unsigned long xcall_call_function;
  * hardware interrupt handler or from a bottom half handler.
  */
 static int sparc64_smp_call_function_mask(void (*func)(void *info), void *info,
-					  int nonatomic, int wait,
-					  cpumask_t mask)
+					  int wait, cpumask_t mask)
 {
 	struct call_data_struct data;
 	int cpus;
@@ -853,11 +851,9 @@ out_unlock:
 	return 0;
 }
 
-int smp_call_function(void (*func)(void *info), void *info,
-		      int nonatomic, int wait)
+int smp_call_function(void (*func)(void *info), void *info, int wait)
 {
-	return sparc64_smp_call_function_mask(func, info, nonatomic, wait,
-						cpu_online_map);
+	return sparc64_smp_call_function_mask(func, info, wait, cpu_online_map);
 }
 
 void smp_call_function_client(int irq, struct pt_regs *regs)
@@ -894,7 +890,7 @@ static void tsb_sync(void *info)
 
 void smp_tsb_sync(struct mm_struct *mm)
 {
-	sparc64_smp_call_function_mask(tsb_sync, mm, 0, 1, mm->cpu_vm_mask);
+	sparc64_smp_call_function_mask(tsb_sync, mm, 1, mm->cpu_vm_mask);
 }
 
 extern unsigned long xcall_flush_tlb_mm;
diff --git a/arch/um/kernel/smp.c b/arch/um/kernel/smp.c
index e1062ec36d40..be2d50c3aa95 100644
--- a/arch/um/kernel/smp.c
+++ b/arch/um/kernel/smp.c
@@ -214,8 +214,7 @@ void smp_call_function_slave(int cpu)
 	atomic_inc(&scf_finished);
 }
 
-int smp_call_function(void (*_func)(void *info), void *_info, int nonatomic,
-		      int wait)
+int smp_call_function(void (*_func)(void *info), void *_info, int wait)
 {
 	int cpus = num_online_cpus() - 1;
 	int i;
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 6a1e278d9323..290652cefddb 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -222,7 +222,7 @@ static void set_mtrr(unsigned int reg, unsigned long base,
 	atomic_set(&data.gate,0);
 
 	/*  Start the ball rolling on other CPUs  */
-	if (smp_call_function(ipi_handler, &data, 1, 0) != 0)
+	if (smp_call_function(ipi_handler, &data, 0) != 0)
 		panic("mtrr: timed out waiting for other CPUs\n");
 
 	local_irq_save(flags);
@@ -822,7 +822,7 @@ void mtrr_ap_init(void)
  */
 void mtrr_save_state(void)
 {
-	smp_call_function_single(0, mtrr_save_fixed_ranges, NULL, 1, 1);
+	smp_call_function_single(0, mtrr_save_fixed_ranges, NULL, 1);
 }
 
 static int __init mtrr_init_finialize(void)
diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c
index daff52a62248..336dd43c9158 100644
--- a/arch/x86/kernel/cpuid.c
+++ b/arch/x86/kernel/cpuid.c
@@ -95,7 +95,7 @@ static ssize_t cpuid_read(struct file *file, char __user *buf,
 	for (; count; count -= 16) {
 		cmd.eax = pos;
 		cmd.ecx = pos >> 32;
-		smp_call_function_single(cpu, cpuid_smp_cpuid, &cmd, 1, 1);
+		smp_call_function_single(cpu, cpuid_smp_cpuid, &cmd, 1);
 		if (copy_to_user(tmp, &cmd, 16))
 			return -EFAULT;
 		tmp += 16;
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index 0224c3637c73..cb0a6398c64b 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -68,7 +68,7 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
 		load_LDT(pc);
 		mask = cpumask_of_cpu(smp_processor_id());
 		if (!cpus_equal(current->mm->cpu_vm_mask, mask))
-			smp_call_function(flush_ldt, NULL, 1, 1);
+			smp_call_function(flush_ldt, NULL, 1);
 		preempt_enable();
 #else
 		load_LDT(pc);
diff --git a/arch/x86/kernel/nmi_32.c b/arch/x86/kernel/nmi_32.c
index 84160f74eeb0..5562dab0bd20 100644
--- a/arch/x86/kernel/nmi_32.c
+++ b/arch/x86/kernel/nmi_32.c
@@ -87,7 +87,7 @@ int __init check_nmi_watchdog(void)
 
 #ifdef CONFIG_SMP
 	if (nmi_watchdog == NMI_LOCAL_APIC)
-		smp_call_function(nmi_cpu_busy, (void *)&endflag, 0, 0);
+		smp_call_function(nmi_cpu_busy, (void *)&endflag, 0);
 #endif
 
 	for_each_possible_cpu(cpu)
diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c
index 5a29ded994fa..2f1e4f503c9e 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -96,7 +96,7 @@ int __init check_nmi_watchdog(void)
 
 #ifdef CONFIG_SMP
 	if (nmi_watchdog == NMI_LOCAL_APIC)
-		smp_call_function(nmi_cpu_busy, (void *)&endflag, 0, 0);
+		smp_call_function(nmi_cpu_busy, (void *)&endflag, 0);
 #endif
 
 	for (cpu = 0; cpu < NR_CPUS; cpu++)
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 575aa3d7248a..56546e8a13ac 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -164,7 +164,7 @@ static void native_smp_send_stop(void)
 	if (reboot_force)
 		return;
 
-	smp_call_function(stop_this_cpu, NULL, 0, 0);
+	smp_call_function(stop_this_cpu, NULL, 0);
 	local_irq_save(flags);
 	disable_local_APIC();
 	local_irq_restore(flags);
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index 61efa2f7d564..0a03d57f9b3b 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -278,7 +278,7 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg)
 {
 	long cpu = (long)arg;
 	if (action == CPU_ONLINE || action == CPU_ONLINE_FROZEN)
-		smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1);
+		smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 1);
 	return NOTIFY_DONE;
 }
 
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 540e95179074..5534fe59b5fc 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -335,7 +335,7 @@ static void vcpu_clear(struct vcpu_vmx *vmx)
 {
 	if (vmx->vcpu.cpu == -1)
 		return;
-	smp_call_function_single(vmx->vcpu.cpu, __vcpu_clear, vmx, 0, 1);
+	smp_call_function_single(vmx->vcpu.cpu, __vcpu_clear, vmx, 1);
 	vmx->launched = 0;
 }
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 63a77caa59f1..0faa2546b1cd 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -4044,6 +4044,6 @@ void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
 	 * So need not to call smp_call_function_single() in that case.
 	 */
 	if (vcpu->guest_mode && vcpu->cpu != cpu)
-		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
+		smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0);
 	put_cpu();
 }
diff --git a/arch/x86/lib/msr-on-cpu.c b/arch/x86/lib/msr-on-cpu.c
index 57d043fa893e..d5a2b39f882b 100644
--- a/arch/x86/lib/msr-on-cpu.c
+++ b/arch/x86/lib/msr-on-cpu.c
@@ -30,10 +30,10 @@ static int _rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h, int safe)
 
 	rv.msr_no = msr_no;
 	if (safe) {
-		smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __rdmsr_safe_on_cpu, &rv, 1);
 		err = rv.err;
 	} else {
-		smp_call_function_single(cpu, __rdmsr_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __rdmsr_on_cpu, &rv, 1);
 	}
 	*l = rv.l;
 	*h = rv.h;
@@ -64,10 +64,10 @@ static int _wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h, int safe)
 	rv.l = l;
 	rv.h = h;
 	if (safe) {
-		smp_call_function_single(cpu, __wrmsr_safe_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __wrmsr_safe_on_cpu, &rv, 1);
 		err = rv.err;
 	} else {
-		smp_call_function_single(cpu, __wrmsr_on_cpu, &rv, 0, 1);
+		smp_call_function_single(cpu, __wrmsr_on_cpu, &rv, 1);
 	}
 
 	return err;
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index cb34407a9930..04f596eab749 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -1113,7 +1113,7 @@ int safe_smp_processor_id(void)
 /* broadcast a halt to all other CPUs */
 static void voyager_smp_send_stop(void)
 {
-	smp_call_function(smp_stop_cpu_function, NULL, 1, 1);
+	smp_call_function(smp_stop_cpu_function, NULL, 1);
 }
 
 /* this function is triggered in time.c when a clock tick fires
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index b3786e749b8e..a1651d029ea8 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -331,7 +331,7 @@ static void stop_self(void *v)
 
 void xen_smp_send_stop(void)
 {
-	smp_call_function(stop_self, NULL, 0, 0);
+	smp_call_function(stop_self, NULL, 0);
 }
 
 void xen_smp_send_reschedule(int cpu)
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 556ee1585192..4976e5db2b3f 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -1339,7 +1339,7 @@ static void smp_callback(void *v)
 static int acpi_processor_latency_notify(struct notifier_block *b,
 		unsigned long l, void *v)
 {
-	smp_call_function(smp_callback, NULL, 0, 1);
+	smp_call_function(smp_callback, NULL, 1);
 	return NOTIFY_OK;
 }
 
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 23554b676d6e..5405769020a1 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -340,7 +340,7 @@ static void smp_callback(void *v)
 static int cpuidle_latency_notify(struct notifier_block *b,
 		unsigned long l, void *v)
 {
-	smp_call_function(smp_callback, NULL, 0, 1);
+	smp_call_function(smp_callback, NULL, 1);
 	return NOTIFY_OK;
 }
 
diff --git a/include/asm-alpha/smp.h b/include/asm-alpha/smp.h
index 2f60a362d75e..544c69af8168 100644
--- a/include/asm-alpha/smp.h
+++ b/include/asm-alpha/smp.h
@@ -53,7 +53,7 @@ extern void arch_send_call_function_ipi(cpumask_t mask);
 #else /* CONFIG_SMP */
 
 #define hard_smp_processor_id()		0
-#define smp_call_function_on_cpu(func,info,retry,wait,cpu)    ({ 0; })
+#define smp_call_function_on_cpu(func,info,wait,cpu)    ({ 0; })
 
 #endif /* CONFIG_SMP */
 
diff --git a/include/asm-sparc/smp.h b/include/asm-sparc/smp.h
index e6d561599726..b61e74bea06a 100644
--- a/include/asm-sparc/smp.h
+++ b/include/asm-sparc/smp.h
@@ -72,7 +72,7 @@ static inline void xc5(smpfunc_t func, unsigned long arg1, unsigned long arg2,
 			   unsigned long arg3, unsigned long arg4, unsigned long arg5)
 { smp_cross_call(func, arg1, arg2, arg3, arg4, arg5); }
 
-static inline int smp_call_function(void (*func)(void *info), void *info, int nonatomic, int wait)
+static inline int smp_call_function(void (*func)(void *info), void *info, int wait)
 {
 	xc1((smpfunc_t)func, (unsigned long)info);
 	return 0;
diff --git a/include/linux/smp.h b/include/linux/smp.h
index eac3e062250f..338cad1b9548 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -62,11 +62,11 @@ extern void smp_cpus_done(unsigned int max_cpus);
 /*
  * Call a function on all other processors
  */
-int smp_call_function(void(*func)(void *info), void *info, int retry, int wait);
+int smp_call_function(void(*func)(void *info), void *info, int wait);
 int smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
 				int wait);
 int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
-				int retry, int wait);
+				int wait);
 void __smp_call_function_single(int cpuid, struct call_single_data *data);
 
 /*
@@ -119,7 +119,7 @@ static inline int up_smp_call_function(void (*func)(void *), void *info)
 {
 	return 0;
 }
-#define smp_call_function(func, info, retry, wait) \
+#define smp_call_function(func, info, wait) \
 			(up_smp_call_function(func, info))
 #define on_each_cpu(func,info,retry,wait)	\
 	({					\
@@ -131,7 +131,7 @@ static inline int up_smp_call_function(void (*func)(void *), void *info)
 static inline void smp_send_reschedule(int cpu) { }
 #define num_booting_cpus()			1
 #define smp_prepare_boot_cpu()			do {} while (0)
-#define smp_call_function_single(cpuid, func, info, retry, wait) \
+#define smp_call_function_single(cpuid, func, info, wait) \
 ({ \
 	WARN_ON(cpuid != 0);	\
 	local_irq_disable();	\
diff --git a/kernel/smp.c b/kernel/smp.c
index f77b75c027ad..7e0432a4a0e2 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -195,7 +195,6 @@ void generic_smp_call_function_single_interrupt(void)
  * smp_call_function_single - Run a function on a specific CPU
  * @func: The function to run. This must be fast and non-blocking.
  * @info: An arbitrary pointer to pass to the function.
- * @retry: Unused
  * @wait: If true, wait until function has completed on other CPUs.
  *
  * Returns 0 on success, else a negative status code. Note that @wait
@@ -203,7 +202,7 @@ void generic_smp_call_function_single_interrupt(void)
  * we fall back to on-stack allocation.
  */
 int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
-			     int retry, int wait)
+			     int wait)
 {
 	struct call_single_data d;
 	unsigned long flags;
@@ -339,7 +338,6 @@ EXPORT_SYMBOL(smp_call_function_mask);
  * smp_call_function(): Run a function on all other CPUs.
  * @func: The function to run. This must be fast and non-blocking.
  * @info: An arbitrary pointer to pass to the function.
- * @natomic: Unused
  * @wait: If true, wait (atomically) until function has completed on other CPUs.
  *
  * Returns 0 on success, else a negative status code.
@@ -351,7 +349,7 @@ EXPORT_SYMBOL(smp_call_function_mask);
  * You must not call this function with disabled interrupts or from a
  * hardware interrupt handler or from a bottom half handler.
  */
-int smp_call_function(void (*func)(void *), void *info, int natomic, int wait)
+int smp_call_function(void (*func)(void *), void *info, int wait)
 {
 	int ret;
 
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 36e061740047..d73afb4764ef 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -679,7 +679,7 @@ int on_each_cpu(void (*func) (void *info), void *info, int retry, int wait)
 	int ret = 0;
 
 	preempt_disable();
-	ret = smp_call_function(func, info, retry, wait);
+	ret = smp_call_function(func, info, wait);
 	local_irq_disable();
 	func(info);
 	local_irq_enable();
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 57a1f02e5ec0..75e718539dcb 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -266,7 +266,7 @@ void tick_broadcast_on_off(unsigned long reason, int *oncpu)
 		       "offline CPU #%d\n", *oncpu);
 	else
 		smp_call_function_single(*oncpu, tick_do_broadcast_on_off,
-					 &reason, 1, 1);
+					 &reason, 1);
 }
 
 /*
diff --git a/net/core/flow.c b/net/core/flow.c
index 19991175fdeb..5cf81052d044 100644
--- a/net/core/flow.c
+++ b/net/core/flow.c
@@ -298,7 +298,7 @@ void flow_cache_flush(void)
 	init_completion(&info.completion);
 
 	local_bh_disable();
-	smp_call_function(flow_cache_flush_per_cpu, &info, 1, 0);
+	smp_call_function(flow_cache_flush_per_cpu, &info, 0);
 	flow_cache_flush_tasklet((unsigned long)&info);
 	local_bh_enable();
 
diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c
index 918970762131..94d5a45c3a57 100644
--- a/net/iucv/iucv.c
+++ b/net/iucv/iucv.c
@@ -480,7 +480,7 @@ static void iucv_setmask_mp(void)
 		if (cpu_isset(cpu, iucv_buffer_cpumask) &&
 		    !cpu_isset(cpu, iucv_irq_cpumask))
 			smp_call_function_single(cpu, iucv_allow_cpu,
-						 NULL, 0, 1);
+						 NULL, 1);
 	preempt_enable();
 }
 
@@ -498,7 +498,7 @@ static void iucv_setmask_up(void)
 	cpumask = iucv_irq_cpumask;
 	cpu_clear(first_cpu(iucv_irq_cpumask), cpumask);
 	for_each_cpu_mask(cpu, cpumask)
-		smp_call_function_single(cpu, iucv_block_cpu, NULL, 0, 1);
+		smp_call_function_single(cpu, iucv_block_cpu, NULL, 1);
 }
 
 /**
@@ -523,7 +523,7 @@ static int iucv_enable(void)
 	rc = -EIO;
 	preempt_disable();
 	for_each_online_cpu(cpu)
-		smp_call_function_single(cpu, iucv_declare_cpu, NULL, 0, 1);
+		smp_call_function_single(cpu, iucv_declare_cpu, NULL, 1);
 	preempt_enable();
 	if (cpus_empty(iucv_buffer_cpumask))
 		/* No cpu could declare an iucv buffer. */
@@ -580,7 +580,7 @@ static int __cpuinit iucv_cpu_notify(struct notifier_block *self,
 	case CPU_ONLINE_FROZEN:
 	case CPU_DOWN_FAILED:
 	case CPU_DOWN_FAILED_FROZEN:
-		smp_call_function_single(cpu, iucv_declare_cpu, NULL, 0, 1);
+		smp_call_function_single(cpu, iucv_declare_cpu, NULL, 1);
 		break;
 	case CPU_DOWN_PREPARE:
 	case CPU_DOWN_PREPARE_FROZEN:
@@ -589,10 +589,10 @@ static int __cpuinit iucv_cpu_notify(struct notifier_block *self,
 		if (cpus_empty(cpumask))
 			/* Can't offline last IUCV enabled cpu. */
 			return NOTIFY_BAD;
-		smp_call_function_single(cpu, iucv_retrieve_cpu, NULL, 0, 1);
+		smp_call_function_single(cpu, iucv_retrieve_cpu, NULL, 1);
 		if (cpus_empty(iucv_irq_cpumask))
 			smp_call_function_single(first_cpu(iucv_buffer_cpumask),
-						 iucv_allow_cpu, NULL, 0, 1);
+						 iucv_allow_cpu, NULL, 1);
 		break;
 	}
 	return NOTIFY_OK;
@@ -652,7 +652,7 @@ static void iucv_cleanup_queue(void)
 	 * pending interrupts force them to the work queue by calling
 	 * an empty function on all cpus.
 	 */
-	smp_call_function(__iucv_cleanup_queue, NULL, 0, 1);
+	smp_call_function(__iucv_cleanup_queue, NULL, 1);
 	spin_lock_irq(&iucv_queue_lock);
 	list_for_each_entry_safe(p, n, &iucv_task_queue, list) {
 		/* Remove stale work items from the task queue. */
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 2d29e260da3d..ea1f595f8a87 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1266,12 +1266,12 @@ static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val,
 	case CPU_UP_CANCELED:
 		printk(KERN_INFO "kvm: disabling virtualization on CPU%d\n",
 		       cpu);
-		smp_call_function_single(cpu, hardware_disable, NULL, 0, 1);
+		smp_call_function_single(cpu, hardware_disable, NULL, 1);
 		break;
 	case CPU_ONLINE:
 		printk(KERN_INFO "kvm: enabling virtualization on CPU%d\n",
 		       cpu);
-		smp_call_function_single(cpu, hardware_enable, NULL, 0, 1);
+		smp_call_function_single(cpu, hardware_enable, NULL, 1);
 		break;
 	}
 	return NOTIFY_OK;
@@ -1474,7 +1474,7 @@ int kvm_init(void *opaque, unsigned int vcpu_size,
 	for_each_online_cpu(cpu) {
 		smp_call_function_single(cpu,
 				kvm_arch_check_processor_compat,
-				&r, 0, 1);
+				&r, 1);
 		if (r < 0)
 			goto out_free_1;
 	}
-- 
cgit v1.2.3


From 15c8b6c1aaaf1c4edd67e2f02e4d8e1bd1a51c0d Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 9 May 2008 09:39:44 +0200
Subject: on_each_cpu(): kill unused 'retry' parameter

It's not even passed on to smp_call_function() anymore, since that
was removed. So kill it.

Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 arch/alpha/kernel/process.c            |  2 +-
 arch/alpha/kernel/smp.c                |  4 ++--
 arch/arm/kernel/smp.c                  |  6 +++---
 arch/ia64/kernel/mca.c                 |  4 ++--
 arch/ia64/kernel/perfmon.c             |  4 ++--
 arch/ia64/kernel/smp.c                 |  4 ++--
 arch/mips/kernel/irq-rm9000.c          |  4 ++--
 arch/mips/kernel/smp.c                 |  4 ++--
 arch/mips/oprofile/common.c            |  6 +++---
 arch/parisc/kernel/cache.c             |  6 +++---
 arch/parisc/kernel/smp.c               |  2 +-
 arch/parisc/mm/init.c                  |  2 +-
 arch/powerpc/kernel/rtas.c             |  2 +-
 arch/powerpc/kernel/tau_6xx.c          |  4 ++--
 arch/powerpc/kernel/time.c             |  2 +-
 arch/powerpc/mm/slice.c                |  2 +-
 arch/powerpc/oprofile/common.c         |  6 +++---
 arch/s390/kernel/smp.c                 |  6 +++---
 arch/s390/kernel/time.c                |  2 +-
 arch/sh/kernel/smp.c                   |  4 ++--
 arch/sparc64/mm/hugetlbpage.c          |  2 +-
 arch/x86/kernel/cpu/mcheck/mce_64.c    |  6 +++---
 arch/x86/kernel/cpu/mcheck/non-fatal.c |  2 +-
 arch/x86/kernel/cpu/perfctr-watchdog.c |  4 ++--
 arch/x86/kernel/io_apic_32.c           |  2 +-
 arch/x86/kernel/io_apic_64.c           |  2 +-
 arch/x86/kernel/nmi_32.c               |  4 ++--
 arch/x86/kernel/nmi_64.c               |  4 ++--
 arch/x86/kernel/tlb_32.c               |  2 +-
 arch/x86/kernel/tlb_64.c               |  2 +-
 arch/x86/kernel/vsyscall_64.c          |  2 +-
 arch/x86/kvm/vmx.c                     |  2 +-
 arch/x86/mach-voyager/voyager_smp.c    |  2 +-
 arch/x86/mm/pageattr.c                 |  4 ++--
 arch/x86/oprofile/nmi_int.c            | 10 +++++-----
 drivers/char/agp/generic.c             |  2 +-
 drivers/lguest/x86/core.c              |  4 ++--
 fs/buffer.c                            |  2 +-
 include/linux/smp.h                    |  4 ++--
 kernel/hrtimer.c                       |  2 +-
 kernel/profile.c                       |  6 +++---
 kernel/rcupdate.c                      |  2 +-
 kernel/softirq.c                       |  2 +-
 mm/page_alloc.c                        |  2 +-
 mm/slab.c                              |  4 ++--
 mm/slub.c                              |  2 +-
 net/iucv/iucv.c                        |  2 +-
 virt/kvm/kvm_main.c                    |  8 ++++----
 48 files changed, 84 insertions(+), 84 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c
index 96ed82fd9eef..351407e07e71 100644
--- a/arch/alpha/kernel/process.c
+++ b/arch/alpha/kernel/process.c
@@ -160,7 +160,7 @@ common_shutdown(int mode, char *restart_cmd)
 	struct halt_info args;
 	args.mode = mode;
 	args.restart_cmd = restart_cmd;
-	on_each_cpu(common_shutdown_1, &args, 1, 0);
+	on_each_cpu(common_shutdown_1, &args, 0);
 }
 
 void
diff --git a/arch/alpha/kernel/smp.c b/arch/alpha/kernel/smp.c
index 44114c8dbb2a..83df541650fc 100644
--- a/arch/alpha/kernel/smp.c
+++ b/arch/alpha/kernel/smp.c
@@ -657,7 +657,7 @@ void
 smp_imb(void)
 {
 	/* Must wait other processors to flush their icache before continue. */
-	if (on_each_cpu(ipi_imb, NULL, 1, 1))
+	if (on_each_cpu(ipi_imb, NULL, 1))
 		printk(KERN_CRIT "smp_imb: timed out\n");
 }
 EXPORT_SYMBOL(smp_imb);
@@ -673,7 +673,7 @@ flush_tlb_all(void)
 {
 	/* Although we don't have any data to pass, we do want to
 	   synchronize with the other processors.  */
-	if (on_each_cpu(ipi_flush_tlb_all, NULL, 1, 1)) {
+	if (on_each_cpu(ipi_flush_tlb_all, NULL, 1)) {
 		printk(KERN_CRIT "flush_tlb_all: timed out\n");
 	}
 }
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 6344466b2113..5a7c09564d13 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -604,7 +604,7 @@ static inline void ipi_flush_tlb_kernel_range(void *arg)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(ipi_flush_tlb_all, NULL, 1, 1);
+	on_each_cpu(ipi_flush_tlb_all, NULL, 1);
 }
 
 void flush_tlb_mm(struct mm_struct *mm)
@@ -631,7 +631,7 @@ void flush_tlb_kernel_page(unsigned long kaddr)
 
 	ta.ta_start = kaddr;
 
-	on_each_cpu(ipi_flush_tlb_kernel_page, &ta, 1, 1);
+	on_each_cpu(ipi_flush_tlb_kernel_page, &ta, 1);
 }
 
 void flush_tlb_range(struct vm_area_struct *vma,
@@ -654,5 +654,5 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 	ta.ta_start = start;
 	ta.ta_end = end;
 
-	on_each_cpu(ipi_flush_tlb_kernel_range, &ta, 1, 1);
+	on_each_cpu(ipi_flush_tlb_kernel_range, &ta, 1);
 }
diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index 9cd818cc7008..7dd96c127177 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -707,7 +707,7 @@ ia64_mca_cmc_vector_enable (void *dummy)
 static void
 ia64_mca_cmc_vector_disable_keventd(struct work_struct *unused)
 {
-	on_each_cpu(ia64_mca_cmc_vector_disable, NULL, 1, 0);
+	on_each_cpu(ia64_mca_cmc_vector_disable, NULL, 0);
 }
 
 /*
@@ -719,7 +719,7 @@ ia64_mca_cmc_vector_disable_keventd(struct work_struct *unused)
 static void
 ia64_mca_cmc_vector_enable_keventd(struct work_struct *unused)
 {
-	on_each_cpu(ia64_mca_cmc_vector_enable, NULL, 1, 0);
+	on_each_cpu(ia64_mca_cmc_vector_enable, NULL, 0);
 }
 
 /*
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 9baa48255c12..19d4493c6193 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -6508,7 +6508,7 @@ pfm_install_alt_pmu_interrupt(pfm_intr_handler_desc_t *hdl)
 	}
 
 	/* save the current system wide pmu states */
-	ret = on_each_cpu(pfm_alt_save_pmu_state, NULL, 0, 1);
+	ret = on_each_cpu(pfm_alt_save_pmu_state, NULL, 1);
 	if (ret) {
 		DPRINT(("on_each_cpu() failed: %d\n", ret));
 		goto cleanup_reserve;
@@ -6553,7 +6553,7 @@ pfm_remove_alt_pmu_interrupt(pfm_intr_handler_desc_t *hdl)
 
 	pfm_alt_intr_handler = NULL;
 
-	ret = on_each_cpu(pfm_alt_restore_pmu_state, NULL, 0, 1);
+	ret = on_each_cpu(pfm_alt_restore_pmu_state, NULL, 1);
 	if (ret) {
 		DPRINT(("on_each_cpu() failed: %d\n", ret));
 	}
diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c
index 19152dcbf6e4..3676468612b6 100644
--- a/arch/ia64/kernel/smp.c
+++ b/arch/ia64/kernel/smp.c
@@ -285,7 +285,7 @@ smp_flush_tlb_cpumask(cpumask_t xcpumask)
 void
 smp_flush_tlb_all (void)
 {
-	on_each_cpu((void (*)(void *))local_flush_tlb_all, NULL, 1, 1);
+	on_each_cpu((void (*)(void *))local_flush_tlb_all, NULL, 1);
 }
 
 void
@@ -308,7 +308,7 @@ smp_flush_tlb_mm (struct mm_struct *mm)
 	 * anyhow, and once a CPU is interrupted, the cost of local_flush_tlb_all() is
 	 * rather trivial.
 	 */
-	on_each_cpu((void (*)(void *))local_finish_flush_tlb_mm, mm, 1, 1);
+	on_each_cpu((void (*)(void *))local_finish_flush_tlb_mm, mm, 1);
 }
 
 void arch_send_call_function_single_ipi(int cpu)
diff --git a/arch/mips/kernel/irq-rm9000.c b/arch/mips/kernel/irq-rm9000.c
index ed9febe63d72..b47e4615ec12 100644
--- a/arch/mips/kernel/irq-rm9000.c
+++ b/arch/mips/kernel/irq-rm9000.c
@@ -49,7 +49,7 @@ static void local_rm9k_perfcounter_irq_startup(void *args)
 
 static unsigned int rm9k_perfcounter_irq_startup(unsigned int irq)
 {
-	on_each_cpu(local_rm9k_perfcounter_irq_startup, (void *) irq, 0, 1);
+	on_each_cpu(local_rm9k_perfcounter_irq_startup, (void *) irq, 1);
 
 	return 0;
 }
@@ -66,7 +66,7 @@ static void local_rm9k_perfcounter_irq_shutdown(void *args)
 
 static void rm9k_perfcounter_irq_shutdown(unsigned int irq)
 {
-	on_each_cpu(local_rm9k_perfcounter_irq_shutdown, (void *) irq, 0, 1);
+	on_each_cpu(local_rm9k_perfcounter_irq_shutdown, (void *) irq, 1);
 }
 
 static struct irq_chip rm9k_irq_controller = {
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 7a9ae830be86..4410f172b8ab 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -246,7 +246,7 @@ static void flush_tlb_all_ipi(void *info)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(flush_tlb_all_ipi, NULL, 1, 1);
+	on_each_cpu(flush_tlb_all_ipi, NULL, 1);
 }
 
 static void flush_tlb_mm_ipi(void *mm)
@@ -366,7 +366,7 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 		.addr2 = end,
 	};
 
-	on_each_cpu(flush_tlb_kernel_range_ipi, &fd, 1, 1);
+	on_each_cpu(flush_tlb_kernel_range_ipi, &fd, 1);
 }
 
 static void flush_tlb_page_ipi(void *info)
diff --git a/arch/mips/oprofile/common.c b/arch/mips/oprofile/common.c
index b5f6f71b27bc..dd2fbd6645c1 100644
--- a/arch/mips/oprofile/common.c
+++ b/arch/mips/oprofile/common.c
@@ -27,7 +27,7 @@ static int op_mips_setup(void)
 	model->reg_setup(ctr);
 
 	/* Configure the registers on all cpus.  */
-	on_each_cpu(model->cpu_setup, NULL, 0, 1);
+	on_each_cpu(model->cpu_setup, NULL, 1);
 
         return 0;
 }
@@ -58,7 +58,7 @@ static int op_mips_create_files(struct super_block * sb, struct dentry * root)
 
 static int op_mips_start(void)
 {
-	on_each_cpu(model->cpu_start, NULL, 0, 1);
+	on_each_cpu(model->cpu_start, NULL, 1);
 
 	return 0;
 }
@@ -66,7 +66,7 @@ static int op_mips_start(void)
 static void op_mips_stop(void)
 {
 	/* Disable performance monitoring for all counters.  */
-	on_each_cpu(model->cpu_stop, NULL, 0, 1);
+	on_each_cpu(model->cpu_stop, NULL, 1);
 }
 
 int __init oprofile_arch_init(struct oprofile_operations *ops)
diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c
index e10d25d2d9c9..5259d8c20676 100644
--- a/arch/parisc/kernel/cache.c
+++ b/arch/parisc/kernel/cache.c
@@ -51,12 +51,12 @@ static struct pdc_btlb_info btlb_info __read_mostly;
 void
 flush_data_cache(void)
 {
-	on_each_cpu(flush_data_cache_local, NULL, 1, 1);
+	on_each_cpu(flush_data_cache_local, NULL, 1);
 }
 void 
 flush_instruction_cache(void)
 {
-	on_each_cpu(flush_instruction_cache_local, NULL, 1, 1);
+	on_each_cpu(flush_instruction_cache_local, NULL, 1);
 }
 #endif
 
@@ -515,7 +515,7 @@ static void cacheflush_h_tmp_function(void *dummy)
 
 void flush_cache_all(void)
 {
-	on_each_cpu(cacheflush_h_tmp_function, NULL, 1, 1);
+	on_each_cpu(cacheflush_h_tmp_function, NULL, 1);
 }
 
 void flush_cache_mm(struct mm_struct *mm)
diff --git a/arch/parisc/kernel/smp.c b/arch/parisc/kernel/smp.c
index 126105c76a44..d47f3975c9c6 100644
--- a/arch/parisc/kernel/smp.c
+++ b/arch/parisc/kernel/smp.c
@@ -292,7 +292,7 @@ void arch_send_call_function_single_ipi(int cpu)
 void
 smp_flush_tlb_all(void)
 {
-	on_each_cpu(flush_tlb_all_local, NULL, 1, 1);
+	on_each_cpu(flush_tlb_all_local, NULL, 1);
 }
 
 /*
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index ce0da689a89d..b4d6c8777ed0 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -1053,7 +1053,7 @@ void flush_tlb_all(void)
 	    do_recycle++;
 	}
 	spin_unlock(&sid_lock);
-	on_each_cpu(flush_tlb_all_local, NULL, 1, 1);
+	on_each_cpu(flush_tlb_all_local, NULL, 1);
 	if (do_recycle) {
 	    spin_lock(&sid_lock);
 	    recycle_sids(recycle_ndirty,recycle_dirty_array);
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index 34843c318419..647f3e8677dc 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -747,7 +747,7 @@ static int rtas_ibm_suspend_me(struct rtas_args *args)
 	/* Call function on all CPUs.  One of us will make the
 	 * rtas call
 	 */
-	if (on_each_cpu(rtas_percpu_suspend_me, &data, 1, 0))
+	if (on_each_cpu(rtas_percpu_suspend_me, &data, 0))
 		data.error = -EINVAL;
 
 	wait_for_completion(&done);
diff --git a/arch/powerpc/kernel/tau_6xx.c b/arch/powerpc/kernel/tau_6xx.c
index 368a4934f7ee..c3a56d65c5a9 100644
--- a/arch/powerpc/kernel/tau_6xx.c
+++ b/arch/powerpc/kernel/tau_6xx.c
@@ -192,7 +192,7 @@ static void tau_timeout_smp(unsigned long unused)
 
 	/* schedule ourselves to be run again */
 	mod_timer(&tau_timer, jiffies + shrink_timer) ;
-	on_each_cpu(tau_timeout, NULL, 1, 0);
+	on_each_cpu(tau_timeout, NULL, 0);
 }
 
 /*
@@ -234,7 +234,7 @@ int __init TAU_init(void)
 	tau_timer.expires = jiffies + shrink_timer;
 	add_timer(&tau_timer);
 
-	on_each_cpu(TAU_init_smp, NULL, 1, 0);
+	on_each_cpu(TAU_init_smp, NULL, 0);
 
 	printk("Thermal assist unit ");
 #ifdef CONFIG_TAU_INT
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 73401e83739a..f1a38a6c1e2d 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -322,7 +322,7 @@ void snapshot_timebases(void)
 {
 	if (!cpu_has_feature(CPU_FTR_PURR))
 		return;
-	on_each_cpu(snapshot_tb_and_purr, NULL, 0, 1);
+	on_each_cpu(snapshot_tb_and_purr, NULL, 1);
 }
 
 /*
diff --git a/arch/powerpc/mm/slice.c b/arch/powerpc/mm/slice.c
index ad928edafb0a..2bd12d965db1 100644
--- a/arch/powerpc/mm/slice.c
+++ b/arch/powerpc/mm/slice.c
@@ -218,7 +218,7 @@ static void slice_convert(struct mm_struct *mm, struct slice_mask mask, int psiz
 	mb();
 
 	/* XXX this is sub-optimal but will do for now */
-	on_each_cpu(slice_flush_segments, mm, 0, 1);
+	on_each_cpu(slice_flush_segments, mm, 1);
 #ifdef CONFIG_SPU_BASE
 	spu_flush_all_slbs(mm);
 #endif
diff --git a/arch/powerpc/oprofile/common.c b/arch/powerpc/oprofile/common.c
index 4908dc98f9ca..17807acb05d9 100644
--- a/arch/powerpc/oprofile/common.c
+++ b/arch/powerpc/oprofile/common.c
@@ -65,7 +65,7 @@ static int op_powerpc_setup(void)
 
 	/* Configure the registers on all cpus.	 If an error occurs on one
 	 * of the cpus, op_per_cpu_rc will be set to the error */
-	on_each_cpu(op_powerpc_cpu_setup, NULL, 0, 1);
+	on_each_cpu(op_powerpc_cpu_setup, NULL, 1);
 
 out:	if (op_per_cpu_rc) {
 		/* error on setup release the performance counter hardware */
@@ -100,7 +100,7 @@ static int op_powerpc_start(void)
 	if (model->global_start)
 		return model->global_start(ctr);
 	if (model->start) {
-		on_each_cpu(op_powerpc_cpu_start, NULL, 0, 1);
+		on_each_cpu(op_powerpc_cpu_start, NULL, 1);
 		return op_per_cpu_rc;
 	}
 	return -EIO; /* No start function is defined for this
@@ -115,7 +115,7 @@ static inline void op_powerpc_cpu_stop(void *dummy)
 static void op_powerpc_stop(void)
 {
 	if (model->stop)
-		on_each_cpu(op_powerpc_cpu_stop, NULL, 0, 1);
+		on_each_cpu(op_powerpc_cpu_stop, NULL, 1);
         if (model->global_stop)
                 model->global_stop();
 }
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 276b105fb2a4..b6781030cfbd 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -299,7 +299,7 @@ static void smp_ptlb_callback(void *info)
 
 void smp_ptlb_all(void)
 {
-	on_each_cpu(smp_ptlb_callback, NULL, 0, 1);
+	on_each_cpu(smp_ptlb_callback, NULL, 1);
 }
 EXPORT_SYMBOL(smp_ptlb_all);
 #endif /* ! CONFIG_64BIT */
@@ -347,7 +347,7 @@ void smp_ctl_set_bit(int cr, int bit)
 	memset(&parms.orvals, 0, sizeof(parms.orvals));
 	memset(&parms.andvals, 0xff, sizeof(parms.andvals));
 	parms.orvals[cr] = 1 << bit;
-	on_each_cpu(smp_ctl_bit_callback, &parms, 0, 1);
+	on_each_cpu(smp_ctl_bit_callback, &parms, 1);
 }
 EXPORT_SYMBOL(smp_ctl_set_bit);
 
@@ -361,7 +361,7 @@ void smp_ctl_clear_bit(int cr, int bit)
 	memset(&parms.orvals, 0, sizeof(parms.orvals));
 	memset(&parms.andvals, 0xff, sizeof(parms.andvals));
 	parms.andvals[cr] = ~(1L << bit);
-	on_each_cpu(smp_ctl_bit_callback, &parms, 0, 1);
+	on_each_cpu(smp_ctl_bit_callback, &parms, 1);
 }
 EXPORT_SYMBOL(smp_ctl_clear_bit);
 
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index bf7bf2c2236a..6037ed2b7471 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -909,7 +909,7 @@ static void etr_work_fn(struct work_struct *work)
 	if (!eacr.ea) {
 		/* Both ports offline. Reset everything. */
 		eacr.dp = eacr.es = eacr.sl = 0;
-		on_each_cpu(etr_disable_sync_clock, NULL, 0, 1);
+		on_each_cpu(etr_disable_sync_clock, NULL, 1);
 		del_timer_sync(&etr_timer);
 		etr_update_eacr(eacr);
 		set_bit(ETR_FLAG_EACCES, &etr_flags);
diff --git a/arch/sh/kernel/smp.c b/arch/sh/kernel/smp.c
index 71781ba2675b..60c50841143e 100644
--- a/arch/sh/kernel/smp.c
+++ b/arch/sh/kernel/smp.c
@@ -197,7 +197,7 @@ static void flush_tlb_all_ipi(void *info)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(flush_tlb_all_ipi, 0, 1, 1);
+	on_each_cpu(flush_tlb_all_ipi, 0, 1);
 }
 
 static void flush_tlb_mm_ipi(void *mm)
@@ -284,7 +284,7 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 
 	fd.addr1 = start;
 	fd.addr2 = end;
-	on_each_cpu(flush_tlb_kernel_range_ipi, (void *)&fd, 1, 1);
+	on_each_cpu(flush_tlb_kernel_range_ipi, (void *)&fd, 1);
 }
 
 static void flush_tlb_page_ipi(void *info)
diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c
index 6cfab2e4d340..ebefd2a14375 100644
--- a/arch/sparc64/mm/hugetlbpage.c
+++ b/arch/sparc64/mm/hugetlbpage.c
@@ -344,7 +344,7 @@ void hugetlb_prefault_arch_hook(struct mm_struct *mm)
 			 * also executing in this address space.
 			 */
 			mm->context.sparc64_ctx_val = ctx;
-			on_each_cpu(context_reload, mm, 0, 0);
+			on_each_cpu(context_reload, mm, 0);
 		}
 		spin_unlock(&ctx_alloc_lock);
 	}
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index e07e8c068ae0..43b7cb594912 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -363,7 +363,7 @@ static void mcheck_check_cpu(void *info)
 
 static void mcheck_timer(struct work_struct *work)
 {
-	on_each_cpu(mcheck_check_cpu, NULL, 1, 1);
+	on_each_cpu(mcheck_check_cpu, NULL, 1);
 
 	/*
 	 * Alert userspace if needed.  If we logged an MCE, reduce the
@@ -612,7 +612,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
 	 * Collect entries that were still getting written before the
 	 * synchronize.
 	 */
-	on_each_cpu(collect_tscs, cpu_tsc, 1, 1);
+	on_each_cpu(collect_tscs, cpu_tsc, 1);
 	for (i = next; i < MCE_LOG_LEN; i++) {
 		if (mcelog.entry[i].finished &&
 		    mcelog.entry[i].tsc < cpu_tsc[mcelog.entry[i].cpu]) {
@@ -737,7 +737,7 @@ static void mce_restart(void)
 	if (next_interval)
 		cancel_delayed_work(&mcheck_work);
 	/* Timer race is harmless here */
-	on_each_cpu(mce_init, NULL, 1, 1);
+	on_each_cpu(mce_init, NULL, 1);
 	next_interval = check_interval * HZ;
 	if (next_interval)
 		schedule_delayed_work(&mcheck_work,
diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c
index 00ccb6c14ec2..cc1fccdd31e0 100644
--- a/arch/x86/kernel/cpu/mcheck/non-fatal.c
+++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c
@@ -59,7 +59,7 @@ static DECLARE_DELAYED_WORK(mce_work, mce_work_fn);
 
 static void mce_work_fn(struct work_struct *work)
 {
-	on_each_cpu(mce_checkregs, NULL, 1, 1);
+	on_each_cpu(mce_checkregs, NULL, 1);
 	schedule_delayed_work(&mce_work, round_jiffies_relative(MCE_RATE));
 }
 
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index f9ae93adffe5..58043f06d7e2 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -180,7 +180,7 @@ void disable_lapic_nmi_watchdog(void)
 	if (atomic_read(&nmi_active) <= 0)
 		return;
 
-	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 1);
 	wd_ops->unreserve();
 
 	BUG_ON(atomic_read(&nmi_active) != 0);
@@ -202,7 +202,7 @@ void enable_lapic_nmi_watchdog(void)
 		return;
 	}
 
-	on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+	on_each_cpu(setup_apic_nmi_watchdog, NULL, 1);
 	touch_nmi_watchdog();
 }
 
diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 4dc8600d9d20..720640ff36ca 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -1565,7 +1565,7 @@ void /*__init*/ print_local_APIC(void * dummy)
 
 void print_all_local_APICs (void)
 {
-	on_each_cpu(print_local_APIC, NULL, 1, 1);
+	on_each_cpu(print_local_APIC, NULL, 1);
 }
 
 void /*__init*/ print_PIC(void)
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index ef1a8dfcc529..4504c7f50012 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -1146,7 +1146,7 @@ void __apicdebuginit print_local_APIC(void * dummy)
 
 void print_all_local_APICs (void)
 {
-	on_each_cpu(print_local_APIC, NULL, 1, 1);
+	on_each_cpu(print_local_APIC, NULL, 1);
 }
 
 void __apicdebuginit print_PIC(void)
diff --git a/arch/x86/kernel/nmi_32.c b/arch/x86/kernel/nmi_32.c
index 5562dab0bd20..11008e0857c0 100644
--- a/arch/x86/kernel/nmi_32.c
+++ b/arch/x86/kernel/nmi_32.c
@@ -218,7 +218,7 @@ static void __acpi_nmi_enable(void *__unused)
 void acpi_nmi_enable(void)
 {
 	if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC)
-		on_each_cpu(__acpi_nmi_enable, NULL, 0, 1);
+		on_each_cpu(__acpi_nmi_enable, NULL, 1);
 }
 
 static void __acpi_nmi_disable(void *__unused)
@@ -232,7 +232,7 @@ static void __acpi_nmi_disable(void *__unused)
 void acpi_nmi_disable(void)
 {
 	if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC)
-		on_each_cpu(__acpi_nmi_disable, NULL, 0, 1);
+		on_each_cpu(__acpi_nmi_disable, NULL, 1);
 }
 
 void setup_apic_nmi_watchdog(void *unused)
diff --git a/arch/x86/kernel/nmi_64.c b/arch/x86/kernel/nmi_64.c
index 2f1e4f503c9e..bbdcb17b3dfe 100644
--- a/arch/x86/kernel/nmi_64.c
+++ b/arch/x86/kernel/nmi_64.c
@@ -225,7 +225,7 @@ static void __acpi_nmi_enable(void *__unused)
 void acpi_nmi_enable(void)
 {
 	if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC)
-		on_each_cpu(__acpi_nmi_enable, NULL, 0, 1);
+		on_each_cpu(__acpi_nmi_enable, NULL, 1);
 }
 
 static void __acpi_nmi_disable(void *__unused)
@@ -239,7 +239,7 @@ static void __acpi_nmi_disable(void *__unused)
 void acpi_nmi_disable(void)
 {
 	if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC)
-		on_each_cpu(__acpi_nmi_disable, NULL, 0, 1);
+		on_each_cpu(__acpi_nmi_disable, NULL, 1);
 }
 
 void setup_apic_nmi_watchdog(void *unused)
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index 9bb2363851af..fec1ecedc9b7 100644
--- a/arch/x86/kernel/tlb_32.c
+++ b/arch/x86/kernel/tlb_32.c
@@ -238,6 +238,6 @@ static void do_flush_tlb_all(void *info)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
+	on_each_cpu(do_flush_tlb_all, NULL, 1);
 }
 
diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c
index a1f07d793202..184a367516d3 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -270,5 +270,5 @@ static void do_flush_tlb_all(void *info)
 
 void flush_tlb_all(void)
 {
-	on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
+	on_each_cpu(do_flush_tlb_all, NULL, 1);
 }
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index 0a03d57f9b3b..0dcae19ed627 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -301,7 +301,7 @@ static int __init vsyscall_init(void)
 #ifdef CONFIG_SYSCTL
 	register_sysctl_table(kernel_root_table2);
 #endif
-	on_each_cpu(cpu_vsyscall_init, NULL, 0, 1);
+	on_each_cpu(cpu_vsyscall_init, NULL, 1);
 	hotcpu_notifier(cpu_vsyscall_notifier, 0);
 	return 0;
 }
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 5534fe59b5fc..10ce6ee4c491 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2968,7 +2968,7 @@ static void vmx_free_vmcs(struct kvm_vcpu *vcpu)
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
 
 	if (vmx->vmcs) {
-		on_each_cpu(__vcpu_clear, vmx, 0, 1);
+		on_each_cpu(__vcpu_clear, vmx, 1);
 		free_vmcs(vmx->vmcs);
 		vmx->vmcs = NULL;
 	}
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 04f596eab749..abea08459a73 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -1072,7 +1072,7 @@ static void do_flush_tlb_all(void *info)
 /* flush the TLB of every active CPU in the system */
 void flush_tlb_all(void)
 {
-	on_each_cpu(do_flush_tlb_all, 0, 1, 1);
+	on_each_cpu(do_flush_tlb_all, 0, 1);
 }
 
 /* used to set up the trampoline for other CPUs when the memory manager
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 60bcb5b6a37e..9b836ba9dedd 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -106,7 +106,7 @@ static void cpa_flush_all(unsigned long cache)
 {
 	BUG_ON(irqs_disabled());
 
-	on_each_cpu(__cpa_flush_all, (void *) cache, 1, 1);
+	on_each_cpu(__cpa_flush_all, (void *) cache, 1);
 }
 
 static void __cpa_flush_range(void *arg)
@@ -127,7 +127,7 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache)
 	BUG_ON(irqs_disabled());
 	WARN_ON(PAGE_ALIGN(start) != start);
 
-	on_each_cpu(__cpa_flush_range, NULL, 1, 1);
+	on_each_cpu(__cpa_flush_range, NULL, 1);
 
 	if (!cache)
 		return;
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index cc48d3fde545..3238ad32ffd8 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -218,8 +218,8 @@ static int nmi_setup(void)
 		}
 
 	}
-	on_each_cpu(nmi_save_registers, NULL, 0, 1);
-	on_each_cpu(nmi_cpu_setup, NULL, 0, 1);
+	on_each_cpu(nmi_save_registers, NULL, 1);
+	on_each_cpu(nmi_cpu_setup, NULL, 1);
 	nmi_enabled = 1;
 	return 0;
 }
@@ -271,7 +271,7 @@ static void nmi_shutdown(void)
 {
 	struct op_msrs *msrs = &__get_cpu_var(cpu_msrs);
 	nmi_enabled = 0;
-	on_each_cpu(nmi_cpu_shutdown, NULL, 0, 1);
+	on_each_cpu(nmi_cpu_shutdown, NULL, 1);
 	unregister_die_notifier(&profile_exceptions_nb);
 	model->shutdown(msrs);
 	free_msrs();
@@ -285,7 +285,7 @@ static void nmi_cpu_start(void *dummy)
 
 static int nmi_start(void)
 {
-	on_each_cpu(nmi_cpu_start, NULL, 0, 1);
+	on_each_cpu(nmi_cpu_start, NULL, 1);
 	return 0;
 }
 
@@ -297,7 +297,7 @@ static void nmi_cpu_stop(void *dummy)
 
 static void nmi_stop(void)
 {
-	on_each_cpu(nmi_cpu_stop, NULL, 0, 1);
+	on_each_cpu(nmi_cpu_stop, NULL, 1);
 }
 
 struct op_counter_config counter_config[OP_MAX_COUNTER];
diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c
index 564daaa6c7d0..eaa1a355bb32 100644
--- a/drivers/char/agp/generic.c
+++ b/drivers/char/agp/generic.c
@@ -1249,7 +1249,7 @@ static void ipi_handler(void *null)
 
 void global_cache_flush(void)
 {
-	if (on_each_cpu(ipi_handler, NULL, 1, 1) != 0)
+	if (on_each_cpu(ipi_handler, NULL, 1) != 0)
 		panic(PFX "timed out waiting for the other CPUs!\n");
 }
 EXPORT_SYMBOL(global_cache_flush);
diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c
index 2e554a4ab337..95dfda52b4f9 100644
--- a/drivers/lguest/x86/core.c
+++ b/drivers/lguest/x86/core.c
@@ -478,7 +478,7 @@ void __init lguest_arch_host_init(void)
 		cpu_had_pge = 1;
 		/* adjust_pge is a helper function which sets or unsets the PGE
 		 * bit on its CPU, depending on the argument (0 == unset). */
-		on_each_cpu(adjust_pge, (void *)0, 0, 1);
+		on_each_cpu(adjust_pge, (void *)0, 1);
 		/* Turn off the feature in the global feature set. */
 		clear_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability);
 	}
@@ -493,7 +493,7 @@ void __exit lguest_arch_host_fini(void)
 	if (cpu_had_pge) {
 		set_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability);
 		/* adjust_pge's argument "1" means set PGE. */
-		on_each_cpu(adjust_pge, (void *)1, 0, 1);
+		on_each_cpu(adjust_pge, (void *)1, 1);
 	}
 	put_online_cpus();
 }
diff --git a/fs/buffer.c b/fs/buffer.c
index a073f3f4f013..5c23ef560d01 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1464,7 +1464,7 @@ static void invalidate_bh_lru(void *arg)
 	
 void invalidate_bh_lrus(void)
 {
-	on_each_cpu(invalidate_bh_lru, NULL, 1, 1);
+	on_each_cpu(invalidate_bh_lru, NULL, 1);
 }
 EXPORT_SYMBOL_GPL(invalidate_bh_lrus);
 
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 338cad1b9548..55261101d09a 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -89,7 +89,7 @@ static inline void init_call_single_data(void)
 /*
  * Call a function on all processors
  */
-int on_each_cpu(void (*func) (void *info), void *info, int retry, int wait);
+int on_each_cpu(void (*func) (void *info), void *info, int wait);
 
 #define MSG_ALL_BUT_SELF	0x8000	/* Assume <32768 CPU's */
 #define MSG_ALL			0x8001
@@ -121,7 +121,7 @@ static inline int up_smp_call_function(void (*func)(void *), void *info)
 }
 #define smp_call_function(func, info, wait) \
 			(up_smp_call_function(func, info))
-#define on_each_cpu(func,info,retry,wait)	\
+#define on_each_cpu(func,info,wait)		\
 	({					\
 		local_irq_disable();		\
 		func(info);			\
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 421be5fe5cc7..50e8616d7955 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -623,7 +623,7 @@ static void retrigger_next_event(void *arg)
 void clock_was_set(void)
 {
 	/* Retrigger the CPU local events everywhere */
-	on_each_cpu(retrigger_next_event, NULL, 0, 1);
+	on_each_cpu(retrigger_next_event, NULL, 1);
 }
 
 /*
diff --git a/kernel/profile.c b/kernel/profile.c
index ae7ead82cbc9..58926411eb2a 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -252,7 +252,7 @@ static void profile_flip_buffers(void)
 	mutex_lock(&profile_flip_mutex);
 	j = per_cpu(cpu_profile_flip, get_cpu());
 	put_cpu();
-	on_each_cpu(__profile_flip_buffers, NULL, 0, 1);
+	on_each_cpu(__profile_flip_buffers, NULL, 1);
 	for_each_online_cpu(cpu) {
 		struct profile_hit *hits = per_cpu(cpu_profile_hits, cpu)[j];
 		for (i = 0; i < NR_PROFILE_HIT; ++i) {
@@ -275,7 +275,7 @@ static void profile_discard_flip_buffers(void)
 	mutex_lock(&profile_flip_mutex);
 	i = per_cpu(cpu_profile_flip, get_cpu());
 	put_cpu();
-	on_each_cpu(__profile_flip_buffers, NULL, 0, 1);
+	on_each_cpu(__profile_flip_buffers, NULL, 1);
 	for_each_online_cpu(cpu) {
 		struct profile_hit *hits = per_cpu(cpu_profile_hits, cpu)[i];
 		memset(hits, 0, NR_PROFILE_HIT*sizeof(struct profile_hit));
@@ -558,7 +558,7 @@ static int __init create_hash_tables(void)
 out_cleanup:
 	prof_on = 0;
 	smp_mb();
-	on_each_cpu(profile_nop, NULL, 0, 1);
+	on_each_cpu(profile_nop, NULL, 1);
 	for_each_online_cpu(cpu) {
 		struct page *page;
 
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index c09605f8d16c..6addab5e6d88 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -127,7 +127,7 @@ void rcu_barrier(void)
 	 * until all the callbacks are queued.
 	 */
 	rcu_read_lock();
-	on_each_cpu(rcu_barrier_func, NULL, 0, 1);
+	on_each_cpu(rcu_barrier_func, NULL, 1);
 	rcu_read_unlock();
 	wait_for_completion(&rcu_barrier_completion);
 	mutex_unlock(&rcu_barrier_mutex);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index d73afb4764ef..c159fd094772 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -674,7 +674,7 @@ __init int spawn_ksoftirqd(void)
 /*
  * Call a function on all processors
  */
-int on_each_cpu(void (*func) (void *info), void *info, int retry, int wait)
+int on_each_cpu(void (*func) (void *info), void *info, int wait)
 {
 	int ret = 0;
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 2f552955a02f..53242344a774 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -918,7 +918,7 @@ void drain_local_pages(void *arg)
  */
 void drain_all_pages(void)
 {
-	on_each_cpu(drain_local_pages, NULL, 0, 1);
+	on_each_cpu(drain_local_pages, NULL, 1);
 }
 
 #ifdef CONFIG_HIBERNATION
diff --git a/mm/slab.c b/mm/slab.c
index 046607f05f3e..0772abb412b9 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2454,7 +2454,7 @@ static void drain_cpu_caches(struct kmem_cache *cachep)
 	struct kmem_list3 *l3;
 	int node;
 
-	on_each_cpu(do_drain, cachep, 1, 1);
+	on_each_cpu(do_drain, cachep, 1);
 	check_irq_on();
 	for_each_online_node(node) {
 		l3 = cachep->nodelists[node];
@@ -3939,7 +3939,7 @@ static int do_tune_cpucache(struct kmem_cache *cachep, int limit,
 	}
 	new->cachep = cachep;
 
-	on_each_cpu(do_ccupdate_local, (void *)new, 1, 1);
+	on_each_cpu(do_ccupdate_local, (void *)new, 1);
 
 	check_irq_on();
 	cachep->batchcount = batchcount;
diff --git a/mm/slub.c b/mm/slub.c
index 0987d1cd943c..44715eb70c06 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1497,7 +1497,7 @@ static void flush_cpu_slab(void *d)
 static void flush_all(struct kmem_cache *s)
 {
 #ifdef CONFIG_SMP
-	on_each_cpu(flush_cpu_slab, s, 1, 1);
+	on_each_cpu(flush_cpu_slab, s, 1);
 #else
 	unsigned long flags;
 
diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c
index 94d5a45c3a57..a178e27e7b1a 100644
--- a/net/iucv/iucv.c
+++ b/net/iucv/iucv.c
@@ -545,7 +545,7 @@ out:
  */
 static void iucv_disable(void)
 {
-	on_each_cpu(iucv_retrieve_cpu, NULL, 0, 1);
+	on_each_cpu(iucv_retrieve_cpu, NULL, 1);
 	kfree(iucv_path_table);
 }
 
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index ea1f595f8a87..d4eae6af0738 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1286,7 +1286,7 @@ static int kvm_reboot(struct notifier_block *notifier, unsigned long val,
 		 * in vmx root mode.
 		 */
 		printk(KERN_INFO "kvm: exiting hardware virtualization\n");
-		on_each_cpu(hardware_disable, NULL, 0, 1);
+		on_each_cpu(hardware_disable, NULL, 1);
 	}
 	return NOTIFY_OK;
 }
@@ -1479,7 +1479,7 @@ int kvm_init(void *opaque, unsigned int vcpu_size,
 			goto out_free_1;
 	}
 
-	on_each_cpu(hardware_enable, NULL, 0, 1);
+	on_each_cpu(hardware_enable, NULL, 1);
 	r = register_cpu_notifier(&kvm_cpu_notifier);
 	if (r)
 		goto out_free_2;
@@ -1525,7 +1525,7 @@ out_free_3:
 	unregister_reboot_notifier(&kvm_reboot_notifier);
 	unregister_cpu_notifier(&kvm_cpu_notifier);
 out_free_2:
-	on_each_cpu(hardware_disable, NULL, 0, 1);
+	on_each_cpu(hardware_disable, NULL, 1);
 out_free_1:
 	kvm_arch_hardware_unsetup();
 out_free_0:
@@ -1547,7 +1547,7 @@ void kvm_exit(void)
 	sysdev_class_unregister(&kvm_sysdev_class);
 	unregister_reboot_notifier(&kvm_reboot_notifier);
 	unregister_cpu_notifier(&kvm_cpu_notifier);
-	on_each_cpu(hardware_disable, NULL, 0, 1);
+	on_each_cpu(hardware_disable, NULL, 1);
 	kvm_arch_hardware_unsetup();
 	kvm_arch_exit();
 	kvm_exit_debug();
-- 
cgit v1.2.3


From f3146aff7f283c8699e0c97df6307a705786eeba Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:22:56 -0300
Subject: rfkill: clarify meaning of rfkill states

rfkill really should have been named rfswitch.  As it is, one can get
confused whether RFKILL_STATE_ON means the KILL switch is on (and
therefore, the radio is being *blocked* from operating), or whether it
means the RADIO rf output is on.

Clearly state that RFKILL_STATE_ON means the radio is *unblocked* from
operating (i.e. there is no rf killing going on).

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Cc: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 Documentation/rfkill.txt | 7 +++++++
 include/linux/rfkill.h   | 6 +++---
 2 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/rfkill.txt b/Documentation/rfkill.txt
index a83ff23cd68c..ec75d6d34785 100644
--- a/Documentation/rfkill.txt
+++ b/Documentation/rfkill.txt
@@ -8,6 +8,13 @@ rfkill - RF switch subsystem support
 ===============================================================================
 1: Implementation details
 
+The rfkill switch subsystem exists to add a generic interface to circuitry that
+can enable or disable the RF output of a radio *transmitter* of any type.
+
+When a rfkill switch is in the RFKILL_STATE_ON, the radio transmitter is
+*enabled*.  When the rfkill switch is in the RFKILL_STATE_OFF, the radio
+transmitter is *disabled*.
+
 The rfkill switch subsystem offers support for keys often found on laptops
 to enable wireless devices like WiFi and Bluetooth.
 
diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index e3ab21d7fc7f..ca89ae1b0219 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -44,8 +44,8 @@ enum rfkill_type {
 };
 
 enum rfkill_state {
-	RFKILL_STATE_OFF	= 0,
-	RFKILL_STATE_ON		= 1,
+	RFKILL_STATE_OFF	= 0,	/* Radio output blocked */
+	RFKILL_STATE_ON		= 1,	/* Radio output active */
 };
 
 /**
@@ -53,7 +53,7 @@ enum rfkill_state {
  * @name: Name of the switch.
  * @type: Radio type which the button controls, the value stored
  *	here should be a value from enum rfkill_type.
- * @state: State of the switch (on/off).
+ * @state: State of the switch, "ON" means radio can operate.
  * @user_claim_unsupported: Whether the hardware supports exclusive
  *	RF-kill control by userspace. Set this before registering.
  * @user_claim: Set when the switch is controlled exlusively by userspace.
-- 
cgit v1.2.3


From 801e49af4c1a9b988ba0d25de2b368c99c3bf2b3 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:23:00 -0300
Subject: rfkill: add read-write rfkill switch support

Currently, rfkill support for read/write rfkill switches is hacked through
a round-trip over the input layer and rfkill-input to let a driver sync
rfkill->state to hardware changes.

This is buggy and sub-optimal.  It causes real problems.  It is best to
think of the rfkill class as supporting only write-only switches at the
moment.

In order to implement the read/write functionality properly:

Add a get_state() hook that is called by the class every time it needs to
fetch the current state of the switch.  Add a call to this hook every time
the *current* state of the radio plays a role in a decision.

Also add a force_state() method that can be used to forcefully syncronize
the class' idea of the current state of the switch.  This allows for a
faster implementation of the read/write functionality, as a driver which
get events on switch changes can avoid the need for a get_state() hook.

If the get_state() hook is left as NULL, current behaviour is maintained,
so this change is fully backwards compatible with the current rfkill
drivers.

For hardware that issues events when the rfkill state changes, leave
get_state() NULL in the rfkill struct, set the initial state properly
before registering with the rfkill class, and use the force_state() method
in the driver to keep the rfkill interface up-to-date.

get_state() can be called by the class from atomic context. It must not
sleep.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Cc: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/rfkill.h |  5 +++++
 net/rfkill/rfkill.c    | 53 +++++++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 55 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index ca89ae1b0219..844e96114861 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -61,6 +61,8 @@ enum rfkill_state {
  * @data: Pointer to the RF button drivers private data which will be
  *	passed along when toggling radio state.
  * @toggle_radio(): Mandatory handler to control state of the radio.
+ * @get_state(): handler to read current radio state from hardware,
+ *      may be called from atomic context, should return 0 on success.
  * @led_trigger: A LED trigger for this button's LED.
  * @dev: Device structure integrating the switch into device tree.
  * @node: Used to place switch into list of all switches known to the
@@ -80,6 +82,7 @@ struct rfkill {
 
 	void *data;
 	int (*toggle_radio)(void *data, enum rfkill_state state);
+	int (*get_state)(void *data, enum rfkill_state *state);
 
 #ifdef CONFIG_RFKILL_LEDS
 	struct led_trigger led_trigger;
@@ -95,6 +98,8 @@ void rfkill_free(struct rfkill *rfkill);
 int rfkill_register(struct rfkill *rfkill);
 void rfkill_unregister(struct rfkill *rfkill);
 
+int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state);
+
 /**
  * rfkill_get_led_name - Get the LED trigger name for the button's LED.
  * This function might return a NULL pointer if registering of the
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index 3edc585dcfa6..4ae4486c77ea 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -62,19 +62,39 @@ static void rfkill_led_trigger(struct rfkill *rfkill,
 #endif /* CONFIG_RFKILL_LEDS */
 }
 
+static void update_rfkill_state(struct rfkill *rfkill)
+{
+	enum rfkill_state newstate;
+
+	if (rfkill->get_state) {
+		mutex_lock(&rfkill->mutex);
+		if (!rfkill->get_state(rfkill->data, &newstate))
+			rfkill->state = newstate;
+		mutex_unlock(&rfkill->mutex);
+	}
+}
+
 static int rfkill_toggle_radio(struct rfkill *rfkill,
 				enum rfkill_state state)
 {
 	int retval = 0;
+	enum rfkill_state oldstate, newstate;
+
+	oldstate = rfkill->state;
+
+	if (rfkill->get_state &&
+	    !rfkill->get_state(rfkill->data, &newstate))
+		rfkill->state = newstate;
 
 	if (state != rfkill->state) {
 		retval = rfkill->toggle_radio(rfkill->data, state);
-		if (!retval) {
+		if (!retval)
 			rfkill->state = state;
-			rfkill_led_trigger(rfkill, state);
-		}
 	}
 
+	if (rfkill->state != oldstate)
+		rfkill_led_trigger(rfkill, rfkill->state);
+
 	return retval;
 }
 
@@ -105,6 +125,32 @@ void rfkill_switch_all(enum rfkill_type type, enum rfkill_state state)
 }
 EXPORT_SYMBOL(rfkill_switch_all);
 
+/**
+ * rfkill_force_state - Force the internal rfkill radio state
+ * @rfkill: pointer to the rfkill class to modify.
+ * @state: the current radio state the class should be forced to.
+ *
+ * This function updates the internal state of the radio cached
+ * by the rfkill class.  It should be used when the driver gets
+ * a notification by the firmware/hardware of the current *real*
+ * state of the radio rfkill switch.
+ *
+ * It may not be called from an atomic context.
+ */
+int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state)
+{
+	if (state != RFKILL_STATE_OFF &&
+	    state != RFKILL_STATE_ON)
+		return -EINVAL;
+
+	mutex_lock(&rfkill->mutex);
+	rfkill->state = state;
+	mutex_unlock(&rfkill->mutex);
+
+	return 0;
+}
+EXPORT_SYMBOL(rfkill_force_state);
+
 static ssize_t rfkill_name_show(struct device *dev,
 				struct device_attribute *attr,
 				char *buf)
@@ -147,6 +193,7 @@ static ssize_t rfkill_state_show(struct device *dev,
 {
 	struct rfkill *rfkill = to_rfkill(dev);
 
+	update_rfkill_state(rfkill);
 	return sprintf(buf, "%d\n", rfkill->state);
 }
 
-- 
cgit v1.2.3


From 477576a073699783abb53ae14993d5d41c66301d Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:23:01 -0300
Subject: rfkill: add the WWAN radio type
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unfortunately, instead of adding a generic Wireless WAN type, a technology-
specific type (WiMAX) was added.  That's useless for other WWAN devices,
such as EDGE, UMTS, X-RTT and other such radios.

Add a WWAN rfkill type for generic wireless WAN devices.  No keys are added
as most devices really want to use KEY_WLAN for WWAN control (in a cycle of
none, WLAN, WWAN, WLAN+WWAN) and need no specific keycode added.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Cc: Iñaky Pérez-González <inaky.perez-gonzalez@intel.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/rfkill.h    | 2 ++
 net/rfkill/rfkill-input.c | 4 ++++
 net/rfkill/rfkill.c       | 3 +++
 3 files changed, 9 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index 844e96114861..c0cab7d37828 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -34,12 +34,14 @@
  * RFKILL_TYPE_BLUETOOTH: switch is on a bluetooth device.
  * RFKILL_TYPE_UWB: switch is on a ultra wideband device.
  * RFKILL_TYPE_WIMAX: switch is on a WiMAX device.
+ * RFKILL_TYPE_WWAN: switch is on a wireless WAN device.
  */
 enum rfkill_type {
 	RFKILL_TYPE_WLAN ,
 	RFKILL_TYPE_BLUETOOTH,
 	RFKILL_TYPE_UWB,
 	RFKILL_TYPE_WIMAX,
+	RFKILL_TYPE_WWAN,
 	RFKILL_TYPE_MAX,
 };
 
diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c
index 9d6c9255bf2c..29c13d308b31 100644
--- a/net/rfkill/rfkill-input.c
+++ b/net/rfkill/rfkill-input.c
@@ -101,6 +101,7 @@ static DEFINE_RFKILL_TASK(rfkill_wlan, RFKILL_TYPE_WLAN);
 static DEFINE_RFKILL_TASK(rfkill_bt, RFKILL_TYPE_BLUETOOTH);
 static DEFINE_RFKILL_TASK(rfkill_uwb, RFKILL_TYPE_UWB);
 static DEFINE_RFKILL_TASK(rfkill_wimax, RFKILL_TYPE_WIMAX);
+static DEFINE_RFKILL_TASK(rfkill_wwan, RFKILL_TYPE_WWAN);
 
 static void rfkill_event(struct input_handle *handle, unsigned int type,
 			unsigned int code, int data)
@@ -126,6 +127,9 @@ static void rfkill_event(struct input_handle *handle, unsigned int type,
 		switch (code) {
 		case SW_RFKILL_ALL:
 			/* EVERY radio type. data != 0 means radios ON */
+			rfkill_schedule_set(&rfkill_wwan,
+					    (data)? RFKILL_STATE_ON:
+						    RFKILL_STATE_OFF);
 			rfkill_schedule_set(&rfkill_wimax,
 					    (data)? RFKILL_STATE_ON:
 						    RFKILL_STATE_OFF);
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index 4ae4486c77ea..79f3bbb027ff 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -180,6 +180,9 @@ static ssize_t rfkill_type_show(struct device *dev,
 	case RFKILL_TYPE_WIMAX:
 		type = "wimax";
 		break;
+	case RFKILL_TYPE_WWAN:
+		type = "wwan";
+		break;
 	default:
 		BUG();
 	}
-- 
cgit v1.2.3


From 79399a8d1908f6a406e82d23c5a9937e1722ed3a Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:23:03 -0300
Subject: rfkill: add notifier chains support

Add a notifier chain for use by the rfkill class.  This notifier chain
signals the following events (more to be added when needed):

  1. rfkill: rfkill device state has changed

A pointer to the rfkill struct will be passed as a parameter.

The notifier message types have been added to include/linux/rfkill.h
instead of to include/linux/notifier.h in order to avoid the madness of
modifying a header used globally (and that triggers an almost full tree
rebuild every time it is touched) with information that is of interest only
to code that includes the rfkill.h header.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/rfkill.h |  7 +++++
 net/rfkill/rfkill.c    | 70 +++++++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 74 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index c0cab7d37828..98667becdee4 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -117,4 +117,11 @@ static inline char *rfkill_get_led_name(struct rfkill *rfkill)
 #endif
 }
 
+/* rfkill notification chain */
+#define RFKILL_STATE_CHANGED		0x0001	/* state of a normal rfkill
+						   switch has changed */
+
+int register_rfkill_notifier(struct notifier_block *nb);
+int unregister_rfkill_notifier(struct notifier_block *nb);
+
 #endif /* RFKILL_H */
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index fb566902030a..a561e350a70a 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -46,6 +46,49 @@ MODULE_PARM_DESC(default_state,
 
 static enum rfkill_state rfkill_states[RFKILL_TYPE_MAX];
 
+static BLOCKING_NOTIFIER_HEAD(rfkill_notifier_list);
+
+
+/**
+ * register_rfkill_notifier - Add notifier to rfkill notifier chain
+ * @nb: pointer to the new entry to add to the chain
+ *
+ * See blocking_notifier_chain_register() for return value and further
+ * observations.
+ *
+ * Adds a notifier to the rfkill notifier chain.  The chain will be
+ * called with a pointer to the relevant rfkill structure as a parameter,
+ * refer to include/linux/rfkill.h for the possible events.
+ *
+ * Notifiers added to this chain are to always return NOTIFY_DONE.  This
+ * chain is a blocking notifier chain: notifiers can sleep.
+ *
+ * Calls to this chain may have been done through a workqueue.  One must
+ * assume unordered asynchronous behaviour, there is no way to know if
+ * actions related to the event that generated the notification have been
+ * carried out already.
+ */
+int register_rfkill_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&rfkill_notifier_list, nb);
+}
+EXPORT_SYMBOL_GPL(register_rfkill_notifier);
+
+/**
+ * unregister_rfkill_notifier - remove notifier from rfkill notifier chain
+ * @nb: pointer to the entry to remove from the chain
+ *
+ * See blocking_notifier_chain_unregister() for return value and further
+ * observations.
+ *
+ * Removes a notifier from the rfkill notifier chain.
+ */
+int unregister_rfkill_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_unregister(&rfkill_notifier_list, nb);
+}
+EXPORT_SYMBOL_GPL(unregister_rfkill_notifier);
+
 
 static void rfkill_led_trigger(struct rfkill *rfkill,
 			       enum rfkill_state state)
@@ -62,14 +105,25 @@ static void rfkill_led_trigger(struct rfkill *rfkill,
 #endif /* CONFIG_RFKILL_LEDS */
 }
 
+static void notify_rfkill_state_change(struct rfkill *rfkill)
+{
+	blocking_notifier_call_chain(&rfkill_notifier_list,
+			RFKILL_STATE_CHANGED,
+			rfkill);
+}
+
 static void update_rfkill_state(struct rfkill *rfkill)
 {
-	enum rfkill_state newstate;
+	enum rfkill_state newstate, oldstate;
 
 	if (rfkill->get_state) {
 		mutex_lock(&rfkill->mutex);
-		if (!rfkill->get_state(rfkill->data, &newstate))
+		if (!rfkill->get_state(rfkill->data, &newstate)) {
+			oldstate = rfkill->state;
 			rfkill->state = newstate;
+			if (oldstate != newstate)
+				notify_rfkill_state_change(rfkill);
+		}
 		mutex_unlock(&rfkill->mutex);
 	}
 }
@@ -93,8 +147,10 @@ static int rfkill_toggle_radio(struct rfkill *rfkill,
 			rfkill->state = state;
 	}
 
-	if (force || rfkill->state != oldstate)
+	if (force || rfkill->state != oldstate) {
 		rfkill_led_trigger(rfkill, rfkill->state);
+		notify_rfkill_state_change(rfkill);
+	}
 
 	return retval;
 }
@@ -139,12 +195,20 @@ EXPORT_SYMBOL(rfkill_switch_all);
  */
 int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state)
 {
+	enum rfkill_state oldstate;
+
 	if (state != RFKILL_STATE_OFF &&
 	    state != RFKILL_STATE_ON)
 		return -EINVAL;
 
 	mutex_lock(&rfkill->mutex);
+
+	oldstate = rfkill->state;
 	rfkill->state = state;
+
+	if (state != oldstate)
+		notify_rfkill_state_change(rfkill);
+
 	mutex_unlock(&rfkill->mutex);
 
 	return 0;
-- 
cgit v1.2.3


From 5005657cbd0fd6f277f807c0612a6b6d4396a02c Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 23 Jun 2008 17:46:42 -0300
Subject: rfkill: rename the rfkill_state states and add block-locked state

The current naming of rfkill_state causes a lot of confusion: not only the
"kill" in rfkill suggests negative logic, but also the fact that rfkill cannot
turn anything on (it can just force something off or stop forcing something
off) is often forgotten.

Rename RFKILL_STATE_OFF to RFKILL_STATE_SOFT_BLOCKED (transmitter is blocked
and will not operate; state can be changed by a toggle_radio request), and
RFKILL_STATE_ON to RFKILL_STATE_UNBLOCKED (transmitter is not blocked, and may
operate).

Also, add a new third state, RFKILL_STATE_HARD_BLOCKED (transmitter is blocked
and will not operate; state cannot be changed through a toggle_radio request),
which is used by drivers to indicate a wireless transmiter was blocked by a
hardware rfkill line that accepts no overrides.

Keep the old names as #defines, but document them as deprecated.  This way,
drivers can be converted to the new names *and* verified to actually use rfkill
correctly one by one.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 Documentation/rfkill.txt  | 56 +++++++++++++++++++++++++++++------
 include/linux/rfkill.h    | 32 ++++++++++++++++++--
 net/rfkill/rfkill-input.c | 29 +++++++++---------
 net/rfkill/rfkill.c       | 75 ++++++++++++++++++++++++++++++++++++++---------
 4 files changed, 152 insertions(+), 40 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/rfkill.txt b/Documentation/rfkill.txt
index cf230c1ad9ef..5316cea95ce0 100644
--- a/Documentation/rfkill.txt
+++ b/Documentation/rfkill.txt
@@ -60,9 +60,20 @@ The second option provides an rfkill input handler. This handler will listen to
 all rfkill key events and will toggle the radio accordingly.  With this option
 enabled userspace could either do nothing or simply perform monitoring tasks.
 
-When a rfkill switch is in the RFKILL_STATE_ON, the wireless transmitter (radio
-TX circuit for example) is *enabled*.  When the rfkill switch is in the
-RFKILL_STATE_OFF, the wireless transmitter is to be *blocked* from operating.
+When a rfkill switch is in the RFKILL_STATE_UNBLOCKED, the wireless transmitter
+(radio TX circuit for example) is *enabled*.  When the rfkill switch is in the
+RFKILL_STATE_SOFT_BLOCKED or RFKILL_STATE_HARD_BLOCKED, the wireless
+transmitter is to be *blocked* from operating.
+
+RFKILL_STATE_SOFT_BLOCKED indicates that a call to toggle_radio() can change
+that state.  RFKILL_STATE_HARD_BLOCKED indicates that a call to toggle_radio()
+will not be able to change the state and will return with a suitable error if
+attempts are made to set the state to RFKILL_STATE_UNBLOCKED.
+
+RFKILL_STATE_HARD_BLOCKED is used by drivers to signal that the device is
+locked in the BLOCKED state by a hardwire rfkill line (typically an input pin
+that, when active, forces the transmitter to be disabled) which the driver
+CANNOT override.
 
 Full rfkill functionality requires two different subsystems to cooperate: the
 input layer and the rfkill class.  The input layer issues *commands* to the
@@ -122,10 +133,10 @@ Userspace input handlers (uevents) or kernel input handlers (rfkill-input):
 	  action).
 	* rfkill-input implements EPO by handling EV_SW SW_RFKILL_ALL 0
 	  (power off all transmitters) in a special way: it ignores any
-	  overrides and local state cache and forces all transmitters to
-	  the OFF state (including those which are already supposed to be
-	  OFF).  Note that the opposite event (power on all transmitters)
-	  is handled normally.
+	  overrides and local state cache and forces all transmitters to the
+	  RFKILL_STATE_SOFT_BLOCKED state (including those which are already
+	  supposed to be BLOCKED).  Note that the opposite event (power on all
+	  transmitters) is handled normally.
 
 Userspace uevent handler or kernel platform-specific drivers hooked to the
 rfkill notifier chain:
@@ -284,6 +295,19 @@ You should:
 	  YOU CAN ACCESS state DIRECTLY)
 	- rfkill_register()
 
+The only way to set a device to the RFKILL_STATE_HARD_BLOCKED state is through
+a suitable return of get_state() or through rfkill_force_state().
+
+When a device is in the RFKILL_STATE_HARD_BLOCKED state, the only way to switch
+it to a different state is through a suitable return of get_state() or through
+rfkill_force_state().
+
+If toggle_radio() is called to set a device to state RFKILL_STATE_SOFT_BLOCKED
+when that device is already at the RFKILL_STATE_HARD_BLOCKED state, it should
+not return an error.  Instead, it should try to double-block the transmitter,
+so that its state will change from RFKILL_STATE_HARD_BLOCKED to
+RFKILL_STATE_SOFT_BLOCKED should the hardware blocking cease.
+
 Please refer to the source for more documentation.
 
 ===============================================================================
@@ -322,13 +346,27 @@ change by writing to the "state" attribute in order for anything to happen.
 
 Take particular care to implement EV_SW SW_RFKILL_ALL properly.  When that
 switch is set to OFF, *every* rfkill device *MUST* be immediately put into the
-OFF state, no questions asked.
+RFKILL_STATE_SOFT_BLOCKED state, no questions asked.
 
 The following sysfs entries will be created:
 
 	name: Name assigned by driver to this key (interface or driver name).
 	type: Name of the key type ("wlan", "bluetooth", etc).
-	state: Current state of the key. 1: On, 0: Off.
+	state: Current state of the transmitter
+		0: RFKILL_STATE_SOFT_BLOCKED
+			transmitter is forced off, but you can override it
+			by a write to the state attribute, or through input
+			events (if rfkill-input is loaded).
+		1: RFKILL_STATE_UNBLOCKED
+			transmiter is NOT forced off, and may operate if
+			all other conditions for such operation are met
+			(such as interface is up and configured, etc).
+		2: RFKILL_STATE_HARD_BLOCKED
+			transmitter is forced off by something outside of
+			the driver's control.
+
+			You cannot set a device to this state through
+			writes to the state attribute.
 	claim: 1: Userspace handles events, 0: Kernel handles events
 
 Both the "state" and "claim" entries are also writable. For the "state" entry
diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index 98667becdee4..c5f6e54ec6ae 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -46,16 +46,25 @@ enum rfkill_type {
 };
 
 enum rfkill_state {
-	RFKILL_STATE_OFF	= 0,	/* Radio output blocked */
-	RFKILL_STATE_ON		= 1,	/* Radio output active */
+	RFKILL_STATE_SOFT_BLOCKED = 0,	/* Radio output blocked */
+	RFKILL_STATE_UNBLOCKED    = 1,	/* Radio output allowed */
+	RFKILL_STATE_HARD_BLOCKED = 2,	/* Output blocked, non-overrideable */
 };
 
+/*
+ * These are DEPRECATED, drivers using them should be verified to
+ * comply with the rfkill usage guidelines in Documentation/rfkill.txt
+ * and then converted to use the new names for rfkill_state
+ */
+#define RFKILL_STATE_OFF RFKILL_STATE_SOFT_BLOCKED
+#define RFKILL_STATE_ON  RFKILL_STATE_UNBLOCKED
+
 /**
  * struct rfkill - rfkill control structure.
  * @name: Name of the switch.
  * @type: Radio type which the button controls, the value stored
  *	here should be a value from enum rfkill_type.
- * @state: State of the switch, "ON" means radio can operate.
+ * @state: State of the switch, "UNBLOCKED" means radio can operate.
  * @user_claim_unsupported: Whether the hardware supports exclusive
  *	RF-kill control by userspace. Set this before registering.
  * @user_claim: Set when the switch is controlled exlusively by userspace.
@@ -63,8 +72,12 @@ enum rfkill_state {
  * @data: Pointer to the RF button drivers private data which will be
  *	passed along when toggling radio state.
  * @toggle_radio(): Mandatory handler to control state of the radio.
+ *	only RFKILL_STATE_SOFT_BLOCKED and RFKILL_STATE_UNBLOCKED are
+ *	valid parameters.
  * @get_state(): handler to read current radio state from hardware,
  *      may be called from atomic context, should return 0 on success.
+ *      Either this handler OR judicious use of rfkill_force_state() is
+ *      MANDATORY for any driver capable of RFKILL_STATE_HARD_BLOCKED.
  * @led_trigger: A LED trigger for this button's LED.
  * @dev: Device structure integrating the switch into device tree.
  * @node: Used to place switch into list of all switches known to the
@@ -102,6 +115,19 @@ void rfkill_unregister(struct rfkill *rfkill);
 
 int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state);
 
+/**
+ * rfkill_state_complement - return complementar state
+ * @state: state to return the complement of
+ *
+ * Returns RFKILL_STATE_SOFT_BLOCKED if @state is RFKILL_STATE_UNBLOCKED,
+ * returns RFKILL_STATE_UNBLOCKED otherwise.
+ */
+static inline enum rfkill_state rfkill_state_complement(enum rfkill_state state)
+{
+	return (state == RFKILL_STATE_UNBLOCKED) ?
+		RFKILL_STATE_SOFT_BLOCKED : RFKILL_STATE_UNBLOCKED;
+}
+
 /**
  * rfkill_get_led_name - Get the LED trigger name for the button's LED.
  * This function might return a NULL pointer if registering of the
diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c
index 5d4c8b2446f7..8aa822730145 100644
--- a/net/rfkill/rfkill-input.c
+++ b/net/rfkill/rfkill-input.c
@@ -84,7 +84,8 @@ static void rfkill_schedule_toggle(struct rfkill_task *task)
 	spin_lock_irqsave(&task->lock, flags);
 
 	if (time_after(jiffies, task->last + msecs_to_jiffies(200))) {
-		task->desired_state = !task->desired_state;
+		task->desired_state =
+				rfkill_state_complement(task->desired_state);
 		task->last = jiffies;
 		schedule_work(&task->work);
 	}
@@ -92,14 +93,14 @@ static void rfkill_schedule_toggle(struct rfkill_task *task)
 	spin_unlock_irqrestore(&task->lock, flags);
 }
 
-#define DEFINE_RFKILL_TASK(n, t)			\
-	struct rfkill_task n = {			\
-		.work = __WORK_INITIALIZER(n.work,	\
-				rfkill_task_handler),	\
-		.type = t,				\
-		.mutex = __MUTEX_INITIALIZER(n.mutex),	\
-		.lock = __SPIN_LOCK_UNLOCKED(n.lock),	\
-		.desired_state = RFKILL_STATE_ON,	\
+#define DEFINE_RFKILL_TASK(n, t)				\
+	struct rfkill_task n = {				\
+		.work = __WORK_INITIALIZER(n.work,		\
+				rfkill_task_handler),		\
+		.type = t,					\
+		.mutex = __MUTEX_INITIALIZER(n.mutex),		\
+		.lock = __SPIN_LOCK_UNLOCKED(n.lock),		\
+		.desired_state = RFKILL_STATE_UNBLOCKED,	\
 	}
 
 static DEFINE_RFKILL_TASK(rfkill_wlan, RFKILL_TYPE_WLAN);
@@ -135,15 +136,15 @@ static void rfkill_event(struct input_handle *handle, unsigned int type,
 			/* handle EPO (emergency power off) through shortcut */
 			if (data) {
 				rfkill_schedule_set(&rfkill_wwan,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_wimax,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_uwb,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_bt,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 				rfkill_schedule_set(&rfkill_wlan,
-						    RFKILL_STATE_ON);
+						    RFKILL_STATE_UNBLOCKED);
 			} else
 				rfkill_schedule_epo();
 			break;
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index 7d07175c407f..ce0e23148cdd 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -39,7 +39,7 @@ MODULE_LICENSE("GPL");
 static LIST_HEAD(rfkill_list);	/* list of registered rf switches */
 static DEFINE_MUTEX(rfkill_mutex);
 
-static unsigned int rfkill_default_state = RFKILL_STATE_ON;
+static unsigned int rfkill_default_state = RFKILL_STATE_UNBLOCKED;
 module_param_named(default_state, rfkill_default_state, uint, 0444);
 MODULE_PARM_DESC(default_state,
 		 "Default initial state for all radio types, 0 = radio off");
@@ -98,7 +98,7 @@ static void rfkill_led_trigger(struct rfkill *rfkill,
 
 	if (!led->name)
 		return;
-	if (state == RFKILL_STATE_OFF)
+	if (state != RFKILL_STATE_UNBLOCKED)
 		led_trigger_event(led, LED_OFF);
 	else
 		led_trigger_event(led, LED_FULL);
@@ -128,6 +128,28 @@ static void update_rfkill_state(struct rfkill *rfkill)
 	}
 }
 
+/**
+ * rfkill_toggle_radio - wrapper for toggle_radio hook
+ * calls toggle_radio taking into account a lot of "small"
+ * details.
+ * @rfkill: the rfkill struct to use
+ * @force: calls toggle_radio even if cache says it is not needed,
+ *	and also makes sure notifications of the state will be
+ *	sent even if it didn't change
+ * @state: the new state to call toggle_radio() with
+ *
+ * This wrappen protects and enforces the API for toggle_radio
+ * calls.  Note that @force cannot override a (possibly cached)
+ * state of RFKILL_STATE_HARD_BLOCKED.  Any device making use of
+ * RFKILL_STATE_HARD_BLOCKED implements either get_state() or
+ * rfkill_force_state(), so the cache either is bypassed or valid.
+ *
+ * Note that we do call toggle_radio for RFKILL_STATE_SOFT_BLOCKED
+ * even if the radio is in RFKILL_STATE_HARD_BLOCKED state, so as to
+ * give the driver a hint that it should double-BLOCK the transmitter.
+ *
+ * Caller must have aquired rfkill_mutex.
+ */
 static int rfkill_toggle_radio(struct rfkill *rfkill,
 				enum rfkill_state state,
 				int force)
@@ -141,9 +163,28 @@ static int rfkill_toggle_radio(struct rfkill *rfkill,
 	    !rfkill->get_state(rfkill->data, &newstate))
 		rfkill->state = newstate;
 
+	switch (state) {
+	case RFKILL_STATE_HARD_BLOCKED:
+		/* typically happens when refreshing hardware state,
+		 * such as on resume */
+		state = RFKILL_STATE_SOFT_BLOCKED;
+		break;
+	case RFKILL_STATE_UNBLOCKED:
+		/* force can't override this, only rfkill_force_state() can */
+		if (rfkill->state == RFKILL_STATE_HARD_BLOCKED)
+			return -EPERM;
+		break;
+	case RFKILL_STATE_SOFT_BLOCKED:
+		/* nothing to do, we want to give drivers the hint to double
+		 * BLOCK even a transmitter that is already in state
+		 * RFKILL_STATE_HARD_BLOCKED */
+		break;
+	}
+
 	if (force || state != rfkill->state) {
 		retval = rfkill->toggle_radio(rfkill->data, state);
-		if (!retval)
+		/* never allow a HARD->SOFT downgrade! */
+		if (!retval && rfkill->state != RFKILL_STATE_HARD_BLOCKED)
 			rfkill->state = state;
 	}
 
@@ -184,7 +225,7 @@ EXPORT_SYMBOL(rfkill_switch_all);
 /**
  * rfkill_epo - emergency power off all transmitters
  *
- * This kicks all rfkill devices to RFKILL_STATE_OFF, ignoring
+ * This kicks all rfkill devices to RFKILL_STATE_SOFT_BLOCKED, ignoring
  * everything in its path but rfkill_mutex.
  */
 void rfkill_epo(void)
@@ -193,7 +234,7 @@ void rfkill_epo(void)
 
 	mutex_lock(&rfkill_mutex);
 	list_for_each_entry(rfkill, &rfkill_list, node) {
-		rfkill_toggle_radio(rfkill, RFKILL_STATE_OFF, 1);
+		rfkill_toggle_radio(rfkill, RFKILL_STATE_SOFT_BLOCKED, 1);
 	}
 	mutex_unlock(&rfkill_mutex);
 }
@@ -215,8 +256,9 @@ int rfkill_force_state(struct rfkill *rfkill, enum rfkill_state state)
 {
 	enum rfkill_state oldstate;
 
-	if (state != RFKILL_STATE_OFF &&
-	    state != RFKILL_STATE_ON)
+	if (state != RFKILL_STATE_SOFT_BLOCKED &&
+	    state != RFKILL_STATE_UNBLOCKED &&
+	    state != RFKILL_STATE_HARD_BLOCKED)
 		return -EINVAL;
 
 	mutex_lock(&rfkill->mutex);
@@ -290,11 +332,14 @@ static ssize_t rfkill_state_store(struct device *dev,
 	if (!capable(CAP_NET_ADMIN))
 		return -EPERM;
 
+	/* RFKILL_STATE_HARD_BLOCKED is illegal here... */
+	if (state != RFKILL_STATE_UNBLOCKED &&
+	    state != RFKILL_STATE_SOFT_BLOCKED)
+		return -EINVAL;
+
 	if (mutex_lock_interruptible(&rfkill->mutex))
 		return -ERESTARTSYS;
-	error = rfkill_toggle_radio(rfkill,
-			state ? RFKILL_STATE_ON : RFKILL_STATE_OFF,
-			0);
+	error = rfkill_toggle_radio(rfkill, state, 0);
 	mutex_unlock(&rfkill->mutex);
 
 	return error ? error : count;
@@ -373,7 +418,8 @@ static int rfkill_suspend(struct device *dev, pm_message_t state)
 			update_rfkill_state(rfkill);
 
 			mutex_lock(&rfkill->mutex);
-			rfkill->toggle_radio(rfkill->data, RFKILL_STATE_OFF);
+			rfkill->toggle_radio(rfkill->data,
+						RFKILL_STATE_SOFT_BLOCKED);
 			mutex_unlock(&rfkill->mutex);
 		}
 
@@ -470,7 +516,7 @@ static void rfkill_remove_switch(struct rfkill *rfkill)
 {
 	mutex_lock(&rfkill_mutex);
 	list_del_init(&rfkill->node);
-	rfkill_toggle_radio(rfkill, RFKILL_STATE_OFF, 1);
+	rfkill_toggle_radio(rfkill, RFKILL_STATE_SOFT_BLOCKED, 1);
 	mutex_unlock(&rfkill_mutex);
 }
 
@@ -610,8 +656,9 @@ static int __init rfkill_init(void)
 	int error;
 	int i;
 
-	if (rfkill_default_state != RFKILL_STATE_OFF &&
-	    rfkill_default_state != RFKILL_STATE_ON)
+	/* RFKILL_STATE_HARD_BLOCKED is illegal here... */
+	if (rfkill_default_state != RFKILL_STATE_SOFT_BLOCKED &&
+	    rfkill_default_state != RFKILL_STATE_UNBLOCKED)
 		return -EINVAL;
 
 	for (i = 0; i < ARRAY_SIZE(rfkill_states); i++)
-- 
cgit v1.2.3


From f2df38596a81b6c24f4586b0b4befeaebf3e02db Mon Sep 17 00:00:00 2001
From: Assaf Krauss <assaf.krauss@intel.com>
Date: Sun, 15 Jun 2008 18:23:29 +0300
Subject: mac80211: 11h Infrastructure - Parsing

This patch introduces parsing of 11h and 11d related elements from incoming
management frames.

Signed-off-by: Assaf Krauss <assaf.krauss@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h  | 54 +++++++++++++++++++++++++++++++++++++++++++---
 net/mac80211/ieee80211_i.h |  9 ++++++++
 net/mac80211/mlme.c        | 19 ++++++++++++++++
 3 files changed, 79 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 2998e3b5f166..8546f09e462c 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -469,6 +469,40 @@ struct ieee80211s_hdr {
 	u8 eaddr3[6];
 } __attribute__ ((packed));
 
+/**
+ * struct ieee80211_quiet_ie
+ *
+ * This structure refers to "Quiet information element"
+ */
+struct ieee80211_quiet_ie {
+	u8 count;
+	u8 period;
+	__le16 duration;
+	__le16 offset;
+} __attribute__ ((packed));
+
+/**
+ * struct ieee80211_msrment_ie
+ *
+ * This structure refers to "Measurement Request/Report information element"
+ */
+struct ieee80211_msrment_ie {
+	u8 token;
+	u8 mode;
+	u8 type;
+	u8 request[0];
+} __attribute__ ((packed));
+
+/**
+ * struct ieee80211_channel_sw_ie
+ *
+ * This structure refers to "Channel Switch Announcement information element"
+ */
+struct ieee80211_channel_sw_ie {
+	u8 mode;
+	u8 new_ch_num;
+	u8 count;
+} __attribute__ ((packed));
 
 struct ieee80211_mgmt {
 	__le16 frame_control;
@@ -544,10 +578,15 @@ struct ieee80211_mgmt {
 					u8 action_code;
 					u8 element_id;
 					u8 length;
-					u8 switch_mode;
-					u8 new_chan;
-					u8 switch_count;
+					struct ieee80211_channel_sw_ie sw_elem;
 				} __attribute__((packed)) chan_switch;
+				struct{
+					u8 action_code;
+					u8 dialog_token;
+					u8 element_id;
+					u8 length;
+					struct ieee80211_msrment_ie msr_elem;
+				} __attribute__((packed)) measurement;
 				struct{
 					u8 action_code;
 					u8 dialog_token;
@@ -875,6 +914,15 @@ enum ieee80211_category {
 	WLAN_CATEGORY_WMM = 17,
 };
 
+/* SPECTRUM_MGMT action code */
+enum ieee80211_spectrum_mgmt_actioncode {
+	WLAN_ACTION_SPCT_MSR_REQ = 0,
+	WLAN_ACTION_SPCT_MSR_RPRT = 1,
+	WLAN_ACTION_SPCT_TPC_REQ = 2,
+	WLAN_ACTION_SPCT_TPC_RPRT = 3,
+	WLAN_ACTION_SPCT_CHL_SWITCH = 4,
+};
+
 /* BACK action code */
 enum ieee80211_back_actioncode {
 	WLAN_ACTION_ADDBA_REQ = 0,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 80a9e7c07b47..af352c05c983 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -791,6 +791,10 @@ struct ieee802_11_elems {
 	u8 *preq;
 	u8 *prep;
 	u8 *perr;
+	u8 *ch_switch_elem;
+	u8 *country_elem;
+	u8 *pwr_constr_elem;
+	u8 *quiet_elem; 	/* first quite element */
 
 	/* length of them, respectively */
 	u8 ssid_len;
@@ -815,6 +819,11 @@ struct ieee802_11_elems {
 	u8 preq_len;
 	u8 prep_len;
 	u8 perr_len;
+	u8 ch_switch_elem_len;
+	u8 country_elem_len;
+	u8 pwr_constr_elem_len;
+	u8 quiet_elem_len;
+	u8 num_of_quiet_elem;	/* can be more the one */
 };
 
 static inline struct ieee80211_local *hw_to_local(
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index e06d6450f215..32453561fe32 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -204,6 +204,25 @@ void ieee802_11_parse_elems(u8 *start, size_t len,
 			elems->perr = pos;
 			elems->perr_len = elen;
 			break;
+		case WLAN_EID_CHANNEL_SWITCH:
+			elems->ch_switch_elem = pos;
+			elems->ch_switch_elem_len = elen;
+			break;
+		case WLAN_EID_QUIET:
+			if (!elems->quiet_elem) {
+				elems->quiet_elem = pos;
+				elems->quiet_elem_len = elen;
+			}
+			elems->num_of_quiet_elem++;
+			break;
+		case WLAN_EID_COUNTRY:
+			elems->country_elem = pos;
+			elems->country_elem_len = elen;
+			break;
+		case WLAN_EID_PWR_CONSTRAINT:
+			elems->pwr_constr_elem = pos;
+			elems->pwr_constr_elem_len = elen;
+			break;
 		default:
 			break;
 		}
-- 
cgit v1.2.3


From b662348662f9661f9259c7186c1bdb65620045f1 Mon Sep 17 00:00:00 2001
From: Assaf Krauss <assaf.krauss@intel.com>
Date: Mon, 16 Jun 2008 16:09:49 +0300
Subject: mac80211: 11h - Handling measurement request

This patch handles the 11h measurement request information element.
This is minimal requested implementation - refuse measurement.

Signed-off-by: Assaf Krauss <assaf.krauss@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h | 10 ++++++
 net/mac80211/mlme.c       | 78 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 88 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 8546f09e462c..cffd6d0094f9 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -739,11 +739,21 @@ struct ieee80211_ht_addt_info {
 #define WLAN_CAPABILITY_SHORT_PREAMBLE	(1<<5)
 #define WLAN_CAPABILITY_PBCC		(1<<6)
 #define WLAN_CAPABILITY_CHANNEL_AGILITY	(1<<7)
+
 /* 802.11h */
 #define WLAN_CAPABILITY_SPECTRUM_MGMT	(1<<8)
 #define WLAN_CAPABILITY_QOS		(1<<9)
 #define WLAN_CAPABILITY_SHORT_SLOT_TIME	(1<<10)
 #define WLAN_CAPABILITY_DSSS_OFDM	(1<<13)
+/* measurement */
+#define IEEE80211_SPCT_MSR_RPRT_MODE_LATE	(1<<0)
+#define IEEE80211_SPCT_MSR_RPRT_MODE_INCAPABLE	(1<<1)
+#define IEEE80211_SPCT_MSR_RPRT_MODE_REFUSED	(1<<2)
+
+#define IEEE80211_SPCT_MSR_RPRT_TYPE_BASIC	0
+#define IEEE80211_SPCT_MSR_RPRT_TYPE_CCA	1
+#define IEEE80211_SPCT_MSR_RPRT_TYPE_RPI	2
+
 
 /* 802.11g ERP information element */
 #define WLAN_ERP_NON_ERP_PRESENT (1<<0)
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 32453561fe32..6d9a4facf9dd 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -1720,6 +1720,71 @@ void ieee80211_sta_tear_down_BA_sessions(struct net_device *dev, u8 *addr)
 	}
 }
 
+static void ieee80211_send_refuse_measurement_request(struct net_device *dev,
+					struct ieee80211_msrment_ie *request_ie,
+					const u8 *da, const u8 *bssid,
+					u8 dialog_token)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct sk_buff *skb;
+	struct ieee80211_mgmt *msr_report;
+
+	skb = dev_alloc_skb(sizeof(*msr_report) + local->hw.extra_tx_headroom +
+				sizeof(struct ieee80211_msrment_ie));
+
+	if (!skb) {
+		printk(KERN_ERR "%s: failed to allocate buffer for "
+				"measurement report frame\n", dev->name);
+		return;
+	}
+
+	skb_reserve(skb, local->hw.extra_tx_headroom);
+	msr_report = (struct ieee80211_mgmt *)skb_put(skb, 24);
+	memset(msr_report, 0, 24);
+	memcpy(msr_report->da, da, ETH_ALEN);
+	memcpy(msr_report->sa, dev->dev_addr, ETH_ALEN);
+	memcpy(msr_report->bssid, bssid, ETH_ALEN);
+	msr_report->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT,
+						IEEE80211_STYPE_ACTION);
+
+	skb_put(skb, 1 + sizeof(msr_report->u.action.u.measurement));
+	msr_report->u.action.category = WLAN_CATEGORY_SPECTRUM_MGMT;
+	msr_report->u.action.u.measurement.action_code =
+				WLAN_ACTION_SPCT_MSR_RPRT;
+	msr_report->u.action.u.measurement.dialog_token = dialog_token;
+
+	msr_report->u.action.u.measurement.element_id = WLAN_EID_MEASURE_REPORT;
+	msr_report->u.action.u.measurement.length =
+			sizeof(struct ieee80211_msrment_ie);
+
+	memset(&msr_report->u.action.u.measurement.msr_elem, 0,
+		sizeof(struct ieee80211_msrment_ie));
+	msr_report->u.action.u.measurement.msr_elem.token = request_ie->token;
+	msr_report->u.action.u.measurement.msr_elem.mode |=
+			IEEE80211_SPCT_MSR_RPRT_MODE_REFUSED;
+	msr_report->u.action.u.measurement.msr_elem.type = request_ie->type;
+
+	ieee80211_sta_tx(dev, skb, 0);
+}
+
+static void ieee80211_sta_process_measurement_req(struct net_device *dev,
+						struct ieee80211_mgmt *mgmt,
+						size_t len)
+{
+	/*
+	 * Ignoring measurement request is spec violation.
+	 * Mandatory measurements must be reported optional
+	 * measurements might be refused or reported incapable
+	 * For now just refuse
+	 * TODO: Answer basic measurement as unmeasured
+	 */
+	ieee80211_send_refuse_measurement_request(dev,
+			&mgmt->u.action.u.measurement.msr_elem,
+			mgmt->sa, mgmt->bssid,
+			mgmt->u.action.u.measurement.dialog_token);
+}
+
+
 static void ieee80211_rx_mgmt_auth(struct net_device *dev,
 				   struct ieee80211_if_sta *ifsta,
 				   struct ieee80211_mgmt *mgmt,
@@ -3044,11 +3109,24 @@ static void ieee80211_rx_mgmt_action(struct net_device *dev,
 				     struct ieee80211_rx_status *rx_status)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
 
 	if (len < IEEE80211_MIN_ACTION_SIZE)
 		return;
 
 	switch (mgmt->u.action.category) {
+	case WLAN_CATEGORY_SPECTRUM_MGMT:
+		if (local->hw.conf.channel->band != IEEE80211_BAND_5GHZ)
+			break;
+		switch (mgmt->u.action.u.chan_switch.action_code) {
+		case WLAN_ACTION_SPCT_MSR_REQ:
+			if (len < (IEEE80211_MIN_ACTION_SIZE +
+				   sizeof(mgmt->u.action.u.measurement)))
+				break;
+			ieee80211_sta_process_measurement_req(dev, mgmt, len);
+			break;
+		}
+		break;
 	case WLAN_CATEGORY_BACK:
 		switch (mgmt->u.action.u.addba_req.action_code) {
 		case WLAN_ACTION_ADDBA_REQ:
-- 
cgit v1.2.3


From 1bdad606338debc6384b2844f1b53cc436b3ac90 Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Tue, 3 Jun 2008 14:09:53 +0100
Subject: [GFS2] Remove remote lock dropping code

There are several reasons why this is undesirable:

 1. It never happens during normal operation anyway
 2. If it does happen it causes performance to be very, very poor
 3. It isn't likely to solve the original problem (memory shortage
    on remote DLM node) it was supposed to solve
 4. It uses a bunch of arbitrary constants which are unlikely to be
    correct for any particular situation and for which the tuning seems
    to be a black art.
 5. In an N node cluster, only 1/N of the dropped locked will actually
    contribute to solving the problem on average.

So all in all we are better off without it. This also makes merging
the lock_dlm module into GFS2 a bit easier.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
---
 fs/gfs2/gfs2.h                 |  5 -----
 fs/gfs2/glock.c                | 12 +++---------
 fs/gfs2/glock.h                |  2 +-
 fs/gfs2/locking/dlm/lock_dlm.h |  3 ---
 fs/gfs2/locking/dlm/mount.c    |  3 ---
 fs/gfs2/locking/dlm/sysfs.c    | 13 -------------
 fs/gfs2/locking/dlm/thread.c   | 19 -------------------
 fs/gfs2/ops_fstype.c           |  2 +-
 fs/gfs2/ops_super.c            |  2 +-
 fs/gfs2/sys.c                  | 14 --------------
 include/linux/lm_interface.h   |  4 ----
 11 files changed, 6 insertions(+), 73 deletions(-)

(limited to 'include/linux')

diff --git a/fs/gfs2/gfs2.h b/fs/gfs2/gfs2.h
index 3bb11c0f8b56..ef606e3a5cf4 100644
--- a/fs/gfs2/gfs2.h
+++ b/fs/gfs2/gfs2.h
@@ -15,11 +15,6 @@ enum {
 	CREATE = 1,
 };
 
-enum {
-	NO_WAIT = 0,
-	WAIT = 1,
-};
-
 enum {
 	NO_FORCE = 0,
 	FORCE = 1,
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index be7ed503f012..8d5450f3c3ef 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -1316,11 +1316,6 @@ void gfs2_glock_cb(void *cb_data, unsigned int type, void *data)
 			wake_up_process(sdp->sd_recoverd_process);
 		return;
 
-	case LM_CB_DROPLOCKS:
-		gfs2_gl_hash_clear(sdp, NO_WAIT);
-		gfs2_quota_scan(sdp);
-		return;
-
 	default:
 		gfs2_assert_warn(sdp, 0);
 		return;
@@ -1508,11 +1503,10 @@ static void clear_glock(struct gfs2_glock *gl)
  * @sdp: the filesystem
  * @wait: wait until it's all gone
  *
- * Called when unmounting the filesystem, or when inter-node lock manager
- * requests DROPLOCKS because it is running out of capacity.
+ * Called when unmounting the filesystem.
  */
 
-void gfs2_gl_hash_clear(struct gfs2_sbd *sdp, int wait)
+void gfs2_gl_hash_clear(struct gfs2_sbd *sdp)
 {
 	unsigned long t;
 	unsigned int x;
@@ -1527,7 +1521,7 @@ void gfs2_gl_hash_clear(struct gfs2_sbd *sdp, int wait)
 				cont = 1;
 		}
 
-		if (!wait || !cont)
+		if (!cont)
 			break;
 
 		if (time_after_eq(jiffies,
diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h
index 7389f8ef0a31..971d92af70fc 100644
--- a/fs/gfs2/glock.h
+++ b/fs/gfs2/glock.h
@@ -132,7 +132,7 @@ void gfs2_lvb_unhold(struct gfs2_glock *gl);
 void gfs2_glock_cb(void *cb_data, unsigned int type, void *data);
 void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl);
 void gfs2_reclaim_glock(struct gfs2_sbd *sdp);
-void gfs2_gl_hash_clear(struct gfs2_sbd *sdp, int wait);
+void gfs2_gl_hash_clear(struct gfs2_sbd *sdp);
 
 int __init gfs2_glock_init(void);
 void gfs2_glock_exit(void);
diff --git a/fs/gfs2/locking/dlm/lock_dlm.h b/fs/gfs2/locking/dlm/lock_dlm.h
index ad944c64eab1..845a27fd303e 100644
--- a/fs/gfs2/locking/dlm/lock_dlm.h
+++ b/fs/gfs2/locking/dlm/lock_dlm.h
@@ -79,9 +79,6 @@ struct gdlm_ls {
 	wait_queue_head_t	wait_control;
 	struct task_struct	*thread;
 	wait_queue_head_t	thread_wait;
-	unsigned long		drop_time;
-	int			drop_locks_count;
-	int			drop_locks_period;
 };
 
 enum {
diff --git a/fs/gfs2/locking/dlm/mount.c b/fs/gfs2/locking/dlm/mount.c
index 0628520a445f..fa31c54c2e67 100644
--- a/fs/gfs2/locking/dlm/mount.c
+++ b/fs/gfs2/locking/dlm/mount.c
@@ -22,8 +22,6 @@ static struct gdlm_ls *init_gdlm(lm_callback_t cb, struct gfs2_sbd *sdp,
 	if (!ls)
 		return NULL;
 
-	ls->drop_locks_count = GDLM_DROP_COUNT;
-	ls->drop_locks_period = GDLM_DROP_PERIOD;
 	ls->fscb = cb;
 	ls->sdp = sdp;
 	ls->fsflags = flags;
@@ -33,7 +31,6 @@ static struct gdlm_ls *init_gdlm(lm_callback_t cb, struct gfs2_sbd *sdp,
 	INIT_LIST_HEAD(&ls->all_locks);
 	init_waitqueue_head(&ls->thread_wait);
 	init_waitqueue_head(&ls->wait_control);
-	ls->drop_time = jiffies;
 	ls->jid = -1;
 
 	strncpy(buf, table_name, 256);
diff --git a/fs/gfs2/locking/dlm/sysfs.c b/fs/gfs2/locking/dlm/sysfs.c
index a4ff271df9ee..4ec571c3d8a9 100644
--- a/fs/gfs2/locking/dlm/sysfs.c
+++ b/fs/gfs2/locking/dlm/sysfs.c
@@ -114,17 +114,6 @@ static ssize_t recover_status_show(struct gdlm_ls *ls, char *buf)
 	return sprintf(buf, "%d\n", ls->recover_jid_status);
 }
 
-static ssize_t drop_count_show(struct gdlm_ls *ls, char *buf)
-{
-	return sprintf(buf, "%d\n", ls->drop_locks_count);
-}
-
-static ssize_t drop_count_store(struct gdlm_ls *ls, const char *buf, size_t len)
-{
-	ls->drop_locks_count = simple_strtol(buf, NULL, 0);
-	return len;
-}
-
 struct gdlm_attr {
 	struct attribute attr;
 	ssize_t (*show)(struct gdlm_ls *, char *);
@@ -144,7 +133,6 @@ GDLM_ATTR(first_done,     0444, first_done_show,     NULL);
 GDLM_ATTR(recover,        0644, recover_show,        recover_store);
 GDLM_ATTR(recover_done,   0444, recover_done_show,   NULL);
 GDLM_ATTR(recover_status, 0444, recover_status_show, NULL);
-GDLM_ATTR(drop_count,     0644, drop_count_show,     drop_count_store);
 
 static struct attribute *gdlm_attrs[] = {
 	&gdlm_attr_proto_name.attr,
@@ -157,7 +145,6 @@ static struct attribute *gdlm_attrs[] = {
 	&gdlm_attr_recover.attr,
 	&gdlm_attr_recover_done.attr,
 	&gdlm_attr_recover_status.attr,
-	&gdlm_attr_drop_count.attr,
 	NULL,
 };
 
diff --git a/fs/gfs2/locking/dlm/thread.c b/fs/gfs2/locking/dlm/thread.c
index f30350abd62f..38823efd698c 100644
--- a/fs/gfs2/locking/dlm/thread.c
+++ b/fs/gfs2/locking/dlm/thread.c
@@ -20,19 +20,6 @@ static inline int no_work(struct gdlm_ls *ls)
 	return ret;
 }
 
-static inline int check_drop(struct gdlm_ls *ls)
-{
-	if (!ls->drop_locks_count)
-		return 0;
-
-	if (time_after(jiffies, ls->drop_time + ls->drop_locks_period * HZ)) {
-		ls->drop_time = jiffies;
-		if (ls->all_locks_count >= ls->drop_locks_count)
-			return 1;
-	}
-	return 0;
-}
-
 static int gdlm_thread(void *data)
 {
 	struct gdlm_ls *ls = (struct gdlm_ls *) data;
@@ -52,12 +39,6 @@ static int gdlm_thread(void *data)
 			gdlm_do_lock(lp);
 			spin_lock(&ls->async_lock);
 		}
-		/* Does this ever happen these days? I hope not anyway */
-		if (check_drop(ls)) {
-			spin_unlock(&ls->async_lock);
-			ls->fscb(ls->sdp, LM_CB_DROPLOCKS, NULL);
-			spin_lock(&ls->async_lock);
-		}
 		spin_unlock(&ls->async_lock);
 	}
 
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 9bd97c5543bd..6ba69dd1a729 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -874,7 +874,7 @@ fail_sb:
 fail_locking:
 	init_locking(sdp, &mount_gh, UNDO);
 fail_lm:
-	gfs2_gl_hash_clear(sdp, WAIT);
+	gfs2_gl_hash_clear(sdp);
 	gfs2_lm_unmount(sdp);
 	while (invalidate_inodes(sb))
 		yield();
diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c
index 66907922109f..f66ea0f7a356 100644
--- a/fs/gfs2/ops_super.c
+++ b/fs/gfs2/ops_super.c
@@ -126,7 +126,7 @@ static void gfs2_put_super(struct super_block *sb)
 	gfs2_clear_rgrpd(sdp);
 	gfs2_jindex_free(sdp);
 	/*  Take apart glock structures and buffer lists  */
-	gfs2_gl_hash_clear(sdp, WAIT);
+	gfs2_gl_hash_clear(sdp);
 	/*  Unmount the locking protocol  */
 	gfs2_lm_unmount(sdp);
 
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 9ab9fc85ecd0..6f7e2e5858e0 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -110,18 +110,6 @@ static ssize_t statfs_sync_store(struct gfs2_sbd *sdp, const char *buf,
 	return len;
 }
 
-static ssize_t shrink_store(struct gfs2_sbd *sdp, const char *buf, size_t len)
-{
-	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
-
-	if (simple_strtol(buf, NULL, 0) != 1)
-		return -EINVAL;
-
-	gfs2_gl_hash_clear(sdp, NO_WAIT);
-	return len;
-}
-
 static ssize_t quota_sync_store(struct gfs2_sbd *sdp, const char *buf,
 				size_t len)
 {
@@ -175,7 +163,6 @@ static struct gfs2_attr gfs2_attr_##name = __ATTR(name, mode, show, store)
 GFS2_ATTR(id,                  0444, id_show,       NULL);
 GFS2_ATTR(fsname,              0444, fsname_show,   NULL);
 GFS2_ATTR(freeze,              0644, freeze_show,   freeze_store);
-GFS2_ATTR(shrink,              0200, NULL,          shrink_store);
 GFS2_ATTR(withdraw,            0644, withdraw_show, withdraw_store);
 GFS2_ATTR(statfs_sync,         0200, NULL,          statfs_sync_store);
 GFS2_ATTR(quota_sync,          0200, NULL,          quota_sync_store);
@@ -186,7 +173,6 @@ static struct attribute *gfs2_attrs[] = {
 	&gfs2_attr_id.attr,
 	&gfs2_attr_fsname.attr,
 	&gfs2_attr_freeze.attr,
-	&gfs2_attr_shrink.attr,
 	&gfs2_attr_withdraw.attr,
 	&gfs2_attr_statfs_sync.attr,
 	&gfs2_attr_quota_sync.attr,
diff --git a/include/linux/lm_interface.h b/include/linux/lm_interface.h
index f274997bc283..d0a7112b9719 100644
--- a/include/linux/lm_interface.h
+++ b/include/linux/lm_interface.h
@@ -138,9 +138,6 @@ typedef void (*lm_callback_t) (void *ptr, unsigned int type, void *data);
  * LM_CB_NEED_RECOVERY
  * The given journal needs to be recovered.
  *
- * LM_CB_DROPLOCKS
- * Reduce the number of cached locks.
- *
  * LM_CB_ASYNC
  * The given lock has been granted.
  */
@@ -149,7 +146,6 @@ typedef void (*lm_callback_t) (void *ptr, unsigned int type, void *data);
 #define LM_CB_NEED_D		258
 #define LM_CB_NEED_S		259
 #define LM_CB_NEED_RECOVERY	260
-#define LM_CB_DROPLOCKS		261
 #define LM_CB_ASYNC		262
 
 /*
-- 
cgit v1.2.3


From b2cad26cfc2091050574a460b304ed103a35dbda Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Tue, 3 Jun 2008 14:34:14 +0100
Subject: [GFS2] Remove obsolete conversion deadlock avoidance code

This is only used by GFS1 so can be removed.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
---
 fs/gfs2/locking/dlm/lock.c   | 23 +----------------------
 include/linux/lm_interface.h |  2 --
 2 files changed, 1 insertion(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/fs/gfs2/locking/dlm/lock.c b/fs/gfs2/locking/dlm/lock.c
index 871ffc9578f2..894df4567a03 100644
--- a/fs/gfs2/locking/dlm/lock.c
+++ b/fs/gfs2/locking/dlm/lock.c
@@ -80,7 +80,6 @@ static void process_complete(struct gdlm_lock *lp)
 {
 	struct gdlm_ls *ls = lp->ls;
 	struct lm_async_cb acb;
-	s16 prev_mode = lp->cur;
 
 	memset(&acb, 0, sizeof(acb));
 
@@ -160,15 +159,7 @@ static void process_complete(struct gdlm_lock *lp)
 			 lp->lksb.sb_status, lp->lockname.ln_type,
 			 (unsigned long long)lp->lockname.ln_number,
 			 lp->flags);
-		if (lp->lksb.sb_status == -EDEADLOCK &&
-		    lp->ls->fsflags & LM_MFLAG_CONV_NODROP) {
-			lp->req = lp->cur;
-			acb.lc_ret |= LM_OUT_CONV_DEADLK;
-			if (lp->cur == DLM_LOCK_IV)
-				lp->lksb.sb_lkid = 0;
-			goto out;
-		} else
-			return;
+		return;
 	}
 
 	/*
@@ -268,10 +259,6 @@ out:
 	acb.lc_name = lp->lockname;
 	acb.lc_ret |= gdlm_make_lmstate(lp->cur);
 
-	if (!test_and_clear_bit(LFL_NOCACHE, &lp->flags) &&
-	    (lp->cur > DLM_LOCK_NL) && (prev_mode > DLM_LOCK_NL))
-		acb.lc_ret |= LM_OUT_CACHEABLE;
-
 	ls->fscb(ls->sdp, LM_CB_ASYNC, &acb);
 }
 
@@ -376,14 +363,6 @@ static inline unsigned int make_flags(struct gdlm_lock *lp,
 
 	if (lp->lksb.sb_lkid != 0) {
 		lkf |= DLM_LKF_CONVERT;
-
-		/* Conversion deadlock avoidance by DLM */
-
-		if (!(lp->ls->fsflags & LM_MFLAG_CONV_NODROP) &&
-		    !test_bit(LFL_FORCE_PROMOTE, &lp->flags) &&
-		    !(lkf & DLM_LKF_NOQUEUE) &&
-		    cur > DLM_LOCK_NL && req > DLM_LOCK_NL && cur != req)
-			lkf |= DLM_LKF_CONVDEADLK;
 	}
 
 	if (lp->lvb)
diff --git a/include/linux/lm_interface.h b/include/linux/lm_interface.h
index d0a7112b9719..2ed8fa1b762b 100644
--- a/include/linux/lm_interface.h
+++ b/include/linux/lm_interface.h
@@ -122,11 +122,9 @@ typedef void (*lm_callback_t) (void *ptr, unsigned int type, void *data);
  */
 
 #define LM_OUT_ST_MASK		0x00000003
-#define LM_OUT_CACHEABLE	0x00000004
 #define LM_OUT_CANCELED		0x00000008
 #define LM_OUT_ASYNC		0x00000080
 #define LM_OUT_ERROR		0x00000100
-#define LM_OUT_CONV_DEADLK	0x00000200
 
 /*
  * lm_callback_t types
-- 
cgit v1.2.3


From c09595f63bb1909c5dc4dca288f4fe818561b5f3 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 27 Jun 2008 13:41:14 +0200
Subject: sched: revert revert of: fair-group: SMP-nice for group scheduling

Try again..

Initial commit: 18d95a2832c1392a2d63227a7a6d433cb9f2037e
Revert: 6363ca57c76b7b83639ca8c83fc285fa26a7880e

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |   1 +
 kernel/sched.c        | 430 ++++++++++++++++++++++++++++++++++++++++++++++----
 kernel/sched_debug.c  |   5 +
 kernel/sched_fair.c   | 124 +++++++++------
 kernel/sched_rt.c     |   4 +
 5 files changed, 489 insertions(+), 75 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index eaf821072dbd..97a58b622ee1 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -765,6 +765,7 @@ struct sched_domain {
 	struct sched_domain *child;	/* bottom domain must be null terminated */
 	struct sched_group *groups;	/* the balancing groups of the domain */
 	cpumask_t span;			/* span of all CPUs in this domain */
+	int first_cpu;			/* cache of the first cpu in this domain */
 	unsigned long min_interval;	/* Minimum balance interval ms */
 	unsigned long max_interval;	/* Maximum balance interval ms */
 	unsigned int busy_factor;	/* less balancing by factor if busy */
diff --git a/kernel/sched.c b/kernel/sched.c
index f653af684fb3..874b6da15430 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -403,6 +403,43 @@ struct cfs_rq {
 	 */
 	struct list_head leaf_cfs_rq_list;
 	struct task_group *tg;	/* group that "owns" this runqueue */
+
+#ifdef CONFIG_SMP
+	unsigned long task_weight;
+	unsigned long shares;
+	/*
+	 * We need space to build a sched_domain wide view of the full task
+	 * group tree, in order to avoid depending on dynamic memory allocation
+	 * during the load balancing we place this in the per cpu task group
+	 * hierarchy. This limits the load balancing to one instance per cpu,
+	 * but more should not be needed anyway.
+	 */
+	struct aggregate_struct {
+		/*
+		 *   load = weight(cpus) * f(tg)
+		 *
+		 * Where f(tg) is the recursive weight fraction assigned to
+		 * this group.
+		 */
+		unsigned long load;
+
+		/*
+		 * part of the group weight distributed to this span.
+		 */
+		unsigned long shares;
+
+		/*
+		 * The sum of all runqueue weights within this span.
+		 */
+		unsigned long rq_weight;
+
+		/*
+		 * Weight contributed by tasks; this is the part we can
+		 * influence by moving tasks around.
+		 */
+		unsigned long task_weight;
+	} aggregate;
+#endif
 #endif
 };
 
@@ -1484,6 +1521,326 @@ static unsigned long source_load(int cpu, int type);
 static unsigned long target_load(int cpu, int type);
 static unsigned long cpu_avg_load_per_task(int cpu);
 static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd);
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+
+/*
+ * Group load balancing.
+ *
+ * We calculate a few balance domain wide aggregate numbers; load and weight.
+ * Given the pictures below, and assuming each item has equal weight:
+ *
+ *         root          1 - thread
+ *         / | \         A - group
+ *        A  1  B
+ *       /|\   / \
+ *      C 2 D 3   4
+ *      |   |
+ *      5   6
+ *
+ * load:
+ *    A and B get 1/3-rd of the total load. C and D get 1/3-rd of A's 1/3-rd,
+ *    which equals 1/9-th of the total load.
+ *
+ * shares:
+ *    The weight of this group on the selected cpus.
+ *
+ * rq_weight:
+ *    Direct sum of all the cpu's their rq weight, e.g. A would get 3 while
+ *    B would get 2.
+ *
+ * task_weight:
+ *    Part of the rq_weight contributed by tasks; all groups except B would
+ *    get 1, B gets 2.
+ */
+
+static inline struct aggregate_struct *
+aggregate(struct task_group *tg, struct sched_domain *sd)
+{
+	return &tg->cfs_rq[sd->first_cpu]->aggregate;
+}
+
+typedef void (*aggregate_func)(struct task_group *, struct sched_domain *);
+
+/*
+ * Iterate the full tree, calling @down when first entering a node and @up when
+ * leaving it for the final time.
+ */
+static
+void aggregate_walk_tree(aggregate_func down, aggregate_func up,
+			 struct sched_domain *sd)
+{
+	struct task_group *parent, *child;
+
+	rcu_read_lock();
+	parent = &root_task_group;
+down:
+	(*down)(parent, sd);
+	list_for_each_entry_rcu(child, &parent->children, siblings) {
+		parent = child;
+		goto down;
+
+up:
+		continue;
+	}
+	(*up)(parent, sd);
+
+	child = parent;
+	parent = parent->parent;
+	if (parent)
+		goto up;
+	rcu_read_unlock();
+}
+
+/*
+ * Calculate the aggregate runqueue weight.
+ */
+static
+void aggregate_group_weight(struct task_group *tg, struct sched_domain *sd)
+{
+	unsigned long rq_weight = 0;
+	unsigned long task_weight = 0;
+	int i;
+
+	for_each_cpu_mask(i, sd->span) {
+		rq_weight += tg->cfs_rq[i]->load.weight;
+		task_weight += tg->cfs_rq[i]->task_weight;
+	}
+
+	aggregate(tg, sd)->rq_weight = rq_weight;
+	aggregate(tg, sd)->task_weight = task_weight;
+}
+
+/*
+ * Compute the weight of this group on the given cpus.
+ */
+static
+void aggregate_group_shares(struct task_group *tg, struct sched_domain *sd)
+{
+	unsigned long shares = 0;
+	int i;
+
+	for_each_cpu_mask(i, sd->span)
+		shares += tg->cfs_rq[i]->shares;
+
+	if ((!shares && aggregate(tg, sd)->rq_weight) || shares > tg->shares)
+		shares = tg->shares;
+
+	aggregate(tg, sd)->shares = shares;
+}
+
+/*
+ * Compute the load fraction assigned to this group, relies on the aggregate
+ * weight and this group's parent's load, i.e. top-down.
+ */
+static
+void aggregate_group_load(struct task_group *tg, struct sched_domain *sd)
+{
+	unsigned long load;
+
+	if (!tg->parent) {
+		int i;
+
+		load = 0;
+		for_each_cpu_mask(i, sd->span)
+			load += cpu_rq(i)->load.weight;
+
+	} else {
+		load = aggregate(tg->parent, sd)->load;
+
+		/*
+		 * shares is our weight in the parent's rq so
+		 * shares/parent->rq_weight gives our fraction of the load
+		 */
+		load *= aggregate(tg, sd)->shares;
+		load /= aggregate(tg->parent, sd)->rq_weight + 1;
+	}
+
+	aggregate(tg, sd)->load = load;
+}
+
+static void __set_se_shares(struct sched_entity *se, unsigned long shares);
+
+/*
+ * Calculate and set the cpu's group shares.
+ */
+static void
+__update_group_shares_cpu(struct task_group *tg, struct sched_domain *sd,
+			  int tcpu)
+{
+	int boost = 0;
+	unsigned long shares;
+	unsigned long rq_weight;
+
+	if (!tg->se[tcpu])
+		return;
+
+	rq_weight = tg->cfs_rq[tcpu]->load.weight;
+
+	/*
+	 * If there are currently no tasks on the cpu pretend there is one of
+	 * average load so that when a new task gets to run here it will not
+	 * get delayed by group starvation.
+	 */
+	if (!rq_weight) {
+		boost = 1;
+		rq_weight = NICE_0_LOAD;
+	}
+
+	/*
+	 *           \Sum shares * rq_weight
+	 * shares =  -----------------------
+	 *               \Sum rq_weight
+	 *
+	 */
+	shares = aggregate(tg, sd)->shares * rq_weight;
+	shares /= aggregate(tg, sd)->rq_weight + 1;
+
+	/*
+	 * record the actual number of shares, not the boosted amount.
+	 */
+	tg->cfs_rq[tcpu]->shares = boost ? 0 : shares;
+
+	if (shares < MIN_SHARES)
+		shares = MIN_SHARES;
+	else if (shares > MAX_SHARES)
+		shares = MAX_SHARES;
+
+	__set_se_shares(tg->se[tcpu], shares);
+}
+
+/*
+ * Re-adjust the weights on the cpu the task came from and on the cpu the
+ * task went to.
+ */
+static void
+__move_group_shares(struct task_group *tg, struct sched_domain *sd,
+		    int scpu, int dcpu)
+{
+	unsigned long shares;
+
+	shares = tg->cfs_rq[scpu]->shares + tg->cfs_rq[dcpu]->shares;
+
+	__update_group_shares_cpu(tg, sd, scpu);
+	__update_group_shares_cpu(tg, sd, dcpu);
+
+	/*
+	 * ensure we never loose shares due to rounding errors in the
+	 * above redistribution.
+	 */
+	shares -= tg->cfs_rq[scpu]->shares + tg->cfs_rq[dcpu]->shares;
+	if (shares)
+		tg->cfs_rq[dcpu]->shares += shares;
+}
+
+/*
+ * Because changing a group's shares changes the weight of the super-group
+ * we need to walk up the tree and change all shares until we hit the root.
+ */
+static void
+move_group_shares(struct task_group *tg, struct sched_domain *sd,
+		  int scpu, int dcpu)
+{
+	while (tg) {
+		__move_group_shares(tg, sd, scpu, dcpu);
+		tg = tg->parent;
+	}
+}
+
+static
+void aggregate_group_set_shares(struct task_group *tg, struct sched_domain *sd)
+{
+	unsigned long shares = aggregate(tg, sd)->shares;
+	int i;
+
+	for_each_cpu_mask(i, sd->span) {
+		struct rq *rq = cpu_rq(i);
+		unsigned long flags;
+
+		spin_lock_irqsave(&rq->lock, flags);
+		__update_group_shares_cpu(tg, sd, i);
+		spin_unlock_irqrestore(&rq->lock, flags);
+	}
+
+	aggregate_group_shares(tg, sd);
+
+	/*
+	 * ensure we never loose shares due to rounding errors in the
+	 * above redistribution.
+	 */
+	shares -= aggregate(tg, sd)->shares;
+	if (shares) {
+		tg->cfs_rq[sd->first_cpu]->shares += shares;
+		aggregate(tg, sd)->shares += shares;
+	}
+}
+
+/*
+ * Calculate the accumulative weight and recursive load of each task group
+ * while walking down the tree.
+ */
+static
+void aggregate_get_down(struct task_group *tg, struct sched_domain *sd)
+{
+	aggregate_group_weight(tg, sd);
+	aggregate_group_shares(tg, sd);
+	aggregate_group_load(tg, sd);
+}
+
+/*
+ * Rebalance the cpu shares while walking back up the tree.
+ */
+static
+void aggregate_get_up(struct task_group *tg, struct sched_domain *sd)
+{
+	aggregate_group_set_shares(tg, sd);
+}
+
+static DEFINE_PER_CPU(spinlock_t, aggregate_lock);
+
+static void __init init_aggregate(void)
+{
+	int i;
+
+	for_each_possible_cpu(i)
+		spin_lock_init(&per_cpu(aggregate_lock, i));
+}
+
+static int get_aggregate(struct sched_domain *sd)
+{
+	if (!spin_trylock(&per_cpu(aggregate_lock, sd->first_cpu)))
+		return 0;
+
+	aggregate_walk_tree(aggregate_get_down, aggregate_get_up, sd);
+	return 1;
+}
+
+static void put_aggregate(struct sched_domain *sd)
+{
+	spin_unlock(&per_cpu(aggregate_lock, sd->first_cpu));
+}
+
+static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares)
+{
+	cfs_rq->shares = shares;
+}
+
+#else
+
+static inline void init_aggregate(void)
+{
+}
+
+static inline int get_aggregate(struct sched_domain *sd)
+{
+	return 0;
+}
+
+static inline void put_aggregate(struct sched_domain *sd)
+{
+}
+#endif
+
 #endif
 
 #include "sched_stats.h"
@@ -1498,26 +1855,14 @@ static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd);
 #define for_each_class(class) \
    for (class = sched_class_highest; class; class = class->next)
 
-static inline void inc_load(struct rq *rq, const struct task_struct *p)
-{
-	update_load_add(&rq->load, p->se.load.weight);
-}
-
-static inline void dec_load(struct rq *rq, const struct task_struct *p)
-{
-	update_load_sub(&rq->load, p->se.load.weight);
-}
-
-static void inc_nr_running(struct task_struct *p, struct rq *rq)
+static void inc_nr_running(struct rq *rq)
 {
 	rq->nr_running++;
-	inc_load(rq, p);
 }
 
-static void dec_nr_running(struct task_struct *p, struct rq *rq)
+static void dec_nr_running(struct rq *rq)
 {
 	rq->nr_running--;
-	dec_load(rq, p);
 }
 
 static void set_load_weight(struct task_struct *p)
@@ -1609,7 +1954,7 @@ static void activate_task(struct rq *rq, struct task_struct *p, int wakeup)
 		rq->nr_uninterruptible--;
 
 	enqueue_task(rq, p, wakeup);
-	inc_nr_running(p, rq);
+	inc_nr_running(rq);
 }
 
 /*
@@ -1621,7 +1966,7 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep)
 		rq->nr_uninterruptible++;
 
 	dequeue_task(rq, p, sleep);
-	dec_nr_running(p, rq);
+	dec_nr_running(rq);
 }
 
 /**
@@ -2274,7 +2619,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 		 * management (if any):
 		 */
 		p->sched_class->task_new(rq, p);
-		inc_nr_running(p, rq);
+		inc_nr_running(rq);
 	}
 	check_preempt_curr(rq, p);
 #ifdef CONFIG_SMP
@@ -3265,9 +3610,12 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 	unsigned long imbalance;
 	struct rq *busiest;
 	unsigned long flags;
+	int unlock_aggregate;
 
 	cpus_setall(*cpus);
 
+	unlock_aggregate = get_aggregate(sd);
+
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
 	 * sibling can pick up load irrespective of busy siblings. In this case,
@@ -3383,8 +3731,9 @@ redo:
 
 	if (!ld_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
 	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-		return -1;
-	return ld_moved;
+		ld_moved = -1;
+
+	goto out;
 
 out_balanced:
 	schedstat_inc(sd, lb_balanced[idle]);
@@ -3399,8 +3748,13 @@ out_one_pinned:
 
 	if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER &&
 	    !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
-		return -1;
-	return 0;
+		ld_moved = -1;
+	else
+		ld_moved = 0;
+out:
+	if (unlock_aggregate)
+		put_aggregate(sd);
+	return ld_moved;
 }
 
 /*
@@ -4588,10 +4942,8 @@ void set_user_nice(struct task_struct *p, long nice)
 		goto out_unlock;
 	}
 	on_rq = p->se.on_rq;
-	if (on_rq) {
+	if (on_rq)
 		dequeue_task(rq, p, 0);
-		dec_load(rq, p);
-	}
 
 	p->static_prio = NICE_TO_PRIO(nice);
 	set_load_weight(p);
@@ -4601,7 +4953,6 @@ void set_user_nice(struct task_struct *p, long nice)
 
 	if (on_rq) {
 		enqueue_task(rq, p, 0);
-		inc_load(rq, p);
 		/*
 		 * If the task increased its priority or is running and
 		 * lowered its priority, then reschedule its CPU:
@@ -7016,6 +7367,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 			SD_INIT(sd, ALLNODES);
 			set_domain_attribute(sd, attr);
 			sd->span = *cpu_map;
+			sd->first_cpu = first_cpu(sd->span);
 			cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask);
 			p = sd;
 			sd_allnodes = 1;
@@ -7026,6 +7378,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, NODE);
 		set_domain_attribute(sd, attr);
 		sched_domain_node_span(cpu_to_node(i), &sd->span);
+		sd->first_cpu = first_cpu(sd->span);
 		sd->parent = p;
 		if (p)
 			p->child = sd;
@@ -7037,6 +7390,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, CPU);
 		set_domain_attribute(sd, attr);
 		sd->span = *nodemask;
+		sd->first_cpu = first_cpu(sd->span);
 		sd->parent = p;
 		if (p)
 			p->child = sd;
@@ -7048,6 +7402,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, MC);
 		set_domain_attribute(sd, attr);
 		sd->span = cpu_coregroup_map(i);
+		sd->first_cpu = first_cpu(sd->span);
 		cpus_and(sd->span, sd->span, *cpu_map);
 		sd->parent = p;
 		p->child = sd;
@@ -7060,6 +7415,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, SIBLING);
 		set_domain_attribute(sd, attr);
 		sd->span = per_cpu(cpu_sibling_map, i);
+		sd->first_cpu = first_cpu(sd->span);
 		cpus_and(sd->span, sd->span, *cpu_map);
 		sd->parent = p;
 		p->child = sd;
@@ -7757,6 +8113,7 @@ void __init sched_init(void)
 	}
 
 #ifdef CONFIG_SMP
+	init_aggregate();
 	init_defrootdomain();
 #endif
 
@@ -8322,14 +8679,11 @@ void sched_move_task(struct task_struct *tsk)
 #endif /* CONFIG_GROUP_SCHED */
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
-static void set_se_shares(struct sched_entity *se, unsigned long shares)
+static void __set_se_shares(struct sched_entity *se, unsigned long shares)
 {
 	struct cfs_rq *cfs_rq = se->cfs_rq;
-	struct rq *rq = cfs_rq->rq;
 	int on_rq;
 
-	spin_lock_irq(&rq->lock);
-
 	on_rq = se->on_rq;
 	if (on_rq)
 		dequeue_entity(cfs_rq, se, 0);
@@ -8339,8 +8693,17 @@ static void set_se_shares(struct sched_entity *se, unsigned long shares)
 
 	if (on_rq)
 		enqueue_entity(cfs_rq, se, 0);
+}
 
-	spin_unlock_irq(&rq->lock);
+static void set_se_shares(struct sched_entity *se, unsigned long shares)
+{
+	struct cfs_rq *cfs_rq = se->cfs_rq;
+	struct rq *rq = cfs_rq->rq;
+	unsigned long flags;
+
+	spin_lock_irqsave(&rq->lock, flags);
+	__set_se_shares(se, shares);
+	spin_unlock_irqrestore(&rq->lock, flags);
 }
 
 static DEFINE_MUTEX(shares_mutex);
@@ -8379,8 +8742,13 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares)
 	 * w/o tripping rebalance_share or load_balance_fair.
 	 */
 	tg->shares = shares;
-	for_each_possible_cpu(i)
+	for_each_possible_cpu(i) {
+		/*
+		 * force a rebalance
+		 */
+		cfs_rq_set_shares(tg->cfs_rq[i], 0);
 		set_se_shares(tg->se[i], shares);
+	}
 
 	/*
 	 * Enable load balance activity on this group, by inserting it back on
diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c
index 8e077b9c91cb..04394ccac88d 100644
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -167,6 +167,11 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
 #endif
 	SEQ_printf(m, "  .%-30s: %ld\n", "nr_spread_over",
 			cfs_rq->nr_spread_over);
+#ifdef CONFIG_FAIR_GROUP_SCHED
+#ifdef CONFIG_SMP
+	SEQ_printf(m, "  .%-30s: %lu\n", "shares", cfs_rq->shares);
+#endif
+#endif
 }
 
 void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq)
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 2e197b8e43f1..183388c4dead 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -567,10 +567,27 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se)
  * Scheduling class queueing methods:
  */
 
+#if defined CONFIG_SMP && defined CONFIG_FAIR_GROUP_SCHED
+static void
+add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight)
+{
+	cfs_rq->task_weight += weight;
+}
+#else
+static inline void
+add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight)
+{
+}
+#endif
+
 static void
 account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
 	update_load_add(&cfs_rq->load, se->load.weight);
+	if (!parent_entity(se))
+		inc_cpu_load(rq_of(cfs_rq), se->load.weight);
+	if (entity_is_task(se))
+		add_cfs_task_weight(cfs_rq, se->load.weight);
 	cfs_rq->nr_running++;
 	se->on_rq = 1;
 	list_add(&se->group_node, &cfs_rq->tasks);
@@ -580,6 +597,10 @@ static void
 account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
 	update_load_sub(&cfs_rq->load, se->load.weight);
+	if (!parent_entity(se))
+		dec_cpu_load(rq_of(cfs_rq), se->load.weight);
+	if (entity_is_task(se))
+		add_cfs_task_weight(cfs_rq, -se->load.weight);
 	cfs_rq->nr_running--;
 	se->on_rq = 0;
 	list_del_init(&se->group_node);
@@ -1372,75 +1393,90 @@ static struct task_struct *load_balance_next_fair(void *arg)
 	return __load_balance_iterator(cfs_rq, cfs_rq->balance_iterator);
 }
 
-#ifdef CONFIG_FAIR_GROUP_SCHED
-static int cfs_rq_best_prio(struct cfs_rq *cfs_rq)
+static unsigned long
+__load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
+		unsigned long max_load_move, struct sched_domain *sd,
+		enum cpu_idle_type idle, int *all_pinned, int *this_best_prio,
+		struct cfs_rq *cfs_rq)
 {
-	struct sched_entity *curr;
-	struct task_struct *p;
-
-	if (!cfs_rq->nr_running || !first_fair(cfs_rq))
-		return MAX_PRIO;
-
-	curr = cfs_rq->curr;
-	if (!curr)
-		curr = __pick_next_entity(cfs_rq);
+	struct rq_iterator cfs_rq_iterator;
 
-	p = task_of(curr);
+	cfs_rq_iterator.start = load_balance_start_fair;
+	cfs_rq_iterator.next = load_balance_next_fair;
+	cfs_rq_iterator.arg = cfs_rq;
 
-	return p->prio;
+	return balance_tasks(this_rq, this_cpu, busiest,
+			max_load_move, sd, idle, all_pinned,
+			this_best_prio, &cfs_rq_iterator);
 }
-#endif
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
 static unsigned long
 load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		  unsigned long max_load_move,
 		  struct sched_domain *sd, enum cpu_idle_type idle,
 		  int *all_pinned, int *this_best_prio)
 {
-	struct cfs_rq *busy_cfs_rq;
 	long rem_load_move = max_load_move;
-	struct rq_iterator cfs_rq_iterator;
-
-	cfs_rq_iterator.start = load_balance_start_fair;
-	cfs_rq_iterator.next = load_balance_next_fair;
+	int busiest_cpu = cpu_of(busiest);
+	struct task_group *tg;
 
-	for_each_leaf_cfs_rq(busiest, busy_cfs_rq) {
-#ifdef CONFIG_FAIR_GROUP_SCHED
-		struct cfs_rq *this_cfs_rq;
+	rcu_read_lock();
+	list_for_each_entry(tg, &task_groups, list) {
 		long imbalance;
-		unsigned long maxload;
+		unsigned long this_weight, busiest_weight;
+		long rem_load, max_load, moved_load;
+
+		/*
+		 * empty group
+		 */
+		if (!aggregate(tg, sd)->task_weight)
+			continue;
+
+		rem_load = rem_load_move * aggregate(tg, sd)->rq_weight;
+		rem_load /= aggregate(tg, sd)->load + 1;
+
+		this_weight = tg->cfs_rq[this_cpu]->task_weight;
+		busiest_weight = tg->cfs_rq[busiest_cpu]->task_weight;
 
-		this_cfs_rq = cpu_cfs_rq(busy_cfs_rq, this_cpu);
+		imbalance = (busiest_weight - this_weight) / 2;
 
-		imbalance = busy_cfs_rq->load.weight - this_cfs_rq->load.weight;
-		/* Don't pull if this_cfs_rq has more load than busy_cfs_rq */
-		if (imbalance <= 0)
+		if (imbalance < 0)
+			imbalance = busiest_weight;
+
+		max_load = max(rem_load, imbalance);
+		moved_load = __load_balance_fair(this_rq, this_cpu, busiest,
+				max_load, sd, idle, all_pinned, this_best_prio,
+				tg->cfs_rq[busiest_cpu]);
+
+		if (!moved_load)
 			continue;
 
-		/* Don't pull more than imbalance/2 */
-		imbalance /= 2;
-		maxload = min(rem_load_move, imbalance);
+		move_group_shares(tg, sd, busiest_cpu, this_cpu);
 
-		*this_best_prio = cfs_rq_best_prio(this_cfs_rq);
-#else
-# define maxload rem_load_move
-#endif
-		/*
-		 * pass busy_cfs_rq argument into
-		 * load_balance_[start|next]_fair iterators
-		 */
-		cfs_rq_iterator.arg = busy_cfs_rq;
-		rem_load_move -= balance_tasks(this_rq, this_cpu, busiest,
-					       maxload, sd, idle, all_pinned,
-					       this_best_prio,
-					       &cfs_rq_iterator);
+		moved_load *= aggregate(tg, sd)->load;
+		moved_load /= aggregate(tg, sd)->rq_weight + 1;
 
-		if (rem_load_move <= 0)
+		rem_load_move -= moved_load;
+		if (rem_load_move < 0)
 			break;
 	}
+	rcu_read_unlock();
 
 	return max_load_move - rem_load_move;
 }
+#else
+static unsigned long
+load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
+		  unsigned long max_load_move,
+		  struct sched_domain *sd, enum cpu_idle_type idle,
+		  int *all_pinned, int *this_best_prio)
+{
+	return __load_balance_fair(this_rq, this_cpu, busiest,
+			max_load_move, sd, idle, all_pinned,
+			this_best_prio, &busiest->cfs);
+}
+#endif
 
 static int
 move_one_task_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 6b4a6b5a4167..765932d0399d 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -670,6 +670,8 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
 		rt_se->timeout = 0;
 
 	enqueue_rt_entity(rt_se);
+
+	inc_cpu_load(rq, p->se.load.weight);
 }
 
 static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
@@ -678,6 +680,8 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
 
 	update_curr_rt(rq);
 	dequeue_rt_entity(rt_se);
+
+	dec_cpu_load(rq, p->se.load.weight);
 }
 
 /*
-- 
cgit v1.2.3


From b6a86c746f5b708012809958462234d19e9c8177 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 27 Jun 2008 13:41:18 +0200
Subject: sched: fix sched_domain aggregation

Keeping the aggregate on the first cpu of the sched domain has two problems:
 - it could collide between different sched domains on different cpus
 - it could slow things down because of the remote accesses

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |   1 -
 kernel/sched.c        | 113 ++++++++++++++++++++++++--------------------------
 kernel/sched_fair.c   |  12 +++---
 3 files changed, 60 insertions(+), 66 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 97a58b622ee1..eaf821072dbd 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -765,7 +765,6 @@ struct sched_domain {
 	struct sched_domain *child;	/* bottom domain must be null terminated */
 	struct sched_group *groups;	/* the balancing groups of the domain */
 	cpumask_t span;			/* span of all CPUs in this domain */
-	int first_cpu;			/* cache of the first cpu in this domain */
 	unsigned long min_interval;	/* Minimum balance interval ms */
 	unsigned long max_interval;	/* Maximum balance interval ms */
 	unsigned int busy_factor;	/* less balancing by factor if busy */
diff --git a/kernel/sched.c b/kernel/sched.c
index 7d282c52bd42..160d3c209b8f 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1480,12 +1480,12 @@ static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd);
  */
 
 static inline struct aggregate_struct *
-aggregate(struct task_group *tg, struct sched_domain *sd)
+aggregate(struct task_group *tg, int cpu)
 {
-	return &tg->cfs_rq[sd->first_cpu]->aggregate;
+	return &tg->cfs_rq[cpu]->aggregate;
 }
 
-typedef void (*aggregate_func)(struct task_group *, struct sched_domain *);
+typedef void (*aggregate_func)(struct task_group *, int, struct sched_domain *);
 
 /*
  * Iterate the full tree, calling @down when first entering a node and @up when
@@ -1493,14 +1493,14 @@ typedef void (*aggregate_func)(struct task_group *, struct sched_domain *);
  */
 static
 void aggregate_walk_tree(aggregate_func down, aggregate_func up,
-			 struct sched_domain *sd)
+			 int cpu, struct sched_domain *sd)
 {
 	struct task_group *parent, *child;
 
 	rcu_read_lock();
 	parent = &root_task_group;
 down:
-	(*down)(parent, sd);
+	(*down)(parent, cpu, sd);
 	list_for_each_entry_rcu(child, &parent->children, siblings) {
 		parent = child;
 		goto down;
@@ -1508,7 +1508,7 @@ down:
 up:
 		continue;
 	}
-	(*up)(parent, sd);
+	(*up)(parent, cpu, sd);
 
 	child = parent;
 	parent = parent->parent;
@@ -1520,8 +1520,8 @@ up:
 /*
  * Calculate the aggregate runqueue weight.
  */
-static
-void aggregate_group_weight(struct task_group *tg, struct sched_domain *sd)
+static void
+aggregate_group_weight(struct task_group *tg, int cpu, struct sched_domain *sd)
 {
 	unsigned long rq_weight = 0;
 	unsigned long task_weight = 0;
@@ -1532,15 +1532,15 @@ void aggregate_group_weight(struct task_group *tg, struct sched_domain *sd)
 		task_weight += tg->cfs_rq[i]->task_weight;
 	}
 
-	aggregate(tg, sd)->rq_weight = rq_weight;
-	aggregate(tg, sd)->task_weight = task_weight;
+	aggregate(tg, cpu)->rq_weight = rq_weight;
+	aggregate(tg, cpu)->task_weight = task_weight;
 }
 
 /*
  * Compute the weight of this group on the given cpus.
  */
-static
-void aggregate_group_shares(struct task_group *tg, struct sched_domain *sd)
+static void
+aggregate_group_shares(struct task_group *tg, int cpu, struct sched_domain *sd)
 {
 	unsigned long shares = 0;
 	int i;
@@ -1548,18 +1548,18 @@ void aggregate_group_shares(struct task_group *tg, struct sched_domain *sd)
 	for_each_cpu_mask(i, sd->span)
 		shares += tg->cfs_rq[i]->shares;
 
-	if ((!shares && aggregate(tg, sd)->rq_weight) || shares > tg->shares)
+	if ((!shares && aggregate(tg, cpu)->rq_weight) || shares > tg->shares)
 		shares = tg->shares;
 
-	aggregate(tg, sd)->shares = shares;
+	aggregate(tg, cpu)->shares = shares;
 }
 
 /*
  * Compute the load fraction assigned to this group, relies on the aggregate
  * weight and this group's parent's load, i.e. top-down.
  */
-static
-void aggregate_group_load(struct task_group *tg, struct sched_domain *sd)
+static void
+aggregate_group_load(struct task_group *tg, int cpu, struct sched_domain *sd)
 {
 	unsigned long load;
 
@@ -1571,17 +1571,17 @@ void aggregate_group_load(struct task_group *tg, struct sched_domain *sd)
 			load += cpu_rq(i)->load.weight;
 
 	} else {
-		load = aggregate(tg->parent, sd)->load;
+		load = aggregate(tg->parent, cpu)->load;
 
 		/*
 		 * shares is our weight in the parent's rq so
 		 * shares/parent->rq_weight gives our fraction of the load
 		 */
-		load *= aggregate(tg, sd)->shares;
-		load /= aggregate(tg->parent, sd)->rq_weight + 1;
+		load *= aggregate(tg, cpu)->shares;
+		load /= aggregate(tg->parent, cpu)->rq_weight + 1;
 	}
 
-	aggregate(tg, sd)->load = load;
+	aggregate(tg, cpu)->load = load;
 }
 
 static void __set_se_shares(struct sched_entity *se, unsigned long shares);
@@ -1590,8 +1590,8 @@ static void __set_se_shares(struct sched_entity *se, unsigned long shares);
  * Calculate and set the cpu's group shares.
  */
 static void
-__update_group_shares_cpu(struct task_group *tg, struct sched_domain *sd,
-			  int tcpu)
+__update_group_shares_cpu(struct task_group *tg, int cpu,
+			  struct sched_domain *sd, int tcpu)
 {
 	int boost = 0;
 	unsigned long shares;
@@ -1618,8 +1618,8 @@ __update_group_shares_cpu(struct task_group *tg, struct sched_domain *sd,
 	 *               \Sum rq_weight
 	 *
 	 */
-	shares = aggregate(tg, sd)->shares * rq_weight;
-	shares /= aggregate(tg, sd)->rq_weight + 1;
+	shares = aggregate(tg, cpu)->shares * rq_weight;
+	shares /= aggregate(tg, cpu)->rq_weight + 1;
 
 	/*
 	 * record the actual number of shares, not the boosted amount.
@@ -1639,15 +1639,15 @@ __update_group_shares_cpu(struct task_group *tg, struct sched_domain *sd,
  * task went to.
  */
 static void
-__move_group_shares(struct task_group *tg, struct sched_domain *sd,
+__move_group_shares(struct task_group *tg, int cpu, struct sched_domain *sd,
 		    int scpu, int dcpu)
 {
 	unsigned long shares;
 
 	shares = tg->cfs_rq[scpu]->shares + tg->cfs_rq[dcpu]->shares;
 
-	__update_group_shares_cpu(tg, sd, scpu);
-	__update_group_shares_cpu(tg, sd, dcpu);
+	__update_group_shares_cpu(tg, cpu, sd, scpu);
+	__update_group_shares_cpu(tg, cpu, sd, dcpu);
 
 	/*
 	 * ensure we never loose shares due to rounding errors in the
@@ -1663,19 +1663,19 @@ __move_group_shares(struct task_group *tg, struct sched_domain *sd,
  * we need to walk up the tree and change all shares until we hit the root.
  */
 static void
-move_group_shares(struct task_group *tg, struct sched_domain *sd,
+move_group_shares(struct task_group *tg, int cpu, struct sched_domain *sd,
 		  int scpu, int dcpu)
 {
 	while (tg) {
-		__move_group_shares(tg, sd, scpu, dcpu);
+		__move_group_shares(tg, cpu, sd, scpu, dcpu);
 		tg = tg->parent;
 	}
 }
 
-static
-void aggregate_group_set_shares(struct task_group *tg, struct sched_domain *sd)
+static void
+aggregate_group_set_shares(struct task_group *tg, int cpu, struct sched_domain *sd)
 {
-	unsigned long shares = aggregate(tg, sd)->shares;
+	unsigned long shares = aggregate(tg, cpu)->shares;
 	int i;
 
 	for_each_cpu_mask(i, sd->span) {
@@ -1683,20 +1683,20 @@ void aggregate_group_set_shares(struct task_group *tg, struct sched_domain *sd)
 		unsigned long flags;
 
 		spin_lock_irqsave(&rq->lock, flags);
-		__update_group_shares_cpu(tg, sd, i);
+		__update_group_shares_cpu(tg, cpu, sd, i);
 		spin_unlock_irqrestore(&rq->lock, flags);
 	}
 
-	aggregate_group_shares(tg, sd);
+	aggregate_group_shares(tg, cpu, sd);
 
 	/*
 	 * ensure we never loose shares due to rounding errors in the
 	 * above redistribution.
 	 */
-	shares -= aggregate(tg, sd)->shares;
+	shares -= aggregate(tg, cpu)->shares;
 	if (shares) {
-		tg->cfs_rq[sd->first_cpu]->shares += shares;
-		aggregate(tg, sd)->shares += shares;
+		tg->cfs_rq[cpu]->shares += shares;
+		aggregate(tg, cpu)->shares += shares;
 	}
 }
 
@@ -1704,21 +1704,21 @@ void aggregate_group_set_shares(struct task_group *tg, struct sched_domain *sd)
  * Calculate the accumulative weight and recursive load of each task group
  * while walking down the tree.
  */
-static
-void aggregate_get_down(struct task_group *tg, struct sched_domain *sd)
+static void
+aggregate_get_down(struct task_group *tg, int cpu, struct sched_domain *sd)
 {
-	aggregate_group_weight(tg, sd);
-	aggregate_group_shares(tg, sd);
-	aggregate_group_load(tg, sd);
+	aggregate_group_weight(tg, cpu, sd);
+	aggregate_group_shares(tg, cpu, sd);
+	aggregate_group_load(tg, cpu, sd);
 }
 
 /*
  * Rebalance the cpu shares while walking back up the tree.
  */
-static
-void aggregate_get_up(struct task_group *tg, struct sched_domain *sd)
+static void
+aggregate_get_up(struct task_group *tg, int cpu, struct sched_domain *sd)
 {
-	aggregate_group_set_shares(tg, sd);
+	aggregate_group_set_shares(tg, cpu, sd);
 }
 
 static DEFINE_PER_CPU(spinlock_t, aggregate_lock);
@@ -1731,18 +1731,18 @@ static void __init init_aggregate(void)
 		spin_lock_init(&per_cpu(aggregate_lock, i));
 }
 
-static int get_aggregate(struct sched_domain *sd)
+static int get_aggregate(int cpu, struct sched_domain *sd)
 {
-	if (!spin_trylock(&per_cpu(aggregate_lock, sd->first_cpu)))
+	if (!spin_trylock(&per_cpu(aggregate_lock, cpu)))
 		return 0;
 
-	aggregate_walk_tree(aggregate_get_down, aggregate_get_up, sd);
+	aggregate_walk_tree(aggregate_get_down, aggregate_get_up, cpu, sd);
 	return 1;
 }
 
-static void put_aggregate(struct sched_domain *sd)
+static void put_aggregate(int cpu, struct sched_domain *sd)
 {
-	spin_unlock(&per_cpu(aggregate_lock, sd->first_cpu));
+	spin_unlock(&per_cpu(aggregate_lock, cpu));
 }
 
 static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares)
@@ -1756,12 +1756,12 @@ static inline void init_aggregate(void)
 {
 }
 
-static inline int get_aggregate(struct sched_domain *sd)
+static inline int get_aggregate(int cpu, struct sched_domain *sd)
 {
 	return 0;
 }
 
-static inline void put_aggregate(struct sched_domain *sd)
+static inline void put_aggregate(int cpu, struct sched_domain *sd)
 {
 }
 #endif
@@ -3539,7 +3539,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 
 	cpus_setall(*cpus);
 
-	unlock_aggregate = get_aggregate(sd);
+	unlock_aggregate = get_aggregate(this_cpu, sd);
 
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
@@ -3678,7 +3678,7 @@ out_one_pinned:
 		ld_moved = 0;
 out:
 	if (unlock_aggregate)
-		put_aggregate(sd);
+		put_aggregate(this_cpu, sd);
 	return ld_moved;
 }
 
@@ -7292,7 +7292,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 			SD_INIT(sd, ALLNODES);
 			set_domain_attribute(sd, attr);
 			sd->span = *cpu_map;
-			sd->first_cpu = first_cpu(sd->span);
 			cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask);
 			p = sd;
 			sd_allnodes = 1;
@@ -7303,7 +7302,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, NODE);
 		set_domain_attribute(sd, attr);
 		sched_domain_node_span(cpu_to_node(i), &sd->span);
-		sd->first_cpu = first_cpu(sd->span);
 		sd->parent = p;
 		if (p)
 			p->child = sd;
@@ -7315,7 +7313,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, CPU);
 		set_domain_attribute(sd, attr);
 		sd->span = *nodemask;
-		sd->first_cpu = first_cpu(sd->span);
 		sd->parent = p;
 		if (p)
 			p->child = sd;
@@ -7327,7 +7324,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, MC);
 		set_domain_attribute(sd, attr);
 		sd->span = cpu_coregroup_map(i);
-		sd->first_cpu = first_cpu(sd->span);
 		cpus_and(sd->span, sd->span, *cpu_map);
 		sd->parent = p;
 		p->child = sd;
@@ -7340,7 +7336,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		SD_INIT(sd, SIBLING);
 		set_domain_attribute(sd, attr);
 		sd->span = per_cpu(cpu_sibling_map, i);
-		sd->first_cpu = first_cpu(sd->span);
 		cpus_and(sd->span, sd->span, *cpu_map);
 		sd->parent = p;
 		p->child = sd;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 509092af0330..40cf24ab4de8 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1429,11 +1429,11 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		/*
 		 * empty group
 		 */
-		if (!aggregate(tg, sd)->task_weight)
+		if (!aggregate(tg, this_cpu)->task_weight)
 			continue;
 
-		rem_load = rem_load_move * aggregate(tg, sd)->rq_weight;
-		rem_load /= aggregate(tg, sd)->load + 1;
+		rem_load = rem_load_move * aggregate(tg, this_cpu)->rq_weight;
+		rem_load /= aggregate(tg, this_cpu)->load + 1;
 
 		this_weight = tg->cfs_rq[this_cpu]->task_weight;
 		busiest_weight = tg->cfs_rq[busiest_cpu]->task_weight;
@@ -1451,10 +1451,10 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		if (!moved_load)
 			continue;
 
-		move_group_shares(tg, sd, busiest_cpu, this_cpu);
+		move_group_shares(tg, this_cpu, sd, busiest_cpu, this_cpu);
 
-		moved_load *= aggregate(tg, sd)->load;
-		moved_load /= aggregate(tg, sd)->rq_weight + 1;
+		moved_load *= aggregate(tg, this_cpu)->load;
+		moved_load /= aggregate(tg, this_cpu)->rq_weight + 1;
 
 		rem_load_move -= moved_load;
 		if (rem_load_move < 0)
-- 
cgit v1.2.3


From 2398f2c6d34b43025f274fc42eaca34d23ec2320 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 27 Jun 2008 13:41:35 +0200
Subject: sched: update shares on wakeup

We found that the affine wakeup code needs rather accurate load figures
to be effective. The trouble is that updating the load figures is fairly
expensive with group scheduling. Therefore ratelimit the updating.

Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h   |  3 +++
 kernel/sched.c          | 30 +++++++++++++++++++++++++++++-
 kernel/sched_features.h |  3 ++-
 kernel/sysctl.c         |  8 ++++++++
 4 files changed, 42 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index eaf821072dbd..835b6c6fcc56 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -783,6 +783,8 @@ struct sched_domain {
 	unsigned int balance_interval;	/* initialise to 1. units in ms. */
 	unsigned int nr_balance_failed; /* initialise to 0 */
 
+	u64 last_update;
+
 #ifdef CONFIG_SCHEDSTATS
 	/* load_balance() stats */
 	unsigned int lb_count[CPU_MAX_IDLE_TYPES];
@@ -1605,6 +1607,7 @@ extern unsigned int sysctl_sched_child_runs_first;
 extern unsigned int sysctl_sched_features;
 extern unsigned int sysctl_sched_migration_cost;
 extern unsigned int sysctl_sched_nr_migrate;
+extern unsigned int sysctl_sched_shares_ratelimit;
 
 int sched_nr_latency_handler(struct ctl_table *table, int write,
 		struct file *file, void __user *buffer, size_t *length,
diff --git a/kernel/sched.c b/kernel/sched.c
index 1cff969f6646..62db0891025a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -777,6 +777,12 @@ late_initcall(sched_init_debug);
  */
 const_debug unsigned int sysctl_sched_nr_migrate = 32;
 
+/*
+ * ratelimit for updating the group shares.
+ * default: 0.5ms
+ */
+const_debug unsigned int sysctl_sched_shares_ratelimit = 500000;
+
 /*
  * period over which we measure -rt task cpu usage in us.
  * default: 1s
@@ -1590,7 +1596,13 @@ tg_nop(struct task_group *tg, int cpu, struct sched_domain *sd)
 
 static void update_shares(struct sched_domain *sd)
 {
-	walk_tg_tree(tg_nop, tg_shares_up, 0, sd);
+	u64 now = cpu_clock(raw_smp_processor_id());
+	s64 elapsed = now - sd->last_update;
+
+	if (elapsed >= (s64)(u64)sysctl_sched_shares_ratelimit) {
+		sd->last_update = now;
+		walk_tg_tree(tg_nop, tg_shares_up, 0, sd);
+	}
 }
 
 static void update_shares_locked(struct rq *rq, struct sched_domain *sd)
@@ -2199,6 +2211,22 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 	if (!sched_feat(SYNC_WAKEUPS))
 		sync = 0;
 
+#ifdef CONFIG_SMP
+	if (sched_feat(LB_WAKEUP_UPDATE)) {
+		struct sched_domain *sd;
+
+		this_cpu = raw_smp_processor_id();
+		cpu = task_cpu(p);
+
+		for_each_domain(this_cpu, sd) {
+			if (cpu_isset(cpu, sd->span)) {
+				update_shares(sd);
+				break;
+			}
+		}
+	}
+#endif
+
 	smp_wmb();
 	rq = task_rq_lock(p, &flags);
 	old_state = p->state;
diff --git a/kernel/sched_features.h b/kernel/sched_features.h
index d56e3053e746..7d616d2a2a3f 100644
--- a/kernel/sched_features.h
+++ b/kernel/sched_features.h
@@ -8,4 +8,5 @@ SCHED_FEAT(SYNC_WAKEUPS, 1)
 SCHED_FEAT(HRTICK, 1)
 SCHED_FEAT(DOUBLE_TICK, 0)
 SCHED_FEAT(ASYM_GRAN, 1)
-SCHED_FEAT(LB_BIAS, 0)
\ No newline at end of file
+SCHED_FEAT(LB_BIAS, 0)
+SCHED_FEAT(LB_WAKEUP_UPDATE, 1)
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 29116652dca8..fe8cdc80ff02 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -264,6 +264,14 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &min_wakeup_granularity_ns,
 		.extra2		= &max_wakeup_granularity_ns,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "sched_shares_ratelimit",
+		.data		= &sysctl_sched_shares_ratelimit,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 	{
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "sched_child_runs_first",
-- 
cgit v1.2.3


From f225763a7d6c92c4932dbd528437997078496fcc Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Fri, 20 Jun 2008 11:50:29 +0200
Subject: ssb, b43, b43legacy, b44: Rewrite SSB DMA API

This is a rewrite of the DMA API for SSB devices.
This is needed, because the old (non-existing) "API" made too many bad
assumptions on the API of the host-bus (PCI).
This introduces an almost complete SSB-DMA-API that maps to the lowlevel
bus-API based on the bustype.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/b44.c                    | 140 +++++++++++++++++-----------------
 drivers/net/wireless/b43/dma.c       |  65 ++++++++--------
 drivers/net/wireless/b43legacy/dma.c |  63 ++++++++-------
 drivers/ssb/Kconfig                  |   2 +-
 drivers/ssb/main.c                   |  75 ++++++++++++++----
 include/linux/ssb/ssb.h              | 143 ++++++++++++++++++++++++++++++++++-
 6 files changed, 336 insertions(+), 152 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/b44.c b/drivers/net/b44.c
index 59dce6aa0865..c3bda5ce67c4 100644
--- a/drivers/net/b44.c
+++ b/drivers/net/b44.c
@@ -148,9 +148,9 @@ static inline void b44_sync_dma_desc_for_device(struct ssb_device *sdev,
 						unsigned long offset,
 						enum dma_data_direction dir)
 {
-	dma_sync_single_range_for_device(sdev->dma_dev, dma_base,
-					 offset & dma_desc_align_mask,
-					 dma_desc_sync_size, dir);
+	ssb_dma_sync_single_range_for_device(sdev, dma_base,
+					     offset & dma_desc_align_mask,
+					     dma_desc_sync_size, dir);
 }
 
 static inline void b44_sync_dma_desc_for_cpu(struct ssb_device *sdev,
@@ -158,9 +158,9 @@ static inline void b44_sync_dma_desc_for_cpu(struct ssb_device *sdev,
 					     unsigned long offset,
 					     enum dma_data_direction dir)
 {
-	dma_sync_single_range_for_cpu(sdev->dma_dev, dma_base,
-				      offset & dma_desc_align_mask,
-				      dma_desc_sync_size, dir);
+	ssb_dma_sync_single_range_for_cpu(sdev, dma_base,
+					  offset & dma_desc_align_mask,
+					  dma_desc_sync_size, dir);
 }
 
 static inline unsigned long br32(const struct b44 *bp, unsigned long reg)
@@ -613,10 +613,10 @@ static void b44_tx(struct b44 *bp)
 
 		BUG_ON(skb == NULL);
 
-		dma_unmap_single(bp->sdev->dma_dev,
-				 rp->mapping,
-				 skb->len,
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(bp->sdev,
+				     rp->mapping,
+				     skb->len,
+				     DMA_TO_DEVICE);
 		rp->skb = NULL;
 		dev_kfree_skb_irq(skb);
 	}
@@ -653,29 +653,29 @@ static int b44_alloc_rx_skb(struct b44 *bp, int src_idx, u32 dest_idx_unmasked)
 	if (skb == NULL)
 		return -ENOMEM;
 
-	mapping = dma_map_single(bp->sdev->dma_dev, skb->data,
-				 RX_PKT_BUF_SZ,
-				 DMA_FROM_DEVICE);
+	mapping = ssb_dma_map_single(bp->sdev, skb->data,
+				     RX_PKT_BUF_SZ,
+				     DMA_FROM_DEVICE);
 
 	/* Hardware bug work-around, the chip is unable to do PCI DMA
 	   to/from anything above 1GB :-( */
-	if (dma_mapping_error(mapping) ||
+	if (ssb_dma_mapping_error(bp->sdev, mapping) ||
 		mapping + RX_PKT_BUF_SZ > DMA_30BIT_MASK) {
 		/* Sigh... */
-		if (!dma_mapping_error(mapping))
-			dma_unmap_single(bp->sdev->dma_dev, mapping,
-					RX_PKT_BUF_SZ, DMA_FROM_DEVICE);
+		if (!ssb_dma_mapping_error(bp->sdev, mapping))
+			ssb_dma_unmap_single(bp->sdev, mapping,
+					     RX_PKT_BUF_SZ, DMA_FROM_DEVICE);
 		dev_kfree_skb_any(skb);
 		skb = __netdev_alloc_skb(bp->dev, RX_PKT_BUF_SZ, GFP_ATOMIC|GFP_DMA);
 		if (skb == NULL)
 			return -ENOMEM;
-		mapping = dma_map_single(bp->sdev->dma_dev, skb->data,
-					 RX_PKT_BUF_SZ,
-					 DMA_FROM_DEVICE);
-		if (dma_mapping_error(mapping) ||
+		mapping = ssb_dma_map_single(bp->sdev, skb->data,
+					     RX_PKT_BUF_SZ,
+					     DMA_FROM_DEVICE);
+		if (ssb_dma_mapping_error(bp->sdev, mapping) ||
 			mapping + RX_PKT_BUF_SZ > DMA_30BIT_MASK) {
-			if (!dma_mapping_error(mapping))
-				dma_unmap_single(bp->sdev->dma_dev, mapping, RX_PKT_BUF_SZ,DMA_FROM_DEVICE);
+			if (!ssb_dma_mapping_error(bp->sdev, mapping))
+				ssb_dma_unmap_single(bp->sdev, mapping, RX_PKT_BUF_SZ,DMA_FROM_DEVICE);
 			dev_kfree_skb_any(skb);
 			return -ENOMEM;
 		}
@@ -750,9 +750,9 @@ static void b44_recycle_rx(struct b44 *bp, int src_idx, u32 dest_idx_unmasked)
 					     dest_idx * sizeof(dest_desc),
 					     DMA_BIDIRECTIONAL);
 
-	dma_sync_single_for_device(bp->sdev->dma_dev, le32_to_cpu(src_desc->addr),
-				   RX_PKT_BUF_SZ,
-				   DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_device(bp->sdev, le32_to_cpu(src_desc->addr),
+				       RX_PKT_BUF_SZ,
+				       DMA_FROM_DEVICE);
 }
 
 static int b44_rx(struct b44 *bp, int budget)
@@ -772,7 +772,7 @@ static int b44_rx(struct b44 *bp, int budget)
 		struct rx_header *rh;
 		u16 len;
 
-		dma_sync_single_for_cpu(bp->sdev->dma_dev, map,
+		ssb_dma_sync_single_for_cpu(bp->sdev, map,
 					    RX_PKT_BUF_SZ,
 					    DMA_FROM_DEVICE);
 		rh = (struct rx_header *) skb->data;
@@ -806,8 +806,8 @@ static int b44_rx(struct b44 *bp, int budget)
 			skb_size = b44_alloc_rx_skb(bp, cons, bp->rx_prod);
 			if (skb_size < 0)
 				goto drop_it;
-			dma_unmap_single(bp->sdev->dma_dev, map,
-					 skb_size, DMA_FROM_DEVICE);
+			ssb_dma_unmap_single(bp->sdev, map,
+					     skb_size, DMA_FROM_DEVICE);
 			/* Leave out rx_header */
                 	skb_put(skb, len + RX_PKT_OFFSET);
             	        skb_pull(skb, RX_PKT_OFFSET);
@@ -966,25 +966,25 @@ static int b44_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		goto err_out;
 	}
 
-	mapping = dma_map_single(bp->sdev->dma_dev, skb->data, len, DMA_TO_DEVICE);
-	if (dma_mapping_error(mapping) || mapping + len > DMA_30BIT_MASK) {
+	mapping = ssb_dma_map_single(bp->sdev, skb->data, len, DMA_TO_DEVICE);
+	if (ssb_dma_mapping_error(bp->sdev, mapping) || mapping + len > DMA_30BIT_MASK) {
 		struct sk_buff *bounce_skb;
 
 		/* Chip can't handle DMA to/from >1GB, use bounce buffer */
-		if (!dma_mapping_error(mapping))
-			dma_unmap_single(bp->sdev->dma_dev, mapping, len,
-					DMA_TO_DEVICE);
+		if (!ssb_dma_mapping_error(bp->sdev, mapping))
+			ssb_dma_unmap_single(bp->sdev, mapping, len,
+					     DMA_TO_DEVICE);
 
 		bounce_skb = __dev_alloc_skb(len, GFP_ATOMIC | GFP_DMA);
 		if (!bounce_skb)
 			goto err_out;
 
-		mapping = dma_map_single(bp->sdev->dma_dev, bounce_skb->data,
-					 len, DMA_TO_DEVICE);
-		if (dma_mapping_error(mapping) || mapping + len > DMA_30BIT_MASK) {
-			if (!dma_mapping_error(mapping))
-				dma_unmap_single(bp->sdev->dma_dev, mapping,
-					 len, DMA_TO_DEVICE);
+		mapping = ssb_dma_map_single(bp->sdev, bounce_skb->data,
+					     len, DMA_TO_DEVICE);
+		if (ssb_dma_mapping_error(bp->sdev, mapping) || mapping + len > DMA_30BIT_MASK) {
+			if (!ssb_dma_mapping_error(bp->sdev, mapping))
+				ssb_dma_unmap_single(bp->sdev, mapping,
+						     len, DMA_TO_DEVICE);
 			dev_kfree_skb_any(bounce_skb);
 			goto err_out;
 		}
@@ -1082,8 +1082,8 @@ static void b44_free_rings(struct b44 *bp)
 
 		if (rp->skb == NULL)
 			continue;
-		dma_unmap_single(bp->sdev->dma_dev, rp->mapping, RX_PKT_BUF_SZ,
-					DMA_FROM_DEVICE);
+		ssb_dma_unmap_single(bp->sdev, rp->mapping, RX_PKT_BUF_SZ,
+				     DMA_FROM_DEVICE);
 		dev_kfree_skb_any(rp->skb);
 		rp->skb = NULL;
 	}
@@ -1094,8 +1094,8 @@ static void b44_free_rings(struct b44 *bp)
 
 		if (rp->skb == NULL)
 			continue;
-		dma_unmap_single(bp->sdev->dma_dev, rp->mapping, rp->skb->len,
-					DMA_TO_DEVICE);
+		ssb_dma_unmap_single(bp->sdev, rp->mapping, rp->skb->len,
+				     DMA_TO_DEVICE);
 		dev_kfree_skb_any(rp->skb);
 		rp->skb = NULL;
 	}
@@ -1117,14 +1117,14 @@ static void b44_init_rings(struct b44 *bp)
 	memset(bp->tx_ring, 0, B44_TX_RING_BYTES);
 
 	if (bp->flags & B44_FLAG_RX_RING_HACK)
-		dma_sync_single_for_device(bp->sdev->dma_dev, bp->rx_ring_dma,
-			                  DMA_TABLE_BYTES,
-			                  DMA_BIDIRECTIONAL);
+		ssb_dma_sync_single_for_device(bp->sdev, bp->rx_ring_dma,
+					       DMA_TABLE_BYTES,
+					       DMA_BIDIRECTIONAL);
 
 	if (bp->flags & B44_FLAG_TX_RING_HACK)
-		dma_sync_single_for_device(bp->sdev->dma_dev, bp->tx_ring_dma,
-			                  DMA_TABLE_BYTES,
-			                  DMA_TO_DEVICE);
+		ssb_dma_sync_single_for_device(bp->sdev, bp->tx_ring_dma,
+					       DMA_TABLE_BYTES,
+					       DMA_TO_DEVICE);
 
 	for (i = 0; i < bp->rx_pending; i++) {
 		if (b44_alloc_rx_skb(bp, -1, i) < 0)
@@ -1144,25 +1144,27 @@ static void b44_free_consistent(struct b44 *bp)
 	bp->tx_buffers = NULL;
 	if (bp->rx_ring) {
 		if (bp->flags & B44_FLAG_RX_RING_HACK) {
-			dma_unmap_single(bp->sdev->dma_dev, bp->rx_ring_dma,
-					DMA_TABLE_BYTES,
-					DMA_BIDIRECTIONAL);
+			ssb_dma_unmap_single(bp->sdev, bp->rx_ring_dma,
+					     DMA_TABLE_BYTES,
+					     DMA_BIDIRECTIONAL);
 			kfree(bp->rx_ring);
 		} else
-			dma_free_coherent(bp->sdev->dma_dev, DMA_TABLE_BYTES,
-					    bp->rx_ring, bp->rx_ring_dma);
+			ssb_dma_free_consistent(bp->sdev, DMA_TABLE_BYTES,
+						bp->rx_ring, bp->rx_ring_dma,
+						GFP_KERNEL);
 		bp->rx_ring = NULL;
 		bp->flags &= ~B44_FLAG_RX_RING_HACK;
 	}
 	if (bp->tx_ring) {
 		if (bp->flags & B44_FLAG_TX_RING_HACK) {
-			dma_unmap_single(bp->sdev->dma_dev, bp->tx_ring_dma,
-					DMA_TABLE_BYTES,
-					DMA_TO_DEVICE);
+			ssb_dma_unmap_single(bp->sdev, bp->tx_ring_dma,
+					     DMA_TABLE_BYTES,
+					     DMA_TO_DEVICE);
 			kfree(bp->tx_ring);
 		} else
-			dma_free_coherent(bp->sdev->dma_dev, DMA_TABLE_BYTES,
-					    bp->tx_ring, bp->tx_ring_dma);
+			ssb_dma_free_consistent(bp->sdev, DMA_TABLE_BYTES,
+						bp->tx_ring, bp->tx_ring_dma,
+						GFP_KERNEL);
 		bp->tx_ring = NULL;
 		bp->flags &= ~B44_FLAG_TX_RING_HACK;
 	}
@@ -1187,7 +1189,7 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		goto out_err;
 
 	size = DMA_TABLE_BYTES;
-	bp->rx_ring = dma_alloc_coherent(bp->sdev->dma_dev, size, &bp->rx_ring_dma, gfp);
+	bp->rx_ring = ssb_dma_alloc_consistent(bp->sdev, size, &bp->rx_ring_dma, gfp);
 	if (!bp->rx_ring) {
 		/* Allocation may have failed due to pci_alloc_consistent
 		   insisting on use of GFP_DMA, which is more restrictive
@@ -1199,11 +1201,11 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		if (!rx_ring)
 			goto out_err;
 
-		rx_ring_dma = dma_map_single(bp->sdev->dma_dev, rx_ring,
-			                    DMA_TABLE_BYTES,
-			                    DMA_BIDIRECTIONAL);
+		rx_ring_dma = ssb_dma_map_single(bp->sdev, rx_ring,
+						 DMA_TABLE_BYTES,
+						 DMA_BIDIRECTIONAL);
 
-		if (dma_mapping_error(rx_ring_dma) ||
+		if (ssb_dma_mapping_error(bp->sdev, rx_ring_dma) ||
 			rx_ring_dma + size > DMA_30BIT_MASK) {
 			kfree(rx_ring);
 			goto out_err;
@@ -1214,9 +1216,9 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		bp->flags |= B44_FLAG_RX_RING_HACK;
 	}
 
-	bp->tx_ring = dma_alloc_coherent(bp->sdev->dma_dev, size, &bp->tx_ring_dma, gfp);
+	bp->tx_ring = ssb_dma_alloc_consistent(bp->sdev, size, &bp->tx_ring_dma, gfp);
 	if (!bp->tx_ring) {
-		/* Allocation may have failed due to dma_alloc_coherent
+		/* Allocation may have failed due to ssb_dma_alloc_consistent
 		   insisting on use of GFP_DMA, which is more restrictive
 		   than necessary...  */
 		struct dma_desc *tx_ring;
@@ -1226,11 +1228,11 @@ static int b44_alloc_consistent(struct b44 *bp, gfp_t gfp)
 		if (!tx_ring)
 			goto out_err;
 
-		tx_ring_dma = dma_map_single(bp->sdev->dma_dev, tx_ring,
+		tx_ring_dma = ssb_dma_map_single(bp->sdev, tx_ring,
 			                    DMA_TABLE_BYTES,
 			                    DMA_TO_DEVICE);
 
-		if (dma_mapping_error(tx_ring_dma) ||
+		if (ssb_dma_mapping_error(bp->sdev, tx_ring_dma) ||
 			tx_ring_dma + size > DMA_30BIT_MASK) {
 			kfree(tx_ring);
 			goto out_err;
diff --git a/drivers/net/wireless/b43/dma.c b/drivers/net/wireless/b43/dma.c
index 8a09a1db08db..098f886976f6 100644
--- a/drivers/net/wireless/b43/dma.c
+++ b/drivers/net/wireless/b43/dma.c
@@ -328,11 +328,11 @@ static inline
 	dma_addr_t dmaaddr;
 
 	if (tx) {
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len, DMA_TO_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len, DMA_TO_DEVICE);
 	} else {
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len, DMA_FROM_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len, DMA_FROM_DEVICE);
 	}
 
 	return dmaaddr;
@@ -343,11 +343,11 @@ static inline
 			  dma_addr_t addr, size_t len, int tx)
 {
 	if (tx) {
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len, DMA_TO_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len, DMA_TO_DEVICE);
 	} else {
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len, DMA_FROM_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len, DMA_FROM_DEVICE);
 	}
 }
 
@@ -356,8 +356,8 @@ static inline
 				 dma_addr_t addr, size_t len)
 {
 	B43_WARN_ON(ring->tx);
-	dma_sync_single_for_cpu(ring->dev->dev->dma_dev,
-				addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_cpu(ring->dev->dev,
+				    addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -365,8 +365,8 @@ static inline
 				    dma_addr_t addr, size_t len)
 {
 	B43_WARN_ON(ring->tx);
-	dma_sync_single_for_device(ring->dev->dev->dma_dev,
-				   addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_device(ring->dev->dev,
+				       addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -381,7 +381,6 @@ static inline
 
 static int alloc_ringmemory(struct b43_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
 	gfp_t flags = GFP_KERNEL;
 
 	/* The specs call for 4K buffers for 30- and 32-bit DMA with 4K
@@ -392,11 +391,14 @@ static int alloc_ringmemory(struct b43_dmaring *ring)
 	 * For unknown reasons - possibly a hardware error - the BCM4311 rev
 	 * 02, which uses 64-bit DMA, needs the ring buffer in very low memory,
 	 * which accounts for the GFP_DMA flag below.
+	 *
+	 * The flags here must match the flags in free_ringmemory below!
 	 */
 	if (ring->type == B43_DMA_64BIT)
 		flags |= GFP_DMA;
-	ring->descbase = dma_alloc_coherent(dma_dev, B43_DMA_RINGMEMSIZE,
-					    &(ring->dmabase), flags);
+	ring->descbase = ssb_dma_alloc_consistent(ring->dev->dev,
+						  B43_DMA_RINGMEMSIZE,
+						  &(ring->dmabase), flags);
 	if (!ring->descbase) {
 		b43err(ring->dev->wl, "DMA ringmemory allocation failed\n");
 		return -ENOMEM;
@@ -408,10 +410,13 @@ static int alloc_ringmemory(struct b43_dmaring *ring)
 
 static void free_ringmemory(struct b43_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
+	gfp_t flags = GFP_KERNEL;
+
+	if (ring->type == B43_DMA_64BIT)
+		flags |= GFP_DMA;
 
-	dma_free_coherent(dma_dev, B43_DMA_RINGMEMSIZE,
-			  ring->descbase, ring->dmabase);
+	ssb_dma_free_consistent(ring->dev->dev, B43_DMA_RINGMEMSIZE,
+				ring->descbase, ring->dmabase, flags);
 }
 
 /* Reset the RX DMA channel */
@@ -518,7 +523,7 @@ static bool b43_dma_mapping_error(struct b43_dmaring *ring,
 				  dma_addr_t addr,
 				  size_t buffersize, bool dma_to_device)
 {
-	if (unlikely(dma_mapping_error(addr)))
+	if (unlikely(ssb_dma_mapping_error(ring->dev->dev, addr)))
 		return 1;
 
 	switch (ring->type) {
@@ -844,10 +849,10 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			goto err_kfree_meta;
 
 		/* test for ability to dma to txhdr_cache */
-		dma_test = dma_map_single(dev->dev->dma_dev,
-					  ring->txhdr_cache,
-					  b43_txhdr_size(dev),
-					  DMA_TO_DEVICE);
+		dma_test = ssb_dma_map_single(dev->dev,
+					      ring->txhdr_cache,
+					      b43_txhdr_size(dev),
+					      DMA_TO_DEVICE);
 
 		if (b43_dma_mapping_error(ring, dma_test,
 					  b43_txhdr_size(dev), 1)) {
@@ -859,10 +864,10 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			if (!ring->txhdr_cache)
 				goto err_kfree_meta;
 
-			dma_test = dma_map_single(dev->dev->dma_dev,
-						  ring->txhdr_cache,
-						  b43_txhdr_size(dev),
-						  DMA_TO_DEVICE);
+			dma_test = ssb_dma_map_single(dev->dev,
+						      ring->txhdr_cache,
+						      b43_txhdr_size(dev),
+						      DMA_TO_DEVICE);
 
 			if (b43_dma_mapping_error(ring, dma_test,
 						  b43_txhdr_size(dev), 1)) {
@@ -873,9 +878,9 @@ struct b43_dmaring *b43_setup_dmaring(struct b43_wldev *dev,
 			}
 		}
 
-		dma_unmap_single(dev->dev->dma_dev,
-				 dma_test, b43_txhdr_size(dev),
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(dev->dev,
+				     dma_test, b43_txhdr_size(dev),
+				     DMA_TO_DEVICE);
 	}
 
 	err = alloc_ringmemory(ring);
diff --git a/drivers/net/wireless/b43legacy/dma.c b/drivers/net/wireless/b43legacy/dma.c
index 33cc256c5baf..9736b2f56a75 100644
--- a/drivers/net/wireless/b43legacy/dma.c
+++ b/drivers/net/wireless/b43legacy/dma.c
@@ -393,13 +393,13 @@ dma_addr_t map_descbuffer(struct b43legacy_dmaring *ring,
 	dma_addr_t dmaaddr;
 
 	if (tx)
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len,
-					 DMA_TO_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len,
+					     DMA_TO_DEVICE);
 	else
-		dmaaddr = dma_map_single(ring->dev->dev->dma_dev,
-					 buf, len,
-					 DMA_FROM_DEVICE);
+		dmaaddr = ssb_dma_map_single(ring->dev->dev,
+					     buf, len,
+					     DMA_FROM_DEVICE);
 
 	return dmaaddr;
 }
@@ -411,13 +411,13 @@ void unmap_descbuffer(struct b43legacy_dmaring *ring,
 		      int tx)
 {
 	if (tx)
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len,
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len,
+				     DMA_TO_DEVICE);
 	else
-		dma_unmap_single(ring->dev->dev->dma_dev,
-				 addr, len,
-				 DMA_FROM_DEVICE);
+		ssb_dma_unmap_single(ring->dev->dev,
+				     addr, len,
+				     DMA_FROM_DEVICE);
 }
 
 static inline
@@ -427,8 +427,8 @@ void sync_descbuffer_for_cpu(struct b43legacy_dmaring *ring,
 {
 	B43legacy_WARN_ON(ring->tx);
 
-	dma_sync_single_for_cpu(ring->dev->dev->dma_dev,
-				addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_cpu(ring->dev->dev,
+				    addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -438,8 +438,8 @@ void sync_descbuffer_for_device(struct b43legacy_dmaring *ring,
 {
 	B43legacy_WARN_ON(ring->tx);
 
-	dma_sync_single_for_device(ring->dev->dev->dma_dev,
-				   addr, len, DMA_FROM_DEVICE);
+	ssb_dma_sync_single_for_device(ring->dev->dev,
+				       addr, len, DMA_FROM_DEVICE);
 }
 
 static inline
@@ -458,10 +458,11 @@ void free_descriptor_buffer(struct b43legacy_dmaring *ring,
 
 static int alloc_ringmemory(struct b43legacy_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
-
-	ring->descbase = dma_alloc_coherent(dma_dev, B43legacy_DMA_RINGMEMSIZE,
-					    &(ring->dmabase), GFP_KERNEL);
+	/* GFP flags must match the flags in free_ringmemory()! */
+	ring->descbase = ssb_dma_alloc_consistent(ring->dev->dev,
+						  B43legacy_DMA_RINGMEMSIZE,
+						  &(ring->dmabase),
+						  GFP_KERNEL);
 	if (!ring->descbase) {
 		b43legacyerr(ring->dev->wl, "DMA ringmemory allocation"
 			     " failed\n");
@@ -474,10 +475,8 @@ static int alloc_ringmemory(struct b43legacy_dmaring *ring)
 
 static void free_ringmemory(struct b43legacy_dmaring *ring)
 {
-	struct device *dma_dev = ring->dev->dev->dma_dev;
-
-	dma_free_coherent(dma_dev, B43legacy_DMA_RINGMEMSIZE,
-			  ring->descbase, ring->dmabase);
+	ssb_dma_free_consistent(ring->dev->dev, B43legacy_DMA_RINGMEMSIZE,
+				ring->descbase, ring->dmabase, GFP_KERNEL);
 }
 
 /* Reset the RX DMA channel */
@@ -589,7 +588,7 @@ static bool b43legacy_dma_mapping_error(struct b43legacy_dmaring *ring,
 					 size_t buffersize,
 					 bool dma_to_device)
 {
-	if (unlikely(dma_mapping_error(addr)))
+	if (unlikely(ssb_dma_mapping_error(ring->dev->dev, addr)))
 		return 1;
 
 	switch (ring->type) {
@@ -893,9 +892,9 @@ struct b43legacy_dmaring *b43legacy_setup_dmaring(struct b43legacy_wldev *dev,
 			goto err_kfree_meta;
 
 		/* test for ability to dma to txhdr_cache */
-		dma_test = dma_map_single(dev->dev->dma_dev, ring->txhdr_cache,
-					  sizeof(struct b43legacy_txhdr_fw3),
-					  DMA_TO_DEVICE);
+		dma_test = ssb_dma_map_single(dev->dev, ring->txhdr_cache,
+					      sizeof(struct b43legacy_txhdr_fw3),
+					      DMA_TO_DEVICE);
 
 		if (b43legacy_dma_mapping_error(ring, dma_test,
 					sizeof(struct b43legacy_txhdr_fw3), 1)) {
@@ -907,7 +906,7 @@ struct b43legacy_dmaring *b43legacy_setup_dmaring(struct b43legacy_wldev *dev,
 			if (!ring->txhdr_cache)
 				goto err_kfree_meta;
 
-			dma_test = dma_map_single(dev->dev->dma_dev,
+				dma_test = ssb_dma_map_single(dev->dev,
 					ring->txhdr_cache,
 					sizeof(struct b43legacy_txhdr_fw3),
 					DMA_TO_DEVICE);
@@ -917,9 +916,9 @@ struct b43legacy_dmaring *b43legacy_setup_dmaring(struct b43legacy_wldev *dev,
 				goto err_kfree_txhdr_cache;
 		}
 
-		dma_unmap_single(dev->dev->dma_dev,
-				 dma_test, sizeof(struct b43legacy_txhdr_fw3),
-				 DMA_TO_DEVICE);
+		ssb_dma_unmap_single(dev->dev, dma_test,
+				     sizeof(struct b43legacy_txhdr_fw3),
+				     DMA_TO_DEVICE);
 	}
 
 	ring->dev = dev;
diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig
index cd845b8acd17..307b1f62d949 100644
--- a/drivers/ssb/Kconfig
+++ b/drivers/ssb/Kconfig
@@ -2,7 +2,7 @@ menu "Sonics Silicon Backplane"
 
 config SSB_POSSIBLE
 	bool
-	depends on HAS_IOMEM
+	depends on HAS_IOMEM && HAS_DMA
 	default y
 
 config SSB
diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
index d184f2aea78d..d831a2beff39 100644
--- a/drivers/ssb/main.c
+++ b/drivers/ssb/main.c
@@ -462,18 +462,15 @@ static int ssb_devices_register(struct ssb_bus *bus)
 #ifdef CONFIG_SSB_PCIHOST
 			sdev->irq = bus->host_pci->irq;
 			dev->parent = &bus->host_pci->dev;
-			sdev->dma_dev = &bus->host_pci->dev;
 #endif
 			break;
 		case SSB_BUSTYPE_PCMCIA:
 #ifdef CONFIG_SSB_PCMCIAHOST
 			sdev->irq = bus->host_pcmcia->irq.AssignedIRQ;
 			dev->parent = &bus->host_pcmcia->dev;
-			sdev->dma_dev = &bus->host_pcmcia->dev;
 #endif
 			break;
 		case SSB_BUSTYPE_SSB:
-			sdev->dma_dev = dev;
 			break;
 		}
 
@@ -1156,36 +1153,82 @@ u32 ssb_dma_translation(struct ssb_device *dev)
 {
 	switch (dev->bus->bustype) {
 	case SSB_BUSTYPE_SSB:
-	case SSB_BUSTYPE_PCMCIA:
 		return 0;
 	case SSB_BUSTYPE_PCI:
 		return SSB_PCI_DMA;
+	default:
+		__ssb_dma_not_implemented(dev);
 	}
 	return 0;
 }
 EXPORT_SYMBOL(ssb_dma_translation);
 
-int ssb_dma_set_mask(struct ssb_device *ssb_dev, u64 mask)
+int ssb_dma_set_mask(struct ssb_device *dev, u64 mask)
 {
-	struct device *dma_dev = ssb_dev->dma_dev;
-	int err = 0;
+	int err;
 
-#ifdef CONFIG_SSB_PCIHOST
-	if (ssb_dev->bus->bustype == SSB_BUSTYPE_PCI) {
-		err = pci_set_dma_mask(ssb_dev->bus->host_pci, mask);
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		err = pci_set_dma_mask(dev->bus->host_pci, mask);
 		if (err)
 			return err;
-		err = pci_set_consistent_dma_mask(ssb_dev->bus->host_pci, mask);
+		err = pci_set_consistent_dma_mask(dev->bus->host_pci, mask);
 		return err;
+	case SSB_BUSTYPE_SSB:
+		return dma_set_mask(dev->dev, mask);
+	default:
+		__ssb_dma_not_implemented(dev);
 	}
-#endif
-	dma_dev->coherent_dma_mask = mask;
-	dma_dev->dma_mask = &dma_dev->coherent_dma_mask;
-
-	return err;
+	return -ENOSYS;
 }
 EXPORT_SYMBOL(ssb_dma_set_mask);
 
+void * ssb_dma_alloc_consistent(struct ssb_device *dev, size_t size,
+				dma_addr_t *dma_handle, gfp_t gfp_flags)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		if (gfp_flags & GFP_DMA) {
+			/* Workaround: The PCI API does not support passing
+			 * a GFP flag. */
+			return dma_alloc_coherent(&dev->bus->host_pci->dev,
+						  size, dma_handle, gfp_flags);
+		}
+		return pci_alloc_consistent(dev->bus->host_pci, size, dma_handle);
+	case SSB_BUSTYPE_SSB:
+		return dma_alloc_coherent(dev->dev, size, dma_handle, gfp_flags);
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+	return NULL;
+}
+EXPORT_SYMBOL(ssb_dma_alloc_consistent);
+
+void ssb_dma_free_consistent(struct ssb_device *dev, size_t size,
+			     void *vaddr, dma_addr_t dma_handle,
+			     gfp_t gfp_flags)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		if (gfp_flags & GFP_DMA) {
+			/* Workaround: The PCI API does not support passing
+			 * a GFP flag. */
+			dma_free_coherent(&dev->bus->host_pci->dev,
+					  size, vaddr, dma_handle);
+			return;
+		}
+		pci_free_consistent(dev->bus->host_pci, size,
+				    vaddr, dma_handle);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_free_coherent(dev->dev, size, vaddr, dma_handle);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+EXPORT_SYMBOL(ssb_dma_free_consistent);
+
 int ssb_bus_may_powerdown(struct ssb_bus *bus)
 {
 	struct ssb_chipcommon *cc;
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 50dfd0dc4093..0fe5a0ded3ea 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -137,9 +137,6 @@ struct ssb_device {
 	const struct ssb_bus_ops *ops;
 
 	struct device *dev;
-	/* Pointer to the device that has to be used for
-	 * any DMA related operation. */
-	struct device *dma_dev;
 
 	struct ssb_bus *bus;
 	struct ssb_device_id id;
@@ -399,13 +396,151 @@ static inline void ssb_block_write(struct ssb_device *dev, const void *buffer,
 #endif /* CONFIG_SSB_BLOCKIO */
 
 
+/* The SSB DMA API. Use this API for any DMA operation on the device.
+ * This API basically is a wrapper that calls the correct DMA API for
+ * the host device type the SSB device is attached to. */
+
 /* Translation (routing) bits that need to be ORed to DMA
  * addresses before they are given to a device. */
 extern u32 ssb_dma_translation(struct ssb_device *dev);
 #define SSB_DMA_TRANSLATION_MASK	0xC0000000
 #define SSB_DMA_TRANSLATION_SHIFT	30
 
-extern int ssb_dma_set_mask(struct ssb_device *ssb_dev, u64 mask);
+extern int ssb_dma_set_mask(struct ssb_device *dev, u64 mask);
+
+extern void * ssb_dma_alloc_consistent(struct ssb_device *dev, size_t size,
+				       dma_addr_t *dma_handle, gfp_t gfp_flags);
+extern void ssb_dma_free_consistent(struct ssb_device *dev, size_t size,
+				    void *vaddr, dma_addr_t dma_handle,
+				    gfp_t gfp_flags);
+
+static inline void __cold __ssb_dma_not_implemented(struct ssb_device *dev)
+{
+#ifdef CONFIG_SSB_DEBUG
+	printk(KERN_ERR "SSB: BUG! Calling DMA API for "
+	       "unsupported bustype %d\n", dev->bus->bustype);
+#endif /* DEBUG */
+}
+
+static inline int ssb_dma_mapping_error(struct ssb_device *dev, dma_addr_t addr)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		return pci_dma_mapping_error(addr);
+	case SSB_BUSTYPE_SSB:
+		return dma_mapping_error(addr);
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+	return -ENOSYS;
+}
+
+static inline dma_addr_t ssb_dma_map_single(struct ssb_device *dev, void *p,
+					    size_t size, enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		return pci_map_single(dev->bus->host_pci, p, size, dir);
+	case SSB_BUSTYPE_SSB:
+		return dma_map_single(dev->dev, p, size, dir);
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+	return 0;
+}
+
+static inline void ssb_dma_unmap_single(struct ssb_device *dev, dma_addr_t dma_addr,
+					size_t size, enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		pci_unmap_single(dev->bus->host_pci, dma_addr, size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_unmap_single(dev->dev, dma_addr, size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_for_cpu(struct ssb_device *dev,
+					       dma_addr_t dma_addr,
+					       size_t size,
+					       enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		pci_dma_sync_single_for_cpu(dev->bus->host_pci, dma_addr,
+					    size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_for_cpu(dev->dev, dma_addr, size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_for_device(struct ssb_device *dev,
+						  dma_addr_t dma_addr,
+						  size_t size,
+						  enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		pci_dma_sync_single_for_device(dev->bus->host_pci, dma_addr,
+					       size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_for_device(dev->dev, dma_addr, size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_range_for_cpu(struct ssb_device *dev,
+						     dma_addr_t dma_addr,
+						     unsigned long offset,
+						     size_t size,
+						     enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		/* Just sync everything. That's all the PCI API can do. */
+		pci_dma_sync_single_for_cpu(dev->bus->host_pci, dma_addr,
+					    offset + size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_range_for_cpu(dev->dev, dma_addr, offset,
+					      size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
+
+static inline void ssb_dma_sync_single_range_for_device(struct ssb_device *dev,
+							dma_addr_t dma_addr,
+							unsigned long offset,
+							size_t size,
+							enum dma_data_direction dir)
+{
+	switch (dev->bus->bustype) {
+	case SSB_BUSTYPE_PCI:
+		/* Just sync everything. That's all the PCI API can do. */
+		pci_dma_sync_single_for_device(dev->bus->host_pci, dma_addr,
+					       offset + size, dir);
+		return;
+	case SSB_BUSTYPE_SSB:
+		dma_sync_single_range_for_device(dev->dev, dma_addr, offset,
+						 size, dir);
+		return;
+	default:
+		__ssb_dma_not_implemented(dev);
+	}
+}
 
 
 #ifdef CONFIG_SSB_PCIHOST
-- 
cgit v1.2.3


From ffd7891dc909b3648e87f7cf8f84a6dc12fc1cc6 Mon Sep 17 00:00:00 2001
From: "Luis R. Rodriguez" <lrodriguez@atheros.com>
Date: Sat, 21 Jun 2008 10:02:46 -0400
Subject: mac80211: Let drivers have access to TKIP key offets for TX and RX
 MIC

Some drivers may want to to use the TKIP key offsets for TX and RX
MIC so lets move this out. Lets also clear up a bit how this is used
internally in mac80211.

Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/nl80211.h |  5 ++++-
 include/net/mac80211.h  |  7 ++++++-
 net/mac80211/key.h      | 37 ++++++++++++-------------------------
 net/mac80211/tkip.c     | 10 +++++-----
 net/mac80211/wpa.c      | 20 ++++++++++++++------
 5 files changed, 41 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index aa8411e2a160..2be7c63bc0f2 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -241,7 +241,10 @@ enum nl80211_attrs {
 	NL80211_ATTR_MAX = __NL80211_ATTR_AFTER_LAST - 1
 };
 
-#define NL80211_MAX_SUPP_RATES	32
+#define NL80211_MAX_SUPP_RATES			32
+#define NL80211_TKIP_DATA_OFFSET_ENCR_KEY	0
+#define NL80211_TKIP_DATA_OFFSET_TX_MIC_KEY	16
+#define NL80211_TKIP_DATA_OFFSET_RX_MIC_KEY	24
 
 /**
  * enum nl80211_iftype - (virtual) interface types
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 7ab4ff6159a2..19f1e412a0f0 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -595,7 +595,12 @@ enum ieee80211_key_flags {
  * @flags: key flags, see &enum ieee80211_key_flags.
  * @keyidx: the key index (0-3)
  * @keylen: key material length
- * @key: key material
+ * @key: key material. For ALG_TKIP the key is encoded as a 256-bit (32 byte)
+ * 	data block:
+ * 	- Temporal Encryption Key (128 bits)
+ * 	- Temporal Authenticator Tx MIC Key (64 bits)
+ * 	- Temporal Authenticator Rx MIC Key (64 bits)
+ *
  */
 struct ieee80211_key_conf {
 	enum ieee80211_key_alg alg;
diff --git a/net/mac80211/key.h b/net/mac80211/key.h
index a0f774aafa45..425816e0996c 100644
--- a/net/mac80211/key.h
+++ b/net/mac80211/key.h
@@ -16,31 +16,18 @@
 #include <linux/rcupdate.h>
 #include <net/mac80211.h>
 
-/* ALG_TKIP
- * struct ieee80211_key::key is encoded as a 256-bit (32 byte) data block:
- * Temporal Encryption Key (128 bits)
- * Temporal Authenticator Tx MIC Key (64 bits)
- * Temporal Authenticator Rx MIC Key (64 bits)
- */
-
-#define WEP_IV_LEN 4
-#define WEP_ICV_LEN 4
-
-#define ALG_TKIP_KEY_LEN 32
-/* Starting offsets for each key */
-#define ALG_TKIP_TEMP_ENCR_KEY 0
-#define ALG_TKIP_TEMP_AUTH_TX_MIC_KEY 16
-#define ALG_TKIP_TEMP_AUTH_RX_MIC_KEY 24
-#define TKIP_IV_LEN 8
-#define TKIP_ICV_LEN 4
-
-#define ALG_CCMP_KEY_LEN 16
-#define CCMP_HDR_LEN 8
-#define CCMP_MIC_LEN 8
-#define CCMP_TK_LEN 16
-#define CCMP_PN_LEN 6
-
-#define NUM_RX_DATA_QUEUES 17
+#define WEP_IV_LEN		4
+#define WEP_ICV_LEN		4
+#define ALG_TKIP_KEY_LEN	32
+#define ALG_CCMP_KEY_LEN	16
+#define CCMP_HDR_LEN		8
+#define CCMP_MIC_LEN		8
+#define CCMP_TK_LEN		16
+#define CCMP_PN_LEN		6
+#define TKIP_IV_LEN		8
+#define TKIP_ICV_LEN		4
+
+#define NUM_RX_DATA_QUEUES	17
 
 struct ieee80211_local;
 struct ieee80211_sub_if_data;
diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c
index 699807889988..995f7af3d25e 100644
--- a/net/mac80211/tkip.c
+++ b/net/mac80211/tkip.c
@@ -164,7 +164,7 @@ void ieee80211_get_tkip_key(struct ieee80211_key_conf *keyconf,
 	iv16 = data[2] | (data[0] << 8);
 	iv32 = get_unaligned_le32(&data[4]);
 
-	tk = &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY];
+	tk = &key->conf.key[NL80211_TKIP_DATA_OFFSET_ENCR_KEY];
 	ctx = &key->u.tkip.tx;
 
 #ifdef CONFIG_MAC80211_TKIP_DEBUG
@@ -205,7 +205,7 @@ void ieee80211_tkip_encrypt_data(struct crypto_blkcipher *tfm,
 {
 	u8 rc4key[16];
 	struct tkip_ctx *ctx = &key->u.tkip.tx;
-	const u8 *tk = &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY];
+	const u8 *tk = &key->conf.key[NL80211_TKIP_DATA_OFFSET_ENCR_KEY];
 
 	/* Calculate per-packet key */
 	if (ctx->iv16 == 0 || !ctx->initialized)
@@ -231,7 +231,7 @@ int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm,
 	u32 iv16;
 	u8 rc4key[16], keyid, *pos = payload;
 	int res;
-	const u8 *tk = &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY];
+	const u8 *tk = &key->conf.key[NL80211_TKIP_DATA_OFFSET_ENCR_KEY];
 
 	if (payload_len < 12)
 		return -1;
@@ -286,13 +286,13 @@ int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm,
 #ifdef CONFIG_MAC80211_TKIP_DEBUG
 		{
 			int i;
+			u8 key_offset = NL80211_TKIP_DATA_OFFSET_ENCR_KEY;
 			DECLARE_MAC_BUF(mac);
 			printk(KERN_DEBUG "TKIP decrypt: Phase1 TA=%s"
 			       " TK=", print_mac(mac, ta));
 			for (i = 0; i < 16; i++)
 				printk("%02x ",
-				       key->conf.key[
-						ALG_TKIP_TEMP_ENCR_KEY + i]);
+				       key->conf.key[key_offset + i]);
 			printk("\n");
 			printk(KERN_DEBUG "TKIP decrypt: P1K=");
 			for (i = 0; i < 5; i++)
diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c
index 345e10e9b313..f809761fbfb5 100644
--- a/net/mac80211/wpa.c
+++ b/net/mac80211/wpa.c
@@ -49,7 +49,7 @@ static int ieee80211_get_hdr_info(const struct sk_buff *skb, u8 **sa, u8 **da,
 ieee80211_tx_result
 ieee80211_tx_h_michael_mic_add(struct ieee80211_tx_data *tx)
 {
-	u8 *data, *sa, *da, *key, *mic, qos_tid;
+	u8 *data, *sa, *da, *key, *mic, qos_tid, key_offset;
 	size_t data_len;
 	u16 fc;
 	struct sk_buff *skb = tx->skb;
@@ -88,8 +88,12 @@ ieee80211_tx_h_michael_mic_add(struct ieee80211_tx_data *tx)
 #else
 	authenticator = 1;
 #endif
-	key = &tx->key->conf.key[authenticator ? ALG_TKIP_TEMP_AUTH_TX_MIC_KEY :
-				 ALG_TKIP_TEMP_AUTH_RX_MIC_KEY];
+	/* At this point we know we're using ALG_TKIP. To get the MIC key
+	 * we now will rely on the offset from the ieee80211_key_conf::key */
+	key_offset = authenticator ?
+		NL80211_TKIP_DATA_OFFSET_TX_MIC_KEY :
+		NL80211_TKIP_DATA_OFFSET_RX_MIC_KEY;
+	key = &tx->key->conf.key[key_offset];
 	mic = skb_put(skb, MICHAEL_MIC_LEN);
 	michael_mic(key, da, sa, qos_tid & 0x0f, data, data_len, mic);
 
@@ -100,7 +104,7 @@ ieee80211_tx_h_michael_mic_add(struct ieee80211_tx_data *tx)
 ieee80211_rx_result
 ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx)
 {
-	u8 *data, *sa, *da, *key = NULL, qos_tid;
+	u8 *data, *sa, *da, *key = NULL, qos_tid, key_offset;
 	size_t data_len;
 	u16 fc;
 	u8 mic[MICHAEL_MIC_LEN];
@@ -131,8 +135,12 @@ ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx)
 #else
 	authenticator = 1;
 #endif
-	key = &rx->key->conf.key[authenticator ? ALG_TKIP_TEMP_AUTH_RX_MIC_KEY :
-				 ALG_TKIP_TEMP_AUTH_TX_MIC_KEY];
+	/* At this point we know we're using ALG_TKIP. To get the MIC key
+	 * we now will rely on the offset from the ieee80211_key_conf::key */
+	key_offset = authenticator ?
+		NL80211_TKIP_DATA_OFFSET_RX_MIC_KEY :
+		NL80211_TKIP_DATA_OFFSET_TX_MIC_KEY;
+	key = &rx->key->conf.key[key_offset];
 	michael_mic(key, da, sa, qos_tid & 0x0f, data, data_len, mic);
 	if (memcmp(mic, data + data_len, MICHAEL_MIC_LEN) != 0 || wpa_test) {
 		if (!(rx->flags & IEEE80211_RX_RA_MATCH))
-- 
cgit v1.2.3


From 9433f6dd3a4677e9b92c6e1cd7f98b11598b7c2c Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Thu, 26 Jun 2008 10:50:04 +0800
Subject: PCI: Fix comment of pci_dynids

struct pci_driver has no field of driver_data.
It's in pci_device_id.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/pci.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/pci.h b/include/linux/pci.h
index f1f73f79a180..76c9a4a96152 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -338,7 +338,7 @@ struct pci_bus_region {
 struct pci_dynids {
 	spinlock_t lock;            /* protects list, index */
 	struct list_head list;      /* for IDs added at runtime */
-	unsigned int use_driver_data:1; /* pci_driver->driver_data is used */
+	unsigned int use_driver_data:1; /* pci_device_id->driver_data is used */
 };
 
 /* ---------------------------------------------------------------- */
-- 
cgit v1.2.3


From a0da84f35b25875870270d16b6eccda4884d61a7 Mon Sep 17 00:00:00 2001
From: Neil Brown <neilb@notabene.brown>
Date: Sat, 28 Jun 2008 08:31:22 +1000
Subject: Improve setting of "events_cleared" for write-intent bitmaps.

When an array is degraded, bits in the write-intent bitmap are not
cleared, so that if the missing device is re-added, it can be synced
by only updated those parts of the device that have changed since
it was removed.

The enable this a 'events_cleared' value is stored. It is the event
counter for the array the last time that any bits were cleared.

Sometimes - if a device disappears from an array while it is 'clean' -
the events_cleared value gets updated incorrectly (there are subtle
ordering issues between updateing events in the main metadata and the
bitmap metadata) resulting in the missing device appearing to require
a full resync when it is re-added.

With this patch, we update events_cleared precisely when we are about
to clear a bit in the bitmap.  We record events_cleared when we clear
the bit internally, and copy that to the superblock which is written
out before the bit on storage.  This makes it more "obviously correct".

We also need to update events_cleared when the event_count is going
backwards (as happens on a dirty->clean transition of a non-degraded
array).

Thanks to Mike Snitzer for identifying this problem and testing early
"fixes".

Cc:  "Mike Snitzer" <snitzer@gmail.com>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/bitmap.c         | 29 ++++++++++++++++++++++++-----
 include/linux/raid/bitmap.h |  1 +
 2 files changed, 25 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index b26927ce889c..dedba16d42f7 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -454,8 +454,11 @@ void bitmap_update_sb(struct bitmap *bitmap)
 	spin_unlock_irqrestore(&bitmap->lock, flags);
 	sb = (bitmap_super_t *)kmap_atomic(bitmap->sb_page, KM_USER0);
 	sb->events = cpu_to_le64(bitmap->mddev->events);
-	if (!bitmap->mddev->degraded)
-		sb->events_cleared = cpu_to_le64(bitmap->mddev->events);
+	if (bitmap->mddev->events < bitmap->events_cleared) {
+		/* rocking back to read-only */
+		bitmap->events_cleared = bitmap->mddev->events;
+		sb->events_cleared = cpu_to_le64(bitmap->events_cleared);
+	}
 	kunmap_atomic(sb, KM_USER0);
 	write_page(bitmap, bitmap->sb_page, 1);
 }
@@ -1085,9 +1088,19 @@ void bitmap_daemon_work(struct bitmap *bitmap)
 			} else
 				spin_unlock_irqrestore(&bitmap->lock, flags);
 			lastpage = page;
-/*
-			printk("bitmap clean at page %lu\n", j);
-*/
+
+			/* We are possibly going to clear some bits, so make
+			 * sure that events_cleared is up-to-date.
+			 */
+			if (bitmap->need_sync) {
+				bitmap_super_t *sb;
+				bitmap->need_sync = 0;
+				sb = kmap_atomic(bitmap->sb_page, KM_USER0);
+				sb->events_cleared =
+					cpu_to_le64(bitmap->events_cleared);
+				kunmap_atomic(sb, KM_USER0);
+				write_page(bitmap, bitmap->sb_page, 1);
+			}
 			spin_lock_irqsave(&bitmap->lock, flags);
 			clear_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
 		}
@@ -1257,6 +1270,12 @@ void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long secto
 			return;
 		}
 
+		if (success &&
+		    bitmap->events_cleared < bitmap->mddev->events) {
+			bitmap->events_cleared = bitmap->mddev->events;
+			bitmap->need_sync = 1;
+		}
+
 		if (!success && ! (*bmc & NEEDED_MASK))
 			*bmc |= NEEDED_MASK;
 
diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h
index 78bfdea24a8e..e98900671ca9 100644
--- a/include/linux/raid/bitmap.h
+++ b/include/linux/raid/bitmap.h
@@ -221,6 +221,7 @@ struct bitmap {
 	unsigned long syncchunk;
 
 	__u64	events_cleared;
+	int need_sync;
 
 	/* bitmap spinlock */
 	spinlock_t lock;
-- 
cgit v1.2.3


From 5e96ee65c8bd629ce093da67a066d3946468298a Mon Sep 17 00:00:00 2001
From: Neil Brown <neilb@notabene.brown>
Date: Sat, 28 Jun 2008 08:31:24 +1000
Subject: Allow setting start point for requested check/repair

This makes it possible to just resync a small part of an array.
e.g. if a drive reports that it has questionable sectors,
a 'repair' of just the region covering those sectors will
cause them to be read and, if there is an error, re-written
with correct data.

Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/md.c           | 47 ++++++++++++++++++++++++++++++++++++++++++-----
 include/linux/raid/md_k.h |  2 ++
 2 files changed, 44 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 2580ac1b9b0f..261322722c19 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -278,6 +278,7 @@ static mddev_t * mddev_find(dev_t unit)
 	init_waitqueue_head(&new->sb_wait);
 	init_waitqueue_head(&new->recovery_wait);
 	new->reshape_position = MaxSector;
+	new->resync_min = 0;
 	new->resync_max = MaxSector;
 	new->level = LEVEL_NONE;
 
@@ -3074,6 +3075,36 @@ sync_completed_show(mddev_t *mddev, char *page)
 
 static struct md_sysfs_entry md_sync_completed = __ATTR_RO(sync_completed);
 
+static ssize_t
+min_sync_show(mddev_t *mddev, char *page)
+{
+	return sprintf(page, "%llu\n",
+		       (unsigned long long)mddev->resync_min);
+}
+static ssize_t
+min_sync_store(mddev_t *mddev, const char *buf, size_t len)
+{
+	unsigned long long min;
+	if (strict_strtoull(buf, 10, &min))
+		return -EINVAL;
+	if (min > mddev->resync_max)
+		return -EINVAL;
+	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
+		return -EBUSY;
+
+	/* Must be a multiple of chunk_size */
+	if (mddev->chunk_size) {
+		if (min & (sector_t)((mddev->chunk_size>>9)-1))
+			return -EINVAL;
+	}
+	mddev->resync_min = min;
+
+	return len;
+}
+
+static struct md_sysfs_entry md_min_sync =
+__ATTR(sync_min, S_IRUGO|S_IWUSR, min_sync_show, min_sync_store);
+
 static ssize_t
 max_sync_show(mddev_t *mddev, char *page)
 {
@@ -3089,9 +3120,10 @@ max_sync_store(mddev_t *mddev, const char *buf, size_t len)
 	if (strncmp(buf, "max", 3) == 0)
 		mddev->resync_max = MaxSector;
 	else {
-		char *ep;
-		unsigned long long max = simple_strtoull(buf, &ep, 10);
-		if (ep == buf || (*ep != 0 && *ep != '\n'))
+		unsigned long long max;
+		if (strict_strtoull(buf, 10, &max))
+			return -EINVAL;
+		if (max < mddev->resync_min)
 			return -EINVAL;
 		if (max < mddev->resync_max &&
 		    test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
@@ -3222,6 +3254,7 @@ static struct attribute *md_redundancy_attrs[] = {
 	&md_sync_speed.attr,
 	&md_sync_force_parallel.attr,
 	&md_sync_completed.attr,
+	&md_min_sync.attr,
 	&md_max_sync.attr,
 	&md_suspend_lo.attr,
 	&md_suspend_hi.attr,
@@ -3777,6 +3810,7 @@ static int do_md_stop(mddev_t * mddev, int mode)
 		mddev->size = 0;
 		mddev->raid_disks = 0;
 		mddev->recovery_cp = 0;
+		mddev->resync_min = 0;
 		mddev->resync_max = MaxSector;
 		mddev->reshape_position = MaxSector;
 		mddev->external = 0;
@@ -5625,9 +5659,11 @@ void md_do_sync(mddev_t *mddev)
 		max_sectors = mddev->resync_max_sectors;
 		mddev->resync_mismatches = 0;
 		/* we don't use the checkpoint if there's a bitmap */
-		if (!mddev->bitmap &&
-		    !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
+		if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
+			j = mddev->resync_min;
+		else if (!mddev->bitmap)
 			j = mddev->recovery_cp;
+
 	} else if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
 		max_sectors = mddev->size << 1;
 	else {
@@ -5796,6 +5832,7 @@ void md_do_sync(mddev_t *mddev)
 
  skip:
 	mddev->curr_resync = 0;
+	mddev->resync_min = 0;
 	mddev->resync_max = MaxSector;
 	sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 	wake_up(&resync_wait);
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 3dea9f545c8f..780e0613e6d5 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -227,6 +227,8 @@ struct mddev_s
 	atomic_t			recovery_active; /* blocks scheduled, but not written */
 	wait_queue_head_t		recovery_wait;
 	sector_t			recovery_cp;
+	sector_t			resync_min;	/* user requested sync
+							 * starts here */
 	sector_t			resync_max;	/* resync should pause
 							 * when it gets here */
 
-- 
cgit v1.2.3


From 72a23c211e4587859d5bf61ac4962d76e593fb02 Mon Sep 17 00:00:00 2001
From: Neil Brown <neilb@notabene.brown>
Date: Sat, 28 Jun 2008 08:31:41 +1000
Subject: Make sure all changes to md/sync_action are notified.

When the 'resync' thread starts or stops, when we explicitly
set sync_action, or when we determine that there is definitely nothing
to do, we notify sync_action.

To stop "sync_action" from occasionally showing the wrong value,
we introduce a new flags - MD_RECOVERY_RECOVER - to say that a
recovery is probably needed or happening, and we make sure
that we set MD_RECOVERY_RUNNING before clearing MD_RECOVERY_NEEDED.

Signed-off-by: Neil Brown <neilb@suse.de>
---
 Documentation/md.txt      |  6 ++++++
 drivers/md/md.c           | 34 ++++++++++++++++++++++++++++------
 include/linux/raid/md_k.h |  2 ++
 3 files changed, 36 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/md.txt b/Documentation/md.txt
index dca97ba4944a..c05bfb55659e 100644
--- a/Documentation/md.txt
+++ b/Documentation/md.txt
@@ -386,6 +386,12 @@ also have
 	'check' and 'repair' will start the appropriate process
            providing the current state is 'idle'.
 
+      This file responds to select/poll.  Any important change in the value
+      triggers a poll event.  Sometimes the value will briefly be
+      "recover" if a recovery seems to be needed, but cannot be
+      achieved. In that case, the transition to "recover" isn't
+      notified, but the transition away is.
+
    mismatch_count
       When performing 'check' and 'repair', and possibly when
       performing 'resync', md will count the number of errors that are
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 5b9d4fe4e6e4..c26dcad8a3ac 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -169,7 +169,6 @@ void md_new_event(mddev_t *mddev)
 {
 	atomic_inc(&md_event_count);
 	wake_up(&md_event_waiters);
-	sysfs_notify(&mddev->kobj, NULL, "sync_action");
 }
 EXPORT_SYMBOL_GPL(md_new_event);
 
@@ -2936,7 +2935,7 @@ action_show(mddev_t *mddev, char *page)
 				type = "check";
 			else
 				type = "repair";
-		} else
+		} else if (test_bit(MD_RECOVERY_RECOVER, &mddev->recovery))
 			type = "recover";
 	}
 	return sprintf(page, "%s\n", type);
@@ -2958,9 +2957,12 @@ action_store(mddev_t *mddev, const char *page, size_t len)
 	} else if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) ||
 		   test_bit(MD_RECOVERY_NEEDED, &mddev->recovery))
 		return -EBUSY;
-	else if (cmd_match(page, "resync") || cmd_match(page, "recover"))
+	else if (cmd_match(page, "resync"))
+		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
+	else if (cmd_match(page, "recover")) {
+		set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
-	else if (cmd_match(page, "reshape")) {
+	} else if (cmd_match(page, "reshape")) {
 		int err;
 		if (mddev->pers->start_reshape == NULL)
 			return -EINVAL;
@@ -2977,6 +2979,7 @@ action_store(mddev_t *mddev, const char *page, size_t len)
 	}
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
+	sysfs_notify(&mddev->kobj, NULL, "sync_action");
 	return len;
 }
 
@@ -3682,6 +3685,7 @@ static int do_md_run(mddev_t * mddev)
 	mddev->changed = 1;
 	md_new_event(mddev);
 	sysfs_notify(&mddev->kobj, NULL, "array_state");
+	sysfs_notify(&mddev->kobj, NULL, "sync_action");
 	kobject_uevent(&mddev->gendisk->dev.kobj, KOBJ_CHANGE);
 	return 0;
 }
@@ -4252,6 +4256,8 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
 			export_rdev(rdev);
 
 		md_update_sb(mddev, 1);
+		if (mddev->degraded)
+			set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 		md_wakeup_thread(mddev->thread);
 		return err;
@@ -5105,6 +5111,8 @@ void md_error(mddev_t *mddev, mdk_rdev_t *rdev)
 	if (!mddev->pers->error_handler)
 		return;
 	mddev->pers->error_handler(mddev,rdev);
+	if (mddev->degraded)
+		set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
@@ -6055,13 +6063,18 @@ void md_check_recovery(mddev_t *mddev)
 			mddev->recovery = 0;
 			/* flag recovery needed just to double check */
 			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
+			sysfs_notify(&mddev->kobj, NULL, "sync_action");
 			md_new_event(mddev);
 			goto unlock;
 		}
+		/* Set RUNNING before clearing NEEDED to avoid
+		 * any transients in the value of "sync_action".
+		 */
+		set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
+		clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 		/* Clear some bits that don't mean anything, but
 		 * might be left set
 		 */
-		clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 		clear_bit(MD_RECOVERY_INTR, &mddev->recovery);
 		clear_bit(MD_RECOVERY_DONE, &mddev->recovery);
 
@@ -6079,17 +6092,19 @@ void md_check_recovery(mddev_t *mddev)
 				/* Cannot proceed */
 				goto unlock;
 			set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
+			clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 		} else if ((spares = remove_and_add_spares(mddev))) {
 			clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
 			clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
+			set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 		} else if (mddev->recovery_cp < MaxSector) {
 			set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
+			clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
 		} else if (!test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
 			/* nothing to be done ... */
 			goto unlock;
 
 		if (mddev->pers->sync_request) {
-			set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
 			if (spares && mddev->bitmap && ! mddev->bitmap->file) {
 				/* We are adding a device or devices to an array
 				 * which has the bitmap stored on all devices.
@@ -6108,9 +6123,16 @@ void md_check_recovery(mddev_t *mddev)
 				mddev->recovery = 0;
 			} else
 				md_wakeup_thread(mddev->sync_thread);
+			sysfs_notify(&mddev->kobj, NULL, "sync_action");
 			md_new_event(mddev);
 		}
 	unlock:
+		if (!mddev->sync_thread) {
+			clear_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
+			if (test_and_clear_bit(MD_RECOVERY_RECOVER,
+					       &mddev->recovery))
+				sysfs_notify(&mddev->kobj, NULL, "sync_action");
+		}
 		mddev_unlock(mddev);
 	}
 }
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 780e0613e6d5..62aa9c9a6ddc 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -188,6 +188,7 @@ struct mddev_s
 	 * NEEDED:   we might need to start a resync/recover
 	 * RUNNING:  a thread is running, or about to be started
 	 * SYNC:     actually doing a resync, not a recovery
+	 * RECOVER:  doing recovery, or need to try it.
 	 * INTR:     resync needs to be aborted for some reason
 	 * DONE:     thread is done and is waiting to be reaped
 	 * REQUEST:  user-space has requested a sync (used with SYNC)
@@ -198,6 +199,7 @@ struct mddev_s
 	 */
 #define	MD_RECOVERY_RUNNING	0
 #define	MD_RECOVERY_SYNC	1
+#define	MD_RECOVERY_RECOVER	2
 #define	MD_RECOVERY_INTR	3
 #define	MD_RECOVERY_DONE	4
 #define	MD_RECOVERY_NEEDED	5
-- 
cgit v1.2.3


From 526647320e696f434647f38421a6ecf65b859c43 Mon Sep 17 00:00:00 2001
From: Neil Brown <neilb@notabene.brown>
Date: Sat, 28 Jun 2008 08:31:44 +1000
Subject: Make sure all changes to md/dev-XX/state are notified

The important state change happens during an interrupt
in md_error.  So just set a flag there and call sysfs_notify
later in process context.

Signed-off-by: Neil Brown <neilb@suse.de>
---
 Documentation/md.txt      | 10 ++++++++++
 drivers/md/md.c           | 14 +++++++++++++-
 include/linux/raid/md_k.h |  3 +++
 3 files changed, 26 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/Documentation/md.txt b/Documentation/md.txt
index eb6e69e3732e..e06cc59437e4 100644
--- a/Documentation/md.txt
+++ b/Documentation/md.txt
@@ -297,6 +297,10 @@ Each directory contains:
 	      writemostly - device will only be subject to read
 		         requests if there are no other options.
 			 This applies only to raid1 arrays.
+	      blocked  - device has failed, metadata is "external",
+	                 and the failure hasn't been acknowledged yet.
+			 Writes that would write to this device if
+			 it were not faulty are blocked.
 	      spare    - device is working, but not a full member.
 			 This includes spares that are in the process
 			 of being recovered to
@@ -306,6 +310,12 @@ Each directory contains:
 	Writing "remove" removes the device from the array.
 	Writing "writemostly" sets the writemostly flag.
 	Writing "-writemostly" clears the writemostly flag.
+	Writing "blocked" sets the "blocked" flag.
+	Writing "-blocked" clear the "blocked" flag and allows writes
+		to complete.
+
+	This file responds to select/poll. Any change to 'faulty'
+	or 'blocked' causes an event.
 
       errors
 	An approximate count of read errors that have been detected on
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 60d4cad88c20..dc99d95a1b6d 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1886,6 +1886,8 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 
 		err = 0;
 	}
+	if (!err)
+		sysfs_notify(&rdev->kobj, NULL, "state");
 	return err ? err : len;
 }
 static struct rdev_sysfs_entry rdev_state =
@@ -1979,7 +1981,8 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 		if (err) {
 			rdev->raid_disk = -1;
 			return err;
-		}
+		} else
+			sysfs_notify(&rdev->kobj, NULL, "state");
 		sprintf(nm, "rd%d", rdev->raid_disk);
 		if (sysfs_create_link(&rdev->mddev->kobj, &rdev->kobj, nm))
 			printk(KERN_WARNING
@@ -1996,6 +1999,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 		clear_bit(Faulty, &rdev->flags);
 		clear_bit(WriteMostly, &rdev->flags);
 		set_bit(In_sync, &rdev->flags);
+		sysfs_notify(&rdev->kobj, NULL, "state");
 	}
 	return len;
 }
@@ -3525,6 +3529,7 @@ static int do_md_run(mddev_t * mddev)
 				return -EINVAL;
 			}
 		}
+		sysfs_notify(&rdev->kobj, NULL, "state");
 	}
 
 	md_probe(mddev->unit, NULL, NULL);
@@ -4256,6 +4261,8 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
 		}
 		if (err)
 			export_rdev(rdev);
+		else
+			sysfs_notify(&rdev->kobj, NULL, "state");
 
 		md_update_sb(mddev, 1);
 		if (mddev->degraded)
@@ -5115,6 +5122,7 @@ void md_error(mddev_t *mddev, mdk_rdev_t *rdev)
 	mddev->pers->error_handler(mddev,rdev);
 	if (mddev->degraded)
 		set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
+	set_bit(StateChanged, &rdev->flags);
 	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
@@ -6037,6 +6045,10 @@ void md_check_recovery(mddev_t *mddev)
 		if (mddev->flags)
 			md_update_sb(mddev, 0);
 
+		rdev_for_each(rdev, rtmp, mddev)
+			if (test_and_clear_bit(StateChanged, &rdev->flags))
+				sysfs_notify(&rdev->kobj, NULL, "state");
+
 
 		if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&
 		    !test_bit(MD_RECOVERY_DONE, &mddev->recovery)) {
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 62aa9c9a6ddc..df30c4395875 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -87,6 +87,9 @@ struct mdk_rdev_s
 #define Blocked		8		/* An error occured on an externally
 					 * managed array, don't allow writes
 					 * until it is cleared */
+#define StateChanged	9		/* Faulty or Blocked has changed during
+					 * interrupt, so it needs to be
+					 * notified by the thread */
 	wait_queue_head_t blocked_wait;
 
 	int desc_nr;			/* descriptor index in the superblock */
-- 
cgit v1.2.3


From b203886edbcaac3ca427cf4dbcb50b18bdb346fd Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Sat, 28 Jun 2008 08:31:50 +1000
Subject: md: kill STRIPE_OP_MOD_DMA in raid5 offload

From: Dan Williams <dan.j.williams@intel.com>

This micro-optimization allowed the raid code to skip a re-read of the
parity block after checking parity.  It took advantage of the fact that
xor-offload-engines have their own internal result buffer and can check
parity without writing to memory.  Remove it for the following reasons:

1/ It is a layering violation for MD to need to manage the DMA and
   non-DMA paths within async_xor_zero_sum
2/ Bad precedent to toggle the 'ops' flags outside the lock
3/ Hard to realize a performance gain as reads will not need an updated
   parity block and writes will dirty it anyways.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/raid5.c         | 10 ----------
 include/linux/raid/raid5.h |  2 --
 2 files changed, 12 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 8c4e6149daea..60e61d2464b5 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -837,15 +837,10 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 static void ops_complete_check(void *stripe_head_ref)
 {
 	struct stripe_head *sh = stripe_head_ref;
-	int pd_idx = sh->pd_idx;
 
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
 
-	if (test_and_clear_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending) &&
-		sh->ops.zero_sum_result == 0)
-		set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
-
 	set_bit(STRIPE_OP_CHECK, &sh->ops.complete);
 	set_bit(STRIPE_HANDLE, &sh->state);
 	release_stripe(sh);
@@ -873,11 +868,6 @@ static void ops_run_check(struct stripe_head *sh)
 	tx = async_xor_zero_sum(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
 		&sh->ops.zero_sum_result, 0, NULL, NULL, NULL);
 
-	if (tx)
-		set_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending);
-	else
-		clear_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending);
-
 	atomic_inc(&sh->count);
 	tx = async_trigger_callback(ASYNC_TX_DEP_ACK | ASYNC_TX_ACK, tx,
 		ops_complete_check, sh);
diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h
index f0827d31ae6f..4ecae31a3dcb 100644
--- a/include/linux/raid/raid5.h
+++ b/include/linux/raid/raid5.h
@@ -267,10 +267,8 @@ struct r6_state {
 
 /* modifiers to the base operations
  * STRIPE_OP_MOD_REPAIR_PD - compute the parity block and write it back
- * STRIPE_OP_MOD_DMA_CHECK - parity is not corrupted by the check
  */
 #define STRIPE_OP_MOD_REPAIR_PD 7
-#define STRIPE_OP_MOD_DMA_CHECK 8
 
 /*
  * Plugging:
-- 
cgit v1.2.3


From 2b7497f0e0a0b9cf21d822e427d5399b2056501a Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Sat, 28 Jun 2008 08:31:52 +1000
Subject: md: kill STRIPE_OP_IO flag

From: Dan Williams <dan.j.williams@intel.com>

The R5_Want{Read,Write} flags already gate i/o.  So, this flag is
superfluous and we can unconditionally call ops_run_io().

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/raid5.c         | 32 +++++---------------------------
 include/linux/raid/raid5.h |  1 -
 2 files changed, 5 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 60e61d2464b5..cac97080b278 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -373,8 +373,6 @@ static unsigned long get_stripe_work(struct stripe_head *sh)
 	test_and_ack_op(STRIPE_OP_BIODRAIN, pending);
 	test_and_ack_op(STRIPE_OP_POSTXOR, pending);
 	test_and_ack_op(STRIPE_OP_CHECK, pending);
-	if (test_and_clear_bit(STRIPE_OP_IO, &sh->ops.pending))
-		ack++;
 
 	sh->ops.count -= ack;
 	if (unlikely(sh->ops.count < 0)) {
@@ -399,7 +397,6 @@ static void ops_run_io(struct stripe_head *sh)
 
 	might_sleep();
 
-	set_bit(STRIPE_IO_STARTED, &sh->state);
 	for (i = disks; i--; ) {
 		int rw;
 		struct bio *bi;
@@ -433,6 +430,8 @@ static void ops_run_io(struct stripe_head *sh)
 				test_bit(STRIPE_EXPAND_READY, &sh->state))
 				md_sync_acct(rdev->bdev, STRIPE_SECTORS);
 
+			set_bit(STRIPE_IO_STARTED, &sh->state);
+
 			bi->bi_bdev = rdev->bdev;
 			pr_debug("%s: for %llu schedule op %ld on disc %d\n",
 				__func__, (unsigned long long)sh->sector,
@@ -900,9 +899,6 @@ static void raid5_run_ops(struct stripe_head *sh, unsigned long pending)
 	if (test_bit(STRIPE_OP_CHECK, &pending))
 		ops_run_check(sh);
 
-	if (test_bit(STRIPE_OP_IO, &pending))
-		ops_run_io(sh);
-
 	if (overlap_clear)
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
@@ -2013,8 +2009,6 @@ static int __handle_issuing_new_read_requests5(struct stripe_head *sh,
 			 */
 			set_bit(R5_LOCKED, &dev->flags);
 			set_bit(R5_Wantread, &dev->flags);
-			if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
-				sh->ops.count++;
 			s->locked++;
 			pr_debug("Reading block %d (sync=%d)\n", disk_idx,
 				s->syncing);
@@ -2208,9 +2202,6 @@ static void handle_issuing_new_write_requests5(raid5_conf_t *conf,
 						"%d for r-m-w\n", i);
 					set_bit(R5_LOCKED, &dev->flags);
 					set_bit(R5_Wantread, &dev->flags);
-					if (!test_and_set_bit(
-						STRIPE_OP_IO, &sh->ops.pending))
-						sh->ops.count++;
 					s->locked++;
 				} else {
 					set_bit(STRIPE_DELAYED, &sh->state);
@@ -2234,9 +2225,6 @@ static void handle_issuing_new_write_requests5(raid5_conf_t *conf,
 						"%d for Reconstruct\n", i);
 					set_bit(R5_LOCKED, &dev->flags);
 					set_bit(R5_Wantread, &dev->flags);
-					if (!test_and_set_bit(
-						STRIPE_OP_IO, &sh->ops.pending))
-						sh->ops.count++;
 					s->locked++;
 				} else {
 					set_bit(STRIPE_DELAYED, &sh->state);
@@ -2444,8 +2432,6 @@ static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
 
 		set_bit(R5_LOCKED, &dev->flags);
 		set_bit(R5_Wantwrite, &dev->flags);
-		if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
-			sh->ops.count++;
 
 		clear_bit(STRIPE_DEGRADED, &sh->state);
 		s->locked++;
@@ -2801,9 +2787,6 @@ static void handle_stripe5(struct stripe_head *sh)
 				(i == sh->pd_idx || dev->written)) {
 				pr_debug("Writing block %d\n", i);
 				set_bit(R5_Wantwrite, &dev->flags);
-				if (!test_and_set_bit(
-				    STRIPE_OP_IO, &sh->ops.pending))
-					sh->ops.count++;
 				if (prexor)
 					continue;
 				if (!test_bit(R5_Insync, &dev->flags) ||
@@ -2857,16 +2840,12 @@ static void handle_stripe5(struct stripe_head *sh)
 		dev = &sh->dev[s.failed_num];
 		if (!test_bit(R5_ReWrite, &dev->flags)) {
 			set_bit(R5_Wantwrite, &dev->flags);
-			if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
-				sh->ops.count++;
 			set_bit(R5_ReWrite, &dev->flags);
 			set_bit(R5_LOCKED, &dev->flags);
 			s.locked++;
 		} else {
 			/* let's read it back */
 			set_bit(R5_Wantread, &dev->flags);
-			if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
-				sh->ops.count++;
 			set_bit(R5_LOCKED, &dev->flags);
 			s.locked++;
 		}
@@ -2884,13 +2863,10 @@ static void handle_stripe5(struct stripe_head *sh)
 		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.ack);
 		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
 
-		for (i = conf->raid_disks; i--; ) {
+		for (i = conf->raid_disks; i--; )
 			set_bit(R5_Wantwrite, &sh->dev[i].flags);
 			set_bit(R5_LOCKED, &dev->flags);
 			s.locked++;
-			if (!test_and_set_bit(STRIPE_OP_IO, &sh->ops.pending))
-				sh->ops.count++;
-		}
 	}
 
 	if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state) &&
@@ -2926,6 +2902,8 @@ static void handle_stripe5(struct stripe_head *sh)
 	if (pending)
 		raid5_run_ops(sh, pending);
 
+	ops_run_io(sh);
+
 	return_io(return_bi);
 
 }
diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h
index 4ecae31a3dcb..1301195abf4b 100644
--- a/include/linux/raid/raid5.h
+++ b/include/linux/raid/raid5.h
@@ -263,7 +263,6 @@ struct r6_state {
 #define STRIPE_OP_BIODRAIN	3
 #define STRIPE_OP_POSTXOR	4
 #define STRIPE_OP_CHECK	5
-#define STRIPE_OP_IO		6
 
 /* modifiers to the base operations
  * STRIPE_OP_MOD_REPAIR_PD - compute the parity block and write it back
-- 
cgit v1.2.3


From ecc65c9b3f9b9d740a5deade3d85b39be56401b6 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Sat, 28 Jun 2008 08:31:57 +1000
Subject: md: replace STRIPE_OP_CHECK with 'check_states'

From: Dan Williams <dan.j.williams@intel.com>

The STRIPE_OP_* flags record the state of stripe operations which are
performed outside the stripe lock.  Their use in indicating which
operations need to be run is straightforward; however, interpolating what
the next state of the stripe should be based on a given combination of
these flags is not straightforward, and has led to bugs.  An easier to read
implementation with minimal degrees of freedom is needed.

Towards this goal, this patch introduces explicit states to replace what was
previously interpolated from the STRIPE_OP_* flags.  For now this only converts
the handle_parity_checks5 path, removing a user of the
ops.{pending,ack,complete,count} fields of struct stripe_operations.

This conversion also found a remaining issue with the current code.  There is
a small window for a drive to fail between when we schedule a repair and when
the parity calculation for that repair completes.  When this happens we will
writeback to 'failed_num' when we really want to write back to 'pd_idx'.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/raid5.c         | 172 ++++++++++++++++++++++-----------------------
 include/linux/raid/raid5.h |  46 ++++++++++--
 2 files changed, 123 insertions(+), 95 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 6f3dd12dd3a4..544e1600f208 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -605,7 +605,11 @@ static void ops_complete_compute5(void *stripe_head_ref)
 	set_bit(R5_UPTODATE, &tgt->flags);
 	BUG_ON(!test_bit(R5_Wantcompute, &tgt->flags));
 	clear_bit(R5_Wantcompute, &tgt->flags);
-	set_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete);
+	clear_bit(STRIPE_COMPUTE_RUN, &sh->state);
+	if (sh->check_state == check_state_compute_run)
+		sh->check_state = check_state_compute_result;
+	else
+		set_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete);
 	set_bit(STRIPE_HANDLE, &sh->state);
 	release_stripe(sh);
 }
@@ -838,7 +842,7 @@ static void ops_complete_check(void *stripe_head_ref)
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
 
-	set_bit(STRIPE_OP_CHECK, &sh->ops.complete);
+	sh->check_state = check_state_check_result;
 	set_bit(STRIPE_HANDLE, &sh->state);
 	release_stripe(sh);
 }
@@ -870,7 +874,8 @@ static void ops_run_check(struct stripe_head *sh)
 		ops_complete_check, sh);
 }
 
-static void raid5_run_ops(struct stripe_head *sh, unsigned long pending)
+static void raid5_run_ops(struct stripe_head *sh, unsigned long pending,
+			  unsigned long ops_request)
 {
 	int overlap_clear = 0, i, disks = sh->disks;
 	struct dma_async_tx_descriptor *tx = NULL;
@@ -880,7 +885,8 @@ static void raid5_run_ops(struct stripe_head *sh, unsigned long pending)
 		overlap_clear++;
 	}
 
-	if (test_bit(STRIPE_OP_COMPUTE_BLK, &pending))
+	if (test_bit(STRIPE_OP_COMPUTE_BLK, &pending) ||
+	    test_bit(STRIPE_OP_COMPUTE_BLK, &ops_request))
 		tx = ops_run_compute5(sh, pending);
 
 	if (test_bit(STRIPE_OP_PREXOR, &pending))
@@ -894,7 +900,7 @@ static void raid5_run_ops(struct stripe_head *sh, unsigned long pending)
 	if (test_bit(STRIPE_OP_POSTXOR, &pending))
 		ops_run_postxor(sh, tx, pending);
 
-	if (test_bit(STRIPE_OP_CHECK, &pending))
+	if (test_bit(STRIPE_OP_CHECK, &ops_request))
 		ops_run_check(sh);
 
 	if (overlap_clear)
@@ -1961,8 +1967,7 @@ static int __handle_issuing_new_read_requests5(struct stripe_head *sh,
 	/* don't schedule compute operations or reads on the parity block while
 	 * a check is in flight
 	 */
-	if ((disk_idx == sh->pd_idx) &&
-	     test_bit(STRIPE_OP_CHECK, &sh->ops.pending))
+	if (disk_idx == sh->pd_idx && sh->check_state)
 		return ~0;
 
 	/* is the data in this block needed, and can we get it? */
@@ -1983,9 +1988,8 @@ static int __handle_issuing_new_read_requests5(struct stripe_head *sh,
 		 * 3/ We hold off parity block re-reads until check operations
 		 * have quiesced.
 		 */
-		if ((s->uptodate == disks - 1) &&
-		    (s->failed && disk_idx == s->failed_num) &&
-		    !test_bit(STRIPE_OP_CHECK, &sh->ops.pending)) {
+		if ((s->uptodate == disks - 1) && !sh->check_state &&
+		    (s->failed && disk_idx == s->failed_num)) {
 			set_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending);
 			set_bit(R5_Wantcompute, &dev->flags);
 			sh->ops.target = disk_idx;
@@ -2021,12 +2025,8 @@ static void handle_issuing_new_read_requests5(struct stripe_head *sh,
 {
 	int i;
 
-	/* Clear completed compute operations.  Parity recovery
-	 * (STRIPE_OP_MOD_REPAIR_PD) implies a write-back which is handled
-	 * later on in this routine
-	 */
-	if (test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete) &&
-		!test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending)) {
+	/* Clear completed compute operations */
+	if (test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete)) {
 		clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete);
 		clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.ack);
 		clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending);
@@ -2350,90 +2350,85 @@ static void handle_issuing_new_write_requests6(raid5_conf_t *conf,
 static void handle_parity_checks5(raid5_conf_t *conf, struct stripe_head *sh,
 				struct stripe_head_state *s, int disks)
 {
-	int canceled_check = 0;
+	struct r5dev *dev = NULL;
 
 	set_bit(STRIPE_HANDLE, &sh->state);
 
-	/* complete a check operation */
-	if (test_and_clear_bit(STRIPE_OP_CHECK, &sh->ops.complete)) {
-		clear_bit(STRIPE_OP_CHECK, &sh->ops.ack);
-		clear_bit(STRIPE_OP_CHECK, &sh->ops.pending);
+	switch (sh->check_state) {
+	case check_state_idle:
+		/* start a new check operation if there are no failures */
 		if (s->failed == 0) {
-			if (sh->ops.zero_sum_result == 0)
-				/* parity is correct (on disc,
-				 * not in buffer any more)
-				 */
-				set_bit(STRIPE_INSYNC, &sh->state);
-			else {
-				conf->mddev->resync_mismatches +=
-					STRIPE_SECTORS;
-				if (test_bit(
-				     MD_RECOVERY_CHECK, &conf->mddev->recovery))
-					/* don't try to repair!! */
-					set_bit(STRIPE_INSYNC, &sh->state);
-				else {
-					set_bit(STRIPE_OP_COMPUTE_BLK,
-						&sh->ops.pending);
-					set_bit(STRIPE_OP_MOD_REPAIR_PD,
-						&sh->ops.pending);
-					set_bit(R5_Wantcompute,
-						&sh->dev[sh->pd_idx].flags);
-					sh->ops.target = sh->pd_idx;
-					sh->ops.count++;
-					s->uptodate++;
-				}
-			}
-		} else
-			canceled_check = 1; /* STRIPE_INSYNC is not set */
-	}
-
-	/* start a new check operation if there are no failures, the stripe is
-	 * not insync, and a repair is not in flight
-	 */
-	if (s->failed == 0 &&
-	    !test_bit(STRIPE_INSYNC, &sh->state) &&
-	    !test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending)) {
-		if (!test_and_set_bit(STRIPE_OP_CHECK, &sh->ops.pending)) {
 			BUG_ON(s->uptodate != disks);
+			sh->check_state = check_state_run;
+			set_bit(STRIPE_OP_CHECK, &s->ops_request);
 			clear_bit(R5_UPTODATE, &sh->dev[sh->pd_idx].flags);
-			sh->ops.count++;
 			s->uptodate--;
+			break;
 		}
-	}
-
-	/* check if we can clear a parity disk reconstruct */
-	if (test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete) &&
-	    test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending)) {
-
-		clear_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending);
-		clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.complete);
-		clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.ack);
-		clear_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending);
-	}
-
+		dev = &sh->dev[s->failed_num];
+		/* fall through */
+	case check_state_compute_result:
+		sh->check_state = check_state_idle;
+		if (!dev)
+			dev = &sh->dev[sh->pd_idx];
+
+		/* check that a write has not made the stripe insync */
+		if (test_bit(STRIPE_INSYNC, &sh->state))
+			break;
 
-	/* Wait for check parity and compute block operations to complete
-	 * before write-back.  If a failure occurred while the check operation
-	 * was in flight we need to cycle this stripe through handle_stripe
-	 * since the parity block may not be uptodate
-	 */
-	if (!canceled_check && !test_bit(STRIPE_INSYNC, &sh->state) &&
-	    !test_bit(STRIPE_OP_CHECK, &sh->ops.pending) &&
-	    !test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending)) {
-		struct r5dev *dev;
 		/* either failed parity check, or recovery is happening */
-		if (s->failed == 0)
-			s->failed_num = sh->pd_idx;
-		dev = &sh->dev[s->failed_num];
 		BUG_ON(!test_bit(R5_UPTODATE, &dev->flags));
 		BUG_ON(s->uptodate != disks);
 
 		set_bit(R5_LOCKED, &dev->flags);
+		s->locked++;
 		set_bit(R5_Wantwrite, &dev->flags);
 
 		clear_bit(STRIPE_DEGRADED, &sh->state);
-		s->locked++;
 		set_bit(STRIPE_INSYNC, &sh->state);
+		break;
+	case check_state_run:
+		break; /* we will be called again upon completion */
+	case check_state_check_result:
+		sh->check_state = check_state_idle;
+
+		/* if a failure occurred during the check operation, leave
+		 * STRIPE_INSYNC not set and let the stripe be handled again
+		 */
+		if (s->failed)
+			break;
+
+		/* handle a successful check operation, if parity is correct
+		 * we are done.  Otherwise update the mismatch count and repair
+		 * parity if !MD_RECOVERY_CHECK
+		 */
+		if (sh->ops.zero_sum_result == 0)
+			/* parity is correct (on disc,
+			 * not in buffer any more)
+			 */
+			set_bit(STRIPE_INSYNC, &sh->state);
+		else {
+			conf->mddev->resync_mismatches += STRIPE_SECTORS;
+			if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery))
+				/* don't try to repair!! */
+				set_bit(STRIPE_INSYNC, &sh->state);
+			else {
+				sh->check_state = check_state_compute_run;
+				set_bit(STRIPE_OP_COMPUTE_BLK, &s->ops_request);
+				set_bit(R5_Wantcompute,
+					&sh->dev[sh->pd_idx].flags);
+				sh->ops.target = sh->pd_idx;
+				s->uptodate++;
+			}
+		}
+		break;
+	case check_state_compute_run:
+		break;
+	default:
+		printk(KERN_ERR "%s: unknown check_state: %d sector: %llu\n",
+		       __func__, sh->check_state,
+		       (unsigned long long) sh->sector);
+		BUG();
 	}
 }
 
@@ -2807,7 +2802,7 @@ static void handle_stripe5(struct stripe_head *sh)
 	 *    block.
 	 */
 	if (s.to_write && !test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending) &&
-			  !test_bit(STRIPE_OP_CHECK, &sh->ops.pending))
+	    !sh->check_state)
 		handle_issuing_new_write_requests5(conf, sh, &s, disks);
 
 	/* maybe we need to check and possibly fix the parity for this stripe
@@ -2815,11 +2810,10 @@ static void handle_stripe5(struct stripe_head *sh)
 	 * data is available.  The parity check is held off while parity
 	 * dependent operations are in flight.
 	 */
-	if ((s.syncing && s.locked == 0 &&
+	if (sh->check_state ||
+	    (s.syncing && s.locked == 0 &&
 	     !test_bit(STRIPE_OP_COMPUTE_BLK, &sh->ops.pending) &&
-	     !test_bit(STRIPE_INSYNC, &sh->state)) ||
-	      test_bit(STRIPE_OP_CHECK, &sh->ops.pending) ||
-	      test_bit(STRIPE_OP_MOD_REPAIR_PD, &sh->ops.pending))
+	     !test_bit(STRIPE_INSYNC, &sh->state)))
 		handle_parity_checks5(conf, sh, &s, disks);
 
 	if (s.syncing && s.locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) {
@@ -2897,8 +2891,8 @@ static void handle_stripe5(struct stripe_head *sh)
 	if (unlikely(blocked_rdev))
 		md_wait_for_blocked_rdev(blocked_rdev, conf->mddev);
 
-	if (pending)
-		raid5_run_ops(sh, pending);
+	if (pending || s.ops_request)
+		raid5_run_ops(sh, pending, s.ops_request);
 
 	ops_run_io(sh, &s);
 
diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h
index 1301195abf4b..2c96d5fd54bf 100644
--- a/include/linux/raid/raid5.h
+++ b/include/linux/raid/raid5.h
@@ -158,6 +158,41 @@
  *    the compute block completes.
  */
 
+/*
+ * Operations state - intermediate states that are visible outside of sh->lock
+ * In general _idle indicates nothing is running, _run indicates a data
+ * processing operation is active, and _result means the data processing result
+ * is stable and can be acted upon.  For simple operations like biofill and
+ * compute that only have an _idle and _run state they are indicated with
+ * sh->state flags (STRIPE_BIOFILL_RUN and STRIPE_COMPUTE_RUN)
+ */
+/**
+ * enum check_states - handles syncing / repairing a stripe
+ * @check_state_idle - check operations are quiesced
+ * @check_state_run - check operation is running
+ * @check_state_result - set outside lock when check result is valid
+ * @check_state_compute_run - check failed and we are repairing
+ * @check_state_compute_result - set outside lock when compute result is valid
+ */
+enum check_states {
+	check_state_idle = 0,
+	check_state_run, /* parity check */
+	check_state_check_result,
+	check_state_compute_run, /* parity repair */
+	check_state_compute_result,
+};
+
+/**
+ * enum reconstruct_states - handles writing or expanding a stripe
+ */
+enum reconstruct_states {
+	reconstruct_state_idle = 0,
+	reconstruct_state_drain_run,		/* write */
+	reconstruct_state_run,			/* expand */
+	reconstruct_state_drain_result,
+	reconstruct_state_result,
+};
+
 struct stripe_head {
 	struct hlist_node	hash;
 	struct list_head	lru;			/* inactive_list or handle_list */
@@ -169,6 +204,7 @@ struct stripe_head {
 	spinlock_t		lock;
 	int			bm_seq;	/* sequence number for bitmap flushes */
 	int			disks;			/* disks in stripe */
+	enum check_states	check_state;
 	/* stripe_operations
 	 * @pending - pending ops flags (set for request->issue->complete)
 	 * @ack - submitted ops flags (set for issue->complete)
@@ -202,6 +238,7 @@ struct stripe_head_state {
 	int locked, uptodate, to_read, to_write, failed, written;
 	int to_fill, compute, req_compute, non_overwrite;
 	int failed_num;
+	unsigned long ops_request;
 };
 
 /* r6_state - extra state data only relevant to r6 */
@@ -254,8 +291,10 @@ struct r6_state {
 #define	STRIPE_EXPAND_READY	11
 #define	STRIPE_IO_STARTED	12 /* do not count towards 'bypass_count' */
 #define	STRIPE_FULL_WRITE	13 /* all blocks are set to be overwritten */
+#define	STRIPE_BIOFILL_RUN	14
+#define	STRIPE_COMPUTE_RUN	15
 /*
- * Operations flags (in issue order)
+ * Operation request flags
  */
 #define STRIPE_OP_BIOFILL	0
 #define STRIPE_OP_COMPUTE_BLK	1
@@ -264,11 +303,6 @@ struct r6_state {
 #define STRIPE_OP_POSTXOR	4
 #define STRIPE_OP_CHECK	5
 
-/* modifiers to the base operations
- * STRIPE_OP_MOD_REPAIR_PD - compute the parity block and write it back
- */
-#define STRIPE_OP_MOD_REPAIR_PD 7
-
 /*
  * Plugging:
  *
-- 
cgit v1.2.3


From 600aa10993012ff2dd5617720dac081e4f992017 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Sat, 28 Jun 2008 08:32:05 +1000
Subject: md: replace STRIPE_OP_{BIODRAIN,PREXOR,POSTXOR} with
 'reconstruct_states'

From: Dan Williams <dan.j.williams@intel.com>

Track the state of reconstruct operations (recalculating the parity block
usually due to incoming writes, or as part of array expansion)  Reduces the
scope of the STRIPE_OP_{BIODRAIN,PREXOR,POSTXOR} flags to only tracking whether
a reconstruct operation has been requested via the ops_request field of struct
stripe_head_state.

This is the final step in the removal of ops.{pending,ack,complete,count}, i.e.
the STRIPE_OP_{BIODRAIN,PREXOR,POSTXOR} flags only request an operation and do
not track the state of the operation.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/raid5.c         | 204 ++++++++++++++-------------------------------
 include/linux/raid/raid5.h |   9 +-
 2 files changed, 63 insertions(+), 150 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 835046bf384e..b9159367491a 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -122,6 +122,13 @@ static void return_io(struct bio *return_bi)
 
 static void print_raid5_conf (raid5_conf_t *conf);
 
+static int stripe_operations_active(struct stripe_head *sh)
+{
+	return sh->check_state || sh->reconstruct_state ||
+	       test_bit(STRIPE_BIOFILL_RUN, &sh->state) ||
+	       test_bit(STRIPE_COMPUTE_RUN, &sh->state);
+}
+
 static void __release_stripe(raid5_conf_t *conf, struct stripe_head *sh)
 {
 	if (atomic_dec_and_test(&sh->count)) {
@@ -141,7 +148,7 @@ static void __release_stripe(raid5_conf_t *conf, struct stripe_head *sh)
 			}
 			md_wakeup_thread(conf->mddev->thread);
 		} else {
-			BUG_ON(sh->ops.pending);
+			BUG_ON(stripe_operations_active(sh));
 			if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
 				atomic_dec(&conf->preread_active_stripes);
 				if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD)
@@ -243,7 +250,7 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int pd_idx, int
 
 	BUG_ON(atomic_read(&sh->count) != 0);
 	BUG_ON(test_bit(STRIPE_HANDLE, &sh->state));
-	BUG_ON(sh->ops.pending || sh->ops.ack || sh->ops.complete);
+	BUG_ON(stripe_operations_active(sh));
 
 	CHECK_DEVLOCK();
 	pr_debug("init_stripe called, stripe %llu\n",
@@ -344,47 +351,6 @@ static struct stripe_head *get_active_stripe(raid5_conf_t *conf, sector_t sector
 	return sh;
 }
 
-/* test_and_ack_op() ensures that we only dequeue an operation once */
-#define test_and_ack_op(op, pend) \
-do {							\
-	if (test_bit(op, &sh->ops.pending) &&		\
-		!test_bit(op, &sh->ops.complete)) {	\
-		if (test_and_set_bit(op, &sh->ops.ack)) \
-			clear_bit(op, &pend);		\
-		else					\
-			ack++;				\
-	} else						\
-		clear_bit(op, &pend);			\
-} while (0)
-
-/* find new work to run, do not resubmit work that is already
- * in flight
- */
-static unsigned long get_stripe_work(struct stripe_head *sh)
-{
-	unsigned long pending;
-	int ack = 0;
-
-	pending = sh->ops.pending;
-
-	test_and_ack_op(STRIPE_OP_BIOFILL, pending);
-	test_and_ack_op(STRIPE_OP_COMPUTE_BLK, pending);
-	test_and_ack_op(STRIPE_OP_PREXOR, pending);
-	test_and_ack_op(STRIPE_OP_BIODRAIN, pending);
-	test_and_ack_op(STRIPE_OP_POSTXOR, pending);
-	test_and_ack_op(STRIPE_OP_CHECK, pending);
-
-	sh->ops.count -= ack;
-	if (unlikely(sh->ops.count < 0)) {
-		printk(KERN_ERR "pending: %#lx ops.pending: %#lx ops.ack: %#lx "
-			"ops.complete: %#lx\n", pending, sh->ops.pending,
-			sh->ops.ack, sh->ops.complete);
-		BUG();
-	}
-
-	return pending;
-}
-
 static void
 raid5_end_read_request(struct bio *bi, int error);
 static void
@@ -609,7 +575,7 @@ static void ops_complete_compute5(void *stripe_head_ref)
 }
 
 static struct dma_async_tx_descriptor *
-ops_run_compute5(struct stripe_head *sh, unsigned long pending)
+ops_run_compute5(struct stripe_head *sh, unsigned long ops_request)
 {
 	/* kernel stack size limits the total number of disks */
 	int disks = sh->disks;
@@ -640,7 +606,7 @@ ops_run_compute5(struct stripe_head *sh, unsigned long pending)
 			ops_complete_compute5, sh);
 
 	/* ack now if postxor is not set to be run */
-	if (tx && !test_bit(STRIPE_OP_POSTXOR, &pending))
+	if (tx && !test_bit(STRIPE_OP_POSTXOR, &ops_request))
 		async_tx_ack(tx);
 
 	return tx;
@@ -652,8 +618,6 @@ static void ops_complete_prexor(void *stripe_head_ref)
 
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
-
-	set_bit(STRIPE_OP_PREXOR, &sh->ops.complete);
 }
 
 static struct dma_async_tx_descriptor *
@@ -686,7 +650,7 @@ ops_run_prexor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 
 static struct dma_async_tx_descriptor *
 ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
-		 unsigned long pending)
+		 unsigned long ops_request)
 {
 	int disks = sh->disks;
 	int pd_idx = sh->pd_idx, i;
@@ -694,7 +658,7 @@ ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 	/* check if prexor is active which means only process blocks
 	 * that are part of a read-modify-write (Wantprexor)
 	 */
-	int prexor = test_bit(STRIPE_OP_PREXOR, &pending);
+	int prexor = test_bit(STRIPE_OP_PREXOR, &ops_request);
 
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
@@ -744,7 +708,7 @@ static void ops_complete_postxor(void *stripe_head_ref)
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
 
-	set_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
+	sh->reconstruct_state = reconstruct_state_result;
 	set_bit(STRIPE_HANDLE, &sh->state);
 	release_stripe(sh);
 }
@@ -763,16 +727,14 @@ static void ops_complete_write(void *stripe_head_ref)
 			set_bit(R5_UPTODATE, &dev->flags);
 	}
 
-	set_bit(STRIPE_OP_BIODRAIN, &sh->ops.complete);
-	set_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
-
+	sh->reconstruct_state = reconstruct_state_drain_result;
 	set_bit(STRIPE_HANDLE, &sh->state);
 	release_stripe(sh);
 }
 
 static void
 ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
-		unsigned long pending)
+		unsigned long ops_request)
 {
 	/* kernel stack size limits the total number of disks */
 	int disks = sh->disks;
@@ -780,7 +742,7 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 
 	int count = 0, pd_idx = sh->pd_idx, i;
 	struct page *xor_dest;
-	int prexor = test_bit(STRIPE_OP_PREXOR, &pending);
+	int prexor = test_bit(STRIPE_OP_PREXOR, &ops_request);
 	unsigned long flags;
 	dma_async_tx_callback callback;
 
@@ -807,7 +769,7 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 	}
 
 	/* check whether this postxor is part of a write */
-	callback = test_bit(STRIPE_OP_BIODRAIN, &pending) ?
+	callback = test_bit(STRIPE_OP_BIODRAIN, &ops_request) ?
 		ops_complete_write : ops_complete_postxor;
 
 	/* 1/ if we prexor'd then the dest is reused as a source
@@ -868,8 +830,7 @@ static void ops_run_check(struct stripe_head *sh)
 		ops_complete_check, sh);
 }
 
-static void raid5_run_ops(struct stripe_head *sh, unsigned long pending,
-			  unsigned long ops_request)
+static void raid5_run_ops(struct stripe_head *sh, unsigned long ops_request)
 {
 	int overlap_clear = 0, i, disks = sh->disks;
 	struct dma_async_tx_descriptor *tx = NULL;
@@ -880,18 +841,18 @@ static void raid5_run_ops(struct stripe_head *sh, unsigned long pending,
 	}
 
 	if (test_bit(STRIPE_OP_COMPUTE_BLK, &ops_request))
-		tx = ops_run_compute5(sh, pending);
+		tx = ops_run_compute5(sh, ops_request);
 
-	if (test_bit(STRIPE_OP_PREXOR, &pending))
+	if (test_bit(STRIPE_OP_PREXOR, &ops_request))
 		tx = ops_run_prexor(sh, tx);
 
-	if (test_bit(STRIPE_OP_BIODRAIN, &pending)) {
-		tx = ops_run_biodrain(sh, tx, pending);
+	if (test_bit(STRIPE_OP_BIODRAIN, &ops_request)) {
+		tx = ops_run_biodrain(sh, tx, ops_request);
 		overlap_clear++;
 	}
 
-	if (test_bit(STRIPE_OP_POSTXOR, &pending))
-		ops_run_postxor(sh, tx, pending);
+	if (test_bit(STRIPE_OP_POSTXOR, &ops_request))
+		ops_run_postxor(sh, tx, ops_request);
 
 	if (test_bit(STRIPE_OP_CHECK, &ops_request))
 		ops_run_check(sh);
@@ -1684,11 +1645,11 @@ static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2)
 	}
 }
 
-static int
-handle_write_operations5(struct stripe_head *sh, int rcw, int expand)
+static void
+handle_write_operations5(struct stripe_head *sh, struct stripe_head_state *s,
+			 int rcw, int expand)
 {
 	int i, pd_idx = sh->pd_idx, disks = sh->disks;
-	int locked = 0;
 
 	if (rcw) {
 		/* if we are not expanding this is a proper write request, and
@@ -1696,12 +1657,12 @@ handle_write_operations5(struct stripe_head *sh, int rcw, int expand)
 		 * stripe cache
 		 */
 		if (!expand) {
-			set_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending);
-			sh->ops.count++;
-		}
+			sh->reconstruct_state = reconstruct_state_drain_run;
+			set_bit(STRIPE_OP_BIODRAIN, &s->ops_request);
+		} else
+			sh->reconstruct_state = reconstruct_state_run;
 
-		set_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
-		sh->ops.count++;
+		set_bit(STRIPE_OP_POSTXOR, &s->ops_request);
 
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
@@ -1710,21 +1671,20 @@ handle_write_operations5(struct stripe_head *sh, int rcw, int expand)
 				set_bit(R5_LOCKED, &dev->flags);
 				if (!expand)
 					clear_bit(R5_UPTODATE, &dev->flags);
-				locked++;
+				s->locked++;
 			}
 		}
-		if (locked + 1 == disks)
+		if (s->locked + 1 == disks)
 			if (!test_and_set_bit(STRIPE_FULL_WRITE, &sh->state))
 				atomic_inc(&sh->raid_conf->pending_full_writes);
 	} else {
 		BUG_ON(!(test_bit(R5_UPTODATE, &sh->dev[pd_idx].flags) ||
 			test_bit(R5_Wantcompute, &sh->dev[pd_idx].flags)));
 
-		set_bit(STRIPE_OP_PREXOR, &sh->ops.pending);
-		set_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending);
-		set_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
-
-		sh->ops.count += 3;
+		sh->reconstruct_state = reconstruct_state_drain_run;
+		set_bit(STRIPE_OP_PREXOR, &s->ops_request);
+		set_bit(STRIPE_OP_BIODRAIN, &s->ops_request);
+		set_bit(STRIPE_OP_POSTXOR, &s->ops_request);
 
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
@@ -1742,7 +1702,7 @@ handle_write_operations5(struct stripe_head *sh, int rcw, int expand)
 				set_bit(R5_Wantprexor, &dev->flags);
 				set_bit(R5_LOCKED, &dev->flags);
 				clear_bit(R5_UPTODATE, &dev->flags);
-				locked++;
+				s->locked++;
 			}
 		}
 	}
@@ -1752,13 +1712,11 @@ handle_write_operations5(struct stripe_head *sh, int rcw, int expand)
 	 */
 	set_bit(R5_LOCKED, &sh->dev[pd_idx].flags);
 	clear_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
-	locked++;
+	s->locked++;
 
-	pr_debug("%s: stripe %llu locked: %d pending: %lx\n",
+	pr_debug("%s: stripe %llu locked: %d ops_request: %lx\n",
 		__func__, (unsigned long long)sh->sector,
-		locked, sh->ops.pending);
-
-	return locked;
+		s->locked, s->ops_request);
 }
 
 /*
@@ -2005,8 +1963,7 @@ static void handle_issuing_new_read_requests5(struct stripe_head *sh,
 	 * midst of changing due to a write
 	 */
 	if (!test_bit(STRIPE_COMPUTE_RUN, &sh->state) && !sh->check_state &&
-	    !test_bit(STRIPE_OP_PREXOR, &sh->ops.pending) &&
-	    !test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending)) {
+	    !sh->reconstruct_state) {
 		for (i = disks; i--; )
 			if (__handle_issuing_new_read_requests5(
 				sh, s, i, disks) == 0)
@@ -2211,7 +2168,7 @@ static void handle_issuing_new_write_requests5(raid5_conf_t *conf,
 	if ((s->req_compute || !test_bit(STRIPE_COMPUTE_RUN, &sh->state)) &&
 	    (s->locked == 0 && (rcw == 0 || rmw == 0) &&
 	    !test_bit(STRIPE_BIT_DELAY, &sh->state)))
-		s->locked += handle_write_operations5(sh, rcw == 0, 0);
+		handle_write_operations5(sh, s, rcw == 0, 0);
 }
 
 static void handle_issuing_new_write_requests6(raid5_conf_t *conf,
@@ -2581,15 +2538,14 @@ static void handle_stripe5(struct stripe_head *sh)
 	struct bio *return_bi = NULL;
 	struct stripe_head_state s;
 	struct r5dev *dev;
-	unsigned long pending = 0;
 	mdk_rdev_t *blocked_rdev = NULL;
 	int prexor;
 
 	memset(&s, 0, sizeof(s));
-	pr_debug("handling stripe %llu, state=%#lx cnt=%d, pd_idx=%d "
-		"ops=%lx:%lx:%lx\n", (unsigned long long)sh->sector, sh->state,
-		atomic_read(&sh->count), sh->pd_idx,
-		sh->ops.pending, sh->ops.ack, sh->ops.complete);
+	pr_debug("handling stripe %llu, state=%#lx cnt=%d, pd_idx=%d check:%d "
+		 "reconstruct:%d\n", (unsigned long long)sh->sector, sh->state,
+		 atomic_read(&sh->count), sh->pd_idx, sh->check_state,
+		 sh->reconstruct_state);
 
 	spin_lock(&sh->lock);
 	clear_bit(STRIPE_HANDLE, &sh->state);
@@ -2703,34 +2659,12 @@ static void handle_stripe5(struct stripe_head *sh)
 	/* Now we check to see if any write operations have recently
 	 * completed
 	 */
-
-	/* leave prexor set until postxor is done, allows us to distinguish
-	 * a rmw from a rcw during biodrain
-	 */
 	prexor = 0;
-	if (test_bit(STRIPE_OP_PREXOR, &sh->ops.complete) &&
-		test_bit(STRIPE_OP_POSTXOR, &sh->ops.complete)) {
-
-		prexor = 1;
-		clear_bit(STRIPE_OP_PREXOR, &sh->ops.complete);
-		clear_bit(STRIPE_OP_PREXOR, &sh->ops.ack);
-		clear_bit(STRIPE_OP_PREXOR, &sh->ops.pending);
-
+	if (sh->reconstruct_state == reconstruct_state_drain_result) {
+		sh->reconstruct_state = reconstruct_state_idle;
 		for (i = disks; i--; )
-			clear_bit(R5_Wantprexor, &sh->dev[i].flags);
-	}
-
-	/* if only POSTXOR is set then this is an 'expand' postxor */
-	if (test_bit(STRIPE_OP_BIODRAIN, &sh->ops.complete) &&
-		test_bit(STRIPE_OP_POSTXOR, &sh->ops.complete)) {
-
-		clear_bit(STRIPE_OP_BIODRAIN, &sh->ops.complete);
-		clear_bit(STRIPE_OP_BIODRAIN, &sh->ops.ack);
-		clear_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending);
-
-		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
-		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.ack);
-		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
+			prexor += test_and_clear_bit(R5_Wantprexor,
+						     &sh->dev[i].flags);
 
 		/* All the 'written' buffers and the parity block are ready to
 		 * be written back to disk
@@ -2763,8 +2697,7 @@ static void handle_stripe5(struct stripe_head *sh)
 	 * 2/ A 'check' operation is in flight, as it may clobber the parity
 	 *    block.
 	 */
-	if (s.to_write && !test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending) &&
-	    !sh->check_state)
+	if (s.to_write && !sh->reconstruct_state && !sh->check_state)
 		handle_issuing_new_write_requests5(conf, sh, &s, disks);
 
 	/* maybe we need to check and possibly fix the parity for this stripe
@@ -2805,18 +2738,10 @@ static void handle_stripe5(struct stripe_head *sh)
 		}
 	}
 
-	/* Finish postxor operations initiated by the expansion
-	 * process
-	 */
-	if (test_bit(STRIPE_OP_POSTXOR, &sh->ops.complete) &&
-		!test_bit(STRIPE_OP_BIODRAIN, &sh->ops.pending)) {
-
+	/* Finish reconstruct operations initiated by the expansion process */
+	if (sh->reconstruct_state == reconstruct_state_result) {
+		sh->reconstruct_state = reconstruct_state_idle;
 		clear_bit(STRIPE_EXPANDING, &sh->state);
-
-		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.pending);
-		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.ack);
-		clear_bit(STRIPE_OP_POSTXOR, &sh->ops.complete);
-
 		for (i = conf->raid_disks; i--; )
 			set_bit(R5_Wantwrite, &sh->dev[i].flags);
 			set_bit(R5_LOCKED, &dev->flags);
@@ -2824,15 +2749,13 @@ static void handle_stripe5(struct stripe_head *sh)
 	}
 
 	if (s.expanded && test_bit(STRIPE_EXPANDING, &sh->state) &&
-		!test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending)) {
+	    !sh->reconstruct_state) {
 		/* Need to write out all blocks after computing parity */
 		sh->disks = conf->raid_disks;
 		sh->pd_idx = stripe_to_pdidx(sh->sector, conf,
 			conf->raid_disks);
-		s.locked += handle_write_operations5(sh, 1, 1);
-	} else if (s.expanded &&
-		   s.locked == 0 &&
-		!test_bit(STRIPE_OP_POSTXOR, &sh->ops.pending)) {
+		handle_write_operations5(sh, &s, 1, 1);
+	} else if (s.expanded && !sh->reconstruct_state && s.locked == 0) {
 		clear_bit(STRIPE_EXPAND_READY, &sh->state);
 		atomic_dec(&conf->reshape_stripes);
 		wake_up(&conf->wait_for_overlap);
@@ -2843,9 +2766,6 @@ static void handle_stripe5(struct stripe_head *sh)
 	    !test_bit(STRIPE_COMPUTE_RUN, &sh->state))
 		handle_stripe_expansion(conf, sh, NULL);
 
-	if (sh->ops.count)
-		pending = get_stripe_work(sh);
-
  unlock:
 	spin_unlock(&sh->lock);
 
@@ -2853,8 +2773,8 @@ static void handle_stripe5(struct stripe_head *sh)
 	if (unlikely(blocked_rdev))
 		md_wait_for_blocked_rdev(blocked_rdev, conf->mddev);
 
-	if (pending || s.ops_request)
-		raid5_run_ops(sh, pending, s.ops_request);
+	if (s.ops_request)
+		raid5_run_ops(sh, s.ops_request);
 
 	ops_run_io(sh, &s);
 
diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h
index 2c96d5fd54bf..5f3e674b87dd 100644
--- a/include/linux/raid/raid5.h
+++ b/include/linux/raid/raid5.h
@@ -205,19 +205,12 @@ struct stripe_head {
 	int			bm_seq;	/* sequence number for bitmap flushes */
 	int			disks;			/* disks in stripe */
 	enum check_states	check_state;
+	enum reconstruct_states reconstruct_state;
 	/* stripe_operations
-	 * @pending - pending ops flags (set for request->issue->complete)
-	 * @ack - submitted ops flags (set for issue->complete)
-	 * @complete - completed ops flags (set for complete)
 	 * @target - STRIPE_OP_COMPUTE_BLK target
-	 * @count - raid5_runs_ops is set to run when this is non-zero
 	 */
 	struct stripe_operations {
-		unsigned long	   pending;
-		unsigned long	   ack;
-		unsigned long	   complete;
 		int		   target;
-		int		   count;
 		u32		   zero_sum_result;
 	} ops;
 	struct r5dev {
-- 
cgit v1.2.3


From d8ee0728b5b30d7a6f62c399a95e953616d31f23 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Sat, 28 Jun 2008 08:32:06 +1000
Subject: md: replace R5_WantPrexor with R5_WantDrain, add 'prexor'
 reconstruct_states

From: Dan Williams <dan.j.williams@intel.com>

Currently ops_run_biodrain and other locations have extra logic to determine
which blocks are processed in the prexor and non-prexor cases.  This can be
eliminated if handle_write_operations5 flags the blocks to be processed in all
cases via R5_Wantdrain.  The presence of the prexor operation is tracked in
sh->reconstruct_state.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/raid5.c         | 89 +++++++++++++++-------------------------------
 include/linux/raid/raid5.h |  6 ++--
 2 files changed, 32 insertions(+), 63 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index b9159367491a..c71246061c0e 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -637,7 +637,7 @@ ops_run_prexor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 	for (i = disks; i--; ) {
 		struct r5dev *dev = &sh->dev[i];
 		/* Only process blocks that are known to be uptodate */
-		if (dev->towrite && test_bit(R5_Wantprexor, &dev->flags))
+		if (test_bit(R5_Wantdrain, &dev->flags))
 			xor_srcs[count++] = dev->page;
 	}
 
@@ -649,16 +649,10 @@ ops_run_prexor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 }
 
 static struct dma_async_tx_descriptor *
-ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
-		 unsigned long ops_request)
+ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 {
 	int disks = sh->disks;
-	int pd_idx = sh->pd_idx, i;
-
-	/* check if prexor is active which means only process blocks
-	 * that are part of a read-modify-write (Wantprexor)
-	 */
-	int prexor = test_bit(STRIPE_OP_PREXOR, &ops_request);
+	int i;
 
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
@@ -666,20 +660,8 @@ ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 	for (i = disks; i--; ) {
 		struct r5dev *dev = &sh->dev[i];
 		struct bio *chosen;
-		int towrite;
 
-		towrite = 0;
-		if (prexor) { /* rmw */
-			if (dev->towrite &&
-			    test_bit(R5_Wantprexor, &dev->flags))
-				towrite = 1;
-		} else { /* rcw */
-			if (i != pd_idx && dev->towrite &&
-				test_bit(R5_LOCKED, &dev->flags))
-				towrite = 1;
-		}
-
-		if (towrite) {
+		if (test_and_clear_bit(R5_Wantdrain, &dev->flags)) {
 			struct bio *wbi;
 
 			spin_lock(&sh->lock);
@@ -702,18 +684,6 @@ ops_run_biodrain(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 }
 
 static void ops_complete_postxor(void *stripe_head_ref)
-{
-	struct stripe_head *sh = stripe_head_ref;
-
-	pr_debug("%s: stripe %llu\n", __func__,
-		(unsigned long long)sh->sector);
-
-	sh->reconstruct_state = reconstruct_state_result;
-	set_bit(STRIPE_HANDLE, &sh->state);
-	release_stripe(sh);
-}
-
-static void ops_complete_write(void *stripe_head_ref)
 {
 	struct stripe_head *sh = stripe_head_ref;
 	int disks = sh->disks, i, pd_idx = sh->pd_idx;
@@ -727,14 +697,21 @@ static void ops_complete_write(void *stripe_head_ref)
 			set_bit(R5_UPTODATE, &dev->flags);
 	}
 
-	sh->reconstruct_state = reconstruct_state_drain_result;
+	if (sh->reconstruct_state == reconstruct_state_drain_run)
+		sh->reconstruct_state = reconstruct_state_drain_result;
+	else if (sh->reconstruct_state == reconstruct_state_prexor_drain_run)
+		sh->reconstruct_state = reconstruct_state_prexor_drain_result;
+	else {
+		BUG_ON(sh->reconstruct_state != reconstruct_state_run);
+		sh->reconstruct_state = reconstruct_state_result;
+	}
+
 	set_bit(STRIPE_HANDLE, &sh->state);
 	release_stripe(sh);
 }
 
 static void
-ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
-		unsigned long ops_request)
+ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx)
 {
 	/* kernel stack size limits the total number of disks */
 	int disks = sh->disks;
@@ -742,9 +719,8 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 
 	int count = 0, pd_idx = sh->pd_idx, i;
 	struct page *xor_dest;
-	int prexor = test_bit(STRIPE_OP_PREXOR, &ops_request);
+	int prexor = 0;
 	unsigned long flags;
-	dma_async_tx_callback callback;
 
 	pr_debug("%s: stripe %llu\n", __func__,
 		(unsigned long long)sh->sector);
@@ -752,7 +728,8 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 	/* check if prexor is active which means only process blocks
 	 * that are part of a read-modify-write (written)
 	 */
-	if (prexor) {
+	if (sh->reconstruct_state == reconstruct_state_prexor_drain_run) {
+		prexor = 1;
 		xor_dest = xor_srcs[count++] = sh->dev[pd_idx].page;
 		for (i = disks; i--; ) {
 			struct r5dev *dev = &sh->dev[i];
@@ -768,10 +745,6 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 		}
 	}
 
-	/* check whether this postxor is part of a write */
-	callback = test_bit(STRIPE_OP_BIODRAIN, &ops_request) ?
-		ops_complete_write : ops_complete_postxor;
-
 	/* 1/ if we prexor'd then the dest is reused as a source
 	 * 2/ if we did not prexor then we are redoing the parity
 	 * set ASYNC_TX_XOR_DROP_DST and ASYNC_TX_XOR_ZERO_DST
@@ -785,10 +758,10 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
 	if (unlikely(count == 1)) {
 		flags &= ~(ASYNC_TX_XOR_DROP_DST | ASYNC_TX_XOR_ZERO_DST);
 		tx = async_memcpy(xor_dest, xor_srcs[0], 0, 0, STRIPE_SIZE,
-			flags, tx, callback, sh);
+			flags, tx, ops_complete_postxor, sh);
 	} else
 		tx = async_xor(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
-			flags, tx, callback, sh);
+			flags, tx, ops_complete_postxor, sh);
 }
 
 static void ops_complete_check(void *stripe_head_ref)
@@ -847,12 +820,12 @@ static void raid5_run_ops(struct stripe_head *sh, unsigned long ops_request)
 		tx = ops_run_prexor(sh, tx);
 
 	if (test_bit(STRIPE_OP_BIODRAIN, &ops_request)) {
-		tx = ops_run_biodrain(sh, tx, ops_request);
+		tx = ops_run_biodrain(sh, tx);
 		overlap_clear++;
 	}
 
 	if (test_bit(STRIPE_OP_POSTXOR, &ops_request))
-		ops_run_postxor(sh, tx, ops_request);
+		ops_run_postxor(sh, tx);
 
 	if (test_bit(STRIPE_OP_CHECK, &ops_request))
 		ops_run_check(sh);
@@ -1669,6 +1642,7 @@ handle_write_operations5(struct stripe_head *sh, struct stripe_head_state *s,
 
 			if (dev->towrite) {
 				set_bit(R5_LOCKED, &dev->flags);
+				set_bit(R5_Wantdrain, &dev->flags);
 				if (!expand)
 					clear_bit(R5_UPTODATE, &dev->flags);
 				s->locked++;
@@ -1681,7 +1655,7 @@ handle_write_operations5(struct stripe_head *sh, struct stripe_head_state *s,
 		BUG_ON(!(test_bit(R5_UPTODATE, &sh->dev[pd_idx].flags) ||
 			test_bit(R5_Wantcompute, &sh->dev[pd_idx].flags)));
 
-		sh->reconstruct_state = reconstruct_state_drain_run;
+		sh->reconstruct_state = reconstruct_state_prexor_drain_run;
 		set_bit(STRIPE_OP_PREXOR, &s->ops_request);
 		set_bit(STRIPE_OP_BIODRAIN, &s->ops_request);
 		set_bit(STRIPE_OP_POSTXOR, &s->ops_request);
@@ -1691,15 +1665,10 @@ handle_write_operations5(struct stripe_head *sh, struct stripe_head_state *s,
 			if (i == pd_idx)
 				continue;
 
-			/* For a read-modify write there may be blocks that are
-			 * locked for reading while others are ready to be
-			 * written so we distinguish these blocks by the
-			 * R5_Wantprexor bit
-			 */
 			if (dev->towrite &&
 			    (test_bit(R5_UPTODATE, &dev->flags) ||
-			    test_bit(R5_Wantcompute, &dev->flags))) {
-				set_bit(R5_Wantprexor, &dev->flags);
+			     test_bit(R5_Wantcompute, &dev->flags))) {
+				set_bit(R5_Wantdrain, &dev->flags);
 				set_bit(R5_LOCKED, &dev->flags);
 				clear_bit(R5_UPTODATE, &dev->flags);
 				s->locked++;
@@ -2660,11 +2629,11 @@ static void handle_stripe5(struct stripe_head *sh)
 	 * completed
 	 */
 	prexor = 0;
-	if (sh->reconstruct_state == reconstruct_state_drain_result) {
+	if (sh->reconstruct_state == reconstruct_state_prexor_drain_result)
+		prexor = 1;
+	if (sh->reconstruct_state == reconstruct_state_drain_result ||
+	    sh->reconstruct_state == reconstruct_state_prexor_drain_result) {
 		sh->reconstruct_state = reconstruct_state_idle;
-		for (i = disks; i--; )
-			prexor += test_and_clear_bit(R5_Wantprexor,
-						     &sh->dev[i].flags);
 
 		/* All the 'written' buffers and the parity block are ready to
 		 * be written back to disk
diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h
index 5f3e674b87dd..3b2672792457 100644
--- a/include/linux/raid/raid5.h
+++ b/include/linux/raid/raid5.h
@@ -187,8 +187,10 @@ enum check_states {
  */
 enum reconstruct_states {
 	reconstruct_state_idle = 0,
+	reconstruct_state_prexor_drain_run,	/* prexor-write */
 	reconstruct_state_drain_run,		/* write */
 	reconstruct_state_run,			/* expand */
+	reconstruct_state_prexor_drain_result,
 	reconstruct_state_drain_result,
 	reconstruct_state_result,
 };
@@ -258,9 +260,7 @@ struct r6_state {
 #define	R5_Wantfill	12 /* dev->toread contains a bio that needs
 				    * filling
 				    */
-#define	R5_Wantprexor	13 /* distinguish blocks ready for rmw from
-				    * other "towrites"
-				    */
+#define R5_Wantdrain	13 /* dev->towrite needs to be drained */
 /*
  * Write method
  */
-- 
cgit v1.2.3


From 818727badc14ce57dc099a075b05505d50b7956e Mon Sep 17 00:00:00 2001
From: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Date: Wed, 18 Jun 2008 15:40:12 +0300
Subject: rndis_host: pass buffer length to rndis_command

Pass buffer length to rndis_command so that rndis_command can read full
response buffer from device instead of max CONTROL_BUFFER_SIZE bytes.

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/net/usb/rndis_host.c      | 14 +++++++-------
 drivers/net/wireless/rndis_wlan.c |  4 ++--
 include/linux/usb/rndis_host.h    |  3 ++-
 3 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c
index ae467f182c40..61c98beb4d17 100644
--- a/drivers/net/usb/rndis_host.c
+++ b/drivers/net/usb/rndis_host.c
@@ -74,7 +74,7 @@ EXPORT_SYMBOL_GPL(rndis_status);
  * Call context is likely probe(), before interface name is known,
  * which is why we won't try to use it in the diagnostics.
  */
-int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf)
+int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf, int buflen)
 {
 	struct cdc_state	*info = (void *) &dev->data;
 	int			master_ifnum;
@@ -121,7 +121,7 @@ int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf)
 			USB_CDC_GET_ENCAPSULATED_RESPONSE,
 			USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE,
 			0, master_ifnum,
-			buf, CONTROL_BUFFER_SIZE,
+			buf, buflen,
 			RNDIS_CONTROL_TIMEOUT_MS);
 		if (likely(retval >= 8)) {
 			msg_len = le32_to_cpu(buf->msg_len);
@@ -239,7 +239,7 @@ static int rndis_query(struct usbnet *dev, struct usb_interface *intf,
 	u.get->len = cpu_to_le32(in_len);
 	u.get->offset = ccpu2(20);
 
-	retval = rndis_command(dev, u.header);
+	retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE);
 	if (unlikely(retval < 0)) {
 		dev_err(&intf->dev, "RNDIS_MSG_QUERY(0x%08x) failed, %d\n",
 				oid, retval);
@@ -328,7 +328,7 @@ generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags)
 	u.init->max_transfer_size = cpu_to_le32(dev->rx_urb_size);
 
 	net->change_mtu = NULL;
-	retval = rndis_command(dev, u.header);
+	retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE);
 	if (unlikely(retval < 0)) {
 		/* it might not even be an RNDIS device!! */
 		dev_err(&intf->dev, "RNDIS init failed, %d\n", retval);
@@ -409,7 +409,7 @@ generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags)
 	u.set->offset = ccpu2((sizeof *u.set) - 8);
 	*(__le32 *)(u.buf + sizeof *u.set) = RNDIS_DEFAULT_FILTER;
 
-	retval = rndis_command(dev, u.header);
+	retval = rndis_command(dev, u.header, CONTROL_BUFFER_SIZE);
 	if (unlikely(retval < 0)) {
 		dev_err(&intf->dev, "rndis set packet filter, %d\n", retval);
 		goto halt_fail_and_release;
@@ -424,7 +424,7 @@ halt_fail_and_release:
 	memset(u.halt, 0, sizeof *u.halt);
 	u.halt->msg_type = RNDIS_MSG_HALT;
 	u.halt->msg_len = ccpu2(sizeof *u.halt);
-	(void) rndis_command(dev, (void *)u.halt);
+	(void) rndis_command(dev, (void *)u.halt, CONTROL_BUFFER_SIZE);
 fail_and_release:
 	usb_set_intfdata(info->data, NULL);
 	usb_driver_release_interface(driver_of(intf), info->data);
@@ -449,7 +449,7 @@ void rndis_unbind(struct usbnet *dev, struct usb_interface *intf)
 	if (halt) {
 		halt->msg_type = RNDIS_MSG_HALT;
 		halt->msg_len = ccpu2(sizeof *halt);
-		(void) rndis_command(dev, (void *)halt);
+		(void) rndis_command(dev, (void *)halt, CONTROL_BUFFER_SIZE);
 		kfree(halt);
 	}
 
diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
index a36d2c85e26e..f001f2afd05e 100644
--- a/drivers/net/wireless/rndis_wlan.c
+++ b/drivers/net/wireless/rndis_wlan.c
@@ -448,7 +448,7 @@ static int rndis_query_oid(struct usbnet *dev, __le32 oid, void *data, int *len)
 	u.get->msg_len = ccpu2(sizeof *u.get);
 	u.get->oid = oid;
 
-	ret = rndis_command(dev, u.header);
+	ret = rndis_command(dev, u.header, buflen);
 	if (ret == 0) {
 		ret = le32_to_cpu(u.get_c->len);
 		*len = (*len > ret) ? ret : *len;
@@ -498,7 +498,7 @@ static int rndis_set_oid(struct usbnet *dev, __le32 oid, void *data, int len)
 	u.set->handle = ccpu2(0);
 	memcpy(u.buf + sizeof(*u.set), data, len);
 
-	ret = rndis_command(dev, u.header);
+	ret = rndis_command(dev, u.header, buflen);
 	if (ret == 0)
 		ret = rndis_error_status(u.set_c->status);
 
diff --git a/include/linux/usb/rndis_host.h b/include/linux/usb/rndis_host.h
index 29d6458ecb8d..0a6e6d4b929a 100644
--- a/include/linux/usb/rndis_host.h
+++ b/include/linux/usb/rndis_host.h
@@ -260,7 +260,8 @@ struct rndis_keepalive_c {	/* IN (optionally OUT) */
 
 
 extern void rndis_status(struct usbnet *dev, struct urb *urb);
-extern int rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf);
+extern int
+rndis_command(struct usbnet *dev, struct rndis_msg_hdr *buf, int buflen);
 extern int
 generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags);
 extern void rndis_unbind(struct usbnet *dev, struct usb_interface *intf);
-- 
cgit v1.2.3


From 3cadd2d98972f806165c634553ac4918b2b7920c Mon Sep 17 00:00:00 2001
From: Richard Lemon <richard@codelemon.com>
Date: Thu, 26 Jun 2008 10:10:41 -0400
Subject: Input: Add driver for iNexio serial touchscreen.

Signed-off-by: Richard Lemon <richard@codelemon.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/touchscreen/Kconfig  |  12 +++
 drivers/input/touchscreen/Makefile |   1 +
 drivers/input/touchscreen/inexio.c | 207 +++++++++++++++++++++++++++++++++++++
 include/linux/serio.h              |   1 +
 4 files changed, 221 insertions(+)
 create mode 100644 drivers/input/touchscreen/inexio.c

(limited to 'include/linux')

diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig
index 030a89734855..4085791e7be2 100644
--- a/drivers/input/touchscreen/Kconfig
+++ b/drivers/input/touchscreen/Kconfig
@@ -103,6 +103,18 @@ config TOUCHSCREEN_MTOUCH
 	  To compile this driver as a module, choose M here: the
 	  module will be called mtouch.
 
+config TOUCHSCREEN_INEXIO
+	tristate "iNexio serial touchscreens"
+	select SERIO
+	help
+	  Say Y here if you have an iNexio serial touchscreen connected to
+	  your system.
+
+	  If unsure, say N.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called inexio.
+
 config TOUCHSCREEN_MK712
 	tristate "ICS MicroClock MK712 touchscreen"
 	help
diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile
index 64a11f443dce..d63bcdcd28ec 100644
--- a/drivers/input/touchscreen/Makefile
+++ b/drivers/input/touchscreen/Makefile
@@ -12,6 +12,7 @@ obj-$(CONFIG_TOUCHSCREEN_CORGI)		+= corgi_ts.o
 obj-$(CONFIG_TOUCHSCREEN_GUNZE)		+= gunze.o
 obj-$(CONFIG_TOUCHSCREEN_ELO)		+= elo.o
 obj-$(CONFIG_TOUCHSCREEN_FUJITSU)	+= fujitsu_ts.o
+obj-$(CONFIG_TOUCHSCREEN_INEXIO)	+= inexio.o
 obj-$(CONFIG_TOUCHSCREEN_MIGOR)		+= migor_ts.o
 obj-$(CONFIG_TOUCHSCREEN_MTOUCH)	+= mtouch.o
 obj-$(CONFIG_TOUCHSCREEN_MK712)		+= mk712.o
diff --git a/drivers/input/touchscreen/inexio.c b/drivers/input/touchscreen/inexio.c
new file mode 100644
index 000000000000..192ade0a0fb9
--- /dev/null
+++ b/drivers/input/touchscreen/inexio.c
@@ -0,0 +1,207 @@
+/*
+ * iNexio serial touchscreen driver
+ *
+ * Copyright (c) 2008 Richard Lemon
+ * Based on the mtouch driver (c) Vojtech Pavlik and Dan Streetman
+ *
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+
+/*
+ * 2008/06/19 Richard Lemon <richard@codelemon.com>
+ *   Copied mtouch.c and edited for iNexio protocol
+ */
+
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/input.h>
+#include <linux/serio.h>
+#include <linux/init.h>
+
+#define DRIVER_DESC	"iNexio serial touchscreen driver"
+
+MODULE_AUTHOR("Richard Lemon <richard@codelemon.com>");
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
+/*
+ * Definitions & global arrays.
+ */
+
+#define INEXIO_FORMAT_TOUCH_BIT 0x01
+#define INEXIO_FORMAT_LENGTH 5
+#define INEXIO_RESPONSE_BEGIN_BYTE 0x80
+
+/* todo: check specs for max length of all responses */
+#define INEXIO_MAX_LENGTH 16
+
+#define INEXIO_MIN_XC 0
+#define INEXIO_MAX_XC 0x3fff
+#define INEXIO_MIN_YC 0
+#define INEXIO_MAX_YC 0x3fff
+
+#define INEXIO_GET_XC(data) (((data[1])<<7) | data[2])
+#define INEXIO_GET_YC(data) (((data[3])<<7) | data[4])
+#define INEXIO_GET_TOUCHED(data) (INEXIO_FORMAT_TOUCH_BIT & data[0])
+
+/*
+ * Per-touchscreen data.
+ */
+
+struct inexio {
+	struct input_dev *dev;
+	struct serio *serio;
+	int idx;
+	unsigned char data[INEXIO_MAX_LENGTH];
+	char phys[32];
+};
+
+static void inexio_process_data(struct inexio *pinexio)
+{
+	struct input_dev *dev = pinexio->dev;
+
+	if (INEXIO_FORMAT_LENGTH == ++pinexio->idx) {
+		input_report_abs(dev, ABS_X, INEXIO_GET_XC(pinexio->data));
+		input_report_abs(dev, ABS_Y, INEXIO_GET_YC(pinexio->data));
+		input_report_key(dev, BTN_TOUCH, INEXIO_GET_TOUCHED(pinexio->data));
+		input_sync(dev);
+
+		pinexio->idx = 0;
+	}
+}
+
+static irqreturn_t inexio_interrupt(struct serio *serio,
+		unsigned char data, unsigned int flags)
+{
+	struct inexio* pinexio = serio_get_drvdata(serio);
+
+	pinexio->data[pinexio->idx] = data;
+
+	if (INEXIO_RESPONSE_BEGIN_BYTE&pinexio->data[0])
+		inexio_process_data(pinexio);
+	else
+		printk(KERN_DEBUG "inexio.c: unknown/unsynchronized data from device, byte %x\n",pinexio->data[0]);
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * inexio_disconnect() is the opposite of inexio_connect()
+ */
+
+static void inexio_disconnect(struct serio *serio)
+{
+	struct inexio* pinexio = serio_get_drvdata(serio);
+
+	input_get_device(pinexio->dev);
+	input_unregister_device(pinexio->dev);
+	serio_close(serio);
+	serio_set_drvdata(serio, NULL);
+	input_put_device(pinexio->dev);
+	kfree(pinexio);
+}
+
+/*
+ * inexio_connect() is the routine that is called when someone adds a
+ * new serio device that supports iNexio protocol and registers it as
+ * an input device. This is usually accomplished using inputattach.
+ */
+
+static int inexio_connect(struct serio *serio, struct serio_driver *drv)
+{
+	struct inexio *pinexio;
+	struct input_dev *input_dev;
+	int err;
+
+	pinexio = kzalloc(sizeof(struct inexio), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!pinexio || !input_dev) {
+		err = -ENOMEM;
+		goto fail1;
+	}
+
+	pinexio->serio = serio;
+	pinexio->dev = input_dev;
+	snprintf(pinexio->phys, sizeof(pinexio->phys), "%s/input0", serio->phys);
+
+	input_dev->name = "iNexio Serial TouchScreen";
+	input_dev->phys = pinexio->phys;
+	input_dev->id.bustype = BUS_RS232;
+	input_dev->id.vendor = SERIO_INEXIO;
+	input_dev->id.product = 0;
+	input_dev->id.version = 0x0001;
+	input_dev->dev.parent = &serio->dev;
+	input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS);
+	input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH);
+	input_set_abs_params(pinexio->dev, ABS_X, INEXIO_MIN_XC, INEXIO_MAX_XC, 0, 0);
+	input_set_abs_params(pinexio->dev, ABS_Y, INEXIO_MIN_YC, INEXIO_MAX_YC, 0, 0);
+
+	serio_set_drvdata(serio, pinexio);
+
+	err = serio_open(serio, drv);
+	if (err)
+		goto fail2;
+
+	err = input_register_device(pinexio->dev);
+	if (err)
+		goto fail3;
+
+	return 0;
+
+ fail3:	serio_close(serio);
+ fail2:	serio_set_drvdata(serio, NULL);
+ fail1:	input_free_device(input_dev);
+	kfree(pinexio);
+	return err;
+}
+
+/*
+ * The serio driver structure.
+ */
+
+static struct serio_device_id inexio_serio_ids[] = {
+	{
+		.type	= SERIO_RS232,
+		.proto	= SERIO_INEXIO,
+		.id	= SERIO_ANY,
+		.extra	= SERIO_ANY,
+	},
+	{ 0 }
+};
+
+MODULE_DEVICE_TABLE(serio, inexio_serio_ids);
+
+static struct serio_driver inexio_drv = {
+	.driver		= {
+		.name	= "inexio",
+	},
+	.description	= DRIVER_DESC,
+	.id_table	= inexio_serio_ids,
+	.interrupt	= inexio_interrupt,
+	.connect	= inexio_connect,
+	.disconnect	= inexio_disconnect,
+};
+
+/*
+ * The functions for inserting/removing us as a module.
+ */
+
+static int __init inexio_init(void)
+{
+	return serio_register_driver(&inexio_drv);
+}
+
+static void __exit inexio_exit(void)
+{
+	serio_unregister_driver(&inexio_drv);
+}
+
+module_init(inexio_init);
+module_exit(inexio_exit);
diff --git a/include/linux/serio.h b/include/linux/serio.h
index 95674d97dabd..36c6ab81c7e0 100644
--- a/include/linux/serio.h
+++ b/include/linux/serio.h
@@ -212,5 +212,6 @@ static inline void serio_unpin_driver(struct serio *serio)
 #define SERIO_TAOSEVM	0x34
 #define SERIO_FUJITSU	0x35
 #define SERIO_ZHENHUA	0x36
+#define SERIO_INEXIO	0x37
 
 #endif
-- 
cgit v1.2.3


From 80be038593dba7aa46fb24a14f0ba83e5ade0edb Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Mon, 30 Jun 2008 11:35:53 -0700
Subject: PCI: add stub for pci_set_consistent_dma_mask()

When CONFIG_PCI=n, there is no stub for pci_set_consistent_dma_mask(),
so add one like other similar stubs.  Otherwise there can be build errors,
as here:

linux-next-20080630/drivers/ssb/main.c:1175: error: implicit declaration of
function 'pci_set_consistent_dma_mask'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/pci.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pci.h b/include/linux/pci.h
index 76c9a4a96152..96ebaa8d80e3 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -853,6 +853,11 @@ static inline int pci_set_dma_mask(struct pci_dev *dev, u64 mask)
 	return -EIO;
 }
 
+static inline int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask)
+{
+	return -EIO;
+}
+
 static inline int pci_set_dma_max_seg_size(struct pci_dev *dev,
 					unsigned int size)
 {
-- 
cgit v1.2.3


From b5470dc5fc18a8ff6517c3bb538d1479e58ecb02 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Fri, 27 Jun 2008 21:44:04 -0700
Subject: md: resolve external metadata handling deadlock in md_allow_write

md_allow_write() marks the metadata dirty while holding mddev->lock and then
waits for the write to complete.  For externally managed metadata this causes a
deadlock as userspace needs to take the lock to communicate that the metadata
update has completed.

Change md_allow_write() in the 'external' case to start the 'mark active'
operation and then return -EAGAIN.  The expected side effects while waiting for
userspace to write 'active' to 'array_state' are holding off reshape (code
currently handles -ENOMEM), cause some 'stripe_cache_size' change requests to
fail, cause some GET_BITMAP_FILE ioctl requests to fall back to GFP_NOIO, and
cause updates to 'raid_disks' to fail.  Except for 'stripe_cache_size' changes
these failures can be mitigated by coordinating with mdmon.

md_write_start() still prevents writes from occurring until the metadata
handler has had a chance to take action as it unconditionally waits for
MD_CHANGE_CLEAN to be cleared.

[neilb@suse.de: return -EAGAIN, try GFP_NOIO]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/md/md.c         | 27 ++++++++++++++++-----------
 drivers/md/raid1.c      |  6 ++++--
 drivers/md/raid5.c      | 12 +++++++++---
 include/linux/raid/md.h |  2 +-
 4 files changed, 30 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index df1230af02cd..43d033d9a05a 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -4172,9 +4172,11 @@ static int get_bitmap_file(mddev_t * mddev, void __user * arg)
 	char *ptr, *buf = NULL;
 	int err = -ENOMEM;
 
-	md_allow_write(mddev);
+	if (md_allow_write(mddev))
+		file = kmalloc(sizeof(*file), GFP_NOIO);
+	else
+		file = kmalloc(sizeof(*file), GFP_KERNEL);
 
-	file = kmalloc(sizeof(*file), GFP_KERNEL);
 	if (!file)
 		goto out;
 
@@ -5667,15 +5669,18 @@ void md_write_end(mddev_t *mddev)
  * may proceed without blocking.  It is important to call this before
  * attempting a GFP_KERNEL allocation while holding the mddev lock.
  * Must be called with mddev_lock held.
+ *
+ * In the ->external case MD_CHANGE_CLEAN can not be cleared until mddev->lock
+ * is dropped, so return -EAGAIN after notifying userspace.
  */
-void md_allow_write(mddev_t *mddev)
+int md_allow_write(mddev_t *mddev)
 {
 	if (!mddev->pers)
-		return;
+		return 0;
 	if (mddev->ro)
-		return;
+		return 0;
 	if (!mddev->pers->sync_request)
-		return;
+		return 0;
 
 	spin_lock_irq(&mddev->write_lock);
 	if (mddev->in_sync) {
@@ -5686,14 +5691,14 @@ void md_allow_write(mddev_t *mddev)
 			mddev->safemode = 1;
 		spin_unlock_irq(&mddev->write_lock);
 		md_update_sb(mddev, 0);
-
 		sysfs_notify(&mddev->kobj, NULL, "array_state");
-		/* wait for the dirty state to be recorded in the metadata */
-		wait_event(mddev->sb_wait,
-			   !test_bit(MD_CHANGE_CLEAN, &mddev->flags) &&
-			   !test_bit(MD_CHANGE_PENDING, &mddev->flags));
 	} else
 		spin_unlock_irq(&mddev->write_lock);
+
+	if (test_bit(MD_CHANGE_CLEAN, &mddev->flags))
+		return -EAGAIN;
+	else
+		return 0;
 }
 EXPORT_SYMBOL_GPL(md_allow_write);
 
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index f05d5983efb6..491dc2d4ad5f 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -2136,7 +2136,7 @@ static int raid1_reshape(mddev_t *mddev)
 	conf_t *conf = mddev_to_conf(mddev);
 	int cnt, raid_disks;
 	unsigned long flags;
-	int d, d2;
+	int d, d2, err;
 
 	/* Cannot change chunk_size, layout, or level */
 	if (mddev->chunk_size != mddev->new_chunk ||
@@ -2148,7 +2148,9 @@ static int raid1_reshape(mddev_t *mddev)
 		return -EINVAL;
 	}
 
-	md_allow_write(mddev);
+	err = md_allow_write(mddev);
+	if (err)
+		return err;
 
 	raid_disks = mddev->raid_disks + mddev->delta_disks;
 
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 442622067cae..8f4c70a53210 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -911,14 +911,16 @@ static int resize_stripes(raid5_conf_t *conf, int newsize)
 	struct stripe_head *osh, *nsh;
 	LIST_HEAD(newstripes);
 	struct disk_info *ndisks;
-	int err = 0;
+	int err;
 	struct kmem_cache *sc;
 	int i;
 
 	if (newsize <= conf->pool_size)
 		return 0; /* never bother to shrink */
 
-	md_allow_write(conf->mddev);
+	err = md_allow_write(conf->mddev);
+	if (err)
+		return err;
 
 	/* Step 1 */
 	sc = kmem_cache_create(conf->cache_name[1-conf->active_name],
@@ -3843,6 +3845,8 @@ raid5_store_stripe_cache_size(mddev_t *mddev, const char *page, size_t len)
 {
 	raid5_conf_t *conf = mddev_to_conf(mddev);
 	unsigned long new;
+	int err;
+
 	if (len >= PAGE_SIZE)
 		return -EINVAL;
 	if (!conf)
@@ -3858,7 +3862,9 @@ raid5_store_stripe_cache_size(mddev_t *mddev, const char *page, size_t len)
 		else
 			break;
 	}
-	md_allow_write(mddev);
+	err = md_allow_write(mddev);
+	if (err)
+		return err;
 	while (new > conf->max_nr_stripes) {
 		if (grow_one_stripe(conf))
 			conf->max_nr_stripes++;
diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h
index b7386ae9d288..dc0e3fcb9f28 100644
--- a/include/linux/raid/md.h
+++ b/include/linux/raid/md.h
@@ -95,7 +95,7 @@ extern int sync_page_io(struct block_device *bdev, sector_t sector, int size,
 			struct page *page, int rw);
 extern void md_do_sync(mddev_t *mddev);
 extern void md_new_event(mddev_t *mddev);
-extern void md_allow_write(mddev_t *mddev);
+extern int md_allow_write(mddev_t *mddev);
 extern void md_wait_for_blocked_rdev(mdk_rdev_t *rdev, mddev_t *mddev);
 
 #endif /* CONFIG_MD */
-- 
cgit v1.2.3


From 9b4a8dd2e9f8af206eb39e3d99c442b0d6158953 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 24 Jun 2008 03:46:57 +1000
Subject: drivers/macintosh: Various cleanups

This contains the following cleanups:
- make the following needlessly global code static:
  - adb.c: adb_controller
  - adb.c: adb_init()
  - adbhid.c: adb_to_linux_keycodes[]  (also make it const)
  - via-pmu68k.c: backlight_level
  - via-pmu68k.c: backlight_enabled
- remove the following unused code:
  - via-pmu68k.c: sleep_notifier_list

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 drivers/macintosh/adb.c        | 5 ++---
 drivers/macintosh/adbhid.c     | 2 +-
 drivers/macintosh/via-pmu68k.c | 5 ++---
 include/linux/adb.h            | 1 -
 4 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c
index dbaad39020a1..61b62a6f681b 100644
--- a/drivers/macintosh/adb.c
+++ b/drivers/macintosh/adb.c
@@ -46,7 +46,6 @@
 #endif
 
 
-EXPORT_SYMBOL(adb_controller);
 EXPORT_SYMBOL(adb_client_list);
 
 extern struct adb_driver via_macii_driver;
@@ -80,7 +79,7 @@ static struct adb_driver *adb_driver_list[] = {
 
 static struct class *adb_dev_class;
 
-struct adb_driver *adb_controller;
+static struct adb_driver *adb_controller;
 BLOCKING_NOTIFIER_HEAD(adb_client_list);
 static int adb_got_sleep;
 static int adb_inited;
@@ -290,7 +289,7 @@ static int adb_resume(struct platform_device *dev)
 }
 #endif /* CONFIG_PM */
 
-int __init adb_init(void)
+static int __init adb_init(void)
 {
 	struct adb_driver *driver;
 	int i;
diff --git a/drivers/macintosh/adbhid.c b/drivers/macintosh/adbhid.c
index ef4c117ea35f..59ea520a5d7a 100644
--- a/drivers/macintosh/adbhid.c
+++ b/drivers/macintosh/adbhid.c
@@ -75,7 +75,7 @@ static struct notifier_block adbhid_adb_notifier = {
 #define ADB_KEY_POWER_OLD	0x7e
 #define ADB_KEY_POWER		0x7f
 
-u16 adb_to_linux_keycodes[128] = {
+static const u16 adb_to_linux_keycodes[128] = {
 	/* 0x00 */ KEY_A, 		/*  30 */
 	/* 0x01 */ KEY_S, 		/*  31 */
 	/* 0x02 */ KEY_D,		/*  32 */
diff --git a/drivers/macintosh/via-pmu68k.c b/drivers/macintosh/via-pmu68k.c
index e2f84da09e7c..b64741c95ac4 100644
--- a/drivers/macintosh/via-pmu68k.c
+++ b/drivers/macintosh/via-pmu68k.c
@@ -101,7 +101,6 @@ static int pmu_kind = PMU_UNKNOWN;
 static int pmu_fully_inited;
 
 int asleep;
-BLOCKING_NOTIFIER_HEAD(sleep_notifier_list);
 
 static int pmu_probe(void);
 static int pmu_init(void);
@@ -741,8 +740,8 @@ pmu_handle_data(unsigned char *data, int len)
 	}
 }
 
-int backlight_level = -1;
-int backlight_enabled = 0;
+static int backlight_level = -1;
+static int backlight_enabled = 0;
 
 #define LEVEL_TO_BRIGHT(lev)	((lev) < 1? 0x7f: 0x4a - ((lev) << 1))
 
diff --git a/include/linux/adb.h b/include/linux/adb.h
index 64d8878e1444..63bca502fa55 100644
--- a/include/linux/adb.h
+++ b/include/linux/adb.h
@@ -84,7 +84,6 @@ enum adb_message {
     ADB_MSG_PRE_RESET,	/* Called before resetting the bus */
     ADB_MSG_POST_RESET	/* Called after resetting the bus (re-do init & register) */
 };
-extern struct adb_driver *adb_controller;
 extern struct blocking_notifier_head adb_client_list;
 
 int adb_request(struct adb_request *req, void (*done)(struct adb_request *),
-- 
cgit v1.2.3


From f3e909c2750eb20536bacacc867dc9047b70546a Mon Sep 17 00:00:00 2001
From: Michael Neuling <mikey@neuling.org>
Date: Tue, 1 Jul 2008 14:01:39 +1000
Subject: powerpc: Update for VSX core file and ptrace

This correctly hooks the VSX dump into Roland McGrath core file
infrastructure.  It adds the VSX dump information as an additional elf
note in the core file (after talking more to the tool chain/gdb guys).
This also ensures the formats are consistent between signals, ptrace
and core files.

Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/process.c | 19 +------------------
 arch/powerpc/kernel/ptrace.c  | 21 +++++++++++++--------
 include/asm-powerpc/elf.h     |  4 ++--
 include/linux/elf.h           |  1 +
 4 files changed, 17 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index d52ded366f14..1924b57bd241 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -162,7 +162,7 @@ void flush_altivec_to_thread(struct task_struct *tsk)
 	}
 }
 
-int dump_task_altivec(struct task_struct *tsk, elf_vrreg_t *vrregs)
+int dump_task_altivec(struct task_struct *tsk, elf_vrregset_t *vrregs)
 {
 	/* ELF_NVRREG includes the VSCR and VRSAVE which we need to save
 	 * separately, see below */
@@ -249,23 +249,6 @@ int dump_task_vsx(struct task_struct *tsk, elf_vrreg_t *vrregs)
 }
 #endif /* CONFIG_VSX */
 
-int dump_task_vector(struct task_struct *tsk, elf_vrregset_t *vrregs)
-{
-	int rc = 0;
-	elf_vrreg_t *regs = (elf_vrreg_t *)vrregs;
-#ifdef CONFIG_ALTIVEC
-	rc = dump_task_altivec(tsk, regs);
-	if (rc)
-		return rc;
-	regs += ELF_NVRREG;
-#endif
-
-#ifdef CONFIG_VSX
-	rc = dump_task_vsx(tsk, regs);
-#endif
-	return rc;
-}
-
 #ifdef CONFIG_SPE
 
 void enable_kernel_spe(void)
diff --git a/arch/powerpc/kernel/ptrace.c b/arch/powerpc/kernel/ptrace.c
index 4e203a89e189..8feb93e7890c 100644
--- a/arch/powerpc/kernel/ptrace.c
+++ b/arch/powerpc/kernel/ptrace.c
@@ -368,13 +368,15 @@ static int vsr_get(struct task_struct *target, const struct user_regset *regset,
 		   unsigned int pos, unsigned int count,
 		   void *kbuf, void __user *ubuf)
 {
-	int ret;
+	double buf[32];
+	int ret, i;
 
 	flush_vsx_to_thread(target);
 
+	for (i = 0; i < 32 ; i++)
+		buf[i] = current->thread.fpr[i][TS_VSRLOWOFFSET];
 	ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf,
-				  target->thread.fpr, 0,
-				  32 * sizeof(vector128));
+				  buf, 0, 32 * sizeof(double));
 
 	return ret;
 }
@@ -383,13 +385,16 @@ static int vsr_set(struct task_struct *target, const struct user_regset *regset,
 		   unsigned int pos, unsigned int count,
 		   const void *kbuf, const void __user *ubuf)
 {
-	int ret;
+	double buf[32];
+	int ret,i;
 
 	flush_vsx_to_thread(target);
 
 	ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
-				 target->thread.fpr, 0,
-				 32 * sizeof(vector128));
+				 buf, 0, 32 * sizeof(double));
+	for (i = 0; i < 32 ; i++)
+		current->thread.fpr[i][TS_VSRLOWOFFSET] = buf[i];
+
 
 	return ret;
 }
@@ -499,8 +504,8 @@ static const struct user_regset native_regsets[] = {
 #endif
 #ifdef CONFIG_VSX
 	[REGSET_VSX] = {
-		.n = 32,
-		.size = sizeof(vector128), .align = sizeof(vector128),
+		.core_note_type = NT_PPC_VSX, .n = 32,
+		.size = sizeof(double), .align = sizeof(double),
 		.active = vsr_active, .get = vsr_get, .set = vsr_set
 	},
 #endif
diff --git a/include/asm-powerpc/elf.h b/include/asm-powerpc/elf.h
index b6a874db801d..38a51728406f 100644
--- a/include/asm-powerpc/elf.h
+++ b/include/asm-powerpc/elf.h
@@ -221,8 +221,8 @@ extern int dump_task_fpu(struct task_struct *, elf_fpregset_t *);
 typedef elf_vrregset_t elf_fpxregset_t;
 
 #ifdef CONFIG_ALTIVEC
-extern int dump_task_vector(struct task_struct *, elf_vrregset_t *vrregs);
-#define ELF_CORE_COPY_XFPREGS(tsk, regs) dump_task_vector(tsk, regs)
+extern int dump_task_altivec(struct task_struct *, elf_vrregset_t *vrregs);
+#define ELF_CORE_COPY_XFPREGS(tsk, regs) dump_task_altivec(tsk, regs)
 #define ELF_CORE_XFPREG_TYPE NT_PPC_VMX
 #endif
 
diff --git a/include/linux/elf.h b/include/linux/elf.h
index ff9fbed90123..edc3dac3f02f 100644
--- a/include/linux/elf.h
+++ b/include/linux/elf.h
@@ -358,6 +358,7 @@ typedef struct elf64_shdr {
 #define NT_PRXFPREG     0x46e62b7f      /* copied from gdb5.1/include/elf/common.h */
 #define NT_PPC_VMX	0x100		/* PowerPC Altivec/VMX registers */
 #define NT_PPC_SPE	0x101		/* PowerPC SPE/EVR registers */
+#define NT_PPC_VSX	0x102		/* PowerPC VSX registers */
 #define NT_386_TLS	0x200		/* i386 TLS slots (struct user_desc) */
 
 
-- 
cgit v1.2.3


From 0853ad66b14feb12acde7ac13b7c3b75770a0adc Mon Sep 17 00:00:00 2001
From: Santwona Behera <santwona.behera@sun.com>
Date: Wed, 2 Jul 2008 03:47:41 -0700
Subject: netdev: Add support for rx flow hash configuration, using ethtool.

Added new interfaces to ethtool to configure receive network flow
distribution across multiple rx rings using hashing.

Signed-off-by: Santwona Behera <santwona.behera@sun.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 33 +++++++++++++++++++++++++++++++++
 net/core/ethtool.c      | 37 +++++++++++++++++++++++++++++++++++++
 2 files changed, 70 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index c8d216357865..8bb5e87df365 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -272,6 +272,12 @@ enum ethtool_flags {
 	ETH_FLAG_LRO		= (1 << 15),	/* LRO is enabled */
 };
 
+struct ethtool_rxnfc {
+	__u32		cmd;
+	__u32		flow_type;
+	__u64		data;
+};
+
 #ifdef __KERNEL__
 
 struct net_device;
@@ -396,6 +402,8 @@ struct ethtool_ops {
 	/* the following hooks are obsolete */
 	int	(*self_test_count)(struct net_device *);/* use get_sset_count */
 	int	(*get_stats_count)(struct net_device *);/* use get_sset_count */
+	int	(*get_rxhash)(struct net_device *, struct ethtool_rxnfc *);
+	int	(*set_rxhash)(struct net_device *, struct ethtool_rxnfc *);
 };
 #endif /* __KERNEL__ */
 
@@ -442,6 +450,9 @@ struct ethtool_ops {
 #define ETHTOOL_GPFLAGS		0x00000027 /* Get driver-private flags bitmap */
 #define ETHTOOL_SPFLAGS		0x00000028 /* Set driver-private flags bitmap */
 
+#define	ETHTOOL_GRXFH		0x00000029 /* Get RX flow hash configuration */
+#define	ETHTOOL_SRXFH		0x0000002a /* Set RX flow hash configuration */
+
 /* compatibility with older code */
 #define SPARC_ETH_GSET		ETHTOOL_GSET
 #define SPARC_ETH_SSET		ETHTOOL_SSET
@@ -528,4 +539,26 @@ struct ethtool_ops {
 #define WAKE_MAGIC		(1 << 5)
 #define WAKE_MAGICSECURE	(1 << 6) /* only meaningful if WAKE_MAGIC */
 
+/* L3-L4 network traffic flow types */
+#define	TCP_V4_FLOW	0x01
+#define	UDP_V4_FLOW	0x02
+#define	SCTP_V4_FLOW	0x03
+#define	AH_ESP_V4_FLOW	0x04
+#define	TCP_V6_FLOW	0x05
+#define	UDP_V6_FLOW	0x06
+#define	SCTP_V6_FLOW	0x07
+#define	AH_ESP_V6_FLOW	0x08
+
+/* L3-L4 network traffic flow hash options */
+#define	RXH_DEV_PORT	(1 << 0)
+#define	RXH_L2DA	(1 << 1)
+#define	RXH_VLAN	(1 << 2)
+#define	RXH_L3_PROTO	(1 << 3)
+#define	RXH_IP_SRC	(1 << 4)
+#define	RXH_IP_DST	(1 << 5)
+#define	RXH_L4_B_0_1	(1 << 6) /* src port in case of TCP/UDP/SCTP */
+#define	RXH_L4_B_2_3	(1 << 7) /* dst port in case of TCP/UDP/SCTP */
+#define	RXH_DISCARD	(1 << 31)
+
+
 #endif /* _LINUX_ETHTOOL_H */
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 0133b5ebd545..14ada537f895 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -209,6 +209,36 @@ static int ethtool_get_drvinfo(struct net_device *dev, void __user *useraddr)
 	return 0;
 }
 
+static int ethtool_set_rxhash(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_rxnfc cmd;
+
+	if (!dev->ethtool_ops->set_rxhash)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&cmd, useraddr, sizeof(cmd)))
+		return -EFAULT;
+
+	return dev->ethtool_ops->set_rxhash(dev, &cmd);
+}
+
+static int ethtool_get_rxhash(struct net_device *dev, void __user *useraddr)
+{
+	struct ethtool_rxnfc info;
+
+	if (!dev->ethtool_ops->get_rxhash)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&info, useraddr, sizeof(info)))
+		return -EFAULT;
+
+	dev->ethtool_ops->get_rxhash(dev, &info);
+
+	if (copy_to_user(useraddr, &info, sizeof(info)))
+		return -EFAULT;
+	return 0;
+}
+
 static int ethtool_get_regs(struct net_device *dev, char __user *useraddr)
 {
 	struct ethtool_regs regs;
@@ -826,6 +856,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_GGSO:
 	case ETHTOOL_GFLAGS:
 	case ETHTOOL_GPFLAGS:
+	case ETHTOOL_GRXFH:
 		break;
 	default:
 		if (!capable(CAP_NET_ADMIN))
@@ -977,6 +1008,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 		rc = ethtool_set_value(dev, useraddr,
 				       dev->ethtool_ops->set_priv_flags);
 		break;
+	case ETHTOOL_GRXFH:
+		rc = ethtool_get_rxhash(dev, useraddr);
+		break;
+	case ETHTOOL_SRXFH:
+		rc = ethtool_set_rxhash(dev, useraddr);
+		break;
 	default:
 		rc = -EOPNOTSUPP;
 	}
-- 
cgit v1.2.3


From ab96dddbedf4bb8a7a0fe44012efc1d99598c36f Mon Sep 17 00:00:00 2001
From: Tom Tucker <tom@opengridcomputing.com>
Date: Wed, 28 May 2008 13:54:04 -0500
Subject: svcrdma: Add a type for keeping NFS RPC mapping

Create a new data structure to hold the remote client address space
to local server address space mapping.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
---
 include/linux/sunrpc/svc_rdma.h          | 27 +++++++++++++++++++++++++++
 net/sunrpc/xprtrdma/svc_rdma.c           | 19 +++++++++++++++++++
 net/sunrpc/xprtrdma/svc_rdma_transport.c | 26 ++++++++++++++++++++++++++
 3 files changed, 72 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h
index 05eb4664d0dd..bd8749cc8084 100644
--- a/include/linux/sunrpc/svc_rdma.h
+++ b/include/linux/sunrpc/svc_rdma.h
@@ -86,6 +86,31 @@ struct svc_rdma_op_ctxt {
 	struct page *pages[RPCSVC_MAXPAGES];
 };
 
+/*
+ * NFS_ requests are mapped on the client side by the chunk lists in
+ * the RPCRDMA header. During the fetching of the RPC from the client
+ * and the writing of the reply to the client, the memory in the
+ * client and the memory in the server must be mapped as contiguous
+ * vaddr/len for access by the hardware. These data strucures keep
+ * these mappings.
+ *
+ * For an RDMA_WRITE, the 'sge' maps the RPC REPLY. For RDMA_READ, the
+ * 'sge' in the svc_rdma_req_map maps the server side RPC reply and the
+ * 'ch' field maps the read-list of the RPCRDMA header to the 'sge'
+ * mapping of the reply.
+ */
+struct svc_rdma_chunk_sge {
+	int start;		/* sge no for this chunk */
+	int count;		/* sge count for this chunk */
+};
+struct svc_rdma_req_map {
+	unsigned long count;
+	union {
+		struct kvec sge[RPCSVC_MAXPAGES];
+		struct svc_rdma_chunk_sge ch[RPCSVC_MAXPAGES];
+	};
+};
+
 #define RDMACTXT_F_LAST_CTXT	2
 
 struct svcxprt_rdma {
@@ -173,6 +198,8 @@ extern int svc_rdma_post_recv(struct svcxprt_rdma *);
 extern int svc_rdma_create_listen(struct svc_serv *, int, struct sockaddr *);
 extern struct svc_rdma_op_ctxt *svc_rdma_get_context(struct svcxprt_rdma *);
 extern void svc_rdma_put_context(struct svc_rdma_op_ctxt *, int);
+extern struct svc_rdma_req_map *svc_rdma_get_req_map(void);
+extern void svc_rdma_put_req_map(struct svc_rdma_req_map *);
 extern void svc_sq_reap(struct svcxprt_rdma *);
 extern void svc_rq_reap(struct svcxprt_rdma *);
 extern struct svc_xprt_class svc_rdma_class;
diff --git a/net/sunrpc/xprtrdma/svc_rdma.c b/net/sunrpc/xprtrdma/svc_rdma.c
index 88c0ca20bb1e..171f2053e90c 100644
--- a/net/sunrpc/xprtrdma/svc_rdma.c
+++ b/net/sunrpc/xprtrdma/svc_rdma.c
@@ -69,6 +69,9 @@ atomic_t rdma_stat_rq_prod;
 atomic_t rdma_stat_sq_poll;
 atomic_t rdma_stat_sq_prod;
 
+/* Temporary NFS request map cache */
+struct kmem_cache *svc_rdma_map_cachep;
+
 /*
  * This function implements reading and resetting an atomic_t stat
  * variable through read/write to a proc file. Any write to the file
@@ -241,6 +244,7 @@ void svc_rdma_cleanup(void)
 		svcrdma_table_header = NULL;
 	}
 	svc_unreg_xprt_class(&svc_rdma_class);
+	kmem_cache_destroy(svc_rdma_map_cachep);
 }
 
 int svc_rdma_init(void)
@@ -255,9 +259,24 @@ int svc_rdma_init(void)
 		svcrdma_table_header =
 			register_sysctl_table(svcrdma_root_table);
 
+	/* Create the temporary map cache */
+	svc_rdma_map_cachep = kmem_cache_create("svc_rdma_map_cache",
+						sizeof(struct svc_rdma_req_map),
+						0,
+						SLAB_HWCACHE_ALIGN,
+						NULL);
+	if (!svc_rdma_map_cachep) {
+		printk(KERN_INFO "Could not allocate map cache.\n");
+		goto err;
+	}
+
 	/* Register RDMA with the SVC transport switch */
 	svc_reg_xprt_class(&svc_rdma_class);
 	return 0;
+
+ err:
+	unregister_sysctl_table(svcrdma_table_header);
+	return -ENOMEM;
 }
 MODULE_AUTHOR("Tom Tucker <tom@opengridcomputing.com>");
 MODULE_DESCRIPTION("SVC RDMA Transport");
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index e132509d1db0..ae90758d8e9b 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -173,6 +173,32 @@ void svc_rdma_put_context(struct svc_rdma_op_ctxt *ctxt, int free_pages)
 	atomic_dec(&xprt->sc_ctxt_used);
 }
 
+/* Temporary NFS request map cache. Created in svc_rdma.c  */
+extern struct kmem_cache *svc_rdma_map_cachep;
+
+/*
+ * Temporary NFS req mappings are shared across all transport
+ * instances. These are short lived and should be bounded by the number
+ * of concurrent server threads * depth of the SQ.
+ */
+struct svc_rdma_req_map *svc_rdma_get_req_map(void)
+{
+	struct svc_rdma_req_map *map;
+	while (1) {
+		map = kmem_cache_alloc(svc_rdma_map_cachep, GFP_KERNEL);
+		if (map)
+			break;
+		schedule_timeout_uninterruptible(msecs_to_jiffies(500));
+	}
+	map->count = 0;
+	return map;
+}
+
+void svc_rdma_put_req_map(struct svc_rdma_req_map *map)
+{
+	kmem_cache_free(svc_rdma_map_cachep, map);
+}
+
 /* ib_cq event handler */
 static void cq_event_handler(struct ib_event *event, void *context)
 {
-- 
cgit v1.2.3


From f820c57ebf5493d4602cc00577c8b0fadd27a7b8 Mon Sep 17 00:00:00 2001
From: Tom Tucker <tom@opengridcomputing.com>
Date: Tue, 27 May 2008 17:03:14 -0500
Subject: svcrdma: Use reply and chunk map for RDMA_READ processing

Modify the RDMA_READ processing to use the reply and chunk list mapping data
types. Also add a special purpose 'hdr_count' field in in the context to hold
the header page count instead of overloading the SGE length field and
corrupting the DMA map length.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
---
 include/linux/sunrpc/svc_rdma.h         |  1 +
 net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 83 +++++++++++++++------------------
 2 files changed, 39 insertions(+), 45 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h
index bd8749cc8084..fd5e8a1c17de 100644
--- a/include/linux/sunrpc/svc_rdma.h
+++ b/include/linux/sunrpc/svc_rdma.h
@@ -72,6 +72,7 @@ extern atomic_t rdma_stat_sq_prod;
  */
 struct svc_rdma_op_ctxt {
 	struct svc_rdma_op_ctxt *read_hdr;
+	int hdr_count;
 	struct list_head free_list;
 	struct xdr_buf arg;
 	struct list_head dto_q;
diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index 06ab4841537b..d25971b42a74 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -112,11 +112,6 @@ static void rdma_build_arg_xdr(struct svc_rqst *rqstp,
 	rqstp->rq_arg.tail[0].iov_len = 0;
 }
 
-struct chunk_sge {
-	int start;		/* sge no for this chunk */
-	int count;		/* sge count for this chunk */
-};
-
 /* Encode a read-chunk-list as an array of IB SGE
  *
  * Assumptions:
@@ -134,8 +129,8 @@ static int rdma_rcl_to_sge(struct svcxprt_rdma *xprt,
 			   struct svc_rqst *rqstp,
 			   struct svc_rdma_op_ctxt *head,
 			   struct rpcrdma_msg *rmsgp,
-			   struct ib_sge *sge,
-			   struct chunk_sge *ch_sge_ary,
+			   struct svc_rdma_req_map *rpl_map,
+			   struct svc_rdma_req_map *chl_map,
 			   int ch_count,
 			   int byte_count)
 {
@@ -156,22 +151,18 @@ static int rdma_rcl_to_sge(struct svcxprt_rdma *xprt,
 	head->arg.head[0] = rqstp->rq_arg.head[0];
 	head->arg.tail[0] = rqstp->rq_arg.tail[0];
 	head->arg.pages = &head->pages[head->count];
-	head->sge[0].length = head->count; /* save count of hdr pages */
+	head->hdr_count = head->count; /* save count of hdr pages */
 	head->arg.page_base = 0;
 	head->arg.page_len = ch_bytes;
 	head->arg.len = rqstp->rq_arg.len + ch_bytes;
 	head->arg.buflen = rqstp->rq_arg.buflen + ch_bytes;
 	head->count++;
-	ch_sge_ary[0].start = 0;
+	chl_map->ch[0].start = 0;
 	while (byte_count) {
+		rpl_map->sge[sge_no].iov_base =
+			page_address(rqstp->rq_arg.pages[page_no]) + page_off;
 		sge_bytes = min_t(int, PAGE_SIZE-page_off, ch_bytes);
-		sge[sge_no].addr =
-			ib_dma_map_page(xprt->sc_cm_id->device,
-					rqstp->rq_arg.pages[page_no],
-					page_off, sge_bytes,
-					DMA_FROM_DEVICE);
-		sge[sge_no].length = sge_bytes;
-		sge[sge_no].lkey = xprt->sc_phys_mr->lkey;
+		rpl_map->sge[sge_no].iov_len = sge_bytes;
 		/*
 		 * Don't bump head->count here because the same page
 		 * may be used by multiple SGE.
@@ -187,11 +178,11 @@ static int rdma_rcl_to_sge(struct svcxprt_rdma *xprt,
 		 * SGE, move to the next SGE
 		 */
 		if (ch_bytes == 0) {
-			ch_sge_ary[ch_no].count =
-				sge_no - ch_sge_ary[ch_no].start;
+			chl_map->ch[ch_no].count =
+				sge_no - chl_map->ch[ch_no].start;
 			ch_no++;
 			ch++;
-			ch_sge_ary[ch_no].start = sge_no;
+			chl_map->ch[ch_no].start = sge_no;
 			ch_bytes = ch->rc_target.rs_length;
 			/* If bytes remaining account for next chunk */
 			if (byte_count) {
@@ -220,18 +211,24 @@ static int rdma_rcl_to_sge(struct svcxprt_rdma *xprt,
 	return sge_no;
 }
 
-static void rdma_set_ctxt_sge(struct svc_rdma_op_ctxt *ctxt,
-			      struct ib_sge *sge,
+static void rdma_set_ctxt_sge(struct svcxprt_rdma *xprt,
+			      struct svc_rdma_op_ctxt *ctxt,
+			      struct kvec *vec,
 			      u64 *sgl_offset,
 			      int count)
 {
 	int i;
 
 	ctxt->count = count;
+	ctxt->direction = DMA_FROM_DEVICE;
 	for (i = 0; i < count; i++) {
-		ctxt->sge[i].addr = sge[i].addr;
-		ctxt->sge[i].length = sge[i].length;
-		*sgl_offset = *sgl_offset + sge[i].length;
+		ctxt->sge[i].addr =
+			ib_dma_map_single(xprt->sc_cm_id->device,
+					  vec[i].iov_base, vec[i].iov_len,
+					  DMA_FROM_DEVICE);
+		ctxt->sge[i].length = vec[i].iov_len;
+		ctxt->sge[i].lkey = xprt->sc_phys_mr->lkey;
+		*sgl_offset = *sgl_offset + vec[i].iov_len;
 	}
 }
 
@@ -282,34 +279,29 @@ static int rdma_read_xdr(struct svcxprt_rdma *xprt,
 	struct ib_send_wr read_wr;
 	int err = 0;
 	int ch_no;
-	struct ib_sge *sge;
 	int ch_count;
 	int byte_count;
 	int sge_count;
 	u64 sgl_offset;
 	struct rpcrdma_read_chunk *ch;
 	struct svc_rdma_op_ctxt *ctxt = NULL;
-	struct svc_rdma_op_ctxt *tmp_sge_ctxt;
-	struct svc_rdma_op_ctxt *tmp_ch_ctxt;
-	struct chunk_sge *ch_sge_ary;
+	struct svc_rdma_req_map *rpl_map;
+	struct svc_rdma_req_map *chl_map;
 
 	/* If no read list is present, return 0 */
 	ch = svc_rdma_get_read_chunk(rmsgp);
 	if (!ch)
 		return 0;
 
-	/* Allocate temporary contexts to keep SGE */
-	BUG_ON(sizeof(struct ib_sge) < sizeof(struct chunk_sge));
-	tmp_sge_ctxt = svc_rdma_get_context(xprt);
-	sge = tmp_sge_ctxt->sge;
-	tmp_ch_ctxt = svc_rdma_get_context(xprt);
-	ch_sge_ary = (struct chunk_sge *)tmp_ch_ctxt->sge;
+	/* Allocate temporary reply and chunk maps */
+	rpl_map = svc_rdma_get_req_map();
+	chl_map = svc_rdma_get_req_map();
 
 	svc_rdma_rcl_chunk_counts(ch, &ch_count, &byte_count);
 	if (ch_count > RPCSVC_MAXPAGES)
 		return -EINVAL;
 	sge_count = rdma_rcl_to_sge(xprt, rqstp, hdr_ctxt, rmsgp,
-				    sge, ch_sge_ary,
+				    rpl_map, chl_map,
 				    ch_count, byte_count);
 	sgl_offset = 0;
 	ch_no = 0;
@@ -331,14 +323,15 @@ next_sge:
 		read_wr.wr.rdma.remote_addr =
 			get_unaligned(&(ch->rc_target.rs_offset)) +
 			sgl_offset;
-		read_wr.sg_list = &sge[ch_sge_ary[ch_no].start];
+		read_wr.sg_list = ctxt->sge;
 		read_wr.num_sge =
-			rdma_read_max_sge(xprt, ch_sge_ary[ch_no].count);
-		rdma_set_ctxt_sge(ctxt, &sge[ch_sge_ary[ch_no].start],
+			rdma_read_max_sge(xprt, chl_map->ch[ch_no].count);
+		rdma_set_ctxt_sge(xprt, ctxt,
+				  &rpl_map->sge[chl_map->ch[ch_no].start],
 				  &sgl_offset,
 				  read_wr.num_sge);
 		if (((ch+1)->rc_discrim == 0) &&
-		    (read_wr.num_sge == ch_sge_ary[ch_no].count)) {
+		    (read_wr.num_sge == chl_map->ch[ch_no].count)) {
 			/*
 			 * Mark the last RDMA_READ with a bit to
 			 * indicate all RPC data has been fetched from
@@ -358,9 +351,9 @@ next_sge:
 		}
 		atomic_inc(&rdma_stat_read);
 
-		if (read_wr.num_sge < ch_sge_ary[ch_no].count) {
-			ch_sge_ary[ch_no].count -= read_wr.num_sge;
-			ch_sge_ary[ch_no].start += read_wr.num_sge;
+		if (read_wr.num_sge < chl_map->ch[ch_no].count) {
+			chl_map->ch[ch_no].count -= read_wr.num_sge;
+			chl_map->ch[ch_no].start += read_wr.num_sge;
 			goto next_sge;
 		}
 		sgl_offset = 0;
@@ -368,8 +361,8 @@ next_sge:
 	}
 
  out:
-	svc_rdma_put_context(tmp_sge_ctxt, 0);
-	svc_rdma_put_context(tmp_ch_ctxt, 0);
+	svc_rdma_put_req_map(rpl_map);
+	svc_rdma_put_req_map(chl_map);
 
 	/* Detach arg pages. svc_recv will replenish them */
 	for (ch_no = 0; &rqstp->rq_pages[ch_no] < rqstp->rq_respages; ch_no++)
@@ -399,7 +392,7 @@ static int rdma_read_complete(struct svc_rqst *rqstp,
 		rqstp->rq_pages[page_no] = head->pages[page_no];
 	}
 	/* Point rq_arg.pages past header */
-	rqstp->rq_arg.pages = &rqstp->rq_pages[head->sge[0].length];
+	rqstp->rq_arg.pages = &rqstp->rq_pages[head->hdr_count];
 	rqstp->rq_arg.page_len = head->arg.page_len;
 	rqstp->rq_arg.page_base = head->arg.page_base;
 
-- 
cgit v1.2.3


From 87295b6c5c7fd7bbc0ce3e7f42d2adbbac7352b9 Mon Sep 17 00:00:00 2001
From: Tom Tucker <tom@opengridcomputing.com>
Date: Wed, 28 May 2008 13:17:44 -0500
Subject: svcrdma: Add dma map count and WARN_ON

Add a dma map count in order to verify that all DMA mapping resources
have been freed when the transport is closed.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
---
 include/linux/sunrpc/svc_rdma.h          | 1 +
 net/sunrpc/xprtrdma/svc_rdma_recvfrom.c  | 1 +
 net/sunrpc/xprtrdma/svc_rdma_sendto.c    | 3 +++
 net/sunrpc/xprtrdma/svc_rdma_transport.c | 5 +++++
 4 files changed, 10 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h
index fd5e8a1c17de..ab93afc03c43 100644
--- a/include/linux/sunrpc/svc_rdma.h
+++ b/include/linux/sunrpc/svc_rdma.h
@@ -130,6 +130,7 @@ struct svcxprt_rdma {
 
 	struct ib_pd         *sc_pd;
 
+	atomic_t	     sc_dma_used;
 	atomic_t	     sc_ctxt_used;
 	struct list_head     sc_ctxt_free;
 	int		     sc_ctxt_cnt;
diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index d25971b42a74..b4b17f44cb29 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -222,6 +222,7 @@ static void rdma_set_ctxt_sge(struct svcxprt_rdma *xprt,
 	ctxt->count = count;
 	ctxt->direction = DMA_FROM_DEVICE;
 	for (i = 0; i < count; i++) {
+		atomic_inc(&xprt->sc_dma_used);
 		ctxt->sge[i].addr =
 			ib_dma_map_single(xprt->sc_cm_id->device,
 					  vec[i].iov_base, vec[i].iov_len,
diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
index bdc11a30e937..a19b22b452a3 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
@@ -163,6 +163,7 @@ static int send_write(struct svcxprt_rdma *xprt, struct svc_rqst *rqstp,
 		sge_bytes = min((size_t)bc,
 				(size_t)(vec->sge[xdr_sge_no].iov_len-sge_off));
 		sge[sge_no].length = sge_bytes;
+		atomic_inc(&xprt->sc_dma_used);
 		sge[sge_no].addr =
 			ib_dma_map_single(xprt->sc_cm_id->device,
 					  (void *)
@@ -385,6 +386,7 @@ static int send_reply(struct svcxprt_rdma *rdma,
 	ctxt->count = 1;
 
 	/* Prepare the SGE for the RPCRDMA Header */
+	atomic_inc(&rdma->sc_dma_used);
 	ctxt->sge[0].addr =
 		ib_dma_map_page(rdma->sc_cm_id->device,
 				page, 0, PAGE_SIZE, DMA_TO_DEVICE);
@@ -396,6 +398,7 @@ static int send_reply(struct svcxprt_rdma *rdma,
 	for (sge_no = 1; byte_count && sge_no < vec->count; sge_no++) {
 		sge_bytes = min_t(size_t, vec->sge[sge_no].iov_len, byte_count);
 		byte_count -= sge_bytes;
+		atomic_inc(&rdma->sc_dma_used);
 		ctxt->sge[sge_no].addr =
 			ib_dma_map_single(rdma->sc_cm_id->device,
 					  vec->sge[sge_no].iov_base,
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index 7e8ee66458ea..6fddd588c031 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -155,6 +155,7 @@ static void svc_rdma_unmap_dma(struct svc_rdma_op_ctxt *ctxt)
 	struct svcxprt_rdma *xprt = ctxt->xprt;
 	int i;
 	for (i = 0; i < ctxt->count && ctxt->sge[i].length; i++) {
+		atomic_dec(&xprt->sc_dma_used);
 		ib_dma_unmap_single(xprt->sc_cm_id->device,
 				    ctxt->sge[i].addr,
 				    ctxt->sge[i].length,
@@ -519,6 +520,7 @@ static struct svcxprt_rdma *rdma_create_xprt(struct svc_serv *serv,
 	cma_xprt->sc_max_requests = svcrdma_max_requests;
 	cma_xprt->sc_sq_depth = svcrdma_max_requests * RPCRDMA_SQ_DEPTH_MULT;
 	atomic_set(&cma_xprt->sc_sq_count, 0);
+	atomic_set(&cma_xprt->sc_ctxt_used, 0);
 
 	if (!listener) {
 		int reqs = cma_xprt->sc_max_requests;
@@ -569,6 +571,7 @@ int svc_rdma_post_recv(struct svcxprt_rdma *xprt)
 		BUG_ON(sge_no >= xprt->sc_max_sge);
 		page = svc_rdma_get_page();
 		ctxt->pages[sge_no] = page;
+		atomic_inc(&xprt->sc_dma_used);
 		pa = ib_dma_map_page(xprt->sc_cm_id->device,
 				     page, 0, PAGE_SIZE,
 				     DMA_FROM_DEVICE);
@@ -1049,6 +1052,7 @@ static void __svc_rdma_free(struct work_struct *work)
 
 	/* Warn if we leaked a resource or under-referenced */
 	WARN_ON(atomic_read(&rdma->sc_ctxt_used) != 0);
+	WARN_ON(atomic_read(&rdma->sc_dma_used) != 0);
 
 	/* Destroy the QP if present (not a listener) */
 	if (rdma->sc_qp && !IS_ERR(rdma->sc_qp))
@@ -1169,6 +1173,7 @@ void svc_rdma_send_error(struct svcxprt_rdma *xprt, struct rpcrdma_msg *rmsgp,
 	length = svc_rdma_xdr_encode_error(xprt, rmsgp, err, va);
 
 	/* Prepare SGE for local address */
+	atomic_inc(&xprt->sc_dma_used);
 	sge.addr = ib_dma_map_page(xprt->sc_cm_id->device,
 				   p, 0, PAGE_SIZE, DMA_FROM_DEVICE);
 	sge.lkey = xprt->sc_phys_mr->lkey;
-- 
cgit v1.2.3


From 779a48577ba88b6a7e9748a04b0b739f36c5e6f6 Mon Sep 17 00:00:00 2001
From: Tom Tucker <tom@opengridcomputing.com>
Date: Mon, 19 May 2008 10:17:09 -0500
Subject: svcrdma: Remove unused wait q from svcrdma_xprt structure

The sc_read_wait queue head is no longer used. Remove it.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
---
 include/linux/sunrpc/svc_rdma.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h
index ab93afc03c43..d8d74c4ab504 100644
--- a/include/linux/sunrpc/svc_rdma.h
+++ b/include/linux/sunrpc/svc_rdma.h
@@ -119,7 +119,6 @@ struct svcxprt_rdma {
 	struct rdma_cm_id    *sc_cm_id;		/* RDMA connection id */
 	struct list_head     sc_accept_q;	/* Conn. waiting accept */
 	int		     sc_ord;		/* RDMA read limit */
-	wait_queue_head_t    sc_read_wait;
 	int                  sc_max_sge;
 
 	int                  sc_sq_depth;	/* Depth of SQ */
-- 
cgit v1.2.3


From 8948896c9e098c6fd31a6a698a598a7cbd7fa40e Mon Sep 17 00:00:00 2001
From: Tom Tucker <tom@opengridcomputing.com>
Date: Wed, 28 May 2008 15:14:02 -0500
Subject: svcrdma: Change WR context get/put to use the kmem cache

Change the WR context pool to be shared across mount points. This
reduces the RDMA transport memory footprint significantly since
idle mounts don't consume WR context memory.

Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
---
 include/linux/sunrpc/svc_rdma.h          |   6 --
 net/sunrpc/xprtrdma/svc_rdma_transport.c | 121 +++----------------------------
 2 files changed, 12 insertions(+), 115 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h
index d8d74c4ab504..ef2e3a20bf3b 100644
--- a/include/linux/sunrpc/svc_rdma.h
+++ b/include/linux/sunrpc/svc_rdma.h
@@ -73,7 +73,6 @@ extern atomic_t rdma_stat_sq_prod;
 struct svc_rdma_op_ctxt {
 	struct svc_rdma_op_ctxt *read_hdr;
 	int hdr_count;
-	struct list_head free_list;
 	struct xdr_buf arg;
 	struct list_head dto_q;
 	enum ib_wr_opcode wr_op;
@@ -131,11 +130,6 @@ struct svcxprt_rdma {
 
 	atomic_t	     sc_dma_used;
 	atomic_t	     sc_ctxt_used;
-	struct list_head     sc_ctxt_free;
-	int		     sc_ctxt_cnt;
-	int		     sc_ctxt_bump;
-	int		     sc_ctxt_max;
-	spinlock_t	     sc_ctxt_lock;
 	struct list_head     sc_rq_dto_q;
 	spinlock_t	     sc_rq_dto_lock;
 	struct ib_qp         *sc_qp;
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index 80104f4999d5..19ddc382b777 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -84,69 +84,23 @@ struct svc_xprt_class svc_rdma_class = {
 	.xcl_max_payload = RPCSVC_MAXPAYLOAD_TCP,
 };
 
-static int rdma_bump_context_cache(struct svcxprt_rdma *xprt)
-{
-	int target;
-	int at_least_one = 0;
-	struct svc_rdma_op_ctxt *ctxt;
-
-	target = min(xprt->sc_ctxt_cnt + xprt->sc_ctxt_bump,
-		     xprt->sc_ctxt_max);
-
-	spin_lock_bh(&xprt->sc_ctxt_lock);
-	while (xprt->sc_ctxt_cnt < target) {
-		xprt->sc_ctxt_cnt++;
-		spin_unlock_bh(&xprt->sc_ctxt_lock);
-
-		ctxt = kmalloc(sizeof(*ctxt), GFP_KERNEL);
-
-		spin_lock_bh(&xprt->sc_ctxt_lock);
-		if (ctxt) {
-			at_least_one = 1;
-			INIT_LIST_HEAD(&ctxt->free_list);
-			list_add(&ctxt->free_list, &xprt->sc_ctxt_free);
-		} else {
-			/* kmalloc failed...give up for now */
-			xprt->sc_ctxt_cnt--;
-			break;
-		}
-	}
-	spin_unlock_bh(&xprt->sc_ctxt_lock);
-	dprintk("svcrdma: sc_ctxt_max=%d, sc_ctxt_cnt=%d\n",
-		xprt->sc_ctxt_max, xprt->sc_ctxt_cnt);
-	return at_least_one;
-}
+/* WR context cache. Created in svc_rdma.c  */
+extern struct kmem_cache *svc_rdma_ctxt_cachep;
 
 struct svc_rdma_op_ctxt *svc_rdma_get_context(struct svcxprt_rdma *xprt)
 {
 	struct svc_rdma_op_ctxt *ctxt;
 
 	while (1) {
-		spin_lock_bh(&xprt->sc_ctxt_lock);
-		if (unlikely(list_empty(&xprt->sc_ctxt_free))) {
-			/* Try to bump my cache. */
-			spin_unlock_bh(&xprt->sc_ctxt_lock);
-
-			if (rdma_bump_context_cache(xprt))
-				continue;
-
-			printk(KERN_INFO "svcrdma: sleeping waiting for "
-			       "context memory on xprt=%p\n",
-			       xprt);
-			schedule_timeout_uninterruptible(msecs_to_jiffies(500));
-			continue;
-		}
-		ctxt = list_entry(xprt->sc_ctxt_free.next,
-				  struct svc_rdma_op_ctxt,
-				  free_list);
-		list_del_init(&ctxt->free_list);
-		spin_unlock_bh(&xprt->sc_ctxt_lock);
-		ctxt->xprt = xprt;
-		INIT_LIST_HEAD(&ctxt->dto_q);
-		ctxt->count = 0;
-		atomic_inc(&xprt->sc_ctxt_used);
-		break;
+		ctxt = kmem_cache_alloc(svc_rdma_ctxt_cachep, GFP_KERNEL);
+		if (ctxt)
+			break;
+		schedule_timeout_uninterruptible(msecs_to_jiffies(500));
 	}
+	ctxt->xprt = xprt;
+	INIT_LIST_HEAD(&ctxt->dto_q);
+	ctxt->count = 0;
+	atomic_inc(&xprt->sc_ctxt_used);
 	return ctxt;
 }
 
@@ -174,9 +128,7 @@ void svc_rdma_put_context(struct svc_rdma_op_ctxt *ctxt, int free_pages)
 		for (i = 0; i < ctxt->count; i++)
 			put_page(ctxt->pages[i]);
 
-	spin_lock_bh(&xprt->sc_ctxt_lock);
-	list_add(&ctxt->free_list, &xprt->sc_ctxt_free);
-	spin_unlock_bh(&xprt->sc_ctxt_lock);
+	kmem_cache_free(svc_rdma_ctxt_cachep, ctxt);
 	atomic_dec(&xprt->sc_ctxt_used);
 }
 
@@ -461,40 +413,6 @@ static void sq_comp_handler(struct ib_cq *cq, void *cq_context)
 	tasklet_schedule(&dto_tasklet);
 }
 
-static void create_context_cache(struct svcxprt_rdma *xprt,
-				 int ctxt_count, int ctxt_bump, int ctxt_max)
-{
-	struct svc_rdma_op_ctxt *ctxt;
-	int i;
-
-	xprt->sc_ctxt_max = ctxt_max;
-	xprt->sc_ctxt_bump = ctxt_bump;
-	xprt->sc_ctxt_cnt = 0;
-	atomic_set(&xprt->sc_ctxt_used, 0);
-
-	INIT_LIST_HEAD(&xprt->sc_ctxt_free);
-	for (i = 0; i < ctxt_count; i++) {
-		ctxt = kmalloc(sizeof(*ctxt), GFP_KERNEL);
-		if (ctxt) {
-			INIT_LIST_HEAD(&ctxt->free_list);
-			list_add(&ctxt->free_list, &xprt->sc_ctxt_free);
-			xprt->sc_ctxt_cnt++;
-		}
-	}
-}
-
-static void destroy_context_cache(struct svcxprt_rdma *xprt)
-{
-	while (!list_empty(&xprt->sc_ctxt_free)) {
-		struct svc_rdma_op_ctxt *ctxt;
-		ctxt = list_entry(xprt->sc_ctxt_free.next,
-				  struct svc_rdma_op_ctxt,
-				  free_list);
-		list_del_init(&ctxt->free_list);
-		kfree(ctxt);
-	}
-}
-
 static struct svcxprt_rdma *rdma_create_xprt(struct svc_serv *serv,
 					     int listener)
 {
@@ -511,7 +429,6 @@ static struct svcxprt_rdma *rdma_create_xprt(struct svc_serv *serv,
 
 	spin_lock_init(&cma_xprt->sc_lock);
 	spin_lock_init(&cma_xprt->sc_read_complete_lock);
-	spin_lock_init(&cma_xprt->sc_ctxt_lock);
 	spin_lock_init(&cma_xprt->sc_rq_dto_lock);
 
 	cma_xprt->sc_ord = svcrdma_ord;
@@ -522,20 +439,7 @@ static struct svcxprt_rdma *rdma_create_xprt(struct svc_serv *serv,
 	atomic_set(&cma_xprt->sc_sq_count, 0);
 	atomic_set(&cma_xprt->sc_ctxt_used, 0);
 
-	if (!listener) {
-		int reqs = cma_xprt->sc_max_requests;
-		create_context_cache(cma_xprt,
-				     reqs << 1, /* starting size */
-				     reqs,	/* bump amount */
-				     reqs +
-				     cma_xprt->sc_sq_depth +
-				     RPCRDMA_MAX_THREADS + 1); /* max */
-		if (list_empty(&cma_xprt->sc_ctxt_free)) {
-			kfree(cma_xprt);
-			return NULL;
-		}
-		clear_bit(XPT_LISTENER, &cma_xprt->sc_xprt.xpt_flags);
-	} else
+	if (listener)
 		set_bit(XPT_LISTENER, &cma_xprt->sc_xprt.xpt_flags);
 
 	return cma_xprt;
@@ -1077,7 +981,6 @@ static void __svc_rdma_free(struct work_struct *work)
 	/* Destroy the CM ID */
 	rdma_destroy_id(rdma->sc_cm_id);
 
-	destroy_context_cache(rdma);
 	kfree(rdma);
 }
 
-- 
cgit v1.2.3


From 9465efc9e96135a2cec8154c0c766fa59984a298 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Fri, 27 Jun 2008 11:05:24 +0200
Subject: Remove BKL from remote_llseek v2

- Replace remote_llseek with generic_file_llseek_unlocked (to force compilation
failures in all users)
- Change all users to either use generic_file_llseek_unlocked directly or
take the BKL around. I changed the file systems who don't use the BKL
for anything (CIFS, GFS) to call it directly. NCPFS and SMBFS and NFS
take the BKL, but explicitely in their own source now.

I moved them all over in a single patch to avoid unbisectable sections.

Open problem: 32bit kernels can corrupt fpos because its modification
is not atomic, but they can do that anyways because there's other paths who
modify it without BKL.

Do we need a special lock for the pos/f_version = 0 checks?

Trond says the NFS BKL is likely not needed, but keep it for now
until his full audit.

v2: Use generic_file_llseek_unlocked instead of remote_llseek_unlocked
    and factor duplicated code (suggested by hch)

Cc: Trond.Myklebust@netapp.com
Cc: swhiteho@redhat.com
Cc: sfrench@samba.org
Cc: vandrove@vc.cvut.cz

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
---
 fs/cifs/cifsfs.c   |  2 +-
 fs/gfs2/ops_file.c |  4 ++--
 fs/ncpfs/file.c    | 12 +++++++++++-
 fs/nfs/file.c      |  6 +++++-
 fs/read_write.c    | 38 +++++++++++---------------------------
 fs/smbfs/file.c    | 11 ++++++++++-
 include/linux/fs.h |  3 ++-
 7 files changed, 42 insertions(+), 34 deletions(-)

(limited to 'include/linux')

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 427a7c695896..aeff0fe5b6b9 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -581,7 +581,7 @@ static loff_t cifs_llseek(struct file *file, loff_t offset, int origin)
 		if (retval < 0)
 			return (loff_t)retval;
 	}
-	return remote_llseek(file, offset, origin);
+	return generic_file_llseek_unlocked(file, offset, origin);
 }
 
 struct file_system_type cifs_fs_type = {
diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c
index e1b7d525a066..24dd59450088 100644
--- a/fs/gfs2/ops_file.c
+++ b/fs/gfs2/ops_file.c
@@ -62,11 +62,11 @@ static loff_t gfs2_llseek(struct file *file, loff_t offset, int origin)
 		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY,
 					   &i_gh);
 		if (!error) {
-			error = remote_llseek(file, offset, origin);
+			error = generic_file_llseek_unlocked(file, offset, origin);
 			gfs2_glock_dq_uninit(&i_gh);
 		}
 	} else
-		error = remote_llseek(file, offset, origin);
+		error = generic_file_llseek_unlocked(file, offset, origin);
 
 	return error;
 }
diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c
index 2b145de45b39..6a7d901f1936 100644
--- a/fs/ncpfs/file.c
+++ b/fs/ncpfs/file.c
@@ -18,6 +18,7 @@
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
 #include <linux/sched.h>
+#include <linux/smp_lock.h>
 
 #include <linux/ncp_fs.h>
 #include "ncplib_kernel.h"
@@ -281,9 +282,18 @@ static int ncp_release(struct inode *inode, struct file *file) {
 	return 0;
 }
 
+static loff_t ncp_remote_llseek(struct file *file, loff_t offset, int origin)
+{
+	loff_t ret;
+	lock_kernel();
+	ret = generic_file_llseek_unlocked(file, offset, origin);
+	unlock_kernel();
+	return ret;
+}
+
 const struct file_operations ncp_file_operations =
 {
-	.llseek		= remote_llseek,
+	.llseek 	= ncp_remote_llseek,
 	.read		= ncp_file_read,
 	.write		= ncp_file_write,
 	.ioctl		= ncp_ioctl,
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 3536b01164f9..a34eb78989f5 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -170,6 +170,7 @@ force_reval:
 
 static loff_t nfs_file_llseek(struct file *filp, loff_t offset, int origin)
 {
+	loff_t loff;
 	/* origin == SEEK_END => we must revalidate the cached file length */
 	if (origin == SEEK_END) {
 		struct inode *inode = filp->f_mapping->host;
@@ -177,7 +178,10 @@ static loff_t nfs_file_llseek(struct file *filp, loff_t offset, int origin)
 		if (retval < 0)
 			return (loff_t)retval;
 	}
-	return remote_llseek(filp, offset, origin);
+	lock_kernel();	/* BKL needed? */
+	loff = generic_file_llseek_unlocked(filp, offset, origin);
+	unlock_kernel();
+	return loff;
 }
 
 /*
diff --git a/fs/read_write.c b/fs/read_write.c
index f0d1240a5c69..9ba495d5a29b 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -31,12 +31,12 @@ const struct file_operations generic_ro_fops = {
 
 EXPORT_SYMBOL(generic_ro_fops);
 
-loff_t generic_file_llseek(struct file *file, loff_t offset, int origin)
+loff_t
+generic_file_llseek_unlocked(struct file *file, loff_t offset, int origin)
 {
 	loff_t retval;
 	struct inode *inode = file->f_mapping->host;
 
-	mutex_lock(&inode->i_mutex);
 	switch (origin) {
 		case SEEK_END:
 			offset += inode->i_size;
@@ -46,42 +46,26 @@ loff_t generic_file_llseek(struct file *file, loff_t offset, int origin)
 	}
 	retval = -EINVAL;
 	if (offset>=0 && offset<=inode->i_sb->s_maxbytes) {
+		/* Special lock needed here? */
 		if (offset != file->f_pos) {
 			file->f_pos = offset;
 			file->f_version = 0;
 		}
 		retval = offset;
 	}
-	mutex_unlock(&inode->i_mutex);
 	return retval;
 }
+EXPORT_SYMBOL(generic_file_llseek_unlocked);
 
-EXPORT_SYMBOL(generic_file_llseek);
-
-loff_t remote_llseek(struct file *file, loff_t offset, int origin)
+loff_t generic_file_llseek(struct file *file, loff_t offset, int origin)
 {
-	loff_t retval;
-
-	lock_kernel();
-	switch (origin) {
-		case SEEK_END:
-			offset += i_size_read(file->f_path.dentry->d_inode);
-			break;
-		case SEEK_CUR:
-			offset += file->f_pos;
-	}
-	retval = -EINVAL;
-	if (offset>=0 && offset<=file->f_path.dentry->d_inode->i_sb->s_maxbytes) {
-		if (offset != file->f_pos) {
-			file->f_pos = offset;
-			file->f_version = 0;
-		}
-		retval = offset;
-	}
-	unlock_kernel();
-	return retval;
+	loff_t n;
+	mutex_lock(&file->f_dentry->d_inode->i_mutex);
+	n = generic_file_llseek_unlocked(file, offset, origin);
+	mutex_unlock(&file->f_dentry->d_inode->i_mutex);
+	return n;
 }
-EXPORT_SYMBOL(remote_llseek);
+EXPORT_SYMBOL(generic_file_llseek);
 
 loff_t no_llseek(struct file *file, loff_t offset, int origin)
 {
diff --git a/fs/smbfs/file.c b/fs/smbfs/file.c
index efbe29af3d7a..2294783320cb 100644
--- a/fs/smbfs/file.c
+++ b/fs/smbfs/file.c
@@ -422,9 +422,18 @@ smb_file_permission(struct inode *inode, int mask, struct nameidata *nd)
 	return error;
 }
 
+static loff_t smb_remote_llseek(struct file *file, loff_t offset, int origin)
+{
+	loff_t ret;
+	lock_kernel();
+	ret = generic_file_llseek_unlocked(file, offset, origin);
+	unlock_kernel();
+	return ret;
+}
+
 const struct file_operations smb_file_operations =
 {
-	.llseek		= remote_llseek,
+	.llseek 	= smb_remote_llseek,
 	.read		= do_sync_read,
 	.aio_read	= smb_file_aio_read,
 	.write		= do_sync_write,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index f413085f748e..b158e5161bca 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1871,7 +1871,8 @@ extern void
 file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping);
 extern loff_t no_llseek(struct file *file, loff_t offset, int origin);
 extern loff_t generic_file_llseek(struct file *file, loff_t offset, int origin);
-extern loff_t remote_llseek(struct file *file, loff_t offset, int origin);
+extern loff_t generic_file_llseek_unlocked(struct file *file, loff_t offset,
+			int origin);
 extern int generic_file_open(struct inode * inode, struct file * filp);
 extern int nonseekable_open(struct inode * inode, struct file * filp);
 
-- 
cgit v1.2.3


From 778d80be52699596bf70e0eb0761cf5e1e46088d Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 28 Jun 2008 14:17:11 +0900
Subject: ipv6: Add disable_ipv6 sysctl to disable IPv6 operaion on specific
 interface.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 Documentation/networking/ip-sysctl.txt |  4 ++++
 include/linux/ipv6.h                   |  2 ++
 net/ipv6/addrconf.c                    | 11 +++++++++++
 net/ipv6/ip6_input.c                   |  3 ++-
 net/ipv6/ip6_output.c                  |  7 +++++++
 5 files changed, 26 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 71c7bea97160..dae980e8f1b9 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1025,6 +1025,10 @@ max_addresses - INTEGER
 	autoconfigured addresses.
 	Default: 16
 
+disable_ipv6 - BOOLEAN
+	Disable IPv6 operation.
+	Default: FALSE (enable IPv6 operation)
+
 icmp/*:
 ratelimit - INTEGER
 	Limit the maximal rates for sending ICMPv6 packets.
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index cde056e08181..d9d7f9b69eb4 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -163,6 +163,7 @@ struct ipv6_devconf {
 #ifdef CONFIG_IPV6_MROUTE
 	__s32		mc_forwarding;
 #endif
+	__s32		disable_ipv6;
 	void		*sysctl;
 };
 
@@ -194,6 +195,7 @@ enum {
 	DEVCONF_OPTIMISTIC_DAD,
 	DEVCONF_ACCEPT_SOURCE_ROUTE,
 	DEVCONF_MC_FORWARDING,
+	DEVCONF_DISABLE_IPV6,
 	DEVCONF_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8b6875f02039..8c5cff50bbed 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -183,6 +183,7 @@ struct ipv6_devconf ipv6_devconf __read_mostly = {
 #endif
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
+	.disable_ipv6		= 0,
 };
 
 static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
@@ -215,6 +216,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 #endif
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
+	.disable_ipv6		= 0,
 };
 
 /* IPv6 Wildcard Address and Loopback Address defined by RFC2553 */
@@ -3657,6 +3659,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 #ifdef CONFIG_IPV6_MROUTE
 	array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding;
 #endif
+	array[DEVCONF_DISABLE_IPV6] = cnf->disable_ipv6;
 }
 
 static inline size_t inet6_if_nlmsg_size(void)
@@ -4215,6 +4218,14 @@ static struct addrconf_sysctl_table
 			.proc_handler	=	&proc_dointvec,
 		},
 #endif
+		{
+			.ctl_name	=	CTL_UNNUMBERED,
+			.procname	=	"disable_ipv6",
+			.data		=	&ipv6_devconf.disable_ipv6,
+			.maxlen		=	sizeof(int),
+			.mode		=	0644,
+			.proc_handler	=	&proc_dointvec,
+		},
 		{
 			.ctl_name	=	0,	/* sentinel */
 		}
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 34e5a96623ae..ea81c614dde2 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -71,7 +71,8 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 
 	IP6_INC_STATS_BH(idev, IPSTATS_MIB_INRECEIVES);
 
-	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
+	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL ||
+	    !idev || unlikely(idev->cnf.disable_ipv6)) {
 		IP6_INC_STATS_BH(idev, IPSTATS_MIB_INDISCARDS);
 		rcu_read_unlock();
 		goto out;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 871bdec09edb..0981c1ef3057 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -173,6 +173,13 @@ static inline int ip6_skb_dst_mtu(struct sk_buff *skb)
 
 int ip6_output(struct sk_buff *skb)
 {
+	struct inet6_dev *idev = ip6_dst_idev(skb->dst);
+	if (unlikely(idev->cnf.disable_ipv6)) {
+		IP6_INC_STATS(idev, IPSTATS_MIB_OUTDISCARDS);
+		kfree_skb(skb);
+		return 0;
+	}
+
 	if ((skb->len > ip6_skb_dst_mtu(skb) && !skb_is_gso(skb)) ||
 				dst_allfrag(skb->dst))
 		return ip6_fragment(skb, ip6_output2);
-- 
cgit v1.2.3


From 1b34be74cbf18f5d58cc85c7c4afcd9f7d74accd Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 28 Jun 2008 14:18:38 +0900
Subject: ipv6 addrconf: add accept_dad sysctl to control DAD operation.

- If 0, disable DAD.
- If 1, perform DAD (default).
- If >1, perform DAD and disable IPv6 operation if DAD for MAC-based
  link-local address has been failed (RFC4862 5.4.5).

We do not follow RFC4862 by default.  Refer to the netdev thread entitled
"Linux IPv6 DAD not full conform to RFC 4862 ?"
	http://www.spinics.net/lists/netdev/msg52027.html

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 Documentation/networking/ip-sysctl.txt |  7 +++++++
 include/linux/ipv6.h                   |  2 ++
 net/ipv6/addrconf.c                    | 35 ++++++++++++++++++++++++++++++++++
 3 files changed, 44 insertions(+)

(limited to 'include/linux')

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index dae980e8f1b9..72f6d52e52e6 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -1029,6 +1029,13 @@ disable_ipv6 - BOOLEAN
 	Disable IPv6 operation.
 	Default: FALSE (enable IPv6 operation)
 
+accept_dad - INTEGER
+	Whether to accept DAD (Duplicate Address Detection).
+	0: Disable DAD
+	1: Enable DAD (default)
+	2: Enable DAD, and disable IPv6 operation if MAC-based duplicate
+	   link-local address has been found.
+
 icmp/*:
 ratelimit - INTEGER
 	Limit the maximal rates for sending ICMPv6 packets.
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index d9d7f9b69eb4..391ad0843a46 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -164,6 +164,7 @@ struct ipv6_devconf {
 	__s32		mc_forwarding;
 #endif
 	__s32		disable_ipv6;
+	__s32		accept_dad;
 	void		*sysctl;
 };
 
@@ -196,6 +197,7 @@ enum {
 	DEVCONF_ACCEPT_SOURCE_ROUTE,
 	DEVCONF_MC_FORWARDING,
 	DEVCONF_DISABLE_IPV6,
+	DEVCONF_ACCEPT_DAD,
 	DEVCONF_MAX
 };
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8c5cff50bbed..2ec73e62202c 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -119,6 +119,7 @@ static void ipv6_regen_rndid(unsigned long data);
 static int desync_factor = MAX_DESYNC_FACTOR * HZ;
 #endif
 
+static int ipv6_generate_eui64(u8 *eui, struct net_device *dev);
 static int ipv6_count_addresses(struct inet6_dev *idev);
 
 /*
@@ -184,6 +185,7 @@ struct ipv6_devconf ipv6_devconf __read_mostly = {
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
+	.accept_dad		= 1,
 };
 
 static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
@@ -217,6 +219,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
+	.accept_dad		= 1,
 };
 
 /* IPv6 Wildcard Address and Loopback Address defined by RFC2553 */
@@ -380,6 +383,9 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 	 */
 	in6_dev_hold(ndev);
 
+	if (dev->flags & (IFF_NOARP | IFF_LOOPBACK))
+		ndev->cnf.accept_dad = -1;
+
 #if defined(CONFIG_IPV6_SIT) || defined(CONFIG_IPV6_SIT_MODULE)
 	if (dev->type == ARPHRD_SIT && (dev->priv_flags & IFF_ISATAP)) {
 		printk(KERN_INFO
@@ -1421,6 +1427,20 @@ static void addrconf_dad_stop(struct inet6_ifaddr *ifp)
 
 void addrconf_dad_failure(struct inet6_ifaddr *ifp)
 {
+	struct inet6_dev *idev = ifp->idev;
+	if (idev->cnf.accept_dad > 1 && !idev->cnf.disable_ipv6) {
+		struct in6_addr addr;
+
+		addr.s6_addr32[0] = htonl(0xfe800000);
+		addr.s6_addr32[1] = 0;
+
+		if (!ipv6_generate_eui64(addr.s6_addr + 8, idev->dev) &&
+		    ipv6_addr_equal(&ifp->addr, &addr)) {
+			/* DAD failed for link-local based on MAC address */
+			idev->cnf.disable_ipv6 = 1;
+		}
+	}
+
 	if (net_ratelimit())
 		printk(KERN_INFO "%s: duplicate address detected!\n", ifp->idev->dev->name);
 	addrconf_dad_stop(ifp);
@@ -2753,6 +2773,7 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags)
 	spin_lock_bh(&ifp->lock);
 
 	if (dev->flags&(IFF_NOARP|IFF_LOOPBACK) ||
+	    idev->cnf.accept_dad < 1 ||
 	    !(ifp->flags&IFA_F_TENTATIVE) ||
 	    ifp->flags & IFA_F_NODAD) {
 		ifp->flags &= ~(IFA_F_TENTATIVE|IFA_F_OPTIMISTIC);
@@ -2800,6 +2821,11 @@ static void addrconf_dad_timer(unsigned long data)
 		read_unlock_bh(&idev->lock);
 		goto out;
 	}
+	if (idev->cnf.accept_dad > 1 && idev->cnf.disable_ipv6) {
+		read_unlock_bh(&idev->lock);
+		addrconf_dad_failure(ifp);
+		return;
+	}
 	spin_lock_bh(&ifp->lock);
 	if (ifp->probes == 0) {
 		/*
@@ -3660,6 +3686,7 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding;
 #endif
 	array[DEVCONF_DISABLE_IPV6] = cnf->disable_ipv6;
+	array[DEVCONF_ACCEPT_DAD] = cnf->accept_dad;
 }
 
 static inline size_t inet6_if_nlmsg_size(void)
@@ -4226,6 +4253,14 @@ static struct addrconf_sysctl_table
 			.mode		=	0644,
 			.proc_handler	=	&proc_dointvec,
 		},
+		{
+			.ctl_name	=	CTL_UNNUMBERED,
+			.procname	=	"accept_dad",
+			.data		=	&ipv6_devconf.accept_dad,
+			.maxlen		=	sizeof(int),
+			.mode		=	0644,
+			.proc_handler	=	&proc_dointvec,
+		},
 		{
 			.ctl_name	=	0,	/* sentinel */
 		}
-- 
cgit v1.2.3


From 623d1a1af77bd52a389c6eda5920e28eb2ee468b Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Thu, 3 Jul 2008 12:13:30 +0800
Subject: ipv6: Do cleanup for ip6_mr_init.

If do not do it, we will get following issues:
1. Leaving junks after inet6_init failing halfway.
2. Leaving proc and notifier junks after ipv6 modules unloading.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/mroute6.h |  3 ++-
 net/ipv6/af_inet6.c     | 11 ++++++++++-
 net/ipv6/ip6mr.c        | 38 +++++++++++++++++++++++++++++++++-----
 3 files changed, 45 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
index e7989593142b..4c4d6f57d5c5 100644
--- a/include/linux/mroute6.h
+++ b/include/linux/mroute6.h
@@ -135,7 +135,8 @@ extern int ip6_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip6_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ip6_mr_input(struct sk_buff *skb);
 extern int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg);
-extern void ip6_mr_init(void);
+extern int ip6_mr_init(void);
+extern void ip6_mr_cleanup(void);
 
 struct mif_device
 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 3ce8d2f318c6..6b39af1acb5e 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -953,7 +953,9 @@ static int __init inet6_init(void)
 	if (err)
 		goto icmp_fail;
 #ifdef CONFIG_IPV6_MROUTE
-	ip6_mr_init();
+	err = ip6_mr_init();
+	if (err)
+		goto ipmr_fail;
 #endif
 	err = ndisc_init();
 	if (err)
@@ -1057,6 +1059,10 @@ netfilter_fail:
 igmp_fail:
 	ndisc_cleanup();
 ndisc_fail:
+#ifdef CONFIG_IPV6_MROUTE
+	ip6_mr_cleanup();
+ipmr_fail:
+#endif
 	icmpv6_cleanup();
 icmp_fail:
 	unregister_pernet_subsys(&inet6_net_ops);
@@ -1111,6 +1117,9 @@ static void __exit inet6_exit(void)
 	ipv6_netfilter_fini();
 	igmp6_cleanup();
 	ndisc_cleanup();
+#ifdef CONFIG_IPV6_MROUTE
+	ip6_mr_cleanup();
+#endif
 	icmpv6_cleanup();
 	rawv6_exit();
 
diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 90e763073dc5..cfac26d674ed 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -948,23 +948,51 @@ static struct notifier_block ip6_mr_notifier = {
  *	Setup for IP multicast routing
  */
 
-void __init ip6_mr_init(void)
+int __init ip6_mr_init(void)
 {
+	int err;
+
 	mrt_cachep = kmem_cache_create("ip6_mrt_cache",
 				       sizeof(struct mfc6_cache),
 				       0, SLAB_HWCACHE_ALIGN,
 				       NULL);
 	if (!mrt_cachep)
-		panic("cannot allocate ip6_mrt_cache");
+		return -ENOMEM;
 
 	setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0);
-	register_netdevice_notifier(&ip6_mr_notifier);
+	err = register_netdevice_notifier(&ip6_mr_notifier);
+	if (err)
+		goto reg_notif_fail;
+#ifdef CONFIG_PROC_FS
+	err = -ENOMEM;
+	if (!proc_net_fops_create(&init_net, "ip6_mr_vif", 0, &ip6mr_vif_fops))
+		goto proc_vif_fail;
+	if (!proc_net_fops_create(&init_net, "ip6_mr_cache",
+				     0, &ip6mr_mfc_fops))
+		goto proc_cache_fail;
+#endif
+	return 0;
+reg_notif_fail:
+	kmem_cache_destroy(mrt_cachep);
 #ifdef CONFIG_PROC_FS
-	proc_net_fops_create(&init_net, "ip6_mr_vif", 0, &ip6mr_vif_fops);
-	proc_net_fops_create(&init_net, "ip6_mr_cache", 0, &ip6mr_mfc_fops);
+proc_vif_fail:
+	unregister_netdevice_notifier(&ip6_mr_notifier);
+proc_cache_fail:
+	proc_net_remove(&init_net, "ip6_mr_vif");
 #endif
+	return err;
 }
 
+void ip6_mr_cleanup(void)
+{
+#ifdef CONFIG_PROC_FS
+	proc_net_remove(&init_net, "ip6_mr_cache");
+	proc_net_remove(&init_net, "ip6_mr_vif");
+#endif
+	unregister_netdevice_notifier(&ip6_mr_notifier);
+	del_timer(&ipmr_expire_timer);
+	kmem_cache_destroy(mrt_cachep);
+}
 
 static int ip6mr_mfc_add(struct mf6cctl *mfc, int mrtsock)
 {
-- 
cgit v1.2.3


From 03d2f897e9fb3218989baa2139a951ce7f5414bf Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Thu, 3 Jul 2008 12:13:36 +0800
Subject: ipv4: Do cleanup for ip_mr_init

Same as ip6_mr_init(), make ip_mr_init() return errno if fails.
But do not do error handling in inet_init(), just print a msg.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/igmp.h   |  1 -
 include/linux/mroute.h |  3 +--
 net/ipv4/af_inet.c     |  5 +++--
 net/ipv4/ipmr.c        | 28 ++++++++++++++++++++++++----
 4 files changed, 28 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index f5a1a0db2e8e..7bb3c095c15b 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -228,7 +228,6 @@ extern int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf,
 extern int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf,
 		struct group_filter __user *optval, int __user *optlen);
 extern int ip_mc_sf_allow(struct sock *sk, __be32 local, __be32 rmt, int dif);
-extern void ip_mr_init(void);
 extern void ip_mc_init_dev(struct in_device *);
 extern void ip_mc_destroy_dev(struct in_device *);
 extern void ip_mc_up(struct in_device *);
diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index de4decfa1bfc..df8efd42bf8a 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -147,8 +147,7 @@ static inline int ip_mroute_opt(int opt)
 extern int ip_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg);
-extern void ip_mr_init(void);
-
+extern int ip_mr_init(void);
 
 struct vif_device
 {
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 42bd24b64b57..dc411335c14f 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1479,14 +1479,15 @@ static int __init inet_init(void)
 	 *	Initialise the multicast router
 	 */
 #if defined(CONFIG_IP_MROUTE)
-	ip_mr_init();
+	if (ip_mr_init())
+		printk(KERN_CRIT "inet_init: Cannot init ipv4 mroute\n");
 #endif
 	/*
 	 *	Initialise per-cpu ipv4 mibs
 	 */
 
 	if (init_ipv4_mibs())
-		printk(KERN_CRIT "inet_init: Cannot init ipv4 mibs\n"); ;
+		printk(KERN_CRIT "inet_init: Cannot init ipv4 mibs\n");
 
 	ipv4_proc_init();
 
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 300ab0c2919e..438fab9c62a0 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1878,16 +1878,36 @@ static struct net_protocol pim_protocol = {
  *	Setup for IP multicast routing
  */
 
-void __init ip_mr_init(void)
+int __init ip_mr_init(void)
 {
+	int err;
+
 	mrt_cachep = kmem_cache_create("ip_mrt_cache",
 				       sizeof(struct mfc_cache),
 				       0, SLAB_HWCACHE_ALIGN|SLAB_PANIC,
 				       NULL);
+	if (!mrt_cachep)
+		return -ENOMEM;
+
 	setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0);
-	register_netdevice_notifier(&ip_mr_notifier);
+	err = register_netdevice_notifier(&ip_mr_notifier);
+	if (err)
+		goto reg_notif_fail;
 #ifdef CONFIG_PROC_FS
-	proc_net_fops_create(&init_net, "ip_mr_vif", 0, &ipmr_vif_fops);
-	proc_net_fops_create(&init_net, "ip_mr_cache", 0, &ipmr_mfc_fops);
+	err = -ENOMEM;
+	if (!proc_net_fops_create(&init_net, "ip_mr_vif", 0, &ipmr_vif_fops))
+		goto proc_vif_fail;
+	if (!proc_net_fops_create(&init_net, "ip_mr_cache", 0, &ipmr_mfc_fops))
+		goto proc_cache_fail;
 #endif
+	return 0;
+reg_notif_fail:
+	kmem_cache_destroy(mrt_cachep);
+#ifdef CONFIG_PROC_FS
+proc_vif_fail:
+	unregister_netdevice_notifier(&ip_mr_notifier);
+proc_cache_fail:
+	proc_net_remove(&init_net, "ip_mr_vif");
+#endif
+	return err;
 }
-- 
cgit v1.2.3


From e0835f8fa56d2d308486f8a34cf1c4480cd27f4e Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Thu, 3 Jul 2008 16:51:22 +0900
Subject: ipv4,ipv6 mroute: Add some helper inline functions to remove ugly
 ifdefs.

ip{,v6}_mroute_{set,get}sockopt() should not matter by optimization but
it would be better not to depend on optimization semantically.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
---
 include/linux/mroute.h  | 27 +++++++++++++++++++++++++++
 include/linux/mroute6.h | 32 ++++++++++++++++++++++++++++++++
 net/ipv6/af_inet6.c     |  8 --------
 3 files changed, 59 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index df8efd42bf8a..07112ee9293a 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -144,10 +144,37 @@ static inline int ip_mroute_opt(int opt)
 }
 #endif
 
+#ifdef CONFIG_IP_MROUTE
 extern int ip_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg);
 extern int ip_mr_init(void);
+#else
+static inline
+int ip_mroute_setsockopt(struct sock *sock,
+			 int optname, char __user *optval, int optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ip_mroute_getsockopt(struct sock *sock,
+			 int optname, char __user *optval, int __user *optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg)
+{
+	return -ENOIOCTLCMD;
+}
+
+static inline int ip_mr_init(void)
+{
+	return 0;
+}
+#endif
 
 struct vif_device
 {
diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h
index 4c4d6f57d5c5..5cf50473a10f 100644
--- a/include/linux/mroute6.h
+++ b/include/linux/mroute6.h
@@ -131,12 +131,44 @@ static inline int ip6_mroute_opt(int opt)
 
 struct sock;
 
+#ifdef CONFIG_IPV6_MROUTE
 extern int ip6_mroute_setsockopt(struct sock *, int, char __user *, int);
 extern int ip6_mroute_getsockopt(struct sock *, int, char __user *, int __user *);
 extern int ip6_mr_input(struct sk_buff *skb);
 extern int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg);
 extern int ip6_mr_init(void);
 extern void ip6_mr_cleanup(void);
+#else
+static inline
+int ip6_mroute_setsockopt(struct sock *sock,
+			  int optname, char __user *optval, int optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ip6_mroute_getsockopt(struct sock *sock,
+			  int optname, char __user *optval, int __user *optlen)
+{
+	return -ENOPROTOOPT;
+}
+
+static inline
+int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg)
+{
+	return -ENOIOCTLCMD;
+}
+
+static inline int ip6_mr_init(void)
+{
+	return 0;
+}
+
+static inline void ip6_mr_cleanup(void)
+{
+	return;
+}
+#endif
 
 struct mif_device
 {
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 6b39af1acb5e..3d828bc4b1cf 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -59,9 +59,7 @@
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
-#ifdef CONFIG_IPV6_MROUTE
 #include <linux/mroute6.h>
-#endif
 
 MODULE_AUTHOR("Cast of dozens");
 MODULE_DESCRIPTION("IPv6 protocol stack for Linux");
@@ -952,11 +950,9 @@ static int __init inet6_init(void)
 	err = icmpv6_init();
 	if (err)
 		goto icmp_fail;
-#ifdef CONFIG_IPV6_MROUTE
 	err = ip6_mr_init();
 	if (err)
 		goto ipmr_fail;
-#endif
 	err = ndisc_init();
 	if (err)
 		goto ndisc_fail;
@@ -1059,10 +1055,8 @@ netfilter_fail:
 igmp_fail:
 	ndisc_cleanup();
 ndisc_fail:
-#ifdef CONFIG_IPV6_MROUTE
 	ip6_mr_cleanup();
 ipmr_fail:
-#endif
 	icmpv6_cleanup();
 icmp_fail:
 	unregister_pernet_subsys(&inet6_net_ops);
@@ -1117,9 +1111,7 @@ static void __exit inet6_exit(void)
 	ipv6_netfilter_fini();
 	igmp6_cleanup();
 	ndisc_cleanup();
-#ifdef CONFIG_IPV6_MROUTE
 	ip6_mr_cleanup();
-#endif
 	icmpv6_cleanup();
 	rawv6_exit();
 
-- 
cgit v1.2.3


From 07240fd0902c872f044f523893364a1a24c9f278 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 03:45:32 -0700
Subject: tun: Interface to query tun/tap features.

The problem with introducing checksum offload and gso to tun is they
need to set dev->features to enable GSO and/or checksumming, which is
supposed to be done before register_netdevice(), ie. as part of
TUNSETIFF.

Unfortunately, TUNSETIFF has always just ignored flags it doesn't
understand, so there's no good way of detecting whether the kernel
supports new IFF_ flags.

This patch implements a TUNGETFEATURES ioctl which returns all the valid IFF
flags.  It could be extended later to include other features.

Here's an example program which uses it:

#include <linux/if_tun.h>
#include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <err.h>
#include <stdio.h>

static struct {
	unsigned int flag;
	const char *name;
} known_flags[] = {
	{ IFF_TUN, "TUN" },
	{ IFF_TAP, "TAP" },
	{ IFF_NO_PI, "NO_PI" },
	{ IFF_ONE_QUEUE, "ONE_QUEUE" },
};

int main()
{
	unsigned int features, i;

	int netfd = open("/dev/net/tun", O_RDWR);
	if (netfd < 0)
		err(1, "Opening /dev/net/tun");

	if (ioctl(netfd, TUNGETFEATURES, &features) != 0) {
		printf("Kernel does not support TUNGETFEATURES, guessing\n");
		features = (IFF_TUN|IFF_TAP|IFF_NO_PI|IFF_ONE_QUEUE);
	}
	printf("Available features are: ");
	for (i = 0; i < sizeof(known_flags)/sizeof(known_flags[0]); i++) {
		if (features & known_flags[i].flag) {
			features &= ~known_flags[i].flag;
			printf("%s ", known_flags[i].name);
		}
	}
	if (features)
		printf("(UNKNOWN %#x)", features);
	printf("\n");
	return 0;
}

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 8 ++++++++
 include/linux/if_tun.h | 1 +
 2 files changed, 9 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 7ab94c825b57..3bb991fd2b51 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -640,6 +640,14 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		return 0;
 	}
 
+	if (cmd == TUNGETFEATURES) {
+		/* Currently this just means: "what IFF flags are valid?".
+		 * This is needed because we never checked for invalid flags on
+		 * TUNSETIFF. */
+		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE,
+				(unsigned int __user*)argp);
+	}
+
 	if (!tun)
 		return -EBADFD;
 
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 18f31b6187a3..94f76a112303 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -40,6 +40,7 @@
 #define TUNSETOWNER   _IOW('T', 204, int)
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
+#define TUNGETFEATURES _IOR('T', 207, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001
-- 
cgit v1.2.3


From 5228ddc98fa49b3cedab4024e269d62410a0d806 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 03:46:16 -0700
Subject: tun: TUNSETFEATURES to set gso features.

ethtool is useful for setting (some) device fields, but it's
root-only.  Finer feature control is available through a tun-specific
ioctl.

(Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/if_tun.h |  7 +++++++
 2 files changed, 56 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 3bb991fd2b51..a314955e6994 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -611,6 +611,46 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 	return err;
 }
 
+/* This is like a cut-down ethtool ops, except done via tun fd so no
+ * privs required. */
+static int set_offload(struct net_device *dev, unsigned long arg)
+{
+	unsigned int old_features, features;
+
+	old_features = dev->features;
+	/* Unset features, set them as we chew on the arg. */
+	features = (old_features & ~(NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST
+				    |NETIF_F_TSO_ECN|NETIF_F_TSO|NETIF_F_TSO6));
+
+	if (arg & TUN_F_CSUM) {
+		features |= NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST;
+		arg &= ~TUN_F_CSUM;
+
+		if (arg & (TUN_F_TSO4|TUN_F_TSO6)) {
+			if (arg & TUN_F_TSO_ECN) {
+				features |= NETIF_F_TSO_ECN;
+				arg &= ~TUN_F_TSO_ECN;
+			}
+			if (arg & TUN_F_TSO4)
+				features |= NETIF_F_TSO;
+			if (arg & TUN_F_TSO6)
+				features |= NETIF_F_TSO6;
+			arg &= ~(TUN_F_TSO4|TUN_F_TSO6);
+		}
+	}
+
+	/* This gives the user a way to test for new features in future by
+	 * trying to set them. */
+	if (arg)
+		return -EINVAL;
+
+	dev->features = features;
+	if (old_features != dev->features)
+		netdev_features_change(dev);
+
+	return 0;
+}
+
 static int tun_chr_ioctl(struct inode *inode, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
@@ -715,6 +755,15 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		break;
 #endif
 
+	case TUNSETOFFLOAD:
+	{
+		int ret;
+		rtnl_lock();
+		ret = set_offload(tun->dev, arg);
+		rtnl_unlock();
+		return ret;
+	}
+
 	case SIOCGIFFLAGS:
 		ifr.ifr_flags = tun->if_flags;
 		if (copy_to_user( argp, &ifr, sizeof ifr))
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 94f76a112303..3f0a0995d449 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -41,6 +41,7 @@
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
 #define TUNGETFEATURES _IOR('T', 207, unsigned int)
+#define TUNSETOFFLOAD _IOW('T', 208, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001
@@ -48,6 +49,12 @@
 #define IFF_NO_PI	0x1000
 #define IFF_ONE_QUEUE	0x2000
 
+/* Features for GSO (TUNSETOFFLOAD). */
+#define TUN_F_CSUM	0x01	/* You can hand me unchecksummed packets. */
+#define TUN_F_TSO4	0x02	/* I can handle TSO for IPv4 packets */
+#define TUN_F_TSO6	0x04	/* I can handle TSO for IPv6 packets */
+#define TUN_F_TSO_ECN	0x08	/* I can handle TSO with ECN bits. */
+
 struct tun_pi {
 	unsigned short flags;
 	__be16 proto;
-- 
cgit v1.2.3


From f43798c27684ab925adde7d8acc34c78c6e50df8 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 03:48:02 -0700
Subject: tun: Allow GSO using virtio_net_hdr

Add a IFF_VNET_HDR flag.  This uses the same ABI as virtio_net
(ie. prepending struct virtio_net_hdr to packets) to indicate GSO and
checksum information.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 95 ++++++++++++++++++++++++++++++++++++++++++++++++--
 include/linux/if_tun.h |  2 ++
 2 files changed, 94 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index a314955e6994..aa4ee4439f04 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -63,6 +63,7 @@
 #include <linux/if_tun.h>
 #include <linux/crc32.h>
 #include <linux/nsproxy.h>
+#include <linux/virtio_net.h>
 #include <net/net_namespace.h>
 #include <net/netns/generic.h>
 
@@ -283,6 +284,7 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 	struct tun_pi pi = { 0, __constant_htons(ETH_P_IP) };
 	struct sk_buff *skb;
 	size_t len = count, align = 0;
+	struct virtio_net_hdr gso = { 0 };
 
 	if (!(tun->flags & TUN_NO_PI)) {
 		if ((len -= sizeof(pi)) > count)
@@ -292,6 +294,17 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 			return -EFAULT;
 	}
 
+	if (tun->flags & TUN_VNET_HDR) {
+		if ((len -= sizeof(gso)) > count)
+			return -EINVAL;
+
+		if (memcpy_fromiovec((void *)&gso, iv, sizeof(gso)))
+			return -EFAULT;
+
+		if (gso.hdr_len > len)
+			return -EINVAL;
+	}
+
 	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
 		align = NET_IP_ALIGN;
 		if (unlikely(len < ETH_HLEN))
@@ -311,6 +324,16 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 		return -EFAULT;
 	}
 
+	if (gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
+		if (!skb_partial_csum_set(skb, gso.csum_start,
+					  gso.csum_offset)) {
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+	} else if (tun->flags & TUN_NOCHECKSUM)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
 	switch (tun->flags & TUN_TYPE_MASK) {
 	case TUN_TUN_DEV:
 		if (tun->flags & TUN_NO_PI) {
@@ -337,8 +360,35 @@ static __inline__ ssize_t tun_get_user(struct tun_struct *tun, struct iovec *iv,
 		break;
 	};
 
-	if (tun->flags & TUN_NOCHECKSUM)
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	if (gso.gso_type != VIRTIO_NET_HDR_GSO_NONE) {
+		pr_debug("GSO!\n");
+		switch (gso.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
+		case VIRTIO_NET_HDR_GSO_TCPV4:
+			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+			break;
+		case VIRTIO_NET_HDR_GSO_TCPV6:
+			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
+			break;
+		default:
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+
+		if (gso.gso_type & VIRTIO_NET_HDR_GSO_ECN)
+			skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;
+
+		skb_shinfo(skb)->gso_size = gso.gso_size;
+		if (skb_shinfo(skb)->gso_size == 0) {
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+
+		/* Header must be checked, and gso_segs computed. */
+		skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
+		skb_shinfo(skb)->gso_segs = 0;
+	}
 
 	netif_rx_ni(skb);
 	tun->dev->last_rx = jiffies;
@@ -384,6 +434,39 @@ static __inline__ ssize_t tun_put_user(struct tun_struct *tun,
 		total += sizeof(pi);
 	}
 
+	if (tun->flags & TUN_VNET_HDR) {
+		struct virtio_net_hdr gso = { 0 }; /* no info leak */
+		if ((len -= sizeof(gso)) < 0)
+			return -EINVAL;
+
+		if (skb_is_gso(skb)) {
+			struct skb_shared_info *sinfo = skb_shinfo(skb);
+
+			/* This is a hint as to how much should be linear. */
+			gso.hdr_len = skb_headlen(skb);
+			gso.gso_size = sinfo->gso_size;
+			if (sinfo->gso_type & SKB_GSO_TCPV4)
+				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
+			else if (sinfo->gso_type & SKB_GSO_TCPV6)
+				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
+			else
+				BUG();
+			if (sinfo->gso_type & SKB_GSO_TCP_ECN)
+				gso.gso_type |= VIRTIO_NET_HDR_GSO_ECN;
+		} else
+			gso.gso_type = VIRTIO_NET_HDR_GSO_NONE;
+
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
+			gso.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
+			gso.csum_start = skb->csum_start - skb_headroom(skb);
+			gso.csum_offset = skb->csum_offset;
+		} /* else everything is zero */
+
+		if (unlikely(memcpy_toiovec(iv, (void *)&gso, sizeof(gso))))
+			return -EFAULT;
+		total += sizeof(gso);
+	}
+
 	len = min_t(int, skb->len, len);
 
 	skb_copy_datagram_iovec(skb, 0, iv, len);
@@ -598,6 +681,11 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 	else
 		tun->flags &= ~TUN_ONE_QUEUE;
 
+	if (ifr->ifr_flags & IFF_VNET_HDR)
+		tun->flags |= TUN_VNET_HDR;
+	else
+		tun->flags &= ~TUN_VNET_HDR;
+
 	file->private_data = tun;
 	tun->attached = 1;
 	get_net(dev_net(tun->dev));
@@ -684,7 +772,8 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		/* Currently this just means: "what IFF flags are valid?".
 		 * This is needed because we never checked for invalid flags on
 		 * TUNSETIFF. */
-		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE,
+		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE |
+				IFF_VNET_HDR,
 				(unsigned int __user*)argp);
 	}
 
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 3f0a0995d449..563fae542da6 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -31,6 +31,7 @@
 #define TUN_NO_PI	0x0040
 #define TUN_ONE_QUEUE	0x0080
 #define TUN_PERSIST 	0x0100	
+#define TUN_VNET_HDR 	0x0200
 
 /* Ioctl defines */
 #define TUNSETNOCSUM  _IOW('T', 200, int) 
@@ -48,6 +49,7 @@
 #define IFF_TAP		0x0002
 #define IFF_NO_PI	0x1000
 #define IFF_ONE_QUEUE	0x2000
+#define IFF_VNET_HDR	0x4000
 
 /* Features for GSO (TUNSETOFFLOAD). */
 #define TUN_F_CSUM	0x01	/* You can hand me unchecksummed packets. */
-- 
cgit v1.2.3


From 02c62304e6af60f1963695c6bc1bbffe619aa585 Mon Sep 17 00:00:00 2001
From: "Alan D. Brunelle" <Alan.Brunelle@hp.com>
Date: Wed, 11 Jun 2008 09:12:52 +0200
Subject: Added in user-injected messages into blk traces

This allows a user to annotate the blk trace stream: writing a suitable
message to {/sys/kernel/debug}/block/<dsf>/msg will have it propagated
into the trace stream.

Signed-off-by: Alan D. Brunelle <alan.brunelle@hp.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blktrace.c             | 45 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/blktrace_api.h |  1 +
 2 files changed, 46 insertions(+)

(limited to 'include/linux')

diff --git a/block/blktrace.c b/block/blktrace.c
index 8d3a27780260..eb9651ccb241 100644
--- a/block/blktrace.c
+++ b/block/blktrace.c
@@ -244,6 +244,7 @@ err:
 static void blk_trace_cleanup(struct blk_trace *bt)
 {
 	relay_close(bt->rchan);
+	debugfs_remove(bt->msg_file);
 	debugfs_remove(bt->dropped_file);
 	blk_remove_tree(bt->dir);
 	free_percpu(bt->sequence);
@@ -291,6 +292,44 @@ static const struct file_operations blk_dropped_fops = {
 	.read =		blk_dropped_read,
 };
 
+static int blk_msg_open(struct inode *inode, struct file *filp)
+{
+	filp->private_data = inode->i_private;
+
+	return 0;
+}
+
+static ssize_t blk_msg_write(struct file *filp, const char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	char *msg;
+	struct blk_trace *bt;
+
+	if (count > BLK_TN_MAX_MSG)
+		return -EINVAL;
+
+	msg = kmalloc(count, GFP_KERNEL);
+	if (msg == NULL)
+		return -ENOMEM;
+
+	if (copy_from_user(msg, buffer, count)) {
+		kfree(msg);
+		return -EFAULT;
+	}
+
+	bt = filp->private_data;
+	__trace_note_message(bt, "%s", msg);
+	kfree(msg);
+
+	return count;
+}
+
+static const struct file_operations blk_msg_fops = {
+	.owner =	THIS_MODULE,
+	.open =		blk_msg_open,
+	.write =	blk_msg_write,
+};
+
 /*
  * Keep track of how many times we encountered a full subbuffer, to aid
  * the user space app in telling how many lost events there were.
@@ -380,6 +419,10 @@ int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev,
 	if (!bt->dropped_file)
 		goto err;
 
+	bt->msg_file = debugfs_create_file("msg", 0222, dir, bt, &blk_msg_fops);
+	if (!bt->msg_file)
+		goto err;
+
 	bt->rchan = relay_open("trace", dir, buts->buf_size,
 				buts->buf_nr, &blk_relay_callbacks, bt);
 	if (!bt->rchan)
@@ -409,6 +452,8 @@ err:
 	if (dir)
 		blk_remove_tree(dir);
 	if (bt) {
+		if (bt->msg_file)
+			debugfs_remove(bt->msg_file);
 		if (bt->dropped_file)
 			debugfs_remove(bt->dropped_file);
 		free_percpu(bt->sequence);
diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
index e3ef903aae88..d084b8d227a5 100644
--- a/include/linux/blktrace_api.h
+++ b/include/linux/blktrace_api.h
@@ -129,6 +129,7 @@ struct blk_trace {
 	u32 dev;
 	struct dentry *dir;
 	struct dentry *dropped_file;
+	struct dentry *msg_file;
 	atomic_t dropped;
 };
 
-- 
cgit v1.2.3


From 244b4d56f85bcd11b21ab0b94845a3dabeed5c10 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 12 Jun 2008 20:12:36 +0200
Subject: block: kill request_queue_t

Everything was moved to struct request_queue a few kernel revisions
ago, maintaining the deprecated typedef to avoid breaking things.
Now the time has come to get rid of that typedef.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/blkdev.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d2a1b71e93c3..6a3da6717135 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -23,7 +23,6 @@
 struct scsi_ioctl_command;
 
 struct request_queue;
-typedef struct request_queue request_queue_t __deprecated;
 struct elevator_queue;
 typedef struct elevator_queue elevator_t;
 struct request_pm_state;
-- 
cgit v1.2.3


From 51d654e1d885607a6edd02b337105fa5c28b6d33 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Tue, 17 Jun 2008 18:59:56 +0200
Subject: block: Globalize bio_set and bio_vec_slab

Move struct bio_set and biovec_slab definitions to bio.h so they can
be used outside of bio.c.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 fs/bio.c            | 30 ++----------------------------
 include/linux/bio.h | 29 +++++++++++++++++++++++++++++
 2 files changed, 31 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/fs/bio.c b/fs/bio.c
index 78562574cb52..7a6598abc967 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -28,25 +28,10 @@
 #include <linux/blktrace_api.h>
 #include <scsi/sg.h>		/* for struct sg_iovec */
 
-#define BIO_POOL_SIZE 2
-
 static struct kmem_cache *bio_slab __read_mostly;
 
-#define BIOVEC_NR_POOLS 6
-
-/*
- * a small number of entries is fine, not going to be performance critical.
- * basically we just need to survive
- */
-#define BIO_SPLIT_ENTRIES 2
 mempool_t *bio_split_pool __read_mostly;
 
-struct biovec_slab {
-	int nr_vecs;
-	char *name; 
-	struct kmem_cache *slab;
-};
-
 /*
  * if you change this list, also change bvec_alloc or things will
  * break badly! cannot be bigger than what you can fit into an
@@ -59,24 +44,13 @@ static struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly = {
 };
 #undef BV
 
-/*
- * bio_set is used to allow other portions of the IO system to
- * allocate their own private memory pools for bio and iovec structures.
- * These memory pools in turn all allocate from the bio_slab
- * and the bvec_slabs[].
- */
-struct bio_set {
-	mempool_t *bio_pool;
-	mempool_t *bvec_pools[BIOVEC_NR_POOLS];
-};
-
 /*
  * fs_bio_set is the bio_set containing bio and iovec memory pools used by
  * IO code that does not need private memory pools.
  */
-static struct bio_set *fs_bio_set;
+struct bio_set *fs_bio_set;
 
-static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs)
+struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs)
 {
 	struct bio_vec *bvl;
 
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 61c15eaf3fb3..49dfb3cb7460 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -333,6 +333,35 @@ extern struct bio *bio_copy_user_iov(struct request_queue *, struct sg_iovec *,
 				     int, int);
 extern int bio_uncopy_user(struct bio *);
 void zero_fill_bio(struct bio *bio);
+extern struct bio_vec *bvec_alloc_bs(gfp_t, int, unsigned long *, struct bio_set *);
+
+/*
+ * bio_set is used to allow other portions of the IO system to
+ * allocate their own private memory pools for bio and iovec structures.
+ * These memory pools in turn all allocate from the bio_slab
+ * and the bvec_slabs[].
+ */
+#define BIO_POOL_SIZE 2
+#define BIOVEC_NR_POOLS 6
+
+struct bio_set {
+	mempool_t *bio_pool;
+	mempool_t *bvec_pools[BIOVEC_NR_POOLS];
+};
+
+struct biovec_slab {
+	int nr_vecs;
+	char *name;
+	struct kmem_cache *slab;
+};
+
+extern struct bio_set *fs_bio_set;
+
+/*
+ * a small number of entries is fine, not going to be performance critical.
+ * basically we just need to survive
+ */
+#define BIO_SPLIT_ENTRIES 2
 
 #ifdef CONFIG_HIGHMEM
 /*
-- 
cgit v1.2.3


From 7ba1ba12eeef0aa7113beb16410ef8b7c748e18b Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Mon, 30 Jun 2008 20:04:41 +0200
Subject: block: Block layer data integrity support

Some block devices support verifying the integrity of requests by way
of checksums or other protection information that is submitted along
with the I/O.

This patch implements support for generating and verifying integrity
metadata, as well as correctly merging, splitting and cloning bios and
requests that have this extra information attached.

See Documentation/block/data-integrity.txt for more information.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/Kconfig          |  12 +
 block/Makefile         |   1 +
 block/blk-core.c       |   7 +
 block/blk-integrity.c  | 382 ++++++++++++++++++++++++++
 block/blk-merge.c      |   3 +
 block/blk.h            |   8 +
 block/elevator.c       |   6 +
 fs/Makefile            |   1 +
 fs/bio-integrity.c     | 708 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/bio.c               |  32 ++-
 include/linux/bio.h    |  94 ++++++-
 include/linux/blkdev.h | 105 ++++++++
 include/linux/genhd.h  |   3 +
 13 files changed, 1355 insertions(+), 7 deletions(-)
 create mode 100644 block/blk-integrity.c
 create mode 100644 fs/bio-integrity.c

(limited to 'include/linux')

diff --git a/block/Kconfig b/block/Kconfig
index 3e97f2bc446f..1ab7c15c8d7a 100644
--- a/block/Kconfig
+++ b/block/Kconfig
@@ -81,6 +81,18 @@ config BLK_DEV_BSG
 
 	  If unsure, say N.
 
+config BLK_DEV_INTEGRITY
+	bool "Block layer data integrity support"
+	---help---
+	Some storage devices allow extra information to be
+	stored/retrieved to help protect the data.  The block layer
+	data integrity option provides hooks which can be used by
+	filesystems to ensure better data integrity.
+
+	Say yes here if you have a storage device that provides the
+	T10/SCSI Data Integrity Field or the T13/ATA External Path
+	Protection.  If in doubt, say N.
+
 endif # BLOCK
 
 config BLOCK_COMPAT
diff --git a/block/Makefile b/block/Makefile
index 5a43c7d79594..045f7b62e4bb 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -14,3 +14,4 @@ obj-$(CONFIG_IOSCHED_CFQ)	+= cfq-iosched.o
 
 obj-$(CONFIG_BLK_DEV_IO_TRACE)	+= blktrace.o
 obj-$(CONFIG_BLOCK_COMPAT)	+= compat_ioctl.o
+obj-$(CONFIG_BLK_DEV_INTEGRITY)	+= blk-integrity.o
diff --git a/block/blk-core.c b/block/blk-core.c
index 1905aaba49fb..e0fb0bcc0c17 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -143,6 +143,10 @@ static void req_bio_endio(struct request *rq, struct bio *bio,
 
 		bio->bi_size -= nbytes;
 		bio->bi_sector += (nbytes >> 9);
+
+		if (bio_integrity(bio))
+			bio_integrity_advance(bio, nbytes);
+
 		if (bio->bi_size == 0)
 			bio_endio(bio, error);
 	} else {
@@ -1381,6 +1385,9 @@ end_io:
 		 */
 		blk_partition_remap(bio);
 
+		if (bio_integrity_enabled(bio) && bio_integrity_prep(bio))
+			goto end_io;
+
 		if (old_sector != -1)
 			blk_add_trace_remap(q, bio, old_dev, bio->bi_sector,
 					    old_sector);
diff --git a/block/blk-integrity.c b/block/blk-integrity.c
new file mode 100644
index 000000000000..65f23ef38bbe
--- /dev/null
+++ b/block/blk-integrity.c
@@ -0,0 +1,382 @@
+/*
+ * blk-integrity.c - Block layer data integrity extensions
+ *
+ * Copyright (C) 2007, 2008 Oracle Corporation
+ * Written by: Martin K. Petersen <martin.petersen@oracle.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; see the file COPYING.  If not, write to
+ * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139,
+ * USA.
+ *
+ */
+
+#include <linux/blkdev.h>
+#include <linux/mempool.h>
+#include <linux/bio.h>
+#include <linux/scatterlist.h>
+
+#include "blk.h"
+
+static struct kmem_cache *integrity_cachep;
+
+/**
+ * blk_rq_count_integrity_sg - Count number of integrity scatterlist elements
+ * @rq:		request with integrity metadata attached
+ *
+ * Description: Returns the number of elements required in a
+ * scatterlist corresponding to the integrity metadata in a request.
+ */
+int blk_rq_count_integrity_sg(struct request *rq)
+{
+	struct bio_vec *iv, *ivprv;
+	struct req_iterator iter;
+	unsigned int segments;
+
+	ivprv = NULL;
+	segments = 0;
+
+	rq_for_each_integrity_segment(iv, rq, iter) {
+
+		if (!ivprv || !BIOVEC_PHYS_MERGEABLE(ivprv, iv))
+			segments++;
+
+		ivprv = iv;
+	}
+
+	return segments;
+}
+EXPORT_SYMBOL(blk_rq_count_integrity_sg);
+
+/**
+ * blk_rq_map_integrity_sg - Map integrity metadata into a scatterlist
+ * @rq:		request with integrity metadata attached
+ * @sglist:	target scatterlist
+ *
+ * Description: Map the integrity vectors in request into a
+ * scatterlist.  The scatterlist must be big enough to hold all
+ * elements.  I.e. sized using blk_rq_count_integrity_sg().
+ */
+int blk_rq_map_integrity_sg(struct request *rq, struct scatterlist *sglist)
+{
+	struct bio_vec *iv, *ivprv;
+	struct req_iterator iter;
+	struct scatterlist *sg;
+	unsigned int segments;
+
+	ivprv = NULL;
+	sg = NULL;
+	segments = 0;
+
+	rq_for_each_integrity_segment(iv, rq, iter) {
+
+		if (ivprv) {
+			if (!BIOVEC_PHYS_MERGEABLE(ivprv, iv))
+				goto new_segment;
+
+			sg->length += iv->bv_len;
+		} else {
+new_segment:
+			if (!sg)
+				sg = sglist;
+			else {
+				sg->page_link &= ~0x02;
+				sg = sg_next(sg);
+			}
+
+			sg_set_page(sg, iv->bv_page, iv->bv_len, iv->bv_offset);
+			segments++;
+		}
+
+		ivprv = iv;
+	}
+
+	if (sg)
+		sg_mark_end(sg);
+
+	return segments;
+}
+EXPORT_SYMBOL(blk_rq_map_integrity_sg);
+
+/**
+ * blk_integrity_compare - Compare integrity profile of two block devices
+ * @b1:		Device to compare
+ * @b2:		Device to compare
+ *
+ * Description: Meta-devices like DM and MD need to verify that all
+ * sub-devices use the same integrity format before advertising to
+ * upper layers that they can send/receive integrity metadata.  This
+ * function can be used to check whether two block devices have
+ * compatible integrity formats.
+ */
+int blk_integrity_compare(struct block_device *bd1, struct block_device *bd2)
+{
+	struct blk_integrity *b1 = bd1->bd_disk->integrity;
+	struct blk_integrity *b2 = bd2->bd_disk->integrity;
+
+	BUG_ON(bd1->bd_disk == NULL);
+	BUG_ON(bd2->bd_disk == NULL);
+
+	if (!b1 || !b2)
+		return 0;
+
+	if (b1->sector_size != b2->sector_size) {
+		printk(KERN_ERR "%s: %s/%s sector sz %u != %u\n", __func__,
+		       bd1->bd_disk->disk_name, bd2->bd_disk->disk_name,
+		       b1->sector_size, b2->sector_size);
+		return -1;
+	}
+
+	if (b1->tuple_size != b2->tuple_size) {
+		printk(KERN_ERR "%s: %s/%s tuple sz %u != %u\n", __func__,
+		       bd1->bd_disk->disk_name, bd2->bd_disk->disk_name,
+		       b1->tuple_size, b2->tuple_size);
+		return -1;
+	}
+
+	if (b1->tag_size && b2->tag_size && (b1->tag_size != b2->tag_size)) {
+		printk(KERN_ERR "%s: %s/%s tag sz %u != %u\n", __func__,
+		       bd1->bd_disk->disk_name, bd2->bd_disk->disk_name,
+		       b1->tag_size, b2->tag_size);
+		return -1;
+	}
+
+	if (strcmp(b1->name, b2->name)) {
+		printk(KERN_ERR "%s: %s/%s type %s != %s\n", __func__,
+		       bd1->bd_disk->disk_name, bd2->bd_disk->disk_name,
+		       b1->name, b2->name);
+		return -1;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(blk_integrity_compare);
+
+struct integrity_sysfs_entry {
+	struct attribute attr;
+	ssize_t (*show)(struct blk_integrity *, char *);
+	ssize_t (*store)(struct blk_integrity *, const char *, size_t);
+};
+
+static ssize_t integrity_attr_show(struct kobject *kobj, struct attribute *attr,
+				   char *page)
+{
+	struct blk_integrity *bi =
+		container_of(kobj, struct blk_integrity, kobj);
+	struct integrity_sysfs_entry *entry =
+		container_of(attr, struct integrity_sysfs_entry, attr);
+
+	return entry->show(bi, page);
+}
+
+static ssize_t integrity_attr_store(struct kobject *kobj, struct attribute *attr,
+				    const char *page, size_t count)
+{
+	struct blk_integrity *bi =
+		container_of(kobj, struct blk_integrity, kobj);
+	struct integrity_sysfs_entry *entry =
+		container_of(attr, struct integrity_sysfs_entry, attr);
+	ssize_t ret = 0;
+
+	if (entry->store)
+		ret = entry->store(bi, page, count);
+
+	return ret;
+}
+
+static ssize_t integrity_format_show(struct blk_integrity *bi, char *page)
+{
+	if (bi != NULL && bi->name != NULL)
+		return sprintf(page, "%s\n", bi->name);
+	else
+		return sprintf(page, "none\n");
+}
+
+static ssize_t integrity_tag_size_show(struct blk_integrity *bi, char *page)
+{
+	if (bi != NULL)
+		return sprintf(page, "%u\n", bi->tag_size);
+	else
+		return sprintf(page, "0\n");
+}
+
+static ssize_t integrity_read_store(struct blk_integrity *bi,
+				    const char *page, size_t count)
+{
+	char *p = (char *) page;
+	unsigned long val = simple_strtoul(p, &p, 10);
+
+	if (val)
+		set_bit(INTEGRITY_FLAG_READ, &bi->flags);
+	else
+		clear_bit(INTEGRITY_FLAG_READ, &bi->flags);
+
+	return count;
+}
+
+static ssize_t integrity_read_show(struct blk_integrity *bi, char *page)
+{
+	return sprintf(page, "%d\n",
+		       test_bit(INTEGRITY_FLAG_READ, &bi->flags) ? 1 : 0);
+}
+
+static ssize_t integrity_write_store(struct blk_integrity *bi,
+				     const char *page, size_t count)
+{
+	char *p = (char *) page;
+	unsigned long val = simple_strtoul(p, &p, 10);
+
+	if (val)
+		set_bit(INTEGRITY_FLAG_WRITE, &bi->flags);
+	else
+		clear_bit(INTEGRITY_FLAG_WRITE, &bi->flags);
+
+	return count;
+}
+
+static ssize_t integrity_write_show(struct blk_integrity *bi, char *page)
+{
+	return sprintf(page, "%d\n",
+		       test_bit(INTEGRITY_FLAG_WRITE, &bi->flags) ? 1 : 0);
+}
+
+static struct integrity_sysfs_entry integrity_format_entry = {
+	.attr = { .name = "format", .mode = S_IRUGO },
+	.show = integrity_format_show,
+};
+
+static struct integrity_sysfs_entry integrity_tag_size_entry = {
+	.attr = { .name = "tag_size", .mode = S_IRUGO },
+	.show = integrity_tag_size_show,
+};
+
+static struct integrity_sysfs_entry integrity_read_entry = {
+	.attr = { .name = "read_verify", .mode = S_IRUGO | S_IWUSR },
+	.show = integrity_read_show,
+	.store = integrity_read_store,
+};
+
+static struct integrity_sysfs_entry integrity_write_entry = {
+	.attr = { .name = "write_generate", .mode = S_IRUGO | S_IWUSR },
+	.show = integrity_write_show,
+	.store = integrity_write_store,
+};
+
+static struct attribute *integrity_attrs[] = {
+	&integrity_format_entry.attr,
+	&integrity_tag_size_entry.attr,
+	&integrity_read_entry.attr,
+	&integrity_write_entry.attr,
+	NULL,
+};
+
+static struct sysfs_ops integrity_ops = {
+	.show	= &integrity_attr_show,
+	.store	= &integrity_attr_store,
+};
+
+static int __init blk_dev_integrity_init(void)
+{
+	integrity_cachep = kmem_cache_create("blkdev_integrity",
+					     sizeof(struct blk_integrity),
+					     0, SLAB_PANIC, NULL);
+	return 0;
+}
+subsys_initcall(blk_dev_integrity_init);
+
+static void blk_integrity_release(struct kobject *kobj)
+{
+	struct blk_integrity *bi =
+		container_of(kobj, struct blk_integrity, kobj);
+
+	kmem_cache_free(integrity_cachep, bi);
+}
+
+static struct kobj_type integrity_ktype = {
+	.default_attrs	= integrity_attrs,
+	.sysfs_ops	= &integrity_ops,
+	.release	= blk_integrity_release,
+};
+
+/**
+ * blk_integrity_register - Register a gendisk as being integrity-capable
+ * @disk:	struct gendisk pointer to make integrity-aware
+ * @template:	integrity profile
+ *
+ * Description: When a device needs to advertise itself as being able
+ * to send/receive integrity metadata it must use this function to
+ * register the capability with the block layer.  The template is a
+ * blk_integrity struct with values appropriate for the underlying
+ * hardware.  See Documentation/block/data-integrity.txt.
+ */
+int blk_integrity_register(struct gendisk *disk, struct blk_integrity *template)
+{
+	struct blk_integrity *bi;
+
+	BUG_ON(disk == NULL);
+	BUG_ON(template == NULL);
+
+	if (disk->integrity == NULL) {
+		bi = kmem_cache_alloc(integrity_cachep, GFP_KERNEL | __GFP_ZERO);
+		if (!bi)
+			return -1;
+
+		if (kobject_init_and_add(&bi->kobj, &integrity_ktype,
+					 &disk->dev.kobj, "%s", "integrity")) {
+			kmem_cache_free(integrity_cachep, bi);
+			return -1;
+		}
+
+		kobject_uevent(&bi->kobj, KOBJ_ADD);
+
+		set_bit(INTEGRITY_FLAG_READ, &bi->flags);
+		set_bit(INTEGRITY_FLAG_WRITE, &bi->flags);
+		bi->sector_size = disk->queue->hardsect_size;
+		disk->integrity = bi;
+	} else
+		bi = disk->integrity;
+
+	/* Use the provided profile as template */
+	bi->name = template->name;
+	bi->generate_fn = template->generate_fn;
+	bi->verify_fn = template->verify_fn;
+	bi->tuple_size = template->tuple_size;
+	bi->set_tag_fn = template->set_tag_fn;
+	bi->get_tag_fn = template->get_tag_fn;
+	bi->tag_size = template->tag_size;
+
+	return 0;
+}
+EXPORT_SYMBOL(blk_integrity_register);
+
+/**
+ * blk_integrity_unregister - Remove block integrity profile
+ * @disk:	disk whose integrity profile to deallocate
+ *
+ * Description: This function frees all memory used by the block
+ * integrity profile.  To be called at device teardown.
+ */
+void blk_integrity_unregister(struct gendisk *disk)
+{
+	struct blk_integrity *bi;
+
+	if (!disk || !disk->integrity)
+		return;
+
+	bi = disk->integrity;
+
+	kobject_uevent(&bi->kobj, KOBJ_REMOVE);
+	kobject_del(&bi->kobj);
+	kobject_put(&disk->dev.kobj);
+	kmem_cache_free(integrity_cachep, bi);
+}
+EXPORT_SYMBOL(blk_integrity_unregister);
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 651136aae76e..5efc9e7a68b7 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -441,6 +441,9 @@ static int attempt_merge(struct request_queue *q, struct request *req,
 	    || next->special)
 		return 0;
 
+	if (blk_integrity_rq(req) != blk_integrity_rq(next))
+		return 0;
+
 	/*
 	 * If we are allowed to merge, then append bio list
 	 * from next to rq and release next. merge_requests_fn
diff --git a/block/blk.h b/block/blk.h
index 59776ab4742a..c79f30e1df52 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -51,4 +51,12 @@ static inline int queue_congestion_off_threshold(struct request_queue *q)
 	return q->nr_congestion_off;
 }
 
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+
+#define rq_for_each_integrity_segment(bvl, _rq, _iter)		\
+	__rq_for_each_bio(_iter.bio, _rq)			\
+		bip_for_each_vec(bvl, _iter.bio->bi_integrity, _iter.i)
+
+#endif /* BLK_DEV_INTEGRITY */
+
 #endif
diff --git a/block/elevator.c b/block/elevator.c
index 902dd1344d56..1f5bfe696026 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -86,6 +86,12 @@ int elv_rq_merge_ok(struct request *rq, struct bio *bio)
 	if (rq->rq_disk != bio->bi_bdev->bd_disk || rq->special)
 		return 0;
 
+	/*
+	 * only merge integrity protected bio into ditto rq
+	 */
+	if (bio_integrity(bio) != blk_integrity_rq(rq))
+		return 0;
+
 	if (!elv_iosched_allow_merge(rq, bio))
 		return 0;
 
diff --git a/fs/Makefile b/fs/Makefile
index 1e7a11bd4da1..277b079dec9e 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -19,6 +19,7 @@ else
 obj-y +=	no-block.o
 endif
 
+obj-$(CONFIG_BLK_DEV_INTEGRITY) += bio-integrity.o
 obj-$(CONFIG_INOTIFY)		+= inotify.o
 obj-$(CONFIG_INOTIFY_USER)	+= inotify_user.o
 obj-$(CONFIG_EPOLL)		+= eventpoll.o
diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c
new file mode 100644
index 000000000000..31b08878913d
--- /dev/null
+++ b/fs/bio-integrity.c
@@ -0,0 +1,708 @@
+/*
+ * bio-integrity.c - bio data integrity extensions
+ *
+ * Copyright (C) 2007, 2008 Oracle Corporation
+ * Written by: Martin K. Petersen <martin.petersen@oracle.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; see the file COPYING.  If not, write to
+ * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139,
+ * USA.
+ *
+ */
+
+#include <linux/blkdev.h>
+#include <linux/mempool.h>
+#include <linux/bio.h>
+#include <linux/workqueue.h>
+
+static struct kmem_cache *bio_integrity_slab __read_mostly;
+static struct workqueue_struct *kintegrityd_wq;
+
+/**
+ * bio_integrity_alloc_bioset - Allocate integrity payload and attach it to bio
+ * @bio:	bio to attach integrity metadata to
+ * @gfp_mask:	Memory allocation mask
+ * @nr_vecs:	Number of integrity metadata scatter-gather elements
+ * @bs:		bio_set to allocate from
+ *
+ * Description: This function prepares a bio for attaching integrity
+ * metadata.  nr_vecs specifies the maximum number of pages containing
+ * integrity metadata that can be attached.
+ */
+struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *bio, gfp_t gfp_mask, unsigned int nr_vecs, struct bio_set *bs)
+{
+	struct bio_integrity_payload *bip;
+	struct bio_vec *iv;
+	unsigned long idx;
+
+	BUG_ON(bio == NULL);
+
+	bip = mempool_alloc(bs->bio_integrity_pool, gfp_mask);
+	if (unlikely(bip == NULL)) {
+		printk(KERN_ERR "%s: could not alloc bip\n", __func__);
+		return NULL;
+	}
+
+	memset(bip, 0, sizeof(*bip));
+
+	iv = bvec_alloc_bs(gfp_mask, nr_vecs, &idx, bs);
+	if (unlikely(iv == NULL)) {
+		printk(KERN_ERR "%s: could not alloc bip_vec\n", __func__);
+		mempool_free(bip, bs->bio_integrity_pool);
+		return NULL;
+	}
+
+	bip->bip_pool = idx;
+	bip->bip_vec = iv;
+	bip->bip_bio = bio;
+	bio->bi_integrity = bip;
+
+	return bip;
+}
+EXPORT_SYMBOL(bio_integrity_alloc_bioset);
+
+/**
+ * bio_integrity_alloc - Allocate integrity payload and attach it to bio
+ * @bio:	bio to attach integrity metadata to
+ * @gfp_mask:	Memory allocation mask
+ * @nr_vecs:	Number of integrity metadata scatter-gather elements
+ *
+ * Description: This function prepares a bio for attaching integrity
+ * metadata.  nr_vecs specifies the maximum number of pages containing
+ * integrity metadata that can be attached.
+ */
+struct bio_integrity_payload *bio_integrity_alloc(struct bio *bio, gfp_t gfp_mask, unsigned int nr_vecs)
+{
+	return bio_integrity_alloc_bioset(bio, gfp_mask, nr_vecs, fs_bio_set);
+}
+EXPORT_SYMBOL(bio_integrity_alloc);
+
+/**
+ * bio_integrity_free - Free bio integrity payload
+ * @bio:	bio containing bip to be freed
+ * @bs:		bio_set this bio was allocated from
+ *
+ * Description: Used to free the integrity portion of a bio. Usually
+ * called from bio_free().
+ */
+void bio_integrity_free(struct bio *bio, struct bio_set *bs)
+{
+	struct bio_integrity_payload *bip = bio->bi_integrity;
+
+	BUG_ON(bip == NULL);
+
+	/* A cloned bio doesn't own the integrity metadata */
+	if (!bio_flagged(bio, BIO_CLONED) && bip->bip_buf != NULL)
+		kfree(bip->bip_buf);
+
+	mempool_free(bip->bip_vec, bs->bvec_pools[bip->bip_pool]);
+	mempool_free(bip, bs->bio_integrity_pool);
+
+	bio->bi_integrity = NULL;
+}
+EXPORT_SYMBOL(bio_integrity_free);
+
+/**
+ * bio_integrity_add_page - Attach integrity metadata
+ * @bio:	bio to update
+ * @page:	page containing integrity metadata
+ * @len:	number of bytes of integrity metadata in page
+ * @offset:	start offset within page
+ *
+ * Description: Attach a page containing integrity metadata to bio.
+ */
+int bio_integrity_add_page(struct bio *bio, struct page *page,
+			   unsigned int len, unsigned int offset)
+{
+	struct bio_integrity_payload *bip = bio->bi_integrity;
+	struct bio_vec *iv;
+
+	if (bip->bip_vcnt >= bvec_nr_vecs(bip->bip_pool)) {
+		printk(KERN_ERR "%s: bip_vec full\n", __func__);
+		return 0;
+	}
+
+	iv = bip_vec_idx(bip, bip->bip_vcnt);
+	BUG_ON(iv == NULL);
+	BUG_ON(iv->bv_page != NULL);
+
+	iv->bv_page = page;
+	iv->bv_len = len;
+	iv->bv_offset = offset;
+	bip->bip_vcnt++;
+
+	return len;
+}
+EXPORT_SYMBOL(bio_integrity_add_page);
+
+/**
+ * bio_integrity_enabled - Check whether integrity can be passed
+ * @bio:	bio to check
+ *
+ * Description: Determines whether bio_integrity_prep() can be called
+ * on this bio or not.	bio data direction and target device must be
+ * set prior to calling.  The functions honors the write_generate and
+ * read_verify flags in sysfs.
+ */
+int bio_integrity_enabled(struct bio *bio)
+{
+	/* Already protected? */
+	if (bio_integrity(bio))
+		return 0;
+
+	return bdev_integrity_enabled(bio->bi_bdev, bio_data_dir(bio));
+}
+EXPORT_SYMBOL(bio_integrity_enabled);
+
+/**
+ * bio_integrity_hw_sectors - Convert 512b sectors to hardware ditto
+ * @bi:		blk_integrity profile for device
+ * @sectors:	Number of 512 sectors to convert
+ *
+ * Description: The block layer calculates everything in 512 byte
+ * sectors but integrity metadata is done in terms of the hardware
+ * sector size of the storage device.  Convert the block layer sectors
+ * to physical sectors.
+ */
+static inline unsigned int bio_integrity_hw_sectors(struct blk_integrity *bi, unsigned int sectors)
+{
+	/* At this point there are only 512b or 4096b DIF/EPP devices */
+	if (bi->sector_size == 4096)
+		return sectors >>= 3;
+
+	return sectors;
+}
+
+/**
+ * bio_integrity_tag_size - Retrieve integrity tag space
+ * @bio:	bio to inspect
+ *
+ * Description: Returns the maximum number of tag bytes that can be
+ * attached to this bio. Filesystems can use this to determine how
+ * much metadata to attach to an I/O.
+ */
+unsigned int bio_integrity_tag_size(struct bio *bio)
+{
+	struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev);
+
+	BUG_ON(bio->bi_size == 0);
+
+	return bi->tag_size * (bio->bi_size / bi->sector_size);
+}
+EXPORT_SYMBOL(bio_integrity_tag_size);
+
+int bio_integrity_tag(struct bio *bio, void *tag_buf, unsigned int len, int set)
+{
+	struct bio_integrity_payload *bip = bio->bi_integrity;
+	struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev);
+	unsigned int nr_sectors;
+
+	BUG_ON(bip->bip_buf == NULL);
+
+	if (bi->tag_size == 0)
+		return -1;
+
+	nr_sectors = bio_integrity_hw_sectors(bi, DIV_ROUND_UP(len, bi->tag_size));
+
+	if (nr_sectors * bi->tuple_size > bip->bip_size) {
+		printk(KERN_ERR "%s: tag too big for bio: %u > %u\n",
+		       __func__, nr_sectors * bi->tuple_size, bip->bip_size);
+		return -1;
+	}
+
+	if (set)
+		bi->set_tag_fn(bip->bip_buf, tag_buf, nr_sectors);
+	else
+		bi->get_tag_fn(bip->bip_buf, tag_buf, nr_sectors);
+
+	return 0;
+}
+
+/**
+ * bio_integrity_set_tag - Attach a tag buffer to a bio
+ * @bio:	bio to attach buffer to
+ * @tag_buf:	Pointer to a buffer containing tag data
+ * @len:	Length of the included buffer
+ *
+ * Description: Use this function to tag a bio by leveraging the extra
+ * space provided by devices formatted with integrity protection.  The
+ * size of the integrity buffer must be <= to the size reported by
+ * bio_integrity_tag_size().
+ */
+int bio_integrity_set_tag(struct bio *bio, void *tag_buf, unsigned int len)
+{
+	BUG_ON(bio_data_dir(bio) != WRITE);
+
+	return bio_integrity_tag(bio, tag_buf, len, 1);
+}
+EXPORT_SYMBOL(bio_integrity_set_tag);
+
+/**
+ * bio_integrity_get_tag - Retrieve a tag buffer from a bio
+ * @bio:	bio to retrieve buffer from
+ * @tag_buf:	Pointer to a buffer for the tag data
+ * @len:	Length of the target buffer
+ *
+ * Description: Use this function to retrieve the tag buffer from a
+ * completed I/O. The size of the integrity buffer must be <= to the
+ * size reported by bio_integrity_tag_size().
+ */
+int bio_integrity_get_tag(struct bio *bio, void *tag_buf, unsigned int len)
+{
+	BUG_ON(bio_data_dir(bio) != READ);
+
+	return bio_integrity_tag(bio, tag_buf, len, 0);
+}
+EXPORT_SYMBOL(bio_integrity_get_tag);
+
+/**
+ * bio_integrity_generate - Generate integrity metadata for a bio
+ * @bio:	bio to generate integrity metadata for
+ *
+ * Description: Generates integrity metadata for a bio by calling the
+ * block device's generation callback function.  The bio must have a
+ * bip attached with enough room to accommodate the generated
+ * integrity metadata.
+ */
+static void bio_integrity_generate(struct bio *bio)
+{
+	struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev);
+	struct blk_integrity_exchg bix;
+	struct bio_vec *bv;
+	sector_t sector = bio->bi_sector;
+	unsigned int i, sectors, total;
+	void *prot_buf = bio->bi_integrity->bip_buf;
+
+	total = 0;
+	bix.disk_name = bio->bi_bdev->bd_disk->disk_name;
+	bix.sector_size = bi->sector_size;
+
+	bio_for_each_segment(bv, bio, i) {
+		void *kaddr = kmap_atomic(bv->bv_page, KM_USER0);
+		bix.data_buf = kaddr + bv->bv_offset;
+		bix.data_size = bv->bv_len;
+		bix.prot_buf = prot_buf;
+		bix.sector = sector;
+
+		bi->generate_fn(&bix);
+
+		sectors = bv->bv_len / bi->sector_size;
+		sector += sectors;
+		prot_buf += sectors * bi->tuple_size;
+		total += sectors * bi->tuple_size;
+		BUG_ON(total > bio->bi_integrity->bip_size);
+
+		kunmap_atomic(kaddr, KM_USER0);
+	}
+}
+
+/**
+ * bio_integrity_prep - Prepare bio for integrity I/O
+ * @bio:	bio to prepare
+ *
+ * Description: Allocates a buffer for integrity metadata, maps the
+ * pages and attaches them to a bio.  The bio must have data
+ * direction, target device and start sector set priot to calling.  In
+ * the WRITE case, integrity metadata will be generated using the
+ * block device's integrity function.  In the READ case, the buffer
+ * will be prepared for DMA and a suitable end_io handler set up.
+ */
+int bio_integrity_prep(struct bio *bio)
+{
+	struct bio_integrity_payload *bip;
+	struct blk_integrity *bi;
+	struct request_queue *q;
+	void *buf;
+	unsigned long start, end;
+	unsigned int len, nr_pages;
+	unsigned int bytes, offset, i;
+	unsigned int sectors;
+
+	bi = bdev_get_integrity(bio->bi_bdev);
+	q = bdev_get_queue(bio->bi_bdev);
+	BUG_ON(bi == NULL);
+	BUG_ON(bio_integrity(bio));
+
+	sectors = bio_integrity_hw_sectors(bi, bio_sectors(bio));
+
+	/* Allocate kernel buffer for protection data */
+	len = sectors * blk_integrity_tuple_size(bi);
+	buf = kmalloc(len, GFP_NOIO | __GFP_NOFAIL | q->bounce_gfp);
+	if (unlikely(buf == NULL)) {
+		printk(KERN_ERR "could not allocate integrity buffer\n");
+		return -EIO;
+	}
+
+	end = (((unsigned long) buf) + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	start = ((unsigned long) buf) >> PAGE_SHIFT;
+	nr_pages = end - start;
+
+	/* Allocate bio integrity payload and integrity vectors */
+	bip = bio_integrity_alloc(bio, GFP_NOIO, nr_pages);
+	if (unlikely(bip == NULL)) {
+		printk(KERN_ERR "could not allocate data integrity bioset\n");
+		kfree(buf);
+		return -EIO;
+	}
+
+	bip->bip_buf = buf;
+	bip->bip_size = len;
+	bip->bip_sector = bio->bi_sector;
+
+	/* Map it */
+	offset = offset_in_page(buf);
+	for (i = 0 ; i < nr_pages ; i++) {
+		int ret;
+		bytes = PAGE_SIZE - offset;
+
+		if (len <= 0)
+			break;
+
+		if (bytes > len)
+			bytes = len;
+
+		ret = bio_integrity_add_page(bio, virt_to_page(buf),
+					     bytes, offset);
+
+		if (ret == 0)
+			return 0;
+
+		if (ret < bytes)
+			break;
+
+		buf += bytes;
+		len -= bytes;
+		offset = 0;
+	}
+
+	/* Install custom I/O completion handler if read verify is enabled */
+	if (bio_data_dir(bio) == READ) {
+		bip->bip_end_io = bio->bi_end_io;
+		bio->bi_end_io = bio_integrity_endio;
+	}
+
+	/* Auto-generate integrity metadata if this is a write */
+	if (bio_data_dir(bio) == WRITE)
+		bio_integrity_generate(bio);
+
+	return 0;
+}
+EXPORT_SYMBOL(bio_integrity_prep);
+
+/**
+ * bio_integrity_verify - Verify integrity metadata for a bio
+ * @bio:	bio to verify
+ *
+ * Description: This function is called to verify the integrity of a
+ * bio.	 The data in the bio io_vec is compared to the integrity
+ * metadata returned by the HBA.
+ */
+static int bio_integrity_verify(struct bio *bio)
+{
+	struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev);
+	struct blk_integrity_exchg bix;
+	struct bio_vec *bv;
+	sector_t sector = bio->bi_integrity->bip_sector;
+	unsigned int i, sectors, total, ret;
+	void *prot_buf = bio->bi_integrity->bip_buf;
+
+	ret = total = 0;
+	bix.disk_name = bio->bi_bdev->bd_disk->disk_name;
+	bix.sector_size = bi->sector_size;
+
+	bio_for_each_segment(bv, bio, i) {
+		void *kaddr = kmap_atomic(bv->bv_page, KM_USER0);
+		bix.data_buf = kaddr + bv->bv_offset;
+		bix.data_size = bv->bv_len;
+		bix.prot_buf = prot_buf;
+		bix.sector = sector;
+
+		ret = bi->verify_fn(&bix);
+
+		if (ret) {
+			kunmap_atomic(kaddr, KM_USER0);
+			break;
+		}
+
+		sectors = bv->bv_len / bi->sector_size;
+		sector += sectors;
+		prot_buf += sectors * bi->tuple_size;
+		total += sectors * bi->tuple_size;
+		BUG_ON(total > bio->bi_integrity->bip_size);
+
+		kunmap_atomic(kaddr, KM_USER0);
+	}
+
+	return ret;
+}
+
+/**
+ * bio_integrity_verify_fn - Integrity I/O completion worker
+ * @work:	Work struct stored in bio to be verified
+ *
+ * Description: This workqueue function is called to complete a READ
+ * request.  The function verifies the transferred integrity metadata
+ * and then calls the original bio end_io function.
+ */
+static void bio_integrity_verify_fn(struct work_struct *work)
+{
+	struct bio_integrity_payload *bip = 
+		container_of(work, struct bio_integrity_payload, bip_work);
+	struct bio *bio = bip->bip_bio;
+	int error = bip->bip_error;
+
+	if (bio_integrity_verify(bio)) {
+		clear_bit(BIO_UPTODATE, &bio->bi_flags);
+		error = -EIO;
+	}
+
+	/* Restore original bio completion handler */
+	bio->bi_end_io = bip->bip_end_io;
+
+	if (bio->bi_end_io)
+		bio->bi_end_io(bio, error);
+}
+
+/**
+ * bio_integrity_endio - Integrity I/O completion function
+ * @bio:	Protected bio
+ * @error:	Pointer to errno
+ *
+ * Description: Completion for integrity I/O
+ *
+ * Normally I/O completion is done in interrupt context.  However,
+ * verifying I/O integrity is a time-consuming task which must be run
+ * in process context.	This function postpones completion
+ * accordingly.
+ */
+void bio_integrity_endio(struct bio *bio, int error)
+{
+	struct bio_integrity_payload *bip = bio->bi_integrity;
+
+	BUG_ON(bip->bip_bio != bio);
+
+	bip->bip_error = error;
+	INIT_WORK(&bip->bip_work, bio_integrity_verify_fn);
+	queue_work(kintegrityd_wq, &bip->bip_work);
+}
+EXPORT_SYMBOL(bio_integrity_endio);
+
+/**
+ * bio_integrity_mark_head - Advance bip_vec skip bytes
+ * @bip:	Integrity vector to advance
+ * @skip:	Number of bytes to advance it
+ */
+void bio_integrity_mark_head(struct bio_integrity_payload *bip, unsigned int skip)
+{
+	struct bio_vec *iv;
+	unsigned int i;
+
+	bip_for_each_vec(iv, bip, i) {
+		if (skip == 0) {
+			bip->bip_idx = i;
+			return;
+		} else if (skip >= iv->bv_len) {
+			skip -= iv->bv_len;
+		} else { /* skip < iv->bv_len) */
+			iv->bv_offset += skip;
+			iv->bv_len -= skip;
+			bip->bip_idx = i;
+			return;
+		}
+	}
+}
+
+/**
+ * bio_integrity_mark_tail - Truncate bip_vec to be len bytes long
+ * @bip:	Integrity vector to truncate
+ * @len:	New length of integrity vector
+ */
+void bio_integrity_mark_tail(struct bio_integrity_payload *bip, unsigned int len)
+{
+	struct bio_vec *iv;
+	unsigned int i;
+
+	bip_for_each_vec(iv, bip, i) {
+		if (len == 0) {
+			bip->bip_vcnt = i;
+			return;
+		} else if (len >= iv->bv_len) {
+			len -= iv->bv_len;
+		} else { /* len < iv->bv_len) */
+			iv->bv_len = len;
+			len = 0;
+		}
+	}
+}
+
+/**
+ * bio_integrity_advance - Advance integrity vector
+ * @bio:	bio whose integrity vector to update
+ * @bytes_done:	number of data bytes that have been completed
+ *
+ * Description: This function calculates how many integrity bytes the
+ * number of completed data bytes correspond to and advances the
+ * integrity vector accordingly.
+ */
+void bio_integrity_advance(struct bio *bio, unsigned int bytes_done)
+{
+	struct bio_integrity_payload *bip = bio->bi_integrity;
+	struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev);
+	unsigned int nr_sectors;
+
+	BUG_ON(bip == NULL);
+	BUG_ON(bi == NULL);
+
+	nr_sectors = bio_integrity_hw_sectors(bi, bytes_done >> 9);
+	bio_integrity_mark_head(bip, nr_sectors * bi->tuple_size);
+}
+EXPORT_SYMBOL(bio_integrity_advance);
+
+/**
+ * bio_integrity_trim - Trim integrity vector
+ * @bio:	bio whose integrity vector to update
+ * @offset:	offset to first data sector
+ * @sectors:	number of data sectors
+ *
+ * Description: Used to trim the integrity vector in a cloned bio.
+ * The ivec will be advanced corresponding to 'offset' data sectors
+ * and the length will be truncated corresponding to 'len' data
+ * sectors.
+ */
+void bio_integrity_trim(struct bio *bio, unsigned int offset, unsigned int sectors)
+{
+	struct bio_integrity_payload *bip = bio->bi_integrity;
+	struct blk_integrity *bi = bdev_get_integrity(bio->bi_bdev);
+	unsigned int nr_sectors;
+
+	BUG_ON(bip == NULL);
+	BUG_ON(bi == NULL);
+	BUG_ON(!bio_flagged(bio, BIO_CLONED));
+
+	nr_sectors = bio_integrity_hw_sectors(bi, sectors);
+	bip->bip_sector = bip->bip_sector + offset;
+	bio_integrity_mark_head(bip, offset * bi->tuple_size);
+	bio_integrity_mark_tail(bip, sectors * bi->tuple_size);
+}
+EXPORT_SYMBOL(bio_integrity_trim);
+
+/**
+ * bio_integrity_split - Split integrity metadata
+ * @bio:	Protected bio
+ * @bp:		Resulting bio_pair
+ * @sectors:	Offset
+ *
+ * Description: Splits an integrity page into a bio_pair.
+ */
+void bio_integrity_split(struct bio *bio, struct bio_pair *bp, int sectors)
+{
+	struct blk_integrity *bi;
+	struct bio_integrity_payload *bip = bio->bi_integrity;
+	unsigned int nr_sectors;
+
+	if (bio_integrity(bio) == 0)
+		return;
+
+	bi = bdev_get_integrity(bio->bi_bdev);
+	BUG_ON(bi == NULL);
+	BUG_ON(bip->bip_vcnt != 1);
+
+	nr_sectors = bio_integrity_hw_sectors(bi, sectors);
+
+	bp->bio1.bi_integrity = &bp->bip1;
+	bp->bio2.bi_integrity = &bp->bip2;
+
+	bp->iv1 = bip->bip_vec[0];
+	bp->iv2 = bip->bip_vec[0];
+
+	bp->bip1.bip_vec = &bp->iv1;
+	bp->bip2.bip_vec = &bp->iv2;
+
+	bp->iv1.bv_len = sectors * bi->tuple_size;
+	bp->iv2.bv_offset += sectors * bi->tuple_size;
+	bp->iv2.bv_len -= sectors * bi->tuple_size;
+
+	bp->bip1.bip_sector = bio->bi_integrity->bip_sector;
+	bp->bip2.bip_sector = bio->bi_integrity->bip_sector + nr_sectors;
+
+	bp->bip1.bip_vcnt = bp->bip2.bip_vcnt = 1;
+	bp->bip1.bip_idx = bp->bip2.bip_idx = 0;
+}
+EXPORT_SYMBOL(bio_integrity_split);
+
+/**
+ * bio_integrity_clone - Callback for cloning bios with integrity metadata
+ * @bio:	New bio
+ * @bio_src:	Original bio
+ * @bs:		bio_set to allocate bip from
+ *
+ * Description:	Called to allocate a bip when cloning a bio
+ */
+int bio_integrity_clone(struct bio *bio, struct bio *bio_src, struct bio_set *bs)
+{
+	struct bio_integrity_payload *bip_src = bio_src->bi_integrity;
+	struct bio_integrity_payload *bip;
+
+	BUG_ON(bip_src == NULL);
+
+	bip = bio_integrity_alloc_bioset(bio, GFP_NOIO, bip_src->bip_vcnt, bs);
+
+	if (bip == NULL)
+		return -EIO;
+
+	memcpy(bip->bip_vec, bip_src->bip_vec,
+	       bip_src->bip_vcnt * sizeof(struct bio_vec));
+
+	bip->bip_sector = bip_src->bip_sector;
+	bip->bip_vcnt = bip_src->bip_vcnt;
+	bip->bip_idx = bip_src->bip_idx;
+
+	return 0;
+}
+EXPORT_SYMBOL(bio_integrity_clone);
+
+int bioset_integrity_create(struct bio_set *bs, int pool_size)
+{
+	bs->bio_integrity_pool = mempool_create_slab_pool(pool_size,
+							  bio_integrity_slab);
+	if (!bs->bio_integrity_pool)
+		return -1;
+
+	return 0;
+}
+EXPORT_SYMBOL(bioset_integrity_create);
+
+void bioset_integrity_free(struct bio_set *bs)
+{
+	if (bs->bio_integrity_pool)
+		mempool_destroy(bs->bio_integrity_pool);
+}
+EXPORT_SYMBOL(bioset_integrity_free);
+
+void __init bio_integrity_init_slab(void)
+{
+	bio_integrity_slab = KMEM_CACHE(bio_integrity_payload,
+					SLAB_HWCACHE_ALIGN|SLAB_PANIC);
+}
+EXPORT_SYMBOL(bio_integrity_init_slab);
+
+static int __init integrity_init(void)
+{
+	kintegrityd_wq = create_workqueue("kintegrityd");
+
+	if (!kintegrityd_wq)
+		panic("Failed to create kintegrityd\n");
+
+	return 0;
+}
+subsys_initcall(integrity_init);
diff --git a/fs/bio.c b/fs/bio.c
index 7a6598abc967..7761c84c7032 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -50,6 +50,11 @@ static struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly = {
  */
 struct bio_set *fs_bio_set;
 
+unsigned int bvec_nr_vecs(unsigned short idx)
+{
+	return bvec_slabs[idx].nr_vecs;
+}
+
 struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs)
 {
 	struct bio_vec *bvl;
@@ -91,6 +96,9 @@ void bio_free(struct bio *bio, struct bio_set *bio_set)
 		mempool_free(bio->bi_io_vec, bio_set->bvec_pools[pool_idx]);
 	}
 
+	if (bio_integrity(bio))
+		bio_integrity_free(bio, bio_set);
+
 	mempool_free(bio, bio_set->bio_pool);
 }
 
@@ -249,9 +257,19 @@ struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask)
 {
 	struct bio *b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, fs_bio_set);
 
-	if (b) {
-		b->bi_destructor = bio_fs_destructor;
-		__bio_clone(b, bio);
+	if (!b)
+		return NULL;
+
+	b->bi_destructor = bio_fs_destructor;
+	__bio_clone(b, bio);
+
+	if (bio_integrity(bio)) {
+		int ret;
+
+		ret = bio_integrity_clone(b, bio, fs_bio_set);
+
+		if (ret < 0)
+			return NULL;
 	}
 
 	return b;
@@ -1223,6 +1241,9 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors)
 	bp->bio1.bi_private = bi;
 	bp->bio2.bi_private = pool;
 
+	if (bio_integrity(bi))
+		bio_integrity_split(bi, bp, first_sectors);
+
 	return bp;
 }
 
@@ -1264,6 +1285,7 @@ void bioset_free(struct bio_set *bs)
 	if (bs->bio_pool)
 		mempool_destroy(bs->bio_pool);
 
+	bioset_integrity_free(bs);
 	biovec_free_pools(bs);
 
 	kfree(bs);
@@ -1280,6 +1302,9 @@ struct bio_set *bioset_create(int bio_pool_size, int bvec_pool_size)
 	if (!bs->bio_pool)
 		goto bad;
 
+	if (bioset_integrity_create(bs, bio_pool_size))
+		goto bad;
+
 	if (!biovec_create_pools(bs, bvec_pool_size))
 		return bs;
 
@@ -1306,6 +1331,7 @@ static int __init init_bio(void)
 {
 	bio_slab = KMEM_CACHE(bio, SLAB_HWCACHE_ALIGN|SLAB_PANIC);
 
+	bio_integrity_init_slab();
 	biovec_init_slabs();
 
 	fs_bio_set = bioset_create(BIO_POOL_SIZE, 2);
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 49dfb3cb7460..6bfc3e8d9d89 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -64,6 +64,7 @@ struct bio_vec {
 
 struct bio_set;
 struct bio;
+struct bio_integrity_payload;
 typedef void (bio_end_io_t) (struct bio *, int);
 typedef void (bio_destructor_t) (struct bio *);
 
@@ -112,6 +113,9 @@ struct bio {
 	atomic_t		bi_cnt;		/* pin count */
 
 	void			*bi_private;
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+	struct bio_integrity_payload *bi_integrity;  /* data integrity */
+#endif
 
 	bio_destructor_t	*bi_destructor;	/* destructor */
 };
@@ -271,6 +275,29 @@ static inline void *bio_data(struct bio *bio)
  */
 #define bio_get(bio)	atomic_inc(&(bio)->bi_cnt)
 
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+/*
+ * bio integrity payload
+ */
+struct bio_integrity_payload {
+	struct bio		*bip_bio;	/* parent bio */
+	struct bio_vec		*bip_vec;	/* integrity data vector */
+
+	sector_t		bip_sector;	/* virtual start sector */
+
+	void			*bip_buf;	/* generated integrity data */
+	bio_end_io_t		*bip_end_io;	/* saved I/O completion fn */
+
+	int			bip_error;	/* saved I/O error */
+	unsigned int		bip_size;
+
+	unsigned short		bip_pool;	/* pool the ivec came from */
+	unsigned short		bip_vcnt;	/* # of integrity bio_vecs */
+	unsigned short		bip_idx;	/* current bip_vec index */
+
+	struct work_struct	bip_work;	/* I/O completion */
+};
+#endif /* CONFIG_BLK_DEV_INTEGRITY */
 
 /*
  * A bio_pair is used when we need to split a bio.
@@ -283,10 +310,14 @@ static inline void *bio_data(struct bio *bio)
  *   in bio2.bi_private
  */
 struct bio_pair {
-	struct bio	bio1, bio2;
-	struct bio_vec	bv1, bv2;
-	atomic_t	cnt;
-	int		error;
+	struct bio			bio1, bio2;
+	struct bio_vec			bv1, bv2;
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+	struct bio_integrity_payload	bip1, bip2;
+	struct bio_vec			iv1, iv2;
+#endif
+	atomic_t			cnt;
+	int				error;
 };
 extern struct bio_pair *bio_split(struct bio *bi, mempool_t *pool,
 				  int first_sectors);
@@ -334,6 +365,7 @@ extern struct bio *bio_copy_user_iov(struct request_queue *, struct sg_iovec *,
 extern int bio_uncopy_user(struct bio *);
 void zero_fill_bio(struct bio *bio);
 extern struct bio_vec *bvec_alloc_bs(gfp_t, int, unsigned long *, struct bio_set *);
+extern unsigned int bvec_nr_vecs(unsigned short idx);
 
 /*
  * bio_set is used to allow other portions of the IO system to
@@ -346,6 +378,9 @@ extern struct bio_vec *bvec_alloc_bs(gfp_t, int, unsigned long *, struct bio_set
 
 struct bio_set {
 	mempool_t *bio_pool;
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+	mempool_t *bio_integrity_pool;
+#endif
 	mempool_t *bvec_pools[BIOVEC_NR_POOLS];
 };
 
@@ -410,5 +445,56 @@ static inline char *__bio_kmap_irq(struct bio *bio, unsigned short idx,
 	__bio_kmap_irq((bio), (bio)->bi_idx, (flags))
 #define bio_kunmap_irq(buf,flags)	__bio_kunmap_irq(buf, flags)
 
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+
+#define bip_vec_idx(bip, idx)	(&(bip->bip_vec[(idx)]))
+#define bip_vec(bip)		bip_vec_idx(bip, 0)
+
+#define __bip_for_each_vec(bvl, bip, i, start_idx)			\
+	for (bvl = bip_vec_idx((bip), (start_idx)), i = (start_idx);	\
+	     i < (bip)->bip_vcnt;					\
+	     bvl++, i++)
+
+#define bip_for_each_vec(bvl, bip, i)					\
+	__bip_for_each_vec(bvl, bip, i, (bip)->bip_idx)
+
+#define bio_integrity(bio)	((bio)->bi_integrity ? 1 : 0)
+
+extern struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *, gfp_t, unsigned int, struct bio_set *);
+extern struct bio_integrity_payload *bio_integrity_alloc(struct bio *, gfp_t, unsigned int);
+extern void bio_integrity_free(struct bio *, struct bio_set *);
+extern int bio_integrity_add_page(struct bio *, struct page *, unsigned int, unsigned int);
+extern int bio_integrity_enabled(struct bio *bio);
+extern int bio_integrity_set_tag(struct bio *, void *, unsigned int);
+extern int bio_integrity_get_tag(struct bio *, void *, unsigned int);
+extern int bio_integrity_prep(struct bio *);
+extern void bio_integrity_endio(struct bio *, int);
+extern void bio_integrity_advance(struct bio *, unsigned int);
+extern void bio_integrity_trim(struct bio *, unsigned int, unsigned int);
+extern void bio_integrity_split(struct bio *, struct bio_pair *, int);
+extern int bio_integrity_clone(struct bio *, struct bio *, struct bio_set *);
+extern int bioset_integrity_create(struct bio_set *, int);
+extern void bioset_integrity_free(struct bio_set *);
+extern void bio_integrity_init_slab(void);
+
+#else /* CONFIG_BLK_DEV_INTEGRITY */
+
+#define bio_integrity(a)		(0)
+#define bioset_integrity_create(a, b)	(0)
+#define bio_integrity_prep(a)		(0)
+#define bio_integrity_enabled(a)	(0)
+#define bio_integrity_clone(a, b, c)	(0)
+#define bioset_integrity_free(a)	do { } while (0)
+#define bio_integrity_free(a, b)	do { } while (0)
+#define bio_integrity_endio(a, b)	do { } while (0)
+#define bio_integrity_advance(a, b)	do { } while (0)
+#define bio_integrity_trim(a, b, c)	do { } while (0)
+#define bio_integrity_split(a, b, c)	do { } while (0)
+#define bio_integrity_set_tag(a, b, c)	do { } while (0)
+#define bio_integrity_get_tag(a, b, c)	do { } while (0)
+#define bio_integrity_init_slab(a)	do { } while (0)
+
+#endif /* CONFIG_BLK_DEV_INTEGRITY */
+
 #endif /* CONFIG_BLOCK */
 #endif /* __LINUX_BIO_H */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 6a3da6717135..4a9ed45270ff 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -112,6 +112,7 @@ enum rq_flag_bits {
 	__REQ_ALLOCED,		/* request came from our alloc pool */
 	__REQ_RW_META,		/* metadata io request */
 	__REQ_COPY_USER,	/* contains copies of user pages */
+	__REQ_INTEGRITY,	/* integrity metadata has been remapped */
 	__REQ_NR_BITS,		/* stops here */
 };
 
@@ -134,6 +135,7 @@ enum rq_flag_bits {
 #define REQ_ALLOCED	(1 << __REQ_ALLOCED)
 #define REQ_RW_META	(1 << __REQ_RW_META)
 #define REQ_COPY_USER	(1 << __REQ_COPY_USER)
+#define REQ_INTEGRITY	(1 << __REQ_INTEGRITY)
 
 #define BLK_MAX_CDB	16
 
@@ -865,6 +867,109 @@ void kblockd_flush_work(struct work_struct *work);
 	MODULE_ALIAS("block-major-" __stringify(major) "-*")
 
 
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+
+#define INTEGRITY_FLAG_READ	1	/* verify data integrity on read */
+#define INTEGRITY_FLAG_WRITE	2	/* generate data integrity on write */
+
+struct blk_integrity_exchg {
+	void			*prot_buf;
+	void			*data_buf;
+	sector_t		sector;
+	unsigned int		data_size;
+	unsigned short		sector_size;
+	const char		*disk_name;
+};
+
+typedef void (integrity_gen_fn) (struct blk_integrity_exchg *);
+typedef int (integrity_vrfy_fn) (struct blk_integrity_exchg *);
+typedef void (integrity_set_tag_fn) (void *, void *, unsigned int);
+typedef void (integrity_get_tag_fn) (void *, void *, unsigned int);
+
+struct blk_integrity {
+	integrity_gen_fn	*generate_fn;
+	integrity_vrfy_fn	*verify_fn;
+	integrity_set_tag_fn	*set_tag_fn;
+	integrity_get_tag_fn	*get_tag_fn;
+
+	unsigned short		flags;
+	unsigned short		tuple_size;
+	unsigned short		sector_size;
+	unsigned short		tag_size;
+
+	const char		*name;
+
+	struct kobject		kobj;
+};
+
+extern int blk_integrity_register(struct gendisk *, struct blk_integrity *);
+extern void blk_integrity_unregister(struct gendisk *);
+extern int blk_integrity_compare(struct block_device *, struct block_device *);
+extern int blk_rq_map_integrity_sg(struct request *, struct scatterlist *);
+extern int blk_rq_count_integrity_sg(struct request *);
+
+static inline unsigned short blk_integrity_tuple_size(struct blk_integrity *bi)
+{
+	if (bi)
+		return bi->tuple_size;
+
+	return 0;
+}
+
+static inline struct blk_integrity *bdev_get_integrity(struct block_device *bdev)
+{
+	return bdev->bd_disk->integrity;
+}
+
+static inline unsigned int bdev_get_tag_size(struct block_device *bdev)
+{
+	struct blk_integrity *bi = bdev_get_integrity(bdev);
+
+	if (bi)
+		return bi->tag_size;
+
+	return 0;
+}
+
+static inline int bdev_integrity_enabled(struct block_device *bdev, int rw)
+{
+	struct blk_integrity *bi = bdev_get_integrity(bdev);
+
+	if (bi == NULL)
+		return 0;
+
+	if (rw == READ && bi->verify_fn != NULL &&
+	    test_bit(INTEGRITY_FLAG_READ, &bi->flags))
+		return 1;
+
+	if (rw == WRITE && bi->generate_fn != NULL &&
+	    test_bit(INTEGRITY_FLAG_WRITE, &bi->flags))
+		return 1;
+
+	return 0;
+}
+
+static inline int blk_integrity_rq(struct request *rq)
+{
+	BUG_ON(rq->bio == NULL);
+
+	return bio_integrity(rq->bio);
+}
+
+#else /* CONFIG_BLK_DEV_INTEGRITY */
+
+#define blk_integrity_rq(rq)			(0)
+#define blk_rq_count_integrity_sg(a)		(0)
+#define blk_rq_map_integrity_sg(a, b)		(0)
+#define bdev_get_integrity(a)			(0)
+#define bdev_get_tag_size(a)			(0)
+#define blk_integrity_compare(a, b)		(0)
+#define blk_integrity_register(a, b)		(0)
+#define blk_integrity_unregister(a)		do { } while (0);
+
+#endif /* CONFIG_BLK_DEV_INTEGRITY */
+
+
 #else /* CONFIG_BLOCK */
 /*
  * stubs for when the block layer is configured out
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index ae7aec3cabee..524ec96f5a23 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -141,6 +141,9 @@ struct gendisk {
 	struct disk_stats dkstats;
 #endif
 	struct work_struct async_notify;
+#ifdef  CONFIG_BLK_DEV_INTEGRITY
+	struct blk_integrity *integrity;
+#endif
 };
 
 /* 
-- 
cgit v1.2.3


From da9cbc87395308a21465bd25441297bbba0477e1 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Mon, 30 Jun 2008 20:42:08 +0200
Subject: block: blkdev.h cleanup, move iocontext stuff to iocontext.h

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/blkdev.h    | 17 -----------------
 include/linux/iocontext.h | 18 ++++++++++++++++++
 kernel/exit.c             |  1 +
 kernel/fork.c             |  1 +
 4 files changed, 20 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 4a9ed45270ff..443df75d2cde 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -33,12 +33,6 @@ struct sg_io_hdr;
 #define BLKDEV_MIN_RQ	4
 #define BLKDEV_MAX_RQ	128	/* Default maximum */
 
-int put_io_context(struct io_context *ioc);
-void exit_io_context(void);
-struct io_context *get_io_context(gfp_t gfp_flags, int node);
-struct io_context *alloc_io_context(gfp_t gfp_flags, int node);
-void copy_io_context(struct io_context **pdst, struct io_context **psrc);
-
 struct request;
 typedef void (rq_end_io_fn)(struct request *, int);
 
@@ -981,17 +975,6 @@ static inline long nr_blockdev_pages(void)
 	return 0;
 }
 
-static inline void exit_io_context(void)
-{
-}
-
-struct io_context;
-static inline int put_io_context(struct io_context *ioc)
-{
-	return 1;
-}
-
-
 #endif /* CONFIG_BLOCK */
 
 #endif
diff --git a/include/linux/iocontext.h b/include/linux/iocontext.h
index 2b7a1187cb29..08b987bccf89 100644
--- a/include/linux/iocontext.h
+++ b/include/linux/iocontext.h
@@ -99,4 +99,22 @@ static inline struct io_context *ioc_task_link(struct io_context *ioc)
 	return NULL;
 }
 
+#ifdef CONFIG_BLOCK
+int put_io_context(struct io_context *ioc);
+void exit_io_context(void);
+struct io_context *get_io_context(gfp_t gfp_flags, int node);
+struct io_context *alloc_io_context(gfp_t gfp_flags, int node);
+void copy_io_context(struct io_context **pdst, struct io_context **psrc);
+#else
+static inline void exit_io_context(void)
+{
+}
+
+struct io_context;
+static inline int put_io_context(struct io_context *ioc)
+{
+	return 1;
+}
+#endif
+
 #endif
diff --git a/kernel/exit.c b/kernel/exit.c
index 8f6185e69b69..ceb258782835 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -13,6 +13,7 @@
 #include <linux/personality.h>
 #include <linux/tty.h>
 #include <linux/mnt_namespace.h>
+#include <linux/iocontext.h>
 #include <linux/key.h>
 #include <linux/security.h>
 #include <linux/cpu.h>
diff --git a/kernel/fork.c b/kernel/fork.c
index 19908b26cf80..b71ccd09fc8d 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -23,6 +23,7 @@
 #include <linux/sem.h>
 #include <linux/file.h>
 #include <linux/fdtable.h>
+#include <linux/iocontext.h>
 #include <linux/key.h>
 #include <linux/binfmts.h>
 #include <linux/mman.h>
-- 
cgit v1.2.3


From 6e2401ad6f33de15ff00f78b88159f00a14f3b35 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Wed, 18 Jun 2008 10:15:02 +0200
Subject: block: integrity cleanups

- No need to check for NULL bio, we'll get an immediate oops anyway.
- Make bio_integrity() a proper function.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/bio.h    | 9 ++++++++-
 include/linux/blkdev.h | 4 ----
 2 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/bio.h b/include/linux/bio.h
index 6bfc3e8d9d89..0933a14e6414 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -458,7 +458,14 @@ static inline char *__bio_kmap_irq(struct bio *bio, unsigned short idx,
 #define bip_for_each_vec(bvl, bip, i)					\
 	__bip_for_each_vec(bvl, bip, i, (bip)->bip_idx)
 
-#define bio_integrity(bio)	((bio)->bi_integrity ? 1 : 0)
+static inline int bio_integrity(struct bio *bio)
+{
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+	return bio->bi_integrity != NULL;
+#else
+	return 0;
+#endif
+}
 
 extern struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *, gfp_t, unsigned int, struct bio_set *);
 extern struct bio_integrity_payload *bio_integrity_alloc(struct bio *, gfp_t, unsigned int);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 443df75d2cde..d3ae9ad97213 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -860,7 +860,6 @@ void kblockd_flush_work(struct work_struct *work);
 #define MODULE_ALIAS_BLOCKDEV_MAJOR(major) \
 	MODULE_ALIAS("block-major-" __stringify(major) "-*")
 
-
 #if defined(CONFIG_BLK_DEV_INTEGRITY)
 
 #define INTEGRITY_FLAG_READ	1	/* verify data integrity on read */
@@ -945,8 +944,6 @@ static inline int bdev_integrity_enabled(struct block_device *bdev, int rw)
 
 static inline int blk_integrity_rq(struct request *rq)
 {
-	BUG_ON(rq->bio == NULL);
-
 	return bio_integrity(rq->bio);
 }
 
@@ -963,7 +960,6 @@ static inline int blk_integrity_rq(struct request *rq)
 
 #endif /* CONFIG_BLK_DEV_INTEGRITY */
 
-
 #else /* CONFIG_BLOCK */
 /*
  * stubs for when the block layer is configured out
-- 
cgit v1.2.3


From 0b07de85a76e1346e675f0e98437378932473df7 Mon Sep 17 00:00:00 2001
From: Adel Gadllah <adel.gadllah@gmail.com>
Date: Thu, 26 Jun 2008 13:48:27 +0200
Subject: allow userspace to modify scsi command filter on per device basis

This patch exports the per-gendisk command filter to user space through
sysfs, so it can be changed by the system administrator.
All users of the old cmd filter have been converted to use the new one.

Original patch from Peter Jones.

Signed-off-by: Adel Gadllah <adel.gadllah@gmail.com>
Signed-off-by: Peter Jones <pjones@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/Makefile         |   3 +-
 block/bsg.c            |  38 ++++--
 block/cmd-filter.c     | 325 +++++++++++++++++++++++++++++++++++++++++++++++++
 block/genhd.c          |   2 +
 block/scsi_ioctl.c     | 121 +-----------------
 drivers/scsi/sg.c      |  40 ++----
 include/linux/blkdev.h |  10 +-
 include/linux/genhd.h  |   9 ++
 8 files changed, 389 insertions(+), 159 deletions(-)
 create mode 100644 block/cmd-filter.c

(limited to 'include/linux')

diff --git a/block/Makefile b/block/Makefile
index 045f7b62e4bb..208000b0750d 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -4,7 +4,8 @@
 
 obj-$(CONFIG_BLOCK) := elevator.o blk-core.o blk-tag.o blk-sysfs.o \
 			blk-barrier.o blk-settings.o blk-ioc.o blk-map.o \
-			blk-exec.o blk-merge.o ioctl.o genhd.o scsi_ioctl.o
+			blk-exec.o blk-merge.o ioctl.o genhd.o scsi_ioctl.o \
+			cmd-filter.o
 
 obj-$(CONFIG_BLK_DEV_BSG)	+= bsg.o
 obj-$(CONFIG_IOSCHED_NOOP)	+= noop-iosched.o
diff --git a/block/bsg.c b/block/bsg.c
index f0b7cd343216..439940c3a1ff 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -44,11 +44,12 @@ struct bsg_device {
 	char name[BUS_ID_SIZE];
 	int max_queue;
 	unsigned long flags;
+	struct blk_scsi_cmd_filter *cmd_filter;
+	mode_t *f_mode;
 };
 
 enum {
 	BSG_F_BLOCK		= 1,
-	BSG_F_WRITE_PERM	= 2,
 };
 
 #define BSG_DEFAULT_CMDS	64
@@ -172,7 +173,7 @@ unlock:
 }
 
 static int blk_fill_sgv4_hdr_rq(struct request_queue *q, struct request *rq,
-				struct sg_io_v4 *hdr, int has_write_perm)
+				struct sg_io_v4 *hdr, struct bsg_device *bd)
 {
 	if (hdr->request_len > BLK_MAX_CDB) {
 		rq->cmd = kzalloc(hdr->request_len, GFP_KERNEL);
@@ -185,7 +186,8 @@ static int blk_fill_sgv4_hdr_rq(struct request_queue *q, struct request *rq,
 		return -EFAULT;
 
 	if (hdr->subprotocol == BSG_SUB_PROTOCOL_SCSI_CMD) {
-		if (blk_verify_command(rq->cmd, has_write_perm))
+		if (blk_cmd_filter_verify_command(bd->cmd_filter, rq->cmd,
+						 bd->f_mode))
 			return -EPERM;
 	} else if (!capable(CAP_SYS_RAWIO))
 		return -EPERM;
@@ -263,8 +265,7 @@ bsg_map_hdr(struct bsg_device *bd, struct sg_io_v4 *hdr)
 	rq = blk_get_request(q, rw, GFP_KERNEL);
 	if (!rq)
 		return ERR_PTR(-ENOMEM);
-	ret = blk_fill_sgv4_hdr_rq(q, rq, hdr, test_bit(BSG_F_WRITE_PERM,
-						       &bd->flags));
+	ret = blk_fill_sgv4_hdr_rq(q, rq, hdr, bd);
 	if (ret)
 		goto out;
 
@@ -566,12 +567,23 @@ static inline void bsg_set_block(struct bsg_device *bd, struct file *file)
 		set_bit(BSG_F_BLOCK, &bd->flags);
 }
 
-static inline void bsg_set_write_perm(struct bsg_device *bd, struct file *file)
+static void bsg_set_cmd_filter(struct bsg_device *bd,
+			   struct file *file)
 {
-	if (file->f_mode & FMODE_WRITE)
-		set_bit(BSG_F_WRITE_PERM, &bd->flags);
-	else
-		clear_bit(BSG_F_WRITE_PERM, &bd->flags);
+	struct inode *inode;
+	struct gendisk *disk;
+
+	if (!file)
+		return;
+
+	inode = file->f_dentry->d_inode;
+	if (!inode)
+		return;
+
+	disk = inode->i_bdev->bd_disk;
+
+	bd->cmd_filter = &disk->cmd_filter;
+	bd->f_mode = &file->f_mode;
 }
 
 /*
@@ -595,6 +607,8 @@ bsg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	dprintk("%s: read %Zd bytes\n", bd->name, count);
 
 	bsg_set_block(bd, file);
+	bsg_set_cmd_filter(bd, file);
+
 	bytes_read = 0;
 	ret = __bsg_read(buf, count, bd, NULL, &bytes_read);
 	*ppos = bytes_read;
@@ -668,7 +682,7 @@ bsg_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 	dprintk("%s: write %Zd bytes\n", bd->name, count);
 
 	bsg_set_block(bd, file);
-	bsg_set_write_perm(bd, file);
+	bsg_set_cmd_filter(bd, file);
 
 	bytes_written = 0;
 	ret = __bsg_write(bd, buf, count, &bytes_written);
@@ -771,7 +785,9 @@ static struct bsg_device *bsg_add_device(struct inode *inode,
 	}
 
 	bd->queue = rq;
+
 	bsg_set_block(bd, file);
+	bsg_set_cmd_filter(bd, file);
 
 	atomic_set(&bd->ref_count, 1);
 	mutex_lock(&bsg_mutex);
diff --git a/block/cmd-filter.c b/block/cmd-filter.c
new file mode 100644
index 000000000000..35e327ceaa97
--- /dev/null
+++ b/block/cmd-filter.c
@@ -0,0 +1,325 @@
+/*
+ * Copyright 2004 Peter M. Jones <pjones@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public Licens
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-
+ *
+ */
+
+#include <linux/list.h>
+#include <linux/genhd.h>
+#include <linux/spinlock.h>
+#include <linux/parser.h>
+#include <linux/capability.h>
+#include <linux/bitops.h>
+
+#include <scsi/scsi.h>
+#include <linux/cdrom.h>
+
+int blk_cmd_filter_verify_command(struct blk_scsi_cmd_filter *filter,
+				  unsigned char *cmd, mode_t *f_mode)
+{
+	/* root can do any command. */
+	if (capable(CAP_SYS_RAWIO))
+		return 0;
+
+	/* if there's no filter set, assume we're filtering everything out */
+	if (!filter)
+		return -EPERM;
+
+	/* Anybody who can open the device can do a read-safe command */
+	if (test_bit(cmd[0], filter->read_ok))
+		return 0;
+
+	/* Write-safe commands require a writable open */
+	if (test_bit(cmd[0], filter->write_ok) && (*f_mode & FMODE_WRITE))
+		return 0;
+
+	return -EPERM;
+}
+EXPORT_SYMBOL(blk_cmd_filter_verify_command);
+
+int blk_verify_command(struct file *file, unsigned char *cmd)
+{
+	struct gendisk *disk;
+	struct inode *inode;
+
+	if (!file)
+		return -EINVAL;
+
+	inode = file->f_dentry->d_inode;
+	if (!inode)
+		return -EINVAL;
+
+	disk = inode->i_bdev->bd_disk;
+
+	return blk_cmd_filter_verify_command(&disk->cmd_filter,
+						 cmd, &file->f_mode);
+}
+EXPORT_SYMBOL(blk_verify_command);
+
+/* and now, the sysfs stuff */
+static ssize_t rcf_cmds_show(struct blk_scsi_cmd_filter *filter, char *page,
+			     int rw)
+{
+	char *npage = page;
+	unsigned long *okbits;
+	int i;
+
+	if (rw == READ)
+		okbits = filter->read_ok;
+	else
+		okbits = filter->write_ok;
+
+	for (i = 0; i < BLK_SCSI_MAX_CMDS; i++) {
+		if (test_bit(i, okbits)) {
+			sprintf(npage, "%02x", i);
+			npage += 2;
+			if (i < BLK_SCSI_MAX_CMDS - 1)
+				sprintf(npage++, " ");
+		}
+	}
+
+	if (npage != page)
+		npage += sprintf(npage, "\n");
+
+	return npage - page;
+}
+
+static ssize_t rcf_readcmds_show(struct blk_scsi_cmd_filter *filter, char *page)
+{
+	return rcf_cmds_show(filter, page, READ);
+}
+
+static ssize_t rcf_writecmds_show(struct blk_scsi_cmd_filter *filter,
+				 char *page)
+{
+	return rcf_cmds_show(filter, page, WRITE);
+}
+
+static ssize_t rcf_cmds_store(struct blk_scsi_cmd_filter *filter,
+			      const char *page, size_t count, int rw)
+{
+	ssize_t ret = 0;
+	unsigned long okbits[BLK_SCSI_CMD_PER_LONG], *target_okbits;
+	int cmd, status, len;
+	substring_t ss;
+
+	memset(&okbits, 0, sizeof(okbits));
+
+	for (len = strlen(page); len > 0; len -= 3) {
+		if (len < 2)
+			break;
+		ss.from = (char *) page + ret;
+		ss.to = (char *) page + ret + 2;
+		ret += 3;
+		status = match_hex(&ss, &cmd);
+		/* either of these cases means invalid input, so do nothing. */
+		if (status || cmd >= BLK_SCSI_MAX_CMDS)
+			return -EINVAL;
+
+		__set_bit(cmd, okbits);
+	}
+
+	if (rw == READ)
+		target_okbits = filter->read_ok;
+	else
+		target_okbits = filter->write_ok;
+
+	memmove(target_okbits, okbits, sizeof(okbits));
+	return count;
+}
+
+static ssize_t rcf_readcmds_store(struct blk_scsi_cmd_filter *filter,
+				  const char *page, size_t count)
+{
+	return rcf_cmds_store(filter, page, count, READ);
+}
+
+static ssize_t rcf_writecmds_store(struct blk_scsi_cmd_filter *filter,
+				   const char *page, size_t count)
+{
+	return rcf_cmds_store(filter, page, count, WRITE);
+}
+
+struct rcf_sysfs_entry {
+	struct attribute attr;
+	ssize_t (*show)(struct blk_scsi_cmd_filter *, char *);
+	ssize_t (*store)(struct blk_scsi_cmd_filter *, const char *, size_t);
+};
+
+static struct rcf_sysfs_entry rcf_readcmds_entry = {
+	.attr = { .name = "read_table", .mode = S_IRUGO | S_IWUSR },
+	.show = rcf_readcmds_show,
+	.store = rcf_readcmds_store,
+};
+
+static struct rcf_sysfs_entry rcf_writecmds_entry = {
+	.attr = {.name = "write_table", .mode = S_IRUGO | S_IWUSR },
+	.show = rcf_writecmds_show,
+	.store = rcf_writecmds_store,
+};
+
+static struct attribute *default_attrs[] = {
+	&rcf_readcmds_entry.attr,
+	&rcf_writecmds_entry.attr,
+	NULL,
+};
+
+#define to_rcf(atr) container_of((atr), struct rcf_sysfs_entry, attr)
+
+static ssize_t
+rcf_attr_show(struct kobject *kobj, struct attribute *attr, char *page)
+{
+	struct rcf_sysfs_entry *entry = to_rcf(attr);
+	struct blk_scsi_cmd_filter *filter;
+
+	filter = container_of(kobj, struct blk_scsi_cmd_filter, kobj);
+	if (entry->show)
+		return entry->show(filter, page);
+
+	return 0;
+}
+
+static ssize_t
+rcf_attr_store(struct kobject *kobj, struct attribute *attr,
+			const char *page, size_t length)
+{
+	struct rcf_sysfs_entry *entry = to_rcf(attr);
+	struct blk_scsi_cmd_filter *filter;
+
+	if (!capable(CAP_SYS_RAWIO))
+		return -EPERM;
+
+	if (!entry->store)
+		return -EINVAL;
+
+	filter = container_of(kobj, struct blk_scsi_cmd_filter, kobj);
+	return entry->store(filter, page, length);
+}
+
+static struct sysfs_ops rcf_sysfs_ops = {
+	.show = rcf_attr_show,
+	.store = rcf_attr_store,
+};
+
+static struct kobj_type rcf_ktype = {
+	.sysfs_ops = &rcf_sysfs_ops,
+	.default_attrs = default_attrs,
+};
+
+static void rcf_set_defaults(struct blk_scsi_cmd_filter *filter)
+{
+	/* Basic read-only commands */
+	__set_bit(TEST_UNIT_READY, filter->read_ok);
+	__set_bit(REQUEST_SENSE, filter->read_ok);
+	__set_bit(READ_6, filter->read_ok);
+	__set_bit(READ_10, filter->read_ok);
+	__set_bit(READ_12, filter->read_ok);
+	__set_bit(READ_16, filter->read_ok);
+	__set_bit(READ_BUFFER, filter->read_ok);
+	__set_bit(READ_DEFECT_DATA, filter->read_ok);
+	__set_bit(READ_LONG, filter->read_ok);
+	__set_bit(INQUIRY, filter->read_ok);
+	__set_bit(MODE_SENSE, filter->read_ok);
+	__set_bit(MODE_SENSE_10, filter->read_ok);
+	__set_bit(LOG_SENSE, filter->read_ok);
+	__set_bit(START_STOP, filter->read_ok);
+	__set_bit(GPCMD_VERIFY_10, filter->read_ok);
+	__set_bit(VERIFY_16, filter->read_ok);
+	__set_bit(GPCMD_READ_BUFFER_CAPACITY, filter->read_ok);
+
+	/* Audio CD commands */
+	__set_bit(GPCMD_PLAY_CD, filter->read_ok);
+	__set_bit(GPCMD_PLAY_AUDIO_10, filter->read_ok);
+	__set_bit(GPCMD_PLAY_AUDIO_MSF, filter->read_ok);
+	__set_bit(GPCMD_PLAY_AUDIO_TI, filter->read_ok);
+	__set_bit(GPCMD_PAUSE_RESUME, filter->read_ok);
+
+	/* CD/DVD data reading */
+	__set_bit(GPCMD_READ_CD, filter->read_ok);
+	__set_bit(GPCMD_READ_CD_MSF, filter->read_ok);
+	__set_bit(GPCMD_READ_DISC_INFO, filter->read_ok);
+	__set_bit(GPCMD_READ_CDVD_CAPACITY, filter->read_ok);
+	__set_bit(GPCMD_READ_DVD_STRUCTURE, filter->read_ok);
+	__set_bit(GPCMD_READ_HEADER, filter->read_ok);
+	__set_bit(GPCMD_READ_TRACK_RZONE_INFO, filter->read_ok);
+	__set_bit(GPCMD_READ_SUBCHANNEL, filter->read_ok);
+	__set_bit(GPCMD_READ_TOC_PMA_ATIP, filter->read_ok);
+	__set_bit(GPCMD_REPORT_KEY, filter->read_ok);
+	__set_bit(GPCMD_SCAN, filter->read_ok);
+	__set_bit(GPCMD_GET_CONFIGURATION, filter->read_ok);
+	__set_bit(GPCMD_READ_FORMAT_CAPACITIES, filter->read_ok);
+	__set_bit(GPCMD_GET_EVENT_STATUS_NOTIFICATION, filter->read_ok);
+	__set_bit(GPCMD_GET_PERFORMANCE, filter->read_ok);
+	__set_bit(GPCMD_SEEK, filter->read_ok);
+	__set_bit(GPCMD_STOP_PLAY_SCAN, filter->read_ok);
+
+	/* Basic writing commands */
+	__set_bit(WRITE_6, filter->write_ok);
+	__set_bit(WRITE_10, filter->write_ok);
+	__set_bit(WRITE_VERIFY, filter->write_ok);
+	__set_bit(WRITE_12, filter->write_ok);
+	__set_bit(WRITE_VERIFY_12, filter->write_ok);
+	__set_bit(WRITE_16, filter->write_ok);
+	__set_bit(WRITE_LONG, filter->write_ok);
+	__set_bit(WRITE_LONG_2, filter->write_ok);
+	__set_bit(ERASE, filter->write_ok);
+	__set_bit(GPCMD_MODE_SELECT_10, filter->write_ok);
+	__set_bit(MODE_SELECT, filter->write_ok);
+	__set_bit(LOG_SELECT, filter->write_ok);
+	__set_bit(GPCMD_BLANK, filter->write_ok);
+	__set_bit(GPCMD_CLOSE_TRACK, filter->write_ok);
+	__set_bit(GPCMD_FLUSH_CACHE, filter->write_ok);
+	__set_bit(GPCMD_FORMAT_UNIT, filter->write_ok);
+	__set_bit(GPCMD_REPAIR_RZONE_TRACK, filter->write_ok);
+	__set_bit(GPCMD_RESERVE_RZONE_TRACK, filter->write_ok);
+	__set_bit(GPCMD_SEND_DVD_STRUCTURE, filter->write_ok);
+	__set_bit(GPCMD_SEND_EVENT, filter->write_ok);
+	__set_bit(GPCMD_SEND_KEY, filter->write_ok);
+	__set_bit(GPCMD_SEND_OPC, filter->write_ok);
+	__set_bit(GPCMD_SEND_CUE_SHEET, filter->write_ok);
+	__set_bit(GPCMD_SET_SPEED, filter->write_ok);
+	__set_bit(GPCMD_PREVENT_ALLOW_MEDIUM_REMOVAL, filter->write_ok);
+	__set_bit(GPCMD_LOAD_UNLOAD, filter->write_ok);
+	__set_bit(GPCMD_SET_STREAMING, filter->write_ok);
+}
+
+int blk_register_filter(struct gendisk *disk)
+{
+	int ret;
+	struct blk_scsi_cmd_filter *filter = &disk->cmd_filter;
+	struct kobject *parent = kobject_get(disk->holder_dir->parent);
+
+	if (!parent)
+		return -ENODEV;
+
+	ret = kobject_init_and_add(&filter->kobj, &rcf_ktype, parent,
+				 "%s", "cmd_filter");
+
+	if (ret < 0)
+		return ret;
+
+	rcf_set_defaults(filter);
+	return 0;
+}
+
+void blk_unregister_filter(struct gendisk *disk)
+{
+	struct blk_scsi_cmd_filter *filter = &disk->cmd_filter;
+
+	kobject_put(&filter->kobj);
+	kobject_put(disk->holder_dir->parent);
+}
+
diff --git a/block/genhd.c b/block/genhd.c
index 43e468ee5993..9074f384b097 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -189,6 +189,7 @@ void add_disk(struct gendisk *disk)
 			    disk->minors, NULL, exact_match, exact_lock, disk);
 	register_disk(disk);
 	blk_register_queue(disk);
+	blk_register_filter(disk);
 
 	bdi = &disk->queue->backing_dev_info;
 	bdi_register_dev(bdi, MKDEV(disk->major, disk->first_minor));
@@ -200,6 +201,7 @@ EXPORT_SYMBOL(del_gendisk);	/* in partitions/check.c */
 
 void unlink_gendisk(struct gendisk *disk)
 {
+	blk_unregister_filter(disk);
 	sysfs_remove_link(&disk->dev.kobj, "bdi");
 	bdi_unregister(&disk->queue->backing_dev_info);
 	blk_unregister_queue(disk);
diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c
index 78199c08ec92..c5b9bcfc0a6d 100644
--- a/block/scsi_ioctl.c
+++ b/block/scsi_ioctl.c
@@ -105,120 +105,12 @@ static int sg_emulated_host(struct request_queue *q, int __user *p)
 	return put_user(1, p);
 }
 
-#define CMD_READ_SAFE	0x01
-#define CMD_WRITE_SAFE	0x02
-#define CMD_WARNED	0x04
-#define safe_for_read(cmd)	[cmd] = CMD_READ_SAFE
-#define safe_for_write(cmd)	[cmd] = CMD_WRITE_SAFE
-
-int blk_verify_command(unsigned char *cmd, int has_write_perm)
-{
-	static unsigned char cmd_type[256] = {
-
-		/* Basic read-only commands */
-		safe_for_read(TEST_UNIT_READY),
-		safe_for_read(REQUEST_SENSE),
-		safe_for_read(READ_6),
-		safe_for_read(READ_10),
-		safe_for_read(READ_12),
-		safe_for_read(READ_16),
-		safe_for_read(READ_BUFFER),
-		safe_for_read(READ_DEFECT_DATA),
-		safe_for_read(READ_LONG),
-		safe_for_read(INQUIRY),
-		safe_for_read(MODE_SENSE),
-		safe_for_read(MODE_SENSE_10),
-		safe_for_read(LOG_SENSE),
-		safe_for_read(START_STOP),
-		safe_for_read(GPCMD_VERIFY_10),
-		safe_for_read(VERIFY_16),
-
-		/* Audio CD commands */
-		safe_for_read(GPCMD_PLAY_CD),
-		safe_for_read(GPCMD_PLAY_AUDIO_10),
-		safe_for_read(GPCMD_PLAY_AUDIO_MSF),
-		safe_for_read(GPCMD_PLAY_AUDIO_TI),
-		safe_for_read(GPCMD_PAUSE_RESUME),
-
-		/* CD/DVD data reading */
-		safe_for_read(GPCMD_READ_BUFFER_CAPACITY),
-		safe_for_read(GPCMD_READ_CD),
-		safe_for_read(GPCMD_READ_CD_MSF),
-		safe_for_read(GPCMD_READ_DISC_INFO),
-		safe_for_read(GPCMD_READ_CDVD_CAPACITY),
-		safe_for_read(GPCMD_READ_DVD_STRUCTURE),
-		safe_for_read(GPCMD_READ_HEADER),
-		safe_for_read(GPCMD_READ_TRACK_RZONE_INFO),
-		safe_for_read(GPCMD_READ_SUBCHANNEL),
-		safe_for_read(GPCMD_READ_TOC_PMA_ATIP),
-		safe_for_read(GPCMD_REPORT_KEY),
-		safe_for_read(GPCMD_SCAN),
-		safe_for_read(GPCMD_GET_CONFIGURATION),
-		safe_for_read(GPCMD_READ_FORMAT_CAPACITIES),
-		safe_for_read(GPCMD_GET_EVENT_STATUS_NOTIFICATION),
-		safe_for_read(GPCMD_GET_PERFORMANCE),
-		safe_for_read(GPCMD_SEEK),
-		safe_for_read(GPCMD_STOP_PLAY_SCAN),
-
-		/* Basic writing commands */
-		safe_for_write(WRITE_6),
-		safe_for_write(WRITE_10),
-		safe_for_write(WRITE_VERIFY),
-		safe_for_write(WRITE_12),
-		safe_for_write(WRITE_VERIFY_12),
-		safe_for_write(WRITE_16),
-		safe_for_write(WRITE_LONG),
-		safe_for_write(WRITE_LONG_2),
-		safe_for_write(ERASE),
-		safe_for_write(GPCMD_MODE_SELECT_10),
-		safe_for_write(MODE_SELECT),
-		safe_for_write(LOG_SELECT),
-		safe_for_write(GPCMD_BLANK),
-		safe_for_write(GPCMD_CLOSE_TRACK),
-		safe_for_write(GPCMD_FLUSH_CACHE),
-		safe_for_write(GPCMD_FORMAT_UNIT),
-		safe_for_write(GPCMD_REPAIR_RZONE_TRACK),
-		safe_for_write(GPCMD_RESERVE_RZONE_TRACK),
-		safe_for_write(GPCMD_SEND_DVD_STRUCTURE),
-		safe_for_write(GPCMD_SEND_EVENT),
-		safe_for_write(GPCMD_SEND_KEY),
-		safe_for_write(GPCMD_SEND_OPC),
-		safe_for_write(GPCMD_SEND_CUE_SHEET),
-		safe_for_write(GPCMD_SET_SPEED),
-		safe_for_write(GPCMD_PREVENT_ALLOW_MEDIUM_REMOVAL),
-		safe_for_write(GPCMD_LOAD_UNLOAD),
-		safe_for_write(GPCMD_SET_STREAMING),
-	};
-	unsigned char type = cmd_type[cmd[0]];
-
-	/* Anybody who can open the device can do a read-safe command */
-	if (type & CMD_READ_SAFE)
-		return 0;
-
-	/* Write-safe commands just require a writable open.. */
-	if ((type & CMD_WRITE_SAFE) && has_write_perm)
-		return 0;
-
-	/* And root can do any command.. */
-	if (capable(CAP_SYS_RAWIO))
-		return 0;
-
-	if (!type) {
-		cmd_type[cmd[0]] = CMD_WARNED;
-		printk(KERN_WARNING "scsi: unknown opcode 0x%02x\n", cmd[0]);
-	}
-
-	/* Otherwise fail it with an "Operation not permitted" */
-	return -EPERM;
-}
-EXPORT_SYMBOL_GPL(blk_verify_command);
-
 static int blk_fill_sghdr_rq(struct request_queue *q, struct request *rq,
-			     struct sg_io_hdr *hdr, int has_write_perm)
+			     struct sg_io_hdr *hdr, struct file *file)
 {
 	if (copy_from_user(rq->cmd, hdr->cmdp, hdr->cmd_len))
 		return -EFAULT;
-	if (blk_verify_command(rq->cmd, has_write_perm))
+	if (blk_verify_command(file, rq->cmd))
 		return -EPERM;
 
 	/*
@@ -287,7 +179,7 @@ static int sg_io(struct file *file, struct request_queue *q,
 		struct gendisk *bd_disk, struct sg_io_hdr *hdr)
 {
 	unsigned long start_time;
-	int writing = 0, ret = 0, has_write_perm = 0;
+	int writing = 0, ret = 0;
 	struct request *rq;
 	char sense[SCSI_SENSE_BUFFERSIZE];
 	struct bio *bio;
@@ -316,10 +208,7 @@ static int sg_io(struct file *file, struct request_queue *q,
 	if (!rq)
 		return -ENOMEM;
 
-	if (file)
-		has_write_perm = file->f_mode & FMODE_WRITE;
-
-	if (blk_fill_sghdr_rq(q, rq, hdr, has_write_perm)) {
+	if (blk_fill_sghdr_rq(q, rq, hdr, file)) {
 		blk_put_request(rq);
 		return -EFAULT;
 	}
@@ -451,7 +340,7 @@ int sg_scsi_ioctl(struct file *file, struct request_queue *q,
 	if (in_len && copy_from_user(buffer, sic->data + cmdlen, in_len))
 		goto error;
 
-	err = blk_verify_command(rq->cmd, file->f_mode & FMODE_WRITE);
+	err = blk_verify_command(file, rq->cmd);
 	if (err)
 		goto error;
 
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index ea0edd1b2e76..f7abccaffaec 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -182,8 +182,9 @@ static int sg_build_sgat(Sg_scatter_hold * schp, const Sg_fd * sfp,
 			 int tablesize);
 static ssize_t sg_new_read(Sg_fd * sfp, char __user *buf, size_t count,
 			   Sg_request * srp);
-static ssize_t sg_new_write(Sg_fd * sfp, const char __user *buf, size_t count,
-			    int blocking, int read_only, Sg_request ** o_srp);
+static ssize_t sg_new_write(Sg_fd *sfp, struct file *file,
+			const char __user *buf, size_t count, int blocking,
+			int read_only, Sg_request **o_srp);
 static int sg_common_write(Sg_fd * sfp, Sg_request * srp,
 			   unsigned char *cmnd, int timeout, int blocking);
 static int sg_u_iovec(sg_io_hdr_t * hp, int sg_num, int ind,
@@ -204,7 +205,6 @@ static Sg_request *sg_get_rq_mark(Sg_fd * sfp, int pack_id);
 static Sg_request *sg_add_request(Sg_fd * sfp);
 static int sg_remove_request(Sg_fd * sfp, Sg_request * srp);
 static int sg_res_in_use(Sg_fd * sfp);
-static int sg_allow_access(unsigned char opcode, char dev_type);
 static int sg_build_direct(Sg_request * srp, Sg_fd * sfp, int dxfer_len);
 static Sg_device *sg_get_dev(int dev);
 #ifdef CONFIG_SCSI_PROC_FS
@@ -544,7 +544,7 @@ sg_write(struct file *filp, const char __user *buf, size_t count, loff_t * ppos)
 		return -EFAULT;
 	blocking = !(filp->f_flags & O_NONBLOCK);
 	if (old_hdr.reply_len < 0)
-		return sg_new_write(sfp, buf, count, blocking, 0, NULL);
+		return sg_new_write(sfp, filp, buf, count, blocking, 0, NULL);
 	if (count < (SZ_SG_HEADER + 6))
 		return -EIO;	/* The minimum scsi command length is 6 bytes. */
 
@@ -621,8 +621,9 @@ sg_write(struct file *filp, const char __user *buf, size_t count, loff_t * ppos)
 }
 
 static ssize_t
-sg_new_write(Sg_fd * sfp, const char __user *buf, size_t count,
-	     int blocking, int read_only, Sg_request ** o_srp)
+sg_new_write(Sg_fd *sfp, struct file *file, const char __user *buf,
+		 size_t count, int blocking, int read_only,
+		 Sg_request **o_srp)
 {
 	int k;
 	Sg_request *srp;
@@ -678,8 +679,7 @@ sg_new_write(Sg_fd * sfp, const char __user *buf, size_t count,
 		sg_remove_request(sfp, srp);
 		return -EFAULT;
 	}
-	if (read_only &&
-	    (!sg_allow_access(cmnd[0], sfp->parentdp->device->type))) {
+	if (read_only && (!blk_verify_command(file, cmnd))) {
 		sg_remove_request(sfp, srp);
 		return -EPERM;
 	}
@@ -799,7 +799,7 @@ sg_ioctl(struct inode *inode, struct file *filp,
 			if (!access_ok(VERIFY_WRITE, p, SZ_SG_IO_HDR))
 				return -EFAULT;
 			result =
-			    sg_new_write(sfp, p, SZ_SG_IO_HDR,
+			    sg_new_write(sfp, filp, p, SZ_SG_IO_HDR,
 					 blocking, read_only, &srp);
 			if (result < 0)
 				return result;
@@ -1048,7 +1048,7 @@ sg_ioctl(struct inode *inode, struct file *filp,
 
 			if (copy_from_user(&opcode, siocp->data, 1))
 				return -EFAULT;
-			if (!sg_allow_access(opcode, sdp->device->type))
+			if (!blk_verify_command(filp, &opcode))
 				return -EPERM;
 		}
 		return sg_scsi_ioctl(filp, sdp->device->request_queue, NULL, p);
@@ -2506,26 +2506,6 @@ sg_page_free(struct page *page, int size)
 #define MAINTENANCE_IN_CMD 0xa3
 #endif
 
-static unsigned char allow_ops[] = { TEST_UNIT_READY, REQUEST_SENSE,
-	INQUIRY, READ_CAPACITY, READ_BUFFER, READ_6, READ_10, READ_12,
-	READ_16, MODE_SENSE, MODE_SENSE_10, LOG_SENSE, REPORT_LUNS,
-	SERVICE_ACTION_IN, RECEIVE_DIAGNOSTIC, READ_LONG, MAINTENANCE_IN_CMD
-};
-
-static int
-sg_allow_access(unsigned char opcode, char dev_type)
-{
-	int k;
-
-	if (TYPE_SCANNER == dev_type)	/* TYPE_ROM maybe burner */
-		return 1;
-	for (k = 0; k < sizeof (allow_ops); ++k) {
-		if (opcode == allow_ops[k])
-			return 1;
-	}
-	return 0;
-}
-
 #ifdef CONFIG_SCSI_PROC_FS
 static int
 sg_idr_max_id(int id, void *p, void *data)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d3ae9ad97213..a842b776d099 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -671,7 +671,6 @@ extern int blk_execute_rq(struct request_queue *, struct gendisk *,
 			  struct request *, int);
 extern void blk_execute_rq_nowait(struct request_queue *, struct gendisk *,
 				  struct request *, int, rq_end_io_fn *);
-extern int blk_verify_command(unsigned char *, int);
 extern void blk_unplug(struct request_queue *q);
 
 static inline struct request_queue *bdev_get_queue(struct block_device *bdev)
@@ -797,6 +796,15 @@ static inline struct request *blk_map_queue_find_tag(struct blk_queue_tag *bqt,
 
 extern int blkdev_issue_flush(struct block_device *, sector_t *);
 
+/*
+* command filter functions
+*/
+extern int blk_verify_command(struct file *file, unsigned char *cmd);
+extern int blk_cmd_filter_verify_command(struct blk_scsi_cmd_filter *filter,
+					 unsigned char *cmd, mode_t *f_mode);
+extern int blk_register_filter(struct gendisk *disk);
+extern void blk_unregister_filter(struct gendisk *disk);
+
 #define MAX_PHYS_SEGMENTS 128
 #define MAX_HW_SEGMENTS 128
 #define SAFE_MAX_SECTORS 255
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 524ec96f5a23..e8787417f65a 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -110,6 +110,14 @@ struct hd_struct {
 #define GENHD_FL_SUPPRESS_PARTITION_INFO	32
 #define GENHD_FL_FAIL				64
 
+#define BLK_SCSI_MAX_CMDS	(256)
+#define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))
+
+struct blk_scsi_cmd_filter {
+	unsigned long read_ok[BLK_SCSI_CMD_PER_LONG];
+	unsigned long write_ok[BLK_SCSI_CMD_PER_LONG];
+	struct kobject kobj;
+};
 
 struct gendisk {
 	int major;			/* major number of driver */
@@ -120,6 +128,7 @@ struct gendisk {
 	struct hd_struct **part;	/* [indexed by minor] */
 	struct block_device_operations *fops;
 	struct request_queue *queue;
+	struct blk_scsi_cmd_filter cmd_filter;
 	void *private_data;
 	sector_t capacity;
 
-- 
cgit v1.2.3


From b24498d477a14680fc3bb3ad884fa9fa76a2d237 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 27 Jun 2008 09:12:09 +0200
Subject: block: integrity flags can't use bit ops on unsigned short

Just use normal open coded bit operations instead, they need not be
atomic.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-integrity.c  | 17 +++++++----------
 include/linux/blkdev.h |  8 ++++----
 2 files changed, 11 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-integrity.c b/block/blk-integrity.c
index 4ffa3814f6a9..3f1a8478cc38 100644
--- a/block/blk-integrity.c
+++ b/block/blk-integrity.c
@@ -217,17 +217,16 @@ static ssize_t integrity_read_store(struct blk_integrity *bi,
 	unsigned long val = simple_strtoul(p, &p, 10);
 
 	if (val)
-		set_bit(INTEGRITY_FLAG_READ, &bi->flags);
+		bi->flags |= INTEGRITY_FLAG_READ;
 	else
-		clear_bit(INTEGRITY_FLAG_READ, &bi->flags);
+		bi->flags &= ~INTEGRITY_FLAG_READ;
 
 	return count;
 }
 
 static ssize_t integrity_read_show(struct blk_integrity *bi, char *page)
 {
-	return sprintf(page, "%d\n",
-		       test_bit(INTEGRITY_FLAG_READ, &bi->flags) ? 1 : 0);
+	return sprintf(page, "%d\n", (bi->flags & INTEGRITY_FLAG_READ) != 0);
 }
 
 static ssize_t integrity_write_store(struct blk_integrity *bi,
@@ -237,17 +236,16 @@ static ssize_t integrity_write_store(struct blk_integrity *bi,
 	unsigned long val = simple_strtoul(p, &p, 10);
 
 	if (val)
-		set_bit(INTEGRITY_FLAG_WRITE, &bi->flags);
+		bi->flags |= INTEGRITY_FLAG_WRITE;
 	else
-		clear_bit(INTEGRITY_FLAG_WRITE, &bi->flags);
+		bi->flags &= ~INTEGRITY_FLAG_WRITE;
 
 	return count;
 }
 
 static ssize_t integrity_write_show(struct blk_integrity *bi, char *page)
 {
-	return sprintf(page, "%d\n",
-		       test_bit(INTEGRITY_FLAG_WRITE, &bi->flags) ? 1 : 0);
+	return sprintf(page, "%d\n", (bi->flags & INTEGRITY_FLAG_WRITE) != 0);
 }
 
 static struct integrity_sysfs_entry integrity_format_entry = {
@@ -340,8 +338,7 @@ int blk_integrity_register(struct gendisk *disk, struct blk_integrity *template)
 
 		kobject_uevent(&bi->kobj, KOBJ_ADD);
 
-		set_bit(INTEGRITY_FLAG_READ, &bi->flags);
-		set_bit(INTEGRITY_FLAG_WRITE, &bi->flags);
+		bi->flags |= INTEGRITY_FLAG_READ | INTEGRITY_FLAG_WRITE;
 		bi->sector_size = disk->queue->hardsect_size;
 		disk->integrity = bi;
 	} else
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index a842b776d099..7ab8acad5b6e 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -870,8 +870,8 @@ void kblockd_flush_work(struct work_struct *work);
 
 #if defined(CONFIG_BLK_DEV_INTEGRITY)
 
-#define INTEGRITY_FLAG_READ	1	/* verify data integrity on read */
-#define INTEGRITY_FLAG_WRITE	2	/* generate data integrity on write */
+#define INTEGRITY_FLAG_READ	2	/* verify data integrity on read */
+#define INTEGRITY_FLAG_WRITE	4	/* generate data integrity on write */
 
 struct blk_integrity_exchg {
 	void			*prot_buf;
@@ -940,11 +940,11 @@ static inline int bdev_integrity_enabled(struct block_device *bdev, int rw)
 		return 0;
 
 	if (rw == READ && bi->verify_fn != NULL &&
-	    test_bit(INTEGRITY_FLAG_READ, &bi->flags))
+	    (bi->flags & INTEGRITY_FLAG_READ))
 		return 1;
 
 	if (rw == WRITE && bi->generate_fn != NULL &&
-	    test_bit(INTEGRITY_FLAG_WRITE, &bi->flags))
+	    (bi->flags & INTEGRITY_FLAG_WRITE))
 		return 1;
 
 	return 0;
-- 
cgit v1.2.3


From cc371e66e340f35eed8dc4651c7c18e754c7fb26 Mon Sep 17 00:00:00 2001
From: Alasdair G Kergon <agk@redhat.com>
Date: Thu, 3 Jul 2008 09:53:43 +0200
Subject: Add bvec_merge_data to handle stacked devices and ->merge_bvec()

When devices are stacked, one device's merge_bvec_fn may need to perform
the mapping and then call one or more functions for its underlying devices.

The following bio fields are used:
  bio->bi_sector
  bio->bi_bdev
  bio->bi_size
  bio->bi_rw  using bio_data_dir()

This patch creates a new struct bvec_merge_data holding a copy of those
fields to avoid having to change them directly in the struct bio when
going down the stack only to have to change them back again on the way
back up.  (And then when the bio gets mapped for real, the whole
exercise gets repeated, but that's a problem for another day...)

Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Milan Broz <mbroz@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 drivers/block/pktcdvd.c |  9 +++++----
 drivers/md/linear.c     | 10 ++++++----
 drivers/md/raid0.c      | 10 ++++++----
 drivers/md/raid10.c     | 15 ++++++++-------
 drivers/md/raid5.c      | 10 ++++++----
 fs/bio.c                | 26 +++++++++++++++++++++-----
 include/linux/blkdev.h  |  9 ++++++++-
 7 files changed, 60 insertions(+), 29 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 3ba1df93e9e3..589850cff359 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -2633,11 +2633,12 @@ end_io:
 
 
 
-static int pkt_merge_bvec(struct request_queue *q, struct bio *bio, struct bio_vec *bvec)
+static int pkt_merge_bvec(struct request_queue *q, struct bvec_merge_data *bmd,
+			  struct bio_vec *bvec)
 {
 	struct pktcdvd_device *pd = q->queuedata;
-	sector_t zone = ZONE(bio->bi_sector, pd);
-	int used = ((bio->bi_sector - zone) << 9) + bio->bi_size;
+	sector_t zone = ZONE(bmd->bi_sector, pd);
+	int used = ((bmd->bi_sector - zone) << 9) + bmd->bi_size;
 	int remaining = (pd->settings.size << 9) - used;
 	int remaining2;
 
@@ -2645,7 +2646,7 @@ static int pkt_merge_bvec(struct request_queue *q, struct bio *bio, struct bio_v
 	 * A bio <= PAGE_SIZE must be allowed. If it crosses a packet
 	 * boundary, pkt_make_request() will split the bio.
 	 */
-	remaining2 = PAGE_SIZE - bio->bi_size;
+	remaining2 = PAGE_SIZE - bmd->bi_size;
 	remaining = max(remaining, remaining2);
 
 	BUG_ON(remaining < 0);
diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 10748240cb2f..6a866d7c8ae5 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -50,17 +50,19 @@ static inline dev_info_t *which_dev(mddev_t *mddev, sector_t sector)
 /**
  *	linear_mergeable_bvec -- tell bio layer if two requests can be merged
  *	@q: request queue
- *	@bio: the buffer head that's been built up so far
+ *	@bvm: properties of new bio
  *	@biovec: the request that could be merged to it.
  *
  *	Return amount of bytes we can take at this offset
  */
-static int linear_mergeable_bvec(struct request_queue *q, struct bio *bio, struct bio_vec *biovec)
+static int linear_mergeable_bvec(struct request_queue *q,
+				 struct bvec_merge_data *bvm,
+				 struct bio_vec *biovec)
 {
 	mddev_t *mddev = q->queuedata;
 	dev_info_t *dev0;
-	unsigned long maxsectors, bio_sectors = bio->bi_size >> 9;
-	sector_t sector = bio->bi_sector + get_start_sect(bio->bi_bdev);
+	unsigned long maxsectors, bio_sectors = bvm->bi_size >> 9;
+	sector_t sector = bvm->bi_sector + get_start_sect(bvm->bi_bdev);
 
 	dev0 = which_dev(mddev, sector);
 	maxsectors = (dev0->size << 1) - (sector - (dev0->offset<<1));
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 914c04ddec7c..bcbb82594a19 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -241,18 +241,20 @@ static int create_strip_zones (mddev_t *mddev)
 /**
  *	raid0_mergeable_bvec -- tell bio layer if a two requests can be merged
  *	@q: request queue
- *	@bio: the buffer head that's been built up so far
+ *	@bvm: properties of new bio
  *	@biovec: the request that could be merged to it.
  *
  *	Return amount of bytes we can accept at this offset
  */
-static int raid0_mergeable_bvec(struct request_queue *q, struct bio *bio, struct bio_vec *biovec)
+static int raid0_mergeable_bvec(struct request_queue *q,
+				struct bvec_merge_data *bvm,
+				struct bio_vec *biovec)
 {
 	mddev_t *mddev = q->queuedata;
-	sector_t sector = bio->bi_sector + get_start_sect(bio->bi_bdev);
+	sector_t sector = bvm->bi_sector + get_start_sect(bvm->bi_bdev);
 	int max;
 	unsigned int chunk_sectors = mddev->chunk_size >> 9;
-	unsigned int bio_sectors = bio->bi_size >> 9;
+	unsigned int bio_sectors = bvm->bi_size >> 9;
 
 	max =  (chunk_sectors - ((sector & (chunk_sectors - 1)) + bio_sectors)) << 9;
 	if (max < 0) max = 0; /* bio_add cannot handle a negative return */
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index a71277b640ab..22bb2b1b886d 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -439,26 +439,27 @@ static sector_t raid10_find_virt(conf_t *conf, sector_t sector, int dev)
 /**
  *	raid10_mergeable_bvec -- tell bio layer if a two requests can be merged
  *	@q: request queue
- *	@bio: the buffer head that's been built up so far
+ *	@bvm: properties of new bio
  *	@biovec: the request that could be merged to it.
  *
  *	Return amount of bytes we can accept at this offset
  *      If near_copies == raid_disk, there are no striping issues,
  *      but in that case, the function isn't called at all.
  */
-static int raid10_mergeable_bvec(struct request_queue *q, struct bio *bio,
-				struct bio_vec *bio_vec)
+static int raid10_mergeable_bvec(struct request_queue *q,
+				 struct bvec_merge_data *bvm,
+				 struct bio_vec *biovec)
 {
 	mddev_t *mddev = q->queuedata;
-	sector_t sector = bio->bi_sector + get_start_sect(bio->bi_bdev);
+	sector_t sector = bvm->bi_sector + get_start_sect(bvm->bi_bdev);
 	int max;
 	unsigned int chunk_sectors = mddev->chunk_size >> 9;
-	unsigned int bio_sectors = bio->bi_size >> 9;
+	unsigned int bio_sectors = bvm->bi_size >> 9;
 
 	max =  (chunk_sectors - ((sector & (chunk_sectors - 1)) + bio_sectors)) << 9;
 	if (max < 0) max = 0; /* bio_add cannot handle a negative return */
-	if (max <= bio_vec->bv_len && bio_sectors == 0)
-		return bio_vec->bv_len;
+	if (max <= biovec->bv_len && bio_sectors == 0)
+		return biovec->bv_len;
 	else
 		return max;
 }
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 54c8ee28fcc4..9b00675dc64f 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3319,15 +3319,17 @@ static int raid5_congested(void *data, int bits)
 /* We want read requests to align with chunks where possible,
  * but write requests don't need to.
  */
-static int raid5_mergeable_bvec(struct request_queue *q, struct bio *bio, struct bio_vec *biovec)
+static int raid5_mergeable_bvec(struct request_queue *q,
+				struct bvec_merge_data *bvm,
+				struct bio_vec *biovec)
 {
 	mddev_t *mddev = q->queuedata;
-	sector_t sector = bio->bi_sector + get_start_sect(bio->bi_bdev);
+	sector_t sector = bvm->bi_sector + get_start_sect(bvm->bi_bdev);
 	int max;
 	unsigned int chunk_sectors = mddev->chunk_size >> 9;
-	unsigned int bio_sectors = bio->bi_size >> 9;
+	unsigned int bio_sectors = bvm->bi_size >> 9;
 
-	if (bio_data_dir(bio) == WRITE)
+	if ((bvm->bi_rw & 1) == WRITE)
 		return biovec->bv_len; /* always allow writes to be mergeable */
 
 	max =  (chunk_sectors - ((sector & (chunk_sectors - 1)) + bio_sectors)) << 9;
diff --git a/fs/bio.c b/fs/bio.c
index 7761c84c7032..88322b066acb 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -325,10 +325,19 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 		if (page == prev->bv_page &&
 		    offset == prev->bv_offset + prev->bv_len) {
 			prev->bv_len += len;
-			if (q->merge_bvec_fn &&
-			    q->merge_bvec_fn(q, bio, prev) < len) {
-				prev->bv_len -= len;
-				return 0;
+
+			if (q->merge_bvec_fn) {
+				struct bvec_merge_data bvm = {
+					.bi_bdev = bio->bi_bdev,
+					.bi_sector = bio->bi_sector,
+					.bi_size = bio->bi_size,
+					.bi_rw = bio->bi_rw,
+				};
+
+				if (q->merge_bvec_fn(q, &bvm, prev) < len) {
+					prev->bv_len -= len;
+					return 0;
+				}
 			}
 
 			goto done;
@@ -369,11 +378,18 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 	 * queue to get further control
 	 */
 	if (q->merge_bvec_fn) {
+		struct bvec_merge_data bvm = {
+			.bi_bdev = bio->bi_bdev,
+			.bi_sector = bio->bi_sector,
+			.bi_size = bio->bi_size,
+			.bi_rw = bio->bi_rw,
+		};
+
 		/*
 		 * merge_bvec_fn() returns number of bytes it can accept
 		 * at this offset
 		 */
-		if (q->merge_bvec_fn(q, bio, bvec) < len) {
+		if (q->merge_bvec_fn(q, &bvm, bvec) < len) {
 			bvec->bv_page = NULL;
 			bvec->bv_len = 0;
 			bvec->bv_offset = 0;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 7ab8acad5b6e..ff9d0bdf2a16 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -254,7 +254,14 @@ typedef int (prep_rq_fn) (struct request_queue *, struct request *);
 typedef void (unplug_fn) (struct request_queue *);
 
 struct bio_vec;
-typedef int (merge_bvec_fn) (struct request_queue *, struct bio *, struct bio_vec *);
+struct bvec_merge_data {
+	struct block_device *bi_bdev;
+	sector_t bi_sector;
+	unsigned bi_size;
+	unsigned long bi_rw;
+};
+typedef int (merge_bvec_fn) (struct request_queue *, struct bvec_merge_data *,
+			     struct bio_vec *);
 typedef void (prepare_flush_fn) (struct request_queue *, struct request *);
 typedef void (softirq_done_fn)(struct request *);
 typedef int (dma_drain_needed_fn)(struct request *);
-- 
cgit v1.2.3


From e48ec69005f02b70b7ecfde1bc39a599086d16ef Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 3 Jul 2008 13:18:54 +0200
Subject: block: extend queue_flag bitops

Add test_and_clear and test_and_set.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-core.c       | 12 ++++--------
 include/linux/blkdev.h | 26 ++++++++++++++++++++++++++
 2 files changed, 30 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index e0fb0bcc0c17..dbc7f42b5d2b 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -205,8 +205,7 @@ void blk_plug_device(struct request_queue *q)
 	if (blk_queue_stopped(q))
 		return;
 
-	if (!test_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags)) {
-		__set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags);
+	if (!queue_flag_test_and_set(QUEUE_FLAG_PLUGGED, q)) {
 		mod_timer(&q->unplug_timer, jiffies + q->unplug_delay);
 		blk_add_trace_generic(q, NULL, 0, BLK_TA_PLUG);
 	}
@@ -221,10 +220,9 @@ int blk_remove_plug(struct request_queue *q)
 {
 	WARN_ON(!irqs_disabled());
 
-	if (!test_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
+	if (!queue_flag_test_and_clear(QUEUE_FLAG_PLUGGED, q))
 		return 0;
 
-	queue_flag_clear(QUEUE_FLAG_PLUGGED, q);
 	del_timer(&q->unplug_timer);
 	return 1;
 }
@@ -328,8 +326,7 @@ void blk_start_queue(struct request_queue *q)
 	 * one level of recursion is ok and is much faster than kicking
 	 * the unplug handling
 	 */
-	if (!test_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) {
-		queue_flag_set(QUEUE_FLAG_REENTER, q);
+	if (!queue_flag_test_and_set(QUEUE_FLAG_REENTER, q)) {
 		q->request_fn(q);
 		queue_flag_clear(QUEUE_FLAG_REENTER, q);
 	} else {
@@ -394,8 +391,7 @@ void __blk_run_queue(struct request_queue *q)
 	 * handling reinvoke the handler shortly if we already got there.
 	 */
 	if (!elv_queue_empty(q)) {
-		if (!test_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) {
-			queue_flag_set(QUEUE_FLAG_REENTER, q);
+		if (!queue_flag_test_and_set(QUEUE_FLAG_REENTER, q)) {
 			q->request_fn(q);
 			queue_flag_clear(QUEUE_FLAG_REENTER, q);
 		} else {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index ff9d0bdf2a16..e04c4ac8a7cf 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -428,6 +428,32 @@ static inline void queue_flag_set_unlocked(unsigned int flag,
 	__set_bit(flag, &q->queue_flags);
 }
 
+static inline int queue_flag_test_and_clear(unsigned int flag,
+					    struct request_queue *q)
+{
+	WARN_ON_ONCE(!queue_is_locked(q));
+
+	if (test_bit(flag, &q->queue_flags)) {
+		__clear_bit(flag, &q->queue_flags);
+		return 1;
+	}
+
+	return 0;
+}
+
+static inline int queue_flag_test_and_set(unsigned int flag,
+					  struct request_queue *q)
+{
+	WARN_ON_ONCE(!queue_is_locked(q));
+
+	if (!test_bit(flag, &q->queue_flags)) {
+		__set_bit(flag, &q->queue_flags);
+		return 0;
+	}
+
+	return 1;
+}
+
 static inline void queue_flag_set(unsigned int flag, struct request_queue *q)
 {
 	WARN_ON_ONCE(!queue_is_locked(q));
-- 
cgit v1.2.3


From 42796d37da6ef4fd851dc6d5d0387baf7e2b0c3c Mon Sep 17 00:00:00 2001
From: eric miao <eric.miao@marvell.com>
Date: Mon, 14 Apr 2008 09:35:08 +0100
Subject: [ARM] pxa: add generic PWM backlight driver

Patch mostly from Eric Miao, with minor edits by rmk to convert
Eric's driver to a generic PWM-based backlight driver.

Signed-off-by: eric miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/video/backlight/Kconfig  |   7 ++
 drivers/video/backlight/Makefile |   1 +
 drivers/video/backlight/pwm_bl.c | 160 +++++++++++++++++++++++++++++++++++++++
 include/linux/pwm_backlight.h    |  14 ++++
 4 files changed, 182 insertions(+)
 create mode 100644 drivers/video/backlight/pwm_bl.c
 create mode 100644 include/linux/pwm_backlight.h

(limited to 'include/linux')

diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index dcd8073c2369..30bf7f2f1635 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -112,3 +112,10 @@ config BACKLIGHT_CARILLO_RANCH
 	help
 	  If you have a Intel LE80578 (Carillo Ranch) say Y to enable the
 	  backlight driver.
+
+config BACKLIGHT_PWM
+	tristate "Generic PWM based Backlight Driver"
+	depends on BACKLIGHT_CLASS_DEVICE && HAVE_PWM
+	help
+	  If you have a LCD backlight adjustable by PWM, say Y to enable
+	  this driver.
diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile
index 33f6c7cecc73..b51a7cd12500 100644
--- a/drivers/video/backlight/Makefile
+++ b/drivers/video/backlight/Makefile
@@ -10,3 +10,4 @@ obj-$(CONFIG_BACKLIGHT_LOCOMO)	+= locomolcd.o
 obj-$(CONFIG_BACKLIGHT_OMAP1)	+= omap1_bl.o
 obj-$(CONFIG_BACKLIGHT_PROGEAR) += progear_bl.o
 obj-$(CONFIG_BACKLIGHT_CARILLO_RANCH) += cr_bllcd.o
+obj-$(CONFIG_BACKLIGHT_PWM)	+= pwm_bl.o
diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/backlight/pwm_bl.c
new file mode 100644
index 000000000000..9637f5e08cb6
--- /dev/null
+++ b/drivers/video/backlight/pwm_bl.c
@@ -0,0 +1,160 @@
+/*
+ * linux/drivers/video/backlight/pwm_bl.c
+ *
+ * simple PWM based backlight control, board code has to setup
+ * 1) pin configuration so PWM waveforms can output
+ * 2) platform_data casts to the PWM id (0/1/2/3 on PXA)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/fb.h>
+#include <linux/backlight.h>
+#include <linux/err.h>
+#include <linux/pwm.h>
+#include <linux/pwm_backlight.h>
+
+struct pwm_bl_data {
+	struct pwm_device	*pwm;
+	unsigned int		period;
+};
+
+static int pwm_backlight_update_status(struct backlight_device *bl)
+{
+	struct pwm_bl_data *pb = dev_get_drvdata(&bl->dev);
+	int brightness = bl->props.brightness;
+	int max = bl->props.max_brightness;
+
+	if (bl->props.power != FB_BLANK_UNBLANK)
+		brightness = 0;
+
+	if (bl->props.fb_blank != FB_BLANK_UNBLANK)
+		brightness = 0;
+
+	if (brightness == 0) {
+		pwm_config(pb->pwm, 0, pb->period);
+		pwm_disable(pb->pwm);
+	} else {
+		pwm_config(pb->pwm, brightness * pb->period / max, pb->period);
+		pwm_enable(pb->pwm);
+	}
+	return 0;
+}
+
+static int pwm_backlight_get_brightness(struct backlight_device *bl)
+{
+	return bl->props.brightness;
+}
+
+static struct backlight_ops pwm_backlight_ops = {
+	.update_status	= pwm_backlight_update_status,
+	.get_brightness	= pwm_backlight_get_brightness,
+};
+
+static int pwm_backlight_probe(struct platform_device *pdev)
+{
+	struct platform_pwm_backlight_data *data = pdev->dev.platform_data;
+	struct backlight_device *bl;
+	struct pwm_bl_data *pb;
+
+	if (!data)
+		return -EINVAL;
+
+	pb = kzalloc(sizeof(*pb), GFP_KERNEL);
+	if (!pb)
+		return -ENOMEM;
+
+	pb->period = data->pwm_period_ns;
+
+	pb->pwm = pwm_request(data->pwm_id, "backlight");
+	if (pb->pwm == NULL) {
+		dev_err(&pdev->dev, "unable to request PWM for backlight\n");
+		kfree(pb);
+		return -EBUSY;
+	}
+
+	bl = backlight_device_register(pdev->name, &pdev->dev,
+			pb, &pwm_backlight_ops);
+	if (IS_ERR(bl)) {
+		dev_err(&pdev->dev, "failed to register backlight\n");
+		pwm_free(pb->pwm);
+		kfree(pb);
+		return PTR_ERR(bl);
+	}
+
+	bl->props.max_brightness = data->max_brightness;
+	bl->props.brightness = data->dft_brightness;
+	backlight_update_status(bl);
+
+	platform_set_drvdata(pdev, bl);
+	return 0;
+}
+
+static int pwm_backlight_remove(struct platform_device *pdev)
+{
+	struct backlight_device *bl = platform_get_drvdata(pdev);
+	struct pwm_bl_data *pb = dev_get_drvdata(&bl->dev);
+
+	backlight_device_unregister(bl);
+	pwm_config(pb->pwm, 0, pb->period);
+	pwm_disable(pb->pwm);
+	pwm_free(pb->pwm);
+	kfree(pb);
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int pwm_backlight_suspend(struct platform_device *pdev,
+				 pm_message_t state)
+{
+	struct backlight_device *bl = platform_get_drvdata(pdev);
+	struct pwm_bl_data *pb = dev_get_drvdata(&bl->dev);
+
+	pwm_config(pb->pwm, 0, pb->period);
+	pwm_disable(pb->pwm);
+	return 0;
+}
+
+static int pwm_backlight_resume(struct platform_device *pdev)
+{
+	struct backlight_device *bl = platform_get_drvdata(pdev);
+
+	backlight_update_status(bl);
+	return 0;
+}
+#else
+#define pwm_backlight_suspend	NULL
+#define pwm_backlight_resume	NULL
+#endif
+
+static struct platform_driver pwm_backlight_driver = {
+	.driver		= {
+		.name	= "pwm-backlight",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= pwm_backlight_probe,
+	.remove		= pwm_backlight_remove,
+	.suspend	= pwm_backlight_suspend,
+	.resume		= pwm_backlight_resume,
+};
+
+static int __init pwm_backlight_init(void)
+{
+	return platform_driver_register(&pwm_backlight_driver);
+}
+module_init(pwm_backlight_init);
+
+static void __exit pwm_backlight_exit(void)
+{
+	platform_driver_unregister(&pwm_backlight_driver);
+}
+module_exit(pwm_backlight_exit);
+
+MODULE_DESCRIPTION("PWM based Backlight Driver");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/pwm_backlight.h b/include/linux/pwm_backlight.h
new file mode 100644
index 000000000000..aeeffedbe822
--- /dev/null
+++ b/include/linux/pwm_backlight.h
@@ -0,0 +1,14 @@
+/*
+ * Generic PWM backlight driver data - see drivers/video/backlight/pwm_bl.c
+ */
+#ifndef __LINUX_PWM_BACKLIGHT_H
+#define __LINUX_PWM_BACKLIGHT_H
+
+struct platform_pwm_backlight_data {
+	int pwm_id;
+	unsigned int max_brightness;
+	unsigned int dft_brightness;
+	unsigned int pwm_period_ns;
+};
+
+#endif
-- 
cgit v1.2.3


From 3b73125af69f93972625f4b655675f42ca4274eb Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 22 May 2008 14:18:40 +0100
Subject: [ARM] 5044/1: pwm_bl: add init/notify/exit callbacks

This allows platform code to manipulate GPIOs and brightness level as
needed.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/video/backlight/pwm_bl.c | 39 ++++++++++++++++++++++++++++++++-------
 include/linux/pwm_backlight.h    |  3 +++
 2 files changed, 35 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/backlight/pwm_bl.c
index 9637f5e08cb6..8346dfc01cf6 100644
--- a/drivers/video/backlight/pwm_bl.c
+++ b/drivers/video/backlight/pwm_bl.c
@@ -23,6 +23,7 @@
 struct pwm_bl_data {
 	struct pwm_device	*pwm;
 	unsigned int		period;
+	int			(*notify)(int brightness);
 };
 
 static int pwm_backlight_update_status(struct backlight_device *bl)
@@ -37,6 +38,9 @@ static int pwm_backlight_update_status(struct backlight_device *bl)
 	if (bl->props.fb_blank != FB_BLANK_UNBLANK)
 		brightness = 0;
 
+	if (pb->notify)
+		brightness = pb->notify(brightness);
+
 	if (brightness == 0) {
 		pwm_config(pb->pwm, 0, pb->period);
 		pwm_disable(pb->pwm);
@@ -62,30 +66,39 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 	struct platform_pwm_backlight_data *data = pdev->dev.platform_data;
 	struct backlight_device *bl;
 	struct pwm_bl_data *pb;
+	int ret;
 
 	if (!data)
 		return -EINVAL;
 
+	if (data->init) {
+		ret = data->init(&pdev->dev);
+		if (ret < 0)
+			return ret;
+	}
+
 	pb = kzalloc(sizeof(*pb), GFP_KERNEL);
-	if (!pb)
-		return -ENOMEM;
+	if (!pb) {
+		ret = -ENOMEM;
+		goto err_alloc;
+	}
 
 	pb->period = data->pwm_period_ns;
+	pb->notify = data->notify;
 
 	pb->pwm = pwm_request(data->pwm_id, "backlight");
 	if (pb->pwm == NULL) {
 		dev_err(&pdev->dev, "unable to request PWM for backlight\n");
-		kfree(pb);
-		return -EBUSY;
+		ret = -EBUSY;
+		goto err_pwm;
 	}
 
 	bl = backlight_device_register(pdev->name, &pdev->dev,
 			pb, &pwm_backlight_ops);
 	if (IS_ERR(bl)) {
 		dev_err(&pdev->dev, "failed to register backlight\n");
-		pwm_free(pb->pwm);
-		kfree(pb);
-		return PTR_ERR(bl);
+		ret = PTR_ERR(bl);
+		goto err_bl;
 	}
 
 	bl->props.max_brightness = data->max_brightness;
@@ -94,10 +107,20 @@ static int pwm_backlight_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, bl);
 	return 0;
+
+err_bl:
+	pwm_free(pb->pwm);
+err_pwm:
+	kfree(pb);
+err_alloc:
+	if (data->exit)
+		data->exit(&pdev->dev);
+	return ret;
 }
 
 static int pwm_backlight_remove(struct platform_device *pdev)
 {
+	struct platform_pwm_backlight_data *data = pdev->dev.platform_data;
 	struct backlight_device *bl = platform_get_drvdata(pdev);
 	struct pwm_bl_data *pb = dev_get_drvdata(&bl->dev);
 
@@ -106,6 +129,8 @@ static int pwm_backlight_remove(struct platform_device *pdev)
 	pwm_disable(pb->pwm);
 	pwm_free(pb->pwm);
 	kfree(pb);
+	if (data->exit)
+		data->exit(&pdev->dev);
 	return 0;
 }
 
diff --git a/include/linux/pwm_backlight.h b/include/linux/pwm_backlight.h
index aeeffedbe822..7a9754c96775 100644
--- a/include/linux/pwm_backlight.h
+++ b/include/linux/pwm_backlight.h
@@ -9,6 +9,9 @@ struct platform_pwm_backlight_data {
 	unsigned int max_brightness;
 	unsigned int dft_brightness;
 	unsigned int pwm_period_ns;
+	int (*init)(struct device *dev);
+	int (*notify)(int brightness);
+	void (*exit)(struct device *dev);
 };
 
 #endif
-- 
cgit v1.2.3


From 27f8221af406e43b529a5425bc99c9b1e9bdf521 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 4 Jul 2008 09:30:03 +0200
Subject: block: add blk_queue_update_dma_pad

This adds blk_queue_update_dma_pad to prevent LLDs from overwriting
the dma pad mask wrongly (we added blk_queue_update_dma_alignment due
to the same reason).

This also converts libata to use blk_queue_update_dma_pad instead of
blk_queue_dma_pad.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Tejun Heo <htejun@gmail.com>
Cc: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-settings.c      | 24 ++++++++++++++++++++----
 drivers/ata/libata-scsi.c |  3 ++-
 include/linux/blkdev.h    |  1 +
 3 files changed, 23 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-settings.c b/block/blk-settings.c
index 8dd86418f35d..dfc77012843f 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -302,11 +302,10 @@ EXPORT_SYMBOL(blk_queue_stack_limits);
  * @q:     the request queue for the device
  * @mask:  pad mask
  *
- * Set pad mask.  Direct IO requests are padded to the mask specified.
+ * Set dma pad mask.
  *
- * Appending pad buffer to a request modifies ->data_len such that it
- * includes the pad buffer.  The original requested data length can be
- * obtained using blk_rq_raw_data_len().
+ * Appending pad buffer to a request modifies the last entry of a
+ * scatter list such that it includes the pad buffer.
  **/
 void blk_queue_dma_pad(struct request_queue *q, unsigned int mask)
 {
@@ -314,6 +313,23 @@ void blk_queue_dma_pad(struct request_queue *q, unsigned int mask)
 }
 EXPORT_SYMBOL(blk_queue_dma_pad);
 
+/**
+ * blk_queue_update_dma_pad - update pad mask
+ * @q:     the request queue for the device
+ * @mask:  pad mask
+ *
+ * Update dma pad mask.
+ *
+ * Appending pad buffer to a request modifies the last entry of a
+ * scatter list such that it includes the pad buffer.
+ **/
+void blk_queue_update_dma_pad(struct request_queue *q, unsigned int mask)
+{
+	if (mask > q->dma_pad_mask)
+		q->dma_pad_mask = mask;
+}
+EXPORT_SYMBOL(blk_queue_update_dma_pad);
+
 /**
  * blk_queue_dma_drain - Set up a drain buffer for excess dma.
  * @q:  the request queue for the device
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 57a43649a461..499ccc628d81 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -885,7 +885,8 @@ static int ata_scsi_dev_config(struct scsi_device *sdev,
 		/* set the min alignment and padding */
 		blk_queue_update_dma_alignment(sdev->request_queue,
 					       ATA_DMA_PAD_SZ - 1);
-		blk_queue_dma_pad(sdev->request_queue, ATA_DMA_PAD_SZ - 1);
+		blk_queue_update_dma_pad(sdev->request_queue,
+					 ATA_DMA_PAD_SZ - 1);
 
 		/* configure draining */
 		buf = kmalloc(ATAPI_MAX_DRAIN, q->bounce_gfp | GFP_KERNEL);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index e04c4ac8a7cf..1ffd8bfdc4c9 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -776,6 +776,7 @@ extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
 extern void blk_queue_hardsect_size(struct request_queue *, unsigned short);
 extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
 extern void blk_queue_dma_pad(struct request_queue *, unsigned int);
+extern void blk_queue_update_dma_pad(struct request_queue *, unsigned int);
 extern int blk_queue_dma_drain(struct request_queue *q,
 			       dma_drain_needed_fn *dma_drain_needed,
 			       void *buf, unsigned int size);
-- 
cgit v1.2.3


From ba8dd03ac09f51a69c154b8cb508b701d713a2cd Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 4 Jul 2008 11:26:40 +0200
Subject: generic-ipi: fix s390 build bug

forgot to remove #include <linux/spinlock.h> from linux/smp.h while
fixing the original s390 build bug.

Patch below fixes this build bug caused by header inclusion dependencies:

  CC      kernel/timer.o
In file included from include/linux/spinlock.h:87,
                 from include/linux/smp.h:11,
                 from include/linux/kernel_stat.h:4,
                 from kernel/timer.c:22:
include/asm/spinlock.h: In function '__raw_spin_lock':
include/asm/spinlock.h:69: error: implicit declaration of function 'smp_processor_id'

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/smp.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/smp.h b/include/linux/smp.h
index 55261101d09a..48262f86c969 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -8,7 +8,6 @@
 
 #include <linux/errno.h>
 #include <linux/list.h>
-#include <linux/spinlock.h>
 #include <linux/cpumask.h>
 
 extern void cpu_idle(void);
-- 
cgit v1.2.3


From 198191c4a7ce4daba379608fb38b9bc5a4eedc61 Mon Sep 17 00:00:00 2001
From: Krzysztof Halasa <khc@pm.waw.pl>
Date: Mon, 30 Jun 2008 23:26:53 +0200
Subject: WAN: convert drivers to use built-in netdev_stats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is no point in using separate net_device_stats structs when
the one in struct net_device is present. Compiles.

Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/char/pcmcia/synclink_cs.c | 28 +++++++--------
 drivers/char/synclink.c           | 33 +++++++++---------
 drivers/char/synclink_gt.c        | 28 +++++++--------
 drivers/char/synclinkmp.c         | 31 ++++++++---------
 drivers/net/wan/c101.c            |  6 ++--
 drivers/net/wan/dscc4.c           | 22 ++++++------
 drivers/net/wan/farsync.c         | 70 +++++++++++++++++---------------------
 drivers/net/wan/hd6457x.c         | 33 +++++++++---------
 drivers/net/wan/hdlc.c            |  2 +-
 drivers/net/wan/hdlc_cisco.c      |  4 +--
 drivers/net/wan/hdlc_fr.c         | 29 ++++++----------
 drivers/net/wan/hdlc_raw_eth.c    |  2 +-
 drivers/net/wan/hdlc_x25.c        |  6 ++--
 drivers/net/wan/pc300_drv.c       | 71 +++++++++++++++++----------------------
 drivers/net/wan/pc300_tty.c       |  6 ++--
 drivers/net/wan/wanxl.c           | 26 +++++++-------
 include/linux/hdlc.h              |  7 ----
 17 files changed, 174 insertions(+), 230 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 1dd0e992c83d..fb2fb159faa3 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -3886,9 +3886,8 @@ static bool rx_get_frame(MGSLPC_INFO *info)
 		framesize = 0;
 #if SYNCLINK_GENERIC_HDLC
 		{
-			struct net_device_stats *stats = hdlc_stats(info->netdev);
-			stats->rx_errors++;
-			stats->rx_frame_errors++;
+			info->netdev->stats.rx_errors++;
+			info->netdev->stats.rx_frame_errors++;
 		}
 #endif
 	} else
@@ -4144,7 +4143,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -4159,8 +4157,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	info->tx_put = info->tx_count = skb->len;
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -4376,14 +4374,13 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	MGSLPC_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_tx_timeout(%s)\n",dev->name);
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->lock,flags);
 	tx_stop(info);
@@ -4416,27 +4413,26 @@ static void hdlcdev_rx(MGSLPC_INFO *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_rx(%s)\n",dev->name);
 
 	if (skb == NULL) {
 		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n", dev->name);
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index ac5080df2565..9d247d8a87a3 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -6640,9 +6640,8 @@ static bool mgsl_get_rx_frame(struct mgsl_struct *info)
 		framesize = 0;
 #if SYNCLINK_GENERIC_HDLC
 		{
-			struct net_device_stats *stats = hdlc_stats(info->netdev);
-			stats->rx_errors++;
-			stats->rx_frame_errors++;
+			info->netdev->stats.rx_errors++;
+			info->netdev->stats.rx_frame_errors++;
 		}
 #endif
 	} else
@@ -7753,7 +7752,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct mgsl_struct *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -7767,8 +7765,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	mgsl_load_tx_dma_buffer(info, skb->data, skb->len);
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -7984,14 +7982,13 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	struct mgsl_struct *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_tx_timeout(%s)\n",dev->name);
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->irq_spinlock,flags);
 	usc_stop_transmitter(info);
@@ -8024,27 +8021,27 @@ static void hdlcdev_rx(struct mgsl_struct *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
-		printk("hdlcdev_rx(%s)\n",dev->name);
+		printk("hdlcdev_rx(%s)\n", dev->name);
 
 	if (skb == NULL) {
-		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n", dev->name);
-		stats->rx_dropped++;
+		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n",
+		       dev->name);
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 55c1653be00c..d88a607e34b7 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -1544,7 +1544,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct slgt_info *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	DBGINFO(("%s hdlc_xmit\n", dev->name));
@@ -1557,8 +1556,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	tx_load(info, skb->data, skb->len);
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -1775,13 +1774,12 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	struct slgt_info *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	DBGINFO(("%s hdlcdev_tx_timeout\n", dev->name));
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->lock,flags);
 	tx_stop(info);
@@ -1814,26 +1812,25 @@ static void hdlcdev_rx(struct slgt_info *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	DBGINFO(("%s hdlcdev_rx\n", dev->name));
 
 	if (skb == NULL) {
 		DBGERR(("%s: can't alloc skb, drop packet\n", dev->name));
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
@@ -4577,9 +4574,8 @@ check_again:
 
 #if SYNCLINK_GENERIC_HDLC
 	if (framesize == 0) {
-		struct net_device_stats *stats = hdlc_stats(info->netdev);
-		stats->rx_errors++;
-		stats->rx_frame_errors++;
+		info->netdev->stats.rx_errors++;
+		info->netdev->stats.rx_frame_errors++;
 	}
 #endif
 
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index bec54866e0bb..10241ed86100 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -1678,7 +1678,6 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding,
 static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	SLMP_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1692,8 +1691,8 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	tx_load_dma_buffer(info, skb->data, skb->len);
 
 	/* update network statistics */
-	stats->tx_packets++;
-	stats->tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
 
 	/* done with socket buffer, so free it */
 	dev_kfree_skb(skb);
@@ -1909,14 +1908,13 @@ static int hdlcdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static void hdlcdev_tx_timeout(struct net_device *dev)
 {
 	SLMP_INFO *info = dev_to_port(dev);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_tx_timeout(%s)\n",dev->name);
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 
 	spin_lock_irqsave(&info->lock,flags);
 	tx_stop(info);
@@ -1949,27 +1947,27 @@ static void hdlcdev_rx(SLMP_INFO *info, char *buf, int size)
 {
 	struct sk_buff *skb = dev_alloc_skb(size);
 	struct net_device *dev = info->netdev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("hdlcdev_rx(%s)\n",dev->name);
 
 	if (skb == NULL) {
-		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n", dev->name);
-		stats->rx_dropped++;
+		printk(KERN_NOTICE "%s: can't alloc skb, dropping packet\n",
+		       dev->name);
+		dev->stats.rx_dropped++;
 		return;
 	}
 
-	memcpy(skb_put(skb, size),buf,size);
+	memcpy(skb_put(skb, size), buf, size);
 
-	skb->protocol = hdlc_type_trans(skb, info->netdev);
+	skb->protocol = hdlc_type_trans(skb, dev);
 
-	stats->rx_packets++;
-	stats->rx_bytes += size;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += size;
 
 	netif_rx(skb);
 
-	info->netdev->last_rx = jiffies;
+	dev->last_rx = jiffies;
 }
 
 /**
@@ -4983,9 +4981,8 @@ CheckAgain:
 		framesize = 0;
 #if SYNCLINK_GENERIC_HDLC
 		{
-			struct net_device_stats *stats = hdlc_stats(info->netdev);
-			stats->rx_errors++;
-			stats->rx_frame_errors++;
+			info->netdev->stats.rx_errors++;
+			info->netdev->stats.rx_frame_errors++;
 		}
 #endif
 	}
diff --git a/drivers/net/wan/c101.c b/drivers/net/wan/c101.c
index c2cc42f723d5..c8e563106a4a 100644
--- a/drivers/net/wan/c101.c
+++ b/drivers/net/wan/c101.c
@@ -133,9 +133,9 @@ static void sca_msci_intr(port_t *port)
 	sca_out(stat & (ST1_UDRN | ST1_CDCD), MSCI0_OFFSET + ST1, port);
 
 	if (stat & ST1_UDRN) {
-		struct net_device_stats *stats = hdlc_stats(port_to_dev(port));
-		stats->tx_errors++; /* TX Underrun error detected */
-		stats->tx_fifo_errors++;
+		/* TX Underrun error detected */
+		port_to_dev(port)->stats.tx_errors++;
+		port_to_dev(port)->stats.tx_fifo_errors++;
 	}
 
 	stat = sca_in(MSCI1_OFFSET + ST1, port); /* read MSCI1 ST1 status */
diff --git a/drivers/net/wan/dscc4.c b/drivers/net/wan/dscc4.c
index c6f26e28e376..50ef5b4efd6d 100644
--- a/drivers/net/wan/dscc4.c
+++ b/drivers/net/wan/dscc4.c
@@ -642,7 +642,6 @@ static inline void dscc4_rx_skb(struct dscc4_dev_priv *dpriv,
 				struct net_device *dev)
 {
 	struct RxFD *rx_fd = dpriv->rx_fd + dpriv->rx_current%RX_RING_SIZE;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	struct pci_dev *pdev = dpriv->pci_priv->pdev;
 	struct sk_buff *skb;
 	int pkt_len;
@@ -656,8 +655,8 @@ static inline void dscc4_rx_skb(struct dscc4_dev_priv *dpriv,
 	pci_unmap_single(pdev, le32_to_cpu(rx_fd->data),
 			 RX_MAX(HDLC_MAX_MRU), PCI_DMA_FROMDEVICE);
 	if ((skb->data[--pkt_len] & FrameOk) == FrameOk) {
-		stats->rx_packets++;
-		stats->rx_bytes += pkt_len;
+		dev->stats.rx_packets++;
+		dev->stats.rx_bytes += pkt_len;
 		skb_put(skb, pkt_len);
 		if (netif_running(dev))
 			skb->protocol = hdlc_type_trans(skb, dev);
@@ -665,13 +664,13 @@ static inline void dscc4_rx_skb(struct dscc4_dev_priv *dpriv,
 		netif_rx(skb);
 	} else {
 		if (skb->data[pkt_len] & FrameRdo)
-			stats->rx_fifo_errors++;
+			dev->stats.rx_fifo_errors++;
 		else if (!(skb->data[pkt_len] | ~FrameCrc))
-			stats->rx_crc_errors++;
+			dev->stats.rx_crc_errors++;
 		else if (!(skb->data[pkt_len] | ~(FrameVfr | FrameRab)))
-			stats->rx_length_errors++;
+			dev->stats.rx_length_errors++;
 		else
-			stats->rx_errors++;
+			dev->stats.rx_errors++;
 		dev_kfree_skb_irq(skb);
 	}
 refill:
@@ -1569,7 +1568,6 @@ try:
 
 	if (state & SccEvt) {
 		if (state & Alls) {
-			struct net_device_stats *stats = hdlc_stats(dev);
 			struct sk_buff *skb;
 			struct TxFD *tx_fd;
 
@@ -1586,8 +1584,8 @@ try:
 				pci_unmap_single(ppriv->pdev, le32_to_cpu(tx_fd->data),
 						 skb->len, PCI_DMA_TODEVICE);
 				if (tx_fd->state & FrameEnd) {
-					stats->tx_packets++;
-					stats->tx_bytes += skb->len;
+					dev->stats.tx_packets++;
+					dev->stats.tx_bytes += skb->len;
 				}
 				dev_kfree_skb_irq(skb);
 				dpriv->tx_skbuff[cur] = NULL;
@@ -1698,7 +1696,7 @@ try:
 		}
 		if (state & Err) {
 			printk(KERN_INFO "%s: Tx ERR\n", dev->name);
-			hdlc_stats(dev)->tx_errors++;
+			dev->stats.tx_errors++;
 			state &= ~Err;
 		}
 	}
@@ -1834,7 +1832,7 @@ try:
 				if (!(rx_fd->state2 & DataComplete))
 					break;
 				if (rx_fd->state2 & FrameAborted) {
-					hdlc_stats(dev)->rx_over_errors++;
+					dev->stats.rx_over_errors++;
 					rx_fd->state1 |= Hold;
 					rx_fd->state2 = 0x00000000;
 					rx_fd->end = cpu_to_le32(0xbabeface);
diff --git a/drivers/net/wan/farsync.c b/drivers/net/wan/farsync.c
index 547368e9633d..754f00809e3e 100644
--- a/drivers/net/wan/farsync.c
+++ b/drivers/net/wan/farsync.c
@@ -845,7 +845,6 @@ fst_tx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 		    int len, int txpos)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	/*
 	 * Everything is now set, just tell the card to go
@@ -853,8 +852,8 @@ fst_tx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 	dbg(DBG_TX, "fst_tx_dma_complete\n");
 	FST_WRB(card, txDescrRing[port->index][txpos].bits,
 		DMA_OWN | TX_STP | TX_ENP);
-	stats->tx_packets++;
-	stats->tx_bytes += len;
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += len;
 	dev->trans_start = jiffies;
 }
 
@@ -876,7 +875,6 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 		    int len, struct sk_buff *skb, int rxp)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int pi;
 	int rx_status;
 
@@ -888,8 +886,8 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 	FST_WRB(card, rxDescrRing[pi][rxp].bits, DMA_OWN);
 
 	/* Update stats */
-	stats->rx_packets++;
-	stats->rx_bytes += len;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += len;
 
 	/* Push upstream */
 	dbg(DBG_RX, "Pushing the frame up the stack\n");
@@ -900,7 +898,7 @@ fst_rx_dma_complete(struct fst_card_info *card, struct fst_port_info *port,
 	rx_status = netif_rx(skb);
 	fst_process_rx_status(rx_status, port_to_dev(port)->name);
 	if (rx_status == NET_RX_DROP)
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 	dev->last_rx = jiffies;
 }
 
@@ -1163,29 +1161,28 @@ fst_log_rx_error(struct fst_card_info *card, struct fst_port_info *port,
 		 unsigned char dmabits, int rxp, unsigned short len)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 
-	/* 
+	/*
 	 * Increment the appropriate error counter
 	 */
-	stats->rx_errors++;
+	dev->stats.rx_errors++;
 	if (dmabits & RX_OFLO) {
-		stats->rx_fifo_errors++;
+		dev->stats.rx_fifo_errors++;
 		dbg(DBG_ASS, "Rx fifo error on card %d port %d buffer %d\n",
 		    card->card_no, port->index, rxp);
 	}
 	if (dmabits & RX_CRC) {
-		stats->rx_crc_errors++;
+		dev->stats.rx_crc_errors++;
 		dbg(DBG_ASS, "Rx crc error on card %d port %d\n",
 		    card->card_no, port->index);
 	}
 	if (dmabits & RX_FRAM) {
-		stats->rx_frame_errors++;
+		dev->stats.rx_frame_errors++;
 		dbg(DBG_ASS, "Rx frame error on card %d port %d\n",
 		    card->card_no, port->index);
 	}
 	if (dmabits == (RX_STP | RX_ENP)) {
-		stats->rx_length_errors++;
+		dev->stats.rx_length_errors++;
 		dbg(DBG_ASS, "Rx length error (%d) on card %d port %d\n",
 		    len, card->card_no, port->index);
 	}
@@ -1242,7 +1239,6 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 	unsigned short len;
 	struct sk_buff *skb;
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	/* Check we have a buffer to process */
 	pi = port->index;
@@ -1291,7 +1287,7 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 	if ((skb = dev_alloc_skb(len)) == NULL) {
 		dbg(DBG_RX, "intr_rx: can't allocate buffer\n");
 
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 
 		/* Return descriptor to card */
 		FST_WRB(card, rxDescrRing[pi][rxp].bits, DMA_OWN);
@@ -1316,8 +1312,8 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 		FST_WRB(card, rxDescrRing[pi][rxp].bits, DMA_OWN);
 
 		/* Update stats */
-		stats->rx_packets++;
-		stats->rx_bytes += len;
+		dev->stats.rx_packets++;
+		dev->stats.rx_bytes += len;
 
 		/* Push upstream */
 		dbg(DBG_RX, "Pushing frame up the stack\n");
@@ -1327,9 +1323,8 @@ fst_intr_rx(struct fst_card_info *card, struct fst_port_info *port)
 			skb->protocol = hdlc_type_trans(skb, dev);
 		rx_status = netif_rx(skb);
 		fst_process_rx_status(rx_status, port_to_dev(port)->name);
-		if (rx_status == NET_RX_DROP) {
-			stats->rx_dropped++;
-		}
+		if (rx_status == NET_RX_DROP)
+			dev->stats.rx_dropped++;
 		dev->last_rx = jiffies;
 	} else {
 		card->dma_skb_rx = skb;
@@ -1361,7 +1356,6 @@ do_bottom_half_tx(struct fst_card_info *card)
 	struct sk_buff *skb;
 	unsigned long flags;
 	struct net_device *dev;
-	struct net_device_stats *stats;
 
 	/*
 	 *  Find a free buffer for the transmit
@@ -1373,12 +1367,10 @@ do_bottom_half_tx(struct fst_card_info *card)
 		if (!port->run)
 			continue;
 
-                dev = port_to_dev(port);
-                stats = hdlc_stats(dev);
-		while (!
-		       (FST_RDB(card, txDescrRing[pi][port->txpos].bits) &
-			DMA_OWN)
-                       && !(card->dmatx_in_progress)) {
+		dev = port_to_dev(port);
+		while (!(FST_RDB(card, txDescrRing[pi][port->txpos].bits) &
+			 DMA_OWN)
+		       && !(card->dmatx_in_progress)) {
 			/*
 			 * There doesn't seem to be a txdone event per-se
 			 * We seem to have to deduce it, by checking the DMA_OWN
@@ -1422,8 +1414,8 @@ do_bottom_half_tx(struct fst_card_info *card)
 						txDescrRing[pi][port->txpos].
 						bits,
 						DMA_OWN | TX_STP | TX_ENP);
-					stats->tx_packets++;
-					stats->tx_bytes += skb->len;
+					dev->stats.tx_packets++;
+					dev->stats.tx_bytes += skb->len;
 					dev->trans_start = jiffies;
 				} else {
 					/* Or do it through dma */
@@ -1628,8 +1620,8 @@ fst_intr(int dummy, void *dev_id)
 			 * always load up the entire packet for DMA.
 			 */
 			dbg(DBG_TX, "Tx underflow port %d\n", port->index);
-                        hdlc_stats(port_to_dev(port))->tx_errors++;
-                        hdlc_stats(port_to_dev(port))->tx_fifo_errors++;
+			port_to_dev(port)->stats.tx_errors++;
+			port_to_dev(port)->stats.tx_fifo_errors++;
 			dbg(DBG_ASS, "Tx underflow on card %d port %d\n",
 			    card->card_no, port->index);
 			break;
@@ -2292,12 +2284,11 @@ fst_tx_timeout(struct net_device *dev)
 {
 	struct fst_port_info *port;
 	struct fst_card_info *card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 
 	port = dev_to_port(dev);
 	card = port->card;
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 	dbg(DBG_ASS, "Tx timeout card %d port %d\n",
 	    card->card_no, port->index);
 	fst_issue_cmd(port, ABORTTX);
@@ -2312,7 +2303,6 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct fst_card_info *card;
 	struct fst_port_info *port;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	unsigned long flags;
 	int txq_length;
 
@@ -2323,8 +2313,8 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* Drop packet with error if we don't have carrier */
 	if (!netif_carrier_ok(dev)) {
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
-		stats->tx_carrier_errors++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_carrier_errors++;
 		dbg(DBG_ASS,
 		    "Tried to transmit but no carrier on card %d port %d\n",
 		    card->card_no, port->index);
@@ -2336,7 +2326,7 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		dbg(DBG_ASS, "Packet too large %d vs %d\n", skb->len,
 		    LEN_TX_BUFFER);
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
+		dev->stats.tx_errors++;
 		return 0;
 	}
 
@@ -2368,7 +2358,7 @@ fst_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * This shouldn't have happened but such is life
 		 */
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
+		dev->stats.tx_errors++;
 		dbg(DBG_ASS, "Tx queue overflow card %d port %d\n",
 		    card->card_no, port->index);
 		return 0;
diff --git a/drivers/net/wan/hd6457x.c b/drivers/net/wan/hd6457x.c
index 8d0a1f2f00e5..591fb45a7c68 100644
--- a/drivers/net/wan/hd6457x.c
+++ b/drivers/net/wan/hd6457x.c
@@ -271,9 +271,9 @@ static inline void sca_msci_intr(port_t *port)
 	sca_out(stat & (ST1_UDRN | ST1_CDCD), msci + ST1, card);
 
 	if (stat & ST1_UDRN) {
-		struct net_device_stats *stats = hdlc_stats(port_to_dev(port));
-		stats->tx_errors++; /* TX Underrun error detected */
-		stats->tx_fifo_errors++;
+		/* TX Underrun error detected */
+		port_to_dev(port)->stats.tx_errors++;
+		port_to_dev(port)->stats.tx_fifo_errors++;
 	}
 
 	if (stat & ST1_CDCD)
@@ -286,7 +286,6 @@ static inline void sca_msci_intr(port_t *port)
 static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u16 rxin)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	struct sk_buff *skb;
 	u16 len;
 	u32 buff;
@@ -298,7 +297,7 @@ static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u1
 	len = readw(&desc->len);
 	skb = dev_alloc_skb(len);
 	if (!skb) {
-		stats->rx_dropped++;
+		dev->stats.rx_dropped++;
 		return;
 	}
 
@@ -327,8 +326,8 @@ static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u1
 	printk(KERN_DEBUG "%s RX(%i):", dev->name, skb->len);
 	debug_frame(skb);
 #endif
-	stats->rx_packets++;
-	stats->rx_bytes += skb->len;
+	dev->stats.rx_packets++;
+	dev->stats.rx_bytes += skb->len;
 	dev->last_rx = jiffies;
 	skb->protocol = hdlc_type_trans(skb, dev);
 	netif_rx(skb);
@@ -339,17 +338,18 @@ static inline void sca_rx(card_t *card, port_t *port, pkt_desc __iomem *desc, u1
 /* Receive DMA interrupt service */
 static inline void sca_rx_intr(port_t *port)
 {
+	struct net_device *dev = port_to_dev(port);
 	u16 dmac = get_dmac_rx(port);
 	card_t *card = port_to_card(port);
 	u8 stat = sca_in(DSR_RX(phy_node(port)), card); /* read DMA Status */
-	struct net_device_stats *stats = hdlc_stats(port_to_dev(port));
 
 	/* Reset DSR status bits */
 	sca_out((stat & (DSR_EOT | DSR_EOM | DSR_BOF | DSR_COF)) | DSR_DWE,
 		DSR_RX(phy_node(port)), card);
 
 	if (stat & DSR_BOF)
-		stats->rx_over_errors++; /* Dropped one or more frames */
+		/* Dropped one or more frames */
+		dev->stats.rx_over_errors++;
 
 	while (1) {
 		u32 desc_off = desc_offset(port, port->rxin, 0);
@@ -364,12 +364,14 @@ static inline void sca_rx_intr(port_t *port)
 		if (!(stat & ST_RX_EOM))
 			port->rxpart = 1; /* partial frame received */
 		else if ((stat & ST_ERROR_MASK) || port->rxpart) {
-			stats->rx_errors++;
-			if (stat & ST_RX_OVERRUN) stats->rx_fifo_errors++;
+			dev->stats.rx_errors++;
+			if (stat & ST_RX_OVERRUN)
+				dev->stats.rx_fifo_errors++;
 			else if ((stat & (ST_RX_SHORT | ST_RX_ABORT |
 					  ST_RX_RESBIT)) || port->rxpart)
-				stats->rx_frame_errors++;
-			else if (stat & ST_RX_CRC) stats->rx_crc_errors++;
+				dev->stats.rx_frame_errors++;
+			else if (stat & ST_RX_CRC)
+				dev->stats.rx_crc_errors++;
 			if (stat & ST_RX_EOM)
 				port->rxpart = 0; /* received last fragment */
 		} else
@@ -390,7 +392,6 @@ static inline void sca_rx_intr(port_t *port)
 static inline void sca_tx_intr(port_t *port)
 {
 	struct net_device *dev = port_to_dev(port);
-	struct net_device_stats *stats = hdlc_stats(dev);
 	u16 dmac = get_dmac_tx(port);
 	card_t* card = port_to_card(port);
 	u8 stat;
@@ -412,8 +413,8 @@ static inline void sca_tx_intr(port_t *port)
 			break;	/* Transmitter is/will_be sending this frame */
 
 		desc = desc_address(port, port->txlast, 1);
-		stats->tx_packets++;
-		stats->tx_bytes += readw(&desc->len);
+		dev->stats.tx_packets++;
+		dev->stats.tx_bytes += readw(&desc->len);
 		writeb(0, &desc->stat);	/* Free descriptor */
 		port->txlast = next_desc(port, port->txlast, 1);
 	}
diff --git a/drivers/net/wan/hdlc.c b/drivers/net/wan/hdlc.c
index 7f984895b0d5..e3a536477c7e 100644
--- a/drivers/net/wan/hdlc.c
+++ b/drivers/net/wan/hdlc.c
@@ -57,7 +57,7 @@ static int hdlc_change_mtu(struct net_device *dev, int new_mtu)
 
 static struct net_device_stats *hdlc_get_stats(struct net_device *dev)
 {
-	return hdlc_stats(dev);
+	return &dev->stats;
 }
 
 
diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c
index 762d21c1c703..849819c2552d 100644
--- a/drivers/net/wan/hdlc_cisco.c
+++ b/drivers/net/wan/hdlc_cisco.c
@@ -252,8 +252,8 @@ static int cisco_rx(struct sk_buff *skb)
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 
- rx_error:
-	dev_to_hdlc(dev)->stats.rx_errors++; /* Mark error */
+rx_error:
+	dev->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c
index 520bb0b1a9a2..cb1b415053eb 100644
--- a/drivers/net/wan/hdlc_fr.c
+++ b/drivers/net/wan/hdlc_fr.c
@@ -136,7 +136,6 @@ typedef struct pvc_device_struct {
 }pvc_device;
 
 struct pvc_desc {
-	struct net_device_stats stats;
 	pvc_device *pvc;
 };
 
@@ -184,11 +183,6 @@ static inline struct pvc_desc* pvcdev_to_desc(struct net_device *dev)
 	return dev->priv;
 }
 
-static inline struct net_device_stats* pvc_get_stats(struct net_device *dev)
-{
-	return &pvcdev_to_desc(dev)->stats;
-}
-
 static inline pvc_device* find_pvc(hdlc_device *hdlc, u16 dlci)
 {
 	pvc_device *pvc = state(hdlc)->first_pvc;
@@ -425,7 +419,6 @@ static int pvc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	pvc_device *pvc = pvcdev_to_desc(dev)->pvc;
-	struct net_device_stats *stats = pvc_get_stats(dev);
 
 	if (pvc->state.active) {
 		if (dev->type == ARPHRD_ETHER) {
@@ -435,7 +428,7 @@ static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 				if (skb_tailroom(skb) < pad)
 					if (pskb_expand_head(skb, 0, pad,
 							     GFP_ATOMIC)) {
-						stats->tx_dropped++;
+						dev->stats.tx_dropped++;
 						dev_kfree_skb(skb);
 						return 0;
 					}
@@ -445,17 +438,17 @@ static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 			skb->protocol = __constant_htons(ETH_P_802_3);
 		}
 		if (!fr_hard_header(&skb, pvc->dlci)) {
-			stats->tx_bytes += skb->len;
-			stats->tx_packets++;
+			dev->stats.tx_bytes += skb->len;
+			dev->stats.tx_packets++;
 			if (pvc->state.fecn) /* TX Congestion counter */
-				stats->tx_compressed++;
+				dev->stats.tx_compressed++;
 			skb->dev = pvc->frad;
 			dev_queue_xmit(skb);
 			return 0;
 		}
 	}
 
-	stats->tx_dropped++;
+	dev->stats.tx_dropped++;
 	dev_kfree_skb(skb);
 	return 0;
 }
@@ -955,7 +948,7 @@ static int fr_rx(struct sk_buff *skb)
 
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		dev_to_hdlc(frad)->stats.rx_dropped++;
+		frad->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
@@ -1003,11 +996,10 @@ static int fr_rx(struct sk_buff *skb)
 	}
 
 	if (dev) {
-		struct net_device_stats *stats = pvc_get_stats(dev);
-		stats->rx_packets++; /* PVC traffic */
-		stats->rx_bytes += skb->len;
+		dev->stats.rx_packets++; /* PVC traffic */
+		dev->stats.rx_bytes += skb->len;
 		if (pvc->state.becn)
-			stats->rx_compressed++;
+			dev->stats.rx_compressed++;
 		netif_rx(skb);
 		return NET_RX_SUCCESS;
 	} else {
@@ -1016,7 +1008,7 @@ static int fr_rx(struct sk_buff *skb)
 	}
 
  rx_error:
-	dev_to_hdlc(frad)->stats.rx_errors++; /* Mark error */
+	frad->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
@@ -1122,7 +1114,6 @@ static int fr_add_pvc(struct net_device *frad, unsigned int dlci, int type)
 		dlci_to_q922(dev->broadcast, dlci);
 	}
 	dev->hard_start_xmit = pvc_xmit;
-	dev->get_stats = pvc_get_stats;
 	dev->open = pvc_open;
 	dev->stop = pvc_close;
 	dev->do_ioctl = pvc_ioctl;
diff --git a/drivers/net/wan/hdlc_raw_eth.c b/drivers/net/wan/hdlc_raw_eth.c
index d20c685f6711..26dee600506f 100644
--- a/drivers/net/wan/hdlc_raw_eth.c
+++ b/drivers/net/wan/hdlc_raw_eth.c
@@ -33,7 +33,7 @@ static int eth_tx(struct sk_buff *skb, struct net_device *dev)
 		int len = skb->len;
 		if (skb_tailroom(skb) < pad)
 			if (pskb_expand_head(skb, 0, pad, GFP_ATOMIC)) {
-				hdlc_stats(dev)->tx_dropped++;
+				dev->stats.tx_dropped++;
 				dev_kfree_skb(skb);
 				return 0;
 			}
diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c
index c15cc11e399b..e808720030ef 100644
--- a/drivers/net/wan/hdlc_x25.c
+++ b/drivers/net/wan/hdlc_x25.c
@@ -164,17 +164,15 @@ static void x25_close(struct net_device *dev)
 
 static int x25_rx(struct sk_buff *skb)
 {
-	struct hdlc_device *hdlc = dev_to_hdlc(skb->dev);
-
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		hdlc->stats.rx_dropped++;
+		skb->dev->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
 	if (lapb_data_received(skb->dev, skb) == LAPB_OK)
 		return NET_RX_SUCCESS;
 
-	hdlc->stats.rx_errors++;
+	skb->dev->stats.rx_errors++;
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
diff --git a/drivers/net/wan/pc300_drv.c b/drivers/net/wan/pc300_drv.c
index 57914fbd41d3..334170527755 100644
--- a/drivers/net/wan/pc300_drv.c
+++ b/drivers/net/wan/pc300_drv.c
@@ -285,7 +285,6 @@ static void rx_dma_buf_init(pc300_t *, int);
 static void tx_dma_buf_check(pc300_t *, int);
 static void rx_dma_buf_check(pc300_t *, int);
 static irqreturn_t cpc_intr(int, void *);
-static struct net_device_stats *cpc_get_stats(struct net_device *);
 static int clock_rate_calc(uclong, uclong, int *);
 static uclong detect_ram(pc300_t *);
 static void plx_init(pc300_t *);
@@ -1775,13 +1774,12 @@ static void cpc_tx_timeout(struct net_device *dev)
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int ch = chan->channel;
 	unsigned long flags;
 	ucchar ilar;
 
-	stats->tx_errors++;
-	stats->tx_aborted_errors++;
+	dev->stats.tx_errors++;
+	dev->stats.tx_aborted_errors++;
 	CPC_LOCK(card, flags);
 	if ((ilar = cpc_readb(card->hw.scabase + ILAR)) != 0) {
 		printk("%s: ILAR=0x%x\n", dev->name, ilar);
@@ -1803,7 +1801,6 @@ static int cpc_queue_xmit(struct sk_buff *skb, struct net_device *dev)
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int ch = chan->channel;
 	unsigned long flags;
 #ifdef PC300_DEBUG_TX
@@ -1817,13 +1814,13 @@ static int cpc_queue_xmit(struct sk_buff *skb, struct net_device *dev)
 	} else if (!netif_carrier_ok(dev)) {
 		/* DCD must be OFF: drop packet */
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
-		stats->tx_carrier_errors++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_carrier_errors++;
 		return 0;
 	} else if (cpc_readb(card->hw.scabase + M_REG(ST3, ch)) & ST3_DCD) {
 		printk("%s: DCD is OFF. Going administrative down.\n", dev->name);
-		stats->tx_errors++;
-		stats->tx_carrier_errors++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_carrier_errors++;
 		dev_kfree_skb(skb);
 		netif_carrier_off(dev);
 		CPC_LOCK(card, flags);
@@ -1843,8 +1840,8 @@ static int cpc_queue_xmit(struct sk_buff *skb, struct net_device *dev)
 //		printk("%s: write error. Dropping TX packet.\n", dev->name);
 		netif_stop_queue(dev);
 		dev_kfree_skb(skb);
-		stats->tx_errors++;
-		stats->tx_dropped++;
+		dev->stats.tx_errors++;
+		dev->stats.tx_dropped++;
 		return 0;
 	}
 #ifdef PC300_DEBUG_TX
@@ -1886,7 +1883,6 @@ static void cpc_net_rx(struct net_device *dev)
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	int ch = chan->channel;
 #ifdef PC300_DEBUG_RX
 	int i;
@@ -1922,24 +1918,24 @@ static void cpc_net_rx(struct net_device *dev)
 #endif
 			if ((skb == NULL) && (rxb > 0)) {
 				/* rxb > dev->mtu */
-				stats->rx_errors++;
-				stats->rx_length_errors++;
+				dev->stats.rx_errors++;
+				dev->stats.rx_length_errors++;
 				continue;
 			}
 
 			if (rxb < 0) {	/* Invalid frame */
 				rxb = -rxb;
 				if (rxb & DST_OVR) {
-					stats->rx_errors++;
-					stats->rx_fifo_errors++;
+					dev->stats.rx_errors++;
+					dev->stats.rx_fifo_errors++;
 				}
 				if (rxb & DST_CRC) {
-					stats->rx_errors++;
-					stats->rx_crc_errors++;
+					dev->stats.rx_errors++;
+					dev->stats.rx_crc_errors++;
 				}
 				if (rxb & (DST_RBIT | DST_SHRT | DST_ABT)) {
-					stats->rx_errors++;
-					stats->rx_frame_errors++;
+					dev->stats.rx_errors++;
+					dev->stats.rx_frame_errors++;
 				}
 			}
 			if (skb) {
@@ -1948,7 +1944,7 @@ static void cpc_net_rx(struct net_device *dev)
 			continue;
 		}
 
-		stats->rx_bytes += rxb;
+		dev->stats.rx_bytes += rxb;
 
 #ifdef PC300_DEBUG_RX
 		printk("%s R:", dev->name);
@@ -1959,7 +1955,7 @@ static void cpc_net_rx(struct net_device *dev)
 		if (d->trace_on) {
 			cpc_trace(dev, skb, 'R');
 		}
-		stats->rx_packets++;
+		dev->stats.rx_packets++;
 		skb->protocol = hdlc_type_trans(skb, dev);
 		netif_rx(skb);
 	}
@@ -1974,16 +1970,15 @@ static void sca_tx_intr(pc300dev_t *dev)
 	pc300_t *card = (pc300_t *)chan->card; 
 	int ch = chan->channel; 
 	volatile pcsca_bd_t __iomem * ptdescr; 
-	struct net_device_stats *stats = hdlc_stats(dev->dev);
 
     /* Clean up descriptors from previous transmission */
 	ptdescr = (card->hw.rambase +
 						TX_BD_ADDR(ch,chan->tx_first_bd));
-	while ((cpc_readl(card->hw.scabase + DTX_REG(CDAL,ch)) != 
-							TX_BD_ADDR(ch,chan->tx_first_bd)) && 
-			(cpc_readb(&ptdescr->status) & DST_OSB)) {
-		stats->tx_packets++;
-		stats->tx_bytes += cpc_readw(&ptdescr->len);
+	while ((cpc_readl(card->hw.scabase + DTX_REG(CDAL,ch)) !=
+		TX_BD_ADDR(ch,chan->tx_first_bd)) &&
+	       (cpc_readb(&ptdescr->status) & DST_OSB)) {
+		dev->dev->stats.tx_packets++;
+		dev->dev->stats.tx_bytes += cpc_readw(&ptdescr->len);
 		cpc_writeb(&ptdescr->status, DST_OSB);
 		cpc_writew(&ptdescr->len, 0);
 		chan->nfree_tx_bd++;
@@ -2048,8 +2043,8 @@ static void sca_intr(pc300_t * card)
 							}
 							cpc_net_rx(dev);
 							/* Discard invalid frames */
-							hdlc_stats(dev)->rx_errors++;
-							hdlc_stats(dev)->rx_over_errors++;
+							dev->stats.rx_errors++;
+							dev->stats.rx_over_errors++;
 							chan->rx_first_bd = 0;
 							chan->rx_last_bd = N_DMA_RX_BUF - 1;
 							rx_dma_start(card, ch);
@@ -2115,8 +2110,8 @@ static void sca_intr(pc300_t * card)
 										   card->hw.cpld_reg2) &
 								   ~ (CPLD_REG2_FALC_LED1 << (2 * ch)));
 						}
-						hdlc_stats(dev)->tx_errors++;
-						hdlc_stats(dev)->tx_fifo_errors++;
+						dev->stats.tx_errors++;
+						dev->stats.tx_fifo_errors++;
 						sca_tx_intr(d);
 					}
 				}
@@ -2604,7 +2599,7 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		case SIOCGPC300UTILSTATS:
 			{
 				if (!arg) {	/* clear statistics */
-					memset(hdlc_stats(dev), 0, sizeof(struct net_device_stats));
+					memset(&dev->stats, 0, sizeof(dev->stats));
 					if (card->hw.type == PC300_TE) {
 						memset(&chan->falc, 0, sizeof(falc_t));
 					}
@@ -2615,8 +2610,8 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 					pc300stats.hw_type = card->hw.type;
 					pc300stats.line_on = card->chan[ch].d.line_on;
 					pc300stats.line_off = card->chan[ch].d.line_off;
-					memcpy(&pc300stats.gen_stats, hdlc_stats(dev),
-					       sizeof(struct net_device_stats));
+					memcpy(&pc300stats.gen_stats, &dev->stats,
+					       sizeof(dev->stats));
 					if (card->hw.type == PC300_TE)
 						memcpy(&pc300stats.te_stats,&chan->falc,sizeof(falc_t));
 				    	if (copy_to_user(arg, &pc300stats, sizeof(pc300stats_t)))
@@ -2823,11 +2818,6 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	}
 }
 
-static struct net_device_stats *cpc_get_stats(struct net_device *dev)
-{
-	return hdlc_stats(dev);
-}
-
 static int clock_rate_calc(uclong rate, uclong clock, int *br_io)
 {
 	int br, tc;
@@ -3394,7 +3384,6 @@ static void cpc_init_card(pc300_t * card)
 		dev->stop = cpc_close;
 		dev->tx_timeout = cpc_tx_timeout;
 		dev->watchdog_timeo = PC300_TX_TIMEOUT;
-		dev->get_stats = cpc_get_stats;
 		dev->set_multicast_list = NULL;
 		dev->set_mac_address = NULL;
 		dev->change_mtu = cpc_change_mtu;
diff --git a/drivers/net/wan/pc300_tty.c b/drivers/net/wan/pc300_tty.c
index e03eef2f2282..eae94ab6b818 100644
--- a/drivers/net/wan/pc300_tty.c
+++ b/drivers/net/wan/pc300_tty.c
@@ -458,7 +458,7 @@ static int cpc_tty_write(struct tty_struct *tty, const unsigned char *buf, int c
 	CPC_TTY_DBG("%s: cpc_tty_write data len=%i\n",cpc_tty->name,count);
 	
 	pc300chan = (pc300ch_t *)((pc300dev_t*)cpc_tty->pc300dev)->chan; 
-	stats = hdlc_stats(((pc300dev_t*)cpc_tty->pc300dev)->dev);
+	stats = &cpc_tty->pc300dev->dev->stats;
 	card = (pc300_t *) pc300chan->card;
 	ch = pc300chan->channel; 
 
@@ -743,7 +743,7 @@ void cpc_tty_receive(pc300dev_t *pc300dev)
 	pc300_t *card = (pc300_t *)pc300chan->card; 
 	int ch = pc300chan->channel; 
 	volatile pcsca_bd_t  __iomem * ptdescr; 
-	struct net_device_stats *stats = hdlc_stats(pc300dev->dev);
+	struct net_device_stats *stats = &pc300dev->dev->stats;
 	int rx_len, rx_aux; 
 	volatile unsigned char status; 
 	unsigned short first_bd = pc300chan->rx_first_bd;
@@ -917,7 +917,7 @@ static int cpc_tty_send_to_card(pc300dev_t *dev,void* buf, int len)
 	pc300ch_t *chan = (pc300ch_t *)dev->chan; 
 	pc300_t *card = (pc300_t *)chan->card; 
 	int ch = chan->channel; 
-	struct net_device_stats *stats = hdlc_stats(dev->dev);
+	struct net_device_stats *stats = &dev->dev->stats;
 	unsigned long flags; 
 	volatile pcsca_bd_t __iomem *ptdescr; 
 	int i, nchar;
diff --git a/drivers/net/wan/wanxl.c b/drivers/net/wan/wanxl.c
index d4aab8a28b61..a8a5ca0ee6c2 100644
--- a/drivers/net/wan/wanxl.c
+++ b/drivers/net/wan/wanxl.c
@@ -161,7 +161,6 @@ static inline void wanxl_cable_intr(port_t *port)
 static inline void wanxl_tx_intr(port_t *port)
 {
 	struct net_device *dev = port->dev;
-	struct net_device_stats *stats = hdlc_stats(dev);
 	while (1) {
                 desc_t *desc = &get_status(port)->tx_descs[port->tx_in];
 		struct sk_buff *skb = port->tx_skbs[port->tx_in];
@@ -173,13 +172,13 @@ static inline void wanxl_tx_intr(port_t *port)
 			return;
 
 		case PACKET_UNDERRUN:
-			stats->tx_errors++;
-			stats->tx_fifo_errors++;
+			dev->stats.tx_errors++;
+			dev->stats.tx_fifo_errors++;
 			break;
 
 		default:
-			stats->tx_packets++;
-			stats->tx_bytes += skb->len;
+			dev->stats.tx_packets++;
+			dev->stats.tx_bytes += skb->len;
 		}
                 desc->stat = PACKET_EMPTY; /* Free descriptor */
 		pci_unmap_single(port->card->pdev, desc->address, skb->len,
@@ -205,10 +204,9 @@ static inline void wanxl_rx_intr(card_t *card)
 			port_t *port = &card->ports[desc->stat &
 						    PACKET_PORT_MASK];
 			struct net_device *dev = port->dev;
-			struct net_device_stats *stats = hdlc_stats(dev);
 
 			if (!skb)
-				stats->rx_dropped++;
+				dev->stats.rx_dropped++;
 			else {
 				pci_unmap_single(card->pdev, desc->address,
 						 BUFFER_LENGTH,
@@ -220,8 +218,8 @@ static inline void wanxl_rx_intr(card_t *card)
 				       skb->len);
 				debug_frame(skb);
 #endif
-				stats->rx_packets++;
-				stats->rx_bytes += skb->len;
+				dev->stats.rx_packets++;
+				dev->stats.rx_bytes += skb->len;
 				dev->last_rx = jiffies;
 				skb->protocol = hdlc_type_trans(skb, dev);
 				netif_rx(skb);
@@ -468,13 +466,13 @@ static int wanxl_close(struct net_device *dev)
 
 static struct net_device_stats *wanxl_get_stats(struct net_device *dev)
 {
-	struct net_device_stats *stats = hdlc_stats(dev);
 	port_t *port = dev_to_port(dev);
 
-	stats->rx_over_errors = get_status(port)->rx_overruns;
-	stats->rx_frame_errors = get_status(port)->rx_frame_errors;
-	stats->rx_errors = stats->rx_over_errors + stats->rx_frame_errors;
-        return stats;
+	dev->stats.rx_over_errors = get_status(port)->rx_overruns;
+	dev->stats.rx_frame_errors = get_status(port)->rx_frame_errors;
+	dev->stats.rx_errors = dev->stats.rx_over_errors +
+		dev->stats.rx_frame_errors;
+	return &dev->stats;
 }
 
 
diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h
index 6115545a5b9c..c59769693bee 100644
--- a/include/linux/hdlc.h
+++ b/include/linux/hdlc.h
@@ -45,7 +45,6 @@ struct hdlc_proto {
 
 /* Pointed to by dev->priv */
 typedef struct hdlc_device {
-	struct net_device_stats stats;
 	/* used by HDLC layer to take control over HDLC device from hw driver*/
 	int (*attach)(struct net_device *dev,
 		      unsigned short encoding, unsigned short parity);
@@ -109,12 +108,6 @@ int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto,
 /* May be used by hardware driver to gain control over HDLC device */
 void detach_hdlc_protocol(struct net_device *dev);
 
-static __inline__ struct net_device_stats *hdlc_stats(struct net_device *dev)
-{
-	return &dev_to_hdlc(dev)->stats;
-}
-
-
 static __inline__ __be16 hdlc_type_trans(struct sk_buff *skb,
 					 struct net_device *dev)
 {
-- 
cgit v1.2.3


From eca9ebac651f774d8b10fce7c5d173c3c3d3394f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Jul 2008 21:26:13 -0700
Subject: net: Add GARP applicant-only participant

Add an implementation of the GARP (Generic Attribute Registration Protocol)
applicant-only participant. This will be used by the following patch to
add GVRP support to the VLAN code.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |   2 +
 include/net/garp.h        | 127 ++++++++++
 net/802/Kconfig           |   4 +
 net/802/Makefile          |   1 +
 net/802/garp.c            | 633 ++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 767 insertions(+)
 create mode 100644 include/net/garp.h
 create mode 100644 net/802/garp.c

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 56dadb528f67..e009c6fbf5cd 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -740,6 +740,8 @@ struct net_device
 	struct net_bridge_port	*br_port;
 	/* macvlan */
 	struct macvlan_port	*macvlan_port;
+	/* GARP */
+	struct garp_port	*garp_port;
 
 	/* class/net/name entry */
 	struct device		dev;
diff --git a/include/net/garp.h b/include/net/garp.h
new file mode 100644
index 000000000000..73c772395f5b
--- /dev/null
+++ b/include/net/garp.h
@@ -0,0 +1,127 @@
+#ifndef _NET_GARP_H
+#define _NET_GARP_H
+
+#include <net/stp.h>
+
+#define GARP_PROTOCOL_ID	0x1
+#define GARP_END_MARK		0x0
+
+struct garp_pdu_hdr {
+	__be16	protocol;
+};
+
+struct garp_msg_hdr {
+	u8	attrtype;
+};
+
+enum garp_attr_event {
+	GARP_LEAVE_ALL,
+	GARP_JOIN_EMPTY,
+	GARP_JOIN_IN,
+	GARP_LEAVE_EMPTY,
+	GARP_LEAVE_IN,
+	GARP_EMPTY,
+};
+
+struct garp_attr_hdr {
+	u8	len;
+	u8	event;
+	u8	data[];
+};
+
+struct garp_skb_cb {
+	u8	cur_type;
+};
+
+static inline struct garp_skb_cb *garp_cb(struct sk_buff *skb)
+{
+	BUILD_BUG_ON(sizeof(struct garp_skb_cb) >
+		     FIELD_SIZEOF(struct sk_buff, cb));
+	return (struct garp_skb_cb *)skb->cb;
+}
+
+enum garp_applicant_state {
+	GARP_APPLICANT_INVALID,
+	GARP_APPLICANT_VA,
+	GARP_APPLICANT_AA,
+	GARP_APPLICANT_QA,
+	GARP_APPLICANT_LA,
+	GARP_APPLICANT_VP,
+	GARP_APPLICANT_AP,
+	GARP_APPLICANT_QP,
+	GARP_APPLICANT_VO,
+	GARP_APPLICANT_AO,
+	GARP_APPLICANT_QO,
+	__GARP_APPLICANT_MAX
+};
+#define GARP_APPLICANT_MAX	(__GARP_APPLICANT_MAX - 1)
+
+enum garp_event {
+	GARP_EVENT_REQ_JOIN,
+	GARP_EVENT_REQ_LEAVE,
+	GARP_EVENT_R_JOIN_IN,
+	GARP_EVENT_R_JOIN_EMPTY,
+	GARP_EVENT_R_EMPTY,
+	GARP_EVENT_R_LEAVE_IN,
+	GARP_EVENT_R_LEAVE_EMPTY,
+	GARP_EVENT_TRANSMIT_PDU,
+	__GARP_EVENT_MAX
+};
+#define GARP_EVENT_MAX		(__GARP_EVENT_MAX - 1)
+
+enum garp_action {
+	GARP_ACTION_NONE,
+	GARP_ACTION_S_JOIN_IN,
+	GARP_ACTION_S_LEAVE_EMPTY,
+};
+
+struct garp_attr {
+	struct rb_node			node;
+	enum garp_applicant_state	state;
+	u8				type;
+	u8				dlen;
+	unsigned char			data[];
+};
+
+enum garp_applications {
+	__GARP_APPLICATION_MAX
+};
+#define GARP_APPLICATION_MAX	(__GARP_APPLICATION_MAX - 1)
+
+struct garp_application {
+	enum garp_applications	type;
+	unsigned int		maxattr;
+	struct stp_proto	proto;
+};
+
+struct garp_applicant {
+	struct garp_application	*app;
+	struct net_device	*dev;
+	struct timer_list	join_timer;
+
+	spinlock_t		lock;
+	struct sk_buff_head	queue;
+	struct sk_buff		*pdu;
+	struct rb_root		gid;
+};
+
+struct garp_port {
+	struct garp_applicant	*applicants[GARP_APPLICATION_MAX + 1];
+};
+
+extern int	garp_register_application(struct garp_application *app);
+extern void	garp_unregister_application(struct garp_application *app);
+
+extern int	garp_init_applicant(struct net_device *dev,
+				    struct garp_application *app);
+extern void	garp_uninit_applicant(struct net_device *dev,
+				      struct garp_application *app);
+
+extern int	garp_request_join(const struct net_device *dev,
+				  const struct garp_application *app,
+				  const void *data, u8 len, u8 type);
+extern void	garp_request_leave(const struct net_device *dev,
+				   const struct garp_application *app,
+				   const void *data, u8 len, u8 type);
+
+#endif /* _NET_GARP_H */
diff --git a/net/802/Kconfig b/net/802/Kconfig
index 01cb0943626b..be33d27c8e69 100644
--- a/net/802/Kconfig
+++ b/net/802/Kconfig
@@ -1,3 +1,7 @@
 config STP
 	tristate
 	select LLC
+
+config GARP
+	tristate
+	select STP
diff --git a/net/802/Makefile b/net/802/Makefile
index c441d895ac25..7893d679910c 100644
--- a/net/802/Makefile
+++ b/net/802/Makefile
@@ -11,3 +11,4 @@ obj-$(CONFIG_HIPPI)	+=                 hippi.o
 obj-$(CONFIG_IPX)	+= p8022.o psnap.o p8023.o
 obj-$(CONFIG_ATALK)	+= p8022.o psnap.o
 obj-$(CONFIG_STP)	+= stp.o
+obj-$(CONFIG_GARP)	+= garp.o
diff --git a/net/802/garp.c b/net/802/garp.c
new file mode 100644
index 000000000000..3b78f7b74fd4
--- /dev/null
+++ b/net/802/garp.c
@@ -0,0 +1,633 @@
+/*
+ *	IEEE 802.1D Generic Attribute Registration Protocol (GARP)
+ *
+ *	Copyright (c) 2008 Patrick McHardy <kaber@trash.net>
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	version 2 as published by the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/timer.h>
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/llc.h>
+#include <net/llc.h>
+#include <net/llc_pdu.h>
+#include <net/garp.h>
+#include <asm/unaligned.h>
+
+static unsigned int garp_join_time __read_mostly = 200;
+module_param(garp_join_time, uint, 0644);
+MODULE_PARM_DESC(garp_join_time, "Join time in ms (default 200ms)");
+MODULE_LICENSE("GPL");
+
+static const struct garp_state_trans {
+	u8	state;
+	u8	action;
+} garp_applicant_state_table[GARP_APPLICANT_MAX + 1][GARP_EVENT_MAX + 1] = {
+	[GARP_APPLICANT_VA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_AA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_AA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_LA },
+	},
+	[GARP_APPLICANT_AA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_QA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_LA },
+	},
+	[GARP_APPLICANT_QA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_LA },
+	},
+	[GARP_APPLICANT_LA] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_VO,
+						    .action = GARP_ACTION_S_LEAVE_EMPTY },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_LA },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_LA },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_LA },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_VA },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+	[GARP_APPLICANT_VP] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_AA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_AP },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_VO },
+	},
+	[GARP_APPLICANT_AP] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_QA,
+						    .action = GARP_ACTION_S_JOIN_IN },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QP },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_AO },
+	},
+	[GARP_APPLICANT_QP] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QP },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_QO },
+	},
+	[GARP_APPLICANT_VO] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_AO },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_VP },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+	[GARP_APPLICANT_AO] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QO },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_AP },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+	[GARP_APPLICANT_QO] = {
+		[GARP_EVENT_TRANSMIT_PDU]	= { .state = GARP_APPLICANT_INVALID },
+		[GARP_EVENT_R_JOIN_IN]		= { .state = GARP_APPLICANT_QO },
+		[GARP_EVENT_R_JOIN_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_EMPTY]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_IN]		= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_R_LEAVE_EMPTY]	= { .state = GARP_APPLICANT_VO },
+		[GARP_EVENT_REQ_JOIN]		= { .state = GARP_APPLICANT_QP },
+		[GARP_EVENT_REQ_LEAVE]		= { .state = GARP_APPLICANT_INVALID },
+	},
+};
+
+static int garp_attr_cmp(const struct garp_attr *attr,
+			 const void *data, u8 len, u8 type)
+{
+	if (attr->type != type)
+		return attr->type - type;
+	if (attr->dlen != len)
+		return attr->dlen - len;
+	return memcmp(attr->data, data, len);
+}
+
+static struct garp_attr *garp_attr_lookup(const struct garp_applicant *app,
+					  const void *data, u8 len, u8 type)
+{
+	struct rb_node *parent = app->gid.rb_node;
+	struct garp_attr *attr;
+	int d;
+
+	while (parent) {
+		attr = rb_entry(parent, struct garp_attr, node);
+		d = garp_attr_cmp(attr, data, len, type);
+		if (d < 0)
+			parent = parent->rb_left;
+		else if (d > 0)
+			parent = parent->rb_right;
+		else
+			return attr;
+	}
+	return NULL;
+}
+
+static void garp_attr_insert(struct garp_applicant *app, struct garp_attr *new)
+{
+	struct rb_node *parent = NULL, **p = &app->gid.rb_node;
+	struct garp_attr *attr;
+	int d;
+
+	while (*p) {
+		parent = *p;
+		attr = rb_entry(parent, struct garp_attr, node);
+		d = garp_attr_cmp(attr, new->data, new->dlen, new->type);
+		if (d < 0)
+			p = &parent->rb_left;
+		else if (d > 0)
+			p = &parent->rb_right;
+	}
+	rb_link_node(&new->node, parent, p);
+	rb_insert_color(&new->node, &app->gid);
+}
+
+static struct garp_attr *garp_attr_create(struct garp_applicant *app,
+					  const void *data, u8 len, u8 type)
+{
+	struct garp_attr *attr;
+
+	attr = kmalloc(sizeof(*attr) + len, GFP_ATOMIC);
+	if (!attr)
+		return attr;
+	attr->state = GARP_APPLICANT_VO;
+	attr->type  = type;
+	attr->dlen  = len;
+	memcpy(attr->data, data, len);
+	garp_attr_insert(app, attr);
+	return attr;
+}
+
+static void garp_attr_destroy(struct garp_applicant *app, struct garp_attr *attr)
+{
+	rb_erase(&attr->node, &app->gid);
+	kfree(attr);
+}
+
+static int garp_pdu_init(struct garp_applicant *app)
+{
+	struct sk_buff *skb;
+	struct garp_pdu_hdr *gp;
+
+#define LLC_RESERVE	sizeof(struct llc_pdu_un)
+	skb = alloc_skb(app->dev->mtu + LL_RESERVED_SPACE(app->dev),
+			GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	skb->dev = app->dev;
+	skb->protocol = htons(ETH_P_802_2);
+	skb_reserve(skb, LL_RESERVED_SPACE(app->dev) + LLC_RESERVE);
+
+	gp = (struct garp_pdu_hdr *)__skb_put(skb, sizeof(*gp));
+	put_unaligned(htons(GARP_PROTOCOL_ID), &gp->protocol);
+
+	app->pdu = skb;
+	return 0;
+}
+
+static int garp_pdu_append_end_mark(struct garp_applicant *app)
+{
+	if (skb_tailroom(app->pdu) < sizeof(u8))
+		return -1;
+	*(u8 *)__skb_put(app->pdu, sizeof(u8)) = GARP_END_MARK;
+	return 0;
+}
+
+static void garp_pdu_queue(struct garp_applicant *app)
+{
+	if (!app->pdu)
+		return;
+
+	garp_pdu_append_end_mark(app);
+	garp_pdu_append_end_mark(app);
+
+	llc_pdu_header_init(app->pdu, LLC_PDU_TYPE_U, LLC_SAP_BSPAN,
+			    LLC_SAP_BSPAN, LLC_PDU_CMD);
+	llc_pdu_init_as_ui_cmd(app->pdu);
+	llc_mac_hdr_init(app->pdu, app->dev->dev_addr,
+			 app->app->proto.group_address);
+
+	skb_queue_tail(&app->queue, app->pdu);
+	app->pdu = NULL;
+}
+
+static void garp_queue_xmit(struct garp_applicant *app)
+{
+	struct sk_buff *skb;
+
+	while ((skb = skb_dequeue(&app->queue)))
+		dev_queue_xmit(skb);
+}
+
+static int garp_pdu_append_msg(struct garp_applicant *app, u8 attrtype)
+{
+	struct garp_msg_hdr *gm;
+
+	if (skb_tailroom(app->pdu) < sizeof(*gm))
+		return -1;
+	gm = (struct garp_msg_hdr *)__skb_put(app->pdu, sizeof(*gm));
+	gm->attrtype = attrtype;
+	garp_cb(app->pdu)->cur_type = attrtype;
+	return 0;
+}
+
+static int garp_pdu_append_attr(struct garp_applicant *app,
+				const struct garp_attr *attr,
+				enum garp_attr_event event)
+{
+	struct garp_attr_hdr *ga;
+	unsigned int len;
+	int err;
+again:
+	if (!app->pdu) {
+		err = garp_pdu_init(app);
+		if (err < 0)
+			return err;
+	}
+
+	if (garp_cb(app->pdu)->cur_type != attr->type) {
+		if (garp_cb(app->pdu)->cur_type &&
+		    garp_pdu_append_end_mark(app) < 0)
+			goto queue;
+		if (garp_pdu_append_msg(app, attr->type) < 0)
+			goto queue;
+	}
+
+	len = sizeof(*ga) + attr->dlen;
+	if (skb_tailroom(app->pdu) < len)
+		goto queue;
+	ga = (struct garp_attr_hdr *)__skb_put(app->pdu, len);
+	ga->len   = len;
+	ga->event = event;
+	memcpy(ga->data, attr->data, attr->dlen);
+	return 0;
+
+queue:
+	garp_pdu_queue(app);
+	goto again;
+}
+
+static void garp_attr_event(struct garp_applicant *app,
+			    struct garp_attr *attr, enum garp_event event)
+{
+	enum garp_applicant_state state;
+
+	state = garp_applicant_state_table[attr->state][event].state;
+	if (state == GARP_APPLICANT_INVALID)
+		return;
+
+	switch (garp_applicant_state_table[attr->state][event].action) {
+	case GARP_ACTION_NONE:
+		break;
+	case GARP_ACTION_S_JOIN_IN:
+		garp_pdu_append_attr(app, attr, GARP_JOIN_IN);
+		break;
+	case GARP_ACTION_S_LEAVE_EMPTY:
+		garp_pdu_append_attr(app, attr, GARP_LEAVE_EMPTY);
+		/* As a pure applicant, sending a leave message implies that
+		 * the attribute was unregistered and can be destroyed. */
+		garp_attr_destroy(app, attr);
+		return;
+	default:
+		WARN_ON(1);
+	}
+
+	attr->state = state;
+}
+
+int garp_request_join(const struct net_device *dev,
+		      const struct garp_application *appl,
+		      const void *data, u8 len, u8 type)
+{
+	struct garp_port *port = dev->garp_port;
+	struct garp_applicant *app = port->applicants[appl->type];
+	struct garp_attr *attr;
+
+	spin_lock_bh(&app->lock);
+	attr = garp_attr_create(app, data, len, type);
+	if (!attr) {
+		spin_unlock_bh(&app->lock);
+		return -ENOMEM;
+	}
+	garp_attr_event(app, attr, GARP_EVENT_REQ_JOIN);
+	spin_unlock_bh(&app->lock);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(garp_request_join);
+
+void garp_request_leave(const struct net_device *dev,
+			const struct garp_application *appl,
+			const void *data, u8 len, u8 type)
+{
+	struct garp_port *port = dev->garp_port;
+	struct garp_applicant *app = port->applicants[appl->type];
+	struct garp_attr *attr;
+
+	spin_lock_bh(&app->lock);
+	attr = garp_attr_lookup(app, data, len, type);
+	if (!attr) {
+		spin_unlock_bh(&app->lock);
+		return;
+	}
+	garp_attr_event(app, attr, GARP_EVENT_REQ_LEAVE);
+	spin_unlock_bh(&app->lock);
+}
+EXPORT_SYMBOL_GPL(garp_request_leave);
+
+static void garp_gid_event(struct garp_applicant *app, enum garp_event event)
+{
+	struct rb_node *node, *next;
+	struct garp_attr *attr;
+
+	for (node = rb_first(&app->gid);
+	     next = node ? rb_next(node) : NULL, node != NULL;
+	     node = next) {
+		attr = rb_entry(node, struct garp_attr, node);
+		garp_attr_event(app, attr, event);
+	}
+}
+
+static void garp_join_timer_arm(struct garp_applicant *app)
+{
+	unsigned long delay;
+
+	delay = (u64)msecs_to_jiffies(garp_join_time) * net_random() >> 32;
+	mod_timer(&app->join_timer, jiffies + delay);
+}
+
+static void garp_join_timer(unsigned long data)
+{
+	struct garp_applicant *app = (struct garp_applicant *)data;
+
+	spin_lock(&app->lock);
+	garp_gid_event(app, GARP_EVENT_TRANSMIT_PDU);
+	garp_pdu_queue(app);
+	spin_unlock(&app->lock);
+
+	garp_queue_xmit(app);
+	garp_join_timer_arm(app);
+}
+
+static int garp_pdu_parse_end_mark(struct sk_buff *skb)
+{
+	if (!pskb_may_pull(skb, sizeof(u8)))
+		return -1;
+	if (*skb->data == GARP_END_MARK) {
+		skb_pull(skb, sizeof(u8));
+		return -1;
+	}
+	return 0;
+}
+
+static int garp_pdu_parse_attr(struct garp_applicant *app, struct sk_buff *skb,
+			       u8 attrtype)
+{
+	const struct garp_attr_hdr *ga;
+	struct garp_attr *attr;
+	enum garp_event event;
+	unsigned int dlen;
+
+	if (!pskb_may_pull(skb, sizeof(*ga)))
+		return -1;
+	ga = (struct garp_attr_hdr *)skb->data;
+	if (ga->len < sizeof(*ga))
+		return -1;
+
+	if (!pskb_may_pull(skb, ga->len))
+		return -1;
+	skb_pull(skb, ga->len);
+	dlen = sizeof(*ga) - ga->len;
+
+	if (attrtype > app->app->maxattr)
+		return 0;
+
+	switch (ga->event) {
+	case GARP_LEAVE_ALL:
+		if (dlen != 0)
+			return -1;
+		garp_gid_event(app, GARP_EVENT_R_LEAVE_EMPTY);
+		return 0;
+	case GARP_JOIN_EMPTY:
+		event = GARP_EVENT_R_JOIN_EMPTY;
+		break;
+	case GARP_JOIN_IN:
+		event = GARP_EVENT_R_JOIN_IN;
+		break;
+	case GARP_LEAVE_EMPTY:
+		event = GARP_EVENT_R_LEAVE_EMPTY;
+		break;
+	case GARP_EMPTY:
+		event = GARP_EVENT_R_EMPTY;
+		break;
+	default:
+		return 0;
+	}
+
+	if (dlen == 0)
+		return -1;
+	attr = garp_attr_lookup(app, ga->data, dlen, attrtype);
+	if (attr == NULL)
+		return 0;
+	garp_attr_event(app, attr, event);
+	return 0;
+}
+
+static int garp_pdu_parse_msg(struct garp_applicant *app, struct sk_buff *skb)
+{
+	const struct garp_msg_hdr *gm;
+
+	if (!pskb_may_pull(skb, sizeof(*gm)))
+		return -1;
+	gm = (struct garp_msg_hdr *)skb->data;
+	if (gm->attrtype == 0)
+		return -1;
+	skb_pull(skb, sizeof(*gm));
+
+	while (skb->len > 0) {
+		if (garp_pdu_parse_attr(app, skb, gm->attrtype) < 0)
+			return -1;
+		if (garp_pdu_parse_end_mark(skb) < 0)
+			break;
+	}
+	return 0;
+}
+
+static void garp_pdu_rcv(const struct stp_proto *proto, struct sk_buff *skb,
+			 struct net_device *dev)
+{
+	struct garp_application *appl = proto->data;
+	struct garp_port *port;
+	struct garp_applicant *app;
+	const struct garp_pdu_hdr *gp;
+
+	port = rcu_dereference(dev->garp_port);
+	if (!port)
+		goto err;
+	app = rcu_dereference(port->applicants[appl->type]);
+	if (!app)
+		goto err;
+
+	if (!pskb_may_pull(skb, sizeof(*gp)))
+		goto err;
+	gp = (struct garp_pdu_hdr *)skb->data;
+	if (get_unaligned(&gp->protocol) != htons(GARP_PROTOCOL_ID))
+		goto err;
+	skb_pull(skb, sizeof(*gp));
+
+	spin_lock(&app->lock);
+	while (skb->len > 0) {
+		if (garp_pdu_parse_msg(app, skb) < 0)
+			break;
+		if (garp_pdu_parse_end_mark(skb) < 0)
+			break;
+	}
+	spin_unlock(&app->lock);
+err:
+	kfree_skb(skb);
+}
+
+static int garp_init_port(struct net_device *dev)
+{
+	struct garp_port *port;
+
+	port = kzalloc(sizeof(*port), GFP_KERNEL);
+	if (!port)
+		return -ENOMEM;
+	rcu_assign_pointer(dev->garp_port, port);
+	return 0;
+}
+
+static void garp_release_port(struct net_device *dev)
+{
+	struct garp_port *port = dev->garp_port;
+	unsigned int i;
+
+	for (i = 0; i <= GARP_APPLICATION_MAX; i++) {
+		if (port->applicants[i])
+			return;
+	}
+	rcu_assign_pointer(dev->garp_port, NULL);
+	synchronize_rcu();
+	kfree(port);
+}
+
+int garp_init_applicant(struct net_device *dev, struct garp_application *appl)
+{
+	struct garp_applicant *app;
+	int err;
+
+	ASSERT_RTNL();
+
+	if (!dev->garp_port) {
+		err = garp_init_port(dev);
+		if (err < 0)
+			goto err1;
+	}
+
+	err = -ENOMEM;
+	app = kzalloc(sizeof(*app), GFP_KERNEL);
+	if (!app)
+		goto err2;
+
+	err = dev_mc_add(dev, appl->proto.group_address, ETH_ALEN, 0);
+	if (err < 0)
+		goto err3;
+
+	app->dev = dev;
+	app->app = appl;
+	app->gid = RB_ROOT;
+	spin_lock_init(&app->lock);
+	skb_queue_head_init(&app->queue);
+	rcu_assign_pointer(dev->garp_port->applicants[appl->type], app);
+	setup_timer(&app->join_timer, garp_join_timer, (unsigned long)app);
+	garp_join_timer_arm(app);
+	return 0;
+
+err3:
+	kfree(app);
+err2:
+	garp_release_port(dev);
+err1:
+	return err;
+}
+EXPORT_SYMBOL_GPL(garp_init_applicant);
+
+void garp_uninit_applicant(struct net_device *dev, struct garp_application *appl)
+{
+	struct garp_port *port = dev->garp_port;
+	struct garp_applicant *app = port->applicants[appl->type];
+
+	ASSERT_RTNL();
+
+	rcu_assign_pointer(port->applicants[appl->type], NULL);
+	synchronize_rcu();
+
+	/* Delete timer and generate a final TRANSMIT_PDU event to flush out
+	 * all pending messages before the applicant is gone. */
+	del_timer_sync(&app->join_timer);
+	garp_gid_event(app, GARP_EVENT_TRANSMIT_PDU);
+	garp_pdu_queue(app);
+	garp_queue_xmit(app);
+
+	dev_mc_delete(dev, appl->proto.group_address, ETH_ALEN, 0);
+	kfree(app);
+	garp_release_port(dev);
+}
+EXPORT_SYMBOL_GPL(garp_uninit_applicant);
+
+int garp_register_application(struct garp_application *appl)
+{
+	appl->proto.rcv = garp_pdu_rcv;
+	appl->proto.data = appl;
+	return stp_proto_register(&appl->proto);
+}
+EXPORT_SYMBOL_GPL(garp_register_application);
+
+void garp_unregister_application(struct garp_application *appl)
+{
+	stp_proto_unregister(&appl->proto);
+}
+EXPORT_SYMBOL_GPL(garp_unregister_application);
-- 
cgit v1.2.3


From 70c03b49b80ba3634958acc31853771019c0ebd3 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 5 Jul 2008 21:26:57 -0700
Subject: vlan: Add GVRP support

Add GVRP support for dynamically registering VLANs with switches.

By default GVRP is disabled because we only support the applicant-only
participant model, which means it should not be enabled on vlans that
are members of a bridge. Since there is currently no way to cleanly
determine that, the user is responsible for enabling it.

The code is pretty small and low impact, its wrapped in a config
option though because it depends on the GARP implementation and
the STP core.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h  |  1 +
 include/net/garp.h       |  1 +
 net/8021q/Kconfig        | 10 ++++++++
 net/8021q/Makefile       |  9 +++----
 net/8021q/vlan.c         | 23 ++++++++++++++---
 net/8021q/vlan.h         | 16 ++++++++++++
 net/8021q/vlan_dev.c     | 18 +++++++++++--
 net/8021q/vlan_gvrp.c    | 66 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/8021q/vlan_netlink.c |  3 ++-
 9 files changed, 134 insertions(+), 13 deletions(-)
 create mode 100644 net/8021q/vlan_gvrp.c

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 15ace02b7b24..5190452ac7dc 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -402,6 +402,7 @@ enum vlan_ioctl_cmds {
 
 enum vlan_flags {
 	VLAN_FLAG_REORDER_HDR	= 0x1,
+	VLAN_FLAG_GVRP		= 0x2,
 };
 
 enum vlan_name_types {
diff --git a/include/net/garp.h b/include/net/garp.h
index 73c772395f5b..825f172caba9 100644
--- a/include/net/garp.h
+++ b/include/net/garp.h
@@ -84,6 +84,7 @@ struct garp_attr {
 };
 
 enum garp_applications {
+	GARP_APPLICATION_GVRP,
 	__GARP_APPLICATION_MAX
 };
 #define GARP_APPLICATION_MAX	(__GARP_APPLICATION_MAX - 1)
diff --git a/net/8021q/Kconfig b/net/8021q/Kconfig
index c4a382e450e2..fa073a54963e 100644
--- a/net/8021q/Kconfig
+++ b/net/8021q/Kconfig
@@ -17,3 +17,13 @@ config VLAN_8021Q
 	  will be called 8021q.
 
 	  If unsure, say N.
+
+config VLAN_8021Q_GVRP
+	bool "GVRP (GARP VLAN Registration Protocol) support"
+	depends on VLAN_8021Q
+	select GARP
+	help
+	  Select this to enable GVRP end-system support. GVRP is used for
+	  automatic propagation of registered VLANs to switches.
+
+	  If unsure, say N.
diff --git a/net/8021q/Makefile b/net/8021q/Makefile
index 10ca7f486c3a..3006e9ed7b08 100644
--- a/net/8021q/Makefile
+++ b/net/8021q/Makefile
@@ -4,9 +4,6 @@
 
 obj-$(CONFIG_VLAN_8021Q) += 8021q.o
 
-8021q-objs := vlan.o vlan_dev.o vlan_netlink.o
-
-ifeq ($(CONFIG_PROC_FS),y)
-8021q-objs += vlanproc.o
-endif
-
+8021q-y				:= vlan.o vlan_dev.o vlan_netlink.o
+8021q-$(CONFIG_VLAN_8021Q_GVRP)	+= vlan_gvrp.o
+8021q-$(CONFIG_PROC_FS)		+= vlanproc.o
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 8cae2daeb1cc..b529110c9355 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -169,6 +169,8 @@ void unregister_vlan_dev(struct net_device *dev)
 
 	/* If the group is now empty, kill off the group. */
 	if (grp->nr_vlans == 0) {
+		vlan_gvrp_uninit_applicant(real_dev);
+
 		if (real_dev->features & NETIF_F_HW_VLAN_RX)
 			real_dev->vlan_rx_register(real_dev, NULL);
 
@@ -249,15 +251,18 @@ int register_vlan_dev(struct net_device *dev)
 		ngrp = grp = vlan_group_alloc(real_dev);
 		if (!grp)
 			return -ENOBUFS;
+		err = vlan_gvrp_init_applicant(real_dev);
+		if (err < 0)
+			goto out_free_group;
 	}
 
 	err = vlan_group_prealloc_vid(grp, vlan_id);
 	if (err < 0)
-		goto out_free_group;
+		goto out_uninit_applicant;
 
 	err = register_netdevice(dev);
 	if (err < 0)
-		goto out_free_group;
+		goto out_uninit_applicant;
 
 	/* Account for reference in struct vlan_dev_info */
 	dev_hold(real_dev);
@@ -278,6 +283,9 @@ int register_vlan_dev(struct net_device *dev)
 
 	return 0;
 
+out_uninit_applicant:
+	if (ngrp)
+		vlan_gvrp_uninit_applicant(real_dev);
 out_free_group:
 	if (ngrp)
 		vlan_group_free(ngrp);
@@ -713,14 +721,20 @@ static int __init vlan_proto_init(void)
 	if (err < 0)
 		goto err2;
 
-	err = vlan_netlink_init();
+	err = vlan_gvrp_init();
 	if (err < 0)
 		goto err3;
 
+	err = vlan_netlink_init();
+	if (err < 0)
+		goto err4;
+
 	dev_add_pack(&vlan_packet_type);
 	vlan_ioctl_set(vlan_ioctl_handler);
 	return 0;
 
+err4:
+	vlan_gvrp_uninit();
 err3:
 	unregister_netdevice_notifier(&vlan_notifier_block);
 err2:
@@ -745,8 +759,9 @@ static void __exit vlan_cleanup_module(void)
 		BUG_ON(!hlist_empty(&vlan_group_hash[i]));
 
 	unregister_pernet_gen_device(vlan_net_id, &vlan_net_ops);
-
 	synchronize_net();
+
+	vlan_gvrp_uninit();
 }
 
 module_init(vlan_proto_init);
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 639e2544a804..097b2e04c928 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -37,6 +37,22 @@ void vlan_setup(struct net_device *dev);
 int register_vlan_dev(struct net_device *dev);
 void unregister_vlan_dev(struct net_device *dev);
 
+#ifdef CONFIG_VLAN_8021Q_GVRP
+extern int vlan_gvrp_request_join(const struct net_device *dev);
+extern void vlan_gvrp_request_leave(const struct net_device *dev);
+extern int vlan_gvrp_init_applicant(struct net_device *dev);
+extern void vlan_gvrp_uninit_applicant(struct net_device *dev);
+extern int vlan_gvrp_init(void);
+extern void vlan_gvrp_uninit(void);
+#else
+static inline int vlan_gvrp_request_join(const struct net_device *dev) { return 0; }
+static inline void vlan_gvrp_request_leave(const struct net_device *dev) {}
+static inline int vlan_gvrp_init_applicant(struct net_device *dev) { return 0; }
+static inline void vlan_gvrp_uninit_applicant(struct net_device *dev) {}
+static inline int vlan_gvrp_init(void) { return 0; }
+static inline void vlan_gvrp_uninit(void) {}
+#endif
+
 int vlan_netlink_init(void);
 void vlan_netlink_fini(void);
 
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 76c665cdab66..a0617bf7cec6 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -512,10 +512,17 @@ int vlan_dev_change_flags(const struct net_device *dev, u32 flags, u32 mask)
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	u32 old_flags = vlan->flags;
 
-	if (mask & ~VLAN_FLAG_REORDER_HDR)
+	if (mask & ~(VLAN_FLAG_REORDER_HDR | VLAN_FLAG_GVRP))
 		return -EINVAL;
 
 	vlan->flags = (old_flags & ~mask) | (flags & mask);
+
+	if (netif_running(dev) && (vlan->flags ^ old_flags) & VLAN_FLAG_GVRP) {
+		if (vlan->flags & VLAN_FLAG_GVRP)
+			vlan_gvrp_request_join(dev);
+		else
+			vlan_gvrp_request_leave(dev);
+	}
 	return 0;
 }
 
@@ -550,12 +557,19 @@ static int vlan_dev_open(struct net_device *dev)
 	if (dev->flags & IFF_PROMISC)
 		dev_set_promiscuity(real_dev, 1);
 
+	if (vlan->flags & VLAN_FLAG_GVRP)
+		vlan_gvrp_request_join(dev);
+
 	return 0;
 }
 
 static int vlan_dev_stop(struct net_device *dev)
 {
-	struct net_device *real_dev = vlan_dev_info(dev)->real_dev;
+	struct vlan_dev_info *vlan = vlan_dev_info(dev);
+	struct net_device *real_dev = vlan->real_dev;
+
+	if (vlan->flags & VLAN_FLAG_GVRP)
+		vlan_gvrp_request_leave(dev);
 
 	dev_mc_unsync(real_dev, dev);
 	dev_unicast_unsync(real_dev, dev);
diff --git a/net/8021q/vlan_gvrp.c b/net/8021q/vlan_gvrp.c
new file mode 100644
index 000000000000..db9781608362
--- /dev/null
+++ b/net/8021q/vlan_gvrp.c
@@ -0,0 +1,66 @@
+/*
+ * 	IEEE 802.1Q GARP VLAN Registration Protocol (GVRP)
+ *
+ * 	Copyright (c) 2008 Patrick McHardy <kaber@trash.net>
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	version 2 as published by the Free Software Foundation.
+ */
+#include <linux/types.h>
+#include <linux/if_vlan.h>
+#include <net/garp.h>
+#include "vlan.h"
+
+#define GARP_GVRP_ADDRESS	{ 0x01, 0x80, 0xc2, 0x00, 0x00, 0x21 }
+
+enum gvrp_attributes {
+	GVRP_ATTR_INVALID,
+	GVRP_ATTR_VID,
+	__GVRP_ATTR_MAX
+};
+#define GVRP_ATTR_MAX	(__GVRP_ATTR_MAX - 1)
+
+static struct garp_application vlan_gvrp_app __read_mostly = {
+	.proto.group_address	= GARP_GVRP_ADDRESS,
+	.maxattr		= GVRP_ATTR_MAX,
+	.type			= GARP_APPLICATION_GVRP,
+};
+
+int vlan_gvrp_request_join(const struct net_device *dev)
+{
+	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
+	__be16 vid = htons(vlan->vlan_id);
+
+	return garp_request_join(vlan->real_dev, &vlan_gvrp_app,
+				 &vid, sizeof(vid), GVRP_ATTR_VID);
+}
+
+void vlan_gvrp_request_leave(const struct net_device *dev)
+{
+	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
+	__be16 vid = htons(vlan->vlan_id);
+
+	garp_request_leave(vlan->real_dev, &vlan_gvrp_app,
+			   &vid, sizeof(vid), GVRP_ATTR_VID);
+}
+
+int vlan_gvrp_init_applicant(struct net_device *dev)
+{
+	return garp_init_applicant(dev, &vlan_gvrp_app);
+}
+
+void vlan_gvrp_uninit_applicant(struct net_device *dev)
+{
+	garp_uninit_applicant(dev, &vlan_gvrp_app);
+}
+
+int __init vlan_gvrp_init(void)
+{
+	return garp_register_application(&vlan_gvrp_app);
+}
+
+void vlan_gvrp_uninit(void)
+{
+	garp_unregister_application(&vlan_gvrp_app);
+}
diff --git a/net/8021q/vlan_netlink.c b/net/8021q/vlan_netlink.c
index fd7cb195d53f..e9c91dcecc9b 100644
--- a/net/8021q/vlan_netlink.c
+++ b/net/8021q/vlan_netlink.c
@@ -59,7 +59,8 @@ static int vlan_validate(struct nlattr *tb[], struct nlattr *data[])
 	}
 	if (data[IFLA_VLAN_FLAGS]) {
 		flags = nla_data(data[IFLA_VLAN_FLAGS]);
-		if ((flags->flags & flags->mask) & ~VLAN_FLAG_REORDER_HDR)
+		if ((flags->flags & flags->mask) &
+		    ~(VLAN_FLAG_REORDER_HDR | VLAN_FLAG_GVRP))
 			return -EINVAL;
 	}
 
-- 
cgit v1.2.3


From 9982fbface82893e77d211fbabfbd229da6bdde6 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 6 Jul 2008 14:24:08 +0200
Subject: Revert "cpumask: introduce new APIs"

This reverts commit acb7669c125676e63cf96582455509216c39745e.

the wrappers are not needed anymore.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/cpumask.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 47418b1b4103..80226e776143 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -394,10 +394,6 @@ int __next_cpu_nr(int n, const cpumask_t *srcp);
 
 #endif /* NR_CPUS > 64 */
 
-#define next_cpu_nr(n, src)		next_cpu(n, src)
-#define cpus_weight_nr(cpumask)		cpus_weight(cpumask)
-#define for_each_cpu_mask_nr(cpu, mask)	for_each_cpu_mask(cpu, mask)
-
 /*
  * The following particular system cpumasks and operations manage
  * possible, present and online cpus.  Each of them is a fixed size
-- 
cgit v1.2.3


From d6315949ac5527efd00d48283a9e33361c86e8e9 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Sun, 22 Jun 2008 12:01:58 +0100
Subject: [ARM] 5096/2: Support Toshiba TC6393XB Mobile I/O Controller.

Add support for Toshiba TC6393XB companion chip. Currently
only GPIO and part of IRQ features of the device are supported.

Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/mfd/Kconfig          |   6 +
 drivers/mfd/Makefile         |   2 +
 drivers/mfd/tc6393xb.c       | 537 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/tc6393xb.h |  47 ++++
 4 files changed, 592 insertions(+)
 create mode 100644 drivers/mfd/tc6393xb.c
 create mode 100644 include/linux/mfd/tc6393xb.h

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 2566479937c9..1a1ac262fc87 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -38,6 +38,12 @@ config HTC_PASIC3
 	  HTC Magician devices, respectively. Actual functionality is
 	  handled by the leds-pasic3 and ds1wm drivers.
 
+config MFD_TC6393XB
+	bool "Support Toshiba TC6393XB"
+	depends on HAVE_GPIO_LIB
+	help
+	  Support for Toshiba Mobile IO Controller TC6393XB
+
 endmenu
 
 menu "Multimedia Capabilities Port drivers"
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index eef4e26807df..b4168442d53c 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -8,6 +8,8 @@ obj-$(CONFIG_MFD_ASIC3)		+= asic3.o
 obj-$(CONFIG_HTC_EGPIO)		+= htc-egpio.o
 obj-$(CONFIG_HTC_PASIC3)	+= htc-pasic3.o
 
+obj-$(CONFIG_MFD_TC6393XB)	+= tc6393xb.o
+
 obj-$(CONFIG_MCP)		+= mcp-core.o
 obj-$(CONFIG_MCP_SA11X0)	+= mcp-sa11x0.o
 obj-$(CONFIG_MCP_UCB1200)	+= ucb1x00-core.o
diff --git a/drivers/mfd/tc6393xb.c b/drivers/mfd/tc6393xb.c
new file mode 100644
index 000000000000..4d7192edefe1
--- /dev/null
+++ b/drivers/mfd/tc6393xb.c
@@ -0,0 +1,537 @@
+/*
+ * Toshiba TC6393XB SoC support
+ *
+ * Copyright(c) 2005-2006 Chris Humbert
+ * Copyright(c) 2005 Dirk Opfer
+ * Copyright(c) 2005 Ian Molton <spyro@f2s.com>
+ * Copyright(c) 2007 Dmitry Baryshkov
+ *
+ * Based on code written by Sharp/Lineo for 2.4 kernels
+ * Based on locomo.c
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+#include <linux/fb.h>
+#include <linux/clk.h>
+#include <linux/mfd/tc6393xb.h>
+#include <linux/gpio.h>
+
+#define SCR_REVID	0x08		/* b Revision ID	*/
+#define SCR_ISR		0x50		/* b Interrupt Status	*/
+#define SCR_IMR		0x52		/* b Interrupt Mask	*/
+#define SCR_IRR		0x54		/* b Interrupt Routing	*/
+#define SCR_GPER	0x60		/* w GP Enable		*/
+#define SCR_GPI_SR(i)	(0x64 + (i))	/* b3 GPI Status	*/
+#define SCR_GPI_IMR(i)	(0x68 + (i))	/* b3 GPI INT Mask	*/
+#define SCR_GPI_EDER(i)	(0x6c + (i))	/* b3 GPI Edge Detect Enable */
+#define SCR_GPI_LIR(i)	(0x70 + (i))	/* b3 GPI Level Invert	*/
+#define SCR_GPO_DSR(i)	(0x78 + (i))	/* b3 GPO Data Set	*/
+#define SCR_GPO_DOECR(i) (0x7c + (i))	/* b3 GPO Data OE Control */
+#define SCR_GP_IARCR(i)	(0x80 + (i))	/* b3 GP Internal Active Register Control */
+#define SCR_GP_IARLCR(i) (0x84 + (i))	/* b3 GP INTERNAL Active Register Level Control */
+#define SCR_GPI_BCR(i)	(0x88 + (i))	/* b3 GPI Buffer Control */
+#define SCR_GPA_IARCR	0x8c		/* w GPa Internal Active Register Control */
+#define SCR_GPA_IARLCR	0x90		/* w GPa Internal Active Register Level Control */
+#define SCR_GPA_BCR	0x94		/* w GPa Buffer Control */
+#define SCR_CCR		0x98		/* w Clock Control	*/
+#define SCR_PLL2CR	0x9a		/* w PLL2 Control	*/
+#define SCR_PLL1CR	0x9c		/* l PLL1 Control	*/
+#define SCR_DIARCR	0xa0		/* b Device Internal Active Register Control */
+#define SCR_DBOCR	0xa1		/* b Device Buffer Off Control */
+#define SCR_FER		0xe0		/* b Function Enable	*/
+#define SCR_MCR		0xe4		/* w Mode Control	*/
+#define SCR_CONFIG	0xfc		/* b Configuration Control */
+#define SCR_DEBUG	0xff		/* b Debug		*/
+
+#define SCR_CCR_CK32K	BIT(0)
+#define SCR_CCR_USBCK	BIT(1)
+#define SCR_CCR_UNK1	BIT(4)
+#define SCR_CCR_MCLK_MASK	(7 << 8)
+#define SCR_CCR_MCLK_OFF	(0 << 8)
+#define SCR_CCR_MCLK_12	(1 << 8)
+#define SCR_CCR_MCLK_24	(2 << 8)
+#define SCR_CCR_MCLK_48	(3 << 8)
+#define SCR_CCR_HCLK_MASK	(3 << 12)
+#define SCR_CCR_HCLK_24	(0 << 12)
+#define SCR_CCR_HCLK_48	(1 << 12)
+
+#define SCR_FER_USBEN		BIT(0)	/* USB host enable */
+#define SCR_FER_LCDCVEN		BIT(1)	/* polysilicon TFT enable */
+#define SCR_FER_SLCDEN		BIT(2)	/* SLCD enable */
+
+#define SCR_MCR_RDY_MASK		(3 << 0)
+#define SCR_MCR_RDY_OPENDRAIN	(0 << 0)
+#define SCR_MCR_RDY_TRISTATE	(1 << 0)
+#define SCR_MCR_RDY_PUSHPULL	(2 << 0)
+#define SCR_MCR_RDY_UNK		BIT(2)
+#define SCR_MCR_RDY_EN		BIT(3)
+#define SCR_MCR_INT_MASK		(3 << 4)
+#define SCR_MCR_INT_OPENDRAIN	(0 << 4)
+#define SCR_MCR_INT_TRISTATE	(1 << 4)
+#define SCR_MCR_INT_PUSHPULL	(2 << 4)
+#define SCR_MCR_INT_UNK		BIT(6)
+#define SCR_MCR_INT_EN		BIT(7)
+/* bits 8 - 16 are unknown */
+
+#define TC_GPIO_BIT(i)		(1 << (i & 0x7))
+
+/*--------------------------------------------------------------------------*/
+
+struct tc6393xb {
+	void __iomem		*scr;
+
+	struct gpio_chip	gpio;
+
+	struct clk		*clk; /* 3,6 Mhz */
+
+	spinlock_t		lock; /* protects RMW cycles */
+
+	struct {
+		u8		fer;
+		u16		ccr;
+		u8		gpi_bcr[3];
+		u8		gpo_dsr[3];
+		u8		gpo_doecr[3];
+	} suspend_state;
+
+	struct resource		rscr;
+	struct resource		*iomem;
+	int			irq;
+	int			irq_base;
+};
+
+/*--------------------------------------------------------------------------*/
+
+static int tc6393xb_gpio_get(struct gpio_chip *chip,
+		unsigned offset)
+{
+	struct tc6393xb *tc6393xb = container_of(chip, struct tc6393xb, gpio);
+
+	/* XXX: does dsr also represent inputs? */
+	return ioread8(tc6393xb->scr + SCR_GPO_DSR(offset / 8))
+		& TC_GPIO_BIT(offset);
+}
+
+static void __tc6393xb_gpio_set(struct gpio_chip *chip,
+		unsigned offset, int value)
+{
+	struct tc6393xb *tc6393xb = container_of(chip, struct tc6393xb, gpio);
+	u8  dsr;
+
+	dsr = ioread8(tc6393xb->scr + SCR_GPO_DSR(offset / 8));
+	if (value)
+		dsr |= TC_GPIO_BIT(offset);
+	else
+		dsr &= ~TC_GPIO_BIT(offset);
+
+	iowrite8(dsr, tc6393xb->scr + SCR_GPO_DSR(offset / 8));
+}
+
+static void tc6393xb_gpio_set(struct gpio_chip *chip,
+		unsigned offset, int value)
+{
+	struct tc6393xb *tc6393xb = container_of(chip, struct tc6393xb, gpio);
+	unsigned long flags;
+
+	spin_lock_irqsave(&tc6393xb->lock, flags);
+
+	__tc6393xb_gpio_set(chip, offset, value);
+
+	spin_unlock_irqrestore(&tc6393xb->lock, flags);
+}
+
+static int tc6393xb_gpio_direction_input(struct gpio_chip *chip,
+			unsigned offset)
+{
+	struct tc6393xb *tc6393xb = container_of(chip, struct tc6393xb, gpio);
+	unsigned long flags;
+	u8 doecr;
+
+	spin_lock_irqsave(&tc6393xb->lock, flags);
+
+	doecr = ioread8(tc6393xb->scr + SCR_GPO_DOECR(offset / 8));
+	doecr &= ~TC_GPIO_BIT(offset);
+	iowrite8(doecr, tc6393xb->scr + SCR_GPO_DOECR(offset / 8));
+
+	spin_unlock_irqrestore(&tc6393xb->lock, flags);
+
+	return 0;
+}
+
+static int tc6393xb_gpio_direction_output(struct gpio_chip *chip,
+			unsigned offset, int value)
+{
+	struct tc6393xb *tc6393xb = container_of(chip, struct tc6393xb, gpio);
+	unsigned long flags;
+	u8 doecr;
+
+	spin_lock_irqsave(&tc6393xb->lock, flags);
+
+	__tc6393xb_gpio_set(chip, offset, value);
+
+	doecr = ioread8(tc6393xb->scr + SCR_GPO_DOECR(offset / 8));
+	doecr |= TC_GPIO_BIT(offset);
+	iowrite8(doecr, tc6393xb->scr + SCR_GPO_DOECR(offset / 8));
+
+	spin_unlock_irqrestore(&tc6393xb->lock, flags);
+
+	return 0;
+}
+
+static int tc6393xb_register_gpio(struct tc6393xb *tc6393xb, int gpio_base)
+{
+	tc6393xb->gpio.label = "tc6393xb";
+	tc6393xb->gpio.base = gpio_base;
+	tc6393xb->gpio.ngpio = 16;
+	tc6393xb->gpio.set = tc6393xb_gpio_set;
+	tc6393xb->gpio.get = tc6393xb_gpio_get;
+	tc6393xb->gpio.direction_input = tc6393xb_gpio_direction_input;
+	tc6393xb->gpio.direction_output = tc6393xb_gpio_direction_output;
+
+	return gpiochip_add(&tc6393xb->gpio);
+}
+
+/*--------------------------------------------------------------------------*/
+
+static void
+tc6393xb_irq(unsigned int irq, struct irq_desc *desc)
+{
+	struct tc6393xb *tc6393xb = get_irq_data(irq);
+	unsigned int isr;
+	unsigned int i, irq_base;
+
+	irq_base = tc6393xb->irq_base;
+
+	while ((isr = ioread8(tc6393xb->scr + SCR_ISR) &
+				~ioread8(tc6393xb->scr + SCR_IMR)))
+		for (i = 0; i < TC6393XB_NR_IRQS; i++) {
+			if (isr & (1 << i))
+				generic_handle_irq(irq_base + i);
+		}
+}
+
+static void tc6393xb_irq_ack(unsigned int irq)
+{
+}
+
+static void tc6393xb_irq_mask(unsigned int irq)
+{
+	struct tc6393xb *tc6393xb = get_irq_chip_data(irq);
+	unsigned long flags;
+	u8 imr;
+
+	spin_lock_irqsave(&tc6393xb->lock, flags);
+	imr = ioread8(tc6393xb->scr + SCR_IMR);
+	imr |= 1 << (irq - tc6393xb->irq_base);
+	iowrite8(imr, tc6393xb->scr + SCR_IMR);
+	spin_unlock_irqrestore(&tc6393xb->lock, flags);
+}
+
+static void tc6393xb_irq_unmask(unsigned int irq)
+{
+	struct tc6393xb *tc6393xb = get_irq_chip_data(irq);
+	unsigned long flags;
+	u8 imr;
+
+	spin_lock_irqsave(&tc6393xb->lock, flags);
+	imr = ioread8(tc6393xb->scr + SCR_IMR);
+	imr &= ~(1 << (irq - tc6393xb->irq_base));
+	iowrite8(imr, tc6393xb->scr + SCR_IMR);
+	spin_unlock_irqrestore(&tc6393xb->lock, flags);
+}
+
+static struct irq_chip tc6393xb_chip = {
+	.name	= "tc6393xb",
+	.ack	= tc6393xb_irq_ack,
+	.mask	= tc6393xb_irq_mask,
+	.unmask	= tc6393xb_irq_unmask,
+};
+
+static void tc6393xb_attach_irq(struct platform_device *dev)
+{
+	struct tc6393xb *tc6393xb = platform_get_drvdata(dev);
+	unsigned int irq, irq_base;
+
+	irq_base = tc6393xb->irq_base;
+
+	for (irq = irq_base; irq < irq_base + TC6393XB_NR_IRQS; irq++) {
+		set_irq_chip(irq, &tc6393xb_chip);
+		set_irq_chip_data(irq, tc6393xb);
+		set_irq_handler(irq, handle_edge_irq);
+		set_irq_flags(irq, IRQF_VALID | IRQF_PROBE);
+	}
+
+	set_irq_type(tc6393xb->irq, IRQT_FALLING);
+	set_irq_data(tc6393xb->irq, tc6393xb);
+	set_irq_chained_handler(tc6393xb->irq, tc6393xb_irq);
+}
+
+static void tc6393xb_detach_irq(struct platform_device *dev)
+{
+	struct tc6393xb *tc6393xb = platform_get_drvdata(dev);
+	unsigned int irq, irq_base;
+
+	set_irq_chained_handler(tc6393xb->irq, NULL);
+	set_irq_data(tc6393xb->irq, NULL);
+
+	irq_base = tc6393xb->irq_base;
+
+	for (irq = irq_base; irq < irq_base + TC6393XB_NR_IRQS; irq++) {
+		set_irq_flags(irq, 0);
+		set_irq_chip(irq, NULL);
+		set_irq_chip_data(irq, NULL);
+	}
+}
+
+/*--------------------------------------------------------------------------*/
+
+static int tc6393xb_hw_init(struct platform_device *dev)
+{
+	struct tc6393xb_platform_data *tcpd = dev->dev.platform_data;
+	struct tc6393xb *tc6393xb = platform_get_drvdata(dev);
+	int i;
+
+	iowrite8(tc6393xb->suspend_state.fer,	tc6393xb->scr + SCR_FER);
+	iowrite16(tcpd->scr_pll2cr,		tc6393xb->scr + SCR_PLL2CR);
+	iowrite16(tc6393xb->suspend_state.ccr,	tc6393xb->scr + SCR_CCR);
+	iowrite16(SCR_MCR_RDY_OPENDRAIN | SCR_MCR_RDY_UNK | SCR_MCR_RDY_EN |
+		  SCR_MCR_INT_OPENDRAIN | SCR_MCR_INT_UNK | SCR_MCR_INT_EN |
+		  BIT(15),			tc6393xb->scr + SCR_MCR);
+	iowrite16(tcpd->scr_gper,		tc6393xb->scr + SCR_GPER);
+	iowrite8(0,				tc6393xb->scr + SCR_IRR);
+	iowrite8(0xbf,				tc6393xb->scr + SCR_IMR);
+
+	for (i = 0; i < 3; i++) {
+		iowrite8(tc6393xb->suspend_state.gpo_dsr[i],
+					tc6393xb->scr + SCR_GPO_DSR(i));
+		iowrite8(tc6393xb->suspend_state.gpo_doecr[i],
+					tc6393xb->scr + SCR_GPO_DOECR(i));
+		iowrite8(tc6393xb->suspend_state.gpi_bcr[i],
+					tc6393xb->scr + SCR_GPI_BCR(i));
+	}
+
+	return 0;
+}
+
+static int __devinit tc6393xb_probe(struct platform_device *dev)
+{
+	struct tc6393xb_platform_data *tcpd = dev->dev.platform_data;
+	struct tc6393xb *tc6393xb;
+	struct resource *iomem;
+	struct resource *rscr;
+	int retval, temp;
+	int i;
+
+	iomem = platform_get_resource(dev, IORESOURCE_MEM, 0);
+	if (!iomem)
+		return -EINVAL;
+
+	tc6393xb = kzalloc(sizeof *tc6393xb, GFP_KERNEL);
+	if (!tc6393xb) {
+		retval = -ENOMEM;
+		goto err_kzalloc;
+	}
+
+	spin_lock_init(&tc6393xb->lock);
+
+	platform_set_drvdata(dev, tc6393xb);
+	tc6393xb->iomem = iomem;
+	tc6393xb->irq = platform_get_irq(dev, 0);
+	tc6393xb->irq_base = tcpd->irq_base;
+
+	tc6393xb->clk = clk_get(&dev->dev, "GPIO27_CLK" /* "CK3P6MI" */);
+	if (IS_ERR(tc6393xb->clk)) {
+		retval = PTR_ERR(tc6393xb->clk);
+		goto err_clk_get;
+	}
+
+	rscr = &tc6393xb->rscr;
+	rscr->name = "tc6393xb-core";
+	rscr->start = iomem->start;
+	rscr->end = iomem->start + 0xff;
+	rscr->flags = IORESOURCE_MEM;
+
+	retval = request_resource(iomem, rscr);
+	if (retval)
+		goto err_request_scr;
+
+	tc6393xb->scr = ioremap(rscr->start, rscr->end - rscr->start + 1);
+	if (!tc6393xb->scr) {
+		retval = -ENOMEM;
+		goto err_ioremap;
+	}
+
+	retval = clk_enable(tc6393xb->clk);
+	if (retval)
+		goto err_clk_enable;
+
+	retval = tcpd->enable(dev);
+	if (retval)
+		goto err_enable;
+
+	tc6393xb->suspend_state.fer = 0;
+	for (i = 0; i < 3; i++) {
+		tc6393xb->suspend_state.gpo_dsr[i] =
+			(tcpd->scr_gpo_dsr >> (8 * i)) & 0xff;
+		tc6393xb->suspend_state.gpo_doecr[i] =
+			(tcpd->scr_gpo_doecr >> (8 * i)) & 0xff;
+	}
+	/*
+	 * It may be necessary to change this back to
+	 * platform-dependant code
+	 */
+	tc6393xb->suspend_state.ccr = SCR_CCR_UNK1 |
+					SCR_CCR_HCLK_48;
+
+	retval = tc6393xb_hw_init(dev);
+	if (retval)
+		goto err_hw_init;
+
+	printk(KERN_INFO "Toshiba tc6393xb revision %d at 0x%08lx, irq %d\n",
+			ioread8(tc6393xb->scr + SCR_REVID),
+			(unsigned long) iomem->start, tc6393xb->irq);
+
+	tc6393xb->gpio.base = -1;
+
+	if (tcpd->gpio_base >= 0) {
+		retval = tc6393xb_register_gpio(tc6393xb, tcpd->gpio_base);
+		if (retval)
+			goto err_gpio_add;
+	}
+
+	if (tc6393xb->irq)
+		tc6393xb_attach_irq(dev);
+
+	return 0;
+
+	if (tc6393xb->irq)
+		tc6393xb_detach_irq(dev);
+
+err_gpio_add:
+	if (tc6393xb->gpio.base != -1)
+		temp = gpiochip_remove(&tc6393xb->gpio);
+err_hw_init:
+	tcpd->disable(dev);
+err_clk_enable:
+	clk_disable(tc6393xb->clk);
+err_enable:
+	iounmap(tc6393xb->scr);
+err_ioremap:
+	release_resource(&tc6393xb->rscr);
+err_request_scr:
+	clk_put(tc6393xb->clk);
+err_clk_get:
+	kfree(tc6393xb);
+err_kzalloc:
+	return retval;
+}
+
+static int __devexit tc6393xb_remove(struct platform_device *dev)
+{
+	struct tc6393xb_platform_data *tcpd = dev->dev.platform_data;
+	struct tc6393xb *tc6393xb = platform_get_drvdata(dev);
+	int ret;
+
+	if (tc6393xb->irq)
+		tc6393xb_detach_irq(dev);
+
+	if (tc6393xb->gpio.base != -1) {
+		ret = gpiochip_remove(&tc6393xb->gpio);
+		if (ret) {
+			dev_err(&dev->dev, "Can't remove gpio chip: %d\n", ret);
+			return ret;
+		}
+	}
+
+	ret = tcpd->disable(dev);
+
+	clk_disable(tc6393xb->clk);
+
+	iounmap(tc6393xb->scr);
+
+	release_resource(&tc6393xb->rscr);
+
+	platform_set_drvdata(dev, NULL);
+
+	clk_put(tc6393xb->clk);
+
+	kfree(tc6393xb);
+
+	return ret;
+}
+
+#ifdef CONFIG_PM
+static int tc6393xb_suspend(struct platform_device *dev, pm_message_t state)
+{
+	struct tc6393xb_platform_data *tcpd = dev->dev.platform_data;
+	struct tc6393xb *tc6393xb = platform_get_drvdata(dev);
+	int i;
+
+
+	tc6393xb->suspend_state.ccr = ioread16(tc6393xb->scr + SCR_CCR);
+	tc6393xb->suspend_state.fer = ioread8(tc6393xb->scr + SCR_FER);
+
+	for (i = 0; i < 3; i++) {
+		tc6393xb->suspend_state.gpo_dsr[i] =
+			ioread8(tc6393xb->scr + SCR_GPO_DSR(i));
+		tc6393xb->suspend_state.gpo_doecr[i] =
+			ioread8(tc6393xb->scr + SCR_GPO_DOECR(i));
+		tc6393xb->suspend_state.gpi_bcr[i] =
+			ioread8(tc6393xb->scr + SCR_GPI_BCR(i));
+	}
+
+	return tcpd->suspend(dev);
+}
+
+static int tc6393xb_resume(struct platform_device *dev)
+{
+	struct tc6393xb_platform_data *tcpd = dev->dev.platform_data;
+	int ret = tcpd->resume(dev);
+
+	if (ret)
+		return ret;
+
+	return tc6393xb_hw_init(dev);
+}
+#else
+#define tc6393xb_suspend NULL
+#define tc6393xb_resume NULL
+#endif
+
+static struct platform_driver tc6393xb_driver = {
+	.probe = tc6393xb_probe,
+	.remove = __devexit_p(tc6393xb_remove),
+	.suspend = tc6393xb_suspend,
+	.resume = tc6393xb_resume,
+
+	.driver = {
+		.name = "tc6393xb",
+		.owner = THIS_MODULE,
+	},
+};
+
+static int __init tc6393xb_init(void)
+{
+	return platform_driver_register(&tc6393xb_driver);
+}
+
+static void __exit tc6393xb_exit(void)
+{
+	platform_driver_unregister(&tc6393xb_driver);
+}
+
+subsys_initcall(tc6393xb_init);
+module_exit(tc6393xb_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Ian Molton, Dmitry Baryshkov and Dirk Opfer");
+MODULE_DESCRIPTION("tc6393xb Toshiba Mobile IO Controller");
+MODULE_ALIAS("platform:tc6393xb");
diff --git a/include/linux/mfd/tc6393xb.h b/include/linux/mfd/tc6393xb.h
new file mode 100644
index 000000000000..0e3dd4ca523b
--- /dev/null
+++ b/include/linux/mfd/tc6393xb.h
@@ -0,0 +1,47 @@
+/*
+ * Toshiba TC6393XB SoC support
+ *
+ * Copyright(c) 2005-2006 Chris Humbert
+ * Copyright(c) 2005 Dirk Opfer
+ * Copyright(c) 2005 Ian Molton <spyro@f2s.com>
+ * Copyright(c) 2007 Dmitry Baryshkov
+ *
+ * Based on code written by Sharp/Lineo for 2.4 kernels
+ * Based on locomo.c
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef TC6393XB_H
+#define TC6393XB_H
+
+/* Also one should provide the CK3P6MI clock */
+struct tc6393xb_platform_data {
+	u16	scr_pll2cr;	/* PLL2 Control */
+	u16	scr_gper;	/* GP Enable */
+	u32	scr_gpo_doecr;	/* GPO Data OE Control */
+	u32	scr_gpo_dsr;	/* GPO Data Set */
+
+	int	(*enable)(struct platform_device *dev);
+	int	(*disable)(struct platform_device *dev);
+	int	(*suspend)(struct platform_device *dev);
+	int	(*resume)(struct platform_device *dev);
+
+	int	irq_base;	/* a base for cascaded irq */
+	int	gpio_base;
+};
+
+/*
+ * Relative to irq_base
+ */
+#define	IRQ_TC6393_NAND		0
+#define	IRQ_TC6393_MMC		1
+#define	IRQ_TC6393_OHCI		2
+#define	IRQ_TC6393_SERIAL	3
+#define	IRQ_TC6393_FB		4
+
+#define	TC6393XB_NR_IRQS	8
+
+#endif
-- 
cgit v1.2.3


From aa613de676986f136fa6f48a4d709b5d264f4f38 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Fri, 27 Jun 2008 10:37:19 +0100
Subject: [ARM] 5127/1: Core MFD support

This patch provides a common subdevice registration system for MFD type
chips, using platfrom device.

Signed-off-by: Ian Molton <spyro@f2s.com>
Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/mfd/Kconfig      |   4 ++
 drivers/mfd/Makefile     |   2 +
 drivers/mfd/mfd-core.c   | 114 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/core.h |  55 +++++++++++++++++++++++
 4 files changed, 175 insertions(+)
 create mode 100644 drivers/mfd/mfd-core.c
 create mode 100644 include/linux/mfd/core.h

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 1a1ac262fc87..8ebc0be10953 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -5,6 +5,10 @@
 menu "Multifunction device drivers"
 	depends on HAS_IOMEM
 
+config MFD_CORE
+	tristate
+	default n
+
 config MFD_SM501
 	tristate "Support for Silicon Motion SM501"
 	 ---help---
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index b4168442d53c..33daa2f45dd8 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -10,6 +10,8 @@ obj-$(CONFIG_HTC_PASIC3)	+= htc-pasic3.o
 
 obj-$(CONFIG_MFD_TC6393XB)	+= tc6393xb.o
 
+obj-$(CONFIG_MFD_CORE)		+= mfd-core.o
+
 obj-$(CONFIG_MCP)		+= mcp-core.o
 obj-$(CONFIG_MCP_SA11X0)	+= mcp-sa11x0.o
 obj-$(CONFIG_MCP_UCB1200)	+= ucb1x00-core.o
diff --git a/drivers/mfd/mfd-core.c b/drivers/mfd/mfd-core.c
new file mode 100644
index 000000000000..d7d88ce053a6
--- /dev/null
+++ b/drivers/mfd/mfd-core.c
@@ -0,0 +1,114 @@
+/*
+ * drivers/mfd/mfd-core.c
+ *
+ * core MFD support
+ * Copyright (c) 2006 Ian Molton
+ * Copyright (c) 2007,2008 Dmitry Baryshkov
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/mfd/core.h>
+
+static int mfd_add_device(struct platform_device *parent,
+		const struct mfd_cell *cell,
+		struct resource *mem_base,
+		int irq_base)
+{
+	struct resource res[cell->num_resources];
+	struct platform_device *pdev;
+	int ret = -ENOMEM;
+	int r;
+
+	pdev = platform_device_alloc(cell->name, parent->id);
+	if (!pdev)
+		goto fail_alloc;
+
+	pdev->dev.parent = &parent->dev;
+
+	ret = platform_device_add_data(pdev,
+			cell, sizeof(struct mfd_cell));
+	if (ret)
+		goto fail_device;
+
+	memzero(res, sizeof(res));
+	for (r = 0; r < cell->num_resources; r++) {
+		res[r].name = cell->resources[r].name;
+		res[r].flags = cell->resources[r].flags;
+
+		/* Find out base to use */
+		if (cell->resources[r].flags & IORESOURCE_MEM) {
+			res[r].parent = mem_base;
+			res[r].start = mem_base->start +
+				cell->resources[r].start;
+			res[r].end = mem_base->start +
+				cell->resources[r].end;
+		} else if (cell->resources[r].flags & IORESOURCE_IRQ) {
+			res[r].start = irq_base +
+				cell->resources[r].start;
+			res[r].end   = irq_base +
+				cell->resources[r].end;
+		} else {
+			res[r].parent = cell->resources[r].parent;
+			res[r].start = cell->resources[r].start;
+			res[r].end   = cell->resources[r].end;
+		}
+	}
+
+	platform_device_add_resources(pdev, res, cell->num_resources);
+
+	ret = platform_device_add(pdev);
+	if (ret)
+		goto fail_device;
+
+	return 0;
+
+/*	platform_device_del(pdev); */
+fail_device:
+	platform_device_put(pdev);
+fail_alloc:
+	return ret;
+}
+
+int mfd_add_devices(
+		struct platform_device *parent,
+		const struct mfd_cell *cells, int n_devs,
+		struct resource *mem_base,
+		int irq_base)
+{
+	int i;
+	int ret = 0;
+
+	for (i = 0; i < n_devs; i++) {
+		ret = mfd_add_device(parent, cells + i, mem_base, irq_base);
+		if (ret)
+			break;
+	}
+
+	if (ret)
+		mfd_remove_devices(parent);
+
+	return ret;
+}
+EXPORT_SYMBOL(mfd_add_devices);
+
+static int mfd_remove_devices_fn(struct device *dev, void *unused)
+{
+	platform_device_unregister(
+			container_of(dev, struct platform_device, dev));
+	return 0;
+}
+
+void mfd_remove_devices(struct platform_device *parent)
+{
+	device_for_each_child(&parent->dev, NULL, mfd_remove_devices_fn);
+}
+EXPORT_SYMBOL(mfd_remove_devices);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Ian Molton, Dmitry Baryshkov");
diff --git a/include/linux/mfd/core.h b/include/linux/mfd/core.h
new file mode 100644
index 000000000000..bb3dd0545928
--- /dev/null
+++ b/include/linux/mfd/core.h
@@ -0,0 +1,55 @@
+#ifndef MFD_CORE_H
+#define MFD_CORE_H
+/*
+ * drivers/mfd/mfd-core.h
+ *
+ * core MFD support
+ * Copyright (c) 2006 Ian Molton
+ * Copyright (c) 2007 Dmitry Baryshkov
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/platform_device.h>
+
+/*
+ * This struct describes the MFD part ("cell").
+ * After registration the copy of this structure will become the platform data
+ * of the resulting platform_device
+ */
+struct mfd_cell {
+	const char		*name;
+
+	int			(*enable)(struct platform_device *dev);
+	int			(*disable)(struct platform_device *dev);
+	int			(*suspend)(struct platform_device *dev);
+	int			(*resume)(struct platform_device *dev);
+
+	void			*driver_data; /* driver-specific data */
+
+	/*
+	 * This resources can be specified relatievly to the parent device.
+	 * For accessing device you should use resources from device
+	 */
+	int			num_resources;
+	const struct resource	*resources;
+};
+
+static inline struct mfd_cell *
+mfd_get_cell(struct platform_device *pdev)
+{
+	return (struct mfd_cell *)pdev->dev.platform_data;
+}
+
+extern int mfd_add_devices(
+		struct platform_device *parent,
+		const struct mfd_cell *cells, int n_devs,
+		struct resource *mem_base,
+		int irq_base);
+
+extern void mfd_remove_devices(struct platform_device *parent);
+
+#endif
-- 
cgit v1.2.3


From f024ff10b1ab13da4d626366019fd05c49721af7 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Fri, 27 Jun 2008 10:37:57 +0100
Subject: [ARM] 5128/1: tc6393xb: tmio-nand support

Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/mfd/Kconfig          |  1 +
 drivers/mfd/tc6393xb.c       | 63 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/tc6393xb.h |  2 ++
 include/linux/mfd/tmio.h     | 17 ++++++++++++
 4 files changed, 83 insertions(+)
 create mode 100644 include/linux/mfd/tmio.h

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 8ebc0be10953..7dff105e8f83 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -45,6 +45,7 @@ config HTC_PASIC3
 config MFD_TC6393XB
 	bool "Support Toshiba TC6393XB"
 	depends on HAVE_GPIO_LIB
+	select MFD_CORE
 	help
 	  Support for Toshiba Mobile IO Controller TC6393XB
 
diff --git a/drivers/mfd/tc6393xb.c b/drivers/mfd/tc6393xb.c
index 4d7192edefe1..2d87501b6fd4 100644
--- a/drivers/mfd/tc6393xb.c
+++ b/drivers/mfd/tc6393xb.c
@@ -21,6 +21,8 @@
 #include <linux/platform_device.h>
 #include <linux/fb.h>
 #include <linux/clk.h>
+#include <linux/mfd/core.h>
+#include <linux/mfd/tmio.h>
 #include <linux/mfd/tc6393xb.h>
 #include <linux/gpio.h>
 
@@ -108,6 +110,59 @@ struct tc6393xb {
 	int			irq_base;
 };
 
+enum {
+	TC6393XB_CELL_NAND,
+};
+
+/*--------------------------------------------------------------------------*/
+
+static int tc6393xb_nand_enable(struct platform_device *nand)
+{
+	struct platform_device *dev = to_platform_device(nand->dev.parent);
+	struct tc6393xb *tc6393xb = platform_get_drvdata(dev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&tc6393xb->lock, flags);
+
+	/* SMD buffer on */
+	dev_dbg(&dev->dev, "SMD buffer on\n");
+	iowrite8(0xff, tc6393xb->scr + SCR_GPI_BCR(1));
+
+	spin_unlock_irqrestore(&tc6393xb->lock, flags);
+
+	return 0;
+}
+
+static struct resource __devinitdata tc6393xb_nand_resources[] = {
+	{
+		.name	= TMIO_NAND_CONFIG,
+		.start	= 0x0100,
+		.end	= 0x01ff,
+		.flags	= IORESOURCE_MEM,
+	},
+	{
+		.name	= TMIO_NAND_CONTROL,
+		.start	= 0x1000,
+		.end	= 0x1007,
+		.flags	= IORESOURCE_MEM,
+	},
+	{
+		.name	= TMIO_NAND_IRQ,
+		.start	= IRQ_TC6393_NAND,
+		.end	= IRQ_TC6393_NAND,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct mfd_cell __devinitdata tc6393xb_cells[] = {
+	[TC6393XB_CELL_NAND] = {
+		.name = "tmio-nand",
+		.enable = tc6393xb_nand_enable,
+		.num_resources = ARRAY_SIZE(tc6393xb_nand_resources),
+		.resources = tc6393xb_nand_resources,
+	},
+};
+
 /*--------------------------------------------------------------------------*/
 
 static int tc6393xb_gpio_get(struct gpio_chip *chip,
@@ -410,6 +465,12 @@ static int __devinit tc6393xb_probe(struct platform_device *dev)
 	if (tc6393xb->irq)
 		tc6393xb_attach_irq(dev);
 
+	tc6393xb_cells[TC6393XB_CELL_NAND].driver_data = tcpd->nand_data;
+
+	retval = mfd_add_devices(dev,
+			tc6393xb_cells, ARRAY_SIZE(tc6393xb_cells),
+			iomem, tcpd->irq_base);
+
 	return 0;
 
 	if (tc6393xb->irq)
@@ -440,6 +501,8 @@ static int __devexit tc6393xb_remove(struct platform_device *dev)
 	struct tc6393xb *tc6393xb = platform_get_drvdata(dev);
 	int ret;
 
+	mfd_remove_devices(dev);
+
 	if (tc6393xb->irq)
 		tc6393xb_detach_irq(dev);
 
diff --git a/include/linux/mfd/tc6393xb.h b/include/linux/mfd/tc6393xb.h
index 0e3dd4ca523b..7cc824a58f7c 100644
--- a/include/linux/mfd/tc6393xb.h
+++ b/include/linux/mfd/tc6393xb.h
@@ -31,6 +31,8 @@ struct tc6393xb_platform_data {
 
 	int	irq_base;	/* a base for cascaded irq */
 	int	gpio_base;
+
+	struct tmio_nand_data	*nand_data;
 };
 
 /*
diff --git a/include/linux/mfd/tmio.h b/include/linux/mfd/tmio.h
new file mode 100644
index 000000000000..9438d8c9ac1c
--- /dev/null
+++ b/include/linux/mfd/tmio.h
@@ -0,0 +1,17 @@
+#ifndef MFD_TMIO_H
+#define MFD_TMIO_H
+
+/*
+ * data for the NAND controller
+ */
+struct tmio_nand_data {
+	struct nand_bbt_descr	*badblock_pattern;
+	struct mtd_partition	*partition;
+	unsigned int		num_partitions;
+};
+
+#define TMIO_NAND_CONFIG	"tmio-nand-config"
+#define TMIO_NAND_CONTROL	"tmio-nand-control"
+#define TMIO_NAND_IRQ		"tmio-nand"
+
+#endif
-- 
cgit v1.2.3


From 73422392734bb68c8ff8bc74ce1bbdc32f1b639a Mon Sep 17 00:00:00 2001
From: Claudio Nieder <private@claudio.ch>
Date: Mon, 7 Jul 2008 11:56:30 -0400
Subject: Input: add driver for Tabletkiosk Sahara TouchIT-213 touchscreen

Signed-off-by: Claudio Nieder <private@claudio.ch>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/touchscreen/Kconfig      |  11 ++
 drivers/input/touchscreen/Makefile     |   1 +
 drivers/input/touchscreen/touchit213.c | 234 +++++++++++++++++++++++++++++++++
 include/linux/serio.h                  |   1 +
 4 files changed, 247 insertions(+)
 create mode 100644 drivers/input/touchscreen/touchit213.c

(limited to 'include/linux')

diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig
index 4085791e7be2..e57366521572 100644
--- a/drivers/input/touchscreen/Kconfig
+++ b/drivers/input/touchscreen/Kconfig
@@ -351,4 +351,15 @@ config TOUCHSCREEN_USB_GOTOP
 	bool "GoTop Super_Q2/GogoPen/PenPower tablet device support" if EMBEDDED
 	depends on TOUCHSCREEN_USB_COMPOSITE
 
+config TOUCHSCREEN_TOUCHIT213
+	tristate "Sahara TouchIT-213 touchscreen"
+	select SERIO
+	help
+	  Say Y here if you have a Sahara TouchIT-213 Tablet PC.
+
+	  If unsure, say N.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called touchit213.
+
 endif
diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile
index d63bcdcd28ec..39a804cd80f1 100644
--- a/drivers/input/touchscreen/Makefile
+++ b/drivers/input/touchscreen/Makefile
@@ -21,6 +21,7 @@ obj-$(CONFIG_TOUCHSCREEN_HP7XX)		+= jornada720_ts.o
 obj-$(CONFIG_TOUCHSCREEN_HTCPEN)	+= htcpen.o
 obj-$(CONFIG_TOUCHSCREEN_USB_COMPOSITE)	+= usbtouchscreen.o
 obj-$(CONFIG_TOUCHSCREEN_PENMOUNT)	+= penmount.o
+obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213)	+= touchit213.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHRIGHT)	+= touchright.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHWIN)	+= touchwin.o
 obj-$(CONFIG_TOUCHSCREEN_UCB1400)	+= ucb1400_ts.o
diff --git a/drivers/input/touchscreen/touchit213.c b/drivers/input/touchscreen/touchit213.c
new file mode 100644
index 000000000000..d1297ba19daf
--- /dev/null
+++ b/drivers/input/touchscreen/touchit213.c
@@ -0,0 +1,234 @@
+/*
+ * Sahara TouchIT-213 serial touchscreen driver
+ *
+ * Copyright (c) 2007-2008 Claudio Nieder <private@claudio.ch>
+ *
+ * Based on Touchright driver (drivers/input/touchscreen/touchright.c)
+ * Copyright (c) 2006 Rick Koch <n1gp@hotmail.com>
+ * Copyright (c) 2004 Vojtech Pavlik
+ * and Dan Streetman <ddstreet@ieee.org>
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/input.h>
+#include <linux/serio.h>
+#include <linux/init.h>
+
+#define DRIVER_DESC	"Sahara TouchIT-213 serial touchscreen driver"
+
+MODULE_AUTHOR("Claudio Nieder <private@claudio.ch>");
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
+/*
+ * Definitions & global arrays.
+ */
+
+/*
+ * Data is received through COM1 at 9600bit/s,8bit,no parity in packets
+ * of 5 byte each.
+ *
+ *   +--------+   +--------+   +--------+   +--------+   +--------+
+ *   |1000000p|   |0xxxxxxx|   |0xxxxxxx|   |0yyyyyyy|   |0yyyyyyy|
+ *   +--------+   +--------+   +--------+   +--------+   +--------+
+ *                    MSB          LSB          MSB          LSB
+ *
+ * The value of p is 1 as long as the screen is touched and 0 when
+ * reporting the location where touching stopped, e.g. where the pen was
+ * lifted from the screen.
+ *
+ * When holding the screen in landscape mode as the BIOS text output is
+ * presented, x is the horizontal axis with values growing from left to
+ * right and y is the vertical axis with values growing from top to
+ * bottom.
+ *
+ * When holding the screen in portrait mode with the Sahara logo in its
+ * correct position, x ist the vertical axis with values growing from
+ * top to bottom and y is the horizontal axis with values growing from
+ * right to left.
+ */
+
+#define T213_FORMAT_TOUCH_BIT	0x01
+#define T213_FORMAT_STATUS_BYTE	0x80
+#define T213_FORMAT_STATUS_MASK	~T213_FORMAT_TOUCH_BIT
+
+/*
+ * On my Sahara Touch-IT 213 I have observed x values from 0 to 0x7f0
+ * and y values from 0x1d to 0x7e9, so the actual measurement is
+ * probably done with an 11 bit precision.
+ */
+#define T213_MIN_XC 0
+#define T213_MAX_XC 0x07ff
+#define T213_MIN_YC 0
+#define T213_MAX_YC 0x07ff
+
+/*
+ * Per-touchscreen data.
+ */
+
+struct touchit213 {
+	struct input_dev *dev;
+	struct serio *serio;
+	int idx;
+	unsigned char csum;
+	unsigned char data[5];
+	char phys[32];
+};
+
+static irqreturn_t touchit213_interrupt(struct serio *serio,
+		unsigned char data, unsigned int flags)
+{
+	struct touchit213 *touchit213 = serio_get_drvdata(serio);
+	struct input_dev *dev = touchit213->dev;
+
+	touchit213->data[touchit213->idx] = data;
+
+	switch (touchit213->idx++) {
+	case 0:
+		if ((touchit213->data[0] & T213_FORMAT_STATUS_MASK) !=
+				T213_FORMAT_STATUS_BYTE) {
+			pr_debug("unsynchronized data: 0x%02x\n", data);
+			touchit213->idx = 0;
+		}
+		break;
+
+	case 4:
+		touchit213->idx = 0;
+		input_report_abs(dev, ABS_X,
+			(touchit213->data[1] << 7) | touchit213->data[2]);
+		input_report_abs(dev, ABS_Y,
+			(touchit213->data[3] << 7) | touchit213->data[4]);
+		input_report_key(dev, BTN_TOUCH,
+			touchit213->data[0] & T213_FORMAT_TOUCH_BIT);
+		input_sync(dev);
+		break;
+	}
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * touchit213_disconnect() is the opposite of touchit213_connect()
+ */
+
+static void touchit213_disconnect(struct serio *serio)
+{
+	struct touchit213 *touchit213 = serio_get_drvdata(serio);
+
+	input_get_device(touchit213->dev);
+	input_unregister_device(touchit213->dev);
+	serio_close(serio);
+	serio_set_drvdata(serio, NULL);
+	input_put_device(touchit213->dev);
+	kfree(touchit213);
+}
+
+/*
+ * touchit213_connect() is the routine that is called when someone adds a
+ * new serio device that supports the Touchright protocol and registers it as
+ * an input device.
+ */
+
+static int touchit213_connect(struct serio *serio, struct serio_driver *drv)
+{
+	struct touchit213 *touchit213;
+	struct input_dev *input_dev;
+	int err;
+
+	touchit213 = kzalloc(sizeof(struct touchit213), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!touchit213 || !input_dev) {
+		err = -ENOMEM;
+		goto fail1;
+	}
+
+	touchit213->serio = serio;
+	touchit213->dev = input_dev;
+	snprintf(touchit213->phys, sizeof(touchit213->phys),
+		 "%s/input0", serio->phys);
+
+	input_dev->name = "Sahara Touch-iT213 Serial TouchScreen";
+	input_dev->phys = touchit213->phys;
+	input_dev->id.bustype = BUS_RS232;
+	input_dev->id.vendor = SERIO_TOUCHIT213;
+	input_dev->id.product = 0;
+	input_dev->id.version = 0x0100;
+	input_dev->dev.parent = &serio->dev;
+	input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS);
+	input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH);
+	input_set_abs_params(touchit213->dev, ABS_X,
+			     T213_MIN_XC, T213_MAX_XC, 0, 0);
+	input_set_abs_params(touchit213->dev, ABS_Y,
+			     T213_MIN_YC, T213_MAX_YC, 0, 0);
+
+	serio_set_drvdata(serio, touchit213);
+
+	err = serio_open(serio, drv);
+	if (err)
+		goto fail2;
+
+	err = input_register_device(touchit213->dev);
+	if (err)
+		goto fail3;
+
+	return 0;
+
+ fail3:	serio_close(serio);
+ fail2:	serio_set_drvdata(serio, NULL);
+ fail1:	input_free_device(input_dev);
+	kfree(touchit213);
+	return err;
+}
+
+/*
+ * The serio driver structure.
+ */
+
+static struct serio_device_id touchit213_serio_ids[] = {
+	{
+		.type	= SERIO_RS232,
+		.proto	= SERIO_TOUCHIT213,
+		.id	= SERIO_ANY,
+		.extra	= SERIO_ANY,
+	},
+	{ 0 }
+};
+
+MODULE_DEVICE_TABLE(serio, touchit213_serio_ids);
+
+static struct serio_driver touchit213_drv = {
+	.driver		= {
+		.name	= "touchit213",
+	},
+	.description	= DRIVER_DESC,
+	.id_table	= touchit213_serio_ids,
+	.interrupt	= touchit213_interrupt,
+	.connect	= touchit213_connect,
+	.disconnect	= touchit213_disconnect,
+};
+
+/*
+ * The functions for inserting/removing us as a module.
+ */
+
+static int __init touchit213_init(void)
+{
+	return serio_register_driver(&touchit213_drv);
+}
+
+static void __exit touchit213_exit(void)
+{
+	serio_unregister_driver(&touchit213_drv);
+}
+
+module_init(touchit213_init);
+module_exit(touchit213_exit);
diff --git a/include/linux/serio.h b/include/linux/serio.h
index 36c6ab81c7e0..48defc4d181e 100644
--- a/include/linux/serio.h
+++ b/include/linux/serio.h
@@ -213,5 +213,6 @@ static inline void serio_unpin_driver(struct serio *serio)
 #define SERIO_FUJITSU	0x35
 #define SERIO_ZHENHUA	0x36
 #define SERIO_INEXIO	0x37
+#define SERIO_TOUCHIT213	0x37
 
 #endif
-- 
cgit v1.2.3


From c6c4f070a61b2b6e5cd317a5fbf25255878688a2 Mon Sep 17 00:00:00 2001
From: Greg KH <gregkh@suse.de>
Date: Thu, 3 Jul 2008 09:49:39 -0700
Subject: PCI: make pci_name use dev_name

Also fixes up the sparc code that was assuming this is not a constant.

Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/sparc64/kernel/pci.c | 2 +-
 include/linux/pci.h       | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c
index 112b09f16f36..d00a3656c287 100644
--- a/arch/sparc64/kernel/pci.c
+++ b/arch/sparc64/kernel/pci.c
@@ -408,7 +408,7 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm,
 	dev->class = class >> 8;
 	dev->revision = class & 0xff;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
+	sprintf(dev->dev.bus_id, "%04x:%02x:%02x.%d", pci_domain_nr(bus),
 		dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
 
 	if (ofpci_verbose)
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 96ebaa8d80e3..4c80dc3f2990 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -996,9 +996,9 @@ static inline void pci_set_drvdata(struct pci_dev *pdev, void *data)
 /* If you want to know what to call your pci_dev, ask this function.
  * Again, it's a wrapper around the generic device.
  */
-static inline char *pci_name(struct pci_dev *pdev)
+static inline const char *pci_name(struct pci_dev *pdev)
 {
-	return pdev->dev.bus_id;
+	return dev_name(&pdev->dev);
 }
 
 
-- 
cgit v1.2.3


From eb9d0fe40e313c0a74115ef456a2e43a6c8da72f Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 7 Jul 2008 03:34:48 +0200
Subject: PCI ACPI: Rework PCI handling of wake-up

* Introduce function acpi_pm_device_sleep_wake() for enabling and
  disabling the system wake-up capability of devices that are power
  manageable by ACPI.

* Introduce function acpi_bus_can_wakeup() allowing other (dependent)
  subsystems to check if ACPI is able to enable the system wake-up
  capability of given device.

* Introduce callback .sleep_wake() in struct pci_platform_pm_ops and
  for the ACPI PCI 'driver' make it use acpi_pm_device_sleep_wake().

* Introduce callback .can_wakeup() in struct pci_platform_pm_ops and
  for the ACPI 'driver' make it use acpi_bus_can_wakeup().

* Move the PME# handlig code out of pci_enable_wake() and split it
  into two functions, pci_pme_capable() and pci_pme_active(),
  allowing the caller to check if given device is capable of
  generating PME# from given power state and to enable/disable the
  device's PME# functionality, respectively.

* Modify pci_enable_wake() to use the new ACPI callbacks and the new
  PME#-related functions.

* Drop the generic .platform_enable_wakeup() callback that is not
  used any more.

* Introduce device_set_wakeup_capable() that will set the
  power.can_wakeup flag of given device.

* Rework PCI device PM initialization so that, if given device is
  capable of generating wake-up events, either natively through the
  PME# mechanism, or with the help of the platform, its
  power.can_wakeup flag is set and its power.should_wakeup flag is
  unset as appropriate.

* Make ACPI set the power.can_wakeup flag for devices found to be
  wake-up capable by it.

* Make the ACPI wake-up code enable/disable GPEs for devices that
  have the wakeup.flags.prepared flag set (which means that their
  wake-up power has been enabled).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/acpi/bus.c          |  11 +++
 drivers/acpi/glue.c         |   2 +
 drivers/acpi/sleep/main.c   |  25 +++++++
 drivers/acpi/sleep/wakeup.c |  11 +--
 drivers/base/power/sysfs.c  |   3 -
 drivers/pci/pci-acpi.c      |  20 ++++++
 drivers/pci/pci.c           | 169 ++++++++++++++++++++++++++++++++------------
 drivers/pci/pci.h           |   8 +++
 drivers/pci/probe.c         |  47 +-----------
 include/acpi/acpi_bus.h     |   2 +
 include/linux/pm_wakeup.h   |  26 ++-----
 11 files changed, 208 insertions(+), 116 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index b9b69d9629b5..fc1110d6a078 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -306,6 +306,17 @@ bool acpi_bus_power_manageable(acpi_handle handle)
 
 EXPORT_SYMBOL(acpi_bus_power_manageable);
 
+bool acpi_bus_can_wakeup(acpi_handle handle)
+{
+	struct acpi_device *device;
+	int result;
+
+	result = acpi_bus_get_device(handle, &device);
+	return result ? false : device->wakeup.flags.valid;
+}
+
+EXPORT_SYMBOL(acpi_bus_can_wakeup);
+
 /* --------------------------------------------------------------------------
                                 Event Management
    -------------------------------------------------------------------------- */
diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 06f8634fe58b..87c5d456e180 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -166,6 +166,8 @@ static int acpi_bind_one(struct device *dev, acpi_handle handle)
 				"firmware_node");
 		ret = sysfs_create_link(&acpi_dev->dev.kobj, &dev->kobj,
 				"physical_node");
+		if (acpi_dev->wakeup.flags.valid)
+			device_set_wakeup_capable(dev, true);
 	}
 
 	return 0;
diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c
index 4addf8ad50ae..af7f4663deaa 100644
--- a/drivers/acpi/sleep/main.c
+++ b/drivers/acpi/sleep/main.c
@@ -468,6 +468,31 @@ int acpi_pm_device_sleep_state(struct device *dev, int *d_min_p)
 		*d_min_p = d_min;
 	return d_max;
 }
+
+/**
+ *	acpi_pm_device_sleep_wake - enable or disable the system wake-up
+ *                                  capability of given device
+ *	@dev: device to handle
+ *	@enable: 'true' - enable, 'false' - disable the wake-up capability
+ */
+int acpi_pm_device_sleep_wake(struct device *dev, bool enable)
+{
+	acpi_handle handle;
+	struct acpi_device *adev;
+
+	if (!device_may_wakeup(dev))
+		return -EINVAL;
+
+	handle = DEVICE_ACPI_HANDLE(dev);
+	if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &adev))) {
+		printk(KERN_DEBUG "ACPI handle has no context!\n");
+		return -ENODEV;
+	}
+
+	return enable ?
+		acpi_enable_wakeup_device_power(adev, acpi_target_sleep_state) :
+		acpi_disable_wakeup_device_power(adev);
+}
 #endif
 
 static void acpi_power_off_prepare(void)
diff --git a/drivers/acpi/sleep/wakeup.c b/drivers/acpi/sleep/wakeup.c
index 7422a2213944..38655eb132dc 100644
--- a/drivers/acpi/sleep/wakeup.c
+++ b/drivers/acpi/sleep/wakeup.c
@@ -66,13 +66,15 @@ void acpi_enable_wakeup_device(u8 sleep_state)
 	list_for_each_safe(node, next, &acpi_wakeup_device_list) {
 		struct acpi_device *dev =
 			container_of(node, struct acpi_device, wakeup_list);
+
 		if (!dev->wakeup.flags.valid)
 			continue;
+
 		/* If users want to disable run-wake GPE,
 		 * we only disable it for wake and leave it for runtime
 		 */
-		if (!dev->wakeup.state.enabled ||
-		    sleep_state > (u32) dev->wakeup.sleep_state) {
+		if ((!dev->wakeup.state.enabled && !dev->wakeup.flags.prepared)
+		    || sleep_state > (u32) dev->wakeup.sleep_state) {
 			if (dev->wakeup.flags.run_wake) {
 				spin_unlock(&acpi_device_lock);
 				/* set_gpe_type will disable GPE, leave it like that */
@@ -110,8 +112,9 @@ void acpi_disable_wakeup_device(u8 sleep_state)
 
 		if (!dev->wakeup.flags.valid)
 			continue;
-		if (!dev->wakeup.state.enabled ||
-		    sleep_state > (u32) dev->wakeup.sleep_state) {
+
+		if ((!dev->wakeup.state.enabled && !dev->wakeup.flags.prepared)
+		    || sleep_state > (u32) dev->wakeup.sleep_state) {
 			if (dev->wakeup.flags.run_wake) {
 				spin_unlock(&acpi_device_lock);
 				acpi_set_gpe_type(dev->wakeup.gpe_device,
diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c
index d11f74b038db..596aeecfdffe 100644
--- a/drivers/base/power/sysfs.c
+++ b/drivers/base/power/sysfs.c
@@ -6,9 +6,6 @@
 #include <linux/string.h>
 #include "power.h"
 
-int (*platform_enable_wakeup)(struct device *dev, int is_on);
-
-
 /*
  *	wakeup - Report/change current wakeup option for device
  *
diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 6bc0d8c870af..7764768b6a0e 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -299,10 +299,30 @@ static int acpi_pci_set_power_state(struct pci_dev *dev, pci_power_t state)
 	return error;
 }
 
+static bool acpi_pci_can_wakeup(struct pci_dev *dev)
+{
+	acpi_handle handle = DEVICE_ACPI_HANDLE(&dev->dev);
+
+	return handle ? acpi_bus_can_wakeup(handle) : false;
+}
+
+static int acpi_pci_sleep_wake(struct pci_dev *dev, bool enable)
+{
+	int error = acpi_pm_device_sleep_wake(&dev->dev, enable);
+
+	if (!error)
+		dev_printk(KERN_INFO, &dev->dev,
+				"wake-up capability %s by ACPI\n",
+				enable ? "enabled" : "disabled");
+	return error;
+}
+
 static struct pci_platform_pm_ops acpi_pci_platform_pm = {
 	.is_manageable = acpi_pci_power_manageable,
 	.set_state = acpi_pci_set_power_state,
 	.choose_state = acpi_pci_choose_state,
+	.can_wakeup = acpi_pci_can_wakeup,
+	.sleep_wake = acpi_pci_sleep_wake,
 };
 
 /* ACPI bus type */
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 20e28077b96d..a6b1b6f96abc 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -380,7 +380,8 @@ static struct pci_platform_pm_ops *pci_platform_pm;
 
 int pci_set_platform_pm(struct pci_platform_pm_ops *ops)
 {
-	if (!ops->is_manageable || !ops->set_state || !ops->choose_state)
+	if (!ops->is_manageable || !ops->set_state || !ops->choose_state
+	    || !ops->sleep_wake || !ops->can_wakeup)
 		return -EINVAL;
 	pci_platform_pm = ops;
 	return 0;
@@ -403,6 +404,17 @@ static inline pci_power_t platform_pci_choose_state(struct pci_dev *dev)
 			pci_platform_pm->choose_state(dev) : PCI_POWER_ERROR;
 }
 
+static inline bool platform_pci_can_wakeup(struct pci_dev *dev)
+{
+	return pci_platform_pm ? pci_platform_pm->can_wakeup(dev) : false;
+}
+
+static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable)
+{
+	return pci_platform_pm ?
+			pci_platform_pm->sleep_wake(dev, enable) : -ENODEV;
+}
+
 /**
  * pci_raw_set_power_state - Use PCI PM registers to set the power state of
  *                           given PCI device
@@ -1035,6 +1047,56 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state)
 	return pcibios_set_pcie_reset_state(dev, state);
 }
 
+/**
+ * pci_pme_capable - check the capability of PCI device to generate PME#
+ * @dev: PCI device to handle.
+ * @pm: PCI PM capability offset of the device.
+ * @state: PCI state from which device will issue PME#.
+ */
+static bool pci_pme_capable(struct pci_dev *dev, int pm, pci_power_t state)
+{
+	u16 pmc;
+
+	if (!pm)
+		return false;
+
+	/* Check device's ability to generate PME# from given state */
+	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
+
+	pmc &= PCI_PM_CAP_PME_MASK;
+	pmc >>= ffs(PCI_PM_CAP_PME_MASK) - 1;   /* First bit of mask */
+
+	return !!(pmc & (1 << state));
+}
+
+/**
+ * pci_pme_active - enable or disable PCI device's PME# function
+ * @dev: PCI device to handle.
+ * @pm: PCI PM capability offset of the device.
+ * @enable: 'true' to enable PME# generation; 'false' to disable it.
+ *
+ * The caller must verify that the device is capable of generating PME# before
+ * calling this function with @enable equal to 'true'.
+ */
+static void pci_pme_active(struct pci_dev *dev, int pm, bool enable)
+{
+	u16 pmcsr;
+
+	if (!pm)
+		return;
+
+	pci_read_config_word(dev, pm + PCI_PM_CTRL, &pmcsr);
+	/* Clear PME_Status by writing 1 to it and enable PME# */
+	pmcsr |= PCI_PM_CTRL_PME_STATUS | PCI_PM_CTRL_PME_ENABLE;
+	if (!enable)
+		pmcsr &= ~PCI_PM_CTRL_PME_ENABLE;
+
+	pci_write_config_word(dev, pm + PCI_PM_CTRL, pmcsr);
+
+	dev_printk(KERN_INFO, &dev->dev, "PME# %s\n",
+			enable ? "enabled" : "disabled");
+}
+
 /**
  * pci_enable_wake - enable PCI device as wakeup event source
  * @dev: PCI device affected
@@ -1046,66 +1108,83 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state)
  * called automatically by this routine.
  *
  * Devices with legacy power management (no standard PCI PM capabilities)
- * always require such platform hooks.  Depending on the platform, devices
- * supporting the standard PCI PME# signal may require such platform hooks;
- * they always update bits in config space to allow PME# generation.
+ * always require such platform hooks.
  *
- * -EIO is returned if the device can't ever be a wakeup event source.
- * -EINVAL is returned if the device can't generate wakeup events from
- * the specified PCI state.  Returns zero if the operation is successful.
+ * RETURN VALUE:
+ * 0 is returned on success
+ * -EINVAL is returned if device is not supposed to wake up the system
+ * Error code depending on the platform is returned if both the platform and
+ * the native mechanism fail to enable the generation of wake-up events
  */
 int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
 {
 	int pm;
-	int status;
-	u16 value;
-
-	/* Note that drivers should verify device_may_wakeup(&dev->dev)
-	 * before calling this function.  Platform code should report
-	 * errors when drivers try to enable wakeup on devices that
-	 * can't issue wakeups, or on which wakeups were disabled by
-	 * userspace updating the /sys/devices.../power/wakeup file.
-	 */
+	int error = 0;
+	bool pme_done = false;
 
-	status = call_platform_enable_wakeup(&dev->dev, enable);
-
-	/* find PCI PM capability in list */
-	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
+	if (!device_may_wakeup(&dev->dev))
+		return -EINVAL;
 
-	/* If device doesn't support PM Capabilities, but caller wants to
-	 * disable wake events, it's a NOP.  Otherwise fail unless the
-	 * platform hooks handled this legacy device already.
+	/*
+	 * According to "PCI System Architecture" 4th ed. by Tom Shanley & Don
+	 * Anderson we should be doing PME# wake enable followed by ACPI wake
+	 * enable.  To disable wake-up we call the platform first, for symmetry.
 	 */
-	if (!pm)
-		return enable ? status : 0;
 
-	/* Check device's ability to generate PME# */
-	pci_read_config_word(dev,pm+PCI_PM_PMC,&value);
+	if (!enable && platform_pci_can_wakeup(dev))
+		error = platform_pci_sleep_wake(dev, false);
 
-	value &= PCI_PM_CAP_PME_MASK;
-	value >>= ffs(PCI_PM_CAP_PME_MASK) - 1;   /* First bit of mask */
-
-	/* Check if it can generate PME# from requested state. */
-	if (!value || !(value & (1 << state))) {
-		/* if it can't, revert what the platform hook changed,
-		 * always reporting the base "EINVAL, can't PME#" error
-		 */
-		if (enable)
-			call_platform_enable_wakeup(&dev->dev, 0);
-		return enable ? -EINVAL : 0;
+	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
+	if (!enable || pci_pme_capable(dev, pm, state)) {
+		pci_pme_active(dev, pm, enable);
+		pme_done = true;
 	}
 
-	pci_read_config_word(dev, pm + PCI_PM_CTRL, &value);
+	if (enable && platform_pci_can_wakeup(dev))
+		error = platform_pci_sleep_wake(dev, true);
 
-	/* Clear PME_Status by writing 1 to it and enable PME# */
-	value |= PCI_PM_CTRL_PME_STATUS | PCI_PM_CTRL_PME_ENABLE;
+	return pme_done ? 0 : error;
+}
 
-	if (!enable)
-		value &= ~PCI_PM_CTRL_PME_ENABLE;
+/**
+ * pci_pm_init - Initialize PM functions of given PCI device
+ * @dev: PCI device to handle.
+ */
+void pci_pm_init(struct pci_dev *dev)
+{
+	int pm;
+	u16 pmc;
 
-	pci_write_config_word(dev, pm + PCI_PM_CTRL, value);
+	/* find PCI PM capability in list */
+	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
+	if (!pm)
+		return;
+	/* Check device's ability to generate PME# */
+	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
 
-	return 0;
+	if ((pmc & PCI_PM_CAP_VER_MASK) > 3) {
+		dev_err(&dev->dev, "unsupported PM cap regs version (%u)\n",
+			pmc & PCI_PM_CAP_VER_MASK);
+		return;
+	}
+
+	if (pmc & PCI_PM_CAP_PME_MASK) {
+		dev_printk(KERN_INFO, &dev->dev,
+			"PME# supported from%s%s%s%s%s\n",
+			(pmc & PCI_PM_CAP_PME_D0) ? " D0" : "",
+			(pmc & PCI_PM_CAP_PME_D1) ? " D1" : "",
+			(pmc & PCI_PM_CAP_PME_D2) ? " D2" : "",
+			(pmc & PCI_PM_CAP_PME_D3) ? " D3hot" : "",
+			(pmc & PCI_PM_CAP_PME_D3cold) ? " D3cold" : "");
+		/*
+		 * Make device's PM flags reflect the wake-up capability, but
+		 * let the user space enable it to wake up the system as needed.
+		 */
+		device_set_wakeup_capable(&dev->dev, true);
+		device_set_wakeup_enable(&dev->dev, false);
+		/* Disable the PME# generation functionality */
+		pci_pme_active(dev, pm, false);
+	}
 }
 
 int
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 0cd2e719933b..b08dfc9746af 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -17,6 +17,11 @@ extern void pci_cleanup_rom(struct pci_dev *dev);
  *                 platform; to be used during system-wide transitions from a
  *                 sleeping state to the working state and vice versa
  *
+ * @can_wakeup - returns 'true' if given device is capable of waking up the
+ *               system from a sleeping state
+ *
+ * @sleep_wake - enables/disables the system wake up capability of given device
+ *
  * If given platform is generally capable of power managing PCI devices, all of
  * these callbacks are mandatory.
  */
@@ -24,9 +29,12 @@ struct pci_platform_pm_ops {
 	bool (*is_manageable)(struct pci_dev *dev);
 	int (*set_state)(struct pci_dev *dev, pci_power_t state);
 	pci_power_t (*choose_state)(struct pci_dev *dev);
+	bool (*can_wakeup)(struct pci_dev *dev);
+	int (*sleep_wake)(struct pci_dev *dev, bool enable);
 };
 
 extern int pci_set_platform_pm(struct pci_platform_pm_ops *ops);
+extern void pci_pm_init(struct pci_dev *dev);
 
 extern int pci_user_read_config_byte(struct pci_dev *dev, int where, u8 *val);
 extern int pci_user_read_config_word(struct pci_dev *dev, int where, u16 *val);
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 2f0ae70710d6..b1724cf31b66 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -860,49 +860,6 @@ int pci_cfg_space_size_ext(struct pci_dev *dev)
 	return PCI_CFG_SPACE_SIZE;
 }
 
-/**
- * pci_disable_pme - Disable the PME function of PCI device
- * @dev: PCI device affected
- * -EINVAL is returned if PCI device doesn't support PME.
- * Zero is returned if the PME is supported and can be disabled.
- */
-static int pci_disable_pme(struct pci_dev *dev)
-{
-	int pm;
-	u16 value;
-
-	/* find PCI PM capability in list */
-	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
-
-	/* If device doesn't support PM Capabilities, it means that PME is
-	 * not supported.
-	 */
-	if (!pm)
-		return -EINVAL;
-	/* Check device's ability to generate PME# */
-	pci_read_config_word(dev, pm + PCI_PM_PMC, &value);
-
-	value &= PCI_PM_CAP_PME_MASK;
-	/* Check if it can generate PME# */
-	if (!value) {
-		/*
-		 * If it is zero, it means that PME is still unsupported
-		 * although there exists the PM capability.
-		 */
-		return -EINVAL;
-	}
-
-	pci_read_config_word(dev, pm + PCI_PM_CTRL, &value);
-
-	/* Clear PME_Status by writing 1 to it */
-	value |= PCI_PM_CTRL_PME_STATUS ;
-	/* Disable PME enable bit */
-	value &= ~PCI_PM_CTRL_PME_ENABLE;
-	pci_write_config_word(dev, pm + PCI_PM_CTRL, value);
-
-	return 0;
-}
-
 int pci_cfg_space_size(struct pci_dev *dev)
 {
 	int pos;
@@ -1010,7 +967,6 @@ static struct pci_dev *pci_scan_device(struct pci_bus *bus, int devfn)
 	}
 
 	pci_vpd_pci22_init(dev);
-	pci_disable_pme(dev);
 
 	return dev;
 }
@@ -1031,6 +987,9 @@ void pci_device_add(struct pci_dev *dev, struct pci_bus *bus)
 	/* Fix up broken headers */
 	pci_fixup_device(pci_fixup_header, dev);
 
+	/* Initialize power management of the device */
+	pci_pm_init(dev);
+
 	/*
 	 * Add the device to our list of discovered devices
 	 * and the bus list for fixup functions, etc.
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index 071daf8db600..7ab5a611e43f 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -337,6 +337,7 @@ int acpi_bus_get_status(struct acpi_device *device);
 int acpi_bus_get_power(acpi_handle handle, int *state);
 int acpi_bus_set_power(acpi_handle handle, int state);
 bool acpi_bus_power_manageable(acpi_handle handle);
+bool acpi_bus_can_wakeup(acpi_handle handle);
 #ifdef CONFIG_ACPI_PROC_EVENT
 int acpi_bus_generate_proc_event(struct acpi_device *device, u8 type, int data);
 int acpi_bus_generate_proc_event4(const char *class, const char *bid, u8 type, int data);
@@ -379,6 +380,7 @@ acpi_handle acpi_get_pci_rootbridge_handle(unsigned int, unsigned int);
 
 #ifdef CONFIG_PM_SLEEP
 int acpi_pm_device_sleep_state(struct device *, int *);
+int acpi_pm_device_sleep_wake(struct device *, bool);
 #else /* !CONFIG_PM_SLEEP */
 static inline int acpi_pm_device_sleep_state(struct device *d, int *p)
 {
diff --git a/include/linux/pm_wakeup.h b/include/linux/pm_wakeup.h
index f0d0b2cb8d20..3af0c8d05cdc 100644
--- a/include/linux/pm_wakeup.h
+++ b/include/linux/pm_wakeup.h
@@ -35,6 +35,11 @@ static inline void device_init_wakeup(struct device *dev, int val)
 	dev->power.can_wakeup = dev->power.should_wakeup = !!val;
 }
 
+static inline void device_set_wakeup_capable(struct device *dev, int val)
+{
+	dev->power.can_wakeup = !!val;
+}
+
 static inline int device_can_wakeup(struct device *dev)
 {
 	return dev->power.can_wakeup;
@@ -47,21 +52,7 @@ static inline void device_set_wakeup_enable(struct device *dev, int val)
 
 static inline int device_may_wakeup(struct device *dev)
 {
-	return dev->power.can_wakeup & dev->power.should_wakeup;
-}
-
-/*
- * Platform hook to activate device wakeup capability, if that's not already
- * handled by enable_irq_wake() etc.
- * Returns zero on success, else negative errno
- */
-extern int (*platform_enable_wakeup)(struct device *dev, int is_on);
-
-static inline int call_platform_enable_wakeup(struct device *dev, int is_on)
-{
-	if (platform_enable_wakeup)
-		return (*platform_enable_wakeup)(dev, is_on);
-	return 0;
+	return dev->power.can_wakeup && dev->power.should_wakeup;
 }
 
 #else /* !CONFIG_PM */
@@ -80,11 +71,6 @@ static inline int device_can_wakeup(struct device *dev)
 #define device_set_wakeup_enable(dev, val)	do {} while (0)
 #define device_may_wakeup(dev)			0
 
-static inline int call_platform_enable_wakeup(struct device *dev, int is_on)
-{
-	return 0;
-}
-
 #endif /* !CONFIG_PM */
 
 #endif /* _LINUX_PM_WAKEUP_H */
-- 
cgit v1.2.3


From 404cc2d8ce41ed4031958fba8e633767e8a2e028 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 7 Jul 2008 03:35:26 +0200
Subject: PCI PM: Introduce pci_prepare_to_sleep and pci_back_from_sleep

Introduce functions pci_prepare_to_sleep() and pci_back_from_sleep(),
to be used by the PCI drivers that want to place their devices into
the lowest power state appropiate for them (PCI_D3hot, if the device
is not supposed to wake up the system, or the deepest state from
which the wake-up is possible, otherwise) while the system is being
prepared to go into a sleeping state and to put them back into D0
during the subsequent transition to the working state.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c   | 83 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pci.h |  2 ++
 2 files changed, 85 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index a6b1b6f96abc..1b0ec45e9934 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1146,6 +1146,87 @@ int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
 	return pme_done ? 0 : error;
 }
 
+/**
+ * pci_prepare_to_sleep - prepare PCI device for system-wide transition into
+ *                        a sleep state
+ * @dev: Device to handle.
+ *
+ * Choose the power state appropriate for the device depending on whether
+ * it can wake up the system and/or is power manageable by the platform
+ * (PCI_D3hot is the default) and put the device into that state.
+ */
+int pci_prepare_to_sleep(struct pci_dev *dev)
+{
+	pci_power_t target_state = PCI_D3hot;
+	int pm = pci_find_capability(dev, PCI_CAP_ID_PM);
+	int error;
+
+	if (platform_pci_power_manageable(dev)) {
+		/*
+		 * Call the platform to choose the target state of the device
+		 * and enable wake-up from this state if supported.
+		 */
+		pci_power_t state = platform_pci_choose_state(dev);
+
+		switch (state) {
+		case PCI_POWER_ERROR:
+		case PCI_UNKNOWN:
+			break;
+		case PCI_D1:
+		case PCI_D2:
+			if (pci_no_d1d2(dev))
+				break;
+		default:
+			target_state = state;
+			pci_enable_wake(dev, target_state, true);
+		}
+	} else if (device_may_wakeup(&dev->dev)) {
+		/*
+		 * Find the deepest state from which the device can generate
+		 * wake-up events, make it the target state and enable device
+		 * to generate PME#.
+		 */
+		u16 pmc;
+
+		if (!pm)
+			return -EIO;
+
+		pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
+		if (pmc & PCI_PM_CAP_PME_MASK) {
+			if (!(pmc & PCI_PM_CAP_PME_D3)) {
+				/* Device cannot generate PME# from D3_hot */
+				if (pmc & PCI_PM_CAP_PME_D2)
+					target_state = PCI_D2;
+				else if (pmc & PCI_PM_CAP_PME_D1)
+					target_state = PCI_D1;
+				else
+					target_state = PCI_D0;
+			}
+			pci_pme_active(dev, pm, true);
+		}
+	}
+
+	error = pci_set_power_state(dev, target_state);
+
+	if (error)
+		pci_enable_wake(dev, target_state, false);
+
+	return error;
+}
+
+/**
+ * pci_back_from_sleep - turn PCI device on during system-wide transition into
+ *                       the working state a sleep state
+ * @dev: Device to handle.
+ *
+ * Disable device's sytem wake-up capability and put it into D0.
+ */
+int pci_back_from_sleep(struct pci_dev *dev)
+{
+	pci_enable_wake(dev, PCI_D0, false);
+	return pci_set_power_state(dev, PCI_D0);
+}
+
 /**
  * pci_pm_init - Initialize PM functions of given PCI device
  * @dev: PCI device to handle.
@@ -1853,5 +1934,7 @@ EXPORT_SYMBOL(pci_set_power_state);
 EXPORT_SYMBOL(pci_save_state);
 EXPORT_SYMBOL(pci_restore_state);
 EXPORT_SYMBOL(pci_enable_wake);
+EXPORT_SYMBOL(pci_prepare_to_sleep);
+EXPORT_SYMBOL(pci_back_from_sleep);
 EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 4c80dc3f2990..52ac06dcce98 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -632,6 +632,8 @@ int pci_restore_state(struct pci_dev *dev);
 int pci_set_power_state(struct pci_dev *dev, pci_power_t state);
 pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state);
 int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable);
+int pci_prepare_to_sleep(struct pci_dev *dev);
+int pci_back_from_sleep(struct pci_dev *dev);
 
 /* Functions for PCI Hotplug drivers to use */
 int pci_bus_find_capability(struct pci_bus *bus, unsigned int devfn, int cap);
-- 
cgit v1.2.3


From 337001b6c42938f49a880b1b8306c3ed771a7e61 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 7 Jul 2008 03:36:24 +0200
Subject: PCI: Simplify PCI device PM code

If the offset of PCI device's PM capability in its configuration space,
the mask of states that the device supports PME# from and the D1 and D2
support bits are cached in the corresponding struct pci_dev, the PCI
device PM code can be simplified quite a bit.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c        | 118 +++++++++++++++++++++--------------------------
 include/linux/pci.h      |   8 +++-
 include/linux/pci_regs.h |   1 +
 3 files changed, 60 insertions(+), 67 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 1b0ec45e9934..e632a58ba5d0 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -419,7 +419,6 @@ static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable)
  * pci_raw_set_power_state - Use PCI PM registers to set the power state of
  *                           given PCI device
  * @dev: PCI device to handle.
- * @pm: PCI PM capability offset of the device.
  * @state: PCI power state (D0, D1, D2, D3hot) to put the device into.
  *
  * RETURN VALUE:
@@ -430,12 +429,12 @@ static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable)
  * 0 if device's power state has been successfully changed.
  */
 static int
-pci_raw_set_power_state(struct pci_dev *dev, int pm, pci_power_t state)
+pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
 {
-	u16 pmcsr, pmc;
+	u16 pmcsr;
 	bool need_restore = false;
 
-	if (!pm)
+	if (!dev->pm_cap)
 		return -EIO;
 
 	if (state < PCI_D0 || state > PCI_D3hot)
@@ -455,20 +454,12 @@ pci_raw_set_power_state(struct pci_dev *dev, int pm, pci_power_t state)
 		return -EINVAL;
 	}
 
-	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
-
-	if ((pmc & PCI_PM_CAP_VER_MASK) > 3) {
-		dev_err(&dev->dev, "unsupported PM cap regs version (%u)\n",
-			pmc & PCI_PM_CAP_VER_MASK);
-		return -EIO;
-	}
-
 	/* check if this device supports the desired state */
-	if ((state == PCI_D1 && !(pmc & PCI_PM_CAP_D1))
-	   || (state == PCI_D2 && !(pmc & PCI_PM_CAP_D2)))
+	if ((state == PCI_D1 && !dev->d1_support)
+	   || (state == PCI_D2 && !dev->d2_support))
 		return -EIO;
 
-	pci_read_config_word(dev, pm + PCI_PM_CTRL, &pmcsr);
+	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
 
 	/* If we're (effectively) in D3, force entire word to 0.
 	 * This doesn't affect PME_Status, disables PME_En, and
@@ -492,7 +483,7 @@ pci_raw_set_power_state(struct pci_dev *dev, int pm, pci_power_t state)
 	}
 
 	/* enter specified state */
-	pci_write_config_word(dev, pm + PCI_PM_CTRL, pmcsr);
+	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
 
 	/* Mandatory power management transition delays */
 	/* see PCI PM 1.1 5.6.1 table 18 */
@@ -528,14 +519,13 @@ pci_raw_set_power_state(struct pci_dev *dev, int pm, pci_power_t state)
  * pci_update_current_state - Read PCI power state of given device from its
  *                            PCI PM registers and cache it
  * @dev: PCI device to handle.
- * @pm: PCI PM capability offset of the device.
  */
-static void pci_update_current_state(struct pci_dev *dev, int pm)
+static void pci_update_current_state(struct pci_dev *dev)
 {
-	if (pm) {
+	if (dev->pm_cap) {
 		u16 pmcsr;
 
-		pci_read_config_word(dev, pm + PCI_PM_CTRL, &pmcsr);
+		pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
 		dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
 	}
 }
@@ -557,7 +547,7 @@ static void pci_update_current_state(struct pci_dev *dev, int pm)
  */
 int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
 {
-	int pm, error;
+	int error;
 
 	/* bound the state we're entering */
 	if (state > PCI_D3hot)
@@ -572,9 +562,6 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
 		 */
 		return 0;
 
-	/* Find PCI PM capability in the list */
-	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
-
 	if (state == PCI_D0 && platform_pci_power_manageable(dev)) {
 		/*
 		 * Allow the platform to change the state, for example via ACPI
@@ -582,16 +569,16 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
 		 */
 		int ret = platform_pci_set_power_state(dev, PCI_D0);
 		if (!ret)
-			pci_update_current_state(dev, pm);
+			pci_update_current_state(dev);
 	}
 
-	error = pci_raw_set_power_state(dev, pm, state);
+	error = pci_raw_set_power_state(dev, state);
 
 	if (state > PCI_D0 && platform_pci_power_manageable(dev)) {
 		/* Allow the platform to finalize the transition */
 		int ret = platform_pci_set_power_state(dev, state);
 		if (!ret) {
-			pci_update_current_state(dev, pm);
+			pci_update_current_state(dev);
 			error = 0;
 		}
 	}
@@ -1050,48 +1037,38 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state)
 /**
  * pci_pme_capable - check the capability of PCI device to generate PME#
  * @dev: PCI device to handle.
- * @pm: PCI PM capability offset of the device.
  * @state: PCI state from which device will issue PME#.
  */
-static bool pci_pme_capable(struct pci_dev *dev, int pm, pci_power_t state)
+static bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
 {
-	u16 pmc;
-
-	if (!pm)
+	if (!dev->pm_cap)
 		return false;
 
-	/* Check device's ability to generate PME# from given state */
-	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
-
-	pmc &= PCI_PM_CAP_PME_MASK;
-	pmc >>= ffs(PCI_PM_CAP_PME_MASK) - 1;   /* First bit of mask */
-
-	return !!(pmc & (1 << state));
+	return !!(dev->pme_support & (1 << state));
 }
 
 /**
  * pci_pme_active - enable or disable PCI device's PME# function
  * @dev: PCI device to handle.
- * @pm: PCI PM capability offset of the device.
  * @enable: 'true' to enable PME# generation; 'false' to disable it.
  *
  * The caller must verify that the device is capable of generating PME# before
  * calling this function with @enable equal to 'true'.
  */
-static void pci_pme_active(struct pci_dev *dev, int pm, bool enable)
+static void pci_pme_active(struct pci_dev *dev, bool enable)
 {
 	u16 pmcsr;
 
-	if (!pm)
+	if (!dev->pm_cap)
 		return;
 
-	pci_read_config_word(dev, pm + PCI_PM_CTRL, &pmcsr);
+	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
 	/* Clear PME_Status by writing 1 to it and enable PME# */
 	pmcsr |= PCI_PM_CTRL_PME_STATUS | PCI_PM_CTRL_PME_ENABLE;
 	if (!enable)
 		pmcsr &= ~PCI_PM_CTRL_PME_ENABLE;
 
-	pci_write_config_word(dev, pm + PCI_PM_CTRL, pmcsr);
+	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
 
 	dev_printk(KERN_INFO, &dev->dev, "PME# %s\n",
 			enable ? "enabled" : "disabled");
@@ -1118,7 +1095,6 @@ static void pci_pme_active(struct pci_dev *dev, int pm, bool enable)
  */
 int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
 {
-	int pm;
 	int error = 0;
 	bool pme_done = false;
 
@@ -1134,9 +1110,8 @@ int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
 	if (!enable && platform_pci_can_wakeup(dev))
 		error = platform_pci_sleep_wake(dev, false);
 
-	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
-	if (!enable || pci_pme_capable(dev, pm, state)) {
-		pci_pme_active(dev, pm, enable);
+	if (!enable || pci_pme_capable(dev, state)) {
+		pci_pme_active(dev, enable);
 		pme_done = true;
 	}
 
@@ -1158,7 +1133,6 @@ int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
 int pci_prepare_to_sleep(struct pci_dev *dev)
 {
 	pci_power_t target_state = PCI_D3hot;
-	int pm = pci_find_capability(dev, PCI_CAP_ID_PM);
 	int error;
 
 	if (platform_pci_power_manageable(dev)) {
@@ -1186,23 +1160,14 @@ int pci_prepare_to_sleep(struct pci_dev *dev)
 		 * wake-up events, make it the target state and enable device
 		 * to generate PME#.
 		 */
-		u16 pmc;
-
-		if (!pm)
+		if (!dev->pm_cap)
 			return -EIO;
 
-		pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
-		if (pmc & PCI_PM_CAP_PME_MASK) {
-			if (!(pmc & PCI_PM_CAP_PME_D3)) {
-				/* Device cannot generate PME# from D3_hot */
-				if (pmc & PCI_PM_CAP_PME_D2)
-					target_state = PCI_D2;
-				else if (pmc & PCI_PM_CAP_PME_D1)
-					target_state = PCI_D1;
-				else
-					target_state = PCI_D0;
-			}
-			pci_pme_active(dev, pm, true);
+		if (dev->pme_support) {
+			while (target_state
+			      && !(dev->pme_support & (1 << target_state)))
+				target_state--;
+			pci_pme_active(dev, true);
 		}
 	}
 
@@ -1236,6 +1201,8 @@ void pci_pm_init(struct pci_dev *dev)
 	int pm;
 	u16 pmc;
 
+	dev->pm_cap = 0;
+
 	/* find PCI PM capability in list */
 	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
 	if (!pm)
@@ -1249,7 +1216,23 @@ void pci_pm_init(struct pci_dev *dev)
 		return;
 	}
 
-	if (pmc & PCI_PM_CAP_PME_MASK) {
+	dev->pm_cap = pm;
+
+	dev->d1_support = false;
+	dev->d2_support = false;
+	if (!pci_no_d1d2(dev)) {
+		if (pmc & PCI_PM_CAP_D1) {
+			dev_printk(KERN_DEBUG, &dev->dev, "supports D1\n");
+			dev->d1_support = true;
+		}
+		if (pmc & PCI_PM_CAP_D2) {
+			dev_printk(KERN_DEBUG, &dev->dev, "supports D2\n");
+			dev->d2_support = true;
+		}
+	}
+
+	pmc &= PCI_PM_CAP_PME_MASK;
+	if (pmc) {
 		dev_printk(KERN_INFO, &dev->dev,
 			"PME# supported from%s%s%s%s%s\n",
 			(pmc & PCI_PM_CAP_PME_D0) ? " D0" : "",
@@ -1257,6 +1240,7 @@ void pci_pm_init(struct pci_dev *dev)
 			(pmc & PCI_PM_CAP_PME_D2) ? " D2" : "",
 			(pmc & PCI_PM_CAP_PME_D3) ? " D3hot" : "",
 			(pmc & PCI_PM_CAP_PME_D3cold) ? " D3cold" : "");
+		dev->pme_support = pmc >> PCI_PM_CAP_PME_SHIFT;
 		/*
 		 * Make device's PM flags reflect the wake-up capability, but
 		 * let the user space enable it to wake up the system as needed.
@@ -1264,7 +1248,9 @@ void pci_pm_init(struct pci_dev *dev)
 		device_set_wakeup_capable(&dev->dev, true);
 		device_set_wakeup_enable(&dev->dev, false);
 		/* Disable the PME# generation functionality */
-		pci_pme_active(dev, pm, false);
+		pci_pme_active(dev, false);
+	} else {
+		dev->pme_support = 0;
 	}
 }
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 52ac06dcce98..68a29f0f2748 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -177,6 +177,13 @@ struct pci_dev {
 	pci_power_t     current_state;  /* Current operating state. In ACPI-speak,
 					   this is D0-D3, D0 being fully functional,
 					   and D3 being off. */
+	int		pm_cap;		/* PM capability offset in the
+					   configuration space */
+	unsigned int	pme_support:5;	/* Bitmask of states from which PME#
+					   can be generated */
+	unsigned int	d1_support:1;	/* Low power state D1 is supported */
+	unsigned int	d2_support:1;	/* Low power state D2 is supported */
+	unsigned int	no_d1d2:1;	/* Only allow D0 and D3 */
 
 #ifdef CONFIG_PCIEASPM
 	struct pcie_link_state	*link_state;	/* ASPM link state. */
@@ -201,7 +208,6 @@ struct pci_dev {
 	unsigned int	is_added:1;
 	unsigned int	is_busmaster:1; /* device is busmaster */
 	unsigned int	no_msi:1;	/* device may not use msi */
-	unsigned int	no_d1d2:1;   /* only allow d0 or d3 */
 	unsigned int	block_ucfg_access:1;	/* userspace config space access is blocked */
 	unsigned int	broken_parity_status:1;	/* Device generates false positive parity */
 	unsigned int 	msi_enabled:1;
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index c0c1223c9194..19958b929905 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -231,6 +231,7 @@
 #define  PCI_PM_CAP_PME_D2	0x2000	/* PME# from D2 */
 #define  PCI_PM_CAP_PME_D3	0x4000	/* PME# from D3 (hot) */
 #define  PCI_PM_CAP_PME_D3cold	0x8000	/* PME# from D3 (cold) */
+#define  PCI_PM_CAP_PME_SHIFT	11	/* Start of the PME Mask in PMC */
 #define PCI_PM_CTRL		4	/* PM control and status register */
 #define  PCI_PM_CTRL_STATE_MASK	0x0003	/* Current power state (D0 to D3) */
 #define  PCI_PM_CTRL_NO_SOFT_RESET	0x0004	/* No reset for D3hot->D0 */
-- 
cgit v1.2.3


From d2dbf343329dc777d77488743465f7be4245971d Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Fri, 13 Jun 2008 02:00:56 -0700
Subject: x86: clean up reserve_bootmem_generic() and port it to 32-bit

1. add reserve_bootmem_generic for 32bit
2. change len to unsigned long
3. make early_res_to_bootmem to use it

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/e820.c    |  2 +-
 arch/x86/kernel/mpparse.c | 18 ++++++------------
 arch/x86/mm/init_32.c     |  6 ++++++
 arch/x86/mm/init_64.c     |  3 ++-
 include/asm-x86/proto.h   |  2 --
 include/linux/bootmem.h   |  2 ++
 6 files changed, 17 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 4f2cd5d179e2..774063f11be0 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -635,7 +635,7 @@ void __init early_res_to_bootmem(u64 start, u64 end)
 			continue;
 		printk(KERN_INFO "  early res: %d [%llx-%llx] %s\n", i,
 			final_start, final_end - 1, r->name);
-		reserve_bootmem(final_start, final_end - final_start,
+		reserve_bootmem_generic(final_start, final_end - final_start,
 				BOOTMEM_DEFAULT);
 	}
 }
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 7ac1b689b70a..b62ac6ba1410 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -859,10 +859,11 @@ static int __init smp_scan_config(unsigned long base, unsigned long length,
 
 			if (!reserve)
 				return 1;
-#ifdef CONFIG_X86_32
-			reserve_bootmem(virt_to_phys(mpf), PAGE_SIZE,
+			reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE,
 					BOOTMEM_DEFAULT);
 			if (mpf->mpf_physptr) {
+				unsigned long size = PAGE_SIZE;
+#ifdef CONFIG_X86_32
 				/*
 				 * We cannot access to MPC table to compute
 				 * table size yet, as only few megabytes from
@@ -872,22 +873,15 @@ static int __init smp_scan_config(unsigned long base, unsigned long length,
 				 * PAGE_SIZE from mpg->mpf_physptr yields BUG()
 				 * in reserve_bootmem.
 				 */
-				unsigned long size = PAGE_SIZE;
 				unsigned long end = max_low_pfn * PAGE_SIZE;
 				if (mpf->mpf_physptr + size > end)
 					size = end - mpf->mpf_physptr;
-				reserve_bootmem(mpf->mpf_physptr, size,
+#endif
+				reserve_bootmem_generic(mpf->mpf_physptr, size,
 						BOOTMEM_DEFAULT);
 			}
 
-#else
-			reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE,
-				BOOTMEM_DEFAULT);
-			if (mpf->mpf_physptr)
-				reserve_bootmem_generic(mpf->mpf_physptr,
-					PAGE_SIZE, BOOTMEM_DEFAULT);
-#endif
-		return 1;
+			return 1;
 		}
 		bp += 4;
 		length -= 16;
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 0e7bb5e81670..abadb1da70df 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -785,3 +785,9 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 	free_init_pages("initrd memory", start, end);
 }
 #endif
+
+int __init reserve_bootmem_generic(unsigned long phys, unsigned long len,
+				   int flags)
+{
+	return reserve_bootmem(phys, len, flags);
+}
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index bf7bf1de6c25..b8c2c1ef7ad5 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -799,7 +799,8 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 }
 #endif
 
-int __init reserve_bootmem_generic(unsigned long phys, unsigned len, int flags)
+int __init reserve_bootmem_generic(unsigned long phys, unsigned long len,
+				   int flags)
 {
 #ifdef CONFIG_NUMA
 	int nid, next_nid;
diff --git a/include/asm-x86/proto.h b/include/asm-x86/proto.h
index a9f51472521e..3dd458c385c0 100644
--- a/include/asm-x86/proto.h
+++ b/include/asm-x86/proto.h
@@ -14,8 +14,6 @@ extern void ia32_syscall(void);
 extern void ia32_cstar_target(void);
 extern void ia32_sysenter_target(void);
 
-extern int reserve_bootmem_generic(unsigned long phys, unsigned len, int flags);
-
 extern void syscall32_cpu_init(void);
 
 extern void check_efer(void);
diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 686895bacd9d..a1d9b79078ea 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -84,6 +84,8 @@ extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags);
 	__alloc_bootmem_low(x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
+extern int reserve_bootmem_generic(unsigned long addr, unsigned long size,
+				   int flags);
 extern unsigned long free_all_bootmem(void);
 extern unsigned long free_all_bootmem_node(pg_data_t *pgdat);
 extern void *__alloc_bootmem_node(pg_data_t *pgdat,
-- 
cgit v1.2.3


From cc1050bafebfb1d7935331282e948b5016318192 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Fri, 13 Jun 2008 19:08:52 -0700
Subject: x86: replace shrink_active_range() with remove_active_range()

in case we have kva before ramdisk on a node, we still need to use
those ranges.

v2: reserve_early kva ram area, in case there are holes in highmem, to avoid
    those area could be treat as free high pages.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/discontig_32.c | 45 ++++++++++++++++++++++++---------------------
 include/linux/mm.h         |  3 ++-
 mm/page_alloc.c            | 29 +++++++++++++++++++++++------
 3 files changed, 49 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c
index accc7c6c57fc..c3f119e99e0d 100644
--- a/arch/x86/mm/discontig_32.c
+++ b/arch/x86/mm/discontig_32.c
@@ -230,8 +230,8 @@ static unsigned long calculate_numa_remap_pages(void)
 	unsigned long size, reserve_pages = 0;
 
 	for_each_online_node(nid) {
-		u64 node_end_target;
-		u64 node_end_final;
+		u64 node_kva_target;
+		u64 node_kva_final;
 
 		/*
 		 * The acpi/srat node info can show hot-add memroy zones
@@ -254,42 +254,45 @@ static unsigned long calculate_numa_remap_pages(void)
 		/* now the roundup is correct, convert to PAGE_SIZE pages */
 		size = size * PTRS_PER_PTE;
 
-		node_end_target = round_down(node_end_pfn[nid] - size,
+		node_kva_target = round_down(node_end_pfn[nid] - size,
 						 PTRS_PER_PTE);
-		node_end_target <<= PAGE_SHIFT;
+		node_kva_target <<= PAGE_SHIFT;
 		do {
-			node_end_final = find_e820_area(node_end_target,
+			node_kva_final = find_e820_area(node_kva_target,
 					((u64)node_end_pfn[nid])<<PAGE_SHIFT,
 						((u64)size)<<PAGE_SHIFT,
 						LARGE_PAGE_BYTES);
-			node_end_target -= LARGE_PAGE_BYTES;
-		} while (node_end_final == -1ULL &&
-			 (node_end_target>>PAGE_SHIFT) > (node_start_pfn[nid]));
+			node_kva_target -= LARGE_PAGE_BYTES;
+		} while (node_kva_final == -1ULL &&
+			 (node_kva_target>>PAGE_SHIFT) > (node_start_pfn[nid]));
 
-		if (node_end_final == -1ULL)
+		if (node_kva_final == -1ULL)
 			panic("Can not get kva ram\n");
 
-		printk("Reserving %ld pages of KVA for lmem_map of node %d\n",
-				size, nid);
 		node_remap_size[nid] = size;
 		node_remap_offset[nid] = reserve_pages;
 		reserve_pages += size;
-		printk("Shrinking node %d from %ld pages to %lld pages\n",
-			nid, node_end_pfn[nid], node_end_final>>PAGE_SHIFT);
+		printk("Reserving %ld pages of KVA for lmem_map of node %d at %llx\n",
+				size, nid, node_kva_final>>PAGE_SHIFT);
 
 		/*
 		 *  prevent kva address below max_low_pfn want it on system
 		 *  with less memory later.
 		 *  layout will be: KVA address , KVA RAM
+		 *
+		 *  we are supposed to only record the one less then max_low_pfn
+		 *  but we could have some hole in high memory, and it will only
+		 *  check page_is_ram(pfn) && !page_is_reserved_early(pfn) to decide
+		 *  to use it as free.
+		 *  So reserve_early here, hope we don't run out of that array
 		 */
-		if ((node_end_final>>PAGE_SHIFT) < max_low_pfn)
-			reserve_early(node_end_final,
-				      node_end_final+(((u64)size)<<PAGE_SHIFT),
-				      "KVA RAM");
-
-		node_end_pfn[nid] = node_end_final>>PAGE_SHIFT;
-		node_remap_start_pfn[nid] = node_end_pfn[nid];
-		shrink_active_range(nid, node_end_pfn[nid]);
+		reserve_early(node_kva_final,
+			      node_kva_final+(((u64)size)<<PAGE_SHIFT),
+			      "KVA RAM");
+
+		node_remap_start_pfn[nid] = node_kva_final>>PAGE_SHIFT;
+		remove_active_range(nid, node_remap_start_pfn[nid],
+					 node_remap_start_pfn[nid] + size);
 	}
 	printk("Reserving total of %ld pages for numa KVA remap\n",
 			reserve_pages);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index ce8e397a61f6..034a3156d2f0 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -998,7 +998,8 @@ extern void free_area_init_node(int nid, pg_data_t *pgdat,
 extern void free_area_init_nodes(unsigned long *max_zone_pfn);
 extern void add_active_range(unsigned int nid, unsigned long start_pfn,
 					unsigned long end_pfn);
-extern void shrink_active_range(unsigned int nid, unsigned long new_end_pfn);
+extern void remove_active_range(unsigned int nid, unsigned long start_pfn,
+					unsigned long end_pfn);
 extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
 					unsigned long end_pfn);
 extern void remove_all_active_ranges(void);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index eee5ba7509c1..d80e1868e570 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3552,30 +3552,47 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn,
 }
 
 /**
- * shrink_active_range - Shrink an existing registered range of PFNs
+ * remove_active_range - Shrink an existing registered range of PFNs
  * @nid: The node id the range is on that should be shrunk
- * @new_end_pfn: The new PFN of the range
+ * @start_pfn: The new PFN of the range
+ * @end_pfn: The new PFN of the range
  *
  * i386 with NUMA use alloc_remap() to store a node_mem_map on a local node.
  * The map is kept near the end physical page range that has already been
  * registered. This function allows an arch to shrink an existing registered
  * range.
  */
-void __init shrink_active_range(unsigned int nid, unsigned long new_end_pfn)
+void __init remove_active_range(unsigned int nid, unsigned long start_pfn,
+				unsigned long end_pfn)
 {
 	int i, j;
 	int removed = 0;
 
+	printk(KERN_DEBUG "remove_active_range (%d, %lu, %lu)\n",
+			  nid, start_pfn, end_pfn);
+
 	/* Find the old active region end and shrink */
 	for_each_active_range_index_in_nid(i, nid) {
-		if (early_node_map[i].start_pfn >= new_end_pfn) {
+		if (early_node_map[i].start_pfn >= start_pfn &&
+		    early_node_map[i].end_pfn <= end_pfn) {
 			/* clear it */
+			early_node_map[i].start_pfn = 0;
 			early_node_map[i].end_pfn = 0;
 			removed = 1;
 			continue;
 		}
-		if (early_node_map[i].end_pfn > new_end_pfn) {
-			early_node_map[i].end_pfn = new_end_pfn;
+		if (early_node_map[i].start_pfn < start_pfn &&
+		    early_node_map[i].end_pfn > start_pfn) {
+			unsigned long temp_end_pfn = early_node_map[i].end_pfn;
+			early_node_map[i].end_pfn = start_pfn;
+			if (temp_end_pfn > end_pfn)
+				add_active_range(nid, end_pfn, temp_end_pfn);
+			continue;
+		}
+		if (early_node_map[i].start_pfn >= start_pfn &&
+		    early_node_map[i].end_pfn > end_pfn &&
+		    early_node_map[i].start_pfn < end_pfn) {
+			early_node_map[i].start_pfn = end_pfn;
 			continue;
 		}
 	}
-- 
cgit v1.2.3


From b5bc6c0e55000dab86b73f838f5ad02908b23755 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Sat, 14 Jun 2008 18:32:52 -0700
Subject: x86, mm: use add_highpages_with_active_regions() for high pages init
 v2

use early_node_map to init high pages, so we can remove page_is_ram() and
page_is_reserved_early() in the big loop with add_one_highpage

also remove page_is_reserved_early(), it is not needed anymore.

v2: fix the build of other platforms

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/e820.c     | 11 --------
 arch/x86/mm/discontig_32.c | 19 ++++++--------
 arch/x86/mm/init_32.c      | 62 ++++++++++++++++++++++++++++++++++++++--------
 include/asm-x86/e820.h     |  1 -
 include/asm-x86/highmem.h  |  3 +++
 include/linux/mm.h         |  2 ++
 mm/page_alloc.c            |  8 ++++++
 7 files changed, 71 insertions(+), 35 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 5051ce744b4e..ed46b7a6bc13 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -612,17 +612,6 @@ void __init free_early(u64 start, u64 end)
 	early_res[j - 1].end = 0;
 }
 
-int __init page_is_reserved_early(unsigned long pagenr)
-{
-	u64 start = (u64)pagenr << PAGE_SHIFT;
-	int i;
-	struct early_res *r;
-
-	i = find_overlapped_early(start, start + PAGE_SIZE);
-	r = &early_res[i];
-	return (i < MAX_EARLY_RES && r->end);
-}
-
 void __init early_res_to_bootmem(u64 start, u64 end)
 {
 	int i;
diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c
index c3f119e99e0d..7c4d0255f8d8 100644
--- a/arch/x86/mm/discontig_32.c
+++ b/arch/x86/mm/discontig_32.c
@@ -100,7 +100,6 @@ unsigned long node_memmap_size_bytes(int nid, unsigned long start_pfn,
 #endif
 
 extern unsigned long find_max_low_pfn(void);
-extern void add_one_highpage_init(struct page *, int, int);
 extern unsigned long highend_pfn, highstart_pfn;
 
 #define LARGE_PAGE_BYTES (PTRS_PER_PTE * PAGE_SIZE)
@@ -432,10 +431,10 @@ void __init set_highmem_pages_init(int bad_ppro)
 {
 #ifdef CONFIG_HIGHMEM
 	struct zone *zone;
-	struct page *page;
+	int nid;
 
 	for_each_zone(zone) {
-		unsigned long node_pfn, zone_start_pfn, zone_end_pfn;
+		unsigned long zone_start_pfn, zone_end_pfn;
 
 		if (!is_highmem(zone))
 			continue;
@@ -443,16 +442,12 @@ void __init set_highmem_pages_init(int bad_ppro)
 		zone_start_pfn = zone->zone_start_pfn;
 		zone_end_pfn = zone_start_pfn + zone->spanned_pages;
 
+		nid = zone_to_nid(zone);
 		printk("Initializing %s for node %d (%08lx:%08lx)\n",
-				zone->name, zone_to_nid(zone),
-				zone_start_pfn, zone_end_pfn);
-
-		for (node_pfn = zone_start_pfn; node_pfn < zone_end_pfn; node_pfn++) {
-			if (!pfn_valid(node_pfn))
-				continue;
-			page = pfn_to_page(node_pfn);
-			add_one_highpage_init(page, node_pfn, bad_ppro);
-		}
+				zone->name, nid, zone_start_pfn, zone_end_pfn);
+
+		add_highpages_with_active_regions(nid, zone_start_pfn,
+				 zone_end_pfn, bad_ppro);
 	}
 	totalram_pages += totalhigh_pages;
 #endif
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index abadb1da70df..ba07a489230e 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -287,10 +287,10 @@ static void __init permanent_kmaps_init(pgd_t *pgd_base)
 	pkmap_page_table = pte;
 }
 
-void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro)
+static void __init
+add_one_highpage_init(struct page *page, int pfn, int bad_ppro)
 {
-	if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn)) &&
-	    !page_is_reserved_early(pfn)) {
+	if (!(bad_ppro && page_kills_ppro(pfn))) {
 		ClearPageReserved(page);
 		init_page_count(page);
 		__free_page(page);
@@ -299,18 +299,58 @@ void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro)
 		SetPageReserved(page);
 }
 
+struct add_highpages_data {
+	unsigned long start_pfn;
+	unsigned long end_pfn;
+	int bad_ppro;
+};
+
+static void __init add_highpages_work_fn(unsigned long start_pfn,
+					 unsigned long end_pfn, void *datax)
+{
+	int node_pfn;
+	struct page *page;
+	unsigned long final_start_pfn, final_end_pfn;
+	struct add_highpages_data *data;
+	int bad_ppro;
+
+	data = (struct add_highpages_data *)datax;
+	bad_ppro = data->bad_ppro;
+
+	final_start_pfn = max(start_pfn, data->start_pfn);
+	final_end_pfn = min(end_pfn, data->end_pfn);
+	if (final_start_pfn >= final_end_pfn)
+		return;
+
+	for (node_pfn = final_start_pfn; node_pfn < final_end_pfn;
+	     node_pfn++) {
+		if (!pfn_valid(node_pfn))
+			continue;
+		page = pfn_to_page(node_pfn);
+		add_one_highpage_init(page, node_pfn, bad_ppro);
+	}
+
+}
+
+void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn,
+					      unsigned long end_pfn,
+					      int bad_ppro)
+{
+	struct add_highpages_data data;
+
+	data.start_pfn = start_pfn;
+	data.end_pfn = end_pfn;
+	data.bad_ppro = bad_ppro;
+
+	work_with_active_regions(nid, add_highpages_work_fn, &data);
+}
+
 #ifndef CONFIG_NUMA
 static void __init set_highmem_pages_init(int bad_ppro)
 {
-	int pfn;
+	add_highpages_with_active_regions(0, highstart_pfn, highend_pfn,
+						bad_ppro);
 
-	for (pfn = highstart_pfn; pfn < highend_pfn; pfn++) {
-		/*
-		 * Holes under sparsemem might not have no mem_map[]:
-		 */
-		if (pfn_valid(pfn))
-			add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
-	}
 	totalram_pages += totalhigh_pages;
 }
 #endif /* !CONFIG_NUMA */
diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h
index 6b0ce745a60c..55d310596907 100644
--- a/include/asm-x86/e820.h
+++ b/include/asm-x86/e820.h
@@ -86,7 +86,6 @@ extern u64 find_e820_area_size(u64 start, u64 *sizep, u64 align);
 extern void reserve_early(u64 start, u64 end, char *name);
 extern void free_early(u64 start, u64 end);
 extern void early_res_to_bootmem(u64 start, u64 end);
-extern int page_is_reserved_early(unsigned long pagenr);
 extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align);
 
 extern unsigned long e820_end_of_ram(void);
diff --git a/include/asm-x86/highmem.h b/include/asm-x86/highmem.h
index e153f3b44774..85c4fea41ff6 100644
--- a/include/asm-x86/highmem.h
+++ b/include/asm-x86/highmem.h
@@ -74,6 +74,9 @@ struct page *kmap_atomic_to_page(void *ptr);
 
 #define flush_cache_kmaps()	do { } while (0)
 
+extern void add_highpages_with_active_regions(int nid, unsigned long start_pfn,
+					unsigned long end_pfn, int bad_ppro);
+
 #endif /* __KERNEL__ */
 
 #endif /* _ASM_HIGHMEM_H */
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 034a3156d2f0..e4de460907c1 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1011,6 +1011,8 @@ extern unsigned long find_min_pfn_with_active_regions(void);
 extern unsigned long find_max_pfn_with_active_regions(void);
 extern void free_bootmem_with_active_regions(int nid,
 						unsigned long max_low_pfn);
+typedef void (*work_fn_t)(unsigned long, unsigned long, void *);
+extern void work_with_active_regions(int nid, work_fn_t work_fn, void *data);
 extern void sparse_memory_present_with_active_regions(int nid);
 #ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
 extern int early_pfn_to_nid(unsigned long pfn);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index d80e1868e570..41c6e3aa059f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2929,6 +2929,14 @@ void __init free_bootmem_with_active_regions(int nid,
 	}
 }
 
+void __init work_with_active_regions(int nid, work_fn_t work_fn, void *data)
+{
+	int i;
+
+	for_each_active_range_index_in_nid(i, nid)
+		work_fn(early_node_map[i].start_pfn, early_node_map[i].end_pfn,
+			data);
+}
 /**
  * sparse_memory_present_with_active_regions - Call memory_present for each active range
  * @nid: The node to call memory_present for. If MAX_NUMNODES, all nodes will be used.
-- 
cgit v1.2.3


From 3461b0af025251bbc6b3d56c821c6ac2de6f7209 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Mon, 12 May 2008 21:21:13 +0200
Subject: x86: remove static boot_cpu_pda array v2

  * Remove the boot_cpu_pda array and pointer table from the data section.
    Allocate the pointer table and array during init.  do_boot_cpu()
    will reallocate the pda in node local memory and if the cpu is being
    brought up before the bootmem array is released (after_bootmem = 0),
    then it will free the initial pda.  This will happen for all cpus
    present at system startup.

    This removes 512k + 32k bytes from the data section.

For inclusion into sched-devel/latest tree.

Based on:
	git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
    +   sched-devel/latest  .../mingo/linux-2.6-sched-devel.git

Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/head64.c  | 26 +++++++++++++++--
 arch/x86/kernel/setup.c   | 73 ++++++++++++++++++++++++++++++++++++-----------
 arch/x86/kernel/setup64.c |  8 ++++--
 arch/x86/kernel/smpboot.c | 59 +++++++++++++++++++++++++++++---------
 include/asm-x86/pda.h     |  6 ++--
 include/linux/mm.h        |  1 +
 6 files changed, 135 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index e25c57b8aa84..0ab59edd7067 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -25,6 +25,24 @@
 #include <asm/e820.h>
 #include <asm/bios_ebda.h>
 
+/* boot cpu pda */
+static struct x8664_pda _boot_cpu_pda __read_mostly;
+
+#ifdef CONFIG_SMP
+#ifdef CONFIG_DEBUG_PER_CPU_MAPS
+/*
+ * We install an empty cpu_pda pointer table to trap references before
+ * the actual cpu_pda pointer table is created in setup_cpu_pda_map().
+ */
+static struct x8664_pda *__cpu_pda[NR_CPUS] __initdata;
+#else
+static struct x8664_pda *__cpu_pda[1] __read_mostly;
+#endif
+
+#else /* !CONFIG_SMP (NR_CPUS will be 1) */
+static struct x8664_pda *__cpu_pda[NR_CPUS] __read_mostly;
+#endif
+
 static void __init zap_identity_mappings(void)
 {
 	pgd_t *pgd = pgd_offset_k(0UL);
@@ -156,10 +174,12 @@ void __init x86_64_start_kernel(char * real_mode_data)
 
 	early_printk("Kernel alive\n");
 
- 	for (i = 0; i < NR_CPUS; i++)
- 		cpu_pda(i) = &boot_cpu_pda[i];
-
+	_cpu_pda = __cpu_pda;
+	cpu_pda(0) = &_boot_cpu_pda;
 	pda_init(0);
+
+	early_printk("Kernel really alive\n");
+
 	copy_bootdata(__va(real_mode_data));
 
 	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 913af838c3c5..dd12c1c84a8f 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -101,6 +101,50 @@ static inline void setup_cpumask_of_cpu(void) { }
  */
 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 EXPORT_SYMBOL(__per_cpu_offset);
+static inline void setup_cpu_pda_map(void) { }
+
+#elif !defined(CONFIG_SMP)
+static inline void setup_cpu_pda_map(void) { }
+
+#else /* CONFIG_SMP && CONFIG_X86_64 */
+
+/*
+ * Allocate cpu_pda pointer table and array via alloc_bootmem.
+ */
+static void __init setup_cpu_pda_map(void)
+{
+	char *pda;
+	struct x8664_pda **new_cpu_pda;
+	unsigned long size;
+	int cpu;
+
+	size = roundup(sizeof(struct x8664_pda), cache_line_size());
+
+	/* allocate cpu_pda array and pointer table */
+	{
+		unsigned long tsize = nr_cpu_ids * sizeof(void *);
+		unsigned long asize = size * (nr_cpu_ids - 1);
+
+		tsize = roundup(tsize, cache_line_size());
+		new_cpu_pda = alloc_bootmem(tsize + asize);
+		pda = (char *)new_cpu_pda + tsize;
+	}
+
+	/* initialize pointer table to static pda's */
+	for_each_possible_cpu(cpu) {
+		if (cpu == 0) {
+			/* leave boot cpu pda in place */
+			new_cpu_pda[0] = cpu_pda(0);
+			continue;
+		}
+		new_cpu_pda[cpu] = (struct x8664_pda *)pda;
+		new_cpu_pda[cpu]->in_bootmem = 1;
+		pda += size;
+	}
+
+	/* point to new pointer table */
+	_cpu_pda = new_cpu_pda;
+}
 #endif
 
 /*
@@ -110,46 +154,43 @@ EXPORT_SYMBOL(__per_cpu_offset);
  */
 void __init setup_per_cpu_areas(void)
 {
-	int i, highest_cpu = 0;
-	unsigned long size;
+	ssize_t size = PERCPU_ENOUGH_ROOM;
+	char *ptr;
+	int cpu;
 
 #ifdef CONFIG_HOTPLUG_CPU
 	prefill_possible_map();
+#else
+	nr_cpu_ids = num_processors;
 #endif
 
+	/* Setup cpu_pda map */
+	setup_cpu_pda_map();
+
 	/* Copy section for each CPU (we discard the original) */
 	size = PERCPU_ENOUGH_ROOM;
 	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n",
 			  size);
 
-	for_each_possible_cpu(i) {
-		char *ptr;
+	for_each_possible_cpu(cpu) {
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 		ptr = alloc_bootmem_pages(size);
 #else
-		int node = early_cpu_to_node(i);
+		int node = early_cpu_to_node(cpu);
 		if (!node_online(node) || !NODE_DATA(node)) {
 			ptr = alloc_bootmem_pages(size);
 			printk(KERN_INFO
 			       "cpu %d has no node %d or node-local memory\n",
-				i, node);
+				cpu, node);
 		}
 		else
 			ptr = alloc_bootmem_pages_node(NODE_DATA(node), size);
 #endif
-		if (!ptr)
-			panic("Cannot allocate cpu data for CPU %d\n", i);
-#ifdef CONFIG_X86_64
-		cpu_pda(i)->data_offset = ptr - __per_cpu_start;
-#else
-		__per_cpu_offset[i] = ptr - __per_cpu_start;
-#endif
+		per_cpu_offset(cpu) = ptr - __per_cpu_start;
 		memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 
-		highest_cpu = i;
 	}
 
-	nr_cpu_ids = highest_cpu + 1;
 	printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d, nr_node_ids %d\n",
 		NR_CPUS, nr_cpu_ids, nr_node_ids);
 
@@ -199,7 +240,7 @@ void __cpuinit numa_set_node(int cpu, int node)
 {
 	int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map);
 
-	if (node != NUMA_NO_NODE)
+	if (cpu_pda(cpu) && node != NUMA_NO_NODE)
 		cpu_pda(cpu)->nodenumber = node;
 
 	if (cpu_to_node_map)
diff --git a/arch/x86/kernel/setup64.c b/arch/x86/kernel/setup64.c
index aee0e8200777..631ea6cc01d8 100644
--- a/arch/x86/kernel/setup64.c
+++ b/arch/x86/kernel/setup64.c
@@ -12,6 +12,7 @@
 #include <linux/bitops.h>
 #include <linux/module.h>
 #include <linux/kgdb.h>
+#include <linux/topology.h>
 #include <asm/pda.h>
 #include <asm/pgtable.h>
 #include <asm/processor.h>
@@ -34,9 +35,8 @@ struct boot_params boot_params;
 
 cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
 
-struct x8664_pda *_cpu_pda[NR_CPUS] __read_mostly;
+struct x8664_pda **_cpu_pda __read_mostly;
 EXPORT_SYMBOL(_cpu_pda);
-struct x8664_pda boot_cpu_pda[NR_CPUS] __cacheline_aligned;
 
 struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
 
@@ -114,8 +114,10 @@ void pda_init(int cpu)
 			__get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
 		if (!pda->irqstackptr)
 			panic("cannot allocate irqstack for cpu %d", cpu); 
-	}
 
+		if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE)
+			pda->nodenumber = cpu_to_node(cpu);
+	}
 
 	pda->irqstackptr += IRQSTACKSIZE-64;
 } 
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 036604d3daed..bf0833487455 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -816,6 +816,43 @@ static void __cpuinit do_fork_idle(struct work_struct *work)
 	complete(&c_idle->done);
 }
 
+/*
+ * Allocate node local memory for the AP pda.
+ *
+ * Must be called after the _cpu_pda pointer table is initialized.
+ */
+static int __cpuinit get_local_pda(int cpu)
+{
+	struct x8664_pda *oldpda, *newpda;
+	unsigned long size = sizeof(struct x8664_pda);
+	int node = cpu_to_node(cpu);
+
+	if (cpu_pda(cpu) && !cpu_pda(cpu)->in_bootmem)
+		return 0;
+
+	oldpda = cpu_pda(cpu);
+	newpda = kmalloc_node(size, GFP_ATOMIC, node);
+	if (!newpda) {
+		printk(KERN_ERR "Could not allocate node local PDA "
+			"for CPU %d on node %d\n", cpu, node);
+
+		if (oldpda)
+			return 0;	/* have a usable pda */
+		else
+			return -1;
+	}
+
+	if (oldpda) {
+		memcpy(newpda, oldpda, size);
+		if (!after_bootmem)
+			free_bootmem((unsigned long)oldpda, size);
+	}
+
+	newpda->in_bootmem = 0;
+	cpu_pda(cpu) = newpda;
+	return 0;
+}
+
 static int __cpuinit do_boot_cpu(int apicid, int cpu)
 /*
  * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
@@ -841,19 +878,11 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu)
 	}
 
 	/* Allocate node local memory for AP pdas */
-	if (cpu_pda(cpu) == &boot_cpu_pda[cpu]) {
-		struct x8664_pda *newpda, *pda;
-		int node = cpu_to_node(cpu);
-		pda = cpu_pda(cpu);
-		newpda = kmalloc_node(sizeof(struct x8664_pda), GFP_ATOMIC,
-				      node);
-		if (newpda) {
-			memcpy(newpda, pda, sizeof(struct x8664_pda));
-			cpu_pda(cpu) = newpda;
-		} else
-			printk(KERN_ERR
-		"Could not allocate node local PDA for CPU %d on node %d\n",
-				cpu, node);
+	if (cpu > 0) {
+		boot_error = get_local_pda(cpu);
+		if (boot_error)
+			goto restore_state;
+			/* if can't get pda memory, can't start cpu */
 	}
 #endif
 
@@ -972,6 +1001,8 @@ do_rest:
 		}
 	}
 
+restore_state:
+
 	if (boot_error) {
 		/* Try to put things back the way they were before ... */
 		unmap_cpu_to_logical_apicid(cpu);
@@ -1347,6 +1378,8 @@ __init void prefill_possible_map(void)
 
 	for (i = 0; i < possible; i++)
 		cpu_set(i, cpu_possible_map);
+
+	nr_cpu_ids = possible;
 }
 
 static void __ref remove_cpu_from_maps(int cpu)
diff --git a/include/asm-x86/pda.h b/include/asm-x86/pda.h
index de2ad9ac35a9..b34e9a7cc80b 100644
--- a/include/asm-x86/pda.h
+++ b/include/asm-x86/pda.h
@@ -22,7 +22,8 @@ struct x8664_pda {
 					   offset 40!!! */
 #endif
 	char *irqstackptr;
-	int nodenumber;			/* number of current node */
+	short nodenumber;		/* number of current node (32k max) */
+	short in_bootmem;		/* pda lives in bootmem */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
 	short mmu_state;
@@ -38,8 +39,7 @@ struct x8664_pda {
 	unsigned irq_spurious_count;
 } ____cacheline_aligned_in_smp;
 
-extern struct x8664_pda *_cpu_pda[];
-extern struct x8664_pda boot_cpu_pda[];
+extern struct x8664_pda **_cpu_pda;
 extern void pda_init(int);
 
 #define cpu_pda(i) (_cpu_pda[i])
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 586a943cab01..0ea48a5af823 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1024,6 +1024,7 @@ extern void mem_init(void);
 extern void show_mem(void);
 extern void si_meminfo(struct sysinfo * val);
 extern void si_meminfo_node(struct sysinfo *val, int nid);
+extern int after_bootmem;
 
 #ifdef CONFIG_NUMA
 extern void setup_per_cpu_pageset(void);
-- 
cgit v1.2.3


From cdf060a5d3c2afc7998af94b26a6c5182419e071 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 8 Jul 2008 02:36:40 -0700
Subject: netfilter: cleanup netfilter_ipv6.h userspace header

Kernel functions are not for userspace.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv6.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h
index fd50988b83ec..d654873aa25a 100644
--- a/include/linux/netfilter_ipv6.h
+++ b/include/linux/netfilter_ipv6.h
@@ -70,6 +70,8 @@ enum nf_ip6_hook_priorities {
 	NF_IP6_PRI_LAST = INT_MAX,
 };
 
+#ifdef  __KERNEL__
+
 #ifdef CONFIG_NETFILTER
 extern int ip6_route_me_harder(struct sk_buff *skb);
 extern __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook,
@@ -82,4 +84,6 @@ static inline int ipv6_netfilter_init(void) { return 0; }
 static inline void ipv6_netfilter_fini(void) { return; }
 #endif /* CONFIG_NETFILTER */
 
+#endif /* __KERNEL__ */
+
 #endif /*__LINUX_IP6_NETFILTER_H*/
-- 
cgit v1.2.3


From b9c796783151678d08b1ec1ef410685b2515ba51 Mon Sep 17 00:00:00 2001
From: Joonwoo Park <joonwpark81@gmail.com>
Date: Tue, 8 Jul 2008 02:37:31 -0700
Subject: textsearch: support for case insensitive searching

The function textsearch_prepare has a new flag to support case
insensitive searching.

Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/textsearch.h | 10 +++++-----
 lib/textsearch.c           | 14 ++++++++------
 2 files changed, 13 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h
index 6f371f24160b..6c34cf28b7aa 100644
--- a/include/linux/textsearch.h
+++ b/include/linux/textsearch.h
@@ -10,10 +10,8 @@
 
 struct ts_config;
 
-/**
- * TS_AUTOLOAD - Automatically load textsearch modules when needed
- */
-#define TS_AUTOLOAD	1
+#define TS_AUTOLOAD	1 /* Automatically load textsearch modules when needed */
+#define TS_IGNORECASE	2 /* Searches string case insensitively */
 
 /**
  * struct ts_state - search state
@@ -39,7 +37,7 @@ struct ts_state
 struct ts_ops
 {
 	const char		*name;
-	struct ts_config *	(*init)(const void *, unsigned int, gfp_t);
+	struct ts_config *	(*init)(const void *, unsigned int, gfp_t, int);
 	unsigned int		(*find)(struct ts_config *,
 					struct ts_state *);
 	void			(*destroy)(struct ts_config *);
@@ -52,12 +50,14 @@ struct ts_ops
 /**
  * struct ts_config - search configuration
  * @ops: operations of chosen algorithm
+ * @flags: flags
  * @get_next_block: callback to fetch the next block to search in
  * @finish: callback to finalize a search
  */
 struct ts_config
 {
 	struct ts_ops		*ops;
+	int 			flags;
 
 	/**
 	 * get_next_block - fetch next block of data
diff --git a/lib/textsearch.c b/lib/textsearch.c
index be8bda3862f5..b451fcc9354c 100644
--- a/lib/textsearch.c
+++ b/lib/textsearch.c
@@ -54,10 +54,13 @@
  * USAGE
  *
  *   Before a search can be performed, a configuration must be created
- *   by calling textsearch_prepare() specyfing the searching algorithm and
- *   the pattern to look for. The returned configuration may then be used
- *   for an arbitary amount of times and even in parallel as long as a
- *   separate struct ts_state variable is provided to every instance.
+ *   by calling textsearch_prepare() specifying the searching algorithm,
+ *   the pattern to look for and flags. As a flag, you can set TS_IGNORECASE
+ *   to perform case insensitive matching. But it might slow down
+ *   performance of algorithm, so you should use it at own your risk.
+ *   The returned configuration may then be used for an arbitary
+ *   amount of times and even in parallel as long as a separate struct
+ *   ts_state variable is provided to every instance.
  *
  *   The actual search is performed by either calling textsearch_find_-
  *   continuous() for linear data or by providing an own get_next_block()
@@ -89,7 +92,6 @@
  *       panic("Oh my god, dancing chickens at %d\n", pos);
  *
  *   textsearch_destroy(conf);
- *
  * ==========================================================================
  */
 
@@ -279,7 +281,7 @@ struct ts_config *textsearch_prepare(const char *algo, const void *pattern,
 	if (ops == NULL)
 		goto errout;
 
-	conf = ops->init(pattern, len, gfp_mask);
+	conf = ops->init(pattern, len, gfp_mask, flags);
 	if (IS_ERR(conf)) {
 		err = PTR_ERR(conf);
 		goto errout;
-- 
cgit v1.2.3


From dde77e604497dada6f224a685278dfb27747ae52 Mon Sep 17 00:00:00 2001
From: Joonwoo Park <joonwpark81@gmail.com>
Date: Tue, 8 Jul 2008 02:38:40 -0700
Subject: textsearch: convert kmalloc + memset to kzalloc

convert kmalloc + memset to kzalloc for alloc_ts_config

Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/textsearch.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h
index 6c34cf28b7aa..d9a85d616385 100644
--- a/include/linux/textsearch.h
+++ b/include/linux/textsearch.h
@@ -162,11 +162,10 @@ static inline struct ts_config *alloc_ts_config(size_t payload,
 {
 	struct ts_config *conf;
 
-	conf = kmalloc(TS_PRIV_ALIGN(sizeof(*conf)) + payload, gfp_mask);
+	conf = kzalloc(TS_PRIV_ALIGN(sizeof(*conf)) + payload, gfp_mask);
 	if (conf == NULL)
 		return ERR_PTR(-ENOMEM);
 
-	memset(conf, 0, TS_PRIV_ALIGN(sizeof(*conf)) + payload);
 	return conf;
 }
 
-- 
cgit v1.2.3


From 4ad3f26162ece5aca3045fd45e15dd99acea4a0e Mon Sep 17 00:00:00 2001
From: Joonwoo Park <joonwpark81@gmail.com>
Date: Tue, 8 Jul 2008 02:38:56 -0700
Subject: netfilter: fix string extension for case insensitive pattern matching

The flag XT_STRING_FLAG_IGNORECASE indicates case insensitive string
matching. netfilter can find cmd.exe, Cmd.exe, cMd.exe and etc easily.

A new revision 1 was added, in the meantime invert of xt_string_info
was moved into flags as a flag. If revision is 1, The flag
XT_STRING_FLAG_INVERT indicates invert matching.

Signed-off-by: Joonwoo Park <joonwpark81@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/xt_string.h | 15 ++++++++++++++-
 net/netfilter/xt_string.c           | 38 +++++++++++++++++++++++++++++++++++--
 2 files changed, 50 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/xt_string.h b/include/linux/netfilter/xt_string.h
index bb21dd1aee2d..8a6ba7bbef9f 100644
--- a/include/linux/netfilter/xt_string.h
+++ b/include/linux/netfilter/xt_string.h
@@ -4,6 +4,11 @@
 #define XT_STRING_MAX_PATTERN_SIZE 128
 #define XT_STRING_MAX_ALGO_NAME_SIZE 16
 
+enum {
+	XT_STRING_FLAG_INVERT		= 0x01,
+	XT_STRING_FLAG_IGNORECASE	= 0x02
+};
+
 struct xt_string_info
 {
 	u_int16_t from_offset;
@@ -11,7 +16,15 @@ struct xt_string_info
 	char	  algo[XT_STRING_MAX_ALGO_NAME_SIZE];
 	char 	  pattern[XT_STRING_MAX_PATTERN_SIZE];
 	u_int8_t  patlen;
-	u_int8_t  invert;
+	union {
+		struct {
+			u_int8_t  invert;
+		} v0;
+
+		struct {
+			u_int8_t  flags;
+		} v1;
+	} u;
 
 	/* Used internally by the kernel */
 	struct ts_config __attribute__((aligned(8))) *config;
diff --git a/net/netfilter/xt_string.c b/net/netfilter/xt_string.c
index 72f694d947f4..4903182a062b 100644
--- a/net/netfilter/xt_string.c
+++ b/net/netfilter/xt_string.c
@@ -29,12 +29,16 @@ string_mt(const struct sk_buff *skb, const struct net_device *in,
 {
 	const struct xt_string_info *conf = matchinfo;
 	struct ts_state state;
+	int invert;
 
 	memset(&state, 0, sizeof(struct ts_state));
 
+	invert = (match->revision == 0 ? conf->u.v0.invert :
+				    conf->u.v1.flags & XT_STRING_FLAG_INVERT);
+
 	return (skb_find_text((struct sk_buff *)skb, conf->from_offset,
 			     conf->to_offset, conf->config, &state)
-			     != UINT_MAX) ^ conf->invert;
+			     != UINT_MAX) ^ invert;
 }
 
 #define STRING_TEXT_PRIV(m) ((struct xt_string_info *)(m))
@@ -46,6 +50,7 @@ string_mt_check(const char *tablename, const void *ip,
 {
 	struct xt_string_info *conf = matchinfo;
 	struct ts_config *ts_conf;
+	int flags = TS_AUTOLOAD;
 
 	/* Damn, can't handle this case properly with iptables... */
 	if (conf->from_offset > conf->to_offset)
@@ -54,8 +59,15 @@ string_mt_check(const char *tablename, const void *ip,
 		return false;
 	if (conf->patlen > XT_STRING_MAX_PATTERN_SIZE)
 		return false;
+	if (match->revision == 1) {
+		if (conf->u.v1.flags &
+		    ~(XT_STRING_FLAG_IGNORECASE | XT_STRING_FLAG_INVERT))
+			return false;
+		if (conf->u.v1.flags & XT_STRING_FLAG_IGNORECASE)
+			flags |= TS_IGNORECASE;
+	}
 	ts_conf = textsearch_prepare(conf->algo, conf->pattern, conf->patlen,
-				     GFP_KERNEL, TS_AUTOLOAD);
+				     GFP_KERNEL, flags);
 	if (IS_ERR(ts_conf))
 		return false;
 
@@ -72,6 +84,17 @@ static void string_mt_destroy(const struct xt_match *match, void *matchinfo)
 static struct xt_match string_mt_reg[] __read_mostly = {
 	{
 		.name 		= "string",
+		.revision	= 0,
+		.family		= AF_INET,
+		.checkentry	= string_mt_check,
+		.match 		= string_mt,
+		.destroy 	= string_mt_destroy,
+		.matchsize	= sizeof(struct xt_string_info),
+		.me 		= THIS_MODULE
+	},
+	{
+		.name 		= "string",
+		.revision	= 1,
 		.family		= AF_INET,
 		.checkentry	= string_mt_check,
 		.match 		= string_mt,
@@ -81,6 +104,17 @@ static struct xt_match string_mt_reg[] __read_mostly = {
 	},
 	{
 		.name 		= "string",
+		.revision	= 0,
+		.family		= AF_INET6,
+		.checkentry	= string_mt_check,
+		.match 		= string_mt,
+		.destroy 	= string_mt_destroy,
+		.matchsize	= sizeof(struct xt_string_info),
+		.me 		= THIS_MODULE
+	},
+	{
+		.name 		= "string",
+		.revision	= 1,
 		.family		= AF_INET6,
 		.checkentry	= string_mt_check,
 		.match 		= string_mt,
-- 
cgit v1.2.3


From 2c693610fe923764fe41b846fb86938a2010da6e Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Tue, 8 Jul 2008 03:03:01 -0700
Subject: net: remove padding from struct socket on 64bit & increase
 objects/cache

remove padding from struct socket reducing its size by 8 bytes.

This allows more objects/cache in sock_inode_cache
12 objects/cache when cacheline size is 128 (generic x86_64)

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/net.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/net.h b/include/linux/net.h
index 71f7dd559285..150a48c68d52 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -106,23 +106,23 @@ enum sock_shutdown_cmd {
 /**
  *  struct socket - general BSD socket
  *  @state: socket state (%SS_CONNECTED, etc)
+ *  @type: socket type (%SOCK_STREAM, etc)
  *  @flags: socket flags (%SOCK_ASYNC_NOSPACE, etc)
  *  @ops: protocol specific socket operations
  *  @fasync_list: Asynchronous wake up list
  *  @file: File back pointer for gc
  *  @sk: internal networking protocol agnostic socket representation
  *  @wait: wait queue for several uses
- *  @type: socket type (%SOCK_STREAM, etc)
  */
 struct socket {
 	socket_state		state;
+	short			type;
 	unsigned long		flags;
 	const struct proto_ops	*ops;
 	struct fasync_struct	*fasync_list;
 	struct file		*file;
 	struct sock		*sk;
 	wait_queue_head_t	wait;
-	short			type;
 };
 
 struct vm_area_struct;
-- 
cgit v1.2.3


From acc81e1465d29e0284008770cc4b8bc90bd93bd7 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:21:27 -0700
Subject: vlan: fix network_header/mac_header adjustments

Lennert Buytenhek points out that the VLAN code incorrectly adjusts
skb->network_header to point in the middle of the VLAN header and
additionally tries to adjust skb->mac_header without checking for
validity.

The network_header should not be touched at all since we're only
adding headers in front of it, mac_header adjustments are not
necessary at all.

Based on patch by Lennert Buytenhek <buytenh@wantstofly.org>.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 2 --
 net/8021q/vlan_dev.c    | 1 -
 2 files changed, 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 5190452ac7dc..8f5bf9b676aa 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -279,8 +279,6 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short
 	veth->h_vlan_TCI = htons(tag);
 
 	skb->protocol = htons(ETH_P_8021Q);
-	skb->mac_header -= VLAN_HLEN;
-	skb->network_header -= VLAN_HLEN;
 
 	return skb;
 }
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index a0617bf7cec6..4a8525927c27 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -308,7 +308,6 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 			vhdr->h_vlan_encapsulated_proto = htons(len);
 
 		skb->protocol = htons(ETH_P_8021Q);
-		skb_reset_network_header(skb);
 	}
 
 	/* Before delegating work to the lower layer, enter our MAC-address */
-- 
cgit v1.2.3


From 7750f403cbe56971336d575b354365190b4e3227 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:23:36 -0700
Subject: vlan: uninline __vlan_hwaccel_rx

The function is huge and included at least once in every VLAN acceleration
capable driver. Uninline it; to avoid having drivers depend on the VLAN
module, the function is always built in statically when VLAN is enabled.

With all VLAN acceleration capable drivers that build on x86_64 enabled,
this results in:

   text    data     bss     dec     hex filename
6515227  854044  343968 7713239  75b1d7 vmlinux.inlined
6505637  854044  343968 7703649  758c61 vmlinux.uninlined
----------------------------------------------------------
  -9590

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 64 +++++++------------------------------------------
 net/8021q/Makefile      |  9 +++----
 net/8021q/vlan.h        |  8 +++++++
 net/8021q/vlan_core.c   | 48 +++++++++++++++++++++++++++++++++++++
 net/Makefile            |  4 +++-
 5 files changed, 72 insertions(+), 61 deletions(-)
 create mode 100644 net/8021q/vlan_core.c

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 8f5bf9b676aa..594cd35b0074 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -150,15 +150,6 @@ static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
 	return netdev_priv(dev);
 }
 
-/* inline functions */
-static inline __u32 vlan_get_ingress_priority(struct net_device *dev,
-					      unsigned short vlan_tag)
-{
-	struct vlan_dev_info *vip = vlan_dev_info(dev);
-
-	return vip->ingress_priority_map[(vlan_tag >> 13) & 0x7];
-}
-
 /* VLAN tx hw acceleration helpers. */
 struct vlan_skb_tx_cookie {
 	u32	magic;
@@ -171,56 +162,17 @@ struct vlan_skb_tx_cookie {
 	(VLAN_TX_SKB_CB(__skb)->magic == VLAN_TX_COOKIE_MAGIC)
 #define vlan_tx_tag_get(__skb)	(VLAN_TX_SKB_CB(__skb)->vlan_tag)
 
-/* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
-static inline int __vlan_hwaccel_rx(struct sk_buff *skb,
-				    struct vlan_group *grp,
+#if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
+			     unsigned short vlan_tag, int polling);
+#else
+static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 				    unsigned short vlan_tag, int polling)
 {
-	struct net_device_stats *stats;
-
-	if (skb_bond_should_drop(skb)) {
-		dev_kfree_skb_any(skb);
-		return NET_RX_DROP;
-	}
-
-	skb->dev = vlan_group_get_device(grp, vlan_tag & VLAN_VID_MASK);
-	if (skb->dev == NULL) {
-		dev_kfree_skb_any(skb);
-
-		/* Not NET_RX_DROP, this is not being dropped
-		 * due to congestion.
-		 */
-		return 0;
-	}
-
-	skb->dev->last_rx = jiffies;
-
-	stats = &skb->dev->stats;
-	stats->rx_packets++;
-	stats->rx_bytes += skb->len;
-
-	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tag);
-	switch (skb->pkt_type) {
-	case PACKET_BROADCAST:
-		break;
-
-	case PACKET_MULTICAST:
-		stats->multicast++;
-		break;
-
-	case PACKET_OTHERHOST:
-		/* Our lower layer thinks this is not local, let's make sure.
-		 * This allows the VLAN to have a different MAC than the underlying
-		 * device, and still route correctly.
-		 */
-		if (!compare_ether_addr(eth_hdr(skb)->h_dest,
-				       	skb->dev->dev_addr))
-			skb->pkt_type = PACKET_HOST;
-		break;
-	};
-
-	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
+	BUG();
+	return NET_XMIT_SUCCESS;
 }
+#endif
 
 static inline int vlan_hwaccel_rx(struct sk_buff *skb,
 				  struct vlan_group *grp,
diff --git a/net/8021q/Makefile b/net/8021q/Makefile
index 3006e9ed7b08..9f4f174ead1c 100644
--- a/net/8021q/Makefile
+++ b/net/8021q/Makefile
@@ -1,9 +1,10 @@
 #
 # Makefile for the Linux VLAN layer.
 #
+obj-$(subst m,y,$(CONFIG_VLAN_8021Q))	+= vlan_core.o
+obj-$(CONFIG_VLAN_8021Q)		+= 8021q.o
 
-obj-$(CONFIG_VLAN_8021Q) += 8021q.o
+8021q-y					:= vlan.o vlan_dev.o vlan_netlink.o
+8021q-$(CONFIG_VLAN_8021Q_GVRP)		+= vlan_gvrp.o
+8021q-$(CONFIG_PROC_FS)			+= vlanproc.o
 
-8021q-y				:= vlan.o vlan_dev.o vlan_netlink.o
-8021q-$(CONFIG_VLAN_8021Q_GVRP)	+= vlan_gvrp.o
-8021q-$(CONFIG_PROC_FS)		+= vlanproc.o
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 097b2e04c928..7cc1a97c42fc 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -37,6 +37,14 @@ void vlan_setup(struct net_device *dev);
 int register_vlan_dev(struct net_device *dev);
 void unregister_vlan_dev(struct net_device *dev);
 
+static inline u32 vlan_get_ingress_priority(struct net_device *dev,
+					    unsigned short vlan_tag)
+{
+	struct vlan_dev_info *vip = vlan_dev_info(dev);
+
+	return vip->ingress_priority_map[(vlan_tag >> 13) & 0x7];
+}
+
 #ifdef CONFIG_VLAN_8021Q_GVRP
 extern int vlan_gvrp_request_join(const struct net_device *dev);
 extern void vlan_gvrp_request_leave(const struct net_device *dev);
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
new file mode 100644
index 000000000000..85c94edb000f
--- /dev/null
+++ b/net/8021q/vlan_core.c
@@ -0,0 +1,48 @@
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/if_vlan.h>
+#include "vlan.h"
+
+/* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
+int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
+		      unsigned short vlan_tag, int polling)
+{
+	struct net_device_stats *stats;
+
+	if (skb_bond_should_drop(skb)) {
+		dev_kfree_skb_any(skb);
+		return NET_RX_DROP;
+	}
+
+	skb->dev = vlan_group_get_device(grp, vlan_tag & VLAN_VID_MASK);
+	if (skb->dev == NULL) {
+		dev_kfree_skb_any(skb);
+		/* Not NET_RX_DROP, this is not being dropped
+		 * due to congestion. */
+		return NET_RX_SUCCESS;
+	}
+	skb->dev->last_rx = jiffies;
+
+	stats = &skb->dev->stats;
+	stats->rx_packets++;
+	stats->rx_bytes += skb->len;
+
+	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tag);
+	switch (skb->pkt_type) {
+	case PACKET_BROADCAST:
+		break;
+	case PACKET_MULTICAST:
+		stats->multicast++;
+		break;
+	case PACKET_OTHERHOST:
+		/* Our lower layer thinks this is not local, let's make sure.
+		 * This allows the VLAN to have a different MAC than the
+		 * underlying device, and still route correctly. */
+		if (!compare_ether_addr(eth_hdr(skb)->h_dest,
+					skb->dev->dev_addr))
+			skb->pkt_type = PACKET_HOST;
+		break;
+	};
+	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
+}
+EXPORT_SYMBOL(__vlan_hwaccel_rx);
diff --git a/net/Makefile b/net/Makefile
index b7a13643b549..4f43e7f874f3 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -42,7 +42,9 @@ obj-$(CONFIG_AF_RXRPC)		+= rxrpc/
 obj-$(CONFIG_ATM)		+= atm/
 obj-$(CONFIG_DECNET)		+= decnet/
 obj-$(CONFIG_ECONET)		+= econet/
-obj-$(CONFIG_VLAN_8021Q)	+= 8021q/
+ifneq ($(CONFIG_VLAN_8021Q),)
+obj-y				+= 8021q/
+endif
 obj-$(CONFIG_IP_DCCP)		+= dccp/
 obj-$(CONFIG_IP_SCTP)		+= sctp/
 obj-y				+= wireless/
-- 
cgit v1.2.3


From 22d1ba74bbafa96d3f425cc12714d3fe8675183f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:23:57 -0700
Subject: vlan: move struct vlan_dev_info to private header

Hide struct vlan_dev_info from drivers to prevent them from growing
more creative ways to use it. Provide accessors for the two drivers
that currently use it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/cxgb3/l2t.c         |  2 +-
 drivers/s390/net/qeth_l3_main.c |  4 +--
 include/linux/if_vlan.h         | 56 +++++++++++------------------------------
 net/8021q/vlan.c                |  4 +--
 net/8021q/vlan.h                | 50 +++++++++++++++++++++++++++++++++++-
 net/8021q/vlan_core.c           | 12 +++++++++
 net/8021q/vlan_dev.c            |  5 ----
 7 files changed, 80 insertions(+), 53 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/cxgb3/l2t.c b/drivers/net/cxgb3/l2t.c
index f510140885ae..825e510bd9ed 100644
--- a/drivers/net/cxgb3/l2t.c
+++ b/drivers/net/cxgb3/l2t.c
@@ -337,7 +337,7 @@ struct l2t_entry *t3_l2t_get(struct t3cdev *cdev, struct neighbour *neigh,
 		atomic_set(&e->refcnt, 1);
 		neigh_replace(e, neigh);
 		if (neigh->dev->priv_flags & IFF_802_1Q_VLAN)
-			e->vlan = vlan_dev_info(neigh->dev)->vlan_id;
+			e->vlan = vlan_dev_vlan_id(neigh->dev);
 		else
 			e->vlan = VLAN_NONE;
 		spin_unlock(&e->lock);
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 999552c83bbe..85be40abdda9 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2020,7 +2020,7 @@ static int qeth_l3_verify_vlan_dev(struct net_device *dev,
 		}
 	}
 
-	if (rc && !(netdev_priv(vlan_dev_info(dev)->real_dev) == (void *)card))
+	if (rc && !(netdev_priv(vlan_dev_real_dev(dev)) == (void *)card))
 		return 0;
 
 	return rc;
@@ -2056,7 +2056,7 @@ static struct qeth_card *qeth_l3_get_card_from_dev(struct net_device *dev)
 	if (rc == QETH_REAL_CARD)
 		card = netdev_priv(dev);
 	else if (rc == QETH_VLAN_CARD)
-		card = netdev_priv(vlan_dev_info(dev)->real_dev);
+		card = netdev_priv(vlan_dev_real_dev(dev));
 	if (card && card->options.layer2)
 		card = NULL;
 	QETH_DBF_TEXT_(TRACE, 4, "%d", rc);
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 594cd35b0074..cb2e6b480882 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -109,47 +109,6 @@ static inline void vlan_group_set_device(struct vlan_group *vg,
 	array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN] = dev;
 }
 
-struct vlan_priority_tci_mapping {
-	u32 priority;
-	unsigned short vlan_qos; /* This should be shifted when first set, so we only do it
-				  * at provisioning time.
-				  * ((skb->priority << 13) & 0xE000)
-				  */
-	struct vlan_priority_tci_mapping *next;
-};
-
-/* Holds information that makes sense if this device is a VLAN device. */
-struct vlan_dev_info {
-	/** This will be the mapping that correlates skb->priority to
-	 * 3 bits of VLAN QOS tags...
-	 */
-	unsigned int nr_ingress_mappings;
-	u32 ingress_priority_map[8];
-
-	unsigned int nr_egress_mappings;
-	struct vlan_priority_tci_mapping *egress_priority_map[16]; /* hash table */
-
-	unsigned short vlan_id;        /*  The VLAN Identifier for this interface. */
-	unsigned short flags;          /* (1 << 0) re_order_header   This option will cause the
-                                        *   VLAN code to move around the ethernet header on
-                                        *   ingress to make the skb look **exactly** like it
-                                        *   came in from an ethernet port.  This destroys some of
-                                        *   the VLAN information in the skb, but it fixes programs
-                                        *   like DHCP that use packet-filtering and don't understand
-                                        *   802.1Q
-                                        */
-	struct net_device *real_dev;    /* the underlying device/interface */
-	unsigned char real_dev_addr[ETH_ALEN];
-	struct proc_dir_entry *dent;    /* Holds the proc data */
-	unsigned long cnt_inc_headroom_on_tx; /* How many times did we have to grow the skb on TX. */
-	unsigned long cnt_encap_on_xmit;      /* How many times did we have to encapsulate the skb on TX. */
-};
-
-static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
-{
-	return netdev_priv(dev);
-}
-
 /* VLAN tx hw acceleration helpers. */
 struct vlan_skb_tx_cookie {
 	u32	magic;
@@ -163,9 +122,24 @@ struct vlan_skb_tx_cookie {
 #define vlan_tx_tag_get(__skb)	(VLAN_TX_SKB_CB(__skb)->vlan_tag)
 
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);
+extern u16 vlan_dev_vlan_id(const struct net_device *dev);
+
 extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 			     unsigned short vlan_tag, int polling);
 #else
+static inline struct net_device *vlan_dev_real_dev(const struct net_device *dev)
+{
+	BUG();
+	return NULL;
+}
+
+static inline u16 vlan_dev_vlan_id(const struct net_device *dev)
+{
+	BUG();
+	return 0;
+}
+
 static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 				    unsigned short vlan_tag, int polling)
 {
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 8141e2dc510b..7a2625d2f9a0 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -543,7 +543,6 @@ static struct notifier_block vlan_notifier_block __read_mostly = {
 static int vlan_ioctl_handler(struct net *net, void __user *arg)
 {
 	int err;
-	unsigned short vid = 0;
 	struct vlan_ioctl_args args;
 	struct net_device *dev = NULL;
 
@@ -644,8 +643,7 @@ static int vlan_ioctl_handler(struct net *net, void __user *arg)
 
 	case GET_VLAN_VID_CMD:
 		err = 0;
-		vlan_dev_get_vid(dev, &vid);
-		args.u.VID = vid;
+		args.u.VID = vlan_dev_vlan_id(dev);
 		if (copy_to_user(arg, &args,
 				 sizeof(struct vlan_ioctl_args)))
 		      err = -EFAULT;
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 7cc1a97c42fc..14c421e033f2 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -3,6 +3,55 @@
 
 #include <linux/if_vlan.h>
 
+
+/**
+ *	struct vlan_priority_tci_mapping - vlan egress priority mappings
+ *	@priority: skb priority
+ *	@vlan_qos: vlan priority: (skb->priority << 13) & 0xE000
+ *	@next: pointer to next struct
+ */
+struct vlan_priority_tci_mapping {
+	u32					priority;
+	unsigned short				vlan_qos;
+	struct vlan_priority_tci_mapping	*next;
+};
+
+/**
+ *	struct vlan_dev_info - VLAN private device data
+ *	@nr_ingress_mappings: number of ingress priority mappings
+ *	@ingress_priority_map: ingress priority mappings
+ *	@nr_egress_mappings: number of egress priority mappings
+ *	@egress_priority_map: hash of egress priority mappings
+ *	@vlan_id: VLAN identifier
+ *	@flags: device flags
+ *	@real_dev: underlying netdevice
+ *	@real_dev_addr: address of underlying netdevice
+ *	@dent: proc dir entry
+ *	@cnt_inc_headroom_on_tx: statistic - number of skb expansions on TX
+ *	@cnt_encap_on_xmit: statistic - number of skb encapsulations on TX
+ */
+struct vlan_dev_info {
+	unsigned int				nr_ingress_mappings;
+	u32					ingress_priority_map[8];
+	unsigned int				nr_egress_mappings;
+	struct vlan_priority_tci_mapping	*egress_priority_map[16];
+
+	unsigned short				vlan_id;
+	unsigned short				flags;
+
+	struct net_device			*real_dev;
+	unsigned char				real_dev_addr[ETH_ALEN];
+
+	struct proc_dir_entry			*dent;
+	unsigned long				cnt_inc_headroom_on_tx;
+	unsigned long				cnt_encap_on_xmit;
+};
+
+static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
+{
+	return netdev_priv(dev);
+}
+
 #define VLAN_GRP_HASH_SHIFT	5
 #define VLAN_GRP_HASH_SIZE	(1 << VLAN_GRP_HASH_SHIFT)
 #define VLAN_GRP_HASH_MASK	(VLAN_GRP_HASH_SIZE - 1)
@@ -30,7 +79,6 @@ int vlan_dev_set_egress_priority(const struct net_device *dev,
 				 u32 skb_prio, short vlan_prio);
 int vlan_dev_change_flags(const struct net_device *dev, u32 flag, u32 mask);
 void vlan_dev_get_realdev_name(const struct net_device *dev, char *result);
-void vlan_dev_get_vid(const struct net_device *dev, unsigned short *result);
 
 int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id);
 void vlan_setup(struct net_device *dev);
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 85c94edb000f..f980b9154cc3 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -46,3 +46,15 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
 }
 EXPORT_SYMBOL(__vlan_hwaccel_rx);
+
+struct net_device *vlan_dev_real_dev(const struct net_device *dev)
+{
+	return vlan_dev_info(dev)->real_dev;
+}
+EXPORT_SYMBOL_GPL(vlan_dev_real_dev);
+
+u16 vlan_dev_vlan_id(const struct net_device *dev)
+{
+	return vlan_dev_info(dev)->vlan_id;
+}
+EXPORT_SYMBOL_GPL(vlan_dev_vlan_id);
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 722697d31e4f..2aab294c5744 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -531,11 +531,6 @@ void vlan_dev_get_realdev_name(const struct net_device *dev, char *result)
 	strncpy(result, vlan_dev_info(dev)->real_dev->name, 23);
 }
 
-void vlan_dev_get_vid(const struct net_device *dev, unsigned short *result)
-{
-	*result = vlan_dev_info(dev)->vlan_id;
-}
-
 static int vlan_dev_open(struct net_device *dev)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
-- 
cgit v1.2.3


From df6b6a0cf62afeacdeb4c1a35b8fba21fda54399 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:24:14 -0700
Subject: vlan: remove useless struct hlist_node declaration from if_vlan.h

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index cb2e6b480882..8e68b05b13df 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -14,10 +14,6 @@
 #define _LINUX_IF_VLAN_H_
 
 #ifdef __KERNEL__
-
-/* externally defined structs */
-struct hlist_node;
-
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 
-- 
cgit v1.2.3


From 9bb8582efb555521c7eec595ebd34e835ddc34b8 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 03:24:44 -0700
Subject: vlan: TCI related type and naming cleanups

The VLAN code contains multiple spots that use tag, id and tci as
identifiers for arguments and variables incorrectly and they actually
contain or are expected to contain something different. Additionally
types are used inconsistently (unsigned short vs u16) and identifiers
are sometimes capitalized.

- consistently use u16 for storing TCI, ID or QoS values
- consistently use vlan_id and vlan_tci for storing the respective values
- remove capitalization
- add kdoc comment to netif_hwaccel_{rx,receive_skb}

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 84 ++++++++++++++++++++++++++++---------------------
 net/8021q/vlan.c        | 34 ++++++++++----------
 net/8021q/vlan.h        | 19 ++++++-----
 net/8021q/vlan_core.c   |  6 ++--
 net/8021q/vlan_dev.c    | 47 ++++++++++++++-------------
 net/8021q/vlan_gvrp.c   |  8 ++---
 6 files changed, 103 insertions(+), 95 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 8e68b05b13df..d36515dae62f 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -87,7 +87,7 @@ struct vlan_group {
 };
 
 static inline struct net_device *vlan_group_get_device(struct vlan_group *vg,
-						       unsigned int vlan_id)
+						       u16 vlan_id)
 {
 	struct net_device **array;
 	array = vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN];
@@ -95,7 +95,7 @@ static inline struct net_device *vlan_group_get_device(struct vlan_group *vg,
 }
 
 static inline void vlan_group_set_device(struct vlan_group *vg,
-					 unsigned int vlan_id,
+					 u16 vlan_id,
 					 struct net_device *dev)
 {
 	struct net_device **array;
@@ -122,7 +122,7 @@ extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);
 extern u16 vlan_dev_vlan_id(const struct net_device *dev);
 
 extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
-			     unsigned short vlan_tag, int polling);
+			     u16 vlan_tci, int polling);
 #else
 static inline struct net_device *vlan_dev_real_dev(const struct net_device *dev)
 {
@@ -137,39 +137,51 @@ static inline u16 vlan_dev_vlan_id(const struct net_device *dev)
 }
 
 static inline int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
-				    unsigned short vlan_tag, int polling)
+				    u16 vlan_tci, int polling)
 {
 	BUG();
 	return NET_XMIT_SUCCESS;
 }
 #endif
 
+/**
+ * vlan_hwaccel_rx - netif_rx wrapper for VLAN RX acceleration
+ * @skb: buffer
+ * @grp: vlan group
+ * @vlan_tci: VLAN TCI as received from the card
+ */
 static inline int vlan_hwaccel_rx(struct sk_buff *skb,
 				  struct vlan_group *grp,
-				  unsigned short vlan_tag)
+				  u16 vlan_tci)
 {
-	return __vlan_hwaccel_rx(skb, grp, vlan_tag, 0);
+	return __vlan_hwaccel_rx(skb, grp, vlan_tci, 0);
 }
 
+/**
+ * vlan_hwaccel_receive_skb - netif_receive_skb wrapper for VLAN RX acceleration
+ * @skb: buffer
+ * @grp: vlan group
+ * @vlan_tci: VLAN TCI as received from the card
+ */
 static inline int vlan_hwaccel_receive_skb(struct sk_buff *skb,
 					   struct vlan_group *grp,
-					   unsigned short vlan_tag)
+					   u16 vlan_tci)
 {
-	return __vlan_hwaccel_rx(skb, grp, vlan_tag, 1);
+	return __vlan_hwaccel_rx(skb, grp, vlan_tci, 1);
 }
 
 /**
  * __vlan_put_tag - regular VLAN tag inserting
  * @skb: skbuff to tag
- * @tag: VLAN tag to insert
+ * @vlan_tci: VLAN TCI to insert
  *
  * Inserts the VLAN tag into @skb as part of the payload
  * Returns a VLAN tagged skb. If a new skb is created, @skb is freed.
- * 
+ *
  * Following the skb_unshare() example, in case of error, the calling function
  * doesn't have to worry about freeing the original skb.
  */
-static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short tag)
+static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
 {
 	struct vlan_ethhdr *veth;
 
@@ -197,8 +209,8 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short
 	/* first, the ethernet type */
 	veth->h_vlan_proto = htons(ETH_P_8021Q);
 
-	/* now, the tag */
-	veth->h_vlan_TCI = htons(tag);
+	/* now, the TCI */
+	veth->h_vlan_TCI = htons(vlan_tci);
 
 	skb->protocol = htons(ETH_P_8021Q);
 
@@ -208,17 +220,18 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, unsigned short
 /**
  * __vlan_hwaccel_put_tag - hardware accelerated VLAN inserting
  * @skb: skbuff to tag
- * @tag: VLAN tag to insert
+ * @vlan_tci: VLAN TCI to insert
  *
- * Puts the VLAN tag in @skb->cb[] and lets the device do the rest
+ * Puts the VLAN TCI in @skb->cb[] and lets the device do the rest
  */
-static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb, unsigned short tag)
+static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb,
+						     u16 vlan_tci)
 {
 	struct vlan_skb_tx_cookie *cookie;
 
 	cookie = VLAN_TX_SKB_CB(skb);
 	cookie->magic = VLAN_TX_COOKIE_MAGIC;
-	cookie->vlan_tag = tag;
+	cookie->vlan_tag = vlan_tci;
 
 	return skb;
 }
@@ -228,28 +241,28 @@ static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb, unsign
 /**
  * vlan_put_tag - inserts VLAN tag according to device features
  * @skb: skbuff to tag
- * @tag: VLAN tag to insert
+ * @vlan_tci: VLAN TCI to insert
  *
  * Assumes skb->dev is the target that will xmit this frame.
  * Returns a VLAN tagged skb.
  */
-static inline struct sk_buff *vlan_put_tag(struct sk_buff *skb, unsigned short tag)
+static inline struct sk_buff *vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
 {
 	if (skb->dev->features & NETIF_F_HW_VLAN_TX) {
-		return __vlan_hwaccel_put_tag(skb, tag);
+		return __vlan_hwaccel_put_tag(skb, vlan_tci);
 	} else {
-		return __vlan_put_tag(skb, tag);
+		return __vlan_put_tag(skb, vlan_tci);
 	}
 }
 
 /**
  * __vlan_get_tag - get the VLAN ID that is part of the payload
  * @skb: skbuff to query
- * @tag: buffer to store vlaue
- * 
+ * @vlan_tci: buffer to store vlaue
+ *
  * Returns error if the skb is not of VLAN type
  */
-static inline int __vlan_get_tag(const struct sk_buff *skb, unsigned short *tag)
+static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
 {
 	struct vlan_ethhdr *veth = (struct vlan_ethhdr *)skb->data;
 
@@ -257,29 +270,28 @@ static inline int __vlan_get_tag(const struct sk_buff *skb, unsigned short *tag)
 		return -EINVAL;
 	}
 
-	*tag = ntohs(veth->h_vlan_TCI);
-
+	*vlan_tci = ntohs(veth->h_vlan_TCI);
 	return 0;
 }
 
 /**
  * __vlan_hwaccel_get_tag - get the VLAN ID that is in @skb->cb[]
  * @skb: skbuff to query
- * @tag: buffer to store vlaue
- * 
+ * @vlan_tci: buffer to store vlaue
+ *
  * Returns error if @skb->cb[] is not set correctly
  */
 static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
-					 unsigned short *tag)
+					 u16 *vlan_tci)
 {
 	struct vlan_skb_tx_cookie *cookie;
 
 	cookie = VLAN_TX_SKB_CB(skb);
 	if (cookie->magic == VLAN_TX_COOKIE_MAGIC) {
-		*tag = cookie->vlan_tag;
+		*vlan_tci = cookie->vlan_tag;
 		return 0;
 	} else {
-		*tag = 0;
+		*vlan_tci = 0;
 		return -EINVAL;
 	}
 }
@@ -289,16 +301,16 @@ static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
 /**
  * vlan_get_tag - get the VLAN ID from the skb
  * @skb: skbuff to query
- * @tag: buffer to store vlaue
- * 
+ * @vlan_tci: buffer to store vlaue
+ *
  * Returns error if the skb is not VLAN tagged
  */
-static inline int vlan_get_tag(const struct sk_buff *skb, unsigned short *tag)
+static inline int vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
 {
 	if (skb->dev->features & NETIF_F_HW_VLAN_TX) {
-		return __vlan_hwaccel_get_tag(skb, tag);
+		return __vlan_hwaccel_get_tag(skb, vlan_tci);
 	} else {
-		return __vlan_get_tag(skb, tag);
+		return __vlan_get_tag(skb, vlan_tci);
 	}
 }
 
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 7a2625d2f9a0..68bdcf4a795c 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -83,13 +83,12 @@ static struct vlan_group *__vlan_find_group(struct net_device *real_dev)
  *
  * Must be invoked with RCU read lock (no preempt)
  */
-struct net_device *__find_vlan_dev(struct net_device *real_dev,
-				   unsigned short VID)
+struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id)
 {
 	struct vlan_group *grp = __vlan_find_group(real_dev);
 
 	if (grp)
-		return vlan_group_get_device(grp, VID);
+		return vlan_group_get_device(grp, vlan_id);
 
 	return NULL;
 }
@@ -117,14 +116,14 @@ static struct vlan_group *vlan_group_alloc(struct net_device *real_dev)
 	return grp;
 }
 
-static int vlan_group_prealloc_vid(struct vlan_group *vg, int vid)
+static int vlan_group_prealloc_vid(struct vlan_group *vg, u16 vlan_id)
 {
 	struct net_device **array;
 	unsigned int size;
 
 	ASSERT_RTNL();
 
-	array = vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN];
+	array = vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN];
 	if (array != NULL)
 		return 0;
 
@@ -133,7 +132,7 @@ static int vlan_group_prealloc_vid(struct vlan_group *vg, int vid)
 	if (array == NULL)
 		return -ENOBUFS;
 
-	vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN] = array;
+	vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN] = array;
 	return 0;
 }
 
@@ -147,7 +146,7 @@ void unregister_vlan_dev(struct net_device *dev)
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev = vlan->real_dev;
 	struct vlan_group *grp;
-	unsigned short vlan_id = vlan->vlan_id;
+	u16 vlan_id = vlan->vlan_id;
 
 	ASSERT_RTNL();
 
@@ -205,7 +204,7 @@ static void vlan_transfer_operstate(const struct net_device *dev,
 	}
 }
 
-int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id)
+int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id)
 {
 	char *name = real_dev->name;
 
@@ -242,7 +241,7 @@ int register_vlan_dev(struct net_device *dev)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct net_device *real_dev = vlan->real_dev;
-	unsigned short vlan_id = vlan->vlan_id;
+	u16 vlan_id = vlan->vlan_id;
 	struct vlan_group *grp, *ngrp = NULL;
 	int err;
 
@@ -295,8 +294,7 @@ out_free_group:
 /*  Attach a VLAN device to a mac address (ie Ethernet Card).
  *  Returns 0 if the device was created or a negative error code otherwise.
  */
-static int register_vlan_device(struct net_device *real_dev,
-				unsigned short VLAN_ID)
+static int register_vlan_device(struct net_device *real_dev, u16 vlan_id)
 {
 	struct net_device *new_dev;
 	struct net *net = dev_net(real_dev);
@@ -304,10 +302,10 @@ static int register_vlan_device(struct net_device *real_dev,
 	char name[IFNAMSIZ];
 	int err;
 
-	if (VLAN_ID >= VLAN_VID_MASK)
+	if (vlan_id >= VLAN_VID_MASK)
 		return -ERANGE;
 
-	err = vlan_check_real_dev(real_dev, VLAN_ID);
+	err = vlan_check_real_dev(real_dev, vlan_id);
 	if (err < 0)
 		return err;
 
@@ -315,26 +313,26 @@ static int register_vlan_device(struct net_device *real_dev,
 	switch (vn->name_type) {
 	case VLAN_NAME_TYPE_RAW_PLUS_VID:
 		/* name will look like:	 eth1.0005 */
-		snprintf(name, IFNAMSIZ, "%s.%.4i", real_dev->name, VLAN_ID);
+		snprintf(name, IFNAMSIZ, "%s.%.4i", real_dev->name, vlan_id);
 		break;
 	case VLAN_NAME_TYPE_PLUS_VID_NO_PAD:
 		/* Put our vlan.VID in the name.
 		 * Name will look like:	 vlan5
 		 */
-		snprintf(name, IFNAMSIZ, "vlan%i", VLAN_ID);
+		snprintf(name, IFNAMSIZ, "vlan%i", vlan_id);
 		break;
 	case VLAN_NAME_TYPE_RAW_PLUS_VID_NO_PAD:
 		/* Put our vlan.VID in the name.
 		 * Name will look like:	 eth0.5
 		 */
-		snprintf(name, IFNAMSIZ, "%s.%i", real_dev->name, VLAN_ID);
+		snprintf(name, IFNAMSIZ, "%s.%i", real_dev->name, vlan_id);
 		break;
 	case VLAN_NAME_TYPE_PLUS_VID:
 		/* Put our vlan.VID in the name.
 		 * Name will look like:	 vlan0005
 		 */
 	default:
-		snprintf(name, IFNAMSIZ, "vlan%.4i", VLAN_ID);
+		snprintf(name, IFNAMSIZ, "vlan%.4i", vlan_id);
 	}
 
 	new_dev = alloc_netdev(sizeof(struct vlan_dev_info), name,
@@ -349,7 +347,7 @@ static int register_vlan_device(struct net_device *real_dev,
 	 */
 	new_dev->mtu = real_dev->mtu;
 
-	vlan_dev_info(new_dev)->vlan_id = VLAN_ID; /* 1 through VLAN_VID_MASK */
+	vlan_dev_info(new_dev)->vlan_id = vlan_id;
 	vlan_dev_info(new_dev)->real_dev = real_dev;
 	vlan_dev_info(new_dev)->dent = NULL;
 	vlan_dev_info(new_dev)->flags = VLAN_FLAG_REORDER_HDR;
diff --git a/net/8021q/vlan.h b/net/8021q/vlan.h
index 14c421e033f2..a6603a4d917f 100644
--- a/net/8021q/vlan.h
+++ b/net/8021q/vlan.h
@@ -12,7 +12,7 @@
  */
 struct vlan_priority_tci_mapping {
 	u32					priority;
-	unsigned short				vlan_qos;
+	u16					vlan_qos;
 	struct vlan_priority_tci_mapping	*next;
 };
 
@@ -36,8 +36,8 @@ struct vlan_dev_info {
 	unsigned int				nr_egress_mappings;
 	struct vlan_priority_tci_mapping	*egress_priority_map[16];
 
-	unsigned short				vlan_id;
-	unsigned short				flags;
+	u16					vlan_id;
+	u16					flags;
 
 	struct net_device			*real_dev;
 	unsigned char				real_dev_addr[ETH_ALEN];
@@ -67,30 +67,29 @@ static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
  *  Must be invoked with rcu_read_lock (ie preempt disabled)
  *  or with RTNL.
  */
-struct net_device *__find_vlan_dev(struct net_device *real_dev,
-				   unsigned short VID); /* vlan.c */
+struct net_device *__find_vlan_dev(struct net_device *real_dev, u16 vlan_id);
 
 /* found in vlan_dev.c */
 int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		  struct packet_type *ptype, struct net_device *orig_dev);
 void vlan_dev_set_ingress_priority(const struct net_device *dev,
-				   u32 skb_prio, short vlan_prio);
+				   u32 skb_prio, u16 vlan_prio);
 int vlan_dev_set_egress_priority(const struct net_device *dev,
-				 u32 skb_prio, short vlan_prio);
+				 u32 skb_prio, u16 vlan_prio);
 int vlan_dev_change_flags(const struct net_device *dev, u32 flag, u32 mask);
 void vlan_dev_get_realdev_name(const struct net_device *dev, char *result);
 
-int vlan_check_real_dev(struct net_device *real_dev, unsigned short vlan_id);
+int vlan_check_real_dev(struct net_device *real_dev, u16 vlan_id);
 void vlan_setup(struct net_device *dev);
 int register_vlan_dev(struct net_device *dev);
 void unregister_vlan_dev(struct net_device *dev);
 
 static inline u32 vlan_get_ingress_priority(struct net_device *dev,
-					    unsigned short vlan_tag)
+					    u16 vlan_tci)
 {
 	struct vlan_dev_info *vip = vlan_dev_info(dev);
 
-	return vip->ingress_priority_map[(vlan_tag >> 13) & 0x7];
+	return vip->ingress_priority_map[(vlan_tci >> 13) & 0x7];
 }
 
 #ifdef CONFIG_VLAN_8021Q_GVRP
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index f980b9154cc3..68df12d3664b 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -5,7 +5,7 @@
 
 /* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
 int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
-		      unsigned short vlan_tag, int polling)
+		      u16 vlan_tci, int polling)
 {
 	struct net_device_stats *stats;
 
@@ -14,7 +14,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_DROP;
 	}
 
-	skb->dev = vlan_group_get_device(grp, vlan_tag & VLAN_VID_MASK);
+	skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
 	if (skb->dev == NULL) {
 		dev_kfree_skb_any(skb);
 		/* Not NET_RX_DROP, this is not being dropped
@@ -27,7 +27,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 	stats->rx_packets++;
 	stats->rx_bytes += skb->len;
 
-	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tag);
+	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tci);
 	switch (skb->pkt_type) {
 	case PACKET_BROADCAST:
 		break;
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 2aab294c5744..2ccac6bea57e 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -150,9 +150,9 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		  struct packet_type *ptype, struct net_device *orig_dev)
 {
 	struct vlan_hdr *vhdr;
-	unsigned short vid;
 	struct net_device_stats *stats;
-	unsigned short vlan_TCI;
+	u16 vlan_id;
+	u16 vlan_tci;
 
 	skb = skb_share_check(skb, GFP_ATOMIC);
 	if (skb == NULL)
@@ -162,14 +162,14 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 		goto err_free;
 
 	vhdr = (struct vlan_hdr *)skb->data;
-	vlan_TCI = ntohs(vhdr->h_vlan_TCI);
-	vid = (vlan_TCI & VLAN_VID_MASK);
+	vlan_tci = ntohs(vhdr->h_vlan_TCI);
+	vlan_id = vlan_tci & VLAN_VID_MASK;
 
 	rcu_read_lock();
-	skb->dev = __find_vlan_dev(dev, vid);
+	skb->dev = __find_vlan_dev(dev, vlan_id);
 	if (!skb->dev) {
 		pr_debug("%s: ERROR: No net_device for VID: %u on dev: %s\n",
-			 __func__, (unsigned int)vid, dev->name);
+			 __func__, vlan_id, dev->name);
 		goto err_unlock;
 	}
 
@@ -181,11 +181,10 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev,
 
 	skb_pull_rcsum(skb, VLAN_HLEN);
 
-	skb->priority = vlan_get_ingress_priority(skb->dev,
-						  ntohs(vhdr->h_vlan_TCI));
+	skb->priority = vlan_get_ingress_priority(skb->dev, vlan_tci);
 
 	pr_debug("%s: priority: %u for TCI: %hu\n",
-		 __func__, skb->priority, ntohs(vhdr->h_vlan_TCI));
+		 __func__, skb->priority, vlan_tci);
 
 	switch (skb->pkt_type) {
 	case PACKET_BROADCAST: /* Yeah, stats collect these together.. */
@@ -228,7 +227,7 @@ err_free:
 	return NET_RX_DROP;
 }
 
-static inline unsigned short
+static inline u16
 vlan_dev_get_egress_qos_mask(struct net_device *dev, struct sk_buff *skb)
 {
 	struct vlan_priority_tci_mapping *mp;
@@ -260,7 +259,7 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 				unsigned int len)
 {
 	struct vlan_hdr *vhdr;
-	unsigned short veth_TCI = 0;
+	u16 vlan_tci = 0;
 	int rc = 0;
 	int build_vlan_header = 0;
 	struct net_device *vdev = dev;
@@ -292,10 +291,10 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 		 * VLAN ID	 12 bits (low bits)
 		 *
 		 */
-		veth_TCI = vlan_dev_info(dev)->vlan_id;
-		veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
+		vlan_tci = vlan_dev_info(dev)->vlan_id;
+		vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
 
-		vhdr->h_vlan_TCI = htons(veth_TCI);
+		vhdr->h_vlan_TCI = htons(vlan_tci);
 
 		/*
 		 *  Set the protocol type. For a packet of type ETH_P_802_3 we
@@ -373,7 +372,7 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (veth->h_vlan_proto != htons(ETH_P_8021Q) ||
 		vlan_dev_info(dev)->flags & VLAN_FLAG_REORDER_HDR) {
 		int orig_headroom = skb_headroom(skb);
-		unsigned short veth_TCI;
+		u16 vlan_tci;
 
 		/* This is not a VLAN frame...but we can fix that! */
 		vlan_dev_info(dev)->cnt_encap_on_xmit++;
@@ -386,10 +385,10 @@ static int vlan_dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * CFI		 1 bit
 		 * VLAN ID	 12 bits (low bits)
 		 */
-		veth_TCI = vlan_dev_info(dev)->vlan_id;
-		veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
+		vlan_tci = vlan_dev_info(dev)->vlan_id;
+		vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
 
-		skb = __vlan_put_tag(skb, veth_TCI);
+		skb = __vlan_put_tag(skb, vlan_tci);
 		if (!skb) {
 			stats->tx_dropped++;
 			return 0;
@@ -422,7 +421,7 @@ static int vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb,
 					    struct net_device *dev)
 {
 	struct net_device_stats *stats = &dev->stats;
-	unsigned short veth_TCI;
+	u16 vlan_tci;
 
 	/* Construct the second two bytes. This field looks something
 	 * like:
@@ -430,9 +429,9 @@ static int vlan_dev_hwaccel_hard_start_xmit(struct sk_buff *skb,
 	 * CFI		 1 bit
 	 * VLAN ID	 12 bits (low bits)
 	 */
-	veth_TCI = vlan_dev_info(dev)->vlan_id;
-	veth_TCI |= vlan_dev_get_egress_qos_mask(dev, skb);
-	skb = __vlan_hwaccel_put_tag(skb, veth_TCI);
+	vlan_tci = vlan_dev_info(dev)->vlan_id;
+	vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb);
+	skb = __vlan_hwaccel_put_tag(skb, vlan_tci);
 
 	stats->tx_packets++;
 	stats->tx_bytes += skb->len;
@@ -457,7 +456,7 @@ static int vlan_dev_change_mtu(struct net_device *dev, int new_mtu)
 }
 
 void vlan_dev_set_ingress_priority(const struct net_device *dev,
-				   u32 skb_prio, short vlan_prio)
+				   u32 skb_prio, u16 vlan_prio)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 
@@ -470,7 +469,7 @@ void vlan_dev_set_ingress_priority(const struct net_device *dev,
 }
 
 int vlan_dev_set_egress_priority(const struct net_device *dev,
-				 u32 skb_prio, short vlan_prio)
+				 u32 skb_prio, u16 vlan_prio)
 {
 	struct vlan_dev_info *vlan = vlan_dev_info(dev);
 	struct vlan_priority_tci_mapping *mp = NULL;
diff --git a/net/8021q/vlan_gvrp.c b/net/8021q/vlan_gvrp.c
index db9781608362..061ceceeef12 100644
--- a/net/8021q/vlan_gvrp.c
+++ b/net/8021q/vlan_gvrp.c
@@ -30,19 +30,19 @@ static struct garp_application vlan_gvrp_app __read_mostly = {
 int vlan_gvrp_request_join(const struct net_device *dev)
 {
 	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
-	__be16 vid = htons(vlan->vlan_id);
+	__be16 vlan_id = htons(vlan->vlan_id);
 
 	return garp_request_join(vlan->real_dev, &vlan_gvrp_app,
-				 &vid, sizeof(vid), GVRP_ATTR_VID);
+				 &vlan_id, sizeof(vlan_id), GVRP_ATTR_VID);
 }
 
 void vlan_gvrp_request_leave(const struct net_device *dev)
 {
 	const struct vlan_dev_info *vlan = vlan_dev_info(dev);
-	__be16 vid = htons(vlan->vlan_id);
+	__be16 vlan_id = htons(vlan->vlan_id);
 
 	garp_request_leave(vlan->real_dev, &vlan_gvrp_app,
-			   &vid, sizeof(vid), GVRP_ATTR_VID);
+			   &vlan_id, sizeof(vlan_id), GVRP_ATTR_VID);
 }
 
 int vlan_gvrp_init_applicant(struct net_device *dev)
-- 
cgit v1.2.3


From a7bf0bd5e6af7fe69342dabf2a3b721f0163469a Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 28 May 2008 15:02:14 +0100
Subject: build: add __page_aligned_data and __page_aligned_bss

Making a variable page-aligned by using
__attribute__((section(".data.page_aligned"))) is fragile because if
sizeof(variable) is not also a multiple of page size, it leaves
variables in the remainder of the section unaligned.

This patch introduces two new qualifiers, __page_aligned_data and
__page_aligned_bss to set the section *and* the alignment of
variables.  This makes page-aligned variables more robust because the
linker will make sure they're aligned properly.  Unfortunately it
requires *all* page-aligned data to use these macros...

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup64.c | 2 +-
 arch/x86/mm/ioremap.c     | 3 +--
 include/linux/linkage.h   | 4 ++++
 3 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/setup64.c b/arch/x86/kernel/setup64.c
index 631ea6cc01d8..fc1a56da8240 100644
--- a/arch/x86/kernel/setup64.c
+++ b/arch/x86/kernel/setup64.c
@@ -40,7 +40,7 @@ EXPORT_SYMBOL(_cpu_pda);
 
 struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
 
-char boot_cpu_stack[IRQSTACKSIZE] __attribute__((section(".bss.page_aligned")));
+char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
 
 unsigned long __supported_pte_mask __read_mostly = ~0UL;
 EXPORT_SYMBOL_GPL(__supported_pte_mask);
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 416ea415f5c2..0561fde56a51 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -394,8 +394,7 @@ static int __init early_ioremap_debug_setup(char *str)
 early_param("early_ioremap_debug", early_ioremap_debug_setup);
 
 static __initdata int after_paging_init;
-static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)]
-		__section(.bss.page_aligned);
+static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __page_aligned_bss;
 
 static inline pmd_t * __init early_ioremap_pmd(unsigned long addr)
 {
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index 2119610b24f8..9fd1f859021b 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_LINKAGE_H
 #define _LINUX_LINKAGE_H
 
+#include <linux/compiler.h>
 #include <asm/linkage.h>
 
 #ifdef __cplusplus
@@ -17,6 +18,9 @@
 # define asmregparm
 #endif
 
+#define __page_aligned_data	__section(.data.page_aligned) __aligned(PAGE_SIZE)
+#define __page_aligned_bss	__section(.bss.page_aligned) __aligned(PAGE_SIZE)
+
 /*
  * This is used by architectures to keep arguments on the stack
  * untouched by the compiler by keeping them live until the end.
-- 
cgit v1.2.3


From d52d53b8a5b258bfaab9223a5e7284fcfdd48577 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Mon, 16 Jun 2008 20:10:55 -0700
Subject: RFC x86: try to remove arch_get_ram_range

want to remove arch_get_ram_range, and use early_node_map instead.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/init_32.c     |  6 ++++--
 drivers/pci/intel-iommu.c | 51 ++++++++++++++++++++++++++++++++++-------------
 include/linux/mm.h        |  2 +-
 mm/page_alloc.c           | 10 +++++++---
 4 files changed, 49 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 65d55056b6e7..a0484adbf59d 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -298,7 +298,7 @@ struct add_highpages_data {
 	unsigned long end_pfn;
 };
 
-static void __init add_highpages_work_fn(unsigned long start_pfn,
+static int __init add_highpages_work_fn(unsigned long start_pfn,
 					 unsigned long end_pfn, void *datax)
 {
 	int node_pfn;
@@ -311,7 +311,7 @@ static void __init add_highpages_work_fn(unsigned long start_pfn,
 	final_start_pfn = max(start_pfn, data->start_pfn);
 	final_end_pfn = min(end_pfn, data->end_pfn);
 	if (final_start_pfn >= final_end_pfn)
-		return;
+		return 0;
 
 	for (node_pfn = final_start_pfn; node_pfn < final_end_pfn;
 	     node_pfn++) {
@@ -321,6 +321,8 @@ static void __init add_highpages_work_fn(unsigned long start_pfn,
 		add_one_highpage_init(page, node_pfn);
 	}
 
+	return 0;
+
 }
 
 void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn,
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 66c0fd21894b..bb0642318a95 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1637,12 +1637,43 @@ static inline int iommu_prepare_rmrr_dev(struct dmar_rmrr_unit *rmrr,
 }
 
 #ifdef CONFIG_DMAR_GFX_WA
-extern int arch_get_ram_range(int slot, u64 *addr, u64 *size);
+struct iommu_prepare_data {
+	struct pci_dev *pdev;
+	int ret;
+};
+
+static int __init iommu_prepare_work_fn(unsigned long start_pfn,
+					 unsigned long end_pfn, void *datax)
+{
+	struct iommu_prepare_data *data;
+
+	data = (struct iommu_prepare_data *)datax;
+
+	data->ret = iommu_prepare_identity_map(data->pdev,
+				start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);
+	return data->ret;
+
+}
+
+static int __init iommu_prepare_with_active_regions(struct pci_dev *pdev)
+{
+	int nid;
+	struct iommu_prepare_data data;
+
+	data.pdev = pdev;
+	data.ret = 0;
+
+	for_each_online_node(nid) {
+		work_with_active_regions(nid, iommu_prepare_work_fn, &data);
+		if (data.ret)
+			return data.ret;
+	}
+	return data.ret;
+}
+
 static void __init iommu_prepare_gfx_mapping(void)
 {
 	struct pci_dev *pdev = NULL;
-	u64 base, size;
-	int slot;
 	int ret;
 
 	for_each_pci_dev(pdev) {
@@ -1651,17 +1682,9 @@ static void __init iommu_prepare_gfx_mapping(void)
 			continue;
 		printk(KERN_INFO "IOMMU: gfx device %s 1-1 mapping\n",
 			pci_name(pdev));
-		slot = arch_get_ram_range(0, &base, &size);
-		while (slot >= 0) {
-			ret = iommu_prepare_identity_map(pdev,
-					base, base + size);
-			if (ret)
-				goto error;
-			slot = arch_get_ram_range(slot, &base, &size);
-		}
-		continue;
-error:
-		printk(KERN_ERR "IOMMU: mapping reserved region failed\n");
+		ret = iommu_prepare_with_active_regions(pdev);
+		if (ret)
+			printk(KERN_ERR "IOMMU: mapping reserved region failed\n");
 	}
 }
 #endif
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 3d647b24041f..cf1cd3a2ed78 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1011,7 +1011,7 @@ extern unsigned long find_min_pfn_with_active_regions(void);
 extern unsigned long find_max_pfn_with_active_regions(void);
 extern void free_bootmem_with_active_regions(int nid,
 						unsigned long max_low_pfn);
-typedef void (*work_fn_t)(unsigned long, unsigned long, void *);
+typedef int (*work_fn_t)(unsigned long, unsigned long, void *);
 extern void work_with_active_regions(int nid, work_fn_t work_fn, void *data);
 extern void sparse_memory_present_with_active_regions(int nid);
 #ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 41c6e3aa059f..e25b6b24f844 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2932,10 +2932,14 @@ void __init free_bootmem_with_active_regions(int nid,
 void __init work_with_active_regions(int nid, work_fn_t work_fn, void *data)
 {
 	int i;
+	int ret;
 
-	for_each_active_range_index_in_nid(i, nid)
-		work_fn(early_node_map[i].start_pfn, early_node_map[i].end_pfn,
-			data);
+	for_each_active_range_index_in_nid(i, nid) {
+		ret = work_fn(early_node_map[i].start_pfn,
+			      early_node_map[i].end_pfn, data);
+		if (ret)
+			break;
+	}
 }
 /**
  * sparse_memory_present_with_active_regions - Call memory_present for each active range
-- 
cgit v1.2.3


From 3c999f142665265afd0fe9190204dd051f17e505 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Fri, 20 Jun 2008 16:11:20 -0700
Subject: x86: check command line when CONFIG_X86_MPPARSE is not set, v2

if acpi=off, acpi=noirq and pci=noacpi, we need to disable apic.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Maciej W. Rozycki" <macro@linux-mips.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/boot.c | 14 ++++++++++++++
 arch/x86/kernel/apic_32.c   |  2 +-
 arch/x86/kernel/setup.c     |  4 ++++
 arch/x86/kernel/setup_32.c  |  5 +++++
 arch/x86/kernel/setup_64.c  |  9 +++++++++
 include/asm-x86/apic.h      |  6 +++++-
 include/linux/acpi.h        |  6 ++++++
 7 files changed, 44 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 6516359922ba..5c0107602b62 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -1787,6 +1787,20 @@ static int __init parse_pci(char *arg)
 }
 early_param("pci", parse_pci);
 
+int __init acpi_mps_check(void)
+{
+#if defined(CONFIG_X86_LOCAL_APIC) && !defined(CONFIG_X86_MPPARSE)
+/* mptable code is not built-in*/
+	if (acpi_disabled || acpi_noirq) {
+		printk(KERN_WARNING "MPS support code is not built-in.\n"
+		       "Using acpi=off or acpi=noirq or pci=noacpi "
+		       "may have problem\n");
+		return 1;
+	}
+#endif
+	return 0;
+}
+
 #ifdef CONFIG_X86_IO_APIC
 static int __init parse_acpi_skip_timer_override(char *arg)
 {
diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index dd8de26b2786..4932d7813bcd 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -57,7 +57,7 @@ unsigned long mp_lapic_addr;
  *
  * -1=force-disable, +1=force-enable
  */
-static int enable_local_apic __initdata;
+int enable_local_apic;
 
 /* Local APIC timer verification ok */
 static int local_apic_timer_verify_ok;
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index c5330f601b68..56aee55cf8dc 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -161,6 +161,10 @@ void __init setup_per_cpu_areas(void)
 	char *ptr;
 	int cpu;
 
+	/* no processor from mptable or madt */
+	if (!num_processors)
+		num_processors = 1;
+
 #ifdef CONFIG_HOTPLUG_CPU
 	prefill_possible_map();
 #else
diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c
index 369d0fe1ff9c..cad4e893df05 100644
--- a/arch/x86/kernel/setup_32.c
+++ b/arch/x86/kernel/setup_32.c
@@ -677,6 +677,11 @@ void __init setup_arch(char **cmdline_p)
 
 	parse_early_param();
 
+	if (acpi_mps_check()){
+		enable_local_apic = -1;
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
+	}
+
 	finish_e820_parsing();
 
 	probe_roms();
diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c
index a93300de4da9..175c696ec536 100644
--- a/arch/x86/kernel/setup_64.c
+++ b/arch/x86/kernel/setup_64.c
@@ -302,6 +302,11 @@ void __init setup_arch(char **cmdline_p)
 
 	parse_early_param();
 
+	if (acpi_mps_check()) {
+		disable_apic = 1;
+		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
+	}
+
 #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
 	if (init_ohci1394_dma_early)
 		init_ohci1394_dma_on_all_controllers();
@@ -723,6 +728,10 @@ static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
 		cpu_devs[c->x86_vendor]->c_early_init(c);
 
 	validate_pat_support(c);
+
+	/* early_param could clear that, but recall get it set again */
+	if (disable_apic)
+		clear_cpu_cap(c, X86_FEATURE_APIC);
 }
 
 /*
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 313bcaf4b6c3..9fe941cd843d 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -39,8 +39,12 @@ extern int apic_verbosity;
 extern int local_apic_timer_c2_ok;
 
 extern int ioapic_force;
-extern int disable_apic;
 
+#ifdef CONFIG_X86_64
+extern int disable_apic;
+#else
+extern int enable_local_apic;
+#endif
 /*
  * Basic functions accessing APICs.
  */
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 41f7ce7edd7a..0601075d09a1 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -82,6 +82,7 @@ char * __acpi_map_table (unsigned long phys_addr, unsigned long size);
 int early_acpi_boot_init(void);
 int acpi_boot_init (void);
 int acpi_boot_table_init (void);
+int acpi_mps_check (void);
 int acpi_numa_init (void);
 
 int acpi_table_init (void);
@@ -250,6 +251,11 @@ static inline int acpi_boot_table_init(void)
 	return 0;
 }
 
+static inline int acpi_mps_check(void)
+{
+	return 0;
+}
+
 static inline int acpi_check_resource_conflict(struct resource *res)
 {
 	return 0;
-- 
cgit v1.2.3


From 429a380571a6e6b8525b93161544eafc9b227e44 Mon Sep 17 00:00:00 2001
From: Ron Rindjunsky <ron.rindjunsky@intel.com>
Date: Tue, 1 Jul 2008 14:16:03 +0300
Subject: mac80211: add block ack request capability

This patch adds block ack request capability

Signed-off-by: Ester Kummer <ester.kummer@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Ron Rindjunsky <ron.rindjunsky@intel.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h  |  4 ++++
 include/net/mac80211.h     |  3 +++
 net/mac80211/ieee80211_i.h |  1 +
 net/mac80211/main.c        | 21 +++++++++++++++++++--
 net/mac80211/mlme.c        | 29 +++++++++++++++++++++++++++++
 5 files changed, 56 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index cffd6d0094f9..aa603c3d76d1 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -658,6 +658,10 @@ struct ieee80211_bar {
 	__le16 start_seq_num;
 } __attribute__((packed));
 
+/* 802.11 BAR control masks */
+#define IEEE80211_BAR_CTRL_ACK_POLICY_NORMAL     0x0000
+#define IEEE80211_BAR_CTRL_CBMTID_COMPRESSED_BA  0x0004
+
 /**
  * struct ieee80211_ht_cap - HT capabilities
  *
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 3a204acad901..0a5de3ef527a 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -235,6 +235,8 @@ struct ieee80211_bss_conf {
  * @IEEE80211_TX_STAT_ACK: Frame was acknowledged
  * @IEEE80211_TX_STAT_AMPDU: The frame was aggregated, so status
  * 	is for the whole aggregation.
+ * @IEEE80211_TX_STAT_AMPDU_NO_BACK: no block ack was returned,
+ * 	so consider using block ack request (BAR).
  */
 enum mac80211_tx_control_flags {
 	IEEE80211_TX_CTL_REQ_TX_STATUS		= BIT(0),
@@ -260,6 +262,7 @@ enum mac80211_tx_control_flags {
 	IEEE80211_TX_STAT_TX_FILTERED		= BIT(20),
 	IEEE80211_TX_STAT_ACK			= BIT(21),
 	IEEE80211_TX_STAT_AMPDU			= BIT(22),
+	IEEE80211_TX_STAT_AMPDU_NO_BACK		= BIT(23),
 };
 
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index f90da1bbec49..175cbdd36d7f 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -904,6 +904,7 @@ void ieee80211_send_addba_request(struct net_device *dev, const u8 *da,
 				  u16 agg_size, u16 timeout);
 void ieee80211_send_delba(struct net_device *dev, const u8 *da, u16 tid,
 				u16 initiator, u16 reason_code);
+void ieee80211_send_bar(struct net_device *dev, u8 *ra, u16 tid, u16 ssn);
 
 void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *da,
 				u16 tid, u16 initiator, u16 reason);
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index f18cfd727872..074f71a62a6d 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1404,14 +1404,15 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb)
 	struct ieee80211_local *local = hw_to_local(hw);
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	u16 frag, type;
+	__le16 fc;
 	struct ieee80211_tx_status_rtap_hdr *rthdr;
 	struct ieee80211_sub_if_data *sdata;
 	struct net_device *prev_dev = NULL;
+	struct sta_info *sta;
 
 	rcu_read_lock();
 
 	if (info->status.excessive_retries) {
-		struct sta_info *sta;
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
 			if (test_sta_flags(sta, WLAN_STA_PS)) {
@@ -1426,8 +1427,24 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb)
 		}
 	}
 
+	fc = hdr->frame_control;
+
+	if ((info->flags & IEEE80211_TX_STAT_AMPDU_NO_BACK) &&
+	    (ieee80211_is_data_qos(fc))) {
+		u16 tid, ssn;
+		u8 *qc;
+		sta = sta_info_get(local, hdr->addr1);
+		if (sta) {
+			qc = ieee80211_get_qos_ctl(hdr);
+			tid = qc[0] & 0xf;
+			ssn = ((le16_to_cpu(hdr->seq_ctrl) + 0x10)
+						& IEEE80211_SCTL_SEQ);
+			ieee80211_send_bar(sta->sdata->dev, hdr->addr1,
+					   tid, ssn);
+		}
+	}
+
 	if (info->flags & IEEE80211_TX_STAT_TX_FILTERED) {
-		struct sta_info *sta;
 		sta = sta_info_get(local, hdr->addr1);
 		if (sta) {
 			ieee80211_handle_filtered_frame(local, sta, skb);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 86abdf96390c..e080482b63cd 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -1536,6 +1536,35 @@ void ieee80211_send_delba(struct net_device *dev, const u8 *da, u16 tid,
 	ieee80211_sta_tx(dev, skb, 0);
 }
 
+void ieee80211_send_bar(struct net_device *dev, u8 *ra, u16 tid, u16 ssn)
+{
+	struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr);
+	struct sk_buff *skb;
+	struct ieee80211_bar *bar;
+	u16 bar_control = 0;
+
+	skb = dev_alloc_skb(sizeof(*bar) + local->hw.extra_tx_headroom);
+	if (!skb) {
+		printk(KERN_ERR "%s: failed to allocate buffer for "
+			"bar frame\n", dev->name);
+		return;
+	}
+	skb_reserve(skb, local->hw.extra_tx_headroom);
+	bar = (struct ieee80211_bar *)skb_put(skb, sizeof(*bar));
+	memset(bar, 0, sizeof(*bar));
+	bar->frame_control = IEEE80211_FC(IEEE80211_FTYPE_CTL,
+					IEEE80211_STYPE_BACK_REQ);
+	memcpy(bar->ra, ra, ETH_ALEN);
+	memcpy(bar->ta, dev->dev_addr, ETH_ALEN);
+	bar_control |= (u16)IEEE80211_BAR_CTRL_ACK_POLICY_NORMAL;
+	bar_control |= (u16)IEEE80211_BAR_CTRL_CBMTID_COMPRESSED_BA;
+	bar_control |= (u16)(tid << 12);
+	bar->control = cpu_to_le16(bar_control);
+	bar->start_seq_num = cpu_to_le16(ssn);
+
+	ieee80211_sta_tx(dev, skb, 0);
+}
+
 void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *ra, u16 tid,
 					u16 initiator, u16 reason)
 {
-- 
cgit v1.2.3


From f3d1eb19abdcb1e740d8ba0e06d606c1d4165438 Mon Sep 17 00:00:00 2001
From: Niels de Vos <niels.devos@wincor-nixdorf.com>
Date: Tue, 8 Jul 2008 10:30:27 -0400
Subject: Input: serio - trivial documentation fix

In include/linux/serio.h two different define-series are documented as
"Serio types". However the second series contains defines for the
different protocols.

Signed-off-by: Niels de Vos <niels.devos@wincor-nixdorf.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 include/linux/serio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/serio.h b/include/linux/serio.h
index 48defc4d181e..e72716cca577 100644
--- a/include/linux/serio.h
+++ b/include/linux/serio.h
@@ -175,7 +175,7 @@ static inline void serio_unpin_driver(struct serio *serio)
 #define SERIO_8042_XL	0x06
 
 /*
- * Serio types
+ * Serio protocols
  */
 #define SERIO_UNKNOWN	0x00
 #define SERIO_MSC	0x01
-- 
cgit v1.2.3


From 69ac9cd629ca96e59f34eb4ccd12d00b2c8276a7 Mon Sep 17 00:00:00 2001
From: Bernhard Walle <bwalle@suse.de>
Date: Fri, 27 Jun 2008 13:12:54 +0200
Subject: sysfs: add /sys/firmware/memmap

This patch adds /sys/firmware/memmap interface that represents the BIOS
(or Firmware) provided memory map. The tree looks like:

    /sys/firmware/memmap/0/start   (hex number)
                           end     (hex number)
                           type    (string)
    ...                 /1/start
                           end
                           type

With the following shell snippet one can print the memory map in the same form
the kernel prints itself when booting on x86 (the E820 map).

  --------- 8< --------------------------
    #!/bin/sh
    cd /sys/firmware/memmap
    for dir in * ; do
        start=$(cat $dir/start)
        end=$(cat $dir/end)
        type=$(cat $dir/type)
        printf "%016x-%016x (%s)\n" $start $[ $end +1] "$type"
    done
  --------- >8 --------------------------

That patch only provides the needed interface:

 1. The sysfs interface.
 2. The structure and enumeration definition.
 3. The function firmware_map_add() and firmware_map_add_early()
    that should be called from architecture code (E820/EFI, for
    example) to add the contents to the interface.

If the kernel is compiled without CONFIG_FIRMWARE_MEMMAP, the interface does
nothing without cluttering the architecture-specific code with #ifdef's.

The purpose of the new interface is kexec: While /proc/iomem represents
the *used* memory map (e.g. modified via kernel parameters like 'memmap'
and 'mem'), the /sys/firmware/memmap tree represents the unmodified memory
map provided via the firmware. So kexec can:

 - use the original memory map for rebooting,
 - use the /proc/iomem for setting up the ELF core headers for kdump
   case that should only represent the memory of the system.

The patch has been tested on i386 and x86_64.

Signed-off-by: Bernhard Walle <bwalle@suse.de>
Acked-by: Greg KH <gregkh@suse.de>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: kexec@lists.infradead.org
Cc: yhlu.kernel@gmail.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 Documentation/ABI/testing/sysfs-firmware-memmap |  71 ++++++++
 drivers/firmware/Kconfig                        |  10 ++
 drivers/firmware/Makefile                       |   1 +
 drivers/firmware/memmap.c                       | 205 ++++++++++++++++++++++++
 include/linux/firmware-map.h                    |  74 +++++++++
 5 files changed, 361 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-firmware-memmap
 create mode 100644 drivers/firmware/memmap.c
 create mode 100644 include/linux/firmware-map.h

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-firmware-memmap b/Documentation/ABI/testing/sysfs-firmware-memmap
new file mode 100644
index 000000000000..0d99ee6ae02e
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-firmware-memmap
@@ -0,0 +1,71 @@
+What:		/sys/firmware/memmap/
+Date:		June 2008
+Contact:	Bernhard Walle <bwalle@suse.de>
+Description:
+		On all platforms, the firmware provides a memory map which the
+		kernel reads. The resources from that memory map are registered
+		in the kernel resource tree and exposed to userspace via
+		/proc/iomem (together with other resources).
+
+		However, on most architectures that firmware-provided memory
+		map is modified afterwards by the kernel itself, either because
+		the kernel merges that memory map with other information or
+		just because the user overwrites that memory map via command
+		line.
+
+		kexec needs the raw firmware-provided memory map to setup the
+		parameter segment of the kernel that should be booted with
+		kexec. Also, the raw memory map is useful for debugging. For
+		that reason, /sys/firmware/memmap is an interface that provides
+		the raw memory map to userspace.
+
+		The structure is as follows: Under /sys/firmware/memmap there
+		are subdirectories with the number of the entry as their name:
+
+			/sys/firmware/memmap/0
+			/sys/firmware/memmap/1
+			/sys/firmware/memmap/2
+			/sys/firmware/memmap/3
+			...
+
+		The maximum depends on the number of memory map entries provided
+		by the firmware. The order is just the order that the firmware
+		provides.
+
+		Each directory contains three files:
+
+		start	: The start address (as hexadecimal number with the
+			  '0x' prefix).
+		end	: The end address, inclusive (regardless whether the
+			  firmware provides inclusive or exclusive ranges).
+		type	: Type of the entry as string. See below for a list of
+			  valid types.
+
+		So, for example:
+
+			/sys/firmware/memmap/0/start
+			/sys/firmware/memmap/0/end
+			/sys/firmware/memmap/0/type
+			/sys/firmware/memmap/1/start
+			...
+
+		Currently following types exist:
+
+		  - System RAM
+		  - ACPI Tables
+		  - ACPI Non-volatile Storage
+		  - reserved
+
+		Following shell snippet can be used to display that memory
+		map in a human-readable format:
+
+		-------------------- 8< ----------------------------------------
+		  #!/bin/bash
+		  cd /sys/firmware/memmap
+		  for dir in * ; do
+		      start=$(cat $dir/start)
+		      end=$(cat $dir/end)
+		      type=$(cat $dir/type)
+		      printf "%016x-%016x (%s)\n" $start $[ $end +1] "$type"
+		  done
+		-------------------- >8 ----------------------------------------
diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig
index dc2cec6127d1..ebb9e51deb0c 100644
--- a/drivers/firmware/Kconfig
+++ b/drivers/firmware/Kconfig
@@ -26,6 +26,16 @@ config EDD_OFF
 	  kernel. Say N if you want EDD enabled by default. EDD can be dynamically set
 	  using the kernel parameter 'edd={on|skipmbr|off}'.
 
+config FIRMWARE_MEMMAP
+    bool "Add firmware-provided memory map to sysfs" if EMBEDDED
+    default (X86_64 || X86_32)
+    help
+      Add the firmware-provided (unmodified) memory map to /sys/firmware/memmap.
+      That memory map is used for example by kexec to set up parameter area
+      for the next kernel, but can also be used for debugging purposes.
+
+      See also Documentation/ABI/testing/sysfs-firmware-memmap.
+
 config EFI_VARS
 	tristate "EFI Variable Support via sysfs"
 	depends on EFI
diff --git a/drivers/firmware/Makefile b/drivers/firmware/Makefile
index 4c9147154df8..1c3c17343dbe 100644
--- a/drivers/firmware/Makefile
+++ b/drivers/firmware/Makefile
@@ -10,3 +10,4 @@ obj-$(CONFIG_DCDBAS)		+= dcdbas.o
 obj-$(CONFIG_DMIID)		+= dmi-id.o
 obj-$(CONFIG_ISCSI_IBFT_FIND)	+= iscsi_ibft_find.o
 obj-$(CONFIG_ISCSI_IBFT)	+= iscsi_ibft.o
+obj-$(CONFIG_FIRMWARE_MEMMAP)	+= memmap.o
diff --git a/drivers/firmware/memmap.c b/drivers/firmware/memmap.c
new file mode 100644
index 000000000000..e23399c7f773
--- /dev/null
+++ b/drivers/firmware/memmap.c
@@ -0,0 +1,205 @@
+/*
+ * linux/drivers/firmware/memmap.c
+ *  Copyright (C) 2008 SUSE LINUX Products GmbH
+ *  by Bernhard Walle <bwalle@suse.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/string.h>
+#include <linux/firmware-map.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/bootmem.h>
+
+/*
+ * Data types ------------------------------------------------------------------
+ */
+
+/*
+ * Firmware map entry. Because firmware memory maps are flat and not
+ * hierarchical, it's ok to organise them in a linked list. No parent
+ * information is necessary as for the resource tree.
+ */
+struct firmware_map_entry {
+	resource_size_t		start;	/* start of the memory range */
+	resource_size_t		end;	/* end of the memory range (incl.) */
+	const char		*type;	/* type of the memory range */
+	struct list_head	list;	/* entry for the linked list */
+	struct kobject		kobj;   /* kobject for each entry */
+};
+
+/*
+ * Forward declarations --------------------------------------------------------
+ */
+static ssize_t memmap_attr_show(struct kobject *kobj,
+				struct attribute *attr, char *buf);
+static ssize_t start_show(struct firmware_map_entry *entry, char *buf);
+static ssize_t end_show(struct firmware_map_entry *entry, char *buf);
+static ssize_t type_show(struct firmware_map_entry *entry, char *buf);
+
+/*
+ * Static data -----------------------------------------------------------------
+ */
+
+struct memmap_attribute {
+	struct attribute attr;
+	ssize_t (*show)(struct firmware_map_entry *entry, char *buf);
+};
+
+struct memmap_attribute memmap_start_attr = __ATTR_RO(start);
+struct memmap_attribute memmap_end_attr   = __ATTR_RO(end);
+struct memmap_attribute memmap_type_attr  = __ATTR_RO(type);
+
+/*
+ * These are default attributes that are added for every memmap entry.
+ */
+static struct attribute *def_attrs[] = {
+	&memmap_start_attr.attr,
+	&memmap_end_attr.attr,
+	&memmap_type_attr.attr,
+	NULL
+};
+
+static struct sysfs_ops memmap_attr_ops = {
+	.show = memmap_attr_show,
+};
+
+static struct kobj_type memmap_ktype = {
+	.sysfs_ops	= &memmap_attr_ops,
+	.default_attrs	= def_attrs,
+};
+
+/*
+ * Registration functions ------------------------------------------------------
+ */
+
+/*
+ * Firmware memory map entries
+ */
+static LIST_HEAD(map_entries);
+
+/**
+ * Common implementation of firmware_map_add() and firmware_map_add_early()
+ * which expects a pre-allocated struct firmware_map_entry.
+ *
+ * @start: Start of the memory range.
+ * @end:   End of the memory range (inclusive).
+ * @type:  Type of the memory range.
+ * @entry: Pre-allocated (either kmalloc() or bootmem allocator), uninitialised
+ *         entry.
+ */
+static int firmware_map_add_entry(resource_size_t start, resource_size_t end,
+				  const char *type,
+				  struct firmware_map_entry *entry)
+{
+	BUG_ON(start > end);
+
+	entry->start = start;
+	entry->end = end;
+	entry->type = type;
+	INIT_LIST_HEAD(&entry->list);
+	kobject_init(&entry->kobj, &memmap_ktype);
+
+	list_add_tail(&entry->list, &map_entries);
+
+	return 0;
+}
+
+/*
+ * See <linux/firmware-map.h> for documentation.
+ */
+int firmware_map_add(resource_size_t start, resource_size_t end,
+		     const char *type)
+{
+	struct firmware_map_entry *entry;
+
+	entry = kmalloc(sizeof(struct firmware_map_entry), GFP_ATOMIC);
+	WARN_ON(!entry);
+	if (!entry)
+		return -ENOMEM;
+
+	return firmware_map_add_entry(start, end, type, entry);
+}
+
+/*
+ * See <linux/firmware-map.h> for documentation.
+ */
+int __init firmware_map_add_early(resource_size_t start, resource_size_t end,
+				  const char *type)
+{
+	struct firmware_map_entry *entry;
+
+	entry = alloc_bootmem_low(sizeof(struct firmware_map_entry));
+	WARN_ON(!entry);
+	if (!entry)
+		return -ENOMEM;
+
+	return firmware_map_add_entry(start, end, type, entry);
+}
+
+/*
+ * Sysfs functions -------------------------------------------------------------
+ */
+
+static ssize_t start_show(struct firmware_map_entry *entry, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "0x%llx\n", entry->start);
+}
+
+static ssize_t end_show(struct firmware_map_entry *entry, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "0x%llx\n", entry->end);
+}
+
+static ssize_t type_show(struct firmware_map_entry *entry, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%s\n", entry->type);
+}
+
+#define to_memmap_attr(_attr) container_of(_attr, struct memmap_attribute, attr)
+#define to_memmap_entry(obj) container_of(obj, struct firmware_map_entry, kobj)
+
+static ssize_t memmap_attr_show(struct kobject *kobj,
+				struct attribute *attr, char *buf)
+{
+	struct firmware_map_entry *entry = to_memmap_entry(kobj);
+	struct memmap_attribute *memmap_attr = to_memmap_attr(attr);
+
+	return memmap_attr->show(entry, buf);
+}
+
+/*
+ * Initialises stuff and adds the entries in the map_entries list to
+ * sysfs. Important is that firmware_map_add() and firmware_map_add_early()
+ * must be called before late_initcall.
+ */
+static int __init memmap_init(void)
+{
+	int i = 0;
+	struct firmware_map_entry *entry;
+	struct kset *memmap_kset;
+
+	memmap_kset = kset_create_and_add("memmap", NULL, firmware_kobj);
+	WARN_ON(!memmap_kset);
+	if (!memmap_kset)
+		return -ENOMEM;
+
+	list_for_each_entry(entry, &map_entries, list) {
+		entry->kobj.kset = memmap_kset;
+		kobject_add(&entry->kobj, NULL, "%d", i++);
+	}
+
+	return 0;
+}
+late_initcall(memmap_init);
+
diff --git a/include/linux/firmware-map.h b/include/linux/firmware-map.h
new file mode 100644
index 000000000000..acbdbcc16051
--- /dev/null
+++ b/include/linux/firmware-map.h
@@ -0,0 +1,74 @@
+/*
+ * include/linux/firmware-map.h:
+ *  Copyright (C) 2008 SUSE LINUX Products GmbH
+ *  by Bernhard Walle <bwalle@suse.de>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License v2.0 as published by
+ * the Free Software Foundation
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+#ifndef _LINUX_FIRMWARE_MAP_H
+#define _LINUX_FIRMWARE_MAP_H
+
+#include <linux/list.h>
+#include <linux/kobject.h>
+
+/*
+ * provide a dummy interface if CONFIG_FIRMWARE_MEMMAP is disabled
+ */
+#ifdef CONFIG_FIRMWARE_MEMMAP
+
+/**
+ * Adds a firmware mapping entry. This function uses kmalloc() for memory
+ * allocation. Use firmware_map_add_early() if you want to use the bootmem
+ * allocator.
+ *
+ * That function must be called before late_initcall.
+ *
+ * @start: Start of the memory range.
+ * @end:   End of the memory range (inclusive).
+ * @type:  Type of the memory range.
+ *
+ * Returns 0 on success, or -ENOMEM if no memory could be allocated.
+ */
+int firmware_map_add(resource_size_t start, resource_size_t end,
+		     const char *type);
+
+/**
+ * Adds a firmware mapping entry. This function uses the bootmem allocator
+ * for memory allocation. Use firmware_map_add() if you want to use kmalloc().
+ *
+ * That function must be called before late_initcall.
+ *
+ * @start: Start of the memory range.
+ * @end:   End of the memory range (inclusive).
+ * @type:  Type of the memory range.
+ *
+ * Returns 0 on success, or -ENOMEM if no memory could be allocated.
+ */
+int firmware_map_add_early(resource_size_t start, resource_size_t end,
+			   const char *type);
+
+#else /* CONFIG_FIRMWARE_MEMMAP */
+
+static inline int firmware_map_add(resource_size_t start, resource_size_t end,
+				   const char *type)
+{
+	return 0;
+}
+
+static inline int firmware_map_add_early(resource_size_t start,
+					 resource_size_t end, const char *type)
+{
+	return 0;
+}
+
+#endif /* CONFIG_FIRMWARE_MEMMAP */
+
+#endif /* _LINUX_FIRMWARE_MAP_H */
-- 
cgit v1.2.3


From 238f74a227fd7de8ea1bc66dcbbd36cf9920d1cb Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 2 Jul 2008 11:05:34 -0700
Subject: mac80211: move QOS control helpers into ieee80211.h

Also remove the WLAN_IS_QOS_DATA inline after removing the last
two users.  This starts moving away from using rx->fc to using
the header frame_control directly.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ieee80211.h |  2 ++
 net/mac80211/rx.c         | 28 ++++++++++++++--------------
 net/mac80211/wme.c        |  4 ++--
 net/mac80211/wme.h        |  8 --------
 4 files changed, 18 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index aa603c3d76d1..a1630ba0b87c 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -99,6 +99,8 @@
 #define IEEE80211_MAX_SSID_LEN		32
 #define IEEE80211_MAX_MESH_ID_LEN	32
 #define IEEE80211_QOS_CTL_LEN		2
+#define IEEE80211_QOS_CTL_TID_MASK	0x000F
+#define IEEE80211_QOS_CTL_TAG1D_MASK	0x0007
 
 struct ieee80211_hdr {
 	__le16 frame_control;
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 6a88e8f9bff0..a3a26e557274 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -321,20 +321,20 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb,
 
 static void ieee80211_parse_qos(struct ieee80211_rx_data *rx)
 {
-	u8 *data = rx->skb->data;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)rx->skb->data;
 	int tid;
 
 	/* does the frame have a qos control field? */
-	if (WLAN_FC_IS_QOS_DATA(rx->fc)) {
-		u8 *qc = data + ieee80211_get_hdrlen(rx->fc) - QOS_CONTROL_LEN;
+	if (ieee80211_is_data_qos(hdr->frame_control)) {
+		u8 *qc = ieee80211_get_qos_ctl(hdr);
 		/* frame has qos control */
-		tid = qc[0] & QOS_CONTROL_TID_MASK;
-		if (qc[0] & IEEE80211_QOS_CONTROL_A_MSDU_PRESENT)
+		tid = *qc & IEEE80211_QOS_CTL_TID_MASK;
+		if (*qc & IEEE80211_QOS_CONTROL_A_MSDU_PRESENT)
 			rx->flags |= IEEE80211_RX_AMSDU;
 		else
 			rx->flags &= ~IEEE80211_RX_AMSDU;
 	} else {
-		if (unlikely((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT)) {
+		if (unlikely(ieee80211_is_mgmt(hdr->frame_control))) {
 			/* Separate TID for management frames */
 			tid = NUM_RX_DATA_QUEUES - 1;
 		} else {
@@ -1037,19 +1037,19 @@ ieee80211_rx_h_ps_poll(struct ieee80211_rx_data *rx)
 static ieee80211_rx_result debug_noinline
 ieee80211_rx_h_remove_qos_control(struct ieee80211_rx_data *rx)
 {
-	u16 fc = rx->fc;
 	u8 *data = rx->skb->data;
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) data;
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)data;
 
-	if (!WLAN_FC_IS_QOS_DATA(fc))
+	if (!ieee80211_is_data_qos(hdr->frame_control))
 		return RX_CONTINUE;
 
 	/* remove the qos control field, update frame type and meta-data */
-	memmove(data + 2, data, ieee80211_get_hdrlen(fc) - 2);
-	hdr = (struct ieee80211_hdr *) skb_pull(rx->skb, 2);
+	memmove(data + IEEE80211_QOS_CTL_LEN, data,
+		ieee80211_hdrlen(hdr->frame_control) - IEEE80211_QOS_CTL_LEN);
+	hdr = (struct ieee80211_hdr *)skb_pull(rx->skb, IEEE80211_QOS_CTL_LEN);
 	/* change frame type to non QOS */
-	rx->fc = fc &= ~IEEE80211_STYPE_QOS_DATA;
-	hdr->frame_control = cpu_to_le16(fc);
+	rx->fc &= ~IEEE80211_STYPE_QOS_DATA;
+	hdr->frame_control &= ~cpu_to_le16(IEEE80211_STYPE_QOS_DATA);
 
 	return RX_CONTINUE;
 }
@@ -2044,7 +2044,7 @@ static u8 ieee80211_rx_reorder_ampdu(struct ieee80211_local *local,
 	if (!ieee80211_is_data_qos(hdr->frame_control))
 		goto end_reorder;
 
-	tid = *ieee80211_get_qos_ctl(hdr) & QOS_CONTROL_TID_MASK;
+	tid = *ieee80211_get_qos_ctl(hdr) & IEEE80211_QOS_CTL_TID_MASK;
 
 	if (sta->ampdu_mlme.tid_state_rx[tid] != HT_AGG_STATE_OPERATIONAL)
 		goto end_reorder;
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index a1a53a4f2b99..5c666f7eda8f 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -154,7 +154,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 		queue = skb_get_queue_mapping(skb);
 		rcu_read_lock();
 		sta = sta_info_get(local, hdr->addr1);
-		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
+		tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
 		if (sta) {
 			int ampdu_queue = sta->tid_to_tx_q[tid];
 			if ((ampdu_queue < QD_NUM(hw)) &&
@@ -181,7 +181,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd)
 	if (ieee80211_is_data_qos(hdr->frame_control)) {
 		u8 *p = ieee80211_get_qos_ctl(hdr);
 		u8 ack_policy = 0;
-		tid = skb->priority & QOS_CONTROL_TAG1D_MASK;
+		tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK;
 		if (local->wifi_wme_noack_test)
 			ack_policy |= QOS_CONTROL_ACK_POLICY_NOACK <<
 					QOS_CONTROL_ACK_POLICY_SHIFT;
diff --git a/net/mac80211/wme.h b/net/mac80211/wme.h
index bbdb53344817..1aca609eccfc 100644
--- a/net/mac80211/wme.h
+++ b/net/mac80211/wme.h
@@ -19,18 +19,10 @@
 #define QOS_CONTROL_ACK_POLICY_NORMAL 0
 #define QOS_CONTROL_ACK_POLICY_NOACK 1
 
-#define QOS_CONTROL_TID_MASK 0x0f
 #define QOS_CONTROL_ACK_POLICY_SHIFT 5
 
-#define QOS_CONTROL_TAG1D_MASK 0x07
-
 extern const int ieee802_1d_to_ac[8];
 
-static inline int WLAN_FC_IS_QOS_DATA(u16 fc)
-{
-	return (fc & 0x8C) == 0x88;
-}
-
 #ifdef CONFIG_MAC80211_QOS
 void ieee80211_install_qdisc(struct net_device *dev);
 int ieee80211_qdisc_installed(struct net_device *dev);
-- 
cgit v1.2.3


From 7cc5bf9a3a84e5a02e23e5739fb894790b37c101 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 8 Jul 2008 11:58:21 -0700
Subject: dmaengine: track the number of clients using a channel

Haavard's dma-slave interface would like to test for exclusive access to a
channel.  The standard channel refcounting is not sufficient in that it
tracks more than just client references, it is also inaccurate as reference
counts are percpu until the channel is removed.

This change also enables a future fix to deallocate resources when a client
declines to use a capable channel.

Acked-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   | 14 ++++++++++----
 include/linux/dmaengine.h |  2 ++
 2 files changed, 12 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 99c22b42bada..10de69eb1a3e 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -183,9 +183,10 @@ static void dma_client_chan_alloc(struct dma_client *client)
 				/* we are done once this client rejects
 				 * an available resource
 				 */
-				if (ack == DMA_ACK)
+				if (ack == DMA_ACK) {
 					dma_chan_get(chan);
-				else if (ack == DMA_NAK)
+					chan->client_count++;
+				} else if (ack == DMA_NAK)
 					return;
 			}
 		}
@@ -272,8 +273,10 @@ static void dma_clients_notify_removed(struct dma_chan *chan)
 		/* client was holding resources for this channel so
 		 * free it
 		 */
-		if (ack == DMA_ACK)
+		if (ack == DMA_ACK) {
 			dma_chan_put(chan);
+			chan->client_count--;
+		}
 	}
 
 	mutex_unlock(&dma_list_mutex);
@@ -313,8 +316,10 @@ void dma_async_client_unregister(struct dma_client *client)
 			ack = client->event_callback(client, chan,
 				DMA_RESOURCE_REMOVED);
 
-			if (ack == DMA_ACK)
+			if (ack == DMA_ACK) {
 				dma_chan_put(chan);
+				chan->client_count--;
+			}
 		}
 
 	list_del(&client->global_node);
@@ -394,6 +399,7 @@ int dma_async_device_register(struct dma_device *device)
 		kref_get(&device->refcount);
 		kref_get(&device->refcount);
 		kref_init(&chan->refcount);
+		chan->client_count = 0;
 		chan->slow_ref = 0;
 		INIT_RCU_HEAD(&chan->rcu);
 	}
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index d08a5c5eb928..6432b8343220 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -139,6 +139,7 @@ struct dma_chan_percpu {
  * @rcu: the DMA channel's RCU head
  * @device_node: used to add this to the device chan list
  * @local: per-cpu pointer to a struct dma_chan_percpu
+ * @client-count: how many clients are using this channel
  */
 struct dma_chan {
 	struct dma_device *device;
@@ -154,6 +155,7 @@ struct dma_chan {
 
 	struct list_head device_node;
 	struct dma_chan_percpu *local;
+	int client_count;
 };
 
 #define to_dma_chan(p) container_of(p, struct dma_chan, dev)
-- 
cgit v1.2.3


From 848c536a37b8db4e461f14ca15fe29850151c822 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Tue, 8 Jul 2008 11:58:58 -0700
Subject: dmaengine: Add dma_client parameter to device_alloc_chan_resources

A DMA controller capable of doing slave transfers may need to know a
few things about the slave when preparing the channel. We don't want
to add this information to struct dma_channel since the channel hasn't
yet been bound to a client at this point.

Instead, pass a reference to the client requesting the channel to the
driver's device_alloc_chan_resources hook so that it can pick the
necessary information from the dma_client struct by itself.

[dan.j.williams@intel.com: fixed up fsldma and mv_xor]
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   | 3 ++-
 drivers/dma/fsldma.c      | 7 ++++---
 drivers/dma/ioat_dma.c    | 5 +++--
 drivers/dma/iop-adma.c    | 7 ++++---
 drivers/dma/mv_xor.c      | 7 ++++---
 include/linux/dmaengine.h | 3 ++-
 6 files changed, 19 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 10de69eb1a3e..7344f5dbd501 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -174,7 +174,8 @@ static void dma_client_chan_alloc(struct dma_client *client)
 			if (!dma_chan_satisfies_mask(chan, client->cap_mask))
 				continue;
 
-			desc = chan->device->device_alloc_chan_resources(chan);
+			desc = chan->device->device_alloc_chan_resources(
+					chan, client);
 			if (desc >= 0) {
 				ack = client->event_callback(client,
 						chan,
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 724f6fdd0af6..c0059ca58340 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -366,7 +366,8 @@ static struct fsl_desc_sw *fsl_dma_alloc_descriptor(
  *
  * Return - The number of descriptors allocated.
  */
-static int fsl_dma_alloc_chan_resources(struct dma_chan *chan)
+static int fsl_dma_alloc_chan_resources(struct dma_chan *chan,
+					struct dma_client *client)
 {
 	struct fsl_dma_chan *fsl_chan = to_fsl_chan(chan);
 	LIST_HEAD(tmp_list);
@@ -819,7 +820,7 @@ static int fsl_dma_self_test(struct fsl_dma_chan *fsl_chan)
 
 	chan = &fsl_chan->common;
 
-	if (fsl_dma_alloc_chan_resources(chan) < 1) {
+	if (fsl_dma_alloc_chan_resources(chan, NULL) < 1) {
 		dev_err(fsl_chan->dev,
 				"selftest: Cannot alloc resources for DMA\n");
 		err = -ENODEV;
@@ -847,7 +848,7 @@ static int fsl_dma_self_test(struct fsl_dma_chan *fsl_chan)
 	/* Test free and re-alloc channel resources */
 	fsl_dma_free_chan_resources(chan);
 
-	if (fsl_dma_alloc_chan_resources(chan) < 1) {
+	if (fsl_dma_alloc_chan_resources(chan, NULL) < 1) {
 		dev_err(fsl_chan->dev,
 				"selftest: Cannot alloc resources for DMA\n");
 		err = -ENODEV;
diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
index 318e8a22d814..90e5b0a28cbf 100644
--- a/drivers/dma/ioat_dma.c
+++ b/drivers/dma/ioat_dma.c
@@ -452,7 +452,8 @@ static void ioat2_dma_massage_chan_desc(struct ioat_dma_chan *ioat_chan)
  * ioat_dma_alloc_chan_resources - returns the number of allocated descriptors
  * @chan: the channel to be filled out
  */
-static int ioat_dma_alloc_chan_resources(struct dma_chan *chan)
+static int ioat_dma_alloc_chan_resources(struct dma_chan *chan,
+					 struct dma_client *client)
 {
 	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
 	struct ioat_desc_sw *desc;
@@ -1049,7 +1050,7 @@ static int ioat_dma_self_test(struct ioatdma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (device->common.device_alloc_chan_resources(dma_chan) < 1) {
+	if (device->common.device_alloc_chan_resources(dma_chan, NULL) < 1) {
 		dev_err(&device->pdev->dev,
 			"selftest cannot allocate chan resource\n");
 		err = -ENODEV;
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index 4e6b052c0654..b57564dd0232 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -444,7 +444,8 @@ static void iop_chan_start_null_memcpy(struct iop_adma_chan *iop_chan);
 static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan);
 
 /* returns the number of allocated descriptors */
-static int iop_adma_alloc_chan_resources(struct dma_chan *chan)
+static int iop_adma_alloc_chan_resources(struct dma_chan *chan,
+					 struct dma_client *client)
 {
 	char *hw_desc;
 	int idx;
@@ -838,7 +839,7 @@ static int __devinit iop_adma_memcpy_self_test(struct iop_adma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
+	if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
@@ -936,7 +937,7 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
+	if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index f0c123ce8ae0..8239cfdbc2e6 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -588,7 +588,8 @@ submit_done:
 }
 
 /* returns the number of allocated descriptors */
-static int mv_xor_alloc_chan_resources(struct dma_chan *chan)
+static int mv_xor_alloc_chan_resources(struct dma_chan *chan,
+				       struct dma_client *client)
 {
 	char *hw_desc;
 	int idx;
@@ -938,7 +939,7 @@ static int __devinit mv_xor_memcpy_self_test(struct mv_xor_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (mv_xor_alloc_chan_resources(dma_chan) < 1) {
+	if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
@@ -1033,7 +1034,7 @@ mv_xor_xor_self_test(struct mv_xor_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (mv_xor_alloc_chan_resources(dma_chan) < 1) {
+	if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 6432b8343220..ba89b0f5056e 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -281,7 +281,8 @@ struct dma_device {
 	int dev_id;
 	struct device *dev;
 
-	int (*device_alloc_chan_resources)(struct dma_chan *chan);
+	int (*device_alloc_chan_resources)(struct dma_chan *chan,
+			struct dma_client *client);
 	void (*device_free_chan_resources)(struct dma_chan *chan);
 
 	struct dma_async_tx_descriptor *(*device_prep_dma_memcpy)(
-- 
cgit v1.2.3


From e1d181efb14a93cf263d6c588a5395518edf3294 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Fri, 4 Jul 2008 00:13:40 -0700
Subject: dmaengine: add DMA_COMPL_SKIP_{SRC,DEST}_UNMAP flags to control dma
 unmap

In some cases client code may need the dma-driver to skip the unmap of source
and/or destination buffers.  Setting these flags indicates to the driver to
skip the unmap step.  In this regard async_xor is currently broken in that it
allows the destination buffer to be unmapped while an operation is still in
progress, i.e. when the number of sources exceeds the hardware channel's
maximum (fixed in a subsequent patch).

Acked-by: Saeed Bishara <saeed@marvell.com>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Acked-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/ioat_dma.c    | 48 +++++++++++++++++++++++------------------------
 drivers/dma/iop-adma.c    | 29 +++++++++++++++++-----------
 drivers/dma/mv_xor.c      | 22 ++++++++++++++++------
 include/linux/dmaengine.h |  4 ++++
 4 files changed, 61 insertions(+), 42 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
index 90e5b0a28cbf..171cad69f318 100644
--- a/drivers/dma/ioat_dma.c
+++ b/drivers/dma/ioat_dma.c
@@ -757,6 +757,27 @@ static void ioat_dma_cleanup_tasklet(unsigned long data)
 	       chan->reg_base + IOAT_CHANCTRL_OFFSET);
 }
 
+static void
+ioat_dma_unmap(struct ioat_dma_chan *ioat_chan, struct ioat_desc_sw *desc)
+{
+	/*
+	 * yes we are unmapping both _page and _single
+	 * alloc'd regions with unmap_page. Is this
+	 * *really* that bad?
+	 */
+	if (!(desc->async_tx.flags & DMA_COMPL_SKIP_DEST_UNMAP))
+		pci_unmap_page(ioat_chan->device->pdev,
+				pci_unmap_addr(desc, dst),
+				pci_unmap_len(desc, len),
+				PCI_DMA_FROMDEVICE);
+
+	if (!(desc->async_tx.flags & DMA_COMPL_SKIP_SRC_UNMAP))
+		pci_unmap_page(ioat_chan->device->pdev,
+				pci_unmap_addr(desc, src),
+				pci_unmap_len(desc, len),
+				PCI_DMA_TODEVICE);
+}
+
 /**
  * ioat_dma_memcpy_cleanup - cleanup up finished descriptors
  * @chan: ioat channel to be cleaned up
@@ -817,21 +838,7 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 			 */
 			if (desc->async_tx.cookie) {
 				cookie = desc->async_tx.cookie;
-
-				/*
-				 * yes we are unmapping both _page and _single
-				 * alloc'd regions with unmap_page. Is this
-				 * *really* that bad?
-				 */
-				pci_unmap_page(ioat_chan->device->pdev,
-						pci_unmap_addr(desc, dst),
-						pci_unmap_len(desc, len),
-						PCI_DMA_FROMDEVICE);
-				pci_unmap_page(ioat_chan->device->pdev,
-						pci_unmap_addr(desc, src),
-						pci_unmap_len(desc, len),
-						PCI_DMA_TODEVICE);
-
+				ioat_dma_unmap(ioat_chan, desc);
 				if (desc->async_tx.callback) {
 					desc->async_tx.callback(desc->async_tx.callback_param);
 					desc->async_tx.callback = NULL;
@@ -890,16 +897,7 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 				if (desc->async_tx.cookie) {
 					cookie = desc->async_tx.cookie;
 					desc->async_tx.cookie = 0;
-
-					pci_unmap_page(ioat_chan->device->pdev,
-						      pci_unmap_addr(desc, dst),
-						      pci_unmap_len(desc, len),
-						      PCI_DMA_FROMDEVICE);
-					pci_unmap_page(ioat_chan->device->pdev,
-						      pci_unmap_addr(desc, src),
-						      pci_unmap_len(desc, len),
-						      PCI_DMA_TODEVICE);
-
+					ioat_dma_unmap(ioat_chan, desc);
 					if (desc->async_tx.callback) {
 						desc->async_tx.callback(desc->async_tx.callback_param);
 						desc->async_tx.callback = NULL;
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index b57564dd0232..434013d41288 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -82,17 +82,24 @@ iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
 			struct device *dev =
 				&iop_chan->device->pdev->dev;
 			u32 len = unmap->unmap_len;
-			u32 src_cnt = unmap->unmap_src_cnt;
-			dma_addr_t addr = iop_desc_get_dest_addr(unmap,
-				iop_chan);
-
-			dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
-			while (src_cnt--) {
-				addr = iop_desc_get_src_addr(unmap,
-							iop_chan,
-							src_cnt);
-				dma_unmap_page(dev, addr, len,
-					DMA_TO_DEVICE);
+			enum dma_ctrl_flags flags = desc->async_tx.flags;
+			u32 src_cnt;
+			dma_addr_t addr;
+
+			if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
+				addr = iop_desc_get_dest_addr(unmap, iop_chan);
+				dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
+			}
+
+			if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
+				src_cnt = unmap->unmap_src_cnt;
+				while (src_cnt--) {
+					addr = iop_desc_get_src_addr(unmap,
+								     iop_chan,
+								     src_cnt);
+					dma_unmap_page(dev, addr, len,
+						       DMA_TO_DEVICE);
+				}
 			}
 			desc->group_head = NULL;
 		}
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index 8239cfdbc2e6..a4e4494663bf 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -308,13 +308,23 @@ mv_xor_run_tx_complete_actions(struct mv_xor_desc_slot *desc,
 			struct device *dev =
 				&mv_chan->device->pdev->dev;
 			u32 len = unmap->unmap_len;
-			u32 src_cnt = unmap->unmap_src_cnt;
-			dma_addr_t addr = mv_desc_get_dest_addr(unmap);
+			enum dma_ctrl_flags flags = desc->async_tx.flags;
+			u32 src_cnt;
+			dma_addr_t addr;
 
-			dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
-			while (src_cnt--) {
-				addr = mv_desc_get_src_addr(unmap, src_cnt);
-				dma_unmap_page(dev, addr, len, DMA_TO_DEVICE);
+			if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
+				addr = mv_desc_get_dest_addr(unmap);
+				dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
+			}
+
+			if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
+				src_cnt = unmap->unmap_src_cnt;
+				while (src_cnt--) {
+					addr = mv_desc_get_src_addr(unmap,
+								    src_cnt);
+					dma_unmap_page(dev, addr, len,
+						       DMA_TO_DEVICE);
+				}
 			}
 			desc->group_head = NULL;
 		}
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index ba89b0f5056e..b058d6360383 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -102,10 +102,14 @@ enum dma_transaction_type {
  * @DMA_CTRL_ACK - the descriptor cannot be reused until the client
  * 	acknowledges receipt, i.e. has has a chance to establish any
  * 	dependency chains
+ * @DMA_COMPL_SKIP_SRC_UNMAP - set to disable dma-unmapping the source buffer(s)
+ * @DMA_COMPL_SKIP_DEST_UNMAP - set to disable dma-unmapping the destination(s)
  */
 enum dma_ctrl_flags {
 	DMA_PREP_INTERRUPT = (1 << 0),
 	DMA_CTRL_ACK = (1 << 1),
+	DMA_COMPL_SKIP_SRC_UNMAP = (1 << 2),
+	DMA_COMPL_SKIP_DEST_UNMAP = (1 << 3),
 };
 
 /**
-- 
cgit v1.2.3


From dc0ee6435cb92ccc81b14ff28d163fecc5a7f120 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Tue, 8 Jul 2008 11:59:35 -0700
Subject: dmaengine: Add slave DMA interface

This patch adds the necessary interfaces to the DMA Engine framework
to use functionality found on most embedded DMA controllers: DMA from
and to I/O registers with hardware handshaking.

In this context, hardware hanshaking means that the peripheral that
owns the I/O registers in question is able to tell the DMA controller
when more data is available for reading, or when there is room for
more data to be written. This usually happens internally on the chip,
but these signals may also be exported outside the chip for things
like IDE DMA, etc.

A new struct dma_slave is introduced. This contains information that
the DMA engine driver needs to set up slave transfers to and from a
slave device. Most engines supporting DMA slave transfers will want to
extend this structure with controller-specific parameters.  This
additional information is usually passed from the platform/board code
through the client driver.

A "slave" pointer is added to the dma_client struct. This must point
to a valid dma_slave structure iff the DMA_SLAVE capability is
requested.  The DMA engine driver may use this information in its
device_alloc_chan_resources hook to configure the DMA controller for
slave transfers from and to the given slave device.

A new operation for preparing slave DMA transfers is added to struct
dma_device. This takes a scatterlist and returns a single descriptor
representing the whole transfer.

Another new operation for terminating all pending transfers is added as
well. The latter is needed because there may be errors outside the scope
of the DMA Engine framework that may require DMA operations to be
terminated prematurely.

DMA Engine drivers may extend the dma_device, dma_chan and/or
dma_slave_descriptor structures to allow controller-specific
operations. The client driver can detect such extensions by looking at
the DMA Engine's struct device, or it can request a specific DMA
Engine device by setting the dma_dev field in struct dma_slave.

dmaslave interface changes since v4:
  * Fix checkpatch errors
  * Fix changelog (there are no slave descriptors anymore)

dmaslave interface changes since v3:
  * Use dma_data_direction instead of a new enum
  * Submit slave transfers as scatterlists
  * Remove the DMA slave descriptor struct

dmaslave interface changes since v2:
  * Add a dma_dev field to struct dma_slave. If set, the client can
    only be bound to the DMA controller that corresponds to this
    device.  This allows controller-specific extensions of the
    dma_slave structure; if the device matches, the controller may
    safely assume its extensions are present.
  * Move reg_width into struct dma_slave as there are currently no
    users that need to be able to set the width on a per-transfer
    basis.

dmaslave interface changes since v1:
  * Drop the set_direction and set_width descriptor hooks. Pass the
    direction and width to the prep function instead.
  * Declare a dma_slave struct with fixed information about a slave,
    i.e. register addresses, handshake interfaces and such.
  * Add pointer to a dma_slave struct to dma_client. Can be NULL if
    the DMA_SLAVE capability isn't requested.
  * Drop the set_slave device hook since the alloc_chan_resources hook
    now has enough information to set up the channel for slave
    transfers.

Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   | 16 ++++++++++++++-
 include/linux/dmaengine.h | 52 ++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 66 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 7344f5dbd501..dc003a3a787d 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -169,7 +169,12 @@ static void dma_client_chan_alloc(struct dma_client *client)
 	enum dma_state_client ack;
 
 	/* Find a channel */
-	list_for_each_entry(device, &dma_device_list, global_node)
+	list_for_each_entry(device, &dma_device_list, global_node) {
+		/* Does the client require a specific DMA controller? */
+		if (client->slave && client->slave->dma_dev
+				&& client->slave->dma_dev != device->dev)
+			continue;
+
 		list_for_each_entry(chan, &device->channels, device_node) {
 			if (!dma_chan_satisfies_mask(chan, client->cap_mask))
 				continue;
@@ -191,6 +196,7 @@ static void dma_client_chan_alloc(struct dma_client *client)
 					return;
 			}
 		}
+	}
 }
 
 enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie)
@@ -289,6 +295,10 @@ static void dma_clients_notify_removed(struct dma_chan *chan)
  */
 void dma_async_client_register(struct dma_client *client)
 {
+	/* validate client data */
+	BUG_ON(dma_has_cap(DMA_SLAVE, client->cap_mask) &&
+		!client->slave);
+
 	mutex_lock(&dma_list_mutex);
 	list_add_tail(&client->global_node, &dma_client_list);
 	mutex_unlock(&dma_list_mutex);
@@ -365,6 +375,10 @@ int dma_async_device_register(struct dma_device *device)
 		!device->device_prep_dma_memset);
 	BUG_ON(dma_has_cap(DMA_INTERRUPT, device->cap_mask) &&
 		!device->device_prep_dma_interrupt);
+	BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) &&
+		!device->device_prep_slave_sg);
+	BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) &&
+		!device->device_terminate_all);
 
 	BUG_ON(!device->device_alloc_chan_resources);
 	BUG_ON(!device->device_free_chan_resources);
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index b058d6360383..9b91d341e1fa 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -89,10 +89,23 @@ enum dma_transaction_type {
 	DMA_MEMSET,
 	DMA_MEMCPY_CRC32C,
 	DMA_INTERRUPT,
+	DMA_SLAVE,
 };
 
 /* last transaction type for creation of the capabilities mask */
-#define DMA_TX_TYPE_END (DMA_INTERRUPT + 1)
+#define DMA_TX_TYPE_END (DMA_SLAVE + 1)
+
+/**
+ * enum dma_slave_width - DMA slave register access width.
+ * @DMA_SLAVE_WIDTH_8BIT: Do 8-bit slave register accesses
+ * @DMA_SLAVE_WIDTH_16BIT: Do 16-bit slave register accesses
+ * @DMA_SLAVE_WIDTH_32BIT: Do 32-bit slave register accesses
+ */
+enum dma_slave_width {
+	DMA_SLAVE_WIDTH_8BIT,
+	DMA_SLAVE_WIDTH_16BIT,
+	DMA_SLAVE_WIDTH_32BIT,
+};
 
 /**
  * enum dma_ctrl_flags - DMA flags to augment operation preparation,
@@ -118,6 +131,32 @@ enum dma_ctrl_flags {
  */
 typedef struct { DECLARE_BITMAP(bits, DMA_TX_TYPE_END); } dma_cap_mask_t;
 
+/**
+ * struct dma_slave - Information about a DMA slave
+ * @dev: device acting as DMA slave
+ * @dma_dev: required DMA master device. If non-NULL, the client can not be
+ *	bound to other masters than this.
+ * @tx_reg: physical address of data register used for
+ *	memory-to-peripheral transfers
+ * @rx_reg: physical address of data register used for
+ *	peripheral-to-memory transfers
+ * @reg_width: peripheral register width
+ *
+ * If dma_dev is non-NULL, the client can not be bound to other DMA
+ * masters than the one corresponding to this device. The DMA master
+ * driver may use this to determine if there is controller-specific
+ * data wrapped around this struct. Drivers of platform code that sets
+ * the dma_dev field must therefore make sure to use an appropriate
+ * controller-specific dma slave structure wrapping this struct.
+ */
+struct dma_slave {
+	struct device		*dev;
+	struct device		*dma_dev;
+	dma_addr_t		tx_reg;
+	dma_addr_t		rx_reg;
+	enum dma_slave_width	reg_width;
+};
+
 /**
  * struct dma_chan_percpu - the per-CPU part of struct dma_chan
  * @refcount: local_t used for open-coded "bigref" counting
@@ -208,11 +247,14 @@ typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client,
  * @event_callback: func ptr to call when something happens
  * @cap_mask: only return channels that satisfy the requested capabilities
  *  a value of zero corresponds to any capability
+ * @slave: data for preparing slave transfer. Must be non-NULL iff the
+ *  DMA_SLAVE capability is requested.
  * @global_node: list_head for global dma_client_list
  */
 struct dma_client {
 	dma_event_callback	event_callback;
 	dma_cap_mask_t		cap_mask;
+	struct dma_slave	*slave;
 	struct list_head	global_node;
 };
 
@@ -269,6 +311,8 @@ struct dma_async_tx_descriptor {
  * @device_prep_dma_zero_sum: prepares a zero_sum operation
  * @device_prep_dma_memset: prepares a memset operation
  * @device_prep_dma_interrupt: prepares an end of chain interrupt operation
+ * @device_prep_slave_sg: prepares a slave dma operation
+ * @device_terminate_all: terminate all pending operations
  * @device_issue_pending: push pending transactions to hardware
  */
 struct dma_device {
@@ -304,6 +348,12 @@ struct dma_device {
 	struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
 		struct dma_chan *chan, unsigned long flags);
 
+	struct dma_async_tx_descriptor *(*device_prep_slave_sg)(
+		struct dma_chan *chan, struct scatterlist *sgl,
+		unsigned int sg_len, enum dma_data_direction direction,
+		unsigned long flags);
+	void (*device_terminate_all)(struct dma_chan *chan);
+
 	enum dma_status (*device_is_tx_complete)(struct dma_chan *chan,
 			dma_cookie_t cookie, dma_cookie_t *last,
 			dma_cookie_t *used);
-- 
cgit v1.2.3


From 3bfb1d20b547a5071d01344581eac5846ea84491 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Tue, 8 Jul 2008 11:59:42 -0700
Subject: dmaengine: Driver for the Synopsys DesignWare DMA controller

This adds a driver for the Synopsys DesignWare DMA controller (aka
DMACA on AVR32 systems.) This DMA controller can be found integrated
on the AT32AP7000 chip and is primarily meant for peripheral DMA
transfer, but can also be used for memory-to-memory transfers.

This patch is based on a driver from David Brownell which was based on
an older version of the DMA Engine framework. It also implements the
proposed extensions to the DMA Engine API for slave DMA operations.

The dmatest client shows no problems, but there may still be room for
improvement performance-wise. DMA slave transfer performance is
definitely "good enough"; reading 100 MiB from an SD card running at ~20
MHz yields ~7.2 MiB/s average transfer rate.

Full documentation for this controller can be found in the Synopsys
DW AHB DMAC Databook:

http://www.synopsys.com/designware/docs/iip/DW_ahb_dmac/latest/doc/dw_ahb_dmac_db.pdf

The controller has lots of implementation options, so it's usually a
good idea to check the data sheet of the chip it's intergrated on as
well. The AT32AP7000 data sheet can be found here:

http://www.atmel.com/dyn/products/datasheets.asp?family_id=682


Changes since v4:
  * Use client_count instead of dma_chan_is_in_use()
  * Add missing include
  * Unmap buffers unless client told us not to

Changes since v3:
  * Update to latest DMA engine and DMA slave APIs
  * Embed the hw descriptor into the sw descriptor
  * Clean up and update MODULE_DESCRIPTION, copyright date, etc.

Changes since v2:
  * Dequeue all pending transfers in terminate_all()
  * Rename dw_dmac.h -> dw_dmac_regs.h
  * Define and use controller-specific dma_slave data
  * Fix up a few outdated comments
  * Define hardware registers as structs (doesn't generate better
    code, unfortunately, but it looks nicer.)
  * Get number of channels from platform_data instead of hardcoding it
    based on CONFIG_WHATEVER_CPU.
  * Give slave clients exclusive access to the channel

Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>,
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 arch/avr32/mach-at32ap/at32ap700x.c        |   27 +-
 drivers/dma/Kconfig                        |    9 +
 drivers/dma/Makefile                       |    1 +
 drivers/dma/dw_dmac.c                      | 1122 ++++++++++++++++++++++++++++
 drivers/dma/dw_dmac_regs.h                 |  225 ++++++
 include/asm-avr32/arch-at32ap/at32ap700x.h |   16 +
 include/linux/dw_dmac.h                    |   62 ++
 7 files changed, 1449 insertions(+), 13 deletions(-)
 create mode 100644 drivers/dma/dw_dmac.c
 create mode 100644 drivers/dma/dw_dmac_regs.h
 create mode 100644 include/linux/dw_dmac.h

(limited to 'include/linux')

diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index 0f24b4f85c17..892e27e0d583 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -7,6 +7,7 @@
  */
 #include <linux/clk.h>
 #include <linux/delay.h>
+#include <linux/dw_dmac.h>
 #include <linux/fb.h>
 #include <linux/init.h>
 #include <linux/platform_device.h>
@@ -599,6 +600,17 @@ static void __init genclk_init_parent(struct clk *clk)
 	clk->parent = parent;
 }
 
+static struct dw_dma_platform_data dw_dmac0_data = {
+	.nr_channels	= 3,
+};
+
+static struct resource dw_dmac0_resource[] = {
+	PBMEM(0xff200000),
+	IRQ(2),
+};
+DEFINE_DEV_DATA(dw_dmac, 0);
+DEV_CLK(hclk, dw_dmac0, hsb, 10);
+
 /* --------------------------------------------------------------------
  *  System peripherals
  * -------------------------------------------------------------------- */
@@ -705,17 +717,6 @@ static struct clk pico_clk = {
 	.users		= 1,
 };
 
-static struct resource dmaca0_resource[] = {
-	{
-		.start	= 0xff200000,
-		.end	= 0xff20ffff,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(2),
-};
-DEFINE_DEV(dmaca, 0);
-DEV_CLK(hclk, dmaca0, hsb, 10);
-
 /* --------------------------------------------------------------------
  * HMATRIX
  * -------------------------------------------------------------------- */
@@ -828,7 +829,7 @@ void __init at32_add_system_devices(void)
 	platform_device_register(&at32_eic0_device);
 	platform_device_register(&smc0_device);
 	platform_device_register(&pdc_device);
-	platform_device_register(&dmaca0_device);
+	platform_device_register(&dw_dmac0_device);
 
 	platform_device_register(&at32_tcb0_device);
 	platform_device_register(&at32_tcb1_device);
@@ -1891,7 +1892,7 @@ struct clk *at32_clock_list[] = {
 	&smc0_mck,
 	&pdc_hclk,
 	&pdc_pclk,
-	&dmaca0_hclk,
+	&dw_dmac0_hclk,
 	&pico_clk,
 	&pio0_mck,
 	&pio1_mck,
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 4b6bd3d099cf..cd303901eb5b 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -38,6 +38,15 @@ config INTEL_IOP_ADMA
 	help
 	  Enable support for the Intel(R) IOP Series RAID engines.
 
+config DW_DMAC
+	tristate "Synopsys DesignWare AHB DMA support"
+	depends on AVR32
+	select DMA_ENGINE
+	default y if CPU_AT32AP7000
+	help
+	  Support the Synopsys DesignWare AHB DMA controller.  This
+	  can be integrated in chips such as the Atmel AT32ap7000.
+
 config FSL_DMA
 	bool "Freescale MPC85xx/MPC83xx DMA support"
 	depends on PPC
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 181e3646fbfe..14f59527d4f6 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -6,3 +6,4 @@ ioatdma-objs := ioat.o ioat_dma.o ioat_dca.o
 obj-$(CONFIG_INTEL_IOP_ADMA) += iop-adma.o
 obj-$(CONFIG_FSL_DMA) += fsldma.o
 obj-$(CONFIG_MV_XOR) += mv_xor.o
+obj-$(CONFIG_DW_DMAC) += dw_dmac.o
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
new file mode 100644
index 000000000000..94df91771243
--- /dev/null
+++ b/drivers/dma/dw_dmac.c
@@ -0,0 +1,1122 @@
+/*
+ * Driver for the Synopsys DesignWare DMA Controller (aka DMACA on
+ * AVR32 systems.)
+ *
+ * Copyright (C) 2007-2008 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include "dw_dmac_regs.h"
+
+/*
+ * This supports the Synopsys "DesignWare AHB Central DMA Controller",
+ * (DW_ahb_dmac) which is used with various AMBA 2.0 systems (not all
+ * of which use ARM any more).  See the "Databook" from Synopsys for
+ * information beyond what licensees probably provide.
+ *
+ * The driver has currently been tested only with the Atmel AT32AP7000,
+ * which does not support descriptor writeback.
+ */
+
+/* NOTE:  DMS+SMS is system-specific. We should get this information
+ * from the platform code somehow.
+ */
+#define DWC_DEFAULT_CTLLO	(DWC_CTLL_DST_MSIZE(0)		\
+				| DWC_CTLL_SRC_MSIZE(0)		\
+				| DWC_CTLL_DMS(0)		\
+				| DWC_CTLL_SMS(1)		\
+				| DWC_CTLL_LLP_D_EN		\
+				| DWC_CTLL_LLP_S_EN)
+
+/*
+ * This is configuration-dependent and usually a funny size like 4095.
+ * Let's round it down to the nearest power of two.
+ *
+ * Note that this is a transfer count, i.e. if we transfer 32-bit
+ * words, we can do 8192 bytes per descriptor.
+ *
+ * This parameter is also system-specific.
+ */
+#define DWC_MAX_COUNT	2048U
+
+/*
+ * Number of descriptors to allocate for each channel. This should be
+ * made configurable somehow; preferably, the clients (at least the
+ * ones using slave transfers) should be able to give us a hint.
+ */
+#define NR_DESCS_PER_CHANNEL	64
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Because we're not relying on writeback from the controller (it may not
+ * even be configured into the core!) we don't need to use dma_pool.  These
+ * descriptors -- and associated data -- are cacheable.  We do need to make
+ * sure their dcache entries are written back before handing them off to
+ * the controller, though.
+ */
+
+static struct dw_desc *dwc_first_active(struct dw_dma_chan *dwc)
+{
+	return list_entry(dwc->active_list.next, struct dw_desc, desc_node);
+}
+
+static struct dw_desc *dwc_first_queued(struct dw_dma_chan *dwc)
+{
+	return list_entry(dwc->queue.next, struct dw_desc, desc_node);
+}
+
+static struct dw_desc *dwc_desc_get(struct dw_dma_chan *dwc)
+{
+	struct dw_desc *desc, *_desc;
+	struct dw_desc *ret = NULL;
+	unsigned int i = 0;
+
+	spin_lock_bh(&dwc->lock);
+	list_for_each_entry_safe(desc, _desc, &dwc->free_list, desc_node) {
+		if (async_tx_test_ack(&desc->txd)) {
+			list_del(&desc->desc_node);
+			ret = desc;
+			break;
+		}
+		dev_dbg(&dwc->chan.dev, "desc %p not ACKed\n", desc);
+		i++;
+	}
+	spin_unlock_bh(&dwc->lock);
+
+	dev_vdbg(&dwc->chan.dev, "scanned %u descriptors on freelist\n", i);
+
+	return ret;
+}
+
+static void dwc_sync_desc_for_cpu(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	struct dw_desc	*child;
+
+	list_for_each_entry(child, &desc->txd.tx_list, desc_node)
+		dma_sync_single_for_cpu(dwc->chan.dev.parent,
+				child->txd.phys, sizeof(child->lli),
+				DMA_TO_DEVICE);
+	dma_sync_single_for_cpu(dwc->chan.dev.parent,
+			desc->txd.phys, sizeof(desc->lli),
+			DMA_TO_DEVICE);
+}
+
+/*
+ * Move a descriptor, including any children, to the free list.
+ * `desc' must not be on any lists.
+ */
+static void dwc_desc_put(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	if (desc) {
+		struct dw_desc *child;
+
+		dwc_sync_desc_for_cpu(dwc, desc);
+
+		spin_lock_bh(&dwc->lock);
+		list_for_each_entry(child, &desc->txd.tx_list, desc_node)
+			dev_vdbg(&dwc->chan.dev,
+					"moving child desc %p to freelist\n",
+					child);
+		list_splice_init(&desc->txd.tx_list, &dwc->free_list);
+		dev_vdbg(&dwc->chan.dev, "moving desc %p to freelist\n", desc);
+		list_add(&desc->desc_node, &dwc->free_list);
+		spin_unlock_bh(&dwc->lock);
+	}
+}
+
+/* Called with dwc->lock held and bh disabled */
+static dma_cookie_t
+dwc_assign_cookie(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	dma_cookie_t cookie = dwc->chan.cookie;
+
+	if (++cookie < 0)
+		cookie = 1;
+
+	dwc->chan.cookie = cookie;
+	desc->txd.cookie = cookie;
+
+	return cookie;
+}
+
+/*----------------------------------------------------------------------*/
+
+/* Called with dwc->lock held and bh disabled */
+static void dwc_dostart(struct dw_dma_chan *dwc, struct dw_desc *first)
+{
+	struct dw_dma	*dw = to_dw_dma(dwc->chan.device);
+
+	/* ASSERT:  channel is idle */
+	if (dma_readl(dw, CH_EN) & dwc->mask) {
+		dev_err(&dwc->chan.dev,
+			"BUG: Attempted to start non-idle channel\n");
+		dev_err(&dwc->chan.dev,
+			"  SAR: 0x%x DAR: 0x%x LLP: 0x%x CTL: 0x%x:%08x\n",
+			channel_readl(dwc, SAR),
+			channel_readl(dwc, DAR),
+			channel_readl(dwc, LLP),
+			channel_readl(dwc, CTL_HI),
+			channel_readl(dwc, CTL_LO));
+
+		/* The tasklet will hopefully advance the queue... */
+		return;
+	}
+
+	channel_writel(dwc, LLP, first->txd.phys);
+	channel_writel(dwc, CTL_LO,
+			DWC_CTLL_LLP_D_EN | DWC_CTLL_LLP_S_EN);
+	channel_writel(dwc, CTL_HI, 0);
+	channel_set_bit(dw, CH_EN, dwc->mask);
+}
+
+/*----------------------------------------------------------------------*/
+
+static void
+dwc_descriptor_complete(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	dma_async_tx_callback		callback;
+	void				*param;
+	struct dma_async_tx_descriptor	*txd = &desc->txd;
+
+	dev_vdbg(&dwc->chan.dev, "descriptor %u complete\n", txd->cookie);
+
+	dwc->completed = txd->cookie;
+	callback = txd->callback;
+	param = txd->callback_param;
+
+	dwc_sync_desc_for_cpu(dwc, desc);
+	list_splice_init(&txd->tx_list, &dwc->free_list);
+	list_move(&desc->desc_node, &dwc->free_list);
+
+	/*
+	 * We use dma_unmap_page() regardless of how the buffers were
+	 * mapped before they were submitted...
+	 */
+	if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP))
+		dma_unmap_page(dwc->chan.dev.parent, desc->lli.dar, desc->len,
+				DMA_FROM_DEVICE);
+	if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP))
+		dma_unmap_page(dwc->chan.dev.parent, desc->lli.sar, desc->len,
+				DMA_TO_DEVICE);
+
+	/*
+	 * The API requires that no submissions are done from a
+	 * callback, so we don't need to drop the lock here
+	 */
+	if (callback)
+		callback(param);
+}
+
+static void dwc_complete_all(struct dw_dma *dw, struct dw_dma_chan *dwc)
+{
+	struct dw_desc *desc, *_desc;
+	LIST_HEAD(list);
+
+	if (dma_readl(dw, CH_EN) & dwc->mask) {
+		dev_err(&dwc->chan.dev,
+			"BUG: XFER bit set, but channel not idle!\n");
+
+		/* Try to continue after resetting the channel... */
+		channel_clear_bit(dw, CH_EN, dwc->mask);
+		while (dma_readl(dw, CH_EN) & dwc->mask)
+			cpu_relax();
+	}
+
+	/*
+	 * Submit queued descriptors ASAP, i.e. before we go through
+	 * the completed ones.
+	 */
+	if (!list_empty(&dwc->queue))
+		dwc_dostart(dwc, dwc_first_queued(dwc));
+	list_splice_init(&dwc->active_list, &list);
+	list_splice_init(&dwc->queue, &dwc->active_list);
+
+	list_for_each_entry_safe(desc, _desc, &list, desc_node)
+		dwc_descriptor_complete(dwc, desc);
+}
+
+static void dwc_scan_descriptors(struct dw_dma *dw, struct dw_dma_chan *dwc)
+{
+	dma_addr_t llp;
+	struct dw_desc *desc, *_desc;
+	struct dw_desc *child;
+	u32 status_xfer;
+
+	/*
+	 * Clear block interrupt flag before scanning so that we don't
+	 * miss any, and read LLP before RAW_XFER to ensure it is
+	 * valid if we decide to scan the list.
+	 */
+	dma_writel(dw, CLEAR.BLOCK, dwc->mask);
+	llp = channel_readl(dwc, LLP);
+	status_xfer = dma_readl(dw, RAW.XFER);
+
+	if (status_xfer & dwc->mask) {
+		/* Everything we've submitted is done */
+		dma_writel(dw, CLEAR.XFER, dwc->mask);
+		dwc_complete_all(dw, dwc);
+		return;
+	}
+
+	dev_vdbg(&dwc->chan.dev, "scan_descriptors: llp=0x%x\n", llp);
+
+	list_for_each_entry_safe(desc, _desc, &dwc->active_list, desc_node) {
+		if (desc->lli.llp == llp)
+			/* This one is currently in progress */
+			return;
+
+		list_for_each_entry(child, &desc->txd.tx_list, desc_node)
+			if (child->lli.llp == llp)
+				/* Currently in progress */
+				return;
+
+		/*
+		 * No descriptors so far seem to be in progress, i.e.
+		 * this one must be done.
+		 */
+		dwc_descriptor_complete(dwc, desc);
+	}
+
+	dev_err(&dwc->chan.dev,
+		"BUG: All descriptors done, but channel not idle!\n");
+
+	/* Try to continue after resetting the channel... */
+	channel_clear_bit(dw, CH_EN, dwc->mask);
+	while (dma_readl(dw, CH_EN) & dwc->mask)
+		cpu_relax();
+
+	if (!list_empty(&dwc->queue)) {
+		dwc_dostart(dwc, dwc_first_queued(dwc));
+		list_splice_init(&dwc->queue, &dwc->active_list);
+	}
+}
+
+static void dwc_dump_lli(struct dw_dma_chan *dwc, struct dw_lli *lli)
+{
+	dev_printk(KERN_CRIT, &dwc->chan.dev,
+			"  desc: s0x%x d0x%x l0x%x c0x%x:%x\n",
+			lli->sar, lli->dar, lli->llp,
+			lli->ctlhi, lli->ctllo);
+}
+
+static void dwc_handle_error(struct dw_dma *dw, struct dw_dma_chan *dwc)
+{
+	struct dw_desc *bad_desc;
+	struct dw_desc *child;
+
+	dwc_scan_descriptors(dw, dwc);
+
+	/*
+	 * The descriptor currently at the head of the active list is
+	 * borked. Since we don't have any way to report errors, we'll
+	 * just have to scream loudly and try to carry on.
+	 */
+	bad_desc = dwc_first_active(dwc);
+	list_del_init(&bad_desc->desc_node);
+	list_splice_init(&dwc->queue, dwc->active_list.prev);
+
+	/* Clear the error flag and try to restart the controller */
+	dma_writel(dw, CLEAR.ERROR, dwc->mask);
+	if (!list_empty(&dwc->active_list))
+		dwc_dostart(dwc, dwc_first_active(dwc));
+
+	/*
+	 * KERN_CRITICAL may seem harsh, but since this only happens
+	 * when someone submits a bad physical address in a
+	 * descriptor, we should consider ourselves lucky that the
+	 * controller flagged an error instead of scribbling over
+	 * random memory locations.
+	 */
+	dev_printk(KERN_CRIT, &dwc->chan.dev,
+			"Bad descriptor submitted for DMA!\n");
+	dev_printk(KERN_CRIT, &dwc->chan.dev,
+			"  cookie: %d\n", bad_desc->txd.cookie);
+	dwc_dump_lli(dwc, &bad_desc->lli);
+	list_for_each_entry(child, &bad_desc->txd.tx_list, desc_node)
+		dwc_dump_lli(dwc, &child->lli);
+
+	/* Pretend the descriptor completed successfully */
+	dwc_descriptor_complete(dwc, bad_desc);
+}
+
+static void dw_dma_tasklet(unsigned long data)
+{
+	struct dw_dma *dw = (struct dw_dma *)data;
+	struct dw_dma_chan *dwc;
+	u32 status_block;
+	u32 status_xfer;
+	u32 status_err;
+	int i;
+
+	status_block = dma_readl(dw, RAW.BLOCK);
+	status_xfer = dma_readl(dw, RAW.BLOCK);
+	status_err = dma_readl(dw, RAW.ERROR);
+
+	dev_vdbg(dw->dma.dev, "tasklet: status_block=%x status_err=%x\n",
+			status_block, status_err);
+
+	for (i = 0; i < dw->dma.chancnt; i++) {
+		dwc = &dw->chan[i];
+		spin_lock(&dwc->lock);
+		if (status_err & (1 << i))
+			dwc_handle_error(dw, dwc);
+		else if ((status_block | status_xfer) & (1 << i))
+			dwc_scan_descriptors(dw, dwc);
+		spin_unlock(&dwc->lock);
+	}
+
+	/*
+	 * Re-enable interrupts. Block Complete interrupts are only
+	 * enabled if the INT_EN bit in the descriptor is set. This
+	 * will trigger a scan before the whole list is done.
+	 */
+	channel_set_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_set_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_set_bit(dw, MASK.ERROR, dw->all_chan_mask);
+}
+
+static irqreturn_t dw_dma_interrupt(int irq, void *dev_id)
+{
+	struct dw_dma *dw = dev_id;
+	u32 status;
+
+	dev_vdbg(dw->dma.dev, "interrupt: status=0x%x\n",
+			dma_readl(dw, STATUS_INT));
+
+	/*
+	 * Just disable the interrupts. We'll turn them back on in the
+	 * softirq handler.
+	 */
+	channel_clear_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.ERROR, dw->all_chan_mask);
+
+	status = dma_readl(dw, STATUS_INT);
+	if (status) {
+		dev_err(dw->dma.dev,
+			"BUG: Unexpected interrupts pending: 0x%x\n",
+			status);
+
+		/* Try to recover */
+		channel_clear_bit(dw, MASK.XFER, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.BLOCK, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.SRC_TRAN, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.DST_TRAN, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.ERROR, (1 << 8) - 1);
+	}
+
+	tasklet_schedule(&dw->tasklet);
+
+	return IRQ_HANDLED;
+}
+
+/*----------------------------------------------------------------------*/
+
+static dma_cookie_t dwc_tx_submit(struct dma_async_tx_descriptor *tx)
+{
+	struct dw_desc		*desc = txd_to_dw_desc(tx);
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(tx->chan);
+	dma_cookie_t		cookie;
+
+	spin_lock_bh(&dwc->lock);
+	cookie = dwc_assign_cookie(dwc, desc);
+
+	/*
+	 * REVISIT: We should attempt to chain as many descriptors as
+	 * possible, perhaps even appending to those already submitted
+	 * for DMA. But this is hard to do in a race-free manner.
+	 */
+	if (list_empty(&dwc->active_list)) {
+		dev_vdbg(&tx->chan->dev, "tx_submit: started %u\n",
+				desc->txd.cookie);
+		dwc_dostart(dwc, desc);
+		list_add_tail(&desc->desc_node, &dwc->active_list);
+	} else {
+		dev_vdbg(&tx->chan->dev, "tx_submit: queued %u\n",
+				desc->txd.cookie);
+
+		list_add_tail(&desc->desc_node, &dwc->queue);
+	}
+
+	spin_unlock_bh(&dwc->lock);
+
+	return cookie;
+}
+
+static struct dma_async_tx_descriptor *
+dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
+		size_t len, unsigned long flags)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_desc		*desc;
+	struct dw_desc		*first;
+	struct dw_desc		*prev;
+	size_t			xfer_count;
+	size_t			offset;
+	unsigned int		src_width;
+	unsigned int		dst_width;
+	u32			ctllo;
+
+	dev_vdbg(&chan->dev, "prep_dma_memcpy d0x%x s0x%x l0x%zx f0x%lx\n",
+			dest, src, len, flags);
+
+	if (unlikely(!len)) {
+		dev_dbg(&chan->dev, "prep_dma_memcpy: length is zero!\n");
+		return NULL;
+	}
+
+	/*
+	 * We can be a lot more clever here, but this should take care
+	 * of the most common optimization.
+	 */
+	if (!((src | dest  | len) & 3))
+		src_width = dst_width = 2;
+	else if (!((src | dest | len) & 1))
+		src_width = dst_width = 1;
+	else
+		src_width = dst_width = 0;
+
+	ctllo = DWC_DEFAULT_CTLLO
+			| DWC_CTLL_DST_WIDTH(dst_width)
+			| DWC_CTLL_SRC_WIDTH(src_width)
+			| DWC_CTLL_DST_INC
+			| DWC_CTLL_SRC_INC
+			| DWC_CTLL_FC_M2M;
+	prev = first = NULL;
+
+	for (offset = 0; offset < len; offset += xfer_count << src_width) {
+		xfer_count = min_t(size_t, (len - offset) >> src_width,
+				DWC_MAX_COUNT);
+
+		desc = dwc_desc_get(dwc);
+		if (!desc)
+			goto err_desc_get;
+
+		desc->lli.sar = src + offset;
+		desc->lli.dar = dest + offset;
+		desc->lli.ctllo = ctllo;
+		desc->lli.ctlhi = xfer_count;
+
+		if (!first) {
+			first = desc;
+		} else {
+			prev->lli.llp = desc->txd.phys;
+			dma_sync_single_for_device(chan->dev.parent,
+					prev->txd.phys, sizeof(prev->lli),
+					DMA_TO_DEVICE);
+			list_add_tail(&desc->desc_node,
+					&first->txd.tx_list);
+		}
+		prev = desc;
+	}
+
+
+	if (flags & DMA_PREP_INTERRUPT)
+		/* Trigger interrupt after last block */
+		prev->lli.ctllo |= DWC_CTLL_INT_EN;
+
+	prev->lli.llp = 0;
+	dma_sync_single_for_device(chan->dev.parent,
+			prev->txd.phys, sizeof(prev->lli),
+			DMA_TO_DEVICE);
+
+	first->txd.flags = flags;
+	first->len = len;
+
+	return &first->txd;
+
+err_desc_get:
+	dwc_desc_put(dwc, first);
+	return NULL;
+}
+
+static struct dma_async_tx_descriptor *
+dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
+		unsigned int sg_len, enum dma_data_direction direction,
+		unsigned long flags)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma_slave	*dws = dwc->dws;
+	struct dw_desc		*prev;
+	struct dw_desc		*first;
+	u32			ctllo;
+	dma_addr_t		reg;
+	unsigned int		reg_width;
+	unsigned int		mem_width;
+	unsigned int		i;
+	struct scatterlist	*sg;
+	size_t			total_len = 0;
+
+	dev_vdbg(&chan->dev, "prep_dma_slave\n");
+
+	if (unlikely(!dws || !sg_len))
+		return NULL;
+
+	reg_width = dws->slave.reg_width;
+	prev = first = NULL;
+
+	sg_len = dma_map_sg(chan->dev.parent, sgl, sg_len, direction);
+
+	switch (direction) {
+	case DMA_TO_DEVICE:
+		ctllo = (DWC_DEFAULT_CTLLO
+				| DWC_CTLL_DST_WIDTH(reg_width)
+				| DWC_CTLL_DST_FIX
+				| DWC_CTLL_SRC_INC
+				| DWC_CTLL_FC_M2P);
+		reg = dws->slave.tx_reg;
+		for_each_sg(sgl, sg, sg_len, i) {
+			struct dw_desc	*desc;
+			u32		len;
+			u32		mem;
+
+			desc = dwc_desc_get(dwc);
+			if (!desc) {
+				dev_err(&chan->dev,
+					"not enough descriptors available\n");
+				goto err_desc_get;
+			}
+
+			mem = sg_phys(sg);
+			len = sg_dma_len(sg);
+			mem_width = 2;
+			if (unlikely(mem & 3 || len & 3))
+				mem_width = 0;
+
+			desc->lli.sar = mem;
+			desc->lli.dar = reg;
+			desc->lli.ctllo = ctllo | DWC_CTLL_SRC_WIDTH(mem_width);
+			desc->lli.ctlhi = len >> mem_width;
+
+			if (!first) {
+				first = desc;
+			} else {
+				prev->lli.llp = desc->txd.phys;
+				dma_sync_single_for_device(chan->dev.parent,
+						prev->txd.phys,
+						sizeof(prev->lli),
+						DMA_TO_DEVICE);
+				list_add_tail(&desc->desc_node,
+						&first->txd.tx_list);
+			}
+			prev = desc;
+			total_len += len;
+		}
+		break;
+	case DMA_FROM_DEVICE:
+		ctllo = (DWC_DEFAULT_CTLLO
+				| DWC_CTLL_SRC_WIDTH(reg_width)
+				| DWC_CTLL_DST_INC
+				| DWC_CTLL_SRC_FIX
+				| DWC_CTLL_FC_P2M);
+
+		reg = dws->slave.rx_reg;
+		for_each_sg(sgl, sg, sg_len, i) {
+			struct dw_desc	*desc;
+			u32		len;
+			u32		mem;
+
+			desc = dwc_desc_get(dwc);
+			if (!desc) {
+				dev_err(&chan->dev,
+					"not enough descriptors available\n");
+				goto err_desc_get;
+			}
+
+			mem = sg_phys(sg);
+			len = sg_dma_len(sg);
+			mem_width = 2;
+			if (unlikely(mem & 3 || len & 3))
+				mem_width = 0;
+
+			desc->lli.sar = reg;
+			desc->lli.dar = mem;
+			desc->lli.ctllo = ctllo | DWC_CTLL_DST_WIDTH(mem_width);
+			desc->lli.ctlhi = len >> reg_width;
+
+			if (!first) {
+				first = desc;
+			} else {
+				prev->lli.llp = desc->txd.phys;
+				dma_sync_single_for_device(chan->dev.parent,
+						prev->txd.phys,
+						sizeof(prev->lli),
+						DMA_TO_DEVICE);
+				list_add_tail(&desc->desc_node,
+						&first->txd.tx_list);
+			}
+			prev = desc;
+			total_len += len;
+		}
+		break;
+	default:
+		return NULL;
+	}
+
+	if (flags & DMA_PREP_INTERRUPT)
+		/* Trigger interrupt after last block */
+		prev->lli.ctllo |= DWC_CTLL_INT_EN;
+
+	prev->lli.llp = 0;
+	dma_sync_single_for_device(chan->dev.parent,
+			prev->txd.phys, sizeof(prev->lli),
+			DMA_TO_DEVICE);
+
+	first->len = total_len;
+
+	return &first->txd;
+
+err_desc_get:
+	dwc_desc_put(dwc, first);
+	return NULL;
+}
+
+static void dwc_terminate_all(struct dma_chan *chan)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
+	struct dw_desc		*desc, *_desc;
+	LIST_HEAD(list);
+
+	/*
+	 * This is only called when something went wrong elsewhere, so
+	 * we don't really care about the data. Just disable the
+	 * channel. We still have to poll the channel enable bit due
+	 * to AHB/HSB limitations.
+	 */
+	spin_lock_bh(&dwc->lock);
+
+	channel_clear_bit(dw, CH_EN, dwc->mask);
+
+	while (dma_readl(dw, CH_EN) & dwc->mask)
+		cpu_relax();
+
+	/* active_list entries will end up before queued entries */
+	list_splice_init(&dwc->queue, &list);
+	list_splice_init(&dwc->active_list, &list);
+
+	spin_unlock_bh(&dwc->lock);
+
+	/* Flush all pending and queued descriptors */
+	list_for_each_entry_safe(desc, _desc, &list, desc_node)
+		dwc_descriptor_complete(dwc, desc);
+}
+
+static enum dma_status
+dwc_is_tx_complete(struct dma_chan *chan,
+		dma_cookie_t cookie,
+		dma_cookie_t *done, dma_cookie_t *used)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	dma_cookie_t		last_used;
+	dma_cookie_t		last_complete;
+	int			ret;
+
+	last_complete = dwc->completed;
+	last_used = chan->cookie;
+
+	ret = dma_async_is_complete(cookie, last_complete, last_used);
+	if (ret != DMA_SUCCESS) {
+		dwc_scan_descriptors(to_dw_dma(chan->device), dwc);
+
+		last_complete = dwc->completed;
+		last_used = chan->cookie;
+
+		ret = dma_async_is_complete(cookie, last_complete, last_used);
+	}
+
+	if (done)
+		*done = last_complete;
+	if (used)
+		*used = last_used;
+
+	return ret;
+}
+
+static void dwc_issue_pending(struct dma_chan *chan)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+
+	spin_lock_bh(&dwc->lock);
+	if (!list_empty(&dwc->queue))
+		dwc_scan_descriptors(to_dw_dma(chan->device), dwc);
+	spin_unlock_bh(&dwc->lock);
+}
+
+static int dwc_alloc_chan_resources(struct dma_chan *chan,
+		struct dma_client *client)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
+	struct dw_desc		*desc;
+	struct dma_slave	*slave;
+	struct dw_dma_slave	*dws;
+	int			i;
+	u32			cfghi;
+	u32			cfglo;
+
+	dev_vdbg(&chan->dev, "alloc_chan_resources\n");
+
+	/* Channels doing slave DMA can only handle one client. */
+	if (dwc->dws || client->slave) {
+		if (chan->client_count)
+			return -EBUSY;
+	}
+
+	/* ASSERT:  channel is idle */
+	if (dma_readl(dw, CH_EN) & dwc->mask) {
+		dev_dbg(&chan->dev, "DMA channel not idle?\n");
+		return -EIO;
+	}
+
+	dwc->completed = chan->cookie = 1;
+
+	cfghi = DWC_CFGH_FIFO_MODE;
+	cfglo = 0;
+
+	slave = client->slave;
+	if (slave) {
+		/*
+		 * We need controller-specific data to set up slave
+		 * transfers.
+		 */
+		BUG_ON(!slave->dma_dev || slave->dma_dev != dw->dma.dev);
+
+		dws = container_of(slave, struct dw_dma_slave, slave);
+
+		dwc->dws = dws;
+		cfghi = dws->cfg_hi;
+		cfglo = dws->cfg_lo;
+	} else {
+		dwc->dws = NULL;
+	}
+
+	channel_writel(dwc, CFG_LO, cfglo);
+	channel_writel(dwc, CFG_HI, cfghi);
+
+	/*
+	 * NOTE: some controllers may have additional features that we
+	 * need to initialize here, like "scatter-gather" (which
+	 * doesn't mean what you think it means), and status writeback.
+	 */
+
+	spin_lock_bh(&dwc->lock);
+	i = dwc->descs_allocated;
+	while (dwc->descs_allocated < NR_DESCS_PER_CHANNEL) {
+		spin_unlock_bh(&dwc->lock);
+
+		desc = kzalloc(sizeof(struct dw_desc), GFP_KERNEL);
+		if (!desc) {
+			dev_info(&chan->dev,
+				"only allocated %d descriptors\n", i);
+			spin_lock_bh(&dwc->lock);
+			break;
+		}
+
+		dma_async_tx_descriptor_init(&desc->txd, chan);
+		desc->txd.tx_submit = dwc_tx_submit;
+		desc->txd.flags = DMA_CTRL_ACK;
+		INIT_LIST_HEAD(&desc->txd.tx_list);
+		desc->txd.phys = dma_map_single(chan->dev.parent, &desc->lli,
+				sizeof(desc->lli), DMA_TO_DEVICE);
+		dwc_desc_put(dwc, desc);
+
+		spin_lock_bh(&dwc->lock);
+		i = ++dwc->descs_allocated;
+	}
+
+	/* Enable interrupts */
+	channel_set_bit(dw, MASK.XFER, dwc->mask);
+	channel_set_bit(dw, MASK.BLOCK, dwc->mask);
+	channel_set_bit(dw, MASK.ERROR, dwc->mask);
+
+	spin_unlock_bh(&dwc->lock);
+
+	dev_dbg(&chan->dev,
+		"alloc_chan_resources allocated %d descriptors\n", i);
+
+	return i;
+}
+
+static void dwc_free_chan_resources(struct dma_chan *chan)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
+	struct dw_desc		*desc, *_desc;
+	LIST_HEAD(list);
+
+	dev_dbg(&chan->dev, "free_chan_resources (descs allocated=%u)\n",
+			dwc->descs_allocated);
+
+	/* ASSERT:  channel is idle */
+	BUG_ON(!list_empty(&dwc->active_list));
+	BUG_ON(!list_empty(&dwc->queue));
+	BUG_ON(dma_readl(to_dw_dma(chan->device), CH_EN) & dwc->mask);
+
+	spin_lock_bh(&dwc->lock);
+	list_splice_init(&dwc->free_list, &list);
+	dwc->descs_allocated = 0;
+	dwc->dws = NULL;
+
+	/* Disable interrupts */
+	channel_clear_bit(dw, MASK.XFER, dwc->mask);
+	channel_clear_bit(dw, MASK.BLOCK, dwc->mask);
+	channel_clear_bit(dw, MASK.ERROR, dwc->mask);
+
+	spin_unlock_bh(&dwc->lock);
+
+	list_for_each_entry_safe(desc, _desc, &list, desc_node) {
+		dev_vdbg(&chan->dev, "  freeing descriptor %p\n", desc);
+		dma_unmap_single(chan->dev.parent, desc->txd.phys,
+				sizeof(desc->lli), DMA_TO_DEVICE);
+		kfree(desc);
+	}
+
+	dev_vdbg(&chan->dev, "free_chan_resources done\n");
+}
+
+/*----------------------------------------------------------------------*/
+
+static void dw_dma_off(struct dw_dma *dw)
+{
+	dma_writel(dw, CFG, 0);
+
+	channel_clear_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.SRC_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.DST_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.ERROR, dw->all_chan_mask);
+
+	while (dma_readl(dw, CFG) & DW_CFG_DMA_EN)
+		cpu_relax();
+}
+
+static int __init dw_probe(struct platform_device *pdev)
+{
+	struct dw_dma_platform_data *pdata;
+	struct resource		*io;
+	struct dw_dma		*dw;
+	size_t			size;
+	int			irq;
+	int			err;
+	int			i;
+
+	pdata = pdev->dev.platform_data;
+	if (!pdata || pdata->nr_channels > DW_DMA_MAX_NR_CHANNELS)
+		return -EINVAL;
+
+	io = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!io)
+		return -EINVAL;
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
+		return irq;
+
+	size = sizeof(struct dw_dma);
+	size += pdata->nr_channels * sizeof(struct dw_dma_chan);
+	dw = kzalloc(size, GFP_KERNEL);
+	if (!dw)
+		return -ENOMEM;
+
+	if (!request_mem_region(io->start, DW_REGLEN, pdev->dev.driver->name)) {
+		err = -EBUSY;
+		goto err_kfree;
+	}
+
+	memset(dw, 0, sizeof *dw);
+
+	dw->regs = ioremap(io->start, DW_REGLEN);
+	if (!dw->regs) {
+		err = -ENOMEM;
+		goto err_release_r;
+	}
+
+	dw->clk = clk_get(&pdev->dev, "hclk");
+	if (IS_ERR(dw->clk)) {
+		err = PTR_ERR(dw->clk);
+		goto err_clk;
+	}
+	clk_enable(dw->clk);
+
+	/* force dma off, just in case */
+	dw_dma_off(dw);
+
+	err = request_irq(irq, dw_dma_interrupt, 0, "dw_dmac", dw);
+	if (err)
+		goto err_irq;
+
+	platform_set_drvdata(pdev, dw);
+
+	tasklet_init(&dw->tasklet, dw_dma_tasklet, (unsigned long)dw);
+
+	dw->all_chan_mask = (1 << pdata->nr_channels) - 1;
+
+	INIT_LIST_HEAD(&dw->dma.channels);
+	for (i = 0; i < pdata->nr_channels; i++, dw->dma.chancnt++) {
+		struct dw_dma_chan	*dwc = &dw->chan[i];
+
+		dwc->chan.device = &dw->dma;
+		dwc->chan.cookie = dwc->completed = 1;
+		dwc->chan.chan_id = i;
+		list_add_tail(&dwc->chan.device_node, &dw->dma.channels);
+
+		dwc->ch_regs = &__dw_regs(dw)->CHAN[i];
+		spin_lock_init(&dwc->lock);
+		dwc->mask = 1 << i;
+
+		INIT_LIST_HEAD(&dwc->active_list);
+		INIT_LIST_HEAD(&dwc->queue);
+		INIT_LIST_HEAD(&dwc->free_list);
+
+		channel_clear_bit(dw, CH_EN, dwc->mask);
+	}
+
+	/* Clear/disable all interrupts on all channels. */
+	dma_writel(dw, CLEAR.XFER, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.BLOCK, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.SRC_TRAN, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.DST_TRAN, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.ERROR, dw->all_chan_mask);
+
+	channel_clear_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.SRC_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.DST_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.ERROR, dw->all_chan_mask);
+
+	dma_cap_set(DMA_MEMCPY, dw->dma.cap_mask);
+	dma_cap_set(DMA_SLAVE, dw->dma.cap_mask);
+	dw->dma.dev = &pdev->dev;
+	dw->dma.device_alloc_chan_resources = dwc_alloc_chan_resources;
+	dw->dma.device_free_chan_resources = dwc_free_chan_resources;
+
+	dw->dma.device_prep_dma_memcpy = dwc_prep_dma_memcpy;
+
+	dw->dma.device_prep_slave_sg = dwc_prep_slave_sg;
+	dw->dma.device_terminate_all = dwc_terminate_all;
+
+	dw->dma.device_is_tx_complete = dwc_is_tx_complete;
+	dw->dma.device_issue_pending = dwc_issue_pending;
+
+	dma_writel(dw, CFG, DW_CFG_DMA_EN);
+
+	printk(KERN_INFO "%s: DesignWare DMA Controller, %d channels\n",
+			pdev->dev.bus_id, dw->dma.chancnt);
+
+	dma_async_device_register(&dw->dma);
+
+	return 0;
+
+err_irq:
+	clk_disable(dw->clk);
+	clk_put(dw->clk);
+err_clk:
+	iounmap(dw->regs);
+	dw->regs = NULL;
+err_release_r:
+	release_resource(io);
+err_kfree:
+	kfree(dw);
+	return err;
+}
+
+static int __exit dw_remove(struct platform_device *pdev)
+{
+	struct dw_dma		*dw = platform_get_drvdata(pdev);
+	struct dw_dma_chan	*dwc, *_dwc;
+	struct resource		*io;
+
+	dw_dma_off(dw);
+	dma_async_device_unregister(&dw->dma);
+
+	free_irq(platform_get_irq(pdev, 0), dw);
+	tasklet_kill(&dw->tasklet);
+
+	list_for_each_entry_safe(dwc, _dwc, &dw->dma.channels,
+			chan.device_node) {
+		list_del(&dwc->chan.device_node);
+		channel_clear_bit(dw, CH_EN, dwc->mask);
+	}
+
+	clk_disable(dw->clk);
+	clk_put(dw->clk);
+
+	iounmap(dw->regs);
+	dw->regs = NULL;
+
+	io = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(io->start, DW_REGLEN);
+
+	kfree(dw);
+
+	return 0;
+}
+
+static void dw_shutdown(struct platform_device *pdev)
+{
+	struct dw_dma	*dw = platform_get_drvdata(pdev);
+
+	dw_dma_off(platform_get_drvdata(pdev));
+	clk_disable(dw->clk);
+}
+
+static int dw_suspend_late(struct platform_device *pdev, pm_message_t mesg)
+{
+	struct dw_dma	*dw = platform_get_drvdata(pdev);
+
+	dw_dma_off(platform_get_drvdata(pdev));
+	clk_disable(dw->clk);
+	return 0;
+}
+
+static int dw_resume_early(struct platform_device *pdev)
+{
+	struct dw_dma	*dw = platform_get_drvdata(pdev);
+
+	clk_enable(dw->clk);
+	dma_writel(dw, CFG, DW_CFG_DMA_EN);
+	return 0;
+
+}
+
+static struct platform_driver dw_driver = {
+	.remove		= __exit_p(dw_remove),
+	.shutdown	= dw_shutdown,
+	.suspend_late	= dw_suspend_late,
+	.resume_early	= dw_resume_early,
+	.driver = {
+		.name	= "dw_dmac",
+	},
+};
+
+static int __init dw_init(void)
+{
+	return platform_driver_probe(&dw_driver, dw_probe);
+}
+module_init(dw_init);
+
+static void __exit dw_exit(void)
+{
+	platform_driver_unregister(&dw_driver);
+}
+module_exit(dw_exit);
+
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("Synopsys DesignWare DMA Controller driver");
+MODULE_AUTHOR("Haavard Skinnemoen <haavard.skinnemoen@atmel.com>");
diff --git a/drivers/dma/dw_dmac_regs.h b/drivers/dma/dw_dmac_regs.h
new file mode 100644
index 000000000000..00fdd187bb0c
--- /dev/null
+++ b/drivers/dma/dw_dmac_regs.h
@@ -0,0 +1,225 @@
+/*
+ * Driver for the Synopsys DesignWare AHB DMA Controller
+ *
+ * Copyright (C) 2005-2007 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/dw_dmac.h>
+
+#define DW_DMA_MAX_NR_CHANNELS	8
+
+/*
+ * Redefine this macro to handle differences between 32- and 64-bit
+ * addressing, big vs. little endian, etc.
+ */
+#define DW_REG(name)		u32 name; u32 __pad_##name
+
+/* Hardware register definitions. */
+struct dw_dma_chan_regs {
+	DW_REG(SAR);		/* Source Address Register */
+	DW_REG(DAR);		/* Destination Address Register */
+	DW_REG(LLP);		/* Linked List Pointer */
+	u32	CTL_LO;		/* Control Register Low */
+	u32	CTL_HI;		/* Control Register High */
+	DW_REG(SSTAT);
+	DW_REG(DSTAT);
+	DW_REG(SSTATAR);
+	DW_REG(DSTATAR);
+	u32	CFG_LO;		/* Configuration Register Low */
+	u32	CFG_HI;		/* Configuration Register High */
+	DW_REG(SGR);
+	DW_REG(DSR);
+};
+
+struct dw_dma_irq_regs {
+	DW_REG(XFER);
+	DW_REG(BLOCK);
+	DW_REG(SRC_TRAN);
+	DW_REG(DST_TRAN);
+	DW_REG(ERROR);
+};
+
+struct dw_dma_regs {
+	/* per-channel registers */
+	struct dw_dma_chan_regs	CHAN[DW_DMA_MAX_NR_CHANNELS];
+
+	/* irq handling */
+	struct dw_dma_irq_regs	RAW;		/* r */
+	struct dw_dma_irq_regs	STATUS;		/* r (raw & mask) */
+	struct dw_dma_irq_regs	MASK;		/* rw (set = irq enabled) */
+	struct dw_dma_irq_regs	CLEAR;		/* w (ack, affects "raw") */
+
+	DW_REG(STATUS_INT);			/* r */
+
+	/* software handshaking */
+	DW_REG(REQ_SRC);
+	DW_REG(REQ_DST);
+	DW_REG(SGL_REQ_SRC);
+	DW_REG(SGL_REQ_DST);
+	DW_REG(LAST_SRC);
+	DW_REG(LAST_DST);
+
+	/* miscellaneous */
+	DW_REG(CFG);
+	DW_REG(CH_EN);
+	DW_REG(ID);
+	DW_REG(TEST);
+
+	/* optional encoded params, 0x3c8..0x3 */
+};
+
+/* Bitfields in CTL_LO */
+#define DWC_CTLL_INT_EN		(1 << 0)	/* irqs enabled? */
+#define DWC_CTLL_DST_WIDTH(n)	((n)<<1)	/* bytes per element */
+#define DWC_CTLL_SRC_WIDTH(n)	((n)<<4)
+#define DWC_CTLL_DST_INC	(0<<7)		/* DAR update/not */
+#define DWC_CTLL_DST_DEC	(1<<7)
+#define DWC_CTLL_DST_FIX	(2<<7)
+#define DWC_CTLL_SRC_INC	(0<<7)		/* SAR update/not */
+#define DWC_CTLL_SRC_DEC	(1<<9)
+#define DWC_CTLL_SRC_FIX	(2<<9)
+#define DWC_CTLL_DST_MSIZE(n)	((n)<<11)	/* burst, #elements */
+#define DWC_CTLL_SRC_MSIZE(n)	((n)<<14)
+#define DWC_CTLL_S_GATH_EN	(1 << 17)	/* src gather, !FIX */
+#define DWC_CTLL_D_SCAT_EN	(1 << 18)	/* dst scatter, !FIX */
+#define DWC_CTLL_FC_M2M		(0 << 20)	/* mem-to-mem */
+#define DWC_CTLL_FC_M2P		(1 << 20)	/* mem-to-periph */
+#define DWC_CTLL_FC_P2M		(2 << 20)	/* periph-to-mem */
+#define DWC_CTLL_FC_P2P		(3 << 20)	/* periph-to-periph */
+/* plus 4 transfer types for peripheral-as-flow-controller */
+#define DWC_CTLL_DMS(n)		((n)<<23)	/* dst master select */
+#define DWC_CTLL_SMS(n)		((n)<<25)	/* src master select */
+#define DWC_CTLL_LLP_D_EN	(1 << 27)	/* dest block chain */
+#define DWC_CTLL_LLP_S_EN	(1 << 28)	/* src block chain */
+
+/* Bitfields in CTL_HI */
+#define DWC_CTLH_DONE		0x00001000
+#define DWC_CTLH_BLOCK_TS_MASK	0x00000fff
+
+/* Bitfields in CFG_LO. Platform-configurable bits are in <linux/dw_dmac.h> */
+#define DWC_CFGL_CH_SUSP	(1 << 8)	/* pause xfer */
+#define DWC_CFGL_FIFO_EMPTY	(1 << 9)	/* pause xfer */
+#define DWC_CFGL_HS_DST		(1 << 10)	/* handshake w/dst */
+#define DWC_CFGL_HS_SRC		(1 << 11)	/* handshake w/src */
+#define DWC_CFGL_MAX_BURST(x)	((x) << 20)
+#define DWC_CFGL_RELOAD_SAR	(1 << 30)
+#define DWC_CFGL_RELOAD_DAR	(1 << 31)
+
+/* Bitfields in CFG_HI. Platform-configurable bits are in <linux/dw_dmac.h> */
+#define DWC_CFGH_DS_UPD_EN	(1 << 5)
+#define DWC_CFGH_SS_UPD_EN	(1 << 6)
+
+/* Bitfields in SGR */
+#define DWC_SGR_SGI(x)		((x) << 0)
+#define DWC_SGR_SGC(x)		((x) << 20)
+
+/* Bitfields in DSR */
+#define DWC_DSR_DSI(x)		((x) << 0)
+#define DWC_DSR_DSC(x)		((x) << 20)
+
+/* Bitfields in CFG */
+#define DW_CFG_DMA_EN		(1 << 0)
+
+#define DW_REGLEN		0x400
+
+struct dw_dma_chan {
+	struct dma_chan		chan;
+	void __iomem		*ch_regs;
+	u8			mask;
+
+	spinlock_t		lock;
+
+	/* these other elements are all protected by lock */
+	dma_cookie_t		completed;
+	struct list_head	active_list;
+	struct list_head	queue;
+	struct list_head	free_list;
+
+	struct dw_dma_slave	*dws;
+
+	unsigned int		descs_allocated;
+};
+
+static inline struct dw_dma_chan_regs __iomem *
+__dwc_regs(struct dw_dma_chan *dwc)
+{
+	return dwc->ch_regs;
+}
+
+#define channel_readl(dwc, name) \
+	__raw_readl(&(__dwc_regs(dwc)->name))
+#define channel_writel(dwc, name, val) \
+	__raw_writel((val), &(__dwc_regs(dwc)->name))
+
+static inline struct dw_dma_chan *to_dw_dma_chan(struct dma_chan *chan)
+{
+	return container_of(chan, struct dw_dma_chan, chan);
+}
+
+
+struct dw_dma {
+	struct dma_device	dma;
+	void __iomem		*regs;
+	struct tasklet_struct	tasklet;
+	struct clk		*clk;
+
+	u8			all_chan_mask;
+
+	struct dw_dma_chan	chan[0];
+};
+
+static inline struct dw_dma_regs __iomem *__dw_regs(struct dw_dma *dw)
+{
+	return dw->regs;
+}
+
+#define dma_readl(dw, name) \
+	__raw_readl(&(__dw_regs(dw)->name))
+#define dma_writel(dw, name, val) \
+	__raw_writel((val), &(__dw_regs(dw)->name))
+
+#define channel_set_bit(dw, reg, mask) \
+	dma_writel(dw, reg, ((mask) << 8) | (mask))
+#define channel_clear_bit(dw, reg, mask) \
+	dma_writel(dw, reg, ((mask) << 8) | 0)
+
+static inline struct dw_dma *to_dw_dma(struct dma_device *ddev)
+{
+	return container_of(ddev, struct dw_dma, dma);
+}
+
+/* LLI == Linked List Item; a.k.a. DMA block descriptor */
+struct dw_lli {
+	/* values that are not changed by hardware */
+	dma_addr_t	sar;
+	dma_addr_t	dar;
+	dma_addr_t	llp;		/* chain to next lli */
+	u32		ctllo;
+	/* values that may get written back: */
+	u32		ctlhi;
+	/* sstat and dstat can snapshot peripheral register state.
+	 * silicon config may discard either or both...
+	 */
+	u32		sstat;
+	u32		dstat;
+};
+
+struct dw_desc {
+	/* FIRST values the hardware uses */
+	struct dw_lli			lli;
+
+	/* THEN values for driver housekeeping */
+	struct list_head		desc_node;
+	struct dma_async_tx_descriptor	txd;
+	size_t				len;
+};
+
+static inline struct dw_desc *
+txd_to_dw_desc(struct dma_async_tx_descriptor *txd)
+{
+	return container_of(txd, struct dw_desc, txd);
+}
diff --git a/include/asm-avr32/arch-at32ap/at32ap700x.h b/include/asm-avr32/arch-at32ap/at32ap700x.h
index 31e48b0e7324..d18a3053be0d 100644
--- a/include/asm-avr32/arch-at32ap/at32ap700x.h
+++ b/include/asm-avr32/arch-at32ap/at32ap700x.h
@@ -30,4 +30,20 @@
 #define GPIO_PIN_PD(N)	(GPIO_PIOD_BASE + (N))
 #define GPIO_PIN_PE(N)	(GPIO_PIOE_BASE + (N))
 
+
+/*
+ * DMAC peripheral hardware handshaking interfaces, used with dw_dmac
+ */
+#define DMAC_MCI_RX		0
+#define DMAC_MCI_TX		1
+#define DMAC_DAC_TX		2
+#define DMAC_AC97_A_RX		3
+#define DMAC_AC97_A_TX		4
+#define DMAC_AC97_B_RX		5
+#define DMAC_AC97_B_TX		6
+#define DMAC_DMAREQ_0		7
+#define DMAC_DMAREQ_1		8
+#define DMAC_DMAREQ_2		9
+#define DMAC_DMAREQ_3		10
+
 #endif /* __ASM_ARCH_AT32AP700X_H__ */
diff --git a/include/linux/dw_dmac.h b/include/linux/dw_dmac.h
new file mode 100644
index 000000000000..04d217b442bf
--- /dev/null
+++ b/include/linux/dw_dmac.h
@@ -0,0 +1,62 @@
+/*
+ * Driver for the Synopsys DesignWare DMA Controller (aka DMACA on
+ * AVR32 systems.)
+ *
+ * Copyright (C) 2007 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef DW_DMAC_H
+#define DW_DMAC_H
+
+#include <linux/dmaengine.h>
+
+/**
+ * struct dw_dma_platform_data - Controller configuration parameters
+ * @nr_channels: Number of channels supported by hardware (max 8)
+ */
+struct dw_dma_platform_data {
+	unsigned int	nr_channels;
+};
+
+/**
+ * struct dw_dma_slave - Controller-specific information about a slave
+ * @slave: Generic information about the slave
+ * @ctl_lo: Platform-specific initializer for the CTL_LO register
+ * @cfg_hi: Platform-specific initializer for the CFG_HI register
+ * @cfg_lo: Platform-specific initializer for the CFG_LO register
+ */
+struct dw_dma_slave {
+	struct dma_slave	slave;
+	u32			cfg_hi;
+	u32			cfg_lo;
+};
+
+/* Platform-configurable bits in CFG_HI */
+#define DWC_CFGH_FCMODE		(1 << 0)
+#define DWC_CFGH_FIFO_MODE	(1 << 1)
+#define DWC_CFGH_PROTCTL(x)	((x) << 2)
+#define DWC_CFGH_SRC_PER(x)	((x) << 7)
+#define DWC_CFGH_DST_PER(x)	((x) << 11)
+
+/* Platform-configurable bits in CFG_LO */
+#define DWC_CFGL_PRIO(x)	((x) << 5)	/* priority */
+#define DWC_CFGL_LOCK_CH_XFER	(0 << 12)	/* scope of LOCK_CH */
+#define DWC_CFGL_LOCK_CH_BLOCK	(1 << 12)
+#define DWC_CFGL_LOCK_CH_XACT	(2 << 12)
+#define DWC_CFGL_LOCK_BUS_XFER	(0 << 14)	/* scope of LOCK_BUS */
+#define DWC_CFGL_LOCK_BUS_BLOCK	(1 << 14)
+#define DWC_CFGL_LOCK_BUS_XACT	(2 << 14)
+#define DWC_CFGL_LOCK_CH	(1 << 15)	/* channel lockout */
+#define DWC_CFGL_LOCK_BUS	(1 << 16)	/* busmaster lockout */
+#define DWC_CFGL_HS_DST_POL	(1 << 18)	/* dst handshake active low */
+#define DWC_CFGL_HS_SRC_POL	(1 << 19)	/* src handshake active low */
+
+static inline struct dw_dma_slave *to_dw_dma_slave(struct dma_slave *slave)
+{
+	return container_of(slave, struct dw_dma_slave, slave);
+}
+
+#endif /* DW_DMAC_H */
-- 
cgit v1.2.3


From 11a100f844f6096787ab20e19f17d72abc957a8f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 8 Jul 2008 15:36:57 -0700
Subject: vlan: avoid header copying and linearisation where possible

- vlan_dev_reorder_header() is only called on the receive path after
  calling skb_share_check(). This means we can use skb_cow() since
  all we need is a writable header.

- vlan_dev_hard_header() includes a work-around for some apparently
  broken out of tree MPLS code. The hard_header functions can expect
  to always have a headroom of at least there own hard_header_len
  available, so the reallocation check is unnecessary.

- __vlan_put_tag() can use skb_cow_head() to avoid the skb_unshare()
  copy when the header is writable.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 18 +++---------------
 net/8021q/vlan_dev.c    | 34 +++++-----------------------------
 2 files changed, 8 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index d36515dae62f..93f5d9b0e9f9 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -185,22 +185,10 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
 {
 	struct vlan_ethhdr *veth;
 
-	if (skb_headroom(skb) < VLAN_HLEN) {
-		struct sk_buff *sk_tmp = skb;
-		skb = skb_realloc_headroom(sk_tmp, VLAN_HLEN);
-		kfree_skb(sk_tmp);
-		if (!skb) {
-			printk(KERN_ERR "vlan: failed to realloc headroom\n");
-			return NULL;
-		}
-	} else {
-		skb = skb_unshare(skb, GFP_ATOMIC);
-		if (!skb) {
-			printk(KERN_ERR "vlan: failed to unshare skbuff\n");
-			return NULL;
-		}
+	if (skb_cow_head(skb, VLAN_HLEN) < 0) {
+		kfree_skb(skb);
+		return NULL;
 	}
-
 	veth = (struct vlan_ethhdr *)skb_push(skb, VLAN_HLEN);
 
 	/* Move the mac addresses to the beginning of the new header. */
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 2ccac6bea57e..b6e52c025fd8 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -74,11 +74,8 @@ static int vlan_dev_rebuild_header(struct sk_buff *skb)
 static inline struct sk_buff *vlan_check_reorder_header(struct sk_buff *skb)
 {
 	if (vlan_dev_info(skb->dev)->flags & VLAN_FLAG_REORDER_HDR) {
-		if (skb_shared(skb) || skb_cloned(skb)) {
-			struct sk_buff *nskb = skb_copy(skb, GFP_ATOMIC);
-			kfree_skb(skb);
-			skb = nskb;
-		}
+		if (skb_cow(skb, skb_headroom(skb)) < 0)
+			skb = NULL;
 		if (skb) {
 			/* Lifted from Gleb's VLAN code... */
 			memmove(skb->data - ETH_HLEN,
@@ -262,12 +259,14 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 	u16 vlan_tci = 0;
 	int rc = 0;
 	int build_vlan_header = 0;
-	struct net_device *vdev = dev;
 
 	pr_debug("%s: skb: %p type: %hx len: %u vlan_id: %hx, daddr: %p\n",
 		 __func__, skb, type, len, vlan_dev_info(dev)->vlan_id,
 		 daddr);
 
+	if (WARN_ON(skb_headroom(skb) < dev->hard_header_len))
+		return -ENOSPC;
+
 	/* build vlan header only if re_order_header flag is NOT set.  This
 	 * fixes some programs that get confused when they see a VLAN device
 	 * sending a frame that is VLAN encoded (the consensus is that the VLAN
@@ -316,29 +315,6 @@ static int vlan_dev_hard_header(struct sk_buff *skb, struct net_device *dev,
 
 	dev = vlan_dev_info(dev)->real_dev;
 
-	/* MPLS can send us skbuffs w/out enough space.	This check will grow
-	 * the skb if it doesn't have enough headroom. Not a beautiful solution,
-	 * so I'll tick a counter so that users can know it's happening...
-	 * If they care...
-	 */
-
-	/* NOTE: This may still break if the underlying device is not the final
-	 * device (and thus there are more headers to add...) It should work for
-	 * good-ole-ethernet though.
-	 */
-	if (skb_headroom(skb) < dev->hard_header_len) {
-		struct sk_buff *sk_tmp = skb;
-		skb = skb_realloc_headroom(sk_tmp, dev->hard_header_len);
-		kfree_skb(sk_tmp);
-		if (skb == NULL) {
-			struct net_device_stats *stats = &vdev->stats;
-			stats->tx_dropped++;
-			return -ENOMEM;
-		}
-		vlan_dev_info(vdev)->cnt_inc_headroom_on_tx++;
-		pr_debug("%s: %s: had to grow skb\n", __func__, vdev->name);
-	}
-
 	if (build_vlan_header) {
 		/* Now make the underlying real hard header */
 		rc = dev_hard_header(skb, dev, ETH_P_8021Q, daddr, saddr,
-- 
cgit v1.2.3


From bb949fbd1878973c3539d9aecff52f284482a937 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 16:55:56 -0700
Subject: netdev: Create netdev_queue abstraction.

A netdev_queue is an entity managed by a qdisc.

Currently there is one RX and one TX queue, and a netdev_queue merely
contains a backpointer to the net_device.

The Qdisc struct is augmented with a netdev_queue pointer as well.

Eventually the 'dev' Qdisc member will go away and we will have the
resulting hierarchy:

	net_device --> netdev_queue --> Qdisc

Also, qdisc_alloc() and qdisc_create_dflt() now take a netdev_queue
pointer argument.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  7 +++++++
 include/net/sch_generic.h |  6 +++++-
 net/core/dev.c            |  8 ++++++++
 net/mac80211/wme.c        |  6 ++++--
 net/sched/sch_api.c       | 12 +++++++-----
 net/sched/sch_atm.c       |  6 ++++--
 net/sched/sch_cbq.c       |  9 ++++++---
 net/sched/sch_dsmark.c    |  6 ++++--
 net/sched/sch_fifo.c      |  3 ++-
 net/sched/sch_generic.c   | 14 ++++++++++----
 net/sched/sch_hfsc.c      |  9 ++++++---
 net/sched/sch_htb.c       | 11 +++++++----
 net/sched/sch_netem.c     |  3 ++-
 net/sched/sch_prio.c      |  3 ++-
 14 files changed, 74 insertions(+), 29 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e009c6fbf5cd..515fd25bf0fc 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -448,6 +448,10 @@ static inline void napi_synchronize(const struct napi_struct *n)
 # define napi_synchronize(n)	barrier()
 #endif
 
+struct netdev_queue {
+	struct net_device	*dev;
+};
+
 /*
  *	The DEVICE structure.
  *	Actually, this whole structure is a big mistake.  It mixes I/O
@@ -624,6 +628,9 @@ struct net_device
 
 	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
 
+	struct netdev_queue	rx_queue;
+	struct netdev_queue	tx_queue;
+
 	/* ingress path synchronizer */
 	spinlock_t		ingress_lock;
 	struct Qdisc		*qdisc_ingress;
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 073f2580b83b..0ab53c575f87 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -37,6 +37,7 @@ struct Qdisc
 	u32			parent;
 	atomic_t		refcnt;
 	struct sk_buff_head	q;
+	struct netdev_queue	*dev_queue;
 	struct net_device	*dev;
 	struct list_head	list;
 
@@ -216,8 +217,11 @@ extern void dev_deactivate(struct net_device *dev);
 extern void qdisc_reset(struct Qdisc *qdisc);
 extern void qdisc_destroy(struct Qdisc *qdisc);
 extern void qdisc_tree_decrease_qlen(struct Qdisc *qdisc, unsigned int n);
-extern struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops);
+extern struct Qdisc *qdisc_alloc(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 struct Qdisc_ops *ops);
 extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
+				       struct netdev_queue *dev_queue,
 				       struct Qdisc_ops *ops, u32 parentid);
 extern void tcf_destroy(struct tcf_proto *tp);
 extern void tcf_destroy_chain(struct tcf_proto **fl);
diff --git a/net/core/dev.c b/net/core/dev.c
index 75933932463d..9b281c906eb0 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4072,6 +4072,12 @@ static struct net_device_stats *internal_stats(struct net_device *dev)
 	return &dev->stats;
 }
 
+static void netdev_init_queues(struct net_device *dev)
+{
+	dev->rx_queue.dev = dev;
+	dev->tx_queue.dev = dev;
+}
+
 /**
  *	alloc_netdev_mq - allocate network device
  *	@sizeof_priv:	size of private data to allocate space for
@@ -4124,6 +4130,8 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 	dev->egress_subqueue_count = queue_count;
 	dev->gso_max_size = GSO_MAX_SIZE;
 
+	netdev_init_queues(dev);
+
 	dev->get_stats = internal_stats;
 	netpoll_netdev_init(dev);
 	setup(dev);
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 5c666f7eda8f..770f1c09b793 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -359,7 +359,8 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt)
 	/* create child queues */
 	for (i = 0; i < QD_NUM(hw); i++) {
 		skb_queue_head_init(&q->requeued[i]);
-		q->queues[i] = qdisc_create_dflt(qd->dev, &pfifo_qdisc_ops,
+		q->queues[i] = qdisc_create_dflt(qd->dev, qd->dev_queue,
+						 &pfifo_qdisc_ops,
 						 qd->handle);
 		if (!q->queues[i]) {
 			q->queues[i] = &noop_qdisc;
@@ -575,7 +576,8 @@ void ieee80211_install_qdisc(struct net_device *dev)
 {
 	struct Qdisc *qdisc;
 
-	qdisc = qdisc_create_dflt(dev, &wme_qdisc_ops, TC_H_ROOT);
+	qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+				  &wme_qdisc_ops, TC_H_ROOT);
 	if (!qdisc) {
 		printk(KERN_ERR "%s: qdisc installation failed\n", dev->name);
 		return;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 69e918bb4278..b86c98bd06a3 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -552,8 +552,8 @@ static int qdisc_graft(struct net_device *dev, struct Qdisc *parent,
  */
 
 static struct Qdisc *
-qdisc_create(struct net_device *dev, u32 parent, u32 handle,
-	   struct nlattr **tca, int *errp)
+qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
+	     u32 parent, u32 handle, struct nlattr **tca, int *errp)
 {
 	int err;
 	struct nlattr *kind = tca[TCA_KIND];
@@ -593,7 +593,7 @@ qdisc_create(struct net_device *dev, u32 parent, u32 handle,
 	if (ops == NULL)
 		goto err_out;
 
-	sch = qdisc_alloc(dev, ops);
+	sch = qdisc_alloc(dev, dev_queue, ops);
 	if (IS_ERR(sch)) {
 		err = PTR_ERR(sch);
 		goto err_out2;
@@ -892,10 +892,12 @@ create_n_graft:
 	if (!(n->nlmsg_flags&NLM_F_CREATE))
 		return -ENOENT;
 	if (clid == TC_H_INGRESS)
-		q = qdisc_create(dev, tcm->tcm_parent, tcm->tcm_parent,
+		q = qdisc_create(dev, &dev->rx_queue,
+				 tcm->tcm_parent, tcm->tcm_parent,
 				 tca, &err);
 	else
-		q = qdisc_create(dev, tcm->tcm_parent, tcm->tcm_handle,
+		q = qdisc_create(dev, &dev->tx_queue,
+				 tcm->tcm_parent, tcm->tcm_handle,
 				 tca, &err);
 	if (q == NULL) {
 		if (err == -EAGAIN)
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index db0e23ae85f8..3dddab531d5a 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -296,7 +296,8 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
 		goto err_out;
 	}
 	flow->filter_list = NULL;
-	flow->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
+	flow->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				    &pfifo_qdisc_ops, classid);
 	if (!flow->q)
 		flow->q = &noop_qdisc;
 	pr_debug("atm_tc_change: qdisc %p\n", flow->q);
@@ -555,7 +556,8 @@ static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt)
 
 	pr_debug("atm_tc_init(sch %p,[qdisc %p],opt %p)\n", sch, p, opt);
 	p->flows = &p->link;
-	p->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, sch->handle);
+	p->link.q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				      &pfifo_qdisc_ops, sch->handle);
 	if (!p->link.q)
 		p->link.q = &noop_qdisc;
 	pr_debug("atm_tc_init: link (%p) qdisc %p\n", &p->link, p->link.q);
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 968b4c73c9c1..d360dcd0818b 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1401,7 +1401,8 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt)
 	q->link.sibling = &q->link;
 	q->link.common.classid = sch->handle;
 	q->link.qdisc = sch;
-	if (!(q->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+	if (!(q->link.q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					    &pfifo_qdisc_ops,
 					    sch->handle)))
 		q->link.q = &noop_qdisc;
 
@@ -1645,7 +1646,8 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	if (cl) {
 		if (new == NULL) {
-			new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+			new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+						&pfifo_qdisc_ops,
 						cl->common.classid);
 			if (new == NULL)
 				return -ENOBUFS;
@@ -1877,7 +1879,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 	cl->R_tab = rtab;
 	rtab = NULL;
 	cl->refcnt = 1;
-	if (!(cl->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid)))
+	if (!(cl->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					&pfifo_qdisc_ops, classid)))
 		cl->q = &noop_qdisc;
 	cl->common.classid = classid;
 	cl->tparent = parent;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index c4c1317cd47d..c955ba24e5cf 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -60,7 +60,8 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg,
 		sch, p, new, old);
 
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+		new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					&pfifo_qdisc_ops,
 					sch->handle);
 		if (new == NULL)
 			new = &noop_qdisc;
@@ -390,7 +391,8 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt)
 	p->default_index = default_index;
 	p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]);
 
-	p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, sch->handle);
+	p->q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				 &pfifo_qdisc_ops, sch->handle);
 	if (p->q == NULL)
 		p->q = &noop_qdisc;
 
diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c
index 82d7d7bbbb16..779eae85faf0 100644
--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -137,7 +137,8 @@ struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops,
 	struct Qdisc *q;
 	int err = -ENOMEM;
 
-	q = qdisc_create_dflt(sch->dev, ops, TC_H_MAKE(sch->handle, 1));
+	q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+			      ops, TC_H_MAKE(sch->handle, 1));
 	if (q) {
 		err = fifo_set_limit(q, limit);
 		if (err < 0) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 13afa7214392..d97086480893 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -440,7 +440,9 @@ static struct Qdisc_ops pfifo_fast_ops __read_mostly = {
 	.owner		=	THIS_MODULE,
 };
 
-struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops)
+struct Qdisc *qdisc_alloc(struct net_device *dev,
+			  struct netdev_queue *dev_queue,
+			  struct Qdisc_ops *ops)
 {
 	void *p;
 	struct Qdisc *sch;
@@ -462,6 +464,7 @@ struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops)
 	sch->ops = ops;
 	sch->enqueue = ops->enqueue;
 	sch->dequeue = ops->dequeue;
+	sch->dev_queue = dev_queue;
 	sch->dev = dev;
 	dev_hold(dev);
 	atomic_set(&sch->refcnt, 1);
@@ -471,12 +474,14 @@ errout:
 	return ERR_PTR(err);
 }
 
-struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops,
+struct Qdisc * qdisc_create_dflt(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 struct Qdisc_ops *ops,
 				 unsigned int parentid)
 {
 	struct Qdisc *sch;
 
-	sch = qdisc_alloc(dev, ops);
+	sch = qdisc_alloc(dev, dev_queue, ops);
 	if (IS_ERR(sch))
 		goto errout;
 	sch->stats_lock = &dev->queue_lock;
@@ -545,7 +550,8 @@ void dev_activate(struct net_device *dev)
 	if (dev->qdisc_sleeping == &noop_qdisc) {
 		struct Qdisc *qdisc;
 		if (dev->tx_queue_len) {
-			qdisc = qdisc_create_dflt(dev, &pfifo_fast_ops,
+			qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+						  &pfifo_fast_ops,
 						  TC_H_ROOT);
 			if (qdisc == NULL) {
 				printk(KERN_INFO "%s: activation failed\n", dev->name);
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 3a8267246a4f..5a22fec4eadd 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1083,7 +1083,8 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	cl->refcnt    = 1;
 	cl->sched     = q;
 	cl->cl_parent = parent;
-	cl->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
+	cl->qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				      &pfifo_qdisc_ops, classid);
 	if (cl->qdisc == NULL)
 		cl->qdisc = &noop_qdisc;
 	INIT_LIST_HEAD(&cl->children);
@@ -1201,7 +1202,8 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (cl->level > 0)
 		return -EINVAL;
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+		new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					&pfifo_qdisc_ops,
 					cl->cl_common.classid);
 		if (new == NULL)
 			new = &noop_qdisc;
@@ -1443,7 +1445,8 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt)
 	q->root.cl_common.classid = sch->handle;
 	q->root.refcnt  = 1;
 	q->root.sched   = q;
-	q->root.qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+	q->root.qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					  &pfifo_qdisc_ops,
 					  sch->handle);
 	if (q->root.qdisc == NULL)
 		q->root.qdisc = &noop_qdisc;
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index ee8b4ffe110c..956a67f66b9c 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1129,7 +1129,8 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	if (cl && !cl->level) {
 		if (new == NULL &&
-		    (new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+		    (new = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					     &pfifo_qdisc_ops,
 					     cl->common.classid))
 		    == NULL)
 			return -ENOBUFS;
@@ -1256,8 +1257,9 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 		return -EBUSY;
 
 	if (!cl->level && htb_parent_last_child(cl)) {
-		new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
-						cl->parent->common.classid);
+		new_q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					  &pfifo_qdisc_ops,
+					  cl->parent->common.classid);
 		last_child = 1;
 	}
 
@@ -1376,7 +1378,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		/* create leaf qdisc early because it uses kmalloc(GFP_KERNEL)
 		   so that can't be used inside of sch_tree_lock
 		   -- thanks to Karlis Peisenieks */
-		new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
+		new_q = qdisc_create_dflt(sch->dev, sch->dev_queue,
+					  &pfifo_qdisc_ops, classid);
 		sch_tree_lock(sch);
 		if (parent && !parent->level) {
 			unsigned int qlen = parent->un.leaf.q->q.qlen;
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 24697667247c..aa7a04e32ae9 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -536,7 +536,8 @@ static int netem_init(struct Qdisc *sch, struct nlattr *opt)
 
 	qdisc_watchdog_init(&q->watchdog, sch);
 
-	q->qdisc = qdisc_create_dflt(sch->dev, &tfifo_qdisc_ops,
+	q->qdisc = qdisc_create_dflt(sch->dev, sch->dev_queue,
+				     &tfifo_qdisc_ops,
 				     TC_H_MAKE(sch->handle, 1));
 	if (!q->qdisc) {
 		pr_debug("netem: qdisc create failed\n");
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 5532f1031ab5..ca58a039208e 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -281,7 +281,8 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 	for (i=0; i<q->bands; i++) {
 		if (q->queues[i] == &noop_qdisc) {
 			struct Qdisc *child;
-			child = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+			child = qdisc_create_dflt(sch->dev, sch->dev_queue,
+						  &pfifo_qdisc_ops,
 						  TC_H_MAKE(sch->handle, i + 1));
 			if (child) {
 				sch_tree_lock(sch);
-- 
cgit v1.2.3


From dc2b48475a0a36f8b3bbb2da60d3a006dc5c2c84 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 17:18:23 -0700
Subject: netdev: Move queue_lock into struct netdev_queue.

The lock is now an attribute of the device queue.

One thing to notice is that "suspicious" places
emerge which will need specific training about
multiple queue handling.  They are so marked with
explicit "netdev->rx_queue" and "netdev->tx_queue"
references.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c         |  8 ++++----
 include/linux/netdevice.h |  4 ++--
 net/core/dev.c            | 33 ++++++++++++++++++++++-----------
 net/mac80211/main.c       | 10 +++++-----
 net/mac80211/wme.c        |  2 +-
 net/sched/sch_api.c       |  2 +-
 net/sched/sch_cbq.c       |  8 ++++----
 net/sched/sch_generic.c   | 40 ++++++++++++++++++++--------------------
 net/sched/sch_hfsc.c      |  4 ++--
 net/sched/sch_htb.c       | 16 ++++++++--------
 net/sched/sch_netem.c     |  4 ++--
 net/sched/sch_teql.c      |  4 ++--
 12 files changed, 73 insertions(+), 62 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index af233b591534..bc3de272a829 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -229,12 +229,12 @@ module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
 /*
- * dev_ifb->queue_lock is usually taken after dev->ingress_lock,
+ * dev_ifb->tx_queue.lock is usually taken after dev->ingress_lock,
  * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev->queue_lock with dev_ifb->ingress_lock.
+ * ifb doesn't take dev->tx_queue.lock with dev_ifb->ingress_lock.
  * But lockdep should know that ifb has different locks from dev.
  */
-static struct lock_class_key ifb_queue_lock_key;
+static struct lock_class_key ifb_tx_queue_lock_key;
 static struct lock_class_key ifb_ingress_lock_key;
 
 
@@ -258,7 +258,7 @@ static int __init ifb_init_one(int index)
 	if (err < 0)
 		goto err;
 
-	lockdep_set_class(&dev_ifb->queue_lock, &ifb_queue_lock_key);
+	lockdep_set_class(&dev_ifb->tx_queue.lock, &ifb_tx_queue_lock_key);
 	lockdep_set_class(&dev_ifb->ingress_lock, &ifb_ingress_lock_key);
 
 	return 0;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 515fd25bf0fc..e835acacb479 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -449,6 +449,7 @@ static inline void napi_synchronize(const struct napi_struct *n)
 #endif
 
 struct netdev_queue {
+	spinlock_t		lock;
 	struct net_device	*dev;
 };
 
@@ -629,7 +630,7 @@ struct net_device
 	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
 
 	struct netdev_queue	rx_queue;
-	struct netdev_queue	tx_queue;
+	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
 
 	/* ingress path synchronizer */
 	spinlock_t		ingress_lock;
@@ -639,7 +640,6 @@ struct net_device
  * Cache line mostly used on queue transmit path (qdisc)
  */
 	/* device queue lock */
-	spinlock_t		queue_lock ____cacheline_aligned_in_smp;
 	struct Qdisc		*qdisc;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
diff --git a/net/core/dev.c b/net/core/dev.c
index 9b281c906eb0..05011048b86c 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1667,6 +1667,7 @@ out_kfree_skb:
 int dev_queue_xmit(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
+	struct netdev_queue *txq;
 	struct Qdisc *q;
 	int rc = -ENOMEM;
 
@@ -1699,14 +1700,15 @@ int dev_queue_xmit(struct sk_buff *skb)
 	}
 
 gso:
-	spin_lock_prefetch(&dev->queue_lock);
+	txq = &dev->tx_queue;
+	spin_lock_prefetch(&txq->lock);
 
 	/* Disable soft irqs for various locks below. Also
 	 * stops preemption for RCU.
 	 */
 	rcu_read_lock_bh();
 
-	/* Updates of qdisc are serialized by queue_lock.
+	/* Updates of qdisc are serialized by queue->lock.
 	 * The struct Qdisc which is pointed to by qdisc is now a
 	 * rcu structure - it may be accessed without acquiring
 	 * a lock (but the structure may be stale.) The freeing of the
@@ -1714,7 +1716,7 @@ gso:
 	 * more references to it.
 	 *
 	 * If the qdisc has an enqueue function, we still need to
-	 * hold the queue_lock before calling it, since queue_lock
+	 * hold the queue->lock before calling it, since queue->lock
 	 * also serializes access to the device queue.
 	 */
 
@@ -1724,19 +1726,19 @@ gso:
 #endif
 	if (q->enqueue) {
 		/* Grab device queue */
-		spin_lock(&dev->queue_lock);
+		spin_lock(&txq->lock);
 		q = dev->qdisc;
 		if (q->enqueue) {
 			/* reset queue_mapping to zero */
 			skb_set_queue_mapping(skb, 0);
 			rc = q->enqueue(skb, q);
 			qdisc_run(dev);
-			spin_unlock(&dev->queue_lock);
+			spin_unlock(&txq->lock);
 
 			rc = rc == NET_XMIT_BYPASS ? NET_XMIT_SUCCESS : rc;
 			goto out;
 		}
-		spin_unlock(&dev->queue_lock);
+		spin_unlock(&txq->lock);
 	}
 
 	/* The device has no queue. Common case for software devices:
@@ -1919,14 +1921,17 @@ static void net_tx_action(struct softirq_action *h)
 
 		while (head) {
 			struct net_device *dev = head;
+			struct netdev_queue *txq;
 			head = head->next_sched;
 
+			txq = &dev->tx_queue;
+
 			smp_mb__before_clear_bit();
 			clear_bit(__LINK_STATE_SCHED, &dev->state);
 
-			if (spin_trylock(&dev->queue_lock)) {
+			if (spin_trylock(&txq->lock)) {
 				qdisc_run(dev);
-				spin_unlock(&dev->queue_lock);
+				spin_unlock(&txq->lock);
 			} else {
 				netif_schedule(dev);
 			}
@@ -3787,7 +3792,6 @@ int register_netdevice(struct net_device *dev)
 	BUG_ON(!dev_net(dev));
 	net = dev_net(dev);
 
-	spin_lock_init(&dev->queue_lock);
 	spin_lock_init(&dev->_xmit_lock);
 	netdev_set_lockdep_class(&dev->_xmit_lock, dev->type);
 	dev->xmit_lock_owner = -1;
@@ -4072,10 +4076,17 @@ static struct net_device_stats *internal_stats(struct net_device *dev)
 	return &dev->stats;
 }
 
+static void netdev_init_one_queue(struct net_device *dev,
+				  struct netdev_queue *queue)
+{
+	spin_lock_init(&queue->lock);
+	queue->dev = dev;
+}
+
 static void netdev_init_queues(struct net_device *dev)
 {
-	dev->rx_queue.dev = dev;
-	dev->tx_queue.dev = dev;
+	netdev_init_one_queue(dev, &dev->rx_queue);
+	netdev_init_one_queue(dev, &dev->tx_queue);
 }
 
 /**
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index cf477ad39dac..12aeaf78ae75 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -636,7 +636,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 
 	/* ensure that TX flow won't interrupt us
 	 * until the end of the call to requeue function */
-	spin_lock_bh(&local->mdev->queue_lock);
+	spin_lock_bh(&local->mdev->tx_queue.lock);
 
 	/* create a new queue for this aggregation */
 	ret = ieee80211_ht_agg_queue_add(local, sta, tid);
@@ -675,7 +675,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 
 	/* Will put all the packets in the new SW queue */
 	ieee80211_requeue(local, ieee802_1d_to_ac[tid]);
-	spin_unlock_bh(&local->mdev->queue_lock);
+	spin_unlock_bh(&local->mdev->tx_queue.lock);
 	spin_unlock_bh(&sta->lock);
 
 	/* send an addBA request */
@@ -701,7 +701,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 err_unlock_queue:
 	kfree(sta->ampdu_mlme.tid_tx[tid]);
 	sta->ampdu_mlme.tid_tx[tid] = NULL;
-	spin_unlock_bh(&local->mdev->queue_lock);
+	spin_unlock_bh(&local->mdev->tx_queue.lock);
 	ret = -EBUSY;
 err_unlock_sta:
 	spin_unlock_bh(&sta->lock);
@@ -875,10 +875,10 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 
 	/* avoid ordering issues: we are the only one that can modify
 	 * the content of the qdiscs */
-	spin_lock_bh(&local->mdev->queue_lock);
+	spin_lock_bh(&local->mdev->tx_queue.lock);
 	/* remove the queue for this aggregation */
 	ieee80211_ht_agg_queue_remove(local, sta, tid, 1);
-	spin_unlock_bh(&local->mdev->queue_lock);
+	spin_unlock_bh(&local->mdev->tx_queue.lock);
 
 	/* we just requeued the all the frames that were in the removed
 	 * queue, and since we might miss a softirq we do netif_schedule.
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 2fbc171130bf..59ed9cae66b9 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -648,7 +648,7 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 }
 
 /**
- * the caller needs to hold local->mdev->queue_lock
+ * the caller needs to hold local->mdev->tx_queue.lock
  */
 void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   struct sta_info *sta, u16 tid,
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 1f893082a4f6..2a1834f8c7d8 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -606,7 +606,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 		sch->stats_lock = &dev->ingress_lock;
 		handle = TC_H_MAKE(TC_H_INGRESS, 0);
 	} else {
-		sch->stats_lock = &dev->queue_lock;
+		sch->stats_lock = &dev_queue->lock;
 		if (handle == 0) {
 			handle = qdisc_alloc_handle(dev);
 			err = -ENOMEM;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 9f2ace585fd6..99ce3da2b0a4 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1746,10 +1746,10 @@ static void cbq_put(struct Qdisc *sch, unsigned long arg)
 #ifdef CONFIG_NET_CLS_ACT
 		struct cbq_sched_data *q = qdisc_priv(sch);
 
-		spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+		spin_lock_bh(&sch->dev_queue->lock);
 		if (q->rx_class == cl)
 			q->rx_class = NULL;
-		spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+		spin_unlock_bh(&sch->dev_queue->lock);
 #endif
 
 		cbq_destroy_class(sch, cl);
@@ -1828,7 +1828,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &qdisc_dev(sch)->queue_lock,
+					      &sch->dev_queue->lock,
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1919,7 +1919,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &qdisc_dev(sch)->queue_lock, tca[TCA_RATE]);
+				  &sch->dev_queue->lock, tca[TCA_RATE]);
 
 	*arg = (unsigned long)cl;
 	return 0;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index b626a4f32b6b..ee8f9f78a095 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -29,31 +29,31 @@
 /* Main transmission queue. */
 
 /* Modifications to data participating in scheduling must be protected with
- * dev->queue_lock spinlock.
+ * queue->lock spinlock.
  *
  * The idea is the following:
  * - enqueue, dequeue are serialized via top level device
- *   spinlock dev->queue_lock.
+ *   spinlock queue->lock.
  * - ingress filtering is serialized via top level device
  *   spinlock dev->ingress_lock.
  * - updates to tree and tree walking are only done under the rtnl mutex.
  */
 
 void qdisc_lock_tree(struct net_device *dev)
-	__acquires(dev->queue_lock)
+	__acquires(dev->tx_queue.lock)
 	__acquires(dev->ingress_lock)
 {
-	spin_lock_bh(&dev->queue_lock);
+	spin_lock_bh(&dev->tx_queue.lock);
 	spin_lock(&dev->ingress_lock);
 }
 EXPORT_SYMBOL(qdisc_lock_tree);
 
 void qdisc_unlock_tree(struct net_device *dev)
 	__releases(dev->ingress_lock)
-	__releases(dev->queue_lock)
+	__releases(dev->tx_queue.lock)
 {
 	spin_unlock(&dev->ingress_lock);
-	spin_unlock_bh(&dev->queue_lock);
+	spin_unlock_bh(&dev->tx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_unlock_tree);
 
@@ -118,15 +118,15 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 }
 
 /*
- * NOTE: Called under dev->queue_lock with locally disabled BH.
+ * NOTE: Called under queue->lock with locally disabled BH.
  *
  * __LINK_STATE_QDISC_RUNNING guarantees only one CPU can process this
- * device at a time. dev->queue_lock serializes queue accesses for
+ * device at a time. queue->lock serializes queue accesses for
  * this device AND dev->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
- *  dev->queue_lock and netif_tx_lock are mutually exclusive,
+ *  queue->lock and netif_tx_lock are mutually exclusive,
  *  if one is grabbed, another must be free.
  *
  * Note, that this procedure can be called by a watchdog timer
@@ -148,14 +148,14 @@ static inline int qdisc_restart(struct net_device *dev)
 
 
 	/* And release queue */
-	spin_unlock(&dev->queue_lock);
+	spin_unlock(&q->dev_queue->lock);
 
 	HARD_TX_LOCK(dev, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
 		ret = dev_hard_start_xmit(skb, dev);
 	HARD_TX_UNLOCK(dev);
 
-	spin_lock(&dev->queue_lock);
+	spin_lock(&q->dev_queue->lock);
 	q = dev->qdisc;
 
 	switch (ret) {
@@ -482,7 +482,7 @@ struct Qdisc * qdisc_create_dflt(struct net_device *dev,
 	sch = qdisc_alloc(dev_queue, ops);
 	if (IS_ERR(sch))
 		goto errout;
-	sch->stats_lock = &dev->queue_lock;
+	sch->stats_lock = &dev_queue->lock;
 	sch->parent = parentid;
 
 	if (!ops->init || ops->init(sch, NULL) == 0)
@@ -494,7 +494,7 @@ errout:
 }
 EXPORT_SYMBOL(qdisc_create_dflt);
 
-/* Under dev->queue_lock and BH! */
+/* Under queue->lock and BH! */
 
 void qdisc_reset(struct Qdisc *qdisc)
 {
@@ -514,7 +514,7 @@ static void __qdisc_destroy(struct rcu_head *head)
 	kfree((char *) qdisc - qdisc->padded);
 }
 
-/* Under dev->queue_lock and BH! */
+/* Under queue->lock and BH! */
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
@@ -566,13 +566,13 @@ void dev_activate(struct net_device *dev)
 		/* Delay activation until next carrier-on event */
 		return;
 
-	spin_lock_bh(&dev->queue_lock);
+	spin_lock_bh(&dev->tx_queue.lock);
 	rcu_assign_pointer(dev->qdisc, dev->qdisc_sleeping);
 	if (dev->qdisc != &noqueue_qdisc) {
 		dev->trans_start = jiffies;
 		dev_watchdog_up(dev);
 	}
-	spin_unlock_bh(&dev->queue_lock);
+	spin_unlock_bh(&dev->tx_queue.lock);
 }
 
 void dev_deactivate(struct net_device *dev)
@@ -581,7 +581,7 @@ void dev_deactivate(struct net_device *dev)
 	struct sk_buff *skb;
 	int running;
 
-	spin_lock_bh(&dev->queue_lock);
+	spin_lock_bh(&dev->tx_queue.lock);
 	qdisc = dev->qdisc;
 	dev->qdisc = &noop_qdisc;
 
@@ -589,7 +589,7 @@ void dev_deactivate(struct net_device *dev)
 
 	skb = dev->gso_skb;
 	dev->gso_skb = NULL;
-	spin_unlock_bh(&dev->queue_lock);
+	spin_unlock_bh(&dev->tx_queue.lock);
 
 	kfree_skb(skb);
 
@@ -607,9 +607,9 @@ void dev_deactivate(struct net_device *dev)
 		 * Double-check inside queue lock to ensure that all effects
 		 * of the queue run are visible when we return.
 		 */
-		spin_lock_bh(&dev->queue_lock);
+		spin_lock_bh(&dev->tx_queue.lock);
 		running = test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state);
-		spin_unlock_bh(&dev->queue_lock);
+		spin_unlock_bh(&dev->tx_queue.lock);
 
 		/*
 		 * The running flag should never be set at this point because
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 333525422f45..997d520ca580 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1045,7 +1045,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &qdisc_dev(sch)->queue_lock,
+					      &sch->dev_queue->lock,
 					      tca[TCA_RATE]);
 		return 0;
 	}
@@ -1104,7 +1104,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 
 	if (tca[TCA_RATE])
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &qdisc_dev(sch)->queue_lock, tca[TCA_RATE]);
+				  &sch->dev_queue->lock, tca[TCA_RATE]);
 	*arg = (unsigned long)cl;
 	return 0;
 }
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 31f7d1536e6d..c8ca54cc26b0 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1043,7 +1043,7 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	struct nlattr *nest;
 	struct tc_htb_glob gopt;
 
-	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_lock_bh(&sch->dev_queue->lock);
 
 	gopt.direct_pkts = q->direct_pkts;
 	gopt.version = HTB_VER;
@@ -1057,11 +1057,11 @@ static int htb_dump(struct Qdisc *sch, struct sk_buff *skb)
 	NLA_PUT(skb, TCA_HTB_INIT, sizeof(gopt), &gopt);
 	nla_nest_end(skb, nest);
 
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1073,7 +1073,7 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	struct nlattr *nest;
 	struct tc_htb_opt opt;
 
-	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_lock_bh(&sch->dev_queue->lock);
 	tcm->tcm_parent = cl->parent ? cl->parent->common.classid : TC_H_ROOT;
 	tcm->tcm_handle = cl->common.classid;
 	if (!cl->level && cl->un.leaf.q)
@@ -1095,11 +1095,11 @@ static int htb_dump_class(struct Qdisc *sch, unsigned long arg,
 	NLA_PUT(skb, TCA_HTB_PARMS, sizeof(opt), &opt);
 
 	nla_nest_end(skb, nest);
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	return skb->len;
 
 nla_put_failure:
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 	nla_nest_cancel(skb, nest);
 	return -1;
 }
@@ -1365,7 +1365,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 			goto failure;
 
 		gen_new_estimator(&cl->bstats, &cl->rate_est,
-				  &qdisc_dev(sch)->queue_lock,
+				  &sch->dev_queue->lock,
 				  tca[TCA_RATE] ? : &est.nla);
 		cl->refcnt = 1;
 		cl->children = 0;
@@ -1420,7 +1420,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 	} else {
 		if (tca[TCA_RATE])
 			gen_replace_estimator(&cl->bstats, &cl->rate_est,
-					      &qdisc_dev(sch)->queue_lock,
+					      &sch->dev_queue->lock,
 					      tca[TCA_RATE]);
 		sch_tree_lock(sch);
 	}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 790582960444..71b73c528f9b 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -333,9 +333,9 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
 	for (i = 0; i < n; i++)
 		d->table[i] = data[i];
 
-	spin_lock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_lock_bh(&sch->dev_queue->lock);
 	d = xchg(&q->delay_dist, d);
-	spin_unlock_bh(&qdisc_dev(sch)->queue_lock);
+	spin_unlock_bh(&sch->dev_queue->lock);
 
 	kfree(d);
 	return 0;
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index b3fc82623fc6..4f3054e8e1ab 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -154,9 +154,9 @@ teql_destroy(struct Qdisc* sch)
 					master->slaves = NEXT_SLAVE(q);
 					if (q == master->slaves) {
 						master->slaves = NULL;
-						spin_lock_bh(&master->dev->queue_lock);
+						spin_lock_bh(&master->dev->tx_queue.lock);
 						qdisc_reset(master->dev->qdisc);
-						spin_unlock_bh(&master->dev->queue_lock);
+						spin_unlock_bh(&master->dev->tx_queue.lock);
 					}
 				}
 				skb_queue_purge(&dat->q);
-- 
cgit v1.2.3


From 555353cfa1aee293de445bfa6de43276138ddd82 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 17:33:13 -0700
Subject: netdev: The ingress_lock member is no longer needed.

Every qdisc is assosciated with a queue, and in the case of ingress
qdiscs that will now be netdev->rx_queue so using that queue's lock is
the thing to do.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c         |  8 ++++----
 include/linux/netdevice.h |  2 --
 net/core/dev.c            | 12 +++++++-----
 net/sched/sch_api.c       |  3 +--
 net/sched/sch_generic.c   | 10 +++++-----
 5 files changed, 17 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index bc3de272a829..ccbd6554f6eb 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -229,13 +229,13 @@ module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
 /*
- * dev_ifb->tx_queue.lock is usually taken after dev->ingress_lock,
+ * dev_ifb->tx_queue.lock is usually taken after dev->rx_queue.lock,
  * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev->tx_queue.lock with dev_ifb->ingress_lock.
+ * ifb doesn't take dev->tx_queue.lock with dev_ifb->rx_queue.lock.
  * But lockdep should know that ifb has different locks from dev.
  */
 static struct lock_class_key ifb_tx_queue_lock_key;
-static struct lock_class_key ifb_ingress_lock_key;
+static struct lock_class_key ifb_rx_queue_lock_key;
 
 
 static int __init ifb_init_one(int index)
@@ -259,7 +259,7 @@ static int __init ifb_init_one(int index)
 		goto err;
 
 	lockdep_set_class(&dev_ifb->tx_queue.lock, &ifb_tx_queue_lock_key);
-	lockdep_set_class(&dev_ifb->ingress_lock, &ifb_ingress_lock_key);
+	lockdep_set_class(&dev_ifb->rx_queue.lock, &ifb_rx_queue_lock_key);
 
 	return 0;
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e835acacb479..633a44c6fa5e 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -632,8 +632,6 @@ struct net_device
 	struct netdev_queue	rx_queue;
 	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
 
-	/* ingress path synchronizer */
-	spinlock_t		ingress_lock;
 	struct Qdisc		*qdisc_ingress;
 
 /*
diff --git a/net/core/dev.c b/net/core/dev.c
index 05011048b86c..2322fb69fd53 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2014,10 +2014,11 @@ static inline struct sk_buff *handle_macvlan(struct sk_buff *skb,
  */
 static int ing_filter(struct sk_buff *skb)
 {
-	struct Qdisc *q;
 	struct net_device *dev = skb->dev;
-	int result = TC_ACT_OK;
 	u32 ttl = G_TC_RTTL(skb->tc_verd);
+	struct netdev_queue *rxq;
+	int result = TC_ACT_OK;
+	struct Qdisc *q;
 
 	if (MAX_RED_LOOP < ttl++) {
 		printk(KERN_WARNING
@@ -2029,10 +2030,12 @@ static int ing_filter(struct sk_buff *skb)
 	skb->tc_verd = SET_TC_RTTL(skb->tc_verd, ttl);
 	skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS);
 
-	spin_lock(&dev->ingress_lock);
+	rxq = &dev->rx_queue;
+
+	spin_lock(&rxq->lock);
 	if ((q = dev->qdisc_ingress) != NULL)
 		result = q->enqueue(skb, q);
-	spin_unlock(&dev->ingress_lock);
+	spin_unlock(&rxq->lock);
 
 	return result;
 }
@@ -3795,7 +3798,6 @@ int register_netdevice(struct net_device *dev)
 	spin_lock_init(&dev->_xmit_lock);
 	netdev_set_lockdep_class(&dev->_xmit_lock, dev->type);
 	dev->xmit_lock_owner = -1;
-	spin_lock_init(&dev->ingress_lock);
 
 	dev->iflink = -1;
 
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 2a1834f8c7d8..570cef2a9c5f 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -601,12 +601,11 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 
 	sch->parent = parent;
 
+	sch->stats_lock = &dev_queue->lock;
 	if (handle == TC_H_INGRESS) {
 		sch->flags |= TCQ_F_INGRESS;
-		sch->stats_lock = &dev->ingress_lock;
 		handle = TC_H_MAKE(TC_H_INGRESS, 0);
 	} else {
-		sch->stats_lock = &dev_queue->lock;
 		if (handle == 0) {
 			handle = qdisc_alloc_handle(dev);
 			err = -ENOMEM;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index ee8f9f78a095..804d44b00348 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -35,24 +35,24 @@
  * - enqueue, dequeue are serialized via top level device
  *   spinlock queue->lock.
  * - ingress filtering is serialized via top level device
- *   spinlock dev->ingress_lock.
+ *   spinlock dev->rx_queue.lock.
  * - updates to tree and tree walking are only done under the rtnl mutex.
  */
 
 void qdisc_lock_tree(struct net_device *dev)
 	__acquires(dev->tx_queue.lock)
-	__acquires(dev->ingress_lock)
+	__acquires(dev->rx_queue.lock)
 {
 	spin_lock_bh(&dev->tx_queue.lock);
-	spin_lock(&dev->ingress_lock);
+	spin_lock(&dev->rx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_lock_tree);
 
 void qdisc_unlock_tree(struct net_device *dev)
-	__releases(dev->ingress_lock)
+	__releases(dev->rx_queue.lock)
 	__releases(dev->tx_queue.lock)
 {
-	spin_unlock(&dev->ingress_lock);
+	spin_unlock(&dev->rx_queue.lock);
 	spin_unlock_bh(&dev->tx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_unlock_tree);
-- 
cgit v1.2.3


From b0e1e6462df3c5944010b3328a546d8fe5d932cd Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 17:42:10 -0700
Subject: netdev: Move rest of qdisc state into struct netdev_queue

Now qdisc, qdisc_sleeping, and qdisc_list also live there.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/isdn/i4l/isdn_net.c    |  2 +-
 include/linux/netdevice.h      | 10 ++---
 include/net/irda/irda_device.h |  2 +-
 net/core/dev.c                 |  4 +-
 net/core/link_watch.c          |  8 +++-
 net/core/rtnetlink.c           |  6 ++-
 net/ipv6/addrconf.c            |  3 +-
 net/mac80211/wme.c             | 20 ++++++----
 net/sched/cls_api.c            |  7 +++-
 net/sched/sch_api.c            | 34 ++++++++++------
 net/sched/sch_generic.c        | 90 ++++++++++++++++++++++++++----------------
 net/sched/sch_netem.c          |  2 +-
 net/sched/sch_teql.c           | 14 ++++---
 13 files changed, 125 insertions(+), 77 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c
index ef1a300068dc..457bbd119f9b 100644
--- a/drivers/isdn/i4l/isdn_net.c
+++ b/drivers/isdn/i4l/isdn_net.c
@@ -287,7 +287,7 @@ isdn_net_unbind_channel(isdn_net_local * lp)
 		   BEWARE! This chunk of code cannot be called from hardware
 		   interrupt handler. I hope it is true. --ANK
 		 */
-		qdisc_reset(lp->netdev->dev->qdisc);
+		qdisc_reset(lp->netdev->dev->tx_queue.qdisc);
 	}
 	lp->dialstate = 0;
 	dev->rx_netdev[isdn_dc2minor(lp->isdn_device, lp->isdn_channel)] = NULL;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 633a44c6fa5e..df702a7b3db5 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -451,6 +451,9 @@ static inline void napi_synchronize(const struct napi_struct *n)
 struct netdev_queue {
 	spinlock_t		lock;
 	struct net_device	*dev;
+	struct Qdisc		*qdisc;
+	struct Qdisc		*qdisc_sleeping;
+	struct list_head	qdisc_list;
 };
 
 /*
@@ -634,13 +637,6 @@ struct net_device
 
 	struct Qdisc		*qdisc_ingress;
 
-/*
- * Cache line mostly used on queue transmit path (qdisc)
- */
-	/* device queue lock */
-	struct Qdisc		*qdisc;
-	struct Qdisc		*qdisc_sleeping;
-	struct list_head	qdisc_list;
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 	/* Partially transmitted GSO packet. */
diff --git a/include/net/irda/irda_device.h b/include/net/irda/irda_device.h
index f70e9b39ebaf..16fbf672e0b2 100644
--- a/include/net/irda/irda_device.h
+++ b/include/net/irda/irda_device.h
@@ -223,7 +223,7 @@ int  irda_device_is_receiving(struct net_device *dev);
 /* Interface for internal use */
 static inline int irda_device_txqueue_empty(const struct net_device *dev)
 {
-	return skb_queue_empty(&dev->qdisc->q);
+	return skb_queue_empty(&dev->tx_queue.qdisc->q);
 }
 int  irda_device_set_raw_mode(struct net_device* self, int status);
 struct net_device *alloc_irdadev(int sizeof_priv);
diff --git a/net/core/dev.c b/net/core/dev.c
index 2322fb69fd53..ce79c28d739d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1720,14 +1720,14 @@ gso:
 	 * also serializes access to the device queue.
 	 */
 
-	q = rcu_dereference(dev->qdisc);
+	q = rcu_dereference(txq->qdisc);
 #ifdef CONFIG_NET_CLS_ACT
 	skb->tc_verd = SET_TC_AT(skb->tc_verd,AT_EGRESS);
 #endif
 	if (q->enqueue) {
 		/* Grab device queue */
 		spin_lock(&txq->lock);
-		q = dev->qdisc;
+		q = txq->qdisc;
 		if (q->enqueue) {
 			/* reset queue_mapping to zero */
 			skb_set_queue_mapping(skb, 0);
diff --git a/net/core/link_watch.c b/net/core/link_watch.c
index a5e372b9ec4d..50218218445b 100644
--- a/net/core/link_watch.c
+++ b/net/core/link_watch.c
@@ -79,8 +79,10 @@ static void rfc2863_policy(struct net_device *dev)
 
 static int linkwatch_urgent_event(struct net_device *dev)
 {
+	struct netdev_queue *txq = &dev->tx_queue;
+
 	return netif_running(dev) && netif_carrier_ok(dev) &&
-	       dev->qdisc != dev->qdisc_sleeping;
+	       txq->qdisc != txq->qdisc_sleeping;
 }
 
 
@@ -181,7 +183,9 @@ static void __linkwatch_run_queue(int urgent_only)
 		rfc2863_policy(dev);
 		if (dev->flags & IFF_UP) {
 			if (netif_carrier_ok(dev)) {
-				WARN_ON(dev->qdisc_sleeping == &noop_qdisc);
+				struct netdev_queue *txq = &dev->tx_queue;
+
+				WARN_ON(txq->qdisc_sleeping == &noop_qdisc);
 				dev_activate(dev);
 			} else
 				dev_deactivate(dev);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 6c8d7f0ea01a..8ef9f1db610e 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -605,6 +605,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
 			    int type, u32 pid, u32 seq, u32 change,
 			    unsigned int flags)
 {
+	struct netdev_queue *txq;
 	struct ifinfomsg *ifm;
 	struct nlmsghdr *nlh;
 	struct net_device_stats *stats;
@@ -635,8 +636,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
 	if (dev->master)
 		NLA_PUT_U32(skb, IFLA_MASTER, dev->master->ifindex);
 
-	if (dev->qdisc_sleeping)
-		NLA_PUT_STRING(skb, IFLA_QDISC, dev->qdisc_sleeping->ops->id);
+	txq = &dev->tx_queue;
+	if (txq->qdisc_sleeping)
+		NLA_PUT_STRING(skb, IFLA_QDISC, txq->qdisc_sleeping->ops->id);
 
 	if (1) {
 		struct rtnl_link_ifmap map = {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8572cb05fc21..5c84c798331d 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -231,7 +231,8 @@ const struct in6_addr in6addr_linklocal_allrouters = IN6ADDR_LINKLOCAL_ALLROUTER
 /* Check if a valid qdisc is available */
 static inline int addrconf_qdisc_ok(struct net_device *dev)
 {
-	return (dev->qdisc != &noop_qdisc);
+	struct netdev_queue *txq = &dev->tx_queue;
+	return (txq->qdisc != &noop_qdisc);
 }
 
 /* Check if a route is valid prefix route */
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 59ed9cae66b9..6ae43a3c7726 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -574,9 +574,10 @@ static struct Qdisc_ops wme_qdisc_ops __read_mostly =
 
 void ieee80211_install_qdisc(struct net_device *dev)
 {
+	struct netdev_queue *txq = &dev->tx_queue;
 	struct Qdisc *qdisc;
 
-	qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+	qdisc = qdisc_create_dflt(dev, txq,
 				  &wme_qdisc_ops, TC_H_ROOT);
 	if (!qdisc) {
 		printk(KERN_ERR "%s: qdisc installation failed\n", dev->name);
@@ -587,15 +588,17 @@ void ieee80211_install_qdisc(struct net_device *dev)
 	qdisc->handle = 0x80010000;
 
 	qdisc_lock_tree(dev);
-	list_add_tail(&qdisc->list, &dev->qdisc_list);
-	dev->qdisc_sleeping = qdisc;
+	list_add_tail(&qdisc->list, &txq->qdisc_list);
+	txq->qdisc_sleeping = qdisc;
 	qdisc_unlock_tree(dev);
 }
 
 
 int ieee80211_qdisc_installed(struct net_device *dev)
 {
-	return dev->qdisc_sleeping->ops == &wme_qdisc_ops;
+	struct netdev_queue *txq = &dev->tx_queue;
+
+	return txq->qdisc_sleeping->ops == &wme_qdisc_ops;
 }
 
 
@@ -614,8 +617,9 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 			struct sta_info *sta, u16 tid)
 {
 	int i;
+	struct netdev_queue *txq = &local->mdev->tx_queue;
 	struct ieee80211_sched_data *q =
-			qdisc_priv(local->mdev->qdisc_sleeping);
+			qdisc_priv(txq->qdisc_sleeping);
 	DECLARE_MAC_BUF(mac);
 
 	/* prepare the filter and save it for the SW queue
@@ -655,8 +659,9 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   u8 requeue)
 {
 	struct ieee80211_hw *hw = &local->hw;
+	struct netdev_queue *txq = &local->mdev->tx_queue;
 	struct ieee80211_sched_data *q =
-		qdisc_priv(local->mdev->qdisc_sleeping);
+		qdisc_priv(txq->qdisc_sleeping);
 	int agg_queue = sta->tid_to_tx_q[tid];
 
 	/* return the qdisc to the pool */
@@ -671,7 +676,8 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 
 void ieee80211_requeue(struct ieee80211_local *local, int queue)
 {
-	struct Qdisc *root_qd = local->mdev->qdisc_sleeping;
+	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct Qdisc *root_qd = txq->qdisc_sleeping;
 	struct ieee80211_sched_data *q = qdisc_priv(root_qd);
 	struct Qdisc *qdisc = q->queues[queue];
 	struct sk_buff *skb = NULL;
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index e2389f161e46..b483bbea6118 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -166,7 +166,8 @@ replay:
 
 	/* Find qdisc */
 	if (!parent) {
-		q = dev->qdisc_sleeping;
+		struct netdev_queue *dev_queue = &dev->tx_queue;
+		q = dev_queue->qdisc_sleeping;
 		parent = q->handle;
 	} else {
 		q = qdisc_lookup(dev, TC_H_MAJ(t->tcm_parent));
@@ -390,6 +391,7 @@ static int tcf_node_dump(struct tcf_proto *tp, unsigned long n,
 static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
+	struct netdev_queue *dev_queue;
 	int t;
 	int s_t;
 	struct net_device *dev;
@@ -408,8 +410,9 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 	if ((dev = dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
 		return skb->len;
 
+	dev_queue = &dev->tx_queue;
 	if (!tcm->tcm_parent)
-		q = dev->qdisc_sleeping;
+		q = dev_queue->qdisc_sleeping;
 	else
 		q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent));
 	if (!q)
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 570cef2a9c5f..2313fa7c97be 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -185,9 +185,10 @@ EXPORT_SYMBOL(unregister_qdisc);
 
 struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 {
+	struct netdev_queue *dev_queue = &dev->tx_queue;
 	struct Qdisc *q;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
+	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 		if (q->handle == handle)
 			return q;
 	}
@@ -441,6 +442,7 @@ static u32 qdisc_alloc_handle(struct net_device *dev)
 static struct Qdisc *
 dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 {
+	struct netdev_queue *dev_queue;
 	struct Qdisc *oqdisc;
 
 	if (dev->flags & IFF_UP)
@@ -459,8 +461,8 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 		}
 
 	} else {
-
-		oqdisc = dev->qdisc_sleeping;
+		dev_queue = &dev->tx_queue;
+		oqdisc = dev_queue->qdisc_sleeping;
 
 		/* Prune old scheduler */
 		if (oqdisc && atomic_read(&oqdisc->refcnt) <= 1)
@@ -469,8 +471,8 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 		/* ... and graft new one */
 		if (qdisc == NULL)
 			qdisc = &noop_qdisc;
-		dev->qdisc_sleeping = qdisc;
-		dev->qdisc = &noop_qdisc;
+		dev_queue->qdisc_sleeping = qdisc;
+		dev_queue->qdisc = &noop_qdisc;
 	}
 
 	qdisc_unlock_tree(dev);
@@ -633,7 +635,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 			}
 		}
 		qdisc_lock_tree(dev);
-		list_add_tail(&sch->list, &dev->qdisc_list);
+		list_add_tail(&sch->list, &dev_queue->qdisc_list);
 		qdisc_unlock_tree(dev);
 
 		return sch;
@@ -740,7 +742,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 				q = dev->qdisc_ingress;
 			}
 		} else {
-			q = dev->qdisc_sleeping;
+			struct netdev_queue *dev_queue = &dev->tx_queue;
+			q = dev_queue->qdisc_sleeping;
 		}
 		if (!q)
 			return -ENOENT;
@@ -814,7 +817,8 @@ replay:
 				q = dev->qdisc_ingress;
 			}
 		} else {
-			q = dev->qdisc_sleeping;
+			struct netdev_queue *dev_queue = &dev->tx_queue;
+			q = dev_queue->qdisc_sleeping;
 		}
 
 		/* It may be default qdisc, ignore it */
@@ -1015,12 +1019,14 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 	read_lock(&dev_base_lock);
 	idx = 0;
 	for_each_netdev(&init_net, dev) {
+		struct netdev_queue *dev_queue;
 		if (idx < s_idx)
 			goto cont;
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		list_for_each_entry(q, &dev->qdisc_list, list) {
+		dev_queue = &dev->tx_queue;
+		list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 			if (q_idx < s_q_idx) {
 				q_idx++;
 				continue;
@@ -1054,6 +1060,7 @@ done:
 static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 {
 	struct net *net = sock_net(skb->sk);
+	struct netdev_queue *dev_queue;
 	struct tcmsg *tcm = NLMSG_DATA(n);
 	struct nlattr *tca[TCA_MAX + 1];
 	struct net_device *dev;
@@ -1091,6 +1098,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 
 	/* Step 1. Determine qdisc handle X:0 */
 
+	dev_queue = &dev->tx_queue;
 	if (pid != TC_H_ROOT) {
 		u32 qid1 = TC_H_MAJ(pid);
 
@@ -1101,7 +1109,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 		} else if (qid1) {
 			qid = qid1;
 		} else if (qid == 0)
-			qid = dev->qdisc_sleeping->handle;
+			qid = dev_queue->qdisc_sleeping->handle;
 
 		/* Now qid is genuine qdisc handle consistent
 		   both with parent and child.
@@ -1112,7 +1120,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 			pid = TC_H_MAKE(qid, pid);
 	} else {
 		if (qid == 0)
-			qid = dev->qdisc_sleeping->handle;
+			qid = dev_queue->qdisc_sleeping->handle;
 	}
 
 	/* OK. Locate qdisc */
@@ -1248,6 +1256,7 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk
 static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
+	struct netdev_queue *dev_queue;
 	int t;
 	int s_t;
 	struct net_device *dev;
@@ -1266,7 +1275,8 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
+	dev_queue = &dev->tx_queue;
+	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 		if (t < s_t || !q->ops->cl_ops ||
 		    (tcm->tcm_parent &&
 		     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 804d44b00348..3223e5ba76aa 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -122,7 +122,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  *
  * __LINK_STATE_QDISC_RUNNING guarantees only one CPU can process this
  * device at a time. queue->lock serializes queue accesses for
- * this device AND dev->qdisc pointer itself.
+ * this device AND txq->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
@@ -138,7 +138,8 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  */
 static inline int qdisc_restart(struct net_device *dev)
 {
-	struct Qdisc *q = dev->qdisc;
+	struct netdev_queue *txq = &dev->tx_queue;
+	struct Qdisc *q = txq->qdisc;
 	struct sk_buff *skb;
 	int ret = NETDEV_TX_BUSY;
 
@@ -148,15 +149,15 @@ static inline int qdisc_restart(struct net_device *dev)
 
 
 	/* And release queue */
-	spin_unlock(&q->dev_queue->lock);
+	spin_unlock(&txq->lock);
 
 	HARD_TX_LOCK(dev, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
 		ret = dev_hard_start_xmit(skb, dev);
 	HARD_TX_UNLOCK(dev);
 
-	spin_lock(&q->dev_queue->lock);
-	q = dev->qdisc;
+	spin_lock(&txq->lock);
+	q = txq->qdisc;
 
 	switch (ret) {
 	case NETDEV_TX_OK:
@@ -207,9 +208,10 @@ void __qdisc_run(struct net_device *dev)
 static void dev_watchdog(unsigned long arg)
 {
 	struct net_device *dev = (struct net_device *)arg;
+	struct netdev_queue *txq = &dev->tx_queue;
 
 	netif_tx_lock(dev);
-	if (dev->qdisc != &noop_qdisc) {
+	if (txq->qdisc != &noop_qdisc) {
 		if (netif_device_present(dev) &&
 		    netif_running(dev) &&
 		    netif_carrier_ok(dev)) {
@@ -539,53 +541,63 @@ EXPORT_SYMBOL(qdisc_destroy);
 
 void dev_activate(struct net_device *dev)
 {
+	struct netdev_queue *txq = &dev->tx_queue;
+
 	/* No queueing discipline is attached to device;
 	   create default one i.e. pfifo_fast for devices,
 	   which need queueing and noqueue_qdisc for
 	   virtual interfaces
 	 */
 
-	if (dev->qdisc_sleeping == &noop_qdisc) {
+	if (txq->qdisc_sleeping == &noop_qdisc) {
 		struct Qdisc *qdisc;
 		if (dev->tx_queue_len) {
-			qdisc = qdisc_create_dflt(dev, &dev->tx_queue,
+			qdisc = qdisc_create_dflt(dev, txq,
 						  &pfifo_fast_ops,
 						  TC_H_ROOT);
 			if (qdisc == NULL) {
 				printk(KERN_INFO "%s: activation failed\n", dev->name);
 				return;
 			}
-			list_add_tail(&qdisc->list, &dev->qdisc_list);
+			list_add_tail(&qdisc->list, &txq->qdisc_list);
 		} else {
 			qdisc =  &noqueue_qdisc;
 		}
-		dev->qdisc_sleeping = qdisc;
+		txq->qdisc_sleeping = qdisc;
 	}
 
 	if (!netif_carrier_ok(dev))
 		/* Delay activation until next carrier-on event */
 		return;
 
-	spin_lock_bh(&dev->tx_queue.lock);
-	rcu_assign_pointer(dev->qdisc, dev->qdisc_sleeping);
-	if (dev->qdisc != &noqueue_qdisc) {
+	spin_lock_bh(&txq->lock);
+	rcu_assign_pointer(txq->qdisc, txq->qdisc_sleeping);
+	if (txq->qdisc != &noqueue_qdisc) {
 		dev->trans_start = jiffies;
 		dev_watchdog_up(dev);
 	}
-	spin_unlock_bh(&dev->tx_queue.lock);
+	spin_unlock_bh(&txq->lock);
+}
+
+static void dev_deactivate_queue(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 struct Qdisc *qdisc_default)
+{
+	struct Qdisc *qdisc = dev_queue->qdisc;
+
+	if (qdisc) {
+		dev_queue->qdisc = qdisc_default;
+		qdisc_reset(qdisc);
+	}
 }
 
 void dev_deactivate(struct net_device *dev)
 {
-	struct Qdisc *qdisc;
 	struct sk_buff *skb;
 	int running;
 
 	spin_lock_bh(&dev->tx_queue.lock);
-	qdisc = dev->qdisc;
-	dev->qdisc = &noop_qdisc;
-
-	qdisc_reset(qdisc);
+	dev_deactivate_queue(dev, &dev->tx_queue, &noop_qdisc);
 
 	skb = dev->gso_skb;
 	dev->gso_skb = NULL;
@@ -622,32 +634,44 @@ void dev_deactivate(struct net_device *dev)
 	} while (WARN_ON_ONCE(running));
 }
 
+static void dev_init_scheduler_queue(struct net_device *dev,
+				     struct netdev_queue *dev_queue,
+				     struct Qdisc *qdisc)
+{
+	dev_queue->qdisc = qdisc;
+	dev_queue->qdisc_sleeping = qdisc;
+	INIT_LIST_HEAD(&dev_queue->qdisc_list);
+}
+
 void dev_init_scheduler(struct net_device *dev)
 {
 	qdisc_lock_tree(dev);
-	dev->qdisc = &noop_qdisc;
-	dev->qdisc_sleeping = &noop_qdisc;
-	INIT_LIST_HEAD(&dev->qdisc_list);
+	dev_init_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
+	dev_init_scheduler_queue(dev, &dev->rx_queue, NULL);
 	qdisc_unlock_tree(dev);
 
 	setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
 }
 
-void dev_shutdown(struct net_device *dev)
+static void dev_shutdown_scheduler_queue(struct net_device *dev,
+					 struct netdev_queue *dev_queue,
+					 struct Qdisc *qdisc_default)
 {
-	struct Qdisc *qdisc;
+	struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
+
+	if (qdisc) {
+		dev_queue->qdisc = qdisc_default;
+		dev_queue->qdisc_sleeping = qdisc_default;
 
-	qdisc_lock_tree(dev);
-	qdisc = dev->qdisc_sleeping;
-	dev->qdisc = &noop_qdisc;
-	dev->qdisc_sleeping = &noop_qdisc;
-	qdisc_destroy(qdisc);
-#if defined(CONFIG_NET_SCH_INGRESS) || defined(CONFIG_NET_SCH_INGRESS_MODULE)
-	if ((qdisc = dev->qdisc_ingress) != NULL) {
-		dev->qdisc_ingress = NULL;
 		qdisc_destroy(qdisc);
 	}
-#endif
+}
+
+void dev_shutdown(struct net_device *dev)
+{
+	qdisc_lock_tree(dev);
+	dev_shutdown_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
+	dev_shutdown_scheduler_queue(dev, &dev->rx_queue, NULL);
 	BUG_TRAP(!timer_pending(&dev->watchdog_timer));
 	qdisc_unlock_tree(dev);
 }
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 71b73c528f9b..4093f1eaaf60 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -180,7 +180,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 	 * skb will be queued.
 	 */
 	if (count > 1 && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) {
-		struct Qdisc *rootq = qdisc_dev(sch)->qdisc;
+		struct Qdisc *rootq = qdisc_dev(sch)->tx_queue.qdisc;
 		u32 dupsave = q->duplicate; /* prevent duplicating a dup... */
 		q->duplicate = 0;
 
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 4f3054e8e1ab..8ac05981be20 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -107,17 +107,19 @@ static struct sk_buff *
 teql_dequeue(struct Qdisc* sch)
 {
 	struct teql_sched_data *dat = qdisc_priv(sch);
+	struct netdev_queue *dat_queue;
 	struct sk_buff *skb;
 
 	skb = __skb_dequeue(&dat->q);
+	dat_queue = &dat->m->dev->tx_queue;
 	if (skb == NULL) {
-		struct net_device *m = qdisc_dev(dat->m->dev->qdisc);
+		struct net_device *m = qdisc_dev(dat_queue->qdisc);
 		if (m) {
 			dat->m->slaves = sch;
 			netif_wake_queue(m);
 		}
 	}
-	sch->q.qlen = dat->q.qlen + dat->m->dev->qdisc->q.qlen;
+	sch->q.qlen = dat->q.qlen + dat_queue->qdisc->q.qlen;
 	return skb;
 }
 
@@ -155,7 +157,7 @@ teql_destroy(struct Qdisc* sch)
 					if (q == master->slaves) {
 						master->slaves = NULL;
 						spin_lock_bh(&master->dev->tx_queue.lock);
-						qdisc_reset(master->dev->qdisc);
+						qdisc_reset(master->dev->tx_queue.qdisc);
 						spin_unlock_bh(&master->dev->tx_queue.lock);
 					}
 				}
@@ -216,7 +218,7 @@ static int teql_qdisc_init(struct Qdisc *sch, struct nlattr *opt)
 static int
 __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *dev)
 {
-	struct teql_sched_data *q = qdisc_priv(dev->qdisc);
+	struct teql_sched_data *q = qdisc_priv(dev->tx_queue.qdisc);
 	struct neighbour *mn = skb->dst->neighbour;
 	struct neighbour *n = q->ncache;
 
@@ -252,7 +254,7 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *
 static inline int teql_resolve(struct sk_buff *skb,
 			       struct sk_buff *skb_res, struct net_device *dev)
 {
-	if (dev->qdisc == &noop_qdisc)
+	if (dev->tx_queue.qdisc == &noop_qdisc)
 		return -ENODEV;
 
 	if (dev->header_ops == NULL ||
@@ -284,7 +286,7 @@ restart:
 	do {
 		struct net_device *slave = qdisc_dev(q);
 
-		if (slave->qdisc_sleeping != q)
+		if (slave->tx_queue.qdisc_sleeping != q)
 			continue;
 		if (netif_queue_stopped(slave) ||
 		    __netif_subqueue_stopped(slave, subq) ||
-- 
cgit v1.2.3


From 816f3258e70db38d6d92c8d871377179fd69160f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 22:49:00 -0700
Subject: netdev: Kill qdisc_ingress, use netdev->rx_queue.qdisc instead.

Now that our qdisc management is bi-directional, per-queue, and fully
orthogonal, there is no reason to have a special ingress qdisc pointer
in struct net_device.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  3 ---
 net/core/dev.c            |  4 ++--
 net/sched/sch_api.c       | 11 ++++++-----
 3 files changed, 8 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index df702a7b3db5..e7c49246fd88 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -634,9 +634,6 @@ struct net_device
 
 	struct netdev_queue	rx_queue;
 	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
-
-	struct Qdisc		*qdisc_ingress;
-
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 	/* Partially transmitted GSO packet. */
diff --git a/net/core/dev.c b/net/core/dev.c
index ce79c28d739d..ab760a954d99 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2033,7 +2033,7 @@ static int ing_filter(struct sk_buff *skb)
 	rxq = &dev->rx_queue;
 
 	spin_lock(&rxq->lock);
-	if ((q = dev->qdisc_ingress) != NULL)
+	if ((q = rxq->qdisc) != NULL)
 		result = q->enqueue(skb, q);
 	spin_unlock(&rxq->lock);
 
@@ -2044,7 +2044,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
 					 struct packet_type **pt_prev,
 					 int *ret, struct net_device *orig_dev)
 {
-	if (!skb->dev->qdisc_ingress)
+	if (!skb->dev->rx_queue.qdisc)
 		goto out;
 
 	if (*pt_prev) {
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 2313fa7c97be..4003c280b69f 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -450,14 +450,15 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 
 	qdisc_lock_tree(dev);
 	if (qdisc && qdisc->flags&TCQ_F_INGRESS) {
-		oqdisc = dev->qdisc_ingress;
+		dev_queue = &dev->rx_queue;
+		oqdisc = dev_queue->qdisc;
 		/* Prune old scheduler */
 		if (oqdisc && atomic_read(&oqdisc->refcnt) <= 1) {
 			/* delete */
 			qdisc_reset(oqdisc);
-			dev->qdisc_ingress = NULL;
+			dev_queue->qdisc = NULL;
 		} else {  /* new */
-			dev->qdisc_ingress = qdisc;
+			dev_queue->qdisc = qdisc;
 		}
 
 	} else {
@@ -739,7 +740,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
 			} else { /* ingress */
-				q = dev->qdisc_ingress;
+				q = dev->rx_queue.qdisc;
 			}
 		} else {
 			struct netdev_queue *dev_queue = &dev->tx_queue;
@@ -814,7 +815,7 @@ replay:
 					return -ENOENT;
 				q = qdisc_leaf(p, clid);
 			} else { /*ingress */
-				q = dev->qdisc_ingress;
+				q = dev->rx_queue.qdisc;
 			}
 		} else {
 			struct netdev_queue *dev_queue = &dev->tx_queue;
-- 
cgit v1.2.3


From ee609cb36220d18c0cf476b066a5ab7e6f6d3a69 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 22:58:37 -0700
Subject: netdev: Move next_sched into struct netdev_queue.

We schedule queues, not the device, for output queue processing in BH.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  5 ++---
 net/core/dev.c            | 15 +++++++--------
 2 files changed, 9 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e7c49246fd88..1379c822e51d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -454,6 +454,7 @@ struct netdev_queue {
 	struct Qdisc		*qdisc;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
+	struct netdev_queue	*next_sched;
 };
 
 /*
@@ -545,8 +546,6 @@ struct net_device
 #define NETIF_F_V6_CSUM		(NETIF_F_GEN_CSUM | NETIF_F_IPV6_CSUM)
 #define NETIF_F_ALL_CSUM	(NETIF_F_V4_CSUM | NETIF_F_V6_CSUM)
 
-	struct net_device	*next_sched;
-
 	/* Interface index. Unique device identifier	*/
 	int			ifindex;
 	int			iflink;
@@ -940,7 +939,7 @@ static inline int unregister_gifconf(unsigned int family)
  */
 struct softnet_data
 {
-	struct net_device	*output_queue;
+	struct netdev_queue	*output_queue;
 	struct sk_buff_head	input_pkt_queue;
 	struct list_head	poll_list;
 	struct sk_buff		*completion_queue;
diff --git a/net/core/dev.c b/net/core/dev.c
index ab760a954d99..d6b8d3c3e6ec 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1323,13 +1323,14 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
 void __netif_schedule(struct net_device *dev)
 {
 	if (!test_and_set_bit(__LINK_STATE_SCHED, &dev->state)) {
+		struct netdev_queue *txq = &dev->tx_queue;
 		unsigned long flags;
 		struct softnet_data *sd;
 
 		local_irq_save(flags);
 		sd = &__get_cpu_var(softnet_data);
-		dev->next_sched = sd->output_queue;
-		sd->output_queue = dev;
+		txq->next_sched = sd->output_queue;
+		sd->output_queue = txq;
 		raise_softirq_irqoff(NET_TX_SOFTIRQ);
 		local_irq_restore(flags);
 	}
@@ -1912,7 +1913,7 @@ static void net_tx_action(struct softirq_action *h)
 	}
 
 	if (sd->output_queue) {
-		struct net_device *head;
+		struct netdev_queue *head;
 
 		local_irq_disable();
 		head = sd->output_queue;
@@ -1920,12 +1921,10 @@ static void net_tx_action(struct softirq_action *h)
 		local_irq_enable();
 
 		while (head) {
-			struct net_device *dev = head;
-			struct netdev_queue *txq;
+			struct netdev_queue *txq = head;
+			struct net_device *dev = txq->dev;
 			head = head->next_sched;
 
-			txq = &dev->tx_queue;
-
 			smp_mb__before_clear_bit();
 			clear_bit(__LINK_STATE_SCHED, &dev->state);
 
@@ -4346,7 +4345,7 @@ static int dev_cpu_callback(struct notifier_block *nfb,
 			    void *ocpu)
 {
 	struct sk_buff **list_skb;
-	struct net_device **list_net;
+	struct netdev_queue **list_net;
 	struct sk_buff *skb;
 	unsigned int cpu, oldcpu = (unsigned long)ocpu;
 	struct softnet_data *sd, *oldsd;
-- 
cgit v1.2.3


From 970565bbad0c7b98db0d14131a69e5a0f4445d49 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:10:33 -0700
Subject: netdev: Move gso_skb into netdev_queue.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 +---
 net/sched/sch_generic.c   | 42 +++++++++++++++++++++++-------------------
 2 files changed, 24 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 1379c822e51d..aae6c6d153f2 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -452,6 +452,7 @@ struct netdev_queue {
 	spinlock_t		lock;
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
+	struct sk_buff		*gso_skb;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
 	struct netdev_queue	*next_sched;
@@ -635,9 +636,6 @@ struct net_device
 	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
-	/* Partially transmitted GSO packet. */
-	struct sk_buff		*gso_skb;
-
 /*
  * One part is mostly used on xmit path (device)
  */
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index dda78ee314ec..8247a406a401 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -63,10 +63,11 @@ static inline int qdisc_qlen(struct Qdisc *q)
 }
 
 static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
+				  struct netdev_queue *dev_queue,
 				  struct Qdisc *q)
 {
 	if (unlikely(skb->next))
-		dev->gso_skb = skb;
+		dev_queue->gso_skb = skb;
 	else
 		q->ops->requeue(skb, q);
 
@@ -75,12 +76,13 @@ static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
 }
 
 static inline struct sk_buff *dev_dequeue_skb(struct net_device *dev,
+					      struct netdev_queue *dev_queue,
 					      struct Qdisc *q)
 {
 	struct sk_buff *skb;
 
-	if ((skb = dev->gso_skb))
-		dev->gso_skb = NULL;
+	if ((skb = dev_queue->gso_skb))
+		dev_queue->gso_skb = NULL;
 	else
 		skb = q->dequeue(q);
 
@@ -89,6 +91,7 @@ static inline struct sk_buff *dev_dequeue_skb(struct net_device *dev,
 
 static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 					   struct net_device *dev,
+					   struct netdev_queue *dev_queue,
 					   struct Qdisc *q)
 {
 	int ret;
@@ -111,7 +114,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		 * some time.
 		 */
 		__get_cpu_var(netdev_rx_stat).cpu_collision++;
-		ret = dev_requeue_skb(skb, dev, q);
+		ret = dev_requeue_skb(skb, dev, dev_queue, q);
 	}
 
 	return ret;
@@ -144,7 +147,7 @@ static inline int qdisc_restart(struct net_device *dev)
 	int ret = NETDEV_TX_BUSY;
 
 	/* Dequeue packet */
-	if (unlikely((skb = dev_dequeue_skb(dev, q)) == NULL))
+	if (unlikely((skb = dev_dequeue_skb(dev, txq, q)) == NULL))
 		return 0;
 
 
@@ -167,7 +170,7 @@ static inline int qdisc_restart(struct net_device *dev)
 
 	case NETDEV_TX_LOCKED:
 		/* Driver try lock failed */
-		ret = handle_dev_cpu_collision(skb, dev, q);
+		ret = handle_dev_cpu_collision(skb, dev, txq, q);
 		break;
 
 	default:
@@ -176,7 +179,7 @@ static inline int qdisc_restart(struct net_device *dev)
 			printk(KERN_WARNING "BUG %s code %d qlen %d\n",
 			       dev->name, ret, q->q.qlen);
 
-		ret = dev_requeue_skb(skb, dev, q);
+		ret = dev_requeue_skb(skb, dev, txq, q);
 		break;
 	}
 
@@ -578,31 +581,32 @@ void dev_activate(struct net_device *dev)
 	spin_unlock_bh(&txq->lock);
 }
 
-static void dev_deactivate_queue(struct net_device *dev,
-				 struct netdev_queue *dev_queue,
+static void dev_deactivate_queue(struct netdev_queue *dev_queue,
 				 struct Qdisc *qdisc_default)
 {
-	struct Qdisc *qdisc = dev_queue->qdisc;
+	struct Qdisc *qdisc;
+	struct sk_buff *skb;
+
+	spin_lock_bh(&dev_queue->lock);
 
+	qdisc = dev_queue->qdisc;
 	if (qdisc) {
 		dev_queue->qdisc = qdisc_default;
 		qdisc_reset(qdisc);
 	}
+	skb = dev_queue->gso_skb;
+	dev_queue->gso_skb = NULL;
+
+	spin_unlock_bh(&dev_queue->lock);
+
+	kfree_skb(skb);
 }
 
 void dev_deactivate(struct net_device *dev)
 {
-	struct sk_buff *skb;
 	int running;
 
-	spin_lock_bh(&dev->tx_queue.lock);
-	dev_deactivate_queue(dev, &dev->tx_queue, &noop_qdisc);
-
-	skb = dev->gso_skb;
-	dev->gso_skb = NULL;
-	spin_unlock_bh(&dev->tx_queue.lock);
-
-	kfree_skb(skb);
+	dev_deactivate_queue(&dev->tx_queue, &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
-- 
cgit v1.2.3


From 86d804e10a37cd86f16bf72386c37e843a98a74b Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:11:25 -0700
Subject: netdev: Make netif_schedule() routines work with netdev_queue
 objects.

Only plain netif_schedule() remains taking a net_device, mostly as a
compatability item while we transition the rest of these interfaces.

Everything else calls netif_schedule_queue() or __netif_schedule(),
both of which take a netdev_queue pointer.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 17 ++++++++++++-----
 net/core/dev.c            |  9 +++++----
 net/mac80211/main.c       |  4 ++--
 net/sched/sch_api.c       |  4 ++--
 net/sched/sch_cbq.c       |  2 +-
 net/sched/sch_generic.c   | 10 +++++-----
 6 files changed, 27 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index aae6c6d153f2..28aa8e77cee9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -952,12 +952,19 @@ DECLARE_PER_CPU(struct softnet_data,softnet_data);
 
 #define HAVE_NETIF_QUEUE
 
-extern void __netif_schedule(struct net_device *dev);
+extern void __netif_schedule(struct netdev_queue *txq);
 
-static inline void netif_schedule(struct net_device *dev)
+static inline void netif_schedule_queue(struct netdev_queue *txq)
 {
+	struct net_device *dev = txq->dev;
+
 	if (!test_bit(__LINK_STATE_XOFF, &dev->state))
-		__netif_schedule(dev);
+		__netif_schedule(txq);
+}
+
+static inline void netif_schedule(struct net_device *dev)
+{
+	netif_schedule_queue(&dev->tx_queue);
 }
 
 /**
@@ -987,7 +994,7 @@ static inline void netif_wake_queue(struct net_device *dev)
 	}
 #endif
 	if (test_and_clear_bit(__LINK_STATE_XOFF, &dev->state))
-		__netif_schedule(dev);
+		__netif_schedule(&dev->tx_queue);
 }
 
 /**
@@ -1103,7 +1110,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 #endif
 	if (test_and_clear_bit(__LINK_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
-		__netif_schedule(dev);
+		__netif_schedule(&dev->tx_queue);
 #endif
 }
 
diff --git a/net/core/dev.c b/net/core/dev.c
index d6b8d3c3e6ec..0dc888ad4217 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1320,12 +1320,13 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
 }
 
 
-void __netif_schedule(struct net_device *dev)
+void __netif_schedule(struct netdev_queue *txq)
 {
+	struct net_device *dev = txq->dev;
+
 	if (!test_and_set_bit(__LINK_STATE_SCHED, &dev->state)) {
-		struct netdev_queue *txq = &dev->tx_queue;
-		unsigned long flags;
 		struct softnet_data *sd;
+		unsigned long flags;
 
 		local_irq_save(flags);
 		sd = &__get_cpu_var(softnet_data);
@@ -1932,7 +1933,7 @@ static void net_tx_action(struct softirq_action *h)
 				qdisc_run(dev);
 				spin_unlock(&txq->lock);
 			} else {
-				netif_schedule(dev);
+				netif_schedule_queue(txq);
 			}
 		}
 	}
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 2968baa66b91..1c4d3ba6b878 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -885,10 +885,10 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 	spin_unlock_bh(&txq->lock);
 
 	/* we just requeued the all the frames that were in the removed
-	 * queue, and since we might miss a softirq we do netif_schedule.
+	 * queue, and since we might miss a softirq we do netif_schedule_queue.
 	 * ieee80211_wake_queue is not used here as this queue is not
 	 * necessarily stopped */
-	netif_schedule(local->mdev);
+	netif_schedule_queue(txq);
 	spin_lock_bh(&sta->lock);
 	*state = HT_AGG_STATE_IDLE;
 	sta->ampdu_mlme.addba_req_num[tid] = 0;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index e73bd68aa7ae..95873f8dd37c 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -282,11 +282,11 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 {
 	struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
 						 timer);
-	struct net_device *dev = qdisc_dev(wd->qdisc);
+	struct netdev_queue *txq = wd->qdisc->dev_queue;
 
 	wd->qdisc->flags &= ~TCQ_F_THROTTLED;
 	smp_wmb();
-	netif_schedule(dev);
+	netif_schedule_queue(txq);
 
 	return HRTIMER_NORESTART;
 }
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 99ce3da2b0a4..4efc836cbf38 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -650,7 +650,7 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer)
 	}
 
 	sch->flags &= ~TCQ_F_THROTTLED;
-	netif_schedule(qdisc_dev(sch));
+	netif_schedule_queue(sch->dev_queue);
 	return HRTIMER_NORESTART;
 }
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 8247a406a401..407dfdb142a4 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -62,7 +62,7 @@ static inline int qdisc_qlen(struct Qdisc *q)
 	return q->q.qlen;
 }
 
-static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
+static inline int dev_requeue_skb(struct sk_buff *skb,
 				  struct netdev_queue *dev_queue,
 				  struct Qdisc *q)
 {
@@ -71,7 +71,7 @@ static inline int dev_requeue_skb(struct sk_buff *skb, struct net_device *dev,
 	else
 		q->ops->requeue(skb, q);
 
-	netif_schedule(dev);
+	netif_schedule_queue(dev_queue);
 	return 0;
 }
 
@@ -114,7 +114,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		 * some time.
 		 */
 		__get_cpu_var(netdev_rx_stat).cpu_collision++;
-		ret = dev_requeue_skb(skb, dev, dev_queue, q);
+		ret = dev_requeue_skb(skb, dev_queue, q);
 	}
 
 	return ret;
@@ -179,7 +179,7 @@ static inline int qdisc_restart(struct net_device *dev)
 			printk(KERN_WARNING "BUG %s code %d qlen %d\n",
 			       dev->name, ret, q->q.qlen);
 
-		ret = dev_requeue_skb(skb, dev, txq, q);
+		ret = dev_requeue_skb(skb, txq, q);
 		break;
 	}
 
@@ -200,7 +200,7 @@ void __qdisc_run(struct net_device *dev)
 		 * 2. we've been doing it for too long.
 		 */
 		if (need_resched() || jiffies != start_time) {
-			netif_schedule(dev);
+			netif_schedule_queue(&dev->tx_queue);
 			break;
 		}
 	}
-- 
cgit v1.2.3


From c773e847ea8f6812804e40f52399c6921a00eab1 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:13:53 -0700
Subject: netdev: Move _xmit_lock and xmit_lock_owner into netdev_queue.

Accesses are mostly structured such that when there are multiple TX
queues the code transformations will be a little bit simpler.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bonding/bond_main.c         | 13 ++++++-
 drivers/net/hamradio/bpqether.c         | 12 ++++++-
 drivers/net/macvlan.c                   | 14 +++++++-
 drivers/net/wireless/hostap/hostap_hw.c | 12 ++++++-
 include/linux/netdevice.h               | 62 +++++++++++++++++++++------------
 net/8021q/vlan_dev.c                    | 15 ++++++--
 net/core/dev.c                          | 28 ++++++++++-----
 net/netrom/af_netrom.c                  | 12 ++++++-
 net/rose/af_rose.c                      | 12 ++++++-
 net/sched/sch_generic.c                 |  9 +++--
 10 files changed, 145 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d57b65dc2c72..dc733d75a5e9 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5019,6 +5019,17 @@ static int bond_check_params(struct bond_params *params)
 
 static struct lock_class_key bonding_netdev_xmit_lock_key;
 
+static void bond_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock,
+			  &bonding_netdev_xmit_lock_key);
+}
+
+static void bond_set_lockdep_class(struct net_device *dev)
+{
+	bond_set_lockdep_class_one(&dev->tx_queue);
+}
+
 /* Create a new bond based on the specified name and bonding parameters.
  * If name is NULL, obtain a suitable "bond%d" name for us.
  * Caller must NOT hold rtnl_lock; we need to release it here before we
@@ -5076,7 +5087,7 @@ int bond_create(char *name, struct bond_params *params)
 		goto out_bond;
 	}
 
-	lockdep_set_class(&bond_dev->_xmit_lock, &bonding_netdev_xmit_lock_key);
+	bond_set_lockdep_class(bond_dev);
 
 	netif_carrier_off(bond_dev);
 
diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c
index 5f4b4c6c9f76..fb186b8c3d4d 100644
--- a/drivers/net/hamradio/bpqether.c
+++ b/drivers/net/hamradio/bpqether.c
@@ -124,6 +124,16 @@ static LIST_HEAD(bpq_devices);
  */
 static struct lock_class_key bpq_netdev_xmit_lock_key;
 
+static void bpq_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock, &bpq_netdev_xmit_lock_key);
+}
+
+static void bpq_set_lockdep_class(struct net_device *dev)
+{
+	bpq_set_lockdep_class_one(&dev->tx_queue);
+}
+
 /* ------------------------------------------------------------------------ */
 
 
@@ -523,7 +533,7 @@ static int bpq_new_device(struct net_device *edev)
 	err = register_netdevice(ndev);
 	if (err)
 		goto error;
-	lockdep_set_class(&ndev->_xmit_lock, &bpq_netdev_xmit_lock_key);
+	bpq_set_lockdep_class(ndev);
 
 	/* List protected by RTNL */
 	list_add_rcu(&bpq->bpq_list, &bpq_devices);
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index c36a03ae9bfb..c02ceaa4a216 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -277,6 +277,17 @@ static struct lock_class_key macvlan_netdev_xmit_lock_key;
 #define MACVLAN_STATE_MASK \
 	((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT))
 
+static void macvlan_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock,
+			  &macvlan_netdev_xmit_lock_key);
+}
+
+static void macvlan_set_lockdep_class(struct net_device *dev)
+{
+	macvlan_set_lockdep_class_one(&dev->tx_queue);
+}
+
 static int macvlan_init(struct net_device *dev)
 {
 	struct macvlan_dev *vlan = netdev_priv(dev);
@@ -287,7 +298,8 @@ static int macvlan_init(struct net_device *dev)
 	dev->features 		= lowerdev->features & MACVLAN_FEATURES;
 	dev->iflink		= lowerdev->ifindex;
 
-	lockdep_set_class(&dev->_xmit_lock, &macvlan_netdev_xmit_lock_key);
+	macvlan_set_lockdep_class(dev);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/hostap/hostap_hw.c b/drivers/net/wireless/hostap/hostap_hw.c
index 09004a632ae7..c1f4bb005d92 100644
--- a/drivers/net/wireless/hostap/hostap_hw.c
+++ b/drivers/net/wireless/hostap/hostap_hw.c
@@ -3102,6 +3102,16 @@ static void prism2_clear_set_tim_queue(local_info_t *local)
  */
 static struct lock_class_key hostap_netdev_xmit_lock_key;
 
+static void prism2_set_lockdep_class_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock,
+			  &hostap_netdev_xmit_lock_key);
+}
+
+static void prism2_set_lockdep_class(struct net_device *dev)
+{
+	prism2_set_lockdep_class_one(&dev->tx_queue);
+}
 
 static struct net_device *
 prism2_init_local_data(struct prism2_helper_functions *funcs, int card_idx,
@@ -3268,7 +3278,7 @@ while (0)
 	if (ret >= 0)
 		ret = register_netdevice(dev);
 
-	lockdep_set_class(&dev->_xmit_lock, &hostap_netdev_xmit_lock_key);
+	prism2_set_lockdep_class(dev);
 	rtnl_unlock();
 	if (ret < 0) {
 		printk(KERN_WARNING "%s: register netdevice failed!\n",
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 28aa8e77cee9..c8d5f128858d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -453,6 +453,8 @@ struct netdev_queue {
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
 	struct sk_buff		*gso_skb;
+	spinlock_t		_xmit_lock;
+	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
 	struct netdev_queue	*next_sched;
@@ -639,12 +641,6 @@ struct net_device
 /*
  * One part is mostly used on xmit path (device)
  */
-	/* hard_start_xmit synchronizer */
-	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
-	/* cpu id of processor entered to hard_start_xmit or -1,
-	   if nobody entered there.
-	 */
-	int			xmit_lock_owner;
 	void			*priv;	/* pointer to private data	*/
 	int			(*hard_start_xmit) (struct sk_buff *skb,
 						    struct net_device *dev);
@@ -1402,52 +1398,72 @@ static inline void netif_rx_complete(struct net_device *dev,
  *
  * Get network device transmit lock
  */
-static inline void __netif_tx_lock(struct net_device *dev, int cpu)
+static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 {
-	spin_lock(&dev->_xmit_lock);
-	dev->xmit_lock_owner = cpu;
+	spin_lock(&txq->_xmit_lock);
+	txq->xmit_lock_owner = cpu;
 }
 
 static inline void netif_tx_lock(struct net_device *dev)
 {
-	__netif_tx_lock(dev, smp_processor_id());
+	__netif_tx_lock(&dev->tx_queue, smp_processor_id());
+}
+
+static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
+{
+	spin_lock_bh(&txq->_xmit_lock);
+	txq->xmit_lock_owner = smp_processor_id();
 }
 
 static inline void netif_tx_lock_bh(struct net_device *dev)
 {
-	spin_lock_bh(&dev->_xmit_lock);
-	dev->xmit_lock_owner = smp_processor_id();
+	__netif_tx_lock_bh(&dev->tx_queue);
 }
 
-static inline int netif_tx_trylock(struct net_device *dev)
+static inline int __netif_tx_trylock(struct netdev_queue *txq)
 {
-	int ok = spin_trylock(&dev->_xmit_lock);
+	int ok = spin_trylock(&txq->_xmit_lock);
 	if (likely(ok))
-		dev->xmit_lock_owner = smp_processor_id();
+		txq->xmit_lock_owner = smp_processor_id();
 	return ok;
 }
 
+static inline int netif_tx_trylock(struct net_device *dev)
+{
+	return __netif_tx_trylock(&dev->tx_queue);
+}
+
+static inline void __netif_tx_unlock(struct netdev_queue *txq)
+{
+	txq->xmit_lock_owner = -1;
+	spin_unlock(&txq->_xmit_lock);
+}
+
 static inline void netif_tx_unlock(struct net_device *dev)
 {
-	dev->xmit_lock_owner = -1;
-	spin_unlock(&dev->_xmit_lock);
+	__netif_tx_unlock(&dev->tx_queue);
+}
+
+static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
+{
+	txq->xmit_lock_owner = -1;
+	spin_unlock_bh(&txq->_xmit_lock);
 }
 
 static inline void netif_tx_unlock_bh(struct net_device *dev)
 {
-	dev->xmit_lock_owner = -1;
-	spin_unlock_bh(&dev->_xmit_lock);
+	__netif_tx_unlock_bh(&dev->tx_queue);
 }
 
-#define HARD_TX_LOCK(dev, cpu) {			\
+#define HARD_TX_LOCK(dev, txq, cpu) {			\
 	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		__netif_tx_lock(dev, cpu);			\
+		__netif_tx_lock(txq, cpu);		\
 	}						\
 }
 
-#define HARD_TX_UNLOCK(dev) {				\
+#define HARD_TX_UNLOCK(dev, txq) {			\
 	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		netif_tx_unlock(dev);			\
+		__netif_tx_unlock(txq);			\
 	}						\
 }
 
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index b6e52c025fd8..8efa399823e3 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -627,6 +627,18 @@ static void vlan_dev_set_rx_mode(struct net_device *vlan_dev)
  */
 static struct lock_class_key vlan_netdev_xmit_lock_key;
 
+static void vlan_dev_set_lockdep_one(struct netdev_queue *txq,
+				     int subclass)
+{
+	lockdep_set_class_and_subclass(&txq->_xmit_lock,
+				       &vlan_netdev_xmit_lock_key, subclass);
+}
+
+static void vlan_dev_set_lockdep_class(struct net_device *dev, int subclass)
+{
+	vlan_dev_set_lockdep_one(&dev->tx_queue, subclass);
+}
+
 static const struct header_ops vlan_header_ops = {
 	.create	 = vlan_dev_hard_header,
 	.rebuild = vlan_dev_rebuild_header,
@@ -668,8 +680,7 @@ static int vlan_dev_init(struct net_device *dev)
 	if (is_vlan_dev(real_dev))
 		subclass = 1;
 
-	lockdep_set_class_and_subclass(&dev->_xmit_lock,
-				&vlan_netdev_xmit_lock_key, subclass);
+	vlan_dev_set_lockdep_class(dev, subclass);
 	return 0;
 }
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 0218b0b9be80..a29a359b15d1 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -258,7 +258,7 @@ DEFINE_PER_CPU(struct softnet_data, softnet_data);
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 /*
- * register_netdevice() inits dev->_xmit_lock and sets lockdep class
+ * register_netdevice() inits txq->_xmit_lock and sets lockdep class
  * according to dev->type
  */
 static const unsigned short netdev_lock_type[] =
@@ -1758,19 +1758,19 @@ gso:
 	if (dev->flags & IFF_UP) {
 		int cpu = smp_processor_id(); /* ok because BHs are off */
 
-		if (dev->xmit_lock_owner != cpu) {
+		if (txq->xmit_lock_owner != cpu) {
 
-			HARD_TX_LOCK(dev, cpu);
+			HARD_TX_LOCK(dev, txq, cpu);
 
 			if (!netif_queue_stopped(dev) &&
 			    !netif_subqueue_stopped(dev, skb)) {
 				rc = 0;
 				if (!dev_hard_start_xmit(skb, dev)) {
-					HARD_TX_UNLOCK(dev);
+					HARD_TX_UNLOCK(dev, txq);
 					goto out;
 				}
 			}
-			HARD_TX_UNLOCK(dev);
+			HARD_TX_UNLOCK(dev, txq);
 			if (net_ratelimit())
 				printk(KERN_CRIT "Virtual device %s asks to "
 				       "queue packet!\n", dev->name);
@@ -3761,6 +3761,20 @@ static void rollback_registered(struct net_device *dev)
 	dev_put(dev);
 }
 
+static void __netdev_init_queue_locks_one(struct netdev_queue *dev_queue,
+					  struct net_device *dev)
+{
+	spin_lock_init(&dev_queue->_xmit_lock);
+	netdev_set_lockdep_class(&dev_queue->_xmit_lock, dev->type);
+	dev_queue->xmit_lock_owner = -1;
+}
+
+static void netdev_init_queue_locks(struct net_device *dev)
+{
+	__netdev_init_queue_locks_one(&dev->tx_queue, dev);
+	__netdev_init_queue_locks_one(&dev->rx_queue, dev);
+}
+
 /**
  *	register_netdevice	- register a network device
  *	@dev: device to register
@@ -3795,9 +3809,7 @@ int register_netdevice(struct net_device *dev)
 	BUG_ON(!dev_net(dev));
 	net = dev_net(dev);
 
-	spin_lock_init(&dev->_xmit_lock);
-	netdev_set_lockdep_class(&dev->_xmit_lock, dev->type);
-	dev->xmit_lock_owner = -1;
+	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
 
diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 74884f4a6255..819afc449e1e 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -74,6 +74,16 @@ static const struct proto_ops nr_proto_ops;
  */
 static struct lock_class_key nr_netdev_xmit_lock_key;
 
+static void nr_set_lockdep_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock, &nr_netdev_xmit_lock_key);
+}
+
+static void nr_set_lockdep_key(struct net_device *dev)
+{
+	nr_set_lockdep_one(&dev->tx_queue);
+}
+
 /*
  *	Socket removal during an interrupt is now safe.
  */
@@ -1430,7 +1440,7 @@ static int __init nr_proto_init(void)
 			free_netdev(dev);
 			goto fail;
 		}
-		lockdep_set_class(&dev->_xmit_lock, &nr_netdev_xmit_lock_key);
+		nr_set_lockdep_key(dev);
 		dev_nr[i] = dev;
 	}
 
diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index 46461a69cd0f..7dbbc0891623 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c
@@ -75,6 +75,16 @@ ax25_address rose_callsign;
  */
 static struct lock_class_key rose_netdev_xmit_lock_key;
 
+static void rose_set_lockdep_one(struct netdev_queue *txq)
+{
+	lockdep_set_class(&txq->_xmit_lock, &rose_netdev_xmit_lock_key);
+}
+
+static void rose_set_lockdep_key(struct net_device *dev)
+{
+	rose_set_lockdep_one(&dev->tx_queue);
+}
+
 /*
  *	Convert a ROSE address into text.
  */
@@ -1576,7 +1586,7 @@ static int __init rose_proto_init(void)
 			free_netdev(dev);
 			goto fail;
 		}
-		lockdep_set_class(&dev->_xmit_lock, &rose_netdev_xmit_lock_key);
+		rose_set_lockdep_key(dev);
 		dev_rose[i] = dev;
 	}
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index fcc7533f0bcc..b6a36d394663 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -92,10 +92,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 					   struct netdev_queue *dev_queue,
 					   struct Qdisc *q)
 {
-	struct net_device *dev = dev_queue->dev;
 	int ret;
 
-	if (unlikely(dev->xmit_lock_owner == smp_processor_id())) {
+	if (unlikely(dev_queue->xmit_lock_owner == smp_processor_id())) {
 		/*
 		 * Same CPU holding the lock. It may be a transient
 		 * configuration error, when hard_start_xmit() recurses. We
@@ -105,7 +104,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		kfree_skb(skb);
 		if (net_ratelimit())
 			printk(KERN_WARNING "Dead loop on netdevice %s, "
-			       "fix it urgently!\n", dev->name);
+			       "fix it urgently!\n", dev_queue->dev->name);
 		ret = qdisc_qlen(q);
 	} else {
 		/*
@@ -155,10 +154,10 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 
 	dev = txq->dev;
 
-	HARD_TX_LOCK(dev, smp_processor_id());
+	HARD_TX_LOCK(dev, txq, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
 		ret = dev_hard_start_xmit(skb, dev);
-	HARD_TX_UNLOCK(dev);
+	HARD_TX_UNLOCK(dev, txq);
 
 	spin_lock(&txq->lock);
 	q = txq->qdisc;
-- 
cgit v1.2.3


From b19fa1fa91845234961c64dbd564671aa7c0fd27 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:14:24 -0700
Subject: net: Delete NETDEVICES_MULTIQUEUE kconfig option.

Multiple TX queue support is a core networking feature.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/multiqueue.txt | 79 +--------------------------------
 drivers/net/Kconfig                     |  8 ----
 drivers/net/cpmac.c                     | 14 ------
 drivers/net/ixgbe/ixgbe_ethtool.c       |  6 ---
 drivers/net/ixgbe/ixgbe_main.c          | 40 -----------------
 drivers/net/s2io.c                      | 47 ++++----------------
 include/linux/netdevice.h               | 14 ------
 include/linux/skbuff.h                  | 10 -----
 net/mac80211/Kconfig                    |  3 --
 9 files changed, 9 insertions(+), 212 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/networking/multiqueue.txt b/Documentation/networking/multiqueue.txt
index ea5a42e8f79f..e6dc1ee9e8f1 100644
--- a/Documentation/networking/multiqueue.txt
+++ b/Documentation/networking/multiqueue.txt
@@ -3,19 +3,11 @@
 		===========================================
 
 Section 1: Base driver requirements for implementing multiqueue support
-Section 2: Qdisc support for multiqueue devices
-Section 3: Brief howto using PRIO or RR for multiqueue devices
-
 
 Intro: Kernel support for multiqueue devices
 ---------------------------------------------------------
 
-Kernel support for multiqueue devices is only an API that is presented to the
-netdevice layer for base drivers to implement.  This feature is part of the
-core networking stack, and all network devices will be running on the
-multiqueue-aware stack.  If a base driver only has one queue, then these
-changes are transparent to that driver.
-
+Kernel support for multiqueue devices is always present.
 
 Section 1: Base driver requirements for implementing multiqueue support
 -----------------------------------------------------------------------
@@ -43,73 +35,4 @@ bitmap on device initialization.  Below is an example from e1000:
 		netdev->features |= NETIF_F_MULTI_QUEUE;
 #endif
 
-
-Section 2: Qdisc support for multiqueue devices
------------------------------------------------
-
-Currently two qdiscs support multiqueue devices.  A new round-robin qdisc,
-sch_rr, and sch_prio. The qdisc is responsible for classifying the skb's to
-bands and queues, and will store the queue mapping into skb->queue_mapping.
-Use this field in the base driver to determine which queue to send the skb
-to.
-
-sch_rr has been added for hardware that doesn't want scheduling policies from
-software, so it's a straight round-robin qdisc.  It uses the same syntax and
-classification priomap that sch_prio uses, so it should be intuitive to
-configure for people who've used sch_prio.
-
-In order to utilitize the multiqueue features of the qdiscs, the network
-device layer needs to enable multiple queue support.  This can be done by
-selecting NETDEVICES_MULTIQUEUE under Drivers.
-
-The PRIO qdisc naturally plugs into a multiqueue device.  If
-NETDEVICES_MULTIQUEUE is selected, then on qdisc load, the number of
-bands requested is compared to the number of queues on the hardware.  If they
-are equal, it sets a one-to-one mapping up between the queues and bands.  If
-they're not equal, it will not load the qdisc.  This is the same behavior
-for RR.  Once the association is made, any skb that is classified will have
-skb->queue_mapping set, which will allow the driver to properly queue skb's
-to multiple queues.
-
-
-Section 3: Brief howto using PRIO and RR for multiqueue devices
----------------------------------------------------------------
-
-The userspace command 'tc,' part of the iproute2 package, is used to configure
-qdiscs.  To add the PRIO qdisc to your network device, assuming the device is
-called eth0, run the following command:
-
-# tc qdisc add dev eth0 root handle 1: prio bands 4 multiqueue
-
-This will create 4 bands, 0 being highest priority, and associate those bands
-to the queues on your NIC.  Assuming eth0 has 4 Tx queues, the band mapping
-would look like:
-
-band 0 => queue 0
-band 1 => queue 1
-band 2 => queue 2
-band 3 => queue 3
-
-Traffic will begin flowing through each queue if your TOS values are assigning
-traffic across the various bands.  For example, ssh traffic will always try to
-go out band 0 based on TOS -> Linux priority conversion (realtime traffic),
-so it will be sent out queue 0.  ICMP traffic (pings) fall into the "normal"
-traffic classification, which is band 1.  Therefore pings will be send out
-queue 1 on the NIC.
-
-Note the use of the multiqueue keyword.  This is only in versions of iproute2
-that support multiqueue networking devices; if this is omitted when loading
-a qdisc onto a multiqueue device, the qdisc will load and operate the same
-if it were loaded onto a single-queue device (i.e. - sends all traffic to
-queue 0).
-
-Another alternative to multiqueue band allocation can be done by using the
-multiqueue option and specify 0 bands.  If this is the case, the qdisc will
-allocate the number of bands to equal the number of queues that the device
-reports, and bring the qdisc online.
-
-The behavior of tc filters remains the same, where it will override TOS priority
-classification.
-
-
 Author: Peter P. Waskiewicz Jr. <peter.p.waskiewicz.jr@intel.com>
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index ef733abc857d..4675c1bd6fb9 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -26,14 +26,6 @@ menuconfig NETDEVICES
 # that for each of the symbols.
 if NETDEVICES
 
-config NETDEVICES_MULTIQUEUE
-	bool "Netdevice multiple hardware queue support"
-	---help---
-	  Say Y here if you want to allow the network stack to use multiple
-	  hardware TX queues on an ethernet device.
-
-	  Most people will say N here.
-
 config IFB
 	tristate "Intermediate Functional Block support"
 	depends on NET_CLS_ACT
diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index 7f3f62e1b113..d630e2a72f42 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -569,11 +569,7 @@ static int cpmac_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	len = max(skb->len, ETH_ZLEN);
 	queue = skb_get_queue_mapping(skb);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netif_stop_subqueue(dev, queue);
-#else
-	netif_stop_queue(dev);
-#endif
 
 	desc = &priv->desc_ring[queue];
 	if (unlikely(desc->dataflags & CPMAC_OWN)) {
@@ -626,24 +622,14 @@ static void cpmac_end_xmit(struct net_device *dev, int queue)
 
 		dev_kfree_skb_irq(desc->skb);
 		desc->skb = NULL;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		if (netif_subqueue_stopped(dev, queue))
 			netif_wake_subqueue(dev, queue);
-#else
-		if (netif_queue_stopped(dev))
-			netif_wake_queue(dev);
-#endif
 	} else {
 		if (netif_msg_tx_err(priv) && net_ratelimit())
 			printk(KERN_WARNING
 			       "%s: end_xmit: spurious interrupt\n", dev->name);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		if (netif_subqueue_stopped(dev, queue))
 			netif_wake_subqueue(dev, queue);
-#else
-		if (netif_queue_stopped(dev))
-			netif_wake_queue(dev);
-#endif
 	}
 }
 
diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index 12990b1fe7e4..81b769093d22 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -252,21 +252,15 @@ static int ixgbe_set_tso(struct net_device *netdev, u32 data)
 		netdev->features |= NETIF_F_TSO;
 		netdev->features |= NETIF_F_TSO6;
 	} else {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		struct ixgbe_adapter *adapter = netdev_priv(netdev);
 		int i;
-#endif
 		netif_stop_queue(netdev);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		for (i = 0; i < adapter->num_tx_queues; i++)
 			netif_stop_subqueue(netdev, i);
-#endif
 		netdev->features &= ~NETIF_F_TSO;
 		netdev->features &= ~NETIF_F_TSO6;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		for (i = 0; i < adapter->num_tx_queues; i++)
 			netif_start_subqueue(netdev, i);
-#endif
 		netif_start_queue(netdev);
 	}
 	return 0;
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index b37d618d8e2a..10a1c8c5cda1 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -266,28 +266,16 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_adapter *adapter,
 		 * sees the new next_to_clean.
 		 */
 		smp_mb();
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 		if (__netif_subqueue_stopped(netdev, tx_ring->queue_index) &&
 		    !test_bit(__IXGBE_DOWN, &adapter->state)) {
 			netif_wake_subqueue(netdev, tx_ring->queue_index);
 			adapter->restart_queue++;
 		}
-#else
-		if (netif_queue_stopped(netdev) &&
-		    !test_bit(__IXGBE_DOWN, &adapter->state)) {
-			netif_wake_queue(netdev);
-			adapter->restart_queue++;
-		}
-#endif
 	}
 
 	if (adapter->detect_tx_hung)
 		if (ixgbe_check_tx_hang(adapter, tx_ring, eop, eop_desc))
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 			netif_stop_subqueue(netdev, tx_ring->queue_index);
-#else
-			netif_stop_queue(netdev);
-#endif
 
 	if (total_tx_packets >= tx_ring->work_limit)
 		IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS, tx_ring->eims_value);
@@ -2192,11 +2180,7 @@ static void __devinit ixgbe_set_num_queues(struct ixgbe_adapter *adapter)
 		case (IXGBE_FLAG_RSS_ENABLED):
 			rss_m = 0xF;
 			nrq = rss_i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 			ntq = rss_i;
-#else
-			ntq = 1;
-#endif
 			break;
 		case 0:
 		default:
@@ -2370,10 +2354,8 @@ try_msi:
 	}
 
 out:
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	/* Notify the stack of the (possibly) reduced Tx Queue count. */
 	adapter->netdev->egress_subqueue_count = adapter->num_tx_queues;
-#endif
 
 	return err;
 }
@@ -2910,9 +2892,7 @@ static void ixgbe_watchdog(unsigned long data)
 	struct net_device *netdev = adapter->netdev;
 	bool link_up;
 	u32 link_speed = 0;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	int i;
-#endif
 
 	adapter->hw.mac.ops.check_link(&adapter->hw, &(link_speed), &link_up);
 
@@ -2934,10 +2914,8 @@ static void ixgbe_watchdog(unsigned long data)
 
 			netif_carrier_on(netdev);
 			netif_wake_queue(netdev);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 			for (i = 0; i < adapter->num_tx_queues; i++)
 				netif_wake_subqueue(netdev, i);
-#endif
 		} else {
 			/* Force detection of hung controller */
 			adapter->detect_tx_hung = true;
@@ -3264,11 +3242,7 @@ static int __ixgbe_maybe_stop_tx(struct net_device *netdev,
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netif_stop_subqueue(netdev, tx_ring->queue_index);
-#else
-	netif_stop_queue(netdev);
-#endif
 	/* Herbert's original patch had:
 	 *  smp_mb__after_netif_stop_queue();
 	 * but since that doesn't exist yet, just open code it. */
@@ -3280,11 +3254,7 @@ static int __ixgbe_maybe_stop_tx(struct net_device *netdev,
 		return -EBUSY;
 
 	/* A reprieve! - use start_queue because it doesn't call schedule */
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netif_wake_subqueue(netdev, tx_ring->queue_index);
-#else
-	netif_wake_queue(netdev);
-#endif
 	++adapter->restart_queue;
 	return 0;
 }
@@ -3312,9 +3282,7 @@ static int ixgbe_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
 	unsigned int f;
 	unsigned int nr_frags = skb_shinfo(skb)->nr_frags;
 	len -= skb->data_len;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	r_idx = (adapter->num_tx_queues - 1) & skb->queue_mapping;
-#endif
 	tx_ring = &adapter->tx_ring[r_idx];
 
 
@@ -3502,11 +3470,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	pci_set_master(pdev);
 	pci_save_state(pdev);
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netdev = alloc_etherdev_mq(sizeof(struct ixgbe_adapter), MAX_TX_QUEUES);
-#else
-	netdev = alloc_etherdev(sizeof(struct ixgbe_adapter));
-#endif
 	if (!netdev) {
 		err = -ENOMEM;
 		goto err_alloc_etherdev;
@@ -3598,9 +3562,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	if (pci_using_dac)
 		netdev->features |= NETIF_F_HIGHDMA;
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	netdev->features |= NETIF_F_MULTI_QUEUE;
-#endif
 
 	/* make sure the EEPROM is good */
 	if (ixgbe_validate_eeprom_checksum(hw, NULL) < 0) {
@@ -3668,10 +3630,8 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 
 	netif_carrier_off(netdev);
 	netif_stop_queue(netdev);
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	for (i = 0; i < adapter->num_tx_queues; i++)
 		netif_stop_subqueue(netdev, i);
-#endif
 
 	ixgbe_napi_add_all(adapter);
 
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index e7a3dbec674c..51a91154125d 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -546,13 +546,10 @@ static struct pci_driver s2io_driver = {
 static inline void s2io_stop_all_tx_queue(struct s2io_nic *sp)
 {
 	int i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			netif_stop_subqueue(sp->dev, i);
-	} else
-#endif
-	{
+	} else {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_STOP;
 		netif_stop_queue(sp->dev);
@@ -561,12 +558,9 @@ static inline void s2io_stop_all_tx_queue(struct s2io_nic *sp)
 
 static inline void s2io_stop_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq)
 		netif_stop_subqueue(sp->dev, fifo_no);
-	else
-#endif
-	{
+	else {
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_STOP;
 		netif_stop_queue(sp->dev);
@@ -576,13 +570,10 @@ static inline void s2io_stop_tx_queue(struct s2io_nic *sp, int fifo_no)
 static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
 {
 	int i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			netif_start_subqueue(sp->dev, i);
-	} else
-#endif
-	{
+	} else {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
 		netif_start_queue(sp->dev);
@@ -591,12 +582,9 @@ static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
 
 static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq)
 		netif_start_subqueue(sp->dev, fifo_no);
-	else
-#endif
-	{
+	else {
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_START;
 		netif_start_queue(sp->dev);
@@ -606,13 +594,10 @@ static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
 static inline void s2io_wake_all_tx_queue(struct s2io_nic *sp)
 {
 	int i;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			netif_wake_subqueue(sp->dev, i);
-	} else
-#endif
-	{
+	} else {
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
 		netif_wake_queue(sp->dev);
@@ -623,13 +608,10 @@ static inline void s2io_wake_tx_queue(
 	struct fifo_info *fifo, int cnt, u8 multiq)
 {
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (multiq) {
 		if (cnt && __netif_subqueue_stopped(fifo->dev, fifo->fifo_no))
 			netif_wake_subqueue(fifo->dev, fifo->fifo_no);
-	} else
-#endif
-	if (cnt && (fifo->queue_state == FIFO_QUEUE_STOP)) {
+	} else if (cnt && (fifo->queue_state == FIFO_QUEUE_STOP)) {
 		if (netif_queue_stopped(fifo->dev)) {
 			fifo->queue_state = FIFO_QUEUE_START;
 			netif_wake_queue(fifo->dev);
@@ -4189,15 +4171,12 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev)
 			return NETDEV_TX_LOCKED;
 	}
 
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (sp->config.multiq) {
 		if (__netif_subqueue_stopped(dev, fifo->fifo_no)) {
 			spin_unlock_irqrestore(&fifo->tx_lock, flags);
 			return NETDEV_TX_BUSY;
 		}
-	} else
-#endif
-	if (unlikely(fifo->queue_state == FIFO_QUEUE_STOP)) {
+	} else if (unlikely(fifo->queue_state == FIFO_QUEUE_STOP)) {
 		if (netif_queue_stopped(dev)) {
 			spin_unlock_irqrestore(&fifo->tx_lock, flags);
 			return NETDEV_TX_BUSY;
@@ -7633,12 +7612,6 @@ static int s2io_verify_parm(struct pci_dev *pdev, u8 *dev_intr_type,
 		DBG_PRINT(ERR_DBG, "tx fifos\n");
 	}
 
-#ifndef CONFIG_NETDEVICES_MULTIQUEUE
-	if (multiq) {
-		DBG_PRINT(ERR_DBG, "s2io: Multiqueue support not enabled\n");
-		multiq = 0;
-	}
-#endif
 	if (multiq)
 		*dev_multiq = multiq;
 
@@ -7783,12 +7756,10 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre)
 		pci_disable_device(pdev);
 		return -ENODEV;
 	}
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (dev_multiq)
 		dev = alloc_etherdev_mq(sizeof(struct s2io_nic), tx_fifo_num);
 	else
-#endif
-	dev = alloc_etherdev(sizeof(struct s2io_nic));
+		dev = alloc_etherdev(sizeof(struct s2io_nic));
 	if (dev == NULL) {
 		DBG_PRINT(ERR_DBG, "Device allocation failed\n");
 		pci_disable_device(pdev);
@@ -7979,10 +7950,8 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre)
 		dev->features |= NETIF_F_UFO;
 		dev->features |= NETIF_F_HW_CSUM;
 	}
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	if (config->multiq)
 		dev->features |= NETIF_F_MULTI_QUEUE;
-#endif
 	dev->tx_timeout = &s2io_tx_watchdog;
 	dev->watchdog_timeo = WATCH_DOG_TIMEOUT;
 	INIT_WORK(&sp->rst_timer_task, s2io_restart_nic);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c8d5f128858d..e2d931f9b700 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1043,9 +1043,7 @@ static inline int netif_running(const struct net_device *dev)
  */
 static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	clear_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
-#endif
 }
 
 /**
@@ -1057,13 +1055,11 @@ static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
  */
 static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
 #endif
 	set_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
-#endif
 }
 
 /**
@@ -1076,12 +1072,8 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 static inline int __netif_subqueue_stopped(const struct net_device *dev,
 					 u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	return test_bit(__LINK_STATE_XOFF,
 			&dev->egress_subqueue[queue_index].state);
-#else
-	return 0;
-#endif
 }
 
 static inline int netif_subqueue_stopped(const struct net_device *dev,
@@ -1099,7 +1091,6 @@ static inline int netif_subqueue_stopped(const struct net_device *dev,
  */
 static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
@@ -1107,7 +1098,6 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 	if (test_and_clear_bit(__LINK_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
 		__netif_schedule(&dev->tx_queue);
-#endif
 }
 
 /**
@@ -1119,11 +1109,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
  */
 static inline int netif_is_multiqueue(const struct net_device *dev)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	return (!!(NETIF_F_MULTI_QUEUE & dev->features));
-#else
-	return 0;
-#endif
 }
 
 /* Use this variant when it is known for sure that it
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 2220b9e2dab0..8f10e3d08fd9 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -305,9 +305,7 @@ struct sk_buff {
 #endif
 
 	int			iif;
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	__u16			queue_mapping;
-#endif
 #ifdef CONFIG_NET_SCHED
 	__u16			tc_index;	/* traffic control index */
 #ifdef CONFIG_NET_CLS_ACT
@@ -1671,25 +1669,17 @@ static inline void skb_init_secmark(struct sk_buff *skb)
 
 static inline void skb_set_queue_mapping(struct sk_buff *skb, u16 queue_mapping)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	skb->queue_mapping = queue_mapping;
-#endif
 }
 
 static inline u16 skb_get_queue_mapping(struct sk_buff *skb)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	return skb->queue_mapping;
-#else
-	return 0;
-#endif
 }
 
 static inline void skb_copy_queue_mapping(struct sk_buff *to, const struct sk_buff *from)
 {
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
 	to->queue_mapping = from->queue_mapping;
-#endif
 }
 
 static inline int skb_is_gso(const struct sk_buff *skb)
diff --git a/net/mac80211/Kconfig b/net/mac80211/Kconfig
index 40f1add17753..d2038418e2bd 100644
--- a/net/mac80211/Kconfig
+++ b/net/mac80211/Kconfig
@@ -15,14 +15,11 @@ config MAC80211_QOS
 	def_bool y
 	depends on MAC80211
 	depends on NET_SCHED
-	depends on NETDEVICES_MULTIQUEUE
 
 comment "QoS/HT support disabled"
 	depends on MAC80211 && !MAC80211_QOS
 comment "QoS/HT support needs CONFIG_NET_SCHED"
 	depends on MAC80211 && !NET_SCHED
-comment "QoS/HT support needs CONFIG_NETDEVICES_MULTIQUEUE"
-	depends on MAC80211 && !NETDEVICES_MULTIQUEUE
 
 menu "Rate control algorithm selection"
 	depends on MAC80211 != n
-- 
cgit v1.2.3


From 79d16385c7f287a33ea771c4dbe60ae43f791b49 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 8 Jul 2008 23:14:46 -0700
Subject: netdev: Move atomic queue state bits into netdev_queue.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 55 +++++++++++++++++++++++++++++++++--------------
 include/net/pkt_sched.h   |  2 +-
 net/sched/sch_generic.c   | 20 +++++++++--------
 3 files changed, 51 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e2d931f9b700..203c5504fe43 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -281,14 +281,12 @@ struct header_ops {
 
 enum netdev_state_t
 {
-	__LINK_STATE_XOFF=0,
 	__LINK_STATE_START,
 	__LINK_STATE_PRESENT,
 	__LINK_STATE_SCHED,
 	__LINK_STATE_NOCARRIER,
 	__LINK_STATE_LINKWATCH_PENDING,
 	__LINK_STATE_DORMANT,
-	__LINK_STATE_QDISC_RUNNING,
 };
 
 
@@ -448,10 +446,17 @@ static inline void napi_synchronize(const struct napi_struct *n)
 # define napi_synchronize(n)	barrier()
 #endif
 
+enum netdev_queue_state_t
+{
+	__QUEUE_STATE_XOFF,
+	__QUEUE_STATE_QDISC_RUNNING,
+};
+
 struct netdev_queue {
 	spinlock_t		lock;
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
+	unsigned long		state;
 	struct sk_buff		*gso_skb;
 	spinlock_t		_xmit_lock;
 	int			xmit_lock_owner;
@@ -952,9 +957,7 @@ extern void __netif_schedule(struct netdev_queue *txq);
 
 static inline void netif_schedule_queue(struct netdev_queue *txq)
 {
-	struct net_device *dev = txq->dev;
-
-	if (!test_bit(__LINK_STATE_XOFF, &dev->state))
+	if (!test_bit(__QUEUE_STATE_XOFF, &txq->state))
 		__netif_schedule(txq);
 }
 
@@ -969,9 +972,14 @@ static inline void netif_schedule(struct net_device *dev)
  *
  *	Allow upper layers to call the device hard_start_xmit routine.
  */
+static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
+{
+	clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 static inline void netif_start_queue(struct net_device *dev)
 {
-	clear_bit(__LINK_STATE_XOFF, &dev->state);
+	netif_tx_start_queue(&dev->tx_queue);
 }
 
 /**
@@ -981,16 +989,21 @@ static inline void netif_start_queue(struct net_device *dev)
  *	Allow upper layers to call the device hard_start_xmit routine.
  *	Used for flow control when transmit resources are available.
  */
-static inline void netif_wake_queue(struct net_device *dev)
+static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 {
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap()) {
-		clear_bit(__LINK_STATE_XOFF, &dev->state);
+		clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
 		return;
 	}
 #endif
-	if (test_and_clear_bit(__LINK_STATE_XOFF, &dev->state))
-		__netif_schedule(&dev->tx_queue);
+	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state))
+		__netif_schedule(dev_queue);
+}
+
+static inline void netif_wake_queue(struct net_device *dev)
+{
+	netif_tx_wake_queue(&dev->tx_queue);
 }
 
 /**
@@ -1000,9 +1013,14 @@ static inline void netif_wake_queue(struct net_device *dev)
  *	Stop upper layers calling the device hard_start_xmit routine.
  *	Used for flow control when transmit resources are unavailable.
  */
+static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
+{
+	set_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 static inline void netif_stop_queue(struct net_device *dev)
 {
-	set_bit(__LINK_STATE_XOFF, &dev->state);
+	netif_tx_stop_queue(&dev->tx_queue);
 }
 
 /**
@@ -1011,9 +1029,14 @@ static inline void netif_stop_queue(struct net_device *dev)
  *
  *	Test if transmit queue on device is currently unable to send.
  */
+static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
+{
+	return test_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 static inline int netif_queue_stopped(const struct net_device *dev)
 {
-	return test_bit(__LINK_STATE_XOFF, &dev->state);
+	return netif_tx_queue_stopped(&dev->tx_queue);
 }
 
 /**
@@ -1043,7 +1066,7 @@ static inline int netif_running(const struct net_device *dev)
  */
 static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
 {
-	clear_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	clear_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
 }
 
 /**
@@ -1059,7 +1082,7 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 	if (netpoll_trap())
 		return;
 #endif
-	set_bit(__LINK_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	set_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
 }
 
 /**
@@ -1072,7 +1095,7 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 static inline int __netif_subqueue_stopped(const struct net_device *dev,
 					 u16 queue_index)
 {
-	return test_bit(__LINK_STATE_XOFF,
+	return test_bit(__QUEUE_STATE_XOFF,
 			&dev->egress_subqueue[queue_index].state);
 }
 
@@ -1095,7 +1118,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 	if (netpoll_trap())
 		return;
 #endif
-	if (test_and_clear_bit(__LINK_STATE_XOFF,
+	if (test_and_clear_bit(__QUEUE_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
 		__netif_schedule(&dev->tx_queue);
 }
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 2311d242bb35..d58c1a5eb845 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -91,7 +91,7 @@ static inline void qdisc_run(struct netdev_queue *txq)
 	struct net_device *dev = txq->dev;
 
 	if (!netif_queue_stopped(dev) &&
-	    !test_and_set_bit(__LINK_STATE_QDISC_RUNNING, &dev->state))
+	    !test_and_set_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state))
 		__qdisc_run(txq);
 }
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index b6a36d394663..243de935b182 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -121,9 +121,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 /*
  * NOTE: Called under queue->lock with locally disabled BH.
  *
- * __LINK_STATE_QDISC_RUNNING guarantees only one CPU can process this
- * device at a time. queue->lock serializes queue accesses for
- * this device AND txq->qdisc pointer itself.
+ * __QUEUE_STATE_QDISC_RUNNING guarantees only one CPU can process
+ * this queue at a time. queue->lock serializes queue accesses for
+ * this queue AND txq->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
@@ -206,7 +206,7 @@ void __qdisc_run(struct netdev_queue *txq)
 		}
 	}
 
-	clear_bit(__LINK_STATE_QDISC_RUNNING, &dev->state);
+	clear_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state);
 }
 
 static void dev_watchdog(unsigned long arg)
@@ -605,9 +605,10 @@ static void dev_deactivate_queue(struct netdev_queue *dev_queue,
 
 void dev_deactivate(struct net_device *dev)
 {
+	struct netdev_queue *dev_queue = &dev->tx_queue;
 	int running;
 
-	dev_deactivate_queue(&dev->tx_queue, &noop_qdisc);
+	dev_deactivate_queue(dev_queue, &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
@@ -616,16 +617,17 @@ void dev_deactivate(struct net_device *dev)
 
 	/* Wait for outstanding qdisc_run calls. */
 	do {
-		while (test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state))
+		while (test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state))
 			yield();
 
 		/*
 		 * Double-check inside queue lock to ensure that all effects
 		 * of the queue run are visible when we return.
 		 */
-		spin_lock_bh(&dev->tx_queue.lock);
-		running = test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state);
-		spin_unlock_bh(&dev->tx_queue.lock);
+		spin_lock_bh(&dev_queue->lock);
+		running = test_bit(__QUEUE_STATE_QDISC_RUNNING,
+				   &dev_queue->state);
+		spin_unlock_bh(&dev_queue->lock);
 
 		/*
 		 * The running flag should never be set at this point because
-- 
cgit v1.2.3


From b845f313d78e4e259ec449909e3bbadf77b53a6d Mon Sep 17 00:00:00 2001
From: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Date: Tue, 8 Jul 2008 00:28:51 +1000
Subject: mm: Allow architectures to define additional protection bits

This patch allows architectures to define functions to deal with
additional protections bits for mmap() and mprotect().

arch_calc_vm_prot_bits() maps additonal protection bits to vm_flags
arch_vm_get_page_prot() maps additional vm_flags to the vma's vm_page_prot
arch_validate_prot() checks for valid values of the protection bits

Note: vm_get_page_prot() is now pretty ugly, but the generated code
should be identical for architectures that don't define additional
protection bits.

Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/mman.h | 29 ++++++++++++++++++++++++++++-
 mm/mmap.c            |  5 +++--
 mm/mprotect.c        |  2 +-
 3 files changed, 32 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mman.h b/include/linux/mman.h
index dab8892e6ff1..30d1073bac3b 100644
--- a/include/linux/mman.h
+++ b/include/linux/mman.h
@@ -33,6 +33,32 @@ static inline void vm_unacct_memory(long pages)
 	vm_acct_memory(-pages);
 }
 
+/*
+ * Allow architectures to handle additional protection bits
+ */
+
+#ifndef arch_calc_vm_prot_bits
+#define arch_calc_vm_prot_bits(prot) 0
+#endif
+
+#ifndef arch_vm_get_page_prot
+#define arch_vm_get_page_prot(vm_flags) __pgprot(0)
+#endif
+
+#ifndef arch_validate_prot
+/*
+ * This is called from mprotect().  PROT_GROWSDOWN and PROT_GROWSUP have
+ * already been masked out.
+ *
+ * Returns true if the prot flags are valid
+ */
+static inline int arch_validate_prot(unsigned long prot)
+{
+	return (prot & ~(PROT_READ | PROT_WRITE | PROT_EXEC | PROT_SEM)) == 0;
+}
+#define arch_validate_prot arch_validate_prot
+#endif
+
 /*
  * Optimisation macro.  It is equivalent to:
  *      (x & bit1) ? bit2 : 0
@@ -51,7 +77,8 @@ calc_vm_prot_bits(unsigned long prot)
 {
 	return _calc_vm_trans(prot, PROT_READ,  VM_READ ) |
 	       _calc_vm_trans(prot, PROT_WRITE, VM_WRITE) |
-	       _calc_vm_trans(prot, PROT_EXEC,  VM_EXEC );
+	       _calc_vm_trans(prot, PROT_EXEC,  VM_EXEC) |
+	       arch_calc_vm_prot_bits(prot);
 }
 
 /*
diff --git a/mm/mmap.c b/mm/mmap.c
index 3354fdd83d4b..1d102b956fd8 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -72,8 +72,9 @@ pgprot_t protection_map[16] = {
 
 pgprot_t vm_get_page_prot(unsigned long vm_flags)
 {
-	return protection_map[vm_flags &
-				(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)];
+	return __pgprot(pgprot_val(protection_map[vm_flags &
+				(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
+			pgprot_val(arch_vm_get_page_prot(vm_flags)));
 }
 EXPORT_SYMBOL(vm_get_page_prot);
 
diff --git a/mm/mprotect.c b/mm/mprotect.c
index a5bf31c27375..ecfaa5844b5f 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -239,7 +239,7 @@ sys_mprotect(unsigned long start, size_t len, unsigned long prot)
 	end = start + len;
 	if (end <= start)
 		return -ENOMEM;
-	if (prot & ~(PROT_READ | PROT_WRITE | PROT_EXEC | PROT_SEM))
+	if (!arch_validate_prot(prot))
 		return -EINVAL;
 
 	reqprot = prot;
-- 
cgit v1.2.3


From aba46c5027cb59d98052231b36efcbbde9c77a1d Mon Sep 17 00:00:00 2001
From: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Date: Tue, 8 Jul 2008 00:28:52 +1000
Subject: powerpc/mm: Define flags for Strong Access Ordering

This patch defines:

- PROT_SAO, which is passed into mmap() and mprotect() in the prot field
- VM_SAO in vma->vm_flags, and
- _PAGE_SAO, the combination of WIMG bits in the pte that enables strong
access ordering for the page.

Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/asm-powerpc/mman.h          | 2 ++
 include/asm-powerpc/pgtable-ppc64.h | 3 +++
 include/linux/mm.h                  | 1 +
 3 files changed, 6 insertions(+)

(limited to 'include/linux')

diff --git a/include/asm-powerpc/mman.h b/include/asm-powerpc/mman.h
index 24cf664a8295..0c46bf2c7d5f 100644
--- a/include/asm-powerpc/mman.h
+++ b/include/asm-powerpc/mman.h
@@ -10,6 +10,8 @@
  * 2 of the License, or (at your option) any later version.
  */
 
+#define PROT_SAO	0x10		/* Strong Access Ordering */
+
 #define MAP_RENAME      MAP_ANONYMOUS   /* In SunOS terminology */
 #define MAP_NORESERVE   0x40            /* don't reserve swap pages */
 #define MAP_LOCKED	0x80
diff --git a/include/asm-powerpc/pgtable-ppc64.h b/include/asm-powerpc/pgtable-ppc64.h
index b2754d46be44..d09599cccb35 100644
--- a/include/asm-powerpc/pgtable-ppc64.h
+++ b/include/asm-powerpc/pgtable-ppc64.h
@@ -93,6 +93,9 @@
 #define _PAGE_RW	0x0200 /* software: user write access allowed */
 #define _PAGE_BUSY	0x0800 /* software: PTE & hash are busy */
 
+/* Strong Access Ordering */
+#define _PAGE_SAO	(_PAGE_WRITETHRU | _PAGE_NO_CACHE | _PAGE_COHERENT)
+
 #define _PAGE_BASE	(_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_COHERENT)
 
 #define _PAGE_WRENABLE	(_PAGE_RW | _PAGE_DIRTY)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 586a943cab01..689184446fc6 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -108,6 +108,7 @@ extern unsigned int kobjsize(const void *objp);
 
 #define VM_CAN_NONLINEAR 0x08000000	/* Has ->fault & does nonlinear pages */
 #define VM_MIXEDMAP	0x10000000	/* Can contain "struct page" and pure PFN pages */
+#define VM_SAO		0x20000000	/* Strong Access Ordering (powerpc) */
 
 #ifndef VM_STACK_DEFAULT_FLAGS		/* arch can override this */
 #define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS
-- 
cgit v1.2.3


From 2116271a347d1181b5497602c2bfada1de8fd53b Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 20 May 2008 19:34:39 -0400
Subject: NFS: Add correct bounds checking to NFSv2 locks

NFSv2 file locking currently fails the Connectathon tests, because the
calls to the VFS locking code do not return an EINVAL error if the
struct file_lock overflows the 32-bit boundaries.

The problem is due to the fact that we occasionally call helpers from
fs/locks.c in order to avoid RPC calls to the server when we know that a
local process holds the lock. These helpers are, of course, always
64-bit enabled, so EINVAL is not returned in cases when it would if
the call had gone to the NLM code.

For consistency, we therefore add support for a bounds-checking helper.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/file.c           | 20 +++++++++++++++-----
 fs/nfs/proc.c           | 24 ++++++++++++++++++++++++
 include/linux/nfs_xdr.h |  1 +
 3 files changed, 40 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index d84a3d8f32af..7c73f06692b6 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -593,6 +593,7 @@ out:
 static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl)
 {
 	struct inode * inode = filp->f_mapping->host;
+	int ret = -ENOLCK;
 
 	dprintk("NFS: nfs_lock(f=%s/%ld, t=%x, fl=%x, r=%Ld:%Ld)\n",
 			inode->i_sb->s_id, inode->i_ino,
@@ -602,13 +603,22 @@ static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl)
 
 	/* No mandatory locks over NFS */
 	if (__mandatory_lock(inode) && fl->fl_type != F_UNLCK)
-		return -ENOLCK;
+		goto out_err;
+
+	if (NFS_PROTO(inode)->lock_check_bounds != NULL) {
+		ret = NFS_PROTO(inode)->lock_check_bounds(fl);
+		if (ret < 0)
+			goto out_err;
+	}
 
 	if (IS_GETLK(cmd))
-		return do_getlk(filp, cmd, fl);
-	if (fl->fl_type == F_UNLCK)
-		return do_unlk(filp, cmd, fl);
-	return do_setlk(filp, cmd, fl);
+		ret = do_getlk(filp, cmd, fl);
+	else if (fl->fl_type == F_UNLCK)
+		ret = do_unlk(filp, cmd, fl);
+	else
+		ret = do_setlk(filp, cmd, fl);
+out_err:
+	return ret;
 }
 
 /*
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 03599bfe81cf..5c35b02857f3 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -598,6 +598,29 @@ nfs_proc_lock(struct file *filp, int cmd, struct file_lock *fl)
 	return nlmclnt_proc(NFS_SERVER(inode)->nlm_host, cmd, fl);
 }
 
+/* Helper functions for NFS lock bounds checking */
+#define NFS_LOCK32_OFFSET_MAX ((__s32)0x7fffffffUL)
+static int nfs_lock_check_bounds(const struct file_lock *fl)
+{
+	__s32 start, end;
+
+	start = (__s32)fl->fl_start;
+	if ((loff_t)start != fl->fl_start)
+		goto out_einval;
+
+	if (fl->fl_end != OFFSET_MAX) {
+		end = (__s32)fl->fl_end;
+		if ((loff_t)end != fl->fl_end)
+			goto out_einval;
+	} else
+		end = NFS_LOCK32_OFFSET_MAX;
+
+	if (start < 0 || start > end)
+		goto out_einval;
+	return 0;
+out_einval:
+	return -EINVAL;
+}
 
 const struct nfs_rpc_ops nfs_v2_clientops = {
 	.version	= 2,		       /* protocol version */
@@ -633,4 +656,5 @@ const struct nfs_rpc_ops nfs_v2_clientops = {
 	.file_open	= nfs_open,
 	.file_release	= nfs_release,
 	.lock		= nfs_proc_lock,
+	.lock_check_bounds = nfs_lock_check_bounds,
 };
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 24263bb8e0be..8d780de371f0 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -832,6 +832,7 @@ struct nfs_rpc_ops {
 	int	(*file_open)   (struct inode *, struct file *);
 	int	(*file_release) (struct inode *, struct file *);
 	int	(*lock)(struct file *, int, struct file_lock *);
+	int	(*lock_check_bounds)(const struct file_lock *);
 	void	(*clear_acl_cache)(struct inode *);
 };
 
-- 
cgit v1.2.3


From b6b6152c46861dd914d0e6cea9c27df057d6e235 Mon Sep 17 00:00:00 2001
From: Olga Kornievskaia <aglo@citi.umich.edu>
Date: Mon, 9 Jun 2008 16:51:31 -0400
Subject: rpc: bring back cl_chatty

The cl_chatty flag alows us to control whether a given rpc client leaves

	"server X not responding, timed out"

messages in the syslog.  Such messages make sense for ordinary nfs
clients (where an unresponsive server means applications on the
mountpoint are probably hanging), but not for the callback client (which
can fail more commonly, with the only result just of disabling some
optimizations).

Previously cl_chatty was removed, do to lack of users; reinstate it, and
use it for the nfsd's callback client.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfsd/nfs4callback.c      |  2 +-
 include/linux/sunrpc/clnt.h |  4 +++-
 net/sunrpc/clnt.c           | 16 +++++++++++-----
 3 files changed, 15 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 4d4760e687c3..702fa577aa6e 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -381,7 +381,7 @@ static int do_probe_callback(void *data)
 		.program	= &cb_program,
 		.version	= nfs_cb_version[1]->number,
 		.authflavor	= RPC_AUTH_UNIX, /* XXX: need AUTH_GSS... */
-		.flags		= (RPC_CLNT_CREATE_NOPING),
+		.flags		= (RPC_CLNT_CREATE_NOPING | RPC_CLNT_CREATE_QUIET),
 	};
 	struct rpc_message msg = {
 		.rpc_proc       = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_NULL],
diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 6fff7f82ef12..764fd4c286e0 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -42,7 +42,8 @@ struct rpc_clnt {
 
 	unsigned int		cl_softrtry : 1,/* soft timeouts */
 				cl_discrtry : 1,/* disconnect before retry */
-				cl_autobind : 1;/* use getport() */
+				cl_autobind : 1,/* use getport() */
+				cl_chatty   : 1;/* be verbose */
 
 	struct rpc_rtt *	cl_rtt;		/* RTO estimator data */
 	const struct rpc_timeout *cl_timeout;	/* Timeout strategy */
@@ -114,6 +115,7 @@ struct rpc_create_args {
 #define RPC_CLNT_CREATE_NONPRIVPORT	(1UL << 3)
 #define RPC_CLNT_CREATE_NOPING		(1UL << 4)
 #define RPC_CLNT_CREATE_DISCRTRY	(1UL << 5)
+#define RPC_CLNT_CREATE_QUIET		(1UL << 6)
 
 struct rpc_clnt *rpc_create(struct rpc_create_args *args);
 struct rpc_clnt	*rpc_bind_new_program(struct rpc_clnt *,
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 0530eea37b59..09631f6e30e9 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -324,6 +324,8 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args)
 		clnt->cl_autobind = 1;
 	if (args->flags & RPC_CLNT_CREATE_DISCRTRY)
 		clnt->cl_discrtry = 1;
+	if (!(args->flags & RPC_CLNT_CREATE_QUIET))
+		clnt->cl_chatty = 1;
 
 	return clnt;
 }
@@ -1149,7 +1151,8 @@ call_status(struct rpc_task *task)
 		rpc_exit(task, status);
 		break;
 	default:
-		printk("%s: RPC call returned error %d\n",
+		if (clnt->cl_chatty)
+			printk("%s: RPC call returned error %d\n",
 			       clnt->cl_protname, -status);
 		rpc_exit(task, status);
 	}
@@ -1174,7 +1177,8 @@ call_timeout(struct rpc_task *task)
 	task->tk_timeouts++;
 
 	if (RPC_IS_SOFT(task)) {
-		printk(KERN_NOTICE "%s: server %s not responding, timed out\n",
+		if (clnt->cl_chatty)
+			printk(KERN_NOTICE "%s: server %s not responding, timed out\n",
 				clnt->cl_protname, clnt->cl_server);
 		rpc_exit(task, -EIO);
 		return;
@@ -1182,7 +1186,8 @@ call_timeout(struct rpc_task *task)
 
 	if (!(task->tk_flags & RPC_CALL_MAJORSEEN)) {
 		task->tk_flags |= RPC_CALL_MAJORSEEN;
-		printk(KERN_NOTICE "%s: server %s not responding, still trying\n",
+		if (clnt->cl_chatty)
+			printk(KERN_NOTICE "%s: server %s not responding, still trying\n",
 			clnt->cl_protname, clnt->cl_server);
 	}
 	rpc_force_rebind(clnt);
@@ -1213,8 +1218,9 @@ call_decode(struct rpc_task *task)
 			task->tk_pid, task->tk_status);
 
 	if (task->tk_flags & RPC_CALL_MAJORSEEN) {
-		printk(KERN_NOTICE "%s: server %s OK\n",
-			clnt->cl_protname, clnt->cl_server);
+		if (clnt->cl_chatty)
+			printk(KERN_NOTICE "%s: server %s OK\n",
+				clnt->cl_protname, clnt->cl_server);
 		task->tk_flags &= ~RPC_CALL_MAJORSEEN;
 	}
 
-- 
cgit v1.2.3


From a486aeda9b2b0d944aecce7871b3186379b898de Mon Sep 17 00:00:00 2001
From: "\\\\\\\"J. Bruce Fields\\\\\\" <bfields@citi.umich.edu>
Date: Mon, 9 Jun 2008 16:51:35 -0400
Subject: rpc: minor cleanup of scheduler callback code

Try to make the comment here a little more clear and concise.

Also, this macro definition seems unnecessary.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/sched.h |  1 -
 net/sunrpc/sched.c           | 14 +++++---------
 2 files changed, 5 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index d1a5c8c1a0f1..64981a2f1cae 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -135,7 +135,6 @@ struct rpc_task_setup {
 #define RPC_IS_SWAPPER(t)	((t)->tk_flags & RPC_TASK_SWAPPER)
 #define RPC_DO_ROOTOVERRIDE(t)	((t)->tk_flags & RPC_TASK_ROOTCREDS)
 #define RPC_ASSASSINATED(t)	((t)->tk_flags & RPC_TASK_KILLED)
-#define RPC_DO_CALLBACK(t)	((t)->tk_callback != NULL)
 #define RPC_IS_SOFT(t)		((t)->tk_flags & RPC_TASK_SOFT)
 
 #define RPC_TASK_RUNNING	0
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index 6eab9bf94baf..6288af05c20f 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -626,19 +626,15 @@ static void __rpc_execute(struct rpc_task *task)
 		/*
 		 * Execute any pending callback.
 		 */
-		if (RPC_DO_CALLBACK(task)) {
-			/* Define a callback save pointer */
+		if (task->tk_callback) {
 			void (*save_callback)(struct rpc_task *);
 
 			/*
-			 * If a callback exists, save it, reset it,
-			 * call it.
-			 * The save is needed to stop from resetting
-			 * another callback set within the callback handler
-			 * - Dave
+			 * We set tk_callback to NULL before calling it,
+			 * in case it sets the tk_callback field itself:
 			 */
-			save_callback=task->tk_callback;
-			task->tk_callback=NULL;
+			save_callback = task->tk_callback;
+			task->tk_callback = NULL;
 			save_callback(task);
 		}
 
-- 
cgit v1.2.3


From 46cb650c224bb8e64a749090105d74b9e8eda669 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 11 Jun 2008 16:32:46 -0400
Subject: NFS: Remove the redundant file_open entry from struct nfs_rpc_ops

All instances are set to nfs_open(), so we should just remove the redundant
indirection. Ditto for the file_release op

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/file.c           | 4 ++--
 fs/nfs/nfs3proc.c       | 2 --
 fs/nfs/nfs4proc.c       | 2 --
 fs/nfs/proc.c           | 2 --
 include/linux/nfs_xdr.h | 2 --
 5 files changed, 2 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 78b26f875eee..509dcb58959e 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -129,7 +129,7 @@ nfs_file_open(struct inode *inode, struct file *filp)
 
 	nfs_inc_stats(inode, NFSIOS_VFSOPEN);
 	lock_kernel();
-	res = NFS_PROTO(inode)->file_open(inode, filp);
+	res = nfs_open(inode, filp);
 	unlock_kernel();
 	return res;
 }
@@ -147,7 +147,7 @@ nfs_file_release(struct inode *inode, struct file *filp)
 	if (filp->f_mode & FMODE_WRITE)
 		nfs_wb_all(dentry->d_inode);
 	nfs_inc_stats(inode, NFSIOS_VFSRELEASE);
-	return NFS_PROTO(inode)->file_release(inode, filp);
+	return nfs_release(inode, filp);
 }
 
 /**
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index b9c2d995332b..1e750e4574a9 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -816,8 +816,6 @@ const struct nfs_rpc_ops nfs_v3_clientops = {
 	.write_done	= nfs3_write_done,
 	.commit_setup	= nfs3_proc_commit_setup,
 	.commit_done	= nfs3_commit_done,
-	.file_open	= nfs_open,
-	.file_release	= nfs_release,
 	.lock		= nfs3_proc_lock,
 	.clear_acl_cache = nfs3_forget_cached_acls,
 };
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 31a7e4c54a12..dfdd19a6d17e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3714,8 +3714,6 @@ const struct nfs_rpc_ops nfs_v4_clientops = {
 	.write_done	= nfs4_write_done,
 	.commit_setup	= nfs4_proc_commit_setup,
 	.commit_done	= nfs4_commit_done,
-	.file_open      = nfs_open,
-	.file_release   = nfs_release,
 	.lock		= nfs4_proc_lock,
 	.clear_acl_cache = nfs4_zap_acl_attr,
 };
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index c7605587d0eb..4dbb84df1b68 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -655,8 +655,6 @@ const struct nfs_rpc_ops nfs_v2_clientops = {
 	.write_setup	= nfs_proc_write_setup,
 	.write_done	= nfs_write_done,
 	.commit_setup	= nfs_proc_commit_setup,
-	.file_open	= nfs_open,
-	.file_release	= nfs_release,
 	.lock		= nfs_proc_lock,
 	.lock_check_bounds = nfs_lock_check_bounds,
 };
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 8d780de371f0..8c77c11224d1 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -829,8 +829,6 @@ struct nfs_rpc_ops {
 	int	(*write_done)  (struct rpc_task *, struct nfs_write_data *);
 	void	(*commit_setup) (struct nfs_write_data *, struct rpc_message *);
 	int	(*commit_done) (struct rpc_task *, struct nfs_write_data *);
-	int	(*file_open)   (struct inode *, struct file *);
-	int	(*file_release) (struct inode *, struct file *);
 	int	(*lock)(struct file *, int, struct file_lock *);
 	int	(*lock_check_bounds)(const struct file_lock *);
 	void	(*clear_acl_cache)(struct inode *);
-- 
cgit v1.2.3


From 34e8f92831cb5c40b3137e47a3daf4c09016ef02 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 12 Jun 2008 12:32:25 -0400
Subject: NFS: Move fs/nfs/iostat.h to include/linux

The fs/nfs/iostat.h header has definitions that were designed to be exposed
to user space.  Move these definitions under include/linux so user space can
use the definitions in applications that read /proc/self/mountstats.

Also address a handful of coding style issues called out by checkpatch.pl in
fs/nfs/iostat.h.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/iostat.h            | 119 +++++----------------------------------------
 include/linux/nfs_iostat.h | 119 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 132 insertions(+), 106 deletions(-)
 create mode 100644 include/linux/nfs_iostat.h

(limited to 'include/linux')

diff --git a/fs/nfs/iostat.h b/fs/nfs/iostat.h
index 6350ecbde589..2ec65e12bfed 100644
--- a/fs/nfs/iostat.h
+++ b/fs/nfs/iostat.h
@@ -5,135 +5,41 @@
  *
  *  Copyright (C) 2005, 2006 Chuck Lever <cel@netapp.com>
  *
- *  NFS client per-mount statistics provide information about the health of
- *  the NFS client and the health of each NFS mount point.  Generally these
- *  are not for detailed problem diagnosis, but simply to indicate that there
- *  is a problem.
- *
- *  These counters are not meant to be human-readable, but are meant to be
- *  integrated into system monitoring tools such as "sar" and "iostat".  As
- *  such, the counters are sampled by the tools over time, and are never
- *  zeroed after a file system is mounted.  Moving averages can be computed
- *  by the tools by taking the difference between two instantaneous samples
- *  and dividing that by the time between the samples.
  */
 
 #ifndef _NFS_IOSTAT
 #define _NFS_IOSTAT
 
-#define NFS_IOSTAT_VERS		"1.0"
-
-/*
- * NFS byte counters
- *
- * 1.  SERVER - the number of payload bytes read from or written to the
- *     server by the NFS client via an NFS READ or WRITE request.
- *
- * 2.  NORMAL - the number of bytes read or written by applications via
- *     the read(2) and write(2) system call interfaces.
- *
- * 3.  DIRECT - the number of bytes read or written from files opened
- *     with the O_DIRECT flag.
- *
- * These counters give a view of the data throughput into and out of the NFS
- * client.  Comparing the number of bytes requested by an application with the
- * number of bytes the client requests from the server can provide an
- * indication of client efficiency (per-op, cache hits, etc).
- *
- * These counters can also help characterize which access methods are in
- * use.  DIRECT by itself shows whether there is any O_DIRECT traffic.
- * NORMAL + DIRECT shows how much data is going through the system call
- * interface.  A large amount of SERVER traffic without much NORMAL or
- * DIRECT traffic shows that applications are using mapped files.
- *
- * NFS page counters
- *
- * These count the number of pages read or written via nfs_readpage(),
- * nfs_readpages(), or their write equivalents.
- */
-enum nfs_stat_bytecounters {
-	NFSIOS_NORMALREADBYTES = 0,
-	NFSIOS_NORMALWRITTENBYTES,
-	NFSIOS_DIRECTREADBYTES,
-	NFSIOS_DIRECTWRITTENBYTES,
-	NFSIOS_SERVERREADBYTES,
-	NFSIOS_SERVERWRITTENBYTES,
-	NFSIOS_READPAGES,
-	NFSIOS_WRITEPAGES,
-	__NFSIOS_BYTESMAX,
-};
-
-/*
- * NFS event counters
- *
- * These counters provide a low-overhead way of monitoring client activity
- * without enabling NFS trace debugging.  The counters show the rate at
- * which VFS requests are made, and how often the client invalidates its
- * data and attribute caches.  This allows system administrators to monitor
- * such things as how close-to-open is working, and answer questions such
- * as "why are there so many GETATTR requests on the wire?"
- *
- * They also count anamolous events such as short reads and writes, silly
- * renames due to close-after-delete, and operations that change the size
- * of a file (such operations can often be the source of data corruption
- * if applications aren't using file locking properly).
- */
-enum nfs_stat_eventcounters {
-	NFSIOS_INODEREVALIDATE = 0,
-	NFSIOS_DENTRYREVALIDATE,
-	NFSIOS_DATAINVALIDATE,
-	NFSIOS_ATTRINVALIDATE,
-	NFSIOS_VFSOPEN,
-	NFSIOS_VFSLOOKUP,
-	NFSIOS_VFSACCESS,
-	NFSIOS_VFSUPDATEPAGE,
-	NFSIOS_VFSREADPAGE,
-	NFSIOS_VFSREADPAGES,
-	NFSIOS_VFSWRITEPAGE,
-	NFSIOS_VFSWRITEPAGES,
-	NFSIOS_VFSGETDENTS,
-	NFSIOS_VFSSETATTR,
-	NFSIOS_VFSFLUSH,
-	NFSIOS_VFSFSYNC,
-	NFSIOS_VFSLOCK,
-	NFSIOS_VFSRELEASE,
-	NFSIOS_CONGESTIONWAIT,
-	NFSIOS_SETATTRTRUNC,
-	NFSIOS_EXTENDWRITE,
-	NFSIOS_SILLYRENAME,
-	NFSIOS_SHORTREAD,
-	NFSIOS_SHORTWRITE,
-	NFSIOS_DELAY,
-	__NFSIOS_COUNTSMAX,
-};
-
-#ifdef __KERNEL__
-
 #include <linux/percpu.h>
 #include <linux/cache.h>
+#include <linux/nfs_iostat.h>
 
 struct nfs_iostats {
 	unsigned long long	bytes[__NFSIOS_BYTESMAX];
 	unsigned long		events[__NFSIOS_COUNTSMAX];
 } ____cacheline_aligned;
 
-static inline void nfs_inc_server_stats(struct nfs_server *server, enum nfs_stat_eventcounters stat)
+static inline void nfs_inc_server_stats(struct nfs_server *server,
+					enum nfs_stat_eventcounters stat)
 {
 	struct nfs_iostats *iostats;
 	int cpu;
 
 	cpu = get_cpu();
 	iostats = per_cpu_ptr(server->io_stats, cpu);
-	iostats->events[stat] ++;
+	iostats->events[stat]++;
 	put_cpu_no_resched();
 }
 
-static inline void nfs_inc_stats(struct inode *inode, enum nfs_stat_eventcounters stat)
+static inline void nfs_inc_stats(struct inode *inode,
+				 enum nfs_stat_eventcounters stat)
 {
 	nfs_inc_server_stats(NFS_SERVER(inode), stat);
 }
 
-static inline void nfs_add_server_stats(struct nfs_server *server, enum nfs_stat_bytecounters stat, unsigned long addend)
+static inline void nfs_add_server_stats(struct nfs_server *server,
+					enum nfs_stat_bytecounters stat,
+					unsigned long addend)
 {
 	struct nfs_iostats *iostats;
 	int cpu;
@@ -144,7 +50,9 @@ static inline void nfs_add_server_stats(struct nfs_server *server, enum nfs_stat
 	put_cpu_no_resched();
 }
 
-static inline void nfs_add_stats(struct inode *inode, enum nfs_stat_bytecounters stat, unsigned long addend)
+static inline void nfs_add_stats(struct inode *inode,
+				 enum nfs_stat_bytecounters stat,
+				 unsigned long addend)
 {
 	nfs_add_server_stats(NFS_SERVER(inode), stat, addend);
 }
@@ -160,5 +68,4 @@ static inline void nfs_free_iostats(struct nfs_iostats *stats)
 		free_percpu(stats);
 }
 
-#endif
-#endif
+#endif /* _NFS_IOSTAT */
diff --git a/include/linux/nfs_iostat.h b/include/linux/nfs_iostat.h
new file mode 100644
index 000000000000..1cb9a3fed2b3
--- /dev/null
+++ b/include/linux/nfs_iostat.h
@@ -0,0 +1,119 @@
+/*
+ *  User-space visible declarations for NFS client per-mount
+ *  point statistics
+ *
+ *  Copyright (C) 2005, 2006 Chuck Lever <cel@netapp.com>
+ *
+ *  NFS client per-mount statistics provide information about the
+ *  health of the NFS client and the health of each NFS mount point.
+ *  Generally these are not for detailed problem diagnosis, but
+ *  simply to indicate that there is a problem.
+ *
+ *  These counters are not meant to be human-readable, but are meant
+ *  to be integrated into system monitoring tools such as "sar" and
+ *  "iostat".  As such, the counters are sampled by the tools over
+ *  time, and are never zeroed after a file system is mounted.
+ *  Moving averages can be computed by the tools by taking the
+ *  difference between two instantaneous samples  and dividing that
+ *  by the time between the samples.
+ */
+
+#ifndef _LINUX_NFS_IOSTAT
+#define _LINUX_NFS_IOSTAT
+
+#define NFS_IOSTAT_VERS		"1.0"
+
+/*
+ * NFS byte counters
+ *
+ * 1.  SERVER - the number of payload bytes read from or written
+ *     to the server by the NFS client via an NFS READ or WRITE
+ *     request.
+ *
+ * 2.  NORMAL - the number of bytes read or written by applications
+ *     via the read(2) and write(2) system call interfaces.
+ *
+ * 3.  DIRECT - the number of bytes read or written from files
+ *     opened with the O_DIRECT flag.
+ *
+ * These counters give a view of the data throughput into and out
+ * of the NFS client.  Comparing the number of bytes requested by
+ * an application with the number of bytes the client requests from
+ * the server can provide an indication of client efficiency
+ * (per-op, cache hits, etc).
+ *
+ * These counters can also help characterize which access methods
+ * are in use.  DIRECT by itself shows whether there is any O_DIRECT
+ * traffic.  NORMAL + DIRECT shows how much data is going through
+ * the system call interface.  A large amount of SERVER traffic
+ * without much NORMAL or DIRECT traffic shows that applications
+ * are using mapped files.
+ *
+ * NFS page counters
+ *
+ * These count the number of pages read or written via nfs_readpage(),
+ * nfs_readpages(), or their write equivalents.
+ *
+ * NB: When adding new byte counters, please include the measured
+ * units in the name of each byte counter to help users of this
+ * interface determine what exactly is being counted.
+ */
+enum nfs_stat_bytecounters {
+	NFSIOS_NORMALREADBYTES = 0,
+	NFSIOS_NORMALWRITTENBYTES,
+	NFSIOS_DIRECTREADBYTES,
+	NFSIOS_DIRECTWRITTENBYTES,
+	NFSIOS_SERVERREADBYTES,
+	NFSIOS_SERVERWRITTENBYTES,
+	NFSIOS_READPAGES,
+	NFSIOS_WRITEPAGES,
+	__NFSIOS_BYTESMAX,
+};
+
+/*
+ * NFS event counters
+ *
+ * These counters provide a low-overhead way of monitoring client
+ * activity without enabling NFS trace debugging.  The counters
+ * show the rate at which VFS requests are made, and how often the
+ * client invalidates its data and attribute caches.  This allows
+ * system administrators to monitor such things as how close-to-open
+ * is working, and answer questions such as "why are there so many
+ * GETATTR requests on the wire?"
+ *
+ * They also count anamolous events such as short reads and writes,
+ * silly renames due to close-after-delete, and operations that
+ * change the size of a file (such operations can often be the
+ * source of data corruption if applications aren't using file
+ * locking properly).
+ */
+enum nfs_stat_eventcounters {
+	NFSIOS_INODEREVALIDATE = 0,
+	NFSIOS_DENTRYREVALIDATE,
+	NFSIOS_DATAINVALIDATE,
+	NFSIOS_ATTRINVALIDATE,
+	NFSIOS_VFSOPEN,
+	NFSIOS_VFSLOOKUP,
+	NFSIOS_VFSACCESS,
+	NFSIOS_VFSUPDATEPAGE,
+	NFSIOS_VFSREADPAGE,
+	NFSIOS_VFSREADPAGES,
+	NFSIOS_VFSWRITEPAGE,
+	NFSIOS_VFSWRITEPAGES,
+	NFSIOS_VFSGETDENTS,
+	NFSIOS_VFSSETATTR,
+	NFSIOS_VFSFLUSH,
+	NFSIOS_VFSFSYNC,
+	NFSIOS_VFSLOCK,
+	NFSIOS_VFSRELEASE,
+	NFSIOS_CONGESTIONWAIT,
+	NFSIOS_SETATTRTRUNC,
+	NFSIOS_EXTENDWRITE,
+	NFSIOS_SILLYRENAME,
+	NFSIOS_SHORTREAD,
+	NFSIOS_SHORTWRITE,
+	NFSIOS_DELAY,
+	__NFSIOS_COUNTSMAX,
+};
+
+#endif	/* _LINUX_NFS_IOSTAT */
-- 
cgit v1.2.3


From e468bae97d243fe0e1515abaa1f7d0edf1476ad0 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Fri, 13 Jun 2008 13:25:22 -0400
Subject: NFS: Allow redirtying of a completed unstable write.

Currently, if an unstable write completes, we cannot redirty the page in
order to reflect a new change in the page data until after we've sent a
COMMIT request.

This patch allows a page rewrite to proceed without the unnecessary COMMIT
step, putting it immediately back onto the dirty page list, undoing the
VM unstable write accounting, and removing the NFS_PAGE_TAG_COMMIT tag from
the NFS radix tree.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/write.c           | 65 ++++++++++++++++++++++++------------------------
 include/linux/nfs_page.h |  9 ++++---
 2 files changed, 38 insertions(+), 36 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 04f51e52e184..feca8c648766 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -242,12 +242,9 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio,
 			return ret;
 		spin_lock(&inode->i_lock);
 	}
-	if (test_bit(PG_NEED_COMMIT, &req->wb_flags)) {
-		/* This request is marked for commit */
+	if (test_bit(PG_CLEAN, &req->wb_flags)) {
 		spin_unlock(&inode->i_lock);
-		nfs_clear_page_tag_locked(req);
-		nfs_pageio_complete(pgio);
-		return 0;
+		BUG();
 	}
 	if (nfs_set_page_writeback(page) != 0) {
 		spin_unlock(&inode->i_lock);
@@ -391,19 +388,6 @@ nfs_mark_request_dirty(struct nfs_page *req)
 	__set_page_dirty_nobuffers(req->wb_page);
 }
 
-/*
- * Check if a request is dirty
- */
-static inline int
-nfs_dirty_request(struct nfs_page *req)
-{
-	struct page *page = req->wb_page;
-
-	if (page == NULL || test_bit(PG_NEED_COMMIT, &req->wb_flags))
-		return 0;
-	return !PageWriteback(page);
-}
-
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 /*
  * Add a request to the inode's commit list.
@@ -416,7 +400,7 @@ nfs_mark_request_commit(struct nfs_page *req)
 
 	spin_lock(&inode->i_lock);
 	nfsi->ncommit++;
-	set_bit(PG_NEED_COMMIT, &(req)->wb_flags);
+	set_bit(PG_CLEAN, &(req)->wb_flags);
 	radix_tree_tag_set(&nfsi->nfs_page_tree,
 			req->wb_index,
 			NFS_PAGE_TAG_COMMIT);
@@ -426,6 +410,19 @@ nfs_mark_request_commit(struct nfs_page *req)
 	__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
 }
 
+static int
+nfs_clear_request_commit(struct nfs_page *req)
+{
+	struct page *page = req->wb_page;
+
+	if (test_and_clear_bit(PG_CLEAN, &(req)->wb_flags)) {
+		dec_zone_page_state(page, NR_UNSTABLE_NFS);
+		dec_bdi_stat(page->mapping->backing_dev_info, BDI_RECLAIMABLE);
+		return 1;
+	}
+	return 0;
+}
+
 static inline
 int nfs_write_need_commit(struct nfs_write_data *data)
 {
@@ -435,7 +432,7 @@ int nfs_write_need_commit(struct nfs_write_data *data)
 static inline
 int nfs_reschedule_unstable_write(struct nfs_page *req)
 {
-	if (test_bit(PG_NEED_COMMIT, &req->wb_flags)) {
+	if (test_and_clear_bit(PG_NEED_COMMIT, &req->wb_flags)) {
 		nfs_mark_request_commit(req);
 		return 1;
 	}
@@ -451,6 +448,12 @@ nfs_mark_request_commit(struct nfs_page *req)
 {
 }
 
+static inline int
+nfs_clear_request_commit(struct nfs_page *req)
+{
+	return 0;
+}
+
 static inline
 int nfs_write_need_commit(struct nfs_write_data *data)
 {
@@ -508,11 +511,8 @@ static void nfs_cancel_commit_list(struct list_head *head)
 
 	while(!list_empty(head)) {
 		req = nfs_list_entry(head->next);
-		dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
-		dec_bdi_stat(req->wb_page->mapping->backing_dev_info,
-				BDI_RECLAIMABLE);
 		nfs_list_remove_request(req);
-		clear_bit(PG_NEED_COMMIT, &(req)->wb_flags);
+		nfs_clear_request_commit(req);
 		nfs_inode_remove_request(req);
 		nfs_unlock_request(req);
 	}
@@ -584,8 +584,7 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode,
 		 * Note: nfs_flush_incompatible() will already
 		 * have flushed out requests having wrong owners.
 		 */
-		if (!nfs_dirty_request(req)
-		    || offset > rqend
+		if (offset > rqend
 		    || end < req->wb_offset)
 			goto out_flushme;
 
@@ -601,6 +600,10 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode,
 		spin_lock(&inode->i_lock);
 	}
 
+	if (nfs_clear_request_commit(req))
+		radix_tree_tag_clear(&NFS_I(inode)->nfs_page_tree,
+				req->wb_index, NFS_PAGE_TAG_COMMIT);
+
 	/* Okay, the request matches. Update the region */
 	if (offset < req->wb_offset) {
 		req->wb_offset = offset;
@@ -682,8 +685,7 @@ int nfs_flush_incompatible(struct file *file, struct page *page)
 		req = nfs_page_find_request(page);
 		if (req == NULL)
 			return 0;
-		do_flush = req->wb_page != page || req->wb_context != ctx
-			|| !nfs_dirty_request(req);
+		do_flush = req->wb_page != page || req->wb_context != ctx;
 		nfs_release_request(req);
 		if (!do_flush)
 			return 0;
@@ -1288,10 +1290,7 @@ static void nfs_commit_release(void *calldata)
 	while (!list_empty(&data->pages)) {
 		req = nfs_list_entry(data->pages.next);
 		nfs_list_remove_request(req);
-		clear_bit(PG_NEED_COMMIT, &(req)->wb_flags);
-		dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
-		dec_bdi_stat(req->wb_page->mapping->backing_dev_info,
-				BDI_RECLAIMABLE);
+		nfs_clear_request_commit(req);
 
 		dprintk("NFS:       commit (%s/%lld %d@%lld)",
 			req->wb_context->path.dentry->d_inode->i_sb->s_id,
@@ -1467,7 +1466,7 @@ int nfs_wb_page_cancel(struct inode *inode, struct page *page)
 		req = nfs_page_find_request(page);
 		if (req == NULL)
 			goto out;
-		if (test_bit(PG_NEED_COMMIT, &req->wb_flags)) {
+		if (test_bit(PG_CLEAN, &req->wb_flags)) {
 			nfs_release_request(req);
 			break;
 		}
diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h
index a1676e19e491..3c60685d972b 100644
--- a/include/linux/nfs_page.h
+++ b/include/linux/nfs_page.h
@@ -27,9 +27,12 @@
 /*
  * Valid flags for a dirty buffer
  */
-#define PG_BUSY			0
-#define PG_NEED_COMMIT		1
-#define PG_NEED_RESCHED		2
+enum {
+	PG_BUSY = 0,
+	PG_CLEAN,
+	PG_NEED_COMMIT,
+	PG_NEED_RESCHED,
+};
 
 struct nfs_inode;
 struct nfs_page {
-- 
cgit v1.2.3


From ce3b7e1906ebbe96753fe090b36de6ffb8e0e0e7 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 23 Jun 2008 12:36:53 -0400
Subject: NFS: Add string length argument to nfs_parse_server_address

To make nfs_parse_server_address() more generally useful, allow it to
accept input strings that are not terminated with '\0'.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/super.c       | 100 +++++++++++++++++++++++++++++++++++----------------
 include/linux/inet.h |   7 ++++
 2 files changed, 77 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index ea4abd266a7a..d27aa1db0074 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -705,38 +705,76 @@ static int nfs_verify_server_address(struct sockaddr *addr)
 	return 0;
 }
 
-/*
- * Parse string addresses passed in via a mount option,
- * and construct a sockaddr based on the result.
- *
- * If address parsing fails, set the sockaddr's address
- * family to AF_UNSPEC to force nfs_verify_server_address()
- * to punt the mount.
- */
-static void nfs_parse_server_address(char *value,
-				     struct sockaddr *sap,
-				     size_t *len)
+static void nfs_parse_ipv4_address(char *string, size_t str_len,
+				   struct sockaddr *sap, size_t *addr_len)
 {
-	if (strchr(value, ':')) {
-		struct sockaddr_in6 *ap = (struct sockaddr_in6 *)sap;
-		u8 *addr = (u8 *)&ap->sin6_addr.in6_u;
+	struct sockaddr_in *sin = (struct sockaddr_in *)sap;
+	u8 *addr = (u8 *)&sin->sin_addr.s_addr;
 
-		ap->sin6_family = AF_INET6;
-		*len = sizeof(*ap);
-		if (in6_pton(value, -1, addr, '\0', NULL))
+	if (str_len <= INET_ADDRSTRLEN) {
+		dfprintk(MOUNT, "NFS: parsing IPv4 address %*s\n",
+				(int)str_len, string);
+
+		sin->sin_family = AF_INET;
+		*addr_len = sizeof(*sin);
+		if (in4_pton(string, str_len, addr, '\0', NULL))
 			return;
-	} else {
-		struct sockaddr_in *ap = (struct sockaddr_in *)sap;
-		u8 *addr = (u8 *)&ap->sin_addr.s_addr;
+	}
+
+	sap->sa_family = AF_UNSPEC;
+	*addr_len = 0;
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+static void nfs_parse_ipv6_address(char *string, size_t str_len,
+				   struct sockaddr *sap, size_t *addr_len)
+{
+	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
+	u8 *addr = (u8 *)&sin6->sin6_addr.in6_u;
+
+	if (str_len <= INET6_ADDRSTRLEN) {
+		dfprintk(MOUNT, "NFS: parsing IPv6 address %*s\n",
+				(int)str_len, string);
 
-		ap->sin_family = AF_INET;
-		*len = sizeof(*ap);
-		if (in4_pton(value, -1, addr, '\0', NULL))
+		sin6->sin6_family = AF_INET6;
+		*addr_len = sizeof(*sin6);
+		if (in6_pton(string, str_len, addr, '\0', NULL))
 			return;
 	}
 
 	sap->sa_family = AF_UNSPEC;
-	*len = 0;
+	*addr_len = 0;
+}
+#else
+static void nfs_parse_ipv6_address(char *string, size_t str_len,
+				   struct sockaddr *sap, size_t *addr_len)
+{
+	sap->sa_family = AF_UNSPEC;
+	*addr_len = 0;
+}
+#endif
+
+/*
+ * Construct a sockaddr based on the contents of a string that contains
+ * an IP address in presentation format.
+ *
+ * If there is a problem constructing the new sockaddr, set the address
+ * family to AF_UNSPEC.
+ */
+static void nfs_parse_ip_address(char *string, size_t str_len,
+				 struct sockaddr *sap, size_t *addr_len)
+{
+	unsigned int i, colons;
+
+	colons = 0;
+	for (i = 0; i < str_len; i++)
+		if (string[i] == ':')
+			colons++;
+
+	if (colons >= 2)
+		nfs_parse_ipv6_address(string, str_len, sap, addr_len);
+	else
+		nfs_parse_ipv4_address(string, str_len, sap, addr_len);
 }
 
 /*
@@ -1070,9 +1108,10 @@ static int nfs_parse_mount_options(char *raw,
 			string = match_strdup(args);
 			if (string == NULL)
 				goto out_nomem;
-			nfs_parse_server_address(string, (struct sockaddr *)
-						 &mnt->nfs_server.address,
-						 &mnt->nfs_server.addrlen);
+			nfs_parse_ip_address(string, strlen(string),
+					     (struct sockaddr *)
+						&mnt->nfs_server.address,
+					     &mnt->nfs_server.addrlen);
 			kfree(string);
 			break;
 		case Opt_clientaddr:
@@ -1093,9 +1132,10 @@ static int nfs_parse_mount_options(char *raw,
 			string = match_strdup(args);
 			if (string == NULL)
 				goto out_nomem;
-			nfs_parse_server_address(string, (struct sockaddr *)
-						 &mnt->mount_server.address,
-						 &mnt->mount_server.addrlen);
+			nfs_parse_ip_address(string, strlen(string),
+					     (struct sockaddr *)
+						&mnt->mount_server.address,
+					     &mnt->mount_server.addrlen);
 			kfree(string);
 			break;
 
diff --git a/include/linux/inet.h b/include/linux/inet.h
index 1354080cf8cf..4cca05c9678e 100644
--- a/include/linux/inet.h
+++ b/include/linux/inet.h
@@ -44,6 +44,13 @@
 
 #include <linux/types.h>
 
+/*
+ * These mimic similar macros defined in user-space for inet_ntop(3).
+ * See /usr/include/netinet/in.h .
+ */
+#define INET_ADDRSTRLEN		(16)
+#define INET6_ADDRSTRLEN	(48)
+
 extern __be32 in_aton(const char *str);
 extern int in4_pton(const char *src, int srclen, u8 *dst, int delim, const char **end);
 extern int in6_pton(const char *src, int srclen, u8 *dst, int delim, const char **end);
-- 
cgit v1.2.3


From 259875efed06d6936f54c9a264e868937f1bc217 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 2 Jul 2008 14:43:47 -0400
Subject: NFS: set transport defaults after mount option parsing is finished

Move the UDP/TCP default timeo/retrans settings for text mounts to
nfs_init_timeout_values(), which was were they were always being
initialised (and sanity checked) for binary mounts.
Document the default timeout values using appropriate #defines.

Ensure that we initialise and sanity check the transport protocols that
may have been specified by the user.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c        | 13 ++++++-----
 fs/nfs/super.c         | 60 +++++++++++++++++++++++++++++++++++---------------
 include/linux/nfs_fs.h |  5 +++++
 3 files changed, 55 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index f2a092ca69b5..5ee23e7058b3 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -431,14 +431,14 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 {
 	to->to_initval = timeo * HZ / 10;
 	to->to_retries = retrans;
-	if (!to->to_retries)
-		to->to_retries = 2;
 
 	switch (proto) {
 	case XPRT_TRANSPORT_TCP:
 	case XPRT_TRANSPORT_RDMA:
+		if (to->to_retries == 0)
+			to->to_retries = NFS_DEF_TCP_RETRANS;
 		if (to->to_initval == 0)
-			to->to_initval = 60 * HZ;
+			to->to_initval = NFS_DEF_TCP_TIMEO * HZ / 10;
 		if (to->to_initval > NFS_MAX_TCP_TIMEOUT)
 			to->to_initval = NFS_MAX_TCP_TIMEOUT;
 		to->to_increment = to->to_initval;
@@ -450,14 +450,17 @@ static void nfs_init_timeout_values(struct rpc_timeout *to, int proto,
 		to->to_exponential = 0;
 		break;
 	case XPRT_TRANSPORT_UDP:
-	default:
+		if (to->to_retries == 0)
+			to->to_retries = NFS_DEF_UDP_RETRANS;
 		if (!to->to_initval)
-			to->to_initval = 11 * HZ / 10;
+			to->to_initval = NFS_DEF_UDP_TIMEO * HZ / 10;
 		if (to->to_initval > NFS_MAX_UDP_TIMEOUT)
 			to->to_initval = NFS_MAX_UDP_TIMEOUT;
 		to->to_maxval = NFS_MAX_UDP_TIMEOUT;
 		to->to_exponential = 1;
 		break;
+	default:
+		BUG();
 	}
 }
 
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 73a8e5970f02..9c1a960f5b94 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -816,6 +816,43 @@ static void nfs_parse_ip_address(char *string, size_t str_len,
 		nfs_parse_ipv4_address(string, str_len, sap, addr_len);
 }
 
+/*
+ * Sanity check the NFS transport protocol.
+ *
+ */
+static void nfs_validate_transport_protocol(struct nfs_parsed_mount_data *mnt)
+{
+	switch (mnt->nfs_server.protocol) {
+	case XPRT_TRANSPORT_UDP:
+	case XPRT_TRANSPORT_TCP:
+	case XPRT_TRANSPORT_RDMA:
+		break;
+	default:
+		mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
+	}
+}
+
+/*
+ * For text based NFSv2/v3 mounts, the mount protocol transport default
+ * settings should depend upon the specified NFS transport.
+ */
+static void nfs_set_mount_transport_protocol(struct nfs_parsed_mount_data *mnt)
+{
+	nfs_validate_transport_protocol(mnt);
+
+	if (mnt->mount_server.protocol == XPRT_TRANSPORT_UDP ||
+	    mnt->mount_server.protocol == XPRT_TRANSPORT_TCP)
+			return;
+	switch (mnt->nfs_server.protocol) {
+	case XPRT_TRANSPORT_UDP:
+		mnt->mount_server.protocol = XPRT_TRANSPORT_UDP;
+		break;
+	case XPRT_TRANSPORT_TCP:
+	case XPRT_TRANSPORT_RDMA:
+		mnt->mount_server.protocol = XPRT_TRANSPORT_TCP;
+	}
+}
+
 /*
  * Error-check and convert a string of mount options from user space into
  * a data structure
@@ -896,20 +933,14 @@ static int nfs_parse_mount_options(char *raw,
 		case Opt_udp:
 			mnt->flags &= ~NFS_MOUNT_TCP;
 			mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP;
-			mnt->timeo = 7;
-			mnt->retrans = 5;
 			break;
 		case Opt_tcp:
 			mnt->flags |= NFS_MOUNT_TCP;
 			mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
-			mnt->timeo = 600;
-			mnt->retrans = 2;
 			break;
 		case Opt_rdma:
 			mnt->flags |= NFS_MOUNT_TCP; /* for side protocols */
 			mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
-			mnt->timeo = 600;
-			mnt->retrans = 2;
 			break;
 		case Opt_acl:
 			mnt->flags &= ~NFS_MOUNT_NOACL;
@@ -1103,21 +1134,15 @@ static int nfs_parse_mount_options(char *raw,
 			case Opt_xprt_udp:
 				mnt->flags &= ~NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP;
-				mnt->timeo = 7;
-				mnt->retrans = 5;
 				break;
 			case Opt_xprt_tcp:
 				mnt->flags |= NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
-				mnt->timeo = 600;
-				mnt->retrans = 2;
 				break;
 			case Opt_xprt_rdma:
 				/* vector side protocols to TCP */
 				mnt->flags |= NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
-				mnt->timeo = 600;
-				mnt->retrans = 2;
 				break;
 			default:
 				goto out_unrec_xprt;
@@ -1438,14 +1463,11 @@ static int nfs_validate_mount_data(void *options,
 	args->flags		= (NFS_MOUNT_VER3 | NFS_MOUNT_TCP);
 	args->rsize		= NFS_MAX_FILE_IO_SIZE;
 	args->wsize		= NFS_MAX_FILE_IO_SIZE;
-	args->timeo		= 600;
-	args->retrans		= 2;
 	args->acregmin		= 3;
 	args->acregmax		= 60;
 	args->acdirmin		= 30;
 	args->acdirmax		= 60;
 	args->mount_server.port	= 0;	/* autobind unless user sets port */
-	args->mount_server.protocol = XPRT_TRANSPORT_UDP;
 	args->nfs_server.port	= 0;	/* autobind unless user sets port */
 	args->nfs_server.protocol = XPRT_TRANSPORT_TCP;
 
@@ -1546,6 +1568,8 @@ static int nfs_validate_mount_data(void *options,
 						&args->nfs_server.address))
 			goto out_no_address;
 
+		nfs_set_mount_transport_protocol(args);
+
 		status = nfs_parse_devname(dev_name,
 					   &args->nfs_server.hostname,
 					   PAGE_SIZE,
@@ -2095,14 +2119,11 @@ static int nfs4_validate_mount_data(void *options,
 
 	args->rsize		= NFS_MAX_FILE_IO_SIZE;
 	args->wsize		= NFS_MAX_FILE_IO_SIZE;
-	args->timeo		= 600;
-	args->retrans		= 2;
 	args->acregmin		= 3;
 	args->acregmax		= 60;
 	args->acdirmin		= 30;
 	args->acdirmax		= 60;
 	args->nfs_server.port	= NFS_PORT; /* 2049 unless user set port= */
-	args->nfs_server.protocol = XPRT_TRANSPORT_TCP;
 
 	switch (data->version) {
 	case 1:
@@ -2163,6 +2184,7 @@ static int nfs4_validate_mount_data(void *options,
 		args->acdirmin	= data->acdirmin;
 		args->acdirmax	= data->acdirmax;
 		args->nfs_server.protocol = data->proto;
+		nfs_validate_transport_protocol(args);
 
 		break;
 	default: {
@@ -2175,6 +2197,8 @@ static int nfs4_validate_mount_data(void *options,
 						&args->nfs_server.address))
 			return -EINVAL;
 
+		nfs_validate_transport_protocol(args);
+
 		switch (args->auth_flavor_len) {
 		case 0:
 			args->auth_flavors[0] = RPC_AUTH_UNIX;
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 27d6a8d98cef..830d9cc8cdce 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -12,6 +12,11 @@
 #include <linux/magic.h>
 
 /* Default timeout values */
+#define NFS_DEF_UDP_TIMEO	(11)
+#define NFS_DEF_UDP_RETRANS	(3)
+#define NFS_DEF_TCP_TIMEO	(600)
+#define NFS_DEF_TCP_RETRANS	(2)
+
 #define NFS_MAX_UDP_TIMEOUT	(60*HZ)
 #define NFS_MAX_TCP_TIMEOUT	(600*HZ)
 
-- 
cgit v1.2.3


From 0e0cab744b17a70ef0f08d818d66935feade7cad Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 26 Jun 2008 17:47:12 -0400
Subject: NFS: use documenting macro constants for initializing ac{reg,
 dir}{min, max}

Clean up.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfsroot.c       |  8 ++++----
 fs/nfs/super.c         | 24 ++++++++++++------------
 include/linux/nfs_fs.h |  5 +++++
 3 files changed, 21 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index 15afe3a6c5be..46763d1cd397 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -295,10 +295,10 @@ static int __init root_nfs_name(char *name)
 	nfs_data.flags    = NFS_MOUNT_NONLM;	/* No lockd in nfs root yet */
 	nfs_data.rsize    = NFS_DEF_FILE_IO_SIZE;
 	nfs_data.wsize    = NFS_DEF_FILE_IO_SIZE;
-	nfs_data.acregmin = 3;
-	nfs_data.acregmax = 60;
-	nfs_data.acdirmin = 30;
-	nfs_data.acdirmax = 60;
+	nfs_data.acregmin = NFS_DEF_ACREGMIN;
+	nfs_data.acregmax = NFS_DEF_ACREGMAX;
+	nfs_data.acdirmin = NFS_DEF_ACDIRMIN;
+	nfs_data.acdirmax = NFS_DEF_ACDIRMAX;
 	strcpy(buf, NFS_ROOT);
 
 	/* Process options received from the remote server */
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index de424d2f3155..ebed63e0ff8e 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -518,13 +518,13 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 	if (nfss->bsize != 0)
 		seq_printf(m, ",bsize=%u", nfss->bsize);
 	seq_printf(m, ",namlen=%u", nfss->namelen);
-	if (nfss->acregmin != 3*HZ || showdefaults)
+	if (nfss->acregmin != NFS_DEF_ACREGMIN*HZ || showdefaults)
 		seq_printf(m, ",acregmin=%u", nfss->acregmin/HZ);
-	if (nfss->acregmax != 60*HZ || showdefaults)
+	if (nfss->acregmax != NFS_DEF_ACREGMAX*HZ || showdefaults)
 		seq_printf(m, ",acregmax=%u", nfss->acregmax/HZ);
-	if (nfss->acdirmin != 30*HZ || showdefaults)
+	if (nfss->acdirmin != NFS_DEF_ACDIRMIN*HZ || showdefaults)
 		seq_printf(m, ",acdirmin=%u", nfss->acdirmin/HZ);
-	if (nfss->acdirmax != 60*HZ || showdefaults)
+	if (nfss->acdirmax != NFS_DEF_ACDIRMAX*HZ || showdefaults)
 		seq_printf(m, ",acdirmax=%u", nfss->acdirmax/HZ);
 	for (nfs_infop = nfs_info; nfs_infop->flag; nfs_infop++) {
 		if (nfss->flags & nfs_infop->flag)
@@ -1460,10 +1460,10 @@ static int nfs_validate_mount_data(void *options,
 	args->flags		= (NFS_MOUNT_VER3 | NFS_MOUNT_TCP);
 	args->rsize		= NFS_MAX_FILE_IO_SIZE;
 	args->wsize		= NFS_MAX_FILE_IO_SIZE;
-	args->acregmin		= 3;
-	args->acregmax		= 60;
-	args->acdirmin		= 30;
-	args->acdirmax		= 60;
+	args->acregmin		= NFS_DEF_ACREGMIN;
+	args->acregmax		= NFS_DEF_ACREGMAX;
+	args->acdirmin		= NFS_DEF_ACDIRMIN;
+	args->acdirmax		= NFS_DEF_ACDIRMAX;
 	args->mount_server.port	= 0;	/* autobind unless user sets port */
 	args->nfs_server.port	= 0;	/* autobind unless user sets port */
 	args->nfs_server.protocol = XPRT_TRANSPORT_TCP;
@@ -2119,10 +2119,10 @@ static int nfs4_validate_mount_data(void *options,
 
 	args->rsize		= NFS_MAX_FILE_IO_SIZE;
 	args->wsize		= NFS_MAX_FILE_IO_SIZE;
-	args->acregmin		= 3;
-	args->acregmax		= 60;
-	args->acdirmin		= 30;
-	args->acdirmax		= 60;
+	args->acregmin		= NFS_DEF_ACREGMIN;
+	args->acregmax		= NFS_DEF_ACREGMAX;
+	args->acdirmin		= NFS_DEF_ACDIRMIN;
+	args->acdirmax		= NFS_DEF_ACDIRMAX;
 	args->nfs_server.port	= NFS_PORT; /* 2049 unless user set port= */
 
 	switch (data->version) {
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 830d9cc8cdce..29d261918734 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -20,6 +20,11 @@
 #define NFS_MAX_UDP_TIMEOUT	(60*HZ)
 #define NFS_MAX_TCP_TIMEOUT	(600*HZ)
 
+#define NFS_DEF_ACREGMIN	(3)
+#define NFS_DEF_ACREGMAX	(60)
+#define NFS_DEF_ACDIRMIN	(30)
+#define NFS_DEF_ACDIRMAX	(60)
+
 /*
  * When flushing a cluster of dirty pages, there can be different
  * strategies:
-- 
cgit v1.2.3


From 004a403c2e954734090a69aedc7f4f822bdcc142 Mon Sep 17 00:00:00 2001
From: Loc Ho <lho@amcc.com>
Date: Wed, 14 May 2008 20:41:47 +0800
Subject: [CRYPTO] hash: Add asynchronous hash support

This patch adds asynchronous hash and digest support.

Signed-off-by: Loc Ho <lho@amcc.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/Makefile         |   1 +
 crypto/ahash.c          | 106 +++++++++++++++++++++++++++
 crypto/api.c            |   8 ++-
 crypto/digest.c         |  81 +++++++++++++++++++++
 crypto/hash.c           | 102 +++++++++++++++++++++++---
 crypto/internal.h       |   1 +
 include/crypto/algapi.h |  36 ++++++++++
 include/linux/crypto.h  | 187 ++++++++++++++++++++++++++++++++++++++++++++++--
 8 files changed, 507 insertions(+), 15 deletions(-)
 create mode 100644 crypto/ahash.c

(limited to 'include/linux')

diff --git a/crypto/Makefile b/crypto/Makefile
index 807656b64e02..d4f3ed857df0 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -19,6 +19,7 @@ obj-$(CONFIG_CRYPTO_BLKCIPHER) += crypto_blkcipher.o
 obj-$(CONFIG_CRYPTO_SEQIV) += seqiv.o
 
 crypto_hash-objs := hash.o
+crypto_hash-objs += ahash.o
 obj-$(CONFIG_CRYPTO_HASH) += crypto_hash.o
 
 obj-$(CONFIG_CRYPTO_MANAGER) += cryptomgr.o
diff --git a/crypto/ahash.c b/crypto/ahash.c
new file mode 100644
index 000000000000..a83e035d9a3f
--- /dev/null
+++ b/crypto/ahash.c
@@ -0,0 +1,106 @@
+/*
+ * Asynchronous Cryptographic Hash operations.
+ *
+ * This is the asynchronous version of hash.c with notification of
+ * completion via a callback.
+ *
+ * Copyright (c) 2008 Loc Ho <lho@amcc.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <crypto/algapi.h>
+#include <linux/err.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/seq_file.h>
+
+#include "internal.h"
+
+static int ahash_setkey_unaligned(struct crypto_ahash *tfm, const u8 *key,
+				unsigned int keylen)
+{
+	struct ahash_alg *ahash = crypto_ahash_alg(tfm);
+	unsigned long alignmask = crypto_ahash_alignmask(tfm);
+	int ret;
+	u8 *buffer, *alignbuffer;
+	unsigned long absize;
+
+	absize = keylen + alignmask;
+	buffer = kmalloc(absize, GFP_ATOMIC);
+	if (!buffer)
+		return -ENOMEM;
+
+	alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1);
+	memcpy(alignbuffer, key, keylen);
+	ret = ahash->setkey(tfm, alignbuffer, keylen);
+	memset(alignbuffer, 0, keylen);
+	kfree(buffer);
+	return ret;
+}
+
+static int ahash_setkey(struct crypto_ahash *tfm, const u8 *key,
+			unsigned int keylen)
+{
+	struct ahash_alg *ahash = crypto_ahash_alg(tfm);
+	unsigned long alignmask = crypto_ahash_alignmask(tfm);
+
+	if ((unsigned long)key & alignmask)
+		return ahash_setkey_unaligned(tfm, key, keylen);
+
+	return ahash->setkey(tfm, key, keylen);
+}
+
+static unsigned int crypto_ahash_ctxsize(struct crypto_alg *alg, u32 type,
+					u32 mask)
+{
+	return alg->cra_ctxsize;
+}
+
+static int crypto_init_ahash_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
+{
+	struct ahash_alg *alg = &tfm->__crt_alg->cra_ahash;
+	struct ahash_tfm *crt   = &tfm->crt_ahash;
+
+	if (alg->digestsize > crypto_tfm_alg_blocksize(tfm))
+		return -EINVAL;
+
+	crt->init = alg->init;
+	crt->update = alg->update;
+	crt->final  = alg->final;
+	crt->digest = alg->digest;
+	crt->setkey = ahash_setkey;
+	crt->base   = __crypto_ahash_cast(tfm);
+	crt->digestsize = alg->digestsize;
+
+	return 0;
+}
+
+static void crypto_ahash_show(struct seq_file *m, struct crypto_alg *alg)
+	__attribute__ ((unused));
+static void crypto_ahash_show(struct seq_file *m, struct crypto_alg *alg)
+{
+	seq_printf(m, "type         : ahash\n");
+	seq_printf(m, "async        : %s\n", alg->cra_flags & CRYPTO_ALG_ASYNC ?
+					     "yes" : "no");
+	seq_printf(m, "blocksize    : %u\n", alg->cra_blocksize);
+	seq_printf(m, "digestsize   : %u\n", alg->cra_hash.digestsize);
+}
+
+const struct crypto_type crypto_ahash_type = {
+	.ctxsize = crypto_ahash_ctxsize,
+	.init = crypto_init_ahash_ops,
+#ifdef CONFIG_PROC_FS
+	.show = crypto_ahash_show,
+#endif
+};
+EXPORT_SYMBOL_GPL(crypto_ahash_type);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Asynchronous cryptographic hash type");
diff --git a/crypto/api.c b/crypto/api.c
index 0a0f41ef255f..d06e33270abe 100644
--- a/crypto/api.c
+++ b/crypto/api.c
@@ -235,8 +235,12 @@ static int crypto_init_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
 		return crypto_init_cipher_ops(tfm);
 		
 	case CRYPTO_ALG_TYPE_DIGEST:
-		return crypto_init_digest_ops(tfm);
-		
+		if ((mask & CRYPTO_ALG_TYPE_HASH_MASK) !=
+		    CRYPTO_ALG_TYPE_HASH_MASK)
+			return crypto_init_digest_ops_async(tfm);
+		else
+			return crypto_init_digest_ops(tfm);
+
 	case CRYPTO_ALG_TYPE_COMPRESS:
 		return crypto_init_compress_ops(tfm);
 	
diff --git a/crypto/digest.c b/crypto/digest.c
index b526cc348b79..025c9aea24ed 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -157,3 +157,84 @@ int crypto_init_digest_ops(struct crypto_tfm *tfm)
 void crypto_exit_digest_ops(struct crypto_tfm *tfm)
 {
 }
+
+static int digest_async_nosetkey(struct crypto_ahash *tfm_async, const u8 *key,
+			unsigned int keylen)
+{
+	crypto_ahash_clear_flags(tfm_async, CRYPTO_TFM_RES_MASK);
+	return -ENOSYS;
+}
+
+static int digest_async_setkey(struct crypto_ahash *tfm_async, const u8 *key,
+			unsigned int keylen)
+{
+	struct crypto_tfm    *tfm        = crypto_ahash_tfm(tfm_async);
+	struct digest_alg    *dalg       = &tfm->__crt_alg->cra_digest;
+
+	crypto_ahash_clear_flags(tfm_async, CRYPTO_TFM_RES_MASK);
+	return dalg->dia_setkey(tfm, key, keylen);
+}
+
+static int digest_async_init(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm  = req->base.tfm;
+	struct digest_alg *dalg = &tfm->__crt_alg->cra_digest;
+
+	dalg->dia_init(tfm);
+	return 0;
+}
+
+static int digest_async_update(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm = req->base.tfm;
+	struct hash_desc  desc = {
+		.tfm   = __crypto_hash_cast(tfm),
+		.flags = req->base.flags,
+	};
+
+	update(&desc, req->src, req->nbytes);
+	return 0;
+}
+
+static int digest_async_final(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm  = req->base.tfm;
+	struct hash_desc  desc = {
+		.tfm   = __crypto_hash_cast(tfm),
+		.flags = req->base.flags,
+	};
+
+	final(&desc, req->result);
+	return 0;
+}
+
+static int digest_async_digest(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm  = req->base.tfm;
+	struct hash_desc  desc = {
+		.tfm   = __crypto_hash_cast(tfm),
+		.flags = req->base.flags,
+	};
+
+	return digest(&desc, req->src, req->nbytes, req->result);
+}
+
+int crypto_init_digest_ops_async(struct crypto_tfm *tfm)
+{
+	struct ahash_tfm  *crt  = &tfm->crt_ahash;
+	struct digest_alg *dalg = &tfm->__crt_alg->cra_digest;
+
+	if (dalg->dia_digestsize > crypto_tfm_alg_blocksize(tfm))
+		return -EINVAL;
+
+	crt->init       = digest_async_init;
+	crt->update     = digest_async_update;
+	crt->final      = digest_async_final;
+	crt->digest     = digest_async_digest;
+	crt->setkey     = dalg->dia_setkey ? digest_async_setkey :
+						digest_async_nosetkey;
+	crt->digestsize = dalg->dia_digestsize;
+	crt->base       = __crypto_ahash_cast(tfm);
+
+	return 0;
+}
diff --git a/crypto/hash.c b/crypto/hash.c
index 7dcff671c19b..f9400a014e74 100644
--- a/crypto/hash.c
+++ b/crypto/hash.c
@@ -59,24 +59,108 @@ static int hash_setkey(struct crypto_hash *crt, const u8 *key,
 	return alg->setkey(crt, key, keylen);
 }
 
-static int crypto_init_hash_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
+static int hash_async_setkey(struct crypto_ahash *tfm_async, const u8 *key,
+			unsigned int keylen)
+{
+	struct crypto_tfm  *tfm      = crypto_ahash_tfm(tfm_async);
+	struct crypto_hash *tfm_hash = __crypto_hash_cast(tfm);
+	struct hash_alg    *alg      = &tfm->__crt_alg->cra_hash;
+
+	return alg->setkey(tfm_hash, key, keylen);
+}
+
+static int hash_async_init(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm = req->base.tfm;
+	struct hash_alg   *alg = &tfm->__crt_alg->cra_hash;
+	struct hash_desc  desc = {
+		.tfm = __crypto_hash_cast(tfm),
+		.flags = req->base.flags,
+	};
+
+	return alg->init(&desc);
+}
+
+static int hash_async_update(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm = req->base.tfm;
+	struct hash_alg   *alg = &tfm->__crt_alg->cra_hash;
+	struct hash_desc  desc = {
+		.tfm = __crypto_hash_cast(tfm),
+		.flags = req->base.flags,
+	};
+
+	return alg->update(&desc, req->src, req->nbytes);
+}
+
+static int hash_async_final(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm = req->base.tfm;
+	struct hash_alg   *alg = &tfm->__crt_alg->cra_hash;
+	struct hash_desc  desc = {
+		.tfm = __crypto_hash_cast(tfm),
+		.flags = req->base.flags,
+	};
+
+	return alg->final(&desc, req->result);
+}
+
+static int hash_async_digest(struct ahash_request *req)
+{
+	struct crypto_tfm *tfm = req->base.tfm;
+	struct hash_alg   *alg = &tfm->__crt_alg->cra_hash;
+	struct hash_desc  desc = {
+		.tfm = __crypto_hash_cast(tfm),
+		.flags = req->base.flags,
+	};
+
+	return alg->digest(&desc, req->src, req->nbytes, req->result);
+}
+
+static int crypto_init_hash_ops_async(struct crypto_tfm *tfm)
+{
+	struct ahash_tfm *crt = &tfm->crt_ahash;
+	struct hash_alg  *alg = &tfm->__crt_alg->cra_hash;
+
+	crt->init       = hash_async_init;
+	crt->update     = hash_async_update;
+	crt->final      = hash_async_final;
+	crt->digest     = hash_async_digest;
+	crt->setkey     = hash_async_setkey;
+	crt->digestsize = alg->digestsize;
+	crt->base       = __crypto_ahash_cast(tfm);
+
+	return 0;
+}
+
+static int crypto_init_hash_ops_sync(struct crypto_tfm *tfm)
 {
 	struct hash_tfm *crt = &tfm->crt_hash;
 	struct hash_alg *alg = &tfm->__crt_alg->cra_hash;
 
-	if (alg->digestsize > crypto_tfm_alg_blocksize(tfm))
-		return -EINVAL;
-
-	crt->init = alg->init;
-	crt->update = alg->update;
-	crt->final = alg->final;
-	crt->digest = alg->digest;
-	crt->setkey = hash_setkey;
+	crt->init       = alg->init;
+	crt->update     = alg->update;
+	crt->final      = alg->final;
+	crt->digest     = alg->digest;
+	crt->setkey     = hash_setkey;
 	crt->digestsize = alg->digestsize;
 
 	return 0;
 }
 
+static int crypto_init_hash_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
+{
+	struct hash_alg *alg = &tfm->__crt_alg->cra_hash;
+
+	if (alg->digestsize > crypto_tfm_alg_blocksize(tfm))
+		return -EINVAL;
+
+	if ((mask & CRYPTO_ALG_TYPE_HASH_MASK) != CRYPTO_ALG_TYPE_HASH_MASK)
+		return crypto_init_hash_ops_async(tfm);
+	else
+		return crypto_init_hash_ops_sync(tfm);
+}
+
 static void crypto_hash_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
 static void crypto_hash_show(struct seq_file *m, struct crypto_alg *alg)
diff --git a/crypto/internal.h b/crypto/internal.h
index 32f4c2145603..683fcb2d91f4 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -86,6 +86,7 @@ struct crypto_alg *__crypto_alg_lookup(const char *name, u32 type, u32 mask);
 struct crypto_alg *crypto_alg_mod_lookup(const char *name, u32 type, u32 mask);
 
 int crypto_init_digest_ops(struct crypto_tfm *tfm);
+int crypto_init_digest_ops_async(struct crypto_tfm *tfm);
 int crypto_init_cipher_ops(struct crypto_tfm *tfm);
 int crypto_init_compress_ops(struct crypto_tfm *tfm);
 
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index 60d06e784be3..fef272a8ceeb 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -98,6 +98,7 @@ extern const struct crypto_type crypto_ablkcipher_type;
 extern const struct crypto_type crypto_aead_type;
 extern const struct crypto_type crypto_blkcipher_type;
 extern const struct crypto_type crypto_hash_type;
+extern const struct crypto_type crypto_ahash_type;
 
 void crypto_mod_put(struct crypto_alg *alg);
 
@@ -314,5 +315,40 @@ static inline int crypto_requires_sync(u32 type, u32 mask)
 	return (type ^ CRYPTO_ALG_ASYNC) & mask & CRYPTO_ALG_ASYNC;
 }
 
+static inline void *crypto_ahash_ctx(struct crypto_ahash *tfm)
+{
+	return crypto_tfm_ctx(&tfm->base);
+}
+
+static inline struct ahash_alg *crypto_ahash_alg(
+	struct crypto_ahash *tfm)
+{
+	return &crypto_ahash_tfm(tfm)->__crt_alg->cra_ahash;
+}
+
+static inline int ahash_enqueue_request(struct crypto_queue *queue,
+					     struct ahash_request *request)
+{
+	return crypto_enqueue_request(queue, &request->base);
+}
+
+static inline struct ahash_request *ahash_dequeue_request(
+	struct crypto_queue *queue)
+{
+	return ahash_request_cast(crypto_dequeue_request(queue));
+}
+
+static inline void *ahash_request_ctx(struct ahash_request *req)
+{
+	return req->__ctx;
+}
+
+static inline int ahash_tfm_in_queue(struct crypto_queue *queue,
+					  struct crypto_ahash *tfm)
+{
+	return crypto_tfm_in_queue(queue, crypto_ahash_tfm(tfm));
+}
+
+
 #endif	/* _CRYPTO_ALGAPI_H */
 
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 425824bd49f3..b6efe569128d 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -30,15 +30,17 @@
  */
 #define CRYPTO_ALG_TYPE_MASK		0x0000000f
 #define CRYPTO_ALG_TYPE_CIPHER		0x00000001
-#define CRYPTO_ALG_TYPE_DIGEST		0x00000002
-#define CRYPTO_ALG_TYPE_HASH		0x00000003
+#define CRYPTO_ALG_TYPE_COMPRESS	0x00000002
+#define CRYPTO_ALG_TYPE_AEAD		0x00000003
 #define CRYPTO_ALG_TYPE_BLKCIPHER	0x00000004
 #define CRYPTO_ALG_TYPE_ABLKCIPHER	0x00000005
 #define CRYPTO_ALG_TYPE_GIVCIPHER	0x00000006
-#define CRYPTO_ALG_TYPE_COMPRESS	0x00000008
-#define CRYPTO_ALG_TYPE_AEAD		0x00000009
+#define CRYPTO_ALG_TYPE_DIGEST		0x00000008
+#define CRYPTO_ALG_TYPE_HASH		0x00000009
+#define CRYPTO_ALG_TYPE_AHASH		0x0000000a
 
 #define CRYPTO_ALG_TYPE_HASH_MASK	0x0000000e
+#define CRYPTO_ALG_TYPE_AHASH_MASK	0x0000000c
 #define CRYPTO_ALG_TYPE_BLKCIPHER_MASK	0x0000000c
 
 #define CRYPTO_ALG_LARVAL		0x00000010
@@ -102,6 +104,7 @@ struct crypto_async_request;
 struct crypto_aead;
 struct crypto_blkcipher;
 struct crypto_hash;
+struct crypto_ahash;
 struct crypto_tfm;
 struct crypto_type;
 struct aead_givcrypt_request;
@@ -131,6 +134,18 @@ struct ablkcipher_request {
 	void *__ctx[] CRYPTO_MINALIGN_ATTR;
 };
 
+struct ahash_request {
+	struct crypto_async_request base;
+
+	void *info;
+
+	unsigned int nbytes;
+	struct scatterlist *src;
+	u8		   *result;
+
+	void *__ctx[] CRYPTO_MINALIGN_ATTR;
+};
+
 /**
  *	struct aead_request - AEAD request
  *	@base: Common attributes for async crypto requests
@@ -195,6 +210,17 @@ struct ablkcipher_alg {
 	unsigned int ivsize;
 };
 
+struct ahash_alg {
+	int (*init)(struct ahash_request *req);
+	int (*update)(struct ahash_request *req);
+	int (*final)(struct ahash_request *req);
+	int (*digest)(struct ahash_request *req);
+	int (*setkey)(struct crypto_ahash *tfm, const u8 *key,
+			unsigned int keylen);
+
+	unsigned int digestsize;
+};
+
 struct aead_alg {
 	int (*setkey)(struct crypto_aead *tfm, const u8 *key,
 	              unsigned int keylen);
@@ -272,6 +298,7 @@ struct compress_alg {
 #define cra_cipher	cra_u.cipher
 #define cra_digest	cra_u.digest
 #define cra_hash	cra_u.hash
+#define cra_ahash	cra_u.ahash
 #define cra_compress	cra_u.compress
 
 struct crypto_alg {
@@ -298,6 +325,7 @@ struct crypto_alg {
 		struct cipher_alg cipher;
 		struct digest_alg digest;
 		struct hash_alg hash;
+		struct ahash_alg ahash;
 		struct compress_alg compress;
 	} cra_u;
 
@@ -383,6 +411,19 @@ struct hash_tfm {
 	unsigned int digestsize;
 };
 
+struct ahash_tfm {
+	int (*init)(struct ahash_request *req);
+	int (*update)(struct ahash_request *req);
+	int (*final)(struct ahash_request *req);
+	int (*digest)(struct ahash_request *req);
+	int (*setkey)(struct crypto_ahash *tfm, const u8 *key,
+			unsigned int keylen);
+
+	unsigned int digestsize;
+	struct crypto_ahash *base;
+	unsigned int reqsize;
+};
+
 struct compress_tfm {
 	int (*cot_compress)(struct crypto_tfm *tfm,
 	                    const u8 *src, unsigned int slen,
@@ -397,6 +438,7 @@ struct compress_tfm {
 #define crt_blkcipher	crt_u.blkcipher
 #define crt_cipher	crt_u.cipher
 #define crt_hash	crt_u.hash
+#define crt_ahash	crt_u.ahash
 #define crt_compress	crt_u.compress
 
 struct crypto_tfm {
@@ -409,6 +451,7 @@ struct crypto_tfm {
 		struct blkcipher_tfm blkcipher;
 		struct cipher_tfm cipher;
 		struct hash_tfm hash;
+		struct ahash_tfm ahash;
 		struct compress_tfm compress;
 	} crt_u;
 	
@@ -441,6 +484,10 @@ struct crypto_hash {
 	struct crypto_tfm base;
 };
 
+struct crypto_ahash {
+	struct crypto_tfm base;
+};
+
 enum {
 	CRYPTOA_UNSPEC,
 	CRYPTOA_ALG,
@@ -1264,5 +1311,137 @@ static inline int crypto_comp_decompress(struct crypto_comp *tfm,
 						    src, slen, dst, dlen);
 }
 
+static inline struct crypto_ahash *__crypto_ahash_cast(struct crypto_tfm *tfm)
+{
+	return (struct crypto_ahash *)tfm;
+}
+
+static inline struct crypto_ahash *crypto_alloc_ahash(const char *alg_name,
+						      u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	mask &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_AHASH;
+	mask |= CRYPTO_ALG_TYPE_AHASH_MASK;
+
+	return __crypto_ahash_cast(crypto_alloc_base(alg_name, type, mask));
+}
+
+static inline struct crypto_tfm *crypto_ahash_tfm(struct crypto_ahash *tfm)
+{
+	return &tfm->base;
+}
+
+static inline void crypto_free_ahash(struct crypto_ahash *tfm)
+{
+	crypto_free_tfm(crypto_ahash_tfm(tfm));
+}
+
+static inline unsigned int crypto_ahash_alignmask(
+	struct crypto_ahash *tfm)
+{
+	return crypto_tfm_alg_alignmask(crypto_ahash_tfm(tfm));
+}
+
+static inline struct ahash_tfm *crypto_ahash_crt(struct crypto_ahash *tfm)
+{
+	return &crypto_ahash_tfm(tfm)->crt_ahash;
+}
+
+static inline unsigned int crypto_ahash_digestsize(struct crypto_ahash *tfm)
+{
+	return crypto_ahash_crt(tfm)->digestsize;
+}
+
+static inline u32 crypto_ahash_get_flags(struct crypto_ahash *tfm)
+{
+	return crypto_tfm_get_flags(crypto_ahash_tfm(tfm));
+}
+
+static inline void crypto_ahash_set_flags(struct crypto_ahash *tfm, u32 flags)
+{
+	crypto_tfm_set_flags(crypto_ahash_tfm(tfm), flags);
+}
+
+static inline void crypto_ahash_clear_flags(struct crypto_ahash *tfm, u32 flags)
+{
+	crypto_tfm_clear_flags(crypto_ahash_tfm(tfm), flags);
+}
+
+static inline struct crypto_ahash *crypto_ahash_reqtfm(
+	struct ahash_request *req)
+{
+	return __crypto_ahash_cast(req->base.tfm);
+}
+
+static inline unsigned int crypto_ahash_reqsize(struct crypto_ahash *tfm)
+{
+	return crypto_ahash_crt(tfm)->reqsize;
+}
+
+static inline int crypto_ahash_setkey(struct crypto_ahash *tfm,
+				      const u8 *key, unsigned int keylen)
+{
+	struct ahash_tfm *crt = crypto_ahash_crt(tfm);
+
+	return crt->setkey(crt->base, key, keylen);
+}
+
+static inline int crypto_ahash_digest(struct ahash_request *req)
+{
+	struct ahash_tfm *crt = crypto_ahash_crt(crypto_ahash_reqtfm(req));
+	return crt->digest(req);
+}
+
+static inline void ahash_request_set_tfm(struct ahash_request *req,
+					 struct crypto_ahash *tfm)
+{
+	req->base.tfm = crypto_ahash_tfm(crypto_ahash_crt(tfm)->base);
+}
+
+static inline struct ahash_request *ahash_request_alloc(
+	struct crypto_ahash *tfm, gfp_t gfp)
+{
+	struct ahash_request *req;
+
+	req = kmalloc(sizeof(struct ahash_request) +
+		      crypto_ahash_reqsize(tfm), gfp);
+
+	if (likely(req))
+		ahash_request_set_tfm(req, tfm);
+
+	return req;
+}
+
+static inline void ahash_request_free(struct ahash_request *req)
+{
+	kfree(req);
+}
+
+static inline struct ahash_request *ahash_request_cast(
+	struct crypto_async_request *req)
+{
+	return container_of(req, struct ahash_request, base);
+}
+
+static inline void ahash_request_set_callback(struct ahash_request *req,
+					      u32 flags,
+					      crypto_completion_t complete,
+					      void *data)
+{
+	req->base.complete = complete;
+	req->base.data = data;
+	req->base.flags = flags;
+}
+
+static inline void ahash_request_set_crypt(struct ahash_request *req,
+					   struct scatterlist *src, u8 *result,
+					   unsigned int nbytes)
+{
+	req->src = src;
+	req->nbytes = nbytes;
+	req->result = result;
+}
+
 #endif	/* _LINUX_CRYPTO_H */
 
-- 
cgit v1.2.3


From 166247f46a9c866e6f7f7d2212be875fb82212a1 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 7 Jul 2008 20:54:35 +0800
Subject: crypto: hash - Removed vestigial ahash fields

The base field in ahash_tfm appears to have been cut-n-pasted from
ablkcipher.  It isn't needed here at all.  Similarly, the info field
in ahash_request also appears to have originated from its cipher
counter-part and is vestigial.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/ahash.c         | 1 -
 crypto/digest.c        | 1 -
 crypto/hash.c          | 1 -
 include/linux/crypto.h | 7 ++-----
 4 files changed, 2 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/ahash.c b/crypto/ahash.c
index 8c1f918a6878..e6e5906ca80a 100644
--- a/crypto/ahash.c
+++ b/crypto/ahash.c
@@ -76,7 +76,6 @@ static int crypto_init_ahash_ops(struct crypto_tfm *tfm, u32 type, u32 mask)
 	crt->final  = alg->final;
 	crt->digest = alg->digest;
 	crt->setkey = ahash_setkey;
-	crt->base   = __crypto_ahash_cast(tfm);
 	crt->digestsize = alg->digestsize;
 
 	return 0;
diff --git a/crypto/digest.c b/crypto/digest.c
index d63d5d96feec..bf332982c50d 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -234,7 +234,6 @@ int crypto_init_digest_ops_async(struct crypto_tfm *tfm)
 	crt->setkey     = dalg->dia_setkey ? digest_async_setkey :
 						digest_async_nosetkey;
 	crt->digestsize = dalg->dia_digestsize;
-	crt->base       = __crypto_ahash_cast(tfm);
 
 	return 0;
 }
diff --git a/crypto/hash.c b/crypto/hash.c
index 0d7caa9ab748..140a75565f15 100644
--- a/crypto/hash.c
+++ b/crypto/hash.c
@@ -128,7 +128,6 @@ static int crypto_init_hash_ops_async(struct crypto_tfm *tfm)
 	crt->digest     = hash_async_digest;
 	crt->setkey     = hash_async_setkey;
 	crt->digestsize = alg->digestsize;
-	crt->base       = __crypto_ahash_cast(tfm);
 
 	return 0;
 }
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index b6efe569128d..68ef293644d3 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -137,8 +137,6 @@ struct ablkcipher_request {
 struct ahash_request {
 	struct crypto_async_request base;
 
-	void *info;
-
 	unsigned int nbytes;
 	struct scatterlist *src;
 	u8		   *result;
@@ -420,7 +418,6 @@ struct ahash_tfm {
 			unsigned int keylen);
 
 	unsigned int digestsize;
-	struct crypto_ahash *base;
 	unsigned int reqsize;
 };
 
@@ -1384,7 +1381,7 @@ static inline int crypto_ahash_setkey(struct crypto_ahash *tfm,
 {
 	struct ahash_tfm *crt = crypto_ahash_crt(tfm);
 
-	return crt->setkey(crt->base, key, keylen);
+	return crt->setkey(tfm, key, keylen);
 }
 
 static inline int crypto_ahash_digest(struct ahash_request *req)
@@ -1396,7 +1393,7 @@ static inline int crypto_ahash_digest(struct ahash_request *req)
 static inline void ahash_request_set_tfm(struct ahash_request *req,
 					 struct crypto_ahash *tfm)
 {
-	req->base.tfm = crypto_ahash_tfm(crypto_ahash_crt(tfm)->base);
+	req->base.tfm = crypto_ahash_tfm(tfm);
 }
 
 static inline struct ahash_request *ahash_request_alloc(
-- 
cgit v1.2.3


From 18e33e6d5cc0495826f5245777cd267732815e01 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 10 Jul 2008 16:01:22 +0800
Subject: crypto: hash - Move ahash functions into crypto/hash.h

All new crypto interfaces should go into individual files as much
as possible in order to ensure that crypto.h does not collapse under
its own weight.

This patch moves the ahash code into crypto/hash.h and crypto/internal/hash.h
respectively.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 crypto/cryptd.c                |   1 +
 crypto/digest.c                |   1 +
 crypto/hash.c                  |   1 +
 crypto/tcrypt.c                |   1 +
 include/crypto/algapi.h        |  36 ----------
 include/crypto/hash.h          | 154 +++++++++++++++++++++++++++++++++++++++++
 include/crypto/internal/hash.h |  37 ++++++++++
 include/linux/crypto.h         | 136 ------------------------------------
 8 files changed, 195 insertions(+), 172 deletions(-)
 create mode 100644 include/crypto/hash.h

(limited to 'include/linux')

diff --git a/crypto/cryptd.c b/crypto/cryptd.c
index d3ecd7e73b7e..d29e06b350ff 100644
--- a/crypto/cryptd.c
+++ b/crypto/cryptd.c
@@ -11,6 +11,7 @@
  */
 
 #include <crypto/algapi.h>
+#include <crypto/internal/hash.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
diff --git a/crypto/digest.c b/crypto/digest.c
index bf332982c50d..ac0919460d14 100644
--- a/crypto/digest.c
+++ b/crypto/digest.c
@@ -12,6 +12,7 @@
  *
  */
 
+#include <crypto/internal/hash.h>
 #include <crypto/scatterwalk.h>
 #include <linux/mm.h>
 #include <linux/errno.h>
diff --git a/crypto/hash.c b/crypto/hash.c
index 140a75565f15..cb86b19fd105 100644
--- a/crypto/hash.c
+++ b/crypto/hash.c
@@ -9,6 +9,7 @@
  * any later version.
  */
 
+#include <crypto/internal/hash.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c
index 87f08f9b0904..59821a22d752 100644
--- a/crypto/tcrypt.c
+++ b/crypto/tcrypt.c
@@ -15,6 +15,7 @@
  *
  */
 
+#include <crypto/hash.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/module.h>
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index fef272a8ceeb..60d06e784be3 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -98,7 +98,6 @@ extern const struct crypto_type crypto_ablkcipher_type;
 extern const struct crypto_type crypto_aead_type;
 extern const struct crypto_type crypto_blkcipher_type;
 extern const struct crypto_type crypto_hash_type;
-extern const struct crypto_type crypto_ahash_type;
 
 void crypto_mod_put(struct crypto_alg *alg);
 
@@ -315,40 +314,5 @@ static inline int crypto_requires_sync(u32 type, u32 mask)
 	return (type ^ CRYPTO_ALG_ASYNC) & mask & CRYPTO_ALG_ASYNC;
 }
 
-static inline void *crypto_ahash_ctx(struct crypto_ahash *tfm)
-{
-	return crypto_tfm_ctx(&tfm->base);
-}
-
-static inline struct ahash_alg *crypto_ahash_alg(
-	struct crypto_ahash *tfm)
-{
-	return &crypto_ahash_tfm(tfm)->__crt_alg->cra_ahash;
-}
-
-static inline int ahash_enqueue_request(struct crypto_queue *queue,
-					     struct ahash_request *request)
-{
-	return crypto_enqueue_request(queue, &request->base);
-}
-
-static inline struct ahash_request *ahash_dequeue_request(
-	struct crypto_queue *queue)
-{
-	return ahash_request_cast(crypto_dequeue_request(queue));
-}
-
-static inline void *ahash_request_ctx(struct ahash_request *req)
-{
-	return req->__ctx;
-}
-
-static inline int ahash_tfm_in_queue(struct crypto_queue *queue,
-					  struct crypto_ahash *tfm)
-{
-	return crypto_tfm_in_queue(queue, crypto_ahash_tfm(tfm));
-}
-
-
 #endif	/* _CRYPTO_ALGAPI_H */
 
diff --git a/include/crypto/hash.h b/include/crypto/hash.h
new file mode 100644
index 000000000000..d12498ec8a4e
--- /dev/null
+++ b/include/crypto/hash.h
@@ -0,0 +1,154 @@
+/*
+ * Hash: Hash algorithms under the crypto API
+ * 
+ * Copyright (c) 2008 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option) 
+ * any later version.
+ *
+ */
+
+#ifndef _CRYPTO_HASH_H
+#define _CRYPTO_HASH_H
+
+#include <linux/crypto.h>
+
+struct crypto_ahash {
+	struct crypto_tfm base;
+};
+
+static inline struct crypto_ahash *__crypto_ahash_cast(struct crypto_tfm *tfm)
+{
+	return (struct crypto_ahash *)tfm;
+}
+
+static inline struct crypto_ahash *crypto_alloc_ahash(const char *alg_name,
+						      u32 type, u32 mask)
+{
+	type &= ~CRYPTO_ALG_TYPE_MASK;
+	mask &= ~CRYPTO_ALG_TYPE_MASK;
+	type |= CRYPTO_ALG_TYPE_AHASH;
+	mask |= CRYPTO_ALG_TYPE_AHASH_MASK;
+
+	return __crypto_ahash_cast(crypto_alloc_base(alg_name, type, mask));
+}
+
+static inline struct crypto_tfm *crypto_ahash_tfm(struct crypto_ahash *tfm)
+{
+	return &tfm->base;
+}
+
+static inline void crypto_free_ahash(struct crypto_ahash *tfm)
+{
+	crypto_free_tfm(crypto_ahash_tfm(tfm));
+}
+
+static inline unsigned int crypto_ahash_alignmask(
+	struct crypto_ahash *tfm)
+{
+	return crypto_tfm_alg_alignmask(crypto_ahash_tfm(tfm));
+}
+
+static inline struct ahash_tfm *crypto_ahash_crt(struct crypto_ahash *tfm)
+{
+	return &crypto_ahash_tfm(tfm)->crt_ahash;
+}
+
+static inline unsigned int crypto_ahash_digestsize(struct crypto_ahash *tfm)
+{
+	return crypto_ahash_crt(tfm)->digestsize;
+}
+
+static inline u32 crypto_ahash_get_flags(struct crypto_ahash *tfm)
+{
+	return crypto_tfm_get_flags(crypto_ahash_tfm(tfm));
+}
+
+static inline void crypto_ahash_set_flags(struct crypto_ahash *tfm, u32 flags)
+{
+	crypto_tfm_set_flags(crypto_ahash_tfm(tfm), flags);
+}
+
+static inline void crypto_ahash_clear_flags(struct crypto_ahash *tfm, u32 flags)
+{
+	crypto_tfm_clear_flags(crypto_ahash_tfm(tfm), flags);
+}
+
+static inline struct crypto_ahash *crypto_ahash_reqtfm(
+	struct ahash_request *req)
+{
+	return __crypto_ahash_cast(req->base.tfm);
+}
+
+static inline unsigned int crypto_ahash_reqsize(struct crypto_ahash *tfm)
+{
+	return crypto_ahash_crt(tfm)->reqsize;
+}
+
+static inline int crypto_ahash_setkey(struct crypto_ahash *tfm,
+				      const u8 *key, unsigned int keylen)
+{
+	struct ahash_tfm *crt = crypto_ahash_crt(tfm);
+
+	return crt->setkey(tfm, key, keylen);
+}
+
+static inline int crypto_ahash_digest(struct ahash_request *req)
+{
+	struct ahash_tfm *crt = crypto_ahash_crt(crypto_ahash_reqtfm(req));
+	return crt->digest(req);
+}
+
+static inline void ahash_request_set_tfm(struct ahash_request *req,
+					 struct crypto_ahash *tfm)
+{
+	req->base.tfm = crypto_ahash_tfm(tfm);
+}
+
+static inline struct ahash_request *ahash_request_alloc(
+	struct crypto_ahash *tfm, gfp_t gfp)
+{
+	struct ahash_request *req;
+
+	req = kmalloc(sizeof(struct ahash_request) +
+		      crypto_ahash_reqsize(tfm), gfp);
+
+	if (likely(req))
+		ahash_request_set_tfm(req, tfm);
+
+	return req;
+}
+
+static inline void ahash_request_free(struct ahash_request *req)
+{
+	kfree(req);
+}
+
+static inline struct ahash_request *ahash_request_cast(
+	struct crypto_async_request *req)
+{
+	return container_of(req, struct ahash_request, base);
+}
+
+static inline void ahash_request_set_callback(struct ahash_request *req,
+					      u32 flags,
+					      crypto_completion_t complete,
+					      void *data)
+{
+	req->base.complete = complete;
+	req->base.data = data;
+	req->base.flags = flags;
+}
+
+static inline void ahash_request_set_crypt(struct ahash_request *req,
+					   struct scatterlist *src, u8 *result,
+					   unsigned int nbytes)
+{
+	req->src = src;
+	req->nbytes = nbytes;
+	req->result = result;
+}
+
+#endif	/* _CRYPTO_HASH_H */
diff --git a/include/crypto/internal/hash.h b/include/crypto/internal/hash.h
index 93ac42280221..917ae57bad4a 100644
--- a/include/crypto/internal/hash.h
+++ b/include/crypto/internal/hash.h
@@ -14,6 +14,7 @@
 #define _CRYPTO_INTERNAL_HASH_H
 
 #include <crypto/algapi.h>
+#include <crypto/hash.h>
 
 struct ahash_request;
 struct scatterlist;
@@ -33,9 +34,45 @@ struct crypto_hash_walk {
 	unsigned int flags;
 };
 
+extern const struct crypto_type crypto_ahash_type;
+
 int crypto_hash_walk_done(struct crypto_hash_walk *walk, int err);
 int crypto_hash_walk_first(struct ahash_request *req,
 			   struct crypto_hash_walk *walk);
 
+static inline void *crypto_ahash_ctx(struct crypto_ahash *tfm)
+{
+	return crypto_tfm_ctx(&tfm->base);
+}
+
+static inline struct ahash_alg *crypto_ahash_alg(
+	struct crypto_ahash *tfm)
+{
+	return &crypto_ahash_tfm(tfm)->__crt_alg->cra_ahash;
+}
+
+static inline int ahash_enqueue_request(struct crypto_queue *queue,
+					     struct ahash_request *request)
+{
+	return crypto_enqueue_request(queue, &request->base);
+}
+
+static inline struct ahash_request *ahash_dequeue_request(
+	struct crypto_queue *queue)
+{
+	return ahash_request_cast(crypto_dequeue_request(queue));
+}
+
+static inline void *ahash_request_ctx(struct ahash_request *req)
+{
+	return req->__ctx;
+}
+
+static inline int ahash_tfm_in_queue(struct crypto_queue *queue,
+					  struct crypto_ahash *tfm)
+{
+	return crypto_tfm_in_queue(queue, crypto_ahash_tfm(tfm));
+}
+
 #endif	/* _CRYPTO_INTERNAL_HASH_H */
 
diff --git a/include/linux/crypto.h b/include/linux/crypto.h
index 68ef293644d3..c43dc47fdf75 100644
--- a/include/linux/crypto.h
+++ b/include/linux/crypto.h
@@ -481,10 +481,6 @@ struct crypto_hash {
 	struct crypto_tfm base;
 };
 
-struct crypto_ahash {
-	struct crypto_tfm base;
-};
-
 enum {
 	CRYPTOA_UNSPEC,
 	CRYPTOA_ALG,
@@ -1308,137 +1304,5 @@ static inline int crypto_comp_decompress(struct crypto_comp *tfm,
 						    src, slen, dst, dlen);
 }
 
-static inline struct crypto_ahash *__crypto_ahash_cast(struct crypto_tfm *tfm)
-{
-	return (struct crypto_ahash *)tfm;
-}
-
-static inline struct crypto_ahash *crypto_alloc_ahash(const char *alg_name,
-						      u32 type, u32 mask)
-{
-	type &= ~CRYPTO_ALG_TYPE_MASK;
-	mask &= ~CRYPTO_ALG_TYPE_MASK;
-	type |= CRYPTO_ALG_TYPE_AHASH;
-	mask |= CRYPTO_ALG_TYPE_AHASH_MASK;
-
-	return __crypto_ahash_cast(crypto_alloc_base(alg_name, type, mask));
-}
-
-static inline struct crypto_tfm *crypto_ahash_tfm(struct crypto_ahash *tfm)
-{
-	return &tfm->base;
-}
-
-static inline void crypto_free_ahash(struct crypto_ahash *tfm)
-{
-	crypto_free_tfm(crypto_ahash_tfm(tfm));
-}
-
-static inline unsigned int crypto_ahash_alignmask(
-	struct crypto_ahash *tfm)
-{
-	return crypto_tfm_alg_alignmask(crypto_ahash_tfm(tfm));
-}
-
-static inline struct ahash_tfm *crypto_ahash_crt(struct crypto_ahash *tfm)
-{
-	return &crypto_ahash_tfm(tfm)->crt_ahash;
-}
-
-static inline unsigned int crypto_ahash_digestsize(struct crypto_ahash *tfm)
-{
-	return crypto_ahash_crt(tfm)->digestsize;
-}
-
-static inline u32 crypto_ahash_get_flags(struct crypto_ahash *tfm)
-{
-	return crypto_tfm_get_flags(crypto_ahash_tfm(tfm));
-}
-
-static inline void crypto_ahash_set_flags(struct crypto_ahash *tfm, u32 flags)
-{
-	crypto_tfm_set_flags(crypto_ahash_tfm(tfm), flags);
-}
-
-static inline void crypto_ahash_clear_flags(struct crypto_ahash *tfm, u32 flags)
-{
-	crypto_tfm_clear_flags(crypto_ahash_tfm(tfm), flags);
-}
-
-static inline struct crypto_ahash *crypto_ahash_reqtfm(
-	struct ahash_request *req)
-{
-	return __crypto_ahash_cast(req->base.tfm);
-}
-
-static inline unsigned int crypto_ahash_reqsize(struct crypto_ahash *tfm)
-{
-	return crypto_ahash_crt(tfm)->reqsize;
-}
-
-static inline int crypto_ahash_setkey(struct crypto_ahash *tfm,
-				      const u8 *key, unsigned int keylen)
-{
-	struct ahash_tfm *crt = crypto_ahash_crt(tfm);
-
-	return crt->setkey(tfm, key, keylen);
-}
-
-static inline int crypto_ahash_digest(struct ahash_request *req)
-{
-	struct ahash_tfm *crt = crypto_ahash_crt(crypto_ahash_reqtfm(req));
-	return crt->digest(req);
-}
-
-static inline void ahash_request_set_tfm(struct ahash_request *req,
-					 struct crypto_ahash *tfm)
-{
-	req->base.tfm = crypto_ahash_tfm(tfm);
-}
-
-static inline struct ahash_request *ahash_request_alloc(
-	struct crypto_ahash *tfm, gfp_t gfp)
-{
-	struct ahash_request *req;
-
-	req = kmalloc(sizeof(struct ahash_request) +
-		      crypto_ahash_reqsize(tfm), gfp);
-
-	if (likely(req))
-		ahash_request_set_tfm(req, tfm);
-
-	return req;
-}
-
-static inline void ahash_request_free(struct ahash_request *req)
-{
-	kfree(req);
-}
-
-static inline struct ahash_request *ahash_request_cast(
-	struct crypto_async_request *req)
-{
-	return container_of(req, struct ahash_request, base);
-}
-
-static inline void ahash_request_set_callback(struct ahash_request *req,
-					      u32 flags,
-					      crypto_completion_t complete,
-					      void *data)
-{
-	req->base.complete = complete;
-	req->base.data = data;
-	req->base.flags = flags;
-}
-
-static inline void ahash_request_set_crypt(struct ahash_request *req,
-					   struct scatterlist *src, u8 *result,
-					   unsigned int nbytes)
-{
-	req->src = src;
-	req->nbytes = nbytes;
-	req->result = result;
-}
-
 #endif	/* _LINUX_CRYPTO_H */
 
-- 
cgit v1.2.3


From b7a39bd0afc4021e8ad2b1189e884551e147427f Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Fri, 23 May 2008 18:38:49 +0100
Subject: firmware: make fw->data const

In preparation for supporting firmware files linked into the static
kernel, make fw->data const to ensure that users aren't modifying it (so
that we can pass a pointer to the original in-kernel copy, rather than
having to copy it).

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/base/firmware_class.c | 2 +-
 include/linux/firmware.h      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c
index 9fd4a8534146..264b3a2cd860 100644
--- a/drivers/base/firmware_class.c
+++ b/drivers/base/firmware_class.c
@@ -257,7 +257,7 @@ firmware_data_write(struct kobject *kobj, struct bin_attribute *bin_attr,
 	if (retval)
 		goto out;
 
-	memcpy(fw->data + offset, buffer, count);
+	memcpy((u8 *)fw->data + offset, buffer, count);
 
 	fw->size = max_t(size_t, offset + count, fw->size);
 	retval = count;
diff --git a/include/linux/firmware.h b/include/linux/firmware.h
index 6c7eff2ebada..88718d60153c 100644
--- a/include/linux/firmware.h
+++ b/include/linux/firmware.h
@@ -8,7 +8,7 @@
 
 struct firmware {
 	size_t size;
-	u8 *data;
+	const u8 *data;
 };
 
 struct device;
-- 
cgit v1.2.3


From 5658c769443d543728b6c5c673dffc2df8676317 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Fri, 23 May 2008 13:52:42 +0100
Subject: firmware: allow firmware files to be built into kernel image

Some drivers have their own hacks to bypass the kernel's firmware loader
and build their firmware into the kernel; this renders those unnecessary.

Other drivers don't use the firmware loader at all, because they always
want the firmware to be available. This allows them to start using the
firmware loader.

A third set of drivers already use the firmware loader, but can't be
used without help from userspace, which sometimes requires an initrd.
This allows them to work in a static kernel.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/base/firmware_class.c     | 33 +++++++++++++++++++++++++++++++--
 include/asm-generic/vmlinux.lds.h |  7 +++++++
 include/linux/firmware.h          | 21 +++++++++++++++++++++
 3 files changed, 59 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c
index 264b3a2cd860..b0be1d18fee2 100644
--- a/drivers/base/firmware_class.c
+++ b/drivers/base/firmware_class.c
@@ -49,6 +49,14 @@ struct firmware_priv {
 	struct timer_list timeout;
 };
 
+#ifdef CONFIG_FW_LOADER
+extern struct builtin_fw __start_builtin_fw[];
+extern struct builtin_fw __end_builtin_fw[];
+#else /* Module case. Avoid ifdefs later; it'll all optimise out */
+static struct builtin_fw *__start_builtin_fw;
+static struct builtin_fw *__end_builtin_fw;
+#endif
+
 static void
 fw_load_abort(struct firmware_priv *fw_priv)
 {
@@ -391,13 +399,12 @@ _request_firmware(const struct firmware **firmware_p, const char *name,
 	struct device *f_dev;
 	struct firmware_priv *fw_priv;
 	struct firmware *firmware;
+	struct builtin_fw *builtin;
 	int retval;
 
 	if (!firmware_p)
 		return -EINVAL;
 
-	printk(KERN_INFO "firmware: requesting %s\n", name);
-
 	*firmware_p = firmware = kzalloc(sizeof(*firmware), GFP_KERNEL);
 	if (!firmware) {
 		printk(KERN_ERR "%s: kmalloc(struct firmware) failed\n",
@@ -406,6 +413,20 @@ _request_firmware(const struct firmware **firmware_p, const char *name,
 		goto out;
 	}
 
+	for (builtin = __start_builtin_fw; builtin != __end_builtin_fw;
+	     builtin++) {
+		if (strcmp(name, builtin->name))
+			continue;
+		printk(KERN_INFO "firmware: using built-in firmware %s\n",
+		       name);
+		firmware->size = builtin->size;
+		firmware->data = builtin->data;
+		return 0;
+	}
+
+	if (uevent)
+		printk(KERN_INFO "firmware: requesting %s\n", name);
+
 	retval = fw_setup_device(firmware, &f_dev, name, device, uevent);
 	if (retval)
 		goto error_kfree_fw;
@@ -473,8 +494,16 @@ request_firmware(const struct firmware **firmware_p, const char *name,
 void
 release_firmware(const struct firmware *fw)
 {
+	struct builtin_fw *builtin;
+
 	if (fw) {
+		for (builtin = __start_builtin_fw; builtin != __end_builtin_fw;
+		     builtin++) {
+			if (fw->data == builtin->data)
+				goto free_fw;
+		}
 		vfree(fw->data);
+	free_fw:
 		kfree(fw);
 	}
 }
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index f054778e916c..8d71a40625f3 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -86,6 +86,13 @@
 		VMLINUX_SYMBOL(__end_pci_fixups_resume) = .;		\
 	}								\
 									\
+	/* Built-in firmware blobs */					\
+	.builtin_fw        : AT(ADDR(.builtin_fw) - LOAD_OFFSET) {	\
+		VMLINUX_SYMBOL(__start_builtin_fw) = .;			\
+		*(.builtin_fw)						\
+		VMLINUX_SYMBOL(__end_builtin_fw) = .;			\
+	}								\
+									\
 	/* RapidIO route ops */						\
 	.rio_route        : AT(ADDR(.rio_route) - LOAD_OFFSET) {	\
 		VMLINUX_SYMBOL(__start_rio_route_ops) = .;		\
diff --git a/include/linux/firmware.h b/include/linux/firmware.h
index 88718d60153c..c8ecf5b2a207 100644
--- a/include/linux/firmware.h
+++ b/include/linux/firmware.h
@@ -1,7 +1,10 @@
 #ifndef _LINUX_FIRMWARE_H
 #define _LINUX_FIRMWARE_H
+
 #include <linux/module.h>
 #include <linux/types.h>
+#include <linux/compiler.h>
+
 #define FIRMWARE_NAME_MAX 30 
 #define FW_ACTION_NOHOTPLUG 0
 #define FW_ACTION_HOTPLUG 1
@@ -13,6 +16,24 @@ struct firmware {
 
 struct device;
 
+struct builtin_fw {
+	char *name;
+	void *data;
+	unsigned long size;
+};
+
+/* We have to play tricks here much like stringify() to get the
+   __COUNTER__ macro to be expanded as we want it */
+#define __fw_concat1(x, y) x##y
+#define __fw_concat(x, y) __fw_concat1(x, y)
+
+#define DECLARE_BUILTIN_FIRMWARE(name, blob)				     \
+	DECLARE_BUILTIN_FIRMWARE_SIZE(name, &(blob), sizeof(blob))
+
+#define DECLARE_BUILTIN_FIRMWARE_SIZE(name, blob, size)			     \
+	static const struct builtin_fw __fw_concat(__builtin_fw,__COUNTER__) \
+	__used __section(.builtin_fw) = { name, blob, size }
+
 #if defined(CONFIG_FW_LOADER) || (defined(CONFIG_FW_LOADER_MODULE) && defined(MODULE))
 int request_firmware(const struct firmware **fw, const char *name,
 		     struct device *device);
-- 
cgit v1.2.3


From bacfe09dd7545467965e8d8f1eab20bc62dce00d Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Fri, 30 May 2008 13:57:27 +0300
Subject: ihex.h: binary representation of ihex records

Some devices need their firmware as a set of {address, len, data...}
records in some specific order rather than a simple blob.

The normal way of doing this kind of thing is 'ihex', which is a text
format and not entirely suitable for use in the kernel.

This provides a binary representation which is very similar, but much
more compact -- and a helper routine to skip to the next record,
because the alignment constraints mean that everybody will screw it up
for themselves otherwise.

Also a helper function which can verify that a 'struct firmware'
contains a valid set of ihex records, and that following them won't run
off the end of the loaded data.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/linux/ihex.h | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 50 insertions(+)
 create mode 100644 include/linux/ihex.h

(limited to 'include/linux')

diff --git a/include/linux/ihex.h b/include/linux/ihex.h
new file mode 100644
index 000000000000..df89edd890ae
--- /dev/null
+++ b/include/linux/ihex.h
@@ -0,0 +1,50 @@
+/*
+ * Compact binary representation of ihex records. Some devices need their
+ * firmware loaded in strange orders rather than a single big blob, but
+ * actually parsing ihex-as-text within the kernel seems silly. Thus,...
+ */
+
+#ifndef __LINUX_IHEX_H__
+#define __LINUX_IHEX_H__
+
+#include <linux/types.h>
+#include <linux/firmware.h>
+
+/* Intel HEX files actually limit the length to 256 bytes, but we have
+   drivers which would benefit from using separate records which are
+   longer than that, so we extend to 16 bits of length */
+struct ihex_binrec {
+	__be32 addr;
+	__be16 len;
+	uint8_t data[0];
+} __attribute__((aligned(4)));
+
+/* Find the next record, taking into account the 4-byte alignment */
+static inline const struct ihex_binrec *
+ihex_next_binrec(const struct ihex_binrec *rec)
+{
+	int next = ((be16_to_cpu(rec->len) + 5) & ~3) - 2;
+	rec = (void *)&rec->data[next];
+
+	return be16_to_cpu(rec->len) ? rec : NULL;
+}
+
+/* Check that ihex_next_binrec() won't take us off the end of the image... */
+static inline int ihex_validate_fw(const struct firmware *fw)
+{
+	const struct ihex_binrec *rec;
+	size_t ofs = 0;
+
+	while (ofs <= fw->size - sizeof(*rec)) {
+		rec = (void *)&fw->data[ofs];
+
+		/* Zero length marks end of records */
+		if (!be16_to_cpu(rec->len))
+			return 0;
+
+		/* Point to next record... */
+		ofs += (sizeof(*rec) + be16_to_cpu(rec->len) + 3) & ~3;
+	}
+	return -EINVAL;
+}
+#endif /* __LINUX_IHEX_H__ */
-- 
cgit v1.2.3


From f1485f3deb89e6ae10c4d34662ec9e692855ab5d Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Sat, 31 May 2008 15:20:37 +0300
Subject: ihex: request_ihex_firmware() function to load and validate firmware

Provide a helper to load the file and validate it in one call, to
simplify error handling in the drivers which are going to use it.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 include/linux/ihex.h | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ihex.h b/include/linux/ihex.h
index df89edd890ae..2baace2788a7 100644
--- a/include/linux/ihex.h
+++ b/include/linux/ihex.h
@@ -9,6 +9,7 @@
 
 #include <linux/types.h>
 #include <linux/firmware.h>
+#include <linux/device.h>
 
 /* Intel HEX files actually limit the length to 256 bytes, but we have
    drivers which would benefit from using separate records which are
@@ -47,4 +48,27 @@ static inline int ihex_validate_fw(const struct firmware *fw)
 	}
 	return -EINVAL;
 }
+
+/* Request firmware and validate it so that we can trust we won't
+ * run off the end while reading records... */
+static inline int request_ihex_firmware(const struct firmware **fw,
+					const char *fw_name,
+					struct device *dev)
+{
+	const struct firmware *lfw;
+	int ret;
+
+	ret = request_firmware(&lfw, fw_name, dev);
+	if (ret)
+		return ret;
+	ret = ihex_validate_fw(lfw);
+	if (ret) {
+		dev_err(dev, "Firmware \"%s\" not valid IHEX records\n",
+			fw_name);
+		release_firmware(lfw);
+		return ret;
+	}
+	*fw = lfw;
+	return 0;
+}
 #endif /* __LINUX_IHEX_H__ */
-- 
cgit v1.2.3


From d8156534040996f6a93a24d3592d5d587f2587e5 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexander.h.duyck@intel.com>
Date: Tue, 8 Jul 2008 15:13:05 -0700
Subject: net: add netif_napi_del function to allow for removal of napistructs

Adds netif_napi_del function which is used to remove the napi struct from
the netdev napi_list in cases where CONFIG_NETPOLL was enabled.
The motivation for adding this is to handle the case in which the number of
queues on a device changes due to a configuration change.  Previously the
napi structs for each queue would be left in the list until the netdev was
freed.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/netdevice.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 203c5504fe43..b54ec16dfbda 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -828,6 +828,19 @@ static inline void netif_napi_add(struct net_device *dev,
 	set_bit(NAPI_STATE_SCHED, &napi->state);
 }
 
+/**
+ *  netif_napi_del - remove a napi context
+ *  @napi: napi context
+ *
+ *  netif_napi_del() removes a napi context from the network device napi list
+ */
+static inline void netif_napi_del(struct napi_struct *napi)
+{
+#ifdef CONFIG_NETPOLL
+	list_del(&napi->dev_list);
+#endif
+}
+
 struct packet_type {
 	__be16			type;	/* This is really htons(ether_type). */
 	struct net_device	*dev;	/* NULL is wildcarded here	     */
-- 
cgit v1.2.3


From 0f420358e3a2abc028320ace7783e2e38cae77bf Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Fri, 11 Jul 2008 22:02:23 +1000
Subject: md: Turn rdev->sb_offset into a sector-based quantity.

Rename it to sb_start to make sure all users have been converted.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 drivers/md/bitmap.c       | 10 +++---
 drivers/md/md.c           | 81 ++++++++++++++++++++++-------------------------
 include/linux/raid/md_k.h |  2 +-
 3 files changed, 44 insertions(+), 49 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index dedba16d42f7..eba83e25b678 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -225,7 +225,7 @@ static struct page *read_sb_page(mddev_t *mddev, long offset, unsigned long inde
 		    || test_bit(Faulty, &rdev->flags))
 			continue;
 
-		target = (rdev->sb_offset << 1) + offset + index * (PAGE_SIZE/512);
+		target = rdev->sb_start + offset + index * (PAGE_SIZE/512);
 
 		if (sync_page_io(rdev->bdev, target, PAGE_SIZE, page, READ)) {
 			page->index = index;
@@ -262,12 +262,12 @@ static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
 					/* bitmap runs in to metadata */
 					return -EINVAL;
 				if (rdev->data_offset + mddev->size*2
-				    > rdev->sb_offset*2 + bitmap->offset)
+				    > rdev->sb_start + bitmap->offset)
 					/* data runs in to bitmap */
 					return -EINVAL;
-			} else if (rdev->sb_offset*2 < rdev->data_offset) {
+			} else if (rdev->sb_start < rdev->data_offset) {
 				/* METADATA BITMAP DATA */
-				if (rdev->sb_offset*2
+				if (rdev->sb_start
 				    + bitmap->offset
 				    + page->index*(PAGE_SIZE/512) + size/512
 				    > rdev->data_offset)
@@ -277,7 +277,7 @@ static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
 				/* DATA METADATA BITMAP - no problems */
 			}
 			md_super_write(mddev, rdev,
-				       (rdev->sb_offset<<1) + bitmap->offset
+				       rdev->sb_start + bitmap->offset
 				       + page->index * (PAGE_SIZE/512),
 				       size,
 				       page);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 3276edde7576..5590cb54b584 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -356,7 +356,7 @@ static inline sector_t calc_dev_sboffset(struct block_device *bdev)
 
 static sector_t calc_num_sectors(mdk_rdev_t *rdev, unsigned chunk_size)
 {
-	sector_t num_sectors = rdev->sb_offset * 2;
+	sector_t num_sectors = rdev->sb_start;
 
 	if (chunk_size)
 		num_sectors &= ~((sector_t)chunk_size/512 - 1);
@@ -383,7 +383,7 @@ static void free_disk_sb(mdk_rdev_t * rdev)
 		put_page(rdev->sb_page);
 		rdev->sb_loaded = 0;
 		rdev->sb_page = NULL;
-		rdev->sb_offset = 0;
+		rdev->sb_start = 0;
 		rdev->size = 0;
 	}
 }
@@ -529,7 +529,7 @@ static int read_disk_sb(mdk_rdev_t * rdev, int size)
 		return 0;
 
 
-	if (!sync_page_io(rdev->bdev, rdev->sb_offset<<1, size, rdev->sb_page, READ))
+	if (!sync_page_io(rdev->bdev, rdev->sb_start, size, rdev->sb_page, READ))
 		goto fail;
 	rdev->sb_loaded = 1;
 	return 0;
@@ -666,16 +666,14 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
 	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
 	mdp_super_t *sb;
 	int ret;
-	sector_t sb_offset;
 
 	/*
-	 * Calculate the position of the superblock,
+	 * Calculate the position of the superblock (512byte sectors),
 	 * it's at the end of the disk.
 	 *
 	 * It also happens to be a multiple of 4Kb.
 	 */
-	sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
-	rdev->sb_offset = sb_offset;
+	rdev->sb_start = calc_dev_sboffset(rdev->bdev);
 
 	ret = read_disk_sb(rdev, MD_SB_BYTES);
 	if (ret) return ret;
@@ -1007,10 +1005,10 @@ super_90_rdev_size_change(mdk_rdev_t *rdev, unsigned long long size)
 	size *= 2; /* convert to sectors */
 	if (rdev->mddev->bitmap_offset)
 		return 0; /* can't move bitmap */
-	rdev->sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
-	if (!size || size > rdev->sb_offset*2)
-		size = rdev->sb_offset*2;
-	md_super_write(rdev->mddev, rdev, rdev->sb_offset << 1, rdev->sb_size,
+	rdev->sb_start = calc_dev_sboffset(rdev->bdev);
+	if (!size || size > rdev->sb_start)
+		size = rdev->sb_start;
+	md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
 		       rdev->sb_page);
 	md_super_wait(rdev->mddev);
 	return size/2; /* kB for sysfs */
@@ -1048,12 +1046,12 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 {
 	struct mdp_superblock_1 *sb;
 	int ret;
-	sector_t sb_offset;
+	sector_t sb_start;
 	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
 	int bmask;
 
 	/*
-	 * Calculate the position of the superblock.
+	 * Calculate the position of the superblock in 512byte sectors.
 	 * It is always aligned to a 4K boundary and
 	 * depeding on minor_version, it can be:
 	 * 0: At least 8K, but less than 12K, from end of device
@@ -1062,22 +1060,20 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 	 */
 	switch(minor_version) {
 	case 0:
-		sb_offset = rdev->bdev->bd_inode->i_size >> 9;
-		sb_offset -= 8*2;
-		sb_offset &= ~(sector_t)(4*2-1);
-		/* convert from sectors to K */
-		sb_offset /= 2;
+		sb_start = rdev->bdev->bd_inode->i_size >> 9;
+		sb_start -= 8*2;
+		sb_start &= ~(sector_t)(4*2-1);
 		break;
 	case 1:
-		sb_offset = 0;
+		sb_start = 0;
 		break;
 	case 2:
-		sb_offset = 4;
+		sb_start = 8;
 		break;
 	default:
 		return -EINVAL;
 	}
-	rdev->sb_offset = sb_offset;
+	rdev->sb_start = sb_start;
 
 	/* superblock is rarely larger than 1K, but it can be larger,
 	 * and it is safe to read 4k, so we do that
@@ -1091,7 +1087,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 	if (sb->magic != cpu_to_le32(MD_SB_MAGIC) ||
 	    sb->major_version != cpu_to_le32(1) ||
 	    le32_to_cpu(sb->max_dev) > (4096-256)/2 ||
-	    le64_to_cpu(sb->super_offset) != (rdev->sb_offset<<1) ||
+	    le64_to_cpu(sb->super_offset) != rdev->sb_start ||
 	    (le32_to_cpu(sb->feature_map) & ~MD_FEATURE_ALL) != 0)
 		return -EINVAL;
 
@@ -1127,7 +1123,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 		rdev->sb_size = (rdev->sb_size | bmask) + 1;
 
 	if (minor_version
-	    && rdev->data_offset < sb_offset + (rdev->sb_size/512))
+	    && rdev->data_offset < sb_start + (rdev->sb_size/512))
 		return -EINVAL;
 
 	if (sb->level == cpu_to_le32(LEVEL_MULTIPATH))
@@ -1163,7 +1159,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version)
 	if (minor_version)
 		rdev->size = ((rdev->bdev->bd_inode->i_size>>9) - le64_to_cpu(sb->data_offset)) / 2;
 	else
-		rdev->size = rdev->sb_offset;
+		rdev->size = rdev->sb_start / 2;
 	if (rdev->size < le64_to_cpu(sb->data_size)/2)
 		return -EINVAL;
 	rdev->size = le64_to_cpu(sb->data_size)/2;
@@ -1350,7 +1346,7 @@ super_1_rdev_size_change(mdk_rdev_t *rdev, unsigned long long size)
 	if (size && size < rdev->mddev->size)
 		return 0; /* component must fit device */
 	size *= 2; /* convert to sectors */
-	if (rdev->sb_offset < rdev->data_offset/2) {
+	if (rdev->sb_start < rdev->data_offset) {
 		/* minor versions 1 and 2; superblock before data */
 		max_size = (rdev->bdev->bd_inode->i_size >> 9);
 		max_size -= rdev->data_offset;
@@ -1361,19 +1357,19 @@ super_1_rdev_size_change(mdk_rdev_t *rdev, unsigned long long size)
 		return 0;
 	} else {
 		/* minor version 0; superblock after data */
-		sector_t sb_offset;
-		sb_offset = (rdev->bdev->bd_inode->i_size >> 9) - 8*2;
-		sb_offset &= ~(sector_t)(4*2 - 1);
-		max_size = rdev->size*2 + sb_offset - rdev->sb_offset*2;
+		sector_t sb_start;
+		sb_start = (rdev->bdev->bd_inode->i_size >> 9) - 8*2;
+		sb_start &= ~(sector_t)(4*2 - 1);
+		max_size = rdev->size*2 + sb_start - rdev->sb_start;
 		if (!size || size > max_size)
 			size = max_size;
-		rdev->sb_offset = sb_offset/2;
+		rdev->sb_start = sb_start;
 	}
 	sb = (struct mdp_superblock_1 *) page_address(rdev->sb_page);
 	sb->data_size = cpu_to_le64(size);
-	sb->super_offset = rdev->sb_offset*2;
+	sb->super_offset = rdev->sb_start;
 	sb->sb_csum = calc_sb_1_csum(sb);
-	md_super_write(rdev->mddev, rdev, rdev->sb_offset << 1, rdev->sb_size,
+	md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
 		       rdev->sb_page);
 	md_super_wait(rdev->mddev);
 	return size/2; /* kB for sysfs */
@@ -1810,11 +1806,11 @@ repeat:
 		dprintk("%s ", bdevname(rdev->bdev,b));
 		if (!test_bit(Faulty, &rdev->flags)) {
 			md_super_write(mddev,rdev,
-				       rdev->sb_offset<<1, rdev->sb_size,
+				       rdev->sb_start, rdev->sb_size,
 				       rdev->sb_page);
 			dprintk(KERN_INFO "(write) %s's sb offset: %llu\n",
 				bdevname(rdev->bdev,b),
-				(unsigned long long)rdev->sb_offset);
+				(unsigned long long)rdev->sb_start);
 			rdev->sb_events = mddev->events;
 
 		} else
@@ -3577,16 +3573,16 @@ static int do_md_run(mddev_t * mddev)
 		 * We don't want the data to overlap the metadata,
 		 * Internal Bitmap issues has handled elsewhere.
 		 */
-		if (rdev->data_offset < rdev->sb_offset) {
+		if (rdev->data_offset < rdev->sb_start) {
 			if (mddev->size &&
 			    rdev->data_offset + mddev->size*2
-			    > rdev->sb_offset*2) {
+			    > rdev->sb_start) {
 				printk("md: %s: data overlaps metadata\n",
 				       mdname(mddev));
 				return -EINVAL;
 			}
 		} else {
-			if (rdev->sb_offset*2 + rdev->sb_size/512
+			if (rdev->sb_start + rdev->sb_size/512
 			    > rdev->data_offset) {
 				printk("md: %s: metadata overlaps data\n",
 				       mdname(mddev));
@@ -4355,9 +4351,9 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
 
 		if (!mddev->persistent) {
 			printk(KERN_INFO "md: nonpersistent superblock ...\n");
-			rdev->sb_offset = rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
+			rdev->sb_start = rdev->bdev->bd_inode->i_size / 512;
 		} else 
-			rdev->sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
+			rdev->sb_start = calc_dev_sboffset(rdev->bdev);
 		rdev->size = calc_num_sectors(rdev, mddev->chunk_size) / 2;
 
 		err = bind_rdev_to_array(rdev, mddev);
@@ -4424,10 +4420,9 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev)
 	}
 
 	if (mddev->persistent)
-		rdev->sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
+		rdev->sb_start = calc_dev_sboffset(rdev->bdev);
 	else
-		rdev->sb_offset =
-			rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
+		rdev->sb_start = rdev->bdev->bd_inode->i_size / 512;
 
 	rdev->size = calc_num_sectors(rdev, mddev->chunk_size) / 2;
 
@@ -4628,7 +4623,7 @@ static int update_size(mddev_t *mddev, sector_t num_sectors)
 	 * linear and raid0 always use whatever space is available. We can only
 	 * consider changing this number if no resync or reconstruction is
 	 * happening, and if the new size is acceptable. It must fit before the
-	 * sb_offset or, if that is <data_offset, it must fit before the size
+	 * sb_start or, if that is <data_offset, it must fit before the size
 	 * of each device.  If num_sectors is zero, we find the largest size
 	 * that fits.
 
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index df30c4395875..e37aaa41efc6 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -59,7 +59,7 @@ struct mdk_rdev_s
 	int		sb_loaded;
 	__u64		sb_events;
 	sector_t	data_offset;	/* start of data in array */
-	sector_t	sb_offset;
+	sector_t 	sb_start;	/* offset of the super block (in 512byte sectors) */
 	int		sb_size;	/* bytes in the superblock */
 	int		preferred_minor;	/* autorun support */
 
-- 
cgit v1.2.3


From 7e93a89251d4ed7bd4475db62616ccd03ddfd01a Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Fri, 11 Jul 2008 22:02:23 +1000
Subject: md: Remove some unused macros.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: Neil Brown <neilb@suse.de>
---
 include/linux/raid/md_p.h | 3 ---
 1 file changed, 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h
index 3f2cd98c508b..8b4de4a41ff1 100644
--- a/include/linux/raid/md_p.h
+++ b/include/linux/raid/md_p.h
@@ -43,14 +43,11 @@
  */
 #define MD_RESERVED_BYTES		(64 * 1024)
 #define MD_RESERVED_SECTORS		(MD_RESERVED_BYTES / 512)
-#define MD_RESERVED_BLOCKS		(MD_RESERVED_BYTES / BLOCK_SIZE)
 
 #define MD_NEW_SIZE_SECTORS(x)		((x & ~(MD_RESERVED_SECTORS - 1)) - MD_RESERVED_SECTORS)
-#define MD_NEW_SIZE_BLOCKS(x)		((x & ~(MD_RESERVED_BLOCKS - 1)) - MD_RESERVED_BLOCKS)
 
 #define MD_SB_BYTES			4096
 #define MD_SB_WORDS			(MD_SB_BYTES / 4)
-#define MD_SB_BLOCKS			(MD_SB_BYTES / BLOCK_SIZE)
 #define MD_SB_SECTORS			(MD_SB_BYTES / 512)
 
 /*
-- 
cgit v1.2.3


From a2bb6a3d85ef3124cd336403a95abc0540d3fbe2 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Thu, 10 Jul 2008 20:58:15 -0400
Subject: ftrace: add ftrace_kill_atomic

It has been suggested that I add a way to disable the function tracer
on an oops. This code adds a ftrace_kill_atomic. It is not meant to be
used in normal situations. It will disable the ftrace tracer, but will
not perform the nice shutdown that requires scheduling.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: Steven Rostedt <srostedt@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/ftrace.h |  1 +
 kernel/trace/ftrace.c  | 15 +++++++++++++++
 2 files changed, 16 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 3121b95443d9..f368d041e02d 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -89,6 +89,7 @@ void ftrace_enable_daemon(void);
 
 /* totally disable ftrace - can not re-enable after this */
 void ftrace_kill(void);
+void ftrace_kill_atomic(void);
 
 static inline void tracer_disable(void)
 {
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 0f271c45cd02..1359632668a4 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1601,6 +1601,21 @@ core_initcall(ftrace_dynamic_init);
 # define ftrace_force_shutdown()	do { } while (0)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
+/**
+ * ftrace_kill_atomic - kill ftrace from critical sections
+ *
+ * This function should be used by panic code. It stops ftrace
+ * but in a not so nice way. If you need to simply kill ftrace
+ * from a non-atomic section, use ftrace_kill.
+ */
+void ftrace_kill_atomic(void)
+{
+	ftrace_disabled = 1;
+	ftrace_enabled = 0;
+	ftraced_suspend = -1;
+	clear_ftrace_function();
+}
+
 /**
  * ftrace_kill - totally shutdown ftrace
  *
-- 
cgit v1.2.3


From af52a90a14cdaa54ecbfb6e6982abb13466a4b56 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Mon, 7 Jul 2008 14:16:52 -0400
Subject: sched_clock: stop maximum check on NO HZ

Working with ftrace I would get large jumps of 11 millisecs or more with
the clock tracer. This killed the latencing timings of ftrace and also
caused the irqoff self tests to fail.

What was happening is with NO_HZ the idle would stop the jiffy counter and
before the jiffy counter was updated the sched_clock would have a bad
delta jiffies to compare with the gtod with the maximum.

The jiffies would stop and the last sched_tick would record the last gtod.
On wakeup, the sched clock update would compare the gtod + delta jiffies
(which would be zero) and compare it to the TSC. The TSC would have
correctly (with a stable TSC) moved forward several jiffies. But because the
jiffies has not been updated yet the clock would be prevented from moving
forward because it would appear that the TSC jumped too far ahead.

The clock would then virtually stop, until the jiffies are updated. Then
the next sched clock update would see that the clock was very much behind
since the delta jiffies is now correct. This would then jump the clock
forward by several jiffies.

This caused ftrace to report several milliseconds of interrupts off
latency at every resume from NO_HZ idle.

This patch adds hooks into the nohz code to disable the checking of the
maximum clock update when nohz is in effect. It resumes the max check
when nohz has updated the jiffies again.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: Steven Rostedt <srostedt@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h    | 17 ++++++++++++++++-
 kernel/sched_clock.c     | 39 ++++++++++++++++++++++++++++++++++++++-
 kernel/time/tick-sched.c |  2 ++
 3 files changed, 56 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index c5d3f847ca8d..33a8f42041fa 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1573,13 +1573,28 @@ static inline void sched_clock_idle_sleep_event(void)
 static inline void sched_clock_idle_wakeup_event(u64 delta_ns)
 {
 }
-#else
+
+#ifdef CONFIG_NO_HZ
+static inline void sched_clock_tick_stop(int cpu)
+{
+}
+
+static inline void sched_clock_tick_start(int cpu)
+{
+}
+#endif
+
+#else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */
 extern void sched_clock_init(void);
 extern u64 sched_clock_cpu(int cpu);
 extern void sched_clock_tick(void);
 extern void sched_clock_idle_sleep_event(void);
 extern void sched_clock_idle_wakeup_event(u64 delta_ns);
+#ifdef CONFIG_NO_HZ
+extern void sched_clock_tick_stop(int cpu);
+extern void sched_clock_tick_start(int cpu);
 #endif
+#endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */
 
 /*
  * For kernel-internal use: high-speed (but slightly incorrect) per-cpu
diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c
index 42b81fa38cbd..97159e225a77 100644
--- a/kernel/sched_clock.c
+++ b/kernel/sched_clock.c
@@ -45,6 +45,9 @@ struct sched_clock_data {
 	u64			tick_raw;
 	u64			tick_gtod;
 	u64			clock;
+#ifdef CONFIG_NO_HZ
+	int			check_max;
+#endif
 };
 
 static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data);
@@ -76,11 +79,45 @@ void sched_clock_init(void)
 		scd->tick_raw = 0;
 		scd->tick_gtod = ktime_now;
 		scd->clock = ktime_now;
+#ifdef CONFIG_NO_HZ
+		scd->check_max = 1;
+#endif
 	}
 
 	sched_clock_running = 1;
 }
 
+#ifdef CONFIG_NO_HZ
+/*
+ * The dynamic ticks makes the delta jiffies inaccurate. This
+ * prevents us from checking the maximum time update.
+ * Disable the maximum check during stopped ticks.
+ */
+void sched_clock_tick_stop(int cpu)
+{
+	struct sched_clock_data *scd = cpu_sdc(cpu);
+
+	scd->check_max = 0;
+}
+
+void sched_clock_tick_start(int cpu)
+{
+	struct sched_clock_data *scd = cpu_sdc(cpu);
+
+	scd->check_max = 1;
+}
+
+static int check_max(struct sched_clock_data *scd)
+{
+	return scd->check_max;
+}
+#else
+static int check_max(struct sched_clock_data *scd)
+{
+	return 1;
+}
+#endif /* CONFIG_NO_HZ */
+
 /*
  * update the percpu scd from the raw @now value
  *
@@ -112,7 +149,7 @@ static void __update_sched_clock(struct sched_clock_data *scd, u64 now)
 	 */
 	max_clock = scd->tick_gtod + (2 + delta_jiffies) * TICK_NSEC;
 
-	if (unlikely(clock + delta > max_clock)) {
+	if (unlikely(clock + delta > max_clock) && check_max(scd)) {
 		if (clock < max_clock)
 			clock = max_clock;
 		else
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index b854a895591e..d63008b09a4c 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -276,6 +276,7 @@ void tick_nohz_stop_sched_tick(void)
 			ts->tick_stopped = 1;
 			ts->idle_jiffies = last_jiffies;
 			rcu_enter_nohz();
+			sched_clock_tick_stop(cpu);
 		}
 
 		/*
@@ -375,6 +376,7 @@ void tick_nohz_restart_sched_tick(void)
 	select_nohz_load_balancer(0);
 	now = ktime_get();
 	tick_do_update_jiffies64(now);
+	sched_clock_tick_start(cpu);
 	cpu_clear(cpu, nohz_cpu_mask);
 
 	/*
-- 
cgit v1.2.3


From 736603ab297506f4396cb5af592004499950fcfd Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Fri, 11 Jul 2008 19:27:31 -0400
Subject: jbd2: Add commit time into the commit block

Carlo Wood has demonstrated that it's possible to recover deleted
files from the journal.  Something that will make this easier is if we
can put the time of the commit into commit block.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/jbd2/commit.c     | 3 +++
 include/linux/jbd2.h | 2 ++
 2 files changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index a2ed72f7ceee..92b6ac3df8ab 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -112,6 +112,7 @@ static int journal_submit_commit_record(journal_t *journal,
 	struct buffer_head *bh;
 	int ret;
 	int barrier_done = 0;
+	struct timespec now = current_kernel_time();
 
 	if (is_journal_aborted(journal))
 		return 0;
@@ -126,6 +127,8 @@ static int journal_submit_commit_record(journal_t *journal,
 	tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER);
 	tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK);
 	tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid);
+	tmp->h_commit_sec = cpu_to_be64(now.tv_sec);
+	tmp->h_commit_nsec = cpu_to_be32(now.tv_nsec);
 
 	if (JBD2_HAS_COMPAT_FEATURE(journal,
 				    JBD2_FEATURE_COMPAT_CHECKSUM)) {
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index d147f0f90360..ec9cadf58227 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -168,6 +168,8 @@ struct commit_header {
 	unsigned char   h_chksum_size;
 	unsigned char 	h_padding[2];
 	__be32 		h_chksum[JBD2_CHECKSUM_BYTES];
+	__be64		h_commit_sec;
+	__be32		h_commit_nsec;
 };
 
 /*
-- 
cgit v1.2.3


From f4c0a0fdfae708f7aa438c27a380ed4071294e11 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 11 Jul 2008 19:27:31 -0400
Subject: vfs: export filemap_fdatawrite_range()

Make filemap_fdatawrite_range() function public, so that it can later
be used in ordered mode rewrite by JBD/JBD2.

Signed-off-by: Jan Kara <jack@suse.cz>
---
 include/linux/fs.h | 2 ++
 mm/filemap.c       | 3 ++-
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index d8e2762ed14d..97f992adc62d 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1740,6 +1740,8 @@ extern int wait_on_page_writeback_range(struct address_space *mapping,
 				pgoff_t start, pgoff_t end);
 extern int __filemap_fdatawrite_range(struct address_space *mapping,
 				loff_t start, loff_t end, int sync_mode);
+extern int filemap_fdatawrite_range(struct address_space *mapping,
+				loff_t start, loff_t end);
 
 extern long do_fsync(struct file *file, int datasync);
 extern void sync_supers(void);
diff --git a/mm/filemap.c b/mm/filemap.c
index 1e6a7d34874f..65d9d9e2b755 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -236,11 +236,12 @@ int filemap_fdatawrite(struct address_space *mapping)
 }
 EXPORT_SYMBOL(filemap_fdatawrite);
 
-static int filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
+int filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
 				loff_t end)
 {
 	return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL);
 }
+EXPORT_SYMBOL(filemap_fdatawrite_range);
 
 /**
  * filemap_flush - mostly a non-blocking flush
-- 
cgit v1.2.3


From c851ed540173736e60d48b53b91a16ea5c903896 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 11 Jul 2008 19:27:31 -0400
Subject: jbd2: Implement data=ordered mode handling via inodes

This patch adds necessary framework into JBD2 to be able to track inodes
with each transaction and write-out their dirty data during transaction
commit time.

This new ordered mode brings all sorts of advantages such as possibility
to get rid of journal heads and buffer heads for data buffers in ordered
mode, better ordering of writes on transaction commit, simplification of
 some JBD code, no more anonymous pages when truncate of data being
committed happens.  Also with this new ordered mode, delayed allocation
on ordered mode is much simpler.

Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/jbd2/commit.c      | 90 +++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/jbd2/journal.c     | 52 +++++++++++++++++++++++++++++
 fs/jbd2/transaction.c | 86 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/jbd2.h  | 42 ++++++++++++++++++++++++
 4 files changed, 270 insertions(+)

(limited to 'include/linux')

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 92b6ac3df8ab..3ca107b5c86b 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -355,6 +355,81 @@ write_out_data:
 	journal_do_submit_data(wbuf, bufs);
 }
 
+/*
+ * Submit all the data buffers of inode associated with the transaction to
+ * disk.
+ *
+ * We are in a committing transaction. Therefore no new inode can be added to
+ * our inode list. We use JI_COMMIT_RUNNING flag to protect inode we currently
+ * operate on from being released while we write out pages.
+ */
+static int journal_submit_inode_data_buffers(journal_t *journal,
+		transaction_t *commit_transaction)
+{
+	struct jbd2_inode *jinode;
+	int err, ret = 0;
+	struct address_space *mapping;
+
+	spin_lock(&journal->j_list_lock);
+	list_for_each_entry(jinode, &commit_transaction->t_inode_list, i_list) {
+		mapping = jinode->i_vfs_inode->i_mapping;
+		jinode->i_flags |= JI_COMMIT_RUNNING;
+		spin_unlock(&journal->j_list_lock);
+		err = filemap_fdatawrite_range(mapping, 0,
+					i_size_read(jinode->i_vfs_inode));
+		if (!ret)
+			ret = err;
+		spin_lock(&journal->j_list_lock);
+		J_ASSERT(jinode->i_transaction == commit_transaction);
+		jinode->i_flags &= ~JI_COMMIT_RUNNING;
+		wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING);
+	}
+	spin_unlock(&journal->j_list_lock);
+	return ret;
+}
+
+/*
+ * Wait for data submitted for writeout, refile inodes to proper
+ * transaction if needed.
+ *
+ */
+static int journal_finish_inode_data_buffers(journal_t *journal,
+		transaction_t *commit_transaction)
+{
+	struct jbd2_inode *jinode, *next_i;
+	int err, ret = 0;
+
+	/* For locking, see the comment in journal_submit_inode_data_buffers() */
+	spin_lock(&journal->j_list_lock);
+	list_for_each_entry(jinode, &commit_transaction->t_inode_list, i_list) {
+		jinode->i_flags |= JI_COMMIT_RUNNING;
+		spin_unlock(&journal->j_list_lock);
+		err = filemap_fdatawait(jinode->i_vfs_inode->i_mapping);
+		if (!ret)
+			ret = err;
+		spin_lock(&journal->j_list_lock);
+		jinode->i_flags &= ~JI_COMMIT_RUNNING;
+		wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING);
+	}
+
+	/* Now refile inode to proper lists */
+	list_for_each_entry_safe(jinode, next_i,
+				 &commit_transaction->t_inode_list, i_list) {
+		list_del(&jinode->i_list);
+		if (jinode->i_next_transaction) {
+			jinode->i_transaction = jinode->i_next_transaction;
+			jinode->i_next_transaction = NULL;
+			list_add(&jinode->i_list,
+				&jinode->i_transaction->t_inode_list);
+		} else {
+			jinode->i_transaction = NULL;
+		}
+	}
+	spin_unlock(&journal->j_list_lock);
+
+	return ret;
+}
+
 static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh)
 {
 	struct page *page = bh->b_page;
@@ -529,6 +604,9 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 */
 	err = 0;
 	journal_submit_data_buffers(journal, commit_transaction);
+	err = journal_submit_inode_data_buffers(journal, commit_transaction);
+	if (err)
+		jbd2_journal_abort(journal, err);
 
 	/*
 	 * Wait for all previously submitted IO to complete if commit
@@ -760,6 +838,17 @@ start_journal_io:
 			__jbd2_journal_abort_hard(journal);
 	}
 
+	/*
+	 * This is the right place to wait for data buffers both for ASYNC
+	 * and !ASYNC commit. If commit is ASYNC, we need to wait only after
+	 * the commit block went to disk (which happens above). If commit is
+	 * SYNC, we need to wait for data buffers before we start writing
+	 * commit block, which happens below in such setting.
+	 */
+	err = journal_finish_inode_data_buffers(journal, commit_transaction);
+	if (err)
+		jbd2_journal_abort(journal, err);
+
 	/* Lo and behold: we have just managed to send a transaction to
            the log.  Before we can commit it, wait for the IO so far to
            complete.  Control buffers being written are on the
@@ -880,6 +969,7 @@ wait_for_iobuf:
 	jbd_debug(3, "JBD: commit phase 7\n");
 
 	J_ASSERT(commit_transaction->t_sync_datalist == NULL);
+	J_ASSERT(list_empty(&commit_transaction->t_inode_list));
 	J_ASSERT(commit_transaction->t_buffers == NULL);
 	J_ASSERT(commit_transaction->t_checkpoint_list == NULL);
 	J_ASSERT(commit_transaction->t_iobuf_list == NULL);
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 2e24567c4a79..78cf7bd7f604 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -82,6 +82,10 @@ EXPORT_SYMBOL(jbd2_journal_blocks_per_page);
 EXPORT_SYMBOL(jbd2_journal_invalidatepage);
 EXPORT_SYMBOL(jbd2_journal_try_to_free_buffers);
 EXPORT_SYMBOL(jbd2_journal_force_commit);
+EXPORT_SYMBOL(jbd2_journal_file_inode);
+EXPORT_SYMBOL(jbd2_journal_init_jbd_inode);
+EXPORT_SYMBOL(jbd2_journal_release_jbd_inode);
+EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate);
 
 static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *);
 static void __journal_abort_soft (journal_t *journal, int errno);
@@ -2194,6 +2198,54 @@ void jbd2_journal_put_journal_head(struct journal_head *jh)
 	jbd_unlock_bh_journal_head(bh);
 }
 
+/*
+ * Initialize jbd inode head
+ */
+void jbd2_journal_init_jbd_inode(struct jbd2_inode *jinode, struct inode *inode)
+{
+	jinode->i_transaction = NULL;
+	jinode->i_next_transaction = NULL;
+	jinode->i_vfs_inode = inode;
+	jinode->i_flags = 0;
+	INIT_LIST_HEAD(&jinode->i_list);
+}
+
+/*
+ * Function to be called before we start removing inode from memory (i.e.,
+ * clear_inode() is a fine place to be called from). It removes inode from
+ * transaction's lists.
+ */
+void jbd2_journal_release_jbd_inode(journal_t *journal,
+				    struct jbd2_inode *jinode)
+{
+	int writeout = 0;
+
+	if (!journal)
+		return;
+restart:
+	spin_lock(&journal->j_list_lock);
+	/* Is commit writing out inode - we have to wait */
+	if (jinode->i_flags & JI_COMMIT_RUNNING) {
+		wait_queue_head_t *wq;
+		DEFINE_WAIT_BIT(wait, &jinode->i_flags, __JI_COMMIT_RUNNING);
+		wq = bit_waitqueue(&jinode->i_flags, __JI_COMMIT_RUNNING);
+		prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE);
+		spin_unlock(&journal->j_list_lock);
+		schedule();
+		finish_wait(wq, &wait.wait);
+		goto restart;
+	}
+
+	/* Do we need to wait for data writeback? */
+	if (journal->j_committing_transaction == jinode->i_transaction)
+		writeout = 1;
+	if (jinode->i_transaction) {
+		list_del(&jinode->i_list);
+		jinode->i_transaction = NULL;
+	}
+	spin_unlock(&journal->j_list_lock);
+}
+
 /*
  * debugfs tunables
  */
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index ba620c4493d2..98b596d23705 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -51,6 +51,7 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
 	transaction->t_tid = journal->j_transaction_sequence++;
 	transaction->t_expires = jiffies + journal->j_commit_interval;
 	spin_lock_init(&transaction->t_handle_lock);
+	INIT_LIST_HEAD(&transaction->t_inode_list);
 
 	/* Set up the commit timer for the new transaction. */
 	journal->j_commit_timer.expires = round_jiffies(transaction->t_expires);
@@ -2195,3 +2196,88 @@ void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh)
 	spin_unlock(&journal->j_list_lock);
 	__brelse(bh);
 }
+
+/*
+ * File inode in the inode list of the handle's transaction
+ */
+int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *jinode)
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+
+	if (is_handle_aborted(handle))
+		return -EIO;
+
+	jbd_debug(4, "Adding inode %lu, tid:%d\n", jinode->i_vfs_inode->i_ino,
+			transaction->t_tid);
+
+	/*
+	 * First check whether inode isn't already on the transaction's
+	 * lists without taking the lock. Note that this check is safe
+	 * without the lock as we cannot race with somebody removing inode
+	 * from the transaction. The reason is that we remove inode from the
+	 * transaction only in journal_release_jbd_inode() and when we commit
+	 * the transaction. We are guarded from the first case by holding
+	 * a reference to the inode. We are safe against the second case
+	 * because if jinode->i_transaction == transaction, commit code
+	 * cannot touch the transaction because we hold reference to it,
+	 * and if jinode->i_next_transaction == transaction, commit code
+	 * will only file the inode where we want it.
+	 */
+	if (jinode->i_transaction == transaction ||
+	    jinode->i_next_transaction == transaction)
+		return 0;
+
+	spin_lock(&journal->j_list_lock);
+
+	if (jinode->i_transaction == transaction ||
+	    jinode->i_next_transaction == transaction)
+		goto done;
+
+	/* On some different transaction's list - should be
+	 * the committing one */
+	if (jinode->i_transaction) {
+		J_ASSERT(jinode->i_next_transaction == NULL);
+		J_ASSERT(jinode->i_transaction ==
+					journal->j_committing_transaction);
+		jinode->i_next_transaction = transaction;
+		goto done;
+	}
+	/* Not on any transaction list... */
+	J_ASSERT(!jinode->i_next_transaction);
+	jinode->i_transaction = transaction;
+	list_add(&jinode->i_list, &transaction->t_inode_list);
+done:
+	spin_unlock(&journal->j_list_lock);
+
+	return 0;
+}
+
+/*
+ * This function must be called when inode is journaled in ordered mode
+ * before truncation happens. It starts writeout of truncated part in
+ * case it is in the committing transaction so that we stand to ordered
+ * mode consistency guarantees.
+ */
+int jbd2_journal_begin_ordered_truncate(struct jbd2_inode *inode,
+					loff_t new_size)
+{
+	journal_t *journal;
+	transaction_t *commit_trans;
+	int ret = 0;
+
+	if (!inode->i_transaction && !inode->i_next_transaction)
+		goto out;
+	journal = inode->i_transaction->t_journal;
+	spin_lock(&journal->j_state_lock);
+	commit_trans = journal->j_committing_transaction;
+	spin_unlock(&journal->j_state_lock);
+	if (inode->i_transaction == commit_trans) {
+		ret = filemap_fdatawrite_range(inode->i_vfs_inode->i_mapping,
+			new_size, LLONG_MAX);
+		if (ret)
+			jbd2_journal_abort(journal, ret);
+	}
+out:
+	return ret;
+}
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index ec9cadf58227..622c3d8ca4ed 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -381,6 +381,38 @@ static inline void jbd_unlock_bh_journal_head(struct buffer_head *bh)
 	bit_spin_unlock(BH_JournalHead, &bh->b_state);
 }
 
+/* Flags in jbd_inode->i_flags */
+#define __JI_COMMIT_RUNNING 0
+/* Commit of the inode data in progress. We use this flag to protect us from
+ * concurrent deletion of inode. We cannot use reference to inode for this
+ * since we cannot afford doing last iput() on behalf of kjournald
+ */
+#define JI_COMMIT_RUNNING (1 << __JI_COMMIT_RUNNING)
+
+/**
+ * struct jbd_inode is the structure linking inodes in ordered mode
+ *   present in a transaction so that we can sync them during commit.
+ */
+struct jbd2_inode {
+	/* Which transaction does this inode belong to? Either the running
+	 * transaction or the committing one. [j_list_lock] */
+	transaction_t *i_transaction;
+
+	/* Pointer to the running transaction modifying inode's data in case
+	 * there is already a committing transaction touching it. [j_list_lock] */
+	transaction_t *i_next_transaction;
+
+	/* List of inodes in the i_transaction [j_list_lock] */
+	struct list_head i_list;
+
+	/* VFS inode this inode belongs to [constant during the lifetime
+	 * of the structure] */
+	struct inode *i_vfs_inode;
+
+	/* Flags of inode [j_list_lock] */
+	unsigned int i_flags;
+};
+
 struct jbd2_revoke_table_s;
 
 /**
@@ -566,6 +598,12 @@ struct transaction_s
 	 */
 	struct journal_head	*t_log_list;
 
+	/*
+	 * List of inodes whose data we've modified in data=ordered mode.
+	 * [j_list_lock]
+	 */
+	struct list_head	t_inode_list;
+
 	/*
 	 * Protects info related to handles
 	 */
@@ -1046,6 +1084,10 @@ extern void	   jbd2_journal_ack_err    (journal_t *);
 extern int	   jbd2_journal_clear_err  (journal_t *);
 extern int	   jbd2_journal_bmap(journal_t *, unsigned long, unsigned long long *);
 extern int	   jbd2_journal_force_commit(journal_t *);
+extern int	   jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *inode);
+extern int	   jbd2_journal_begin_ordered_truncate(struct jbd2_inode *inode, loff_t new_size);
+extern void	   jbd2_journal_init_jbd_inode(struct jbd2_inode *jinode, struct inode *inode);
+extern void	   jbd2_journal_release_jbd_inode(journal_t *journal, struct jbd2_inode *jinode);
 
 /*
  * journal_head management
-- 
cgit v1.2.3


From 87c89c232c8f7b3820c33c3b9bc803e9358027da Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 11 Jul 2008 19:27:31 -0400
Subject: jbd2: Remove data=ordered mode support using jbd buffer heads

Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/jbd2/checkpoint.c  |   1 -
 fs/jbd2/commit.c      | 221 ++------------------------------------------------
 fs/jbd2/journal.c     |   1 -
 fs/jbd2/transaction.c | 217 ++-----------------------------------------------
 include/linux/jbd2.h  |  29 ++-----
 5 files changed, 21 insertions(+), 448 deletions(-)

(limited to 'include/linux')

diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 6914598022ce..91389c8aee8a 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -688,7 +688,6 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact
 
 	J_ASSERT(transaction->t_state == T_FINISHED);
 	J_ASSERT(transaction->t_buffers == NULL);
-	J_ASSERT(transaction->t_sync_datalist == NULL);
 	J_ASSERT(transaction->t_forget == NULL);
 	J_ASSERT(transaction->t_iobuf_list == NULL);
 	J_ASSERT(transaction->t_shadow_list == NULL);
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 3ca107b5c86b..483183d15ed5 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -37,8 +37,8 @@ static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate)
 }
 
 /*
- * When an ext3-ordered file is truncated, it is possible that many pages are
- * not sucessfully freed, because they are attached to a committing transaction.
+ * When an ext4 file is truncated, it is possible that some pages are not
+ * successfully freed, because they are attached to a committing transaction.
  * After the transaction commits, these pages are left on the LRU, with no
  * ->mapping, and with attached buffers.  These pages are trivially reclaimable
  * by the VM, but their apparent absence upsets the VM accounting, and it makes
@@ -79,21 +79,6 @@ nope:
 	__brelse(bh);
 }
 
-/*
- * Try to acquire jbd_lock_bh_state() against the buffer, when j_list_lock is
- * held.  For ranking reasons we must trylock.  If we lose, schedule away and
- * return 0.  j_list_lock is dropped in this case.
- */
-static int inverted_lock(journal_t *journal, struct buffer_head *bh)
-{
-	if (!jbd_trylock_bh_state(bh)) {
-		spin_unlock(&journal->j_list_lock);
-		schedule();
-		return 0;
-	}
-	return 1;
-}
-
 /*
  * Done it all: now submit the commit record.  We should have
  * cleaned up our previous buffers by now, so if we are in abort
@@ -199,162 +184,6 @@ static int journal_wait_on_commit_record(struct buffer_head *bh)
 	return ret;
 }
 
-/*
- * Wait for all submitted IO to complete.
- */
-static int journal_wait_on_locked_list(journal_t *journal,
-				       transaction_t *commit_transaction)
-{
-	int ret = 0;
-	struct journal_head *jh;
-
-	while (commit_transaction->t_locked_list) {
-		struct buffer_head *bh;
-
-		jh = commit_transaction->t_locked_list->b_tprev;
-		bh = jh2bh(jh);
-		get_bh(bh);
-		if (buffer_locked(bh)) {
-			spin_unlock(&journal->j_list_lock);
-			wait_on_buffer(bh);
-			if (unlikely(!buffer_uptodate(bh)))
-				ret = -EIO;
-			spin_lock(&journal->j_list_lock);
-		}
-		if (!inverted_lock(journal, bh)) {
-			put_bh(bh);
-			spin_lock(&journal->j_list_lock);
-			continue;
-		}
-		if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) {
-			__jbd2_journal_unfile_buffer(jh);
-			jbd_unlock_bh_state(bh);
-			jbd2_journal_remove_journal_head(bh);
-			put_bh(bh);
-		} else {
-			jbd_unlock_bh_state(bh);
-		}
-		put_bh(bh);
-		cond_resched_lock(&journal->j_list_lock);
-	}
-	return ret;
-  }
-
-static void journal_do_submit_data(struct buffer_head **wbuf, int bufs)
-{
-	int i;
-
-	for (i = 0; i < bufs; i++) {
-		wbuf[i]->b_end_io = end_buffer_write_sync;
-		/* We use-up our safety reference in submit_bh() */
-		submit_bh(WRITE, wbuf[i]);
-	}
-}
-
-/*
- *  Submit all the data buffers to disk
- */
-static void journal_submit_data_buffers(journal_t *journal,
-				transaction_t *commit_transaction)
-{
-	struct journal_head *jh;
-	struct buffer_head *bh;
-	int locked;
-	int bufs = 0;
-	struct buffer_head **wbuf = journal->j_wbuf;
-
-	/*
-	 * Whenever we unlock the journal and sleep, things can get added
-	 * onto ->t_sync_datalist, so we have to keep looping back to
-	 * write_out_data until we *know* that the list is empty.
-	 *
-	 * Cleanup any flushed data buffers from the data list.  Even in
-	 * abort mode, we want to flush this out as soon as possible.
-	 */
-write_out_data:
-	cond_resched();
-	spin_lock(&journal->j_list_lock);
-
-	while (commit_transaction->t_sync_datalist) {
-		jh = commit_transaction->t_sync_datalist;
-		bh = jh2bh(jh);
-		locked = 0;
-
-		/* Get reference just to make sure buffer does not disappear
-		 * when we are forced to drop various locks */
-		get_bh(bh);
-		/* If the buffer is dirty, we need to submit IO and hence
-		 * we need the buffer lock. We try to lock the buffer without
-		 * blocking. If we fail, we need to drop j_list_lock and do
-		 * blocking lock_buffer().
-		 */
-		if (buffer_dirty(bh)) {
-			if (test_set_buffer_locked(bh)) {
-				BUFFER_TRACE(bh, "needs blocking lock");
-				spin_unlock(&journal->j_list_lock);
-				/* Write out all data to prevent deadlocks */
-				journal_do_submit_data(wbuf, bufs);
-				bufs = 0;
-				lock_buffer(bh);
-				spin_lock(&journal->j_list_lock);
-			}
-			locked = 1;
-		}
-		/* We have to get bh_state lock. Again out of order, sigh. */
-		if (!inverted_lock(journal, bh)) {
-			jbd_lock_bh_state(bh);
-			spin_lock(&journal->j_list_lock);
-		}
-		/* Someone already cleaned up the buffer? */
-		if (!buffer_jbd(bh)
-			|| jh->b_transaction != commit_transaction
-			|| jh->b_jlist != BJ_SyncData) {
-			jbd_unlock_bh_state(bh);
-			if (locked)
-				unlock_buffer(bh);
-			BUFFER_TRACE(bh, "already cleaned up");
-			put_bh(bh);
-			continue;
-		}
-		if (locked && test_clear_buffer_dirty(bh)) {
-			BUFFER_TRACE(bh, "needs writeout, adding to array");
-			wbuf[bufs++] = bh;
-			__jbd2_journal_file_buffer(jh, commit_transaction,
-						BJ_Locked);
-			jbd_unlock_bh_state(bh);
-			if (bufs == journal->j_wbufsize) {
-				spin_unlock(&journal->j_list_lock);
-				journal_do_submit_data(wbuf, bufs);
-				bufs = 0;
-				goto write_out_data;
-			}
-		} else if (!locked && buffer_locked(bh)) {
-			__jbd2_journal_file_buffer(jh, commit_transaction,
-						BJ_Locked);
-			jbd_unlock_bh_state(bh);
-			put_bh(bh);
-		} else {
-			BUFFER_TRACE(bh, "writeout complete: unfile");
-			__jbd2_journal_unfile_buffer(jh);
-			jbd_unlock_bh_state(bh);
-			if (locked)
-				unlock_buffer(bh);
-			jbd2_journal_remove_journal_head(bh);
-			/* Once for our safety reference, once for
-			 * jbd2_journal_remove_journal_head() */
-			put_bh(bh);
-			put_bh(bh);
-		}
-
-		if (need_resched() || spin_needbreak(&journal->j_list_lock)) {
-			spin_unlock(&journal->j_list_lock);
-			goto write_out_data;
-		}
-	}
-	spin_unlock(&journal->j_list_lock);
-	journal_do_submit_data(wbuf, bufs);
-}
-
 /*
  * Submit all the data buffers of inode associated with the transaction to
  * disk.
@@ -602,24 +431,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 * Now start flushing things to disk, in the order they appear
 	 * on the transaction lists.  Data blocks go first.
 	 */
-	err = 0;
-	journal_submit_data_buffers(journal, commit_transaction);
 	err = journal_submit_inode_data_buffers(journal, commit_transaction);
-	if (err)
-		jbd2_journal_abort(journal, err);
-
-	/*
-	 * Wait for all previously submitted IO to complete if commit
-	 * record is to be written synchronously.
-	 */
-	spin_lock(&journal->j_list_lock);
-	if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
-		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT))
-		err = journal_wait_on_locked_list(journal,
-						commit_transaction);
-
-	spin_unlock(&journal->j_list_lock);
-
 	if (err)
 		jbd2_journal_abort(journal, err);
 
@@ -627,16 +439,6 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 
 	jbd_debug(3, "JBD: commit phase 2\n");
 
-	/*
-	 * If we found any dirty or locked buffers, then we should have
-	 * looped back up to the write_out_data label.  If there weren't
-	 * any then journal_clean_data_list should have wiped the list
-	 * clean by now, so check that it is in fact empty.
-	 */
-	J_ASSERT (commit_transaction->t_sync_datalist == NULL);
-
-	jbd_debug (3, "JBD: commit phase 3\n");
-
 	/*
 	 * Way to go: we have now written out all of the data for a
 	 * transaction!  Now comes the tricky part: we need to write out
@@ -655,6 +457,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	J_ASSERT(commit_transaction->t_nr_buffers <=
 		 commit_transaction->t_outstanding_credits);
 
+	err = 0;
 	descriptor = NULL;
 	bufs = 0;
 	while (commit_transaction->t_buffers) {
@@ -829,13 +632,6 @@ start_journal_io:
 						 &cbh, crc32_sum);
 		if (err)
 			__jbd2_journal_abort_hard(journal);
-
-		spin_lock(&journal->j_list_lock);
-		err = journal_wait_on_locked_list(journal,
-						commit_transaction);
-		spin_unlock(&journal->j_list_lock);
-		if (err)
-			__jbd2_journal_abort_hard(journal);
 	}
 
 	/*
@@ -860,7 +656,7 @@ start_journal_io:
 	   so we incur less scheduling load.
 	*/
 
-	jbd_debug(3, "JBD: commit phase 4\n");
+	jbd_debug(3, "JBD: commit phase 3\n");
 
 	/*
 	 * akpm: these are BJ_IO, and j_list_lock is not needed.
@@ -919,7 +715,7 @@ wait_for_iobuf:
 
 	J_ASSERT (commit_transaction->t_shadow_list == NULL);
 
-	jbd_debug(3, "JBD: commit phase 5\n");
+	jbd_debug(3, "JBD: commit phase 4\n");
 
 	/* Here we wait for the revoke record and descriptor record buffers */
  wait_for_ctlbuf:
@@ -946,7 +742,7 @@ wait_for_iobuf:
 		/* AKPM: bforget here */
 	}
 
-	jbd_debug(3, "JBD: commit phase 6\n");
+	jbd_debug(3, "JBD: commit phase 5\n");
 
 	if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
 		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
@@ -966,9 +762,8 @@ wait_for_iobuf:
            transaction can be removed from any checkpoint list it was on
            before. */
 
-	jbd_debug(3, "JBD: commit phase 7\n");
+	jbd_debug(3, "JBD: commit phase 6\n");
 
-	J_ASSERT(commit_transaction->t_sync_datalist == NULL);
 	J_ASSERT(list_empty(&commit_transaction->t_inode_list));
 	J_ASSERT(commit_transaction->t_buffers == NULL);
 	J_ASSERT(commit_transaction->t_checkpoint_list == NULL);
@@ -1090,7 +885,7 @@ restart_loop:
 
 	/* Done with this transaction! */
 
-	jbd_debug(3, "JBD: commit phase 8\n");
+	jbd_debug(3, "JBD: commit phase 7\n");
 
 	J_ASSERT(commit_transaction->t_state == T_COMMIT);
 
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 78cf7bd7f604..b26c6d9fe6ae 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -50,7 +50,6 @@ EXPORT_SYMBOL(jbd2_journal_unlock_updates);
 EXPORT_SYMBOL(jbd2_journal_get_write_access);
 EXPORT_SYMBOL(jbd2_journal_get_create_access);
 EXPORT_SYMBOL(jbd2_journal_get_undo_access);
-EXPORT_SYMBOL(jbd2_journal_dirty_data);
 EXPORT_SYMBOL(jbd2_journal_dirty_metadata);
 EXPORT_SYMBOL(jbd2_journal_release_buffer);
 EXPORT_SYMBOL(jbd2_journal_forget);
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index 98b596d23705..4f7cadbb19fa 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -942,183 +942,6 @@ out:
 	return err;
 }
 
-/**
- * int jbd2_journal_dirty_data() -  mark a buffer as containing dirty data which
- *                             needs to be flushed before we can commit the
- *                             current transaction.
- * @handle: transaction
- * @bh: bufferhead to mark
- *
- * The buffer is placed on the transaction's data list and is marked as
- * belonging to the transaction.
- *
- * Returns error number or 0 on success.
- *
- * jbd2_journal_dirty_data() can be called via page_launder->ext3_writepage
- * by kswapd.
- */
-int jbd2_journal_dirty_data(handle_t *handle, struct buffer_head *bh)
-{
-	journal_t *journal = handle->h_transaction->t_journal;
-	int need_brelse = 0;
-	struct journal_head *jh;
-
-	if (is_handle_aborted(handle))
-		return 0;
-
-	jh = jbd2_journal_add_journal_head(bh);
-	JBUFFER_TRACE(jh, "entry");
-
-	/*
-	 * The buffer could *already* be dirty.  Writeout can start
-	 * at any time.
-	 */
-	jbd_debug(4, "jh: %p, tid:%d\n", jh, handle->h_transaction->t_tid);
-
-	/*
-	 * What if the buffer is already part of a running transaction?
-	 *
-	 * There are two cases:
-	 * 1) It is part of the current running transaction.  Refile it,
-	 *    just in case we have allocated it as metadata, deallocated
-	 *    it, then reallocated it as data.
-	 * 2) It is part of the previous, still-committing transaction.
-	 *    If all we want to do is to guarantee that the buffer will be
-	 *    written to disk before this new transaction commits, then
-	 *    being sure that the *previous* transaction has this same
-	 *    property is sufficient for us!  Just leave it on its old
-	 *    transaction.
-	 *
-	 * In case (2), the buffer must not already exist as metadata
-	 * --- that would violate write ordering (a transaction is free
-	 * to write its data at any point, even before the previous
-	 * committing transaction has committed).  The caller must
-	 * never, ever allow this to happen: there's nothing we can do
-	 * about it in this layer.
-	 */
-	jbd_lock_bh_state(bh);
-	spin_lock(&journal->j_list_lock);
-
-	/* Now that we have bh_state locked, are we really still mapped? */
-	if (!buffer_mapped(bh)) {
-		JBUFFER_TRACE(jh, "unmapped buffer, bailing out");
-		goto no_journal;
-	}
-
-	if (jh->b_transaction) {
-		JBUFFER_TRACE(jh, "has transaction");
-		if (jh->b_transaction != handle->h_transaction) {
-			JBUFFER_TRACE(jh, "belongs to older transaction");
-			J_ASSERT_JH(jh, jh->b_transaction ==
-					journal->j_committing_transaction);
-
-			/* @@@ IS THIS TRUE  ? */
-			/*
-			 * Not any more.  Scenario: someone does a write()
-			 * in data=journal mode.  The buffer's transaction has
-			 * moved into commit.  Then someone does another
-			 * write() to the file.  We do the frozen data copyout
-			 * and set b_next_transaction to point to j_running_t.
-			 * And while we're in that state, someone does a
-			 * writepage() in an attempt to pageout the same area
-			 * of the file via a shared mapping.  At present that
-			 * calls jbd2_journal_dirty_data(), and we get right here.
-			 * It may be too late to journal the data.  Simply
-			 * falling through to the next test will suffice: the
-			 * data will be dirty and wil be checkpointed.  The
-			 * ordering comments in the next comment block still
-			 * apply.
-			 */
-			//J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
-
-			/*
-			 * If we're journalling data, and this buffer was
-			 * subject to a write(), it could be metadata, forget
-			 * or shadow against the committing transaction.  Now,
-			 * someone has dirtied the same darn page via a mapping
-			 * and it is being writepage()'d.
-			 * We *could* just steal the page from commit, with some
-			 * fancy locking there.  Instead, we just skip it -
-			 * don't tie the page's buffers to the new transaction
-			 * at all.
-			 * Implication: if we crash before the writepage() data
-			 * is written into the filesystem, recovery will replay
-			 * the write() data.
-			 */
-			if (jh->b_jlist != BJ_None &&
-					jh->b_jlist != BJ_SyncData &&
-					jh->b_jlist != BJ_Locked) {
-				JBUFFER_TRACE(jh, "Not stealing");
-				goto no_journal;
-			}
-
-			/*
-			 * This buffer may be undergoing writeout in commit.  We
-			 * can't return from here and let the caller dirty it
-			 * again because that can cause the write-out loop in
-			 * commit to never terminate.
-			 */
-			if (buffer_dirty(bh)) {
-				get_bh(bh);
-				spin_unlock(&journal->j_list_lock);
-				jbd_unlock_bh_state(bh);
-				need_brelse = 1;
-				sync_dirty_buffer(bh);
-				jbd_lock_bh_state(bh);
-				spin_lock(&journal->j_list_lock);
-				/* Since we dropped the lock... */
-				if (!buffer_mapped(bh)) {
-					JBUFFER_TRACE(jh, "buffer got unmapped");
-					goto no_journal;
-				}
-				/* The buffer may become locked again at any
-				   time if it is redirtied */
-			}
-
-			/* journal_clean_data_list() may have got there first */
-			if (jh->b_transaction != NULL) {
-				JBUFFER_TRACE(jh, "unfile from commit");
-				__jbd2_journal_temp_unlink_buffer(jh);
-				/* It still points to the committing
-				 * transaction; move it to this one so
-				 * that the refile assert checks are
-				 * happy. */
-				jh->b_transaction = handle->h_transaction;
-			}
-			/* The buffer will be refiled below */
-
-		}
-		/*
-		 * Special case --- the buffer might actually have been
-		 * allocated and then immediately deallocated in the previous,
-		 * committing transaction, so might still be left on that
-		 * transaction's metadata lists.
-		 */
-		if (jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) {
-			JBUFFER_TRACE(jh, "not on correct data list: unfile");
-			J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow);
-			__jbd2_journal_temp_unlink_buffer(jh);
-			jh->b_transaction = handle->h_transaction;
-			JBUFFER_TRACE(jh, "file as data");
-			__jbd2_journal_file_buffer(jh, handle->h_transaction,
-						BJ_SyncData);
-		}
-	} else {
-		JBUFFER_TRACE(jh, "not on a transaction");
-		__jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_SyncData);
-	}
-no_journal:
-	spin_unlock(&journal->j_list_lock);
-	jbd_unlock_bh_state(bh);
-	if (need_brelse) {
-		BUFFER_TRACE(bh, "brelse");
-		__brelse(bh);
-	}
-	JBUFFER_TRACE(jh, "exit");
-	jbd2_journal_put_journal_head(jh);
-	return 0;
-}
-
 /**
  * int jbd2_journal_dirty_metadata() -  mark a buffer as containing dirty metadata
  * @handle: transaction to add buffer to.
@@ -1541,10 +1364,10 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh)
  * Remove a buffer from the appropriate transaction list.
  *
  * Note that this function can *change* the value of
- * bh->b_transaction->t_sync_datalist, t_buffers, t_forget,
- * t_iobuf_list, t_shadow_list, t_log_list or t_reserved_list.  If the caller
- * is holding onto a copy of one of thee pointers, it could go bad.
- * Generally the caller needs to re-read the pointer from the transaction_t.
+ * bh->b_transaction->t_buffers, t_forget, t_iobuf_list, t_shadow_list,
+ * t_log_list or t_reserved_list.  If the caller is holding onto a copy of one
+ * of these pointers, it could go bad.  Generally the caller needs to re-read
+ * the pointer from the transaction_t.
  *
  * Called under j_list_lock.  The journal may not be locked.
  */
@@ -1566,9 +1389,6 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
 	switch (jh->b_jlist) {
 	case BJ_None:
 		return;
-	case BJ_SyncData:
-		list = &transaction->t_sync_datalist;
-		break;
 	case BJ_Metadata:
 		transaction->t_nr_buffers--;
 		J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0);
@@ -1589,9 +1409,6 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
 	case BJ_Reserved:
 		list = &transaction->t_reserved_list;
 		break;
-	case BJ_Locked:
-		list = &transaction->t_locked_list;
-		break;
 	}
 
 	__blist_del_buffer(list, jh);
@@ -1634,15 +1451,7 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh)
 		goto out;
 
 	spin_lock(&journal->j_list_lock);
-	if (jh->b_transaction != NULL && jh->b_cp_transaction == NULL) {
-		if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) {
-			/* A written-back ordered data buffer */
-			JBUFFER_TRACE(jh, "release data");
-			__jbd2_journal_unfile_buffer(jh);
-			jbd2_journal_remove_journal_head(bh);
-			__brelse(bh);
-		}
-	} else if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) {
+	if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) {
 		/* written-back checkpointed metadata buffer */
 		if (jh->b_jlist == BJ_None) {
 			JBUFFER_TRACE(jh, "remove from checkpoint list");
@@ -1878,6 +1687,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 	if (!buffer_jbd(bh))
 		goto zap_buffer_unlocked;
 
+	/* OK, we have data buffer in journaled mode */
 	spin_lock(&journal->j_state_lock);
 	jbd_lock_bh_state(bh);
 	spin_lock(&journal->j_list_lock);
@@ -1941,15 +1751,6 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 		}
 	} else if (transaction == journal->j_committing_transaction) {
 		JBUFFER_TRACE(jh, "on committing transaction");
-		if (jh->b_jlist == BJ_Locked) {
-			/*
-			 * The buffer is on the committing transaction's locked
-			 * list.  We have the buffer locked, so I/O has
-			 * completed.  So we can nail the buffer now.
-			 */
-			may_free = __dispose_buffer(jh, transaction);
-			goto zap_buffer;
-		}
 		/*
 		 * If it is committing, we simply cannot touch it.  We
 		 * can remove it's next_transaction pointer from the
@@ -2082,9 +1883,6 @@ void __jbd2_journal_file_buffer(struct journal_head *jh,
 		J_ASSERT_JH(jh, !jh->b_committed_data);
 		J_ASSERT_JH(jh, !jh->b_frozen_data);
 		return;
-	case BJ_SyncData:
-		list = &transaction->t_sync_datalist;
-		break;
 	case BJ_Metadata:
 		transaction->t_nr_buffers++;
 		list = &transaction->t_buffers;
@@ -2104,9 +1902,6 @@ void __jbd2_journal_file_buffer(struct journal_head *jh,
 	case BJ_Reserved:
 		list = &transaction->t_reserved_list;
 		break;
-	case BJ_Locked:
-		list =  &transaction->t_locked_list;
-		break;
 	}
 
 	__blist_add_buffer(list, jh);
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 622c3d8ca4ed..3dd209007098 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -542,24 +542,12 @@ struct transaction_s
 	 */
 	struct journal_head	*t_reserved_list;
 
-	/*
-	 * Doubly-linked circular list of all buffers under writeout during
-	 * commit [j_list_lock]
-	 */
-	struct journal_head	*t_locked_list;
-
 	/*
 	 * Doubly-linked circular list of all metadata buffers owned by this
 	 * transaction [j_list_lock]
 	 */
 	struct journal_head	*t_buffers;
 
-	/*
-	 * Doubly-linked circular list of all data buffers still to be
-	 * flushed before this transaction can be committed [j_list_lock]
-	 */
-	struct journal_head	*t_sync_datalist;
-
 	/*
 	 * Doubly-linked circular list of all forget buffers (superseded
 	 * buffers which we can un-checkpoint once this transaction commits)
@@ -1044,7 +1032,6 @@ extern int	 jbd2_journal_extend (handle_t *, int nblocks);
 extern int	 jbd2_journal_get_write_access(handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_get_create_access (handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_get_undo_access(handle_t *, struct buffer_head *);
-extern int	 jbd2_journal_dirty_data (handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_dirty_metadata (handle_t *, struct buffer_head *);
 extern void	 jbd2_journal_release_buffer (handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_forget (handle_t *, struct buffer_head *);
@@ -1223,15 +1210,13 @@ static inline int jbd_space_needed(journal_t *journal)
 
 /* journaling buffer types */
 #define BJ_None		0	/* Not journaled */
-#define BJ_SyncData	1	/* Normal data: flush before commit */
-#define BJ_Metadata	2	/* Normal journaled metadata */
-#define BJ_Forget	3	/* Buffer superseded by this transaction */
-#define BJ_IO		4	/* Buffer is for temporary IO use */
-#define BJ_Shadow	5	/* Buffer contents being shadowed to the log */
-#define BJ_LogCtl	6	/* Buffer contains log descriptors */
-#define BJ_Reserved	7	/* Buffer is reserved for access by journal */
-#define BJ_Locked	8	/* Locked for I/O during commit */
-#define BJ_Types	9
+#define BJ_Metadata	1	/* Normal journaled metadata */
+#define BJ_Forget	2	/* Buffer superseded by this transaction */
+#define BJ_IO		3	/* Buffer is for temporary IO use */
+#define BJ_Shadow	4	/* Buffer contents being shadowed to the log */
+#define BJ_LogCtl	5	/* Buffer contains log descriptors */
+#define BJ_Reserved	6	/* Buffer is reserved for access by journal */
+#define BJ_Types	7
 
 extern int jbd_blocks_per_page(struct inode *inode);
 
-- 
cgit v1.2.3


From 29a814d2ee0e43c2980f33f91c1311ec06c0aa35 Mon Sep 17 00:00:00 2001
From: Alex Tomas <alex@clusterfs.com>
Date: Fri, 11 Jul 2008 19:27:31 -0400
Subject: vfs: add hooks for ext4's delayed allocation support

Export mpage_bio_submit() and __mpage_writepage() for the benefit of
ext4's delayed allocation support.   Also change __block_write_full_page
so that if buffers that have the BH_Delay flag set it will call
get_block() to get the physical block allocated, just as in the
!BH_Mapped case.

Signed-off-by: Alex Tomas <alex@clusterfs.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/buffer.c           |  7 +++++--
 fs/mpage.c            | 14 +++++---------
 include/linux/mpage.h | 10 ++++++++++
 3 files changed, 20 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/fs/buffer.c b/fs/buffer.c
index f4b033237a02..5fa1512cd9a2 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1691,11 +1691,13 @@ static int __block_write_full_page(struct inode *inode, struct page *page,
 			 */
 			clear_buffer_dirty(bh);
 			set_buffer_uptodate(bh);
-		} else if (!buffer_mapped(bh) && buffer_dirty(bh)) {
+		} else if ((!buffer_mapped(bh) || buffer_delay(bh)) &&
+			   buffer_dirty(bh)) {
 			WARN_ON(bh->b_size != blocksize);
 			err = get_block(inode, block, bh, 1);
 			if (err)
 				goto recover;
+			clear_buffer_delay(bh);
 			if (buffer_new(bh)) {
 				/* blockdev mappings never come here */
 				clear_buffer_new(bh);
@@ -1774,7 +1776,8 @@ recover:
 	bh = head;
 	/* Recovery: lock and submit the mapped buffers */
 	do {
-		if (buffer_mapped(bh) && buffer_dirty(bh)) {
+		if (buffer_mapped(bh) && buffer_dirty(bh) &&
+		    !buffer_delay(bh)) {
 			lock_buffer(bh);
 			mark_buffer_async_write(bh);
 		} else {
diff --git a/fs/mpage.c b/fs/mpage.c
index 235e4d3873a8..dbcc7af76a15 100644
--- a/fs/mpage.c
+++ b/fs/mpage.c
@@ -82,7 +82,7 @@ static void mpage_end_io_write(struct bio *bio, int err)
 	bio_put(bio);
 }
 
-static struct bio *mpage_bio_submit(int rw, struct bio *bio)
+struct bio *mpage_bio_submit(int rw, struct bio *bio)
 {
 	bio->bi_end_io = mpage_end_io_read;
 	if (rw == WRITE)
@@ -90,6 +90,7 @@ static struct bio *mpage_bio_submit(int rw, struct bio *bio)
 	submit_bio(rw, bio);
 	return NULL;
 }
+EXPORT_SYMBOL(mpage_bio_submit);
 
 static struct bio *
 mpage_alloc(struct block_device *bdev,
@@ -435,15 +436,9 @@ EXPORT_SYMBOL(mpage_readpage);
  * written, so it can intelligently allocate a suitably-sized BIO.  For now,
  * just allocate full-size (16-page) BIOs.
  */
-struct mpage_data {
-	struct bio *bio;
-	sector_t last_block_in_bio;
-	get_block_t *get_block;
-	unsigned use_writepage;
-};
 
-static int __mpage_writepage(struct page *page, struct writeback_control *wbc,
-			     void *data)
+int __mpage_writepage(struct page *page, struct writeback_control *wbc,
+		      void *data)
 {
 	struct mpage_data *mpd = data;
 	struct bio *bio = mpd->bio;
@@ -651,6 +646,7 @@ out:
 	mpd->bio = bio;
 	return ret;
 }
+EXPORT_SYMBOL(__mpage_writepage);
 
 /**
  * mpage_writepages - walk the list of dirty pages of the given address space & writepage() all of them
diff --git a/include/linux/mpage.h b/include/linux/mpage.h
index 068a0c9946af..5c42821da2d1 100644
--- a/include/linux/mpage.h
+++ b/include/linux/mpage.h
@@ -11,11 +11,21 @@
  */
 #ifdef CONFIG_BLOCK
 
+struct mpage_data {
+	struct bio *bio;
+	sector_t last_block_in_bio;
+	get_block_t *get_block;
+	unsigned use_writepage;
+};
+
 struct writeback_control;
 
+struct bio *mpage_bio_submit(int rw, struct bio *bio);
 int mpage_readpages(struct address_space *mapping, struct list_head *pages,
 				unsigned nr_pages, get_block_t get_block);
 int mpage_readpage(struct page *page, get_block_t get_block);
+int __mpage_writepage(struct page *page, struct writeback_control *wbc,
+		      void *data);
 int mpage_writepages(struct address_space *mapping,
 		struct writeback_control *wbc, get_block_t get_block);
 int mpage_writepage(struct page *page, get_block_t *get_block,
-- 
cgit v1.2.3


From e8ced39d5e8911c662d4d69a342b9d053eaaac4e Mon Sep 17 00:00:00 2001
From: Mingming Cao <cmm@us.ibm.com>
Date: Fri, 11 Jul 2008 19:27:31 -0400
Subject: percpu_counter: new function percpu_counter_sum_and_set

Delayed allocation need to check free blocks at every write time.
percpu_counter_read_positive() is not quit accurate. delayed
allocation need a more accurate accounting, but using
percpu_counter_sum_positive() is frequently is quite expensive.

This patch added a new function to update center counter when sum
per-cpu counter, to increase the accurate rate for next
percpu_counter_read() and require less calling expensive
percpu_counter_sum().

Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/ext4/balloc.c               |  2 +-
 include/linux/percpu_counter.h | 12 +++++++++---
 lib/percpu_counter.c           |  7 ++++++-
 3 files changed, 16 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 25f63d8c1b3d..6369bacf0dcb 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -1621,7 +1621,7 @@ ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi,
 #ifdef CONFIG_SMP
 	if (free_blocks - root_blocks < FBC_BATCH)
 		free_blocks =
-			percpu_counter_sum_positive(&sbi->s_freeblocks_counter);
+			percpu_counter_sum_and_set(&sbi->s_freeblocks_counter);
 #endif
 	if (free_blocks - root_blocks < nblocks)
 		return free_blocks - root_blocks;
diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h
index 9007ccdfc112..208388835357 100644
--- a/include/linux/percpu_counter.h
+++ b/include/linux/percpu_counter.h
@@ -35,7 +35,7 @@ int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount);
 void percpu_counter_destroy(struct percpu_counter *fbc);
 void percpu_counter_set(struct percpu_counter *fbc, s64 amount);
 void __percpu_counter_add(struct percpu_counter *fbc, s64 amount, s32 batch);
-s64 __percpu_counter_sum(struct percpu_counter *fbc);
+s64 __percpu_counter_sum(struct percpu_counter *fbc, int set);
 
 static inline void percpu_counter_add(struct percpu_counter *fbc, s64 amount)
 {
@@ -44,13 +44,19 @@ static inline void percpu_counter_add(struct percpu_counter *fbc, s64 amount)
 
 static inline s64 percpu_counter_sum_positive(struct percpu_counter *fbc)
 {
-	s64 ret = __percpu_counter_sum(fbc);
+	s64 ret = __percpu_counter_sum(fbc, 0);
 	return ret < 0 ? 0 : ret;
 }
 
+static inline s64 percpu_counter_sum_and_set(struct percpu_counter *fbc)
+{
+	return __percpu_counter_sum(fbc, 1);
+}
+
+
 static inline s64 percpu_counter_sum(struct percpu_counter *fbc)
 {
-	return __percpu_counter_sum(fbc);
+	return __percpu_counter_sum(fbc, 0);
 }
 
 static inline s64 percpu_counter_read(struct percpu_counter *fbc)
diff --git a/lib/percpu_counter.c b/lib/percpu_counter.c
index 119174494cb5..4a8ba4bf5f6f 100644
--- a/lib/percpu_counter.c
+++ b/lib/percpu_counter.c
@@ -52,7 +52,7 @@ EXPORT_SYMBOL(__percpu_counter_add);
  * Add up all the per-cpu counts, return the result.  This is a more accurate
  * but much slower version of percpu_counter_read_positive()
  */
-s64 __percpu_counter_sum(struct percpu_counter *fbc)
+s64 __percpu_counter_sum(struct percpu_counter *fbc, int set)
 {
 	s64 ret;
 	int cpu;
@@ -62,7 +62,12 @@ s64 __percpu_counter_sum(struct percpu_counter *fbc)
 	for_each_online_cpu(cpu) {
 		s32 *pcount = per_cpu_ptr(fbc->counters, cpu);
 		ret += *pcount;
+		if (set)
+			*pcount = 0;
 	}
+	if (set)
+		fbc->count = ret;
+
 	spin_unlock(&fbc->lock);
 	return ret;
 }
-- 
cgit v1.2.3


From 06d6cf6959d22037fcec598f4f954db5db3d7356 Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Fri, 11 Jul 2008 19:27:31 -0400
Subject: mm: Add range_cont mode for writeback

Filesystems like ext4 needs to start a new transaction in
the writepages for block allocation. This happens with delayed
allocation and there is limit to how many credits we can request
from the journal layer. So we call write_cache_pages multiple
times with wbc->nr_to_write set to the maximum possible value
limitted by the max journal credits available.

Add a new mode to writeback that enables us to handle this
behaviour. In the new mode we update the wbc->range_start
to point to the new offset to be written. Next call to
call to write_cache_pages will start writeout from specified
range_start offset. In the new mode we also limit writing
to the specified wbc->range_end.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 include/linux/writeback.h | 1 +
 mm/page-writeback.c       | 3 +++
 2 files changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index f462439cc288..0d8573e6b9ec 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -63,6 +63,7 @@ struct writeback_control {
 	unsigned for_writepages:1;	/* This is a writepages() call */
 	unsigned range_cyclic:1;	/* range_start is cyclic */
 	unsigned more_io:1;		/* more io to be dispatched */
+	unsigned range_cont:1;
 };
 
 /*
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 789b6adbef37..ded57d528060 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -956,6 +956,9 @@ retry:
 	}
 	if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
 		mapping->writeback_index = index;
+
+	if (wbc->range_cont)
+		wbc->range_start = index << PAGE_CACHE_SHIFT;
 	return ret;
 }
 EXPORT_SYMBOL(write_cache_pages);
-- 
cgit v1.2.3


From f11f594edba7f689af9792a5673ed59d660ad371 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Wed, 25 Jun 2008 11:22:42 -0400
Subject: [SCSI] lib: Add support for the T10 (SCSI) Data Integrity Field CRC

The SCSI Block Protocol uses this 16-bit CRC to verify the integrity
of each data sector.  crc_t10dif() is used by sd_dif.c when performing
I/O to or from disks formatted with protection information.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 include/linux/crc-t10dif.h |  8 ++++++
 lib/Kconfig                |  7 +++++
 lib/Makefile               |  1 +
 lib/crc-t10dif.c           | 67 ++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 83 insertions(+)
 create mode 100644 include/linux/crc-t10dif.h
 create mode 100644 lib/crc-t10dif.c

(limited to 'include/linux')

diff --git a/include/linux/crc-t10dif.h b/include/linux/crc-t10dif.h
new file mode 100644
index 000000000000..a9c96d865ee7
--- /dev/null
+++ b/include/linux/crc-t10dif.h
@@ -0,0 +1,8 @@
+#ifndef _LINUX_CRC_T10DIF_H
+#define _LINUX_CRC_T10DIF_H
+
+#include <linux/types.h>
+
+__u16 crc_t10dif(unsigned char const *, size_t);
+
+#endif
diff --git a/lib/Kconfig b/lib/Kconfig
index 8cc8e8722a3f..c7ad7a5b3535 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -29,6 +29,13 @@ config CRC16
 	  the kernel tree does. Such modules that use library CRC16
 	  functions require M here.
 
+config CRC_T10DIF
+	tristate "CRC calculation for the T10 Data Integrity Field"
+	help
+	  This option is only needed if a module that's not in the
+	  kernel tree needs to calculate CRC checks for use with the
+	  SCSI data integrity subsystem.
+
 config CRC_ITU_T
 	tristate "CRC ITU-T V.41 functions"
 	help
diff --git a/lib/Makefile b/lib/Makefile
index 74b0cfb1fcc3..237a8298f8cb 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -45,6 +45,7 @@ endif
 obj-$(CONFIG_BITREVERSE) += bitrev.o
 obj-$(CONFIG_CRC_CCITT)	+= crc-ccitt.o
 obj-$(CONFIG_CRC16)	+= crc16.o
+obj-$(CONFIG_CRC_T10DIF)+= crc-t10dif.o
 obj-$(CONFIG_CRC_ITU_T)	+= crc-itu-t.o
 obj-$(CONFIG_CRC32)	+= crc32.o
 obj-$(CONFIG_CRC7)	+= crc7.o
diff --git a/lib/crc-t10dif.c b/lib/crc-t10dif.c
new file mode 100644
index 000000000000..fbbd66ed86cd
--- /dev/null
+++ b/lib/crc-t10dif.c
@@ -0,0 +1,67 @@
+/*
+ * T10 Data Integrity Field CRC16 calculation
+ *
+ * Copyright (c) 2007 Oracle Corporation.  All rights reserved.
+ * Written by Martin K. Petersen <martin.petersen@oracle.com>
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2. See the file COPYING for more details.
+ */
+
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/crc-t10dif.h>
+
+/* Table generated using the following polynomium:
+ * x^16 + x^15 + x^11 + x^9 + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1
+ * gt: 0x8bb7
+ */
+static const __u16 t10_dif_crc_table[256] = {
+	0x0000, 0x8BB7, 0x9CD9, 0x176E, 0xB205, 0x39B2, 0x2EDC, 0xA56B,
+	0xEFBD, 0x640A, 0x7364, 0xF8D3, 0x5DB8, 0xD60F, 0xC161, 0x4AD6,
+	0x54CD, 0xDF7A, 0xC814, 0x43A3, 0xE6C8, 0x6D7F, 0x7A11, 0xF1A6,
+	0xBB70, 0x30C7, 0x27A9, 0xAC1E, 0x0975, 0x82C2, 0x95AC, 0x1E1B,
+	0xA99A, 0x222D, 0x3543, 0xBEF4, 0x1B9F, 0x9028, 0x8746, 0x0CF1,
+	0x4627, 0xCD90, 0xDAFE, 0x5149, 0xF422, 0x7F95, 0x68FB, 0xE34C,
+	0xFD57, 0x76E0, 0x618E, 0xEA39, 0x4F52, 0xC4E5, 0xD38B, 0x583C,
+	0x12EA, 0x995D, 0x8E33, 0x0584, 0xA0EF, 0x2B58, 0x3C36, 0xB781,
+	0xD883, 0x5334, 0x445A, 0xCFED, 0x6A86, 0xE131, 0xF65F, 0x7DE8,
+	0x373E, 0xBC89, 0xABE7, 0x2050, 0x853B, 0x0E8C, 0x19E2, 0x9255,
+	0x8C4E, 0x07F9, 0x1097, 0x9B20, 0x3E4B, 0xB5FC, 0xA292, 0x2925,
+	0x63F3, 0xE844, 0xFF2A, 0x749D, 0xD1F6, 0x5A41, 0x4D2F, 0xC698,
+	0x7119, 0xFAAE, 0xEDC0, 0x6677, 0xC31C, 0x48AB, 0x5FC5, 0xD472,
+	0x9EA4, 0x1513, 0x027D, 0x89CA, 0x2CA1, 0xA716, 0xB078, 0x3BCF,
+	0x25D4, 0xAE63, 0xB90D, 0x32BA, 0x97D1, 0x1C66, 0x0B08, 0x80BF,
+	0xCA69, 0x41DE, 0x56B0, 0xDD07, 0x786C, 0xF3DB, 0xE4B5, 0x6F02,
+	0x3AB1, 0xB106, 0xA668, 0x2DDF, 0x88B4, 0x0303, 0x146D, 0x9FDA,
+	0xD50C, 0x5EBB, 0x49D5, 0xC262, 0x6709, 0xECBE, 0xFBD0, 0x7067,
+	0x6E7C, 0xE5CB, 0xF2A5, 0x7912, 0xDC79, 0x57CE, 0x40A0, 0xCB17,
+	0x81C1, 0x0A76, 0x1D18, 0x96AF, 0x33C4, 0xB873, 0xAF1D, 0x24AA,
+	0x932B, 0x189C, 0x0FF2, 0x8445, 0x212E, 0xAA99, 0xBDF7, 0x3640,
+	0x7C96, 0xF721, 0xE04F, 0x6BF8, 0xCE93, 0x4524, 0x524A, 0xD9FD,
+	0xC7E6, 0x4C51, 0x5B3F, 0xD088, 0x75E3, 0xFE54, 0xE93A, 0x628D,
+	0x285B, 0xA3EC, 0xB482, 0x3F35, 0x9A5E, 0x11E9, 0x0687, 0x8D30,
+	0xE232, 0x6985, 0x7EEB, 0xF55C, 0x5037, 0xDB80, 0xCCEE, 0x4759,
+	0x0D8F, 0x8638, 0x9156, 0x1AE1, 0xBF8A, 0x343D, 0x2353, 0xA8E4,
+	0xB6FF, 0x3D48, 0x2A26, 0xA191, 0x04FA, 0x8F4D, 0x9823, 0x1394,
+	0x5942, 0xD2F5, 0xC59B, 0x4E2C, 0xEB47, 0x60F0, 0x779E, 0xFC29,
+	0x4BA8, 0xC01F, 0xD771, 0x5CC6, 0xF9AD, 0x721A, 0x6574, 0xEEC3,
+	0xA415, 0x2FA2, 0x38CC, 0xB37B, 0x1610, 0x9DA7, 0x8AC9, 0x017E,
+	0x1F65, 0x94D2, 0x83BC, 0x080B, 0xAD60, 0x26D7, 0x31B9, 0xBA0E,
+	0xF0D8, 0x7B6F, 0x6C01, 0xE7B6, 0x42DD, 0xC96A, 0xDE04, 0x55B3
+};
+
+__u16 crc_t10dif(const unsigned char *buffer, size_t len)
+{
+	__u16 crc = 0;
+	unsigned int i;
+
+	for (i = 0 ; i < len ; i++)
+		crc = (crc << 8) ^ t10_dif_crc_table[((crc >> 8) ^ buffer[i]) & 0xff];
+
+	return crc;
+}
+EXPORT_SYMBOL(crc_t10dif);
+
+MODULE_DESCRIPTION("T10 DIF CRC calculation");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3


From d280eadc4fba0bf99fb1c3b60e8c5e007f7da02c Mon Sep 17 00:00:00 2001
From: Eric Miao <eric.miao@marvell.com>
Date: Fri, 6 Jun 2008 17:13:02 +0800
Subject: [NET] smc91x: remove "irq_flags" from "struct smc91x_platdata"

IRQ trigger type can be specified in the IRQ resource definition by
IORESOURCE_IRQ_*, we need only one way to specify this.

This also fixes the following small issue:

To allow dynamic support for multiple platforms, when those relevant
macros are not defined for one specific platform, the default case
will be:

	- SMC_DYNAMIC_BUS_CONFIG defined
	- and SMC_IRQ_FLAGS = IRQF_TRIGGER_RISING

While if "irq_flags" is missing when defining the smc91x_platdata,
usually as follows:

  static struct smc91x_platdata xxxx_smc91x_data = {
	.flags	= SMC91X_USE_XXBIT,
  };

The lp->cfg.irq_flags will always be overriden by the above structure
(due to a memcpy), thus rendering lp->cfg.irq_flags to be "0" always.
(regardless of the default SMC_IRQ_FLAGS or IORESOURCE_IRQ_* flags)

Fixes this by forcing to use IORESOURCE_IRQ_* flags if present, and
make the only user of smc91x_platdata.irq_flags (renesas/migor) to
use IORESOURCE_IRQ_*.

Signed-off-by: Eric Miao <eric.miao@marvell.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/sh/boards/renesas/migor/setup.c | 3 +--
 drivers/net/smc91x.c                 | 9 +++++----
 include/linux/smc91x.h               | 1 -
 3 files changed, 6 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/arch/sh/boards/renesas/migor/setup.c b/arch/sh/boards/renesas/migor/setup.c
index 01af44245b57..963c99322095 100644
--- a/arch/sh/boards/renesas/migor/setup.c
+++ b/arch/sh/boards/renesas/migor/setup.c
@@ -30,7 +30,6 @@
 
 static struct smc91x_platdata smc91x_info = {
 	.flags = SMC91X_USE_16BIT,
-	.irq_flags = IRQF_TRIGGER_HIGH,
 };
 
 static struct resource smc91x_eth_resources[] = {
@@ -42,7 +41,7 @@ static struct resource smc91x_eth_resources[] = {
 	},
 	[1] = {
 		.start  = 32, /* IRQ0 */
-		.flags  = IORESOURCE_IRQ,
+		.flags  = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
 	},
 };
 
diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c
index f2051b209da2..e4a6c361995a 100644
--- a/drivers/net/smc91x.c
+++ b/drivers/net/smc91x.c
@@ -2123,6 +2123,7 @@ static int smc_drv_probe(struct platform_device *pdev)
 	struct net_device *ndev;
 	struct resource *res, *ires;
 	unsigned int __iomem *addr;
+	unsigned long irq_flags = SMC_IRQ_FLAGS;
 	int ret;
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-regs");
@@ -2152,7 +2153,6 @@ static int smc_drv_probe(struct platform_device *pdev)
 	 */
 
 	lp = netdev_priv(ndev);
-	lp->cfg.irq_flags = SMC_IRQ_FLAGS;
 
 #ifdef SMC_DYNAMIC_BUS_CONFIG
 	if (pd)
@@ -2177,8 +2177,9 @@ static int smc_drv_probe(struct platform_device *pdev)
 	}
 
 	ndev->irq = ires->start;
-	if (SMC_IRQ_FLAGS == -1)
-		lp->cfg.irq_flags = ires->flags & IRQF_TRIGGER_MASK;
+
+	if (ires->flags & IRQF_TRIGGER_MASK)
+		irq_flags = ires->flags & IRQF_TRIGGER_MASK;
 
 	ret = smc_request_attrib(pdev);
 	if (ret)
@@ -2205,7 +2206,7 @@ static int smc_drv_probe(struct platform_device *pdev)
 	}
 #endif
 
-	ret = smc_probe(ndev, addr, lp->cfg.irq_flags);
+	ret = smc_probe(ndev, addr, irq_flags);
 	if (ret != 0)
 		goto out_iounmap;
 
diff --git a/include/linux/smc91x.h b/include/linux/smc91x.h
index 8e0556b8781c..fc7682f04d89 100644
--- a/include/linux/smc91x.h
+++ b/include/linux/smc91x.h
@@ -7,7 +7,6 @@
 
 struct smc91x_platdata {
 	unsigned long flags;
-	unsigned long irq_flags; /* IRQF_... */
 };
 
 #endif /* __SMC91X_H__ */
-- 
cgit v1.2.3


From c4f0e76747e80578a8f7fddd82fd0ce8127bd2f8 Mon Sep 17 00:00:00 2001
From: Eric Miao <eric.miao@marvell.com>
Date: Thu, 19 Jun 2008 17:39:03 +0800
Subject: [NET] smc91x: add SMC91X_NOWAIT flag to platform data

And also favors the usage of SMC91X_NOWAIT over the hardcoded SMC_NOWAIT
by converting "nowait" (module parameter overridable) to platform flag.

There are several possibilities:

  1. platform data present - preferred and use as is
  2. platform data absent  - use "nowait", it can be:
       a. SMC_NOWAIT if defined
       b. default to 0 if SMC_NOWAIT isn't defined
       c. overriden by module parameter

Signed-off-by: Eric Miao <eric.miao@marvell.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/net/smc91x.c   | 3 ++-
 include/linux/smc91x.h | 2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c
index 1b19022b6c7c..de7a913c487c 100644
--- a/drivers/net/smc91x.c
+++ b/drivers/net/smc91x.c
@@ -308,7 +308,7 @@ static void smc_reset(struct net_device *dev)
 	 * can't handle it then there will be no recovery except for
 	 * a hard reset or power cycle
 	 */
-	if (nowait)
+	if (lp->cfg.flags & SMC91X_NOWAIT)
 		cfg |= CONFIG_NO_WAIT;
 
 	/*
@@ -2160,6 +2160,7 @@ static int smc_drv_probe(struct platform_device *pdev)
 		lp->cfg.flags |= (SMC_CAN_USE_8BIT)  ? SMC91X_USE_8BIT  : 0;
 		lp->cfg.flags |= (SMC_CAN_USE_16BIT) ? SMC91X_USE_16BIT : 0;
 		lp->cfg.flags |= (SMC_CAN_USE_32BIT) ? SMC91X_USE_32BIT : 0;
+		lp->cfg.flags |= (nowait) ? SMC91X_NOWAIT : 0;
 	}
 
 	ndev->dma = (unsigned char)-1;
diff --git a/include/linux/smc91x.h b/include/linux/smc91x.h
index fc7682f04d89..90434db72db2 100644
--- a/include/linux/smc91x.h
+++ b/include/linux/smc91x.h
@@ -5,6 +5,8 @@
 #define SMC91X_USE_16BIT (1 << 1)
 #define SMC91X_USE_32BIT (1 << 2)
 
+#define SMC91X_NOWAIT		(1 << 3)
+
 struct smc91x_platdata {
 	unsigned long flags;
 };
-- 
cgit v1.2.3


From 159198862adad7109bb347bb30a620f67beac45f Mon Sep 17 00:00:00 2001
From: Eric Miao <eric.miao@marvell.com>
Date: Tue, 24 Jun 2008 13:38:50 +0800
Subject: [NET] smc91x: prepare for SMC_IO_SHIFT to be a platform configurable
 variable

Now one can use the following code

  #define SMC_IO_SHIFT	lp->io_shift

to make SMC_IO_SHIFT a variable. This, however, will slightly increase
the CPU overhead and have negative impact on the network performance.
The tradeoff is, this can be specified in the smc91x platform data so
that multiple boards support can be built in a single zImage.

Signed-off-by: Eric Miao <eric.miao@marvell.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/net/smc91x.c   | 57 +++++++++++++++++++++++++++-----------------------
 drivers/net/smc91x.h   |  3 +++
 include/linux/smc91x.h |  7 +++++++
 3 files changed, 41 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c
index de7a913c487c..34bfc60e8074 100644
--- a/drivers/net/smc91x.c
+++ b/drivers/net/smc91x.c
@@ -2050,9 +2050,11 @@ static int smc_enable_device(struct platform_device *pdev)
 	return 0;
 }
 
-static int smc_request_attrib(struct platform_device *pdev)
+static int smc_request_attrib(struct platform_device *pdev,
+			      struct net_device *ndev)
 {
 	struct resource * res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-attrib");
+	struct smc_local *lp = netdev_priv(ndev);
 
 	if (!res)
 		return 0;
@@ -2063,9 +2065,11 @@ static int smc_request_attrib(struct platform_device *pdev)
 	return 0;
 }
 
-static void smc_release_attrib(struct platform_device *pdev)
+static void smc_release_attrib(struct platform_device *pdev,
+			       struct net_device *ndev)
 {
 	struct resource * res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-attrib");
+	struct smc_local *lp = netdev_priv(ndev);
 
 	if (res)
 		release_mem_region(res->start, ATTRIB_SIZE);
@@ -2126,25 +2130,11 @@ static int smc_drv_probe(struct platform_device *pdev)
 	unsigned long irq_flags = SMC_IRQ_FLAGS;
 	int ret;
 
-	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-regs");
-	if (!res)
-		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		ret = -ENODEV;
-		goto out;
-	}
-
-
-	if (!request_mem_region(res->start, SMC_IO_EXTENT, CARDNAME)) {
-		ret = -EBUSY;
-		goto out;
-	}
-
 	ndev = alloc_etherdev(sizeof(struct smc_local));
 	if (!ndev) {
 		printk("%s: could not allocate device.\n", CARDNAME);
 		ret = -ENOMEM;
-		goto out_release_io;
+		goto out;
 	}
 	SET_NETDEV_DEV(ndev, &pdev->dev);
 
@@ -2154,9 +2144,10 @@ static int smc_drv_probe(struct platform_device *pdev)
 
 	lp = netdev_priv(ndev);
 
-	if (pd)
+	if (pd) {
 		memcpy(&lp->cfg, pd, sizeof(lp->cfg));
-	else {
+		lp->io_shift = SMC91X_IO_SHIFT(lp->cfg.flags);
+	} else {
 		lp->cfg.flags |= (SMC_CAN_USE_8BIT)  ? SMC91X_USE_8BIT  : 0;
 		lp->cfg.flags |= (SMC_CAN_USE_16BIT) ? SMC91X_USE_16BIT : 0;
 		lp->cfg.flags |= (SMC_CAN_USE_32BIT) ? SMC91X_USE_32BIT : 0;
@@ -2165,10 +2156,24 @@ static int smc_drv_probe(struct platform_device *pdev)
 
 	ndev->dma = (unsigned char)-1;
 
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-regs");
+	if (!res)
+		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		ret = -ENODEV;
+		goto out_free_netdev;
+	}
+
+
+	if (!request_mem_region(res->start, SMC_IO_EXTENT, CARDNAME)) {
+		ret = -EBUSY;
+		goto out_free_netdev;
+	}
+
 	ires = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
 	if (!ires) {
 		ret = -ENODEV;
-		goto out_free_netdev;
+		goto out_release_io;
 	}
 
 	ndev->irq = ires->start;
@@ -2176,9 +2181,9 @@ static int smc_drv_probe(struct platform_device *pdev)
 	if (ires->flags & IRQF_TRIGGER_MASK)
 		irq_flags = ires->flags & IRQF_TRIGGER_MASK;
 
-	ret = smc_request_attrib(pdev);
+	ret = smc_request_attrib(pdev, ndev);
 	if (ret)
-		goto out_free_netdev;
+		goto out_release_io;
 #if defined(CONFIG_SA1100_ASSABET)
 	NCR_0 |= NCR_ENET_OSC_EN;
 #endif
@@ -2213,11 +2218,11 @@ static int smc_drv_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, NULL);
 	iounmap(addr);
  out_release_attrib:
-	smc_release_attrib(pdev);
- out_free_netdev:
-	free_netdev(ndev);
+	smc_release_attrib(pdev, ndev);
  out_release_io:
 	release_mem_region(res->start, SMC_IO_EXTENT);
+ out_free_netdev:
+	free_netdev(ndev);
  out:
 	printk("%s: not found (%d).\n", CARDNAME, ret);
 
@@ -2243,7 +2248,7 @@ static int smc_drv_remove(struct platform_device *pdev)
 	iounmap(lp->base);
 
 	smc_release_datacs(pdev,ndev);
-	smc_release_attrib(pdev);
+	smc_release_attrib(pdev,ndev);
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-regs");
 	if (!res)
diff --git a/drivers/net/smc91x.h b/drivers/net/smc91x.h
index 8310f1a073d8..80fb80f39200 100644
--- a/drivers/net/smc91x.h
+++ b/drivers/net/smc91x.h
@@ -500,6 +500,9 @@ struct smc_local {
 	void __iomem *base;
 	void __iomem *datacs;
 
+	/* the low address lines on some platforms aren't connected... */
+	int	io_shift;
+
 	struct smc91x_platdata cfg;
 };
 
diff --git a/include/linux/smc91x.h b/include/linux/smc91x.h
index 90434db72db2..0dea9459a8e4 100644
--- a/include/linux/smc91x.h
+++ b/include/linux/smc91x.h
@@ -7,6 +7,13 @@
 
 #define SMC91X_NOWAIT		(1 << 3)
 
+/* two bits for IO_SHIFT, let's hope later designs will keep this sane */
+#define SMC91X_IO_SHIFT_0	(0 << 4)
+#define SMC91X_IO_SHIFT_1	(1 << 4)
+#define SMC91X_IO_SHIFT_2	(2 << 4)
+#define SMC91X_IO_SHIFT_3	(3 << 4)
+#define SMC91X_IO_SHIFT(x)	(((x) >> 4) & 0x3)
+
 struct smc91x_platdata {
 	unsigned long flags;
 };
-- 
cgit v1.2.3


From 52256c0e06e4a4df67134b951a21b50c713a9588 Mon Sep 17 00:00:00 2001
From: Eric Miao <eric.miao@marvell.com>
Date: Tue, 24 Jun 2008 15:36:05 +0800
Subject: [NET] smc91x: prepare SMC_USE_PXA_DMA to be specified in platform
 data

Now that the original SMC_USE_PXA_DMA specific code will always being
built if CONFIG_ARCH_PXA is defined, so to make this part of the code
to be PXA public, and still prevent it from being built if support of
PXA is not selected.

A SMC91X_USE_DMA flag is added to the platform data to allow platform
to choose its usage of DMA. Note this flag itself is so named to be
generic enough (assuming other platforms can also use DMA).

It keeps backward compatibility to set the SMC91X_USE_DMA flag if
SMC_USE_PXA_DMA is still defined.

Signed-off-by: Eric Miao <eric.miao@marvell.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Acked-by: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/net/smc91x.c   | 13 ++++++++-----
 drivers/net/smc91x.h   |  6 +++---
 include/linux/smc91x.h |  2 ++
 3 files changed, 13 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c
index 34bfc60e8074..2040965d7724 100644
--- a/drivers/net/smc91x.c
+++ b/drivers/net/smc91x.c
@@ -1939,8 +1939,11 @@ static int __init smc_probe(struct net_device *dev, void __iomem *ioaddr,
       	if (retval)
       		goto err_out;
 
-#ifdef SMC_USE_PXA_DMA
-	{
+#ifdef CONFIG_ARCH_PXA
+#  ifdef SMC_USE_PXA_DMA
+	lp->cfg.flags |= SMC91X_USE_DMA;
+#  endif
+	if (lp->cfg.flags & SMC91X_USE_DMA) {
 		int dma = pxa_request_dma(dev->name, DMA_PRIO_LOW,
 					  smc_pxa_dma_irq, NULL);
 		if (dma >= 0)
@@ -1980,7 +1983,7 @@ static int __init smc_probe(struct net_device *dev, void __iomem *ioaddr,
 	}
 
 err_out:
-#ifdef SMC_USE_PXA_DMA
+#ifdef CONFIG_ARCH_PXA
 	if (retval && dev->dma != (unsigned char)-1)
 		pxa_free_dma(dev->dma);
 #endif
@@ -2198,7 +2201,7 @@ static int smc_drv_probe(struct platform_device *pdev)
 		goto out_release_attrib;
 	}
 
-#ifdef SMC_USE_PXA_DMA
+#ifdef CONFIG_ARCH_PXA
 	{
 		struct smc_local *lp = netdev_priv(ndev);
 		lp->device = &pdev->dev;
@@ -2241,7 +2244,7 @@ static int smc_drv_remove(struct platform_device *pdev)
 
 	free_irq(ndev->irq, ndev);
 
-#ifdef SMC_USE_PXA_DMA
+#ifdef CONFIG_ARCH_PXA
 	if (ndev->dma != (unsigned char)-1)
 		pxa_free_dma(ndev->dma);
 #endif
diff --git a/drivers/net/smc91x.h b/drivers/net/smc91x.h
index 80fb80f39200..f02cc6ac248b 100644
--- a/drivers/net/smc91x.h
+++ b/drivers/net/smc91x.h
@@ -492,7 +492,7 @@ struct smc_local {
 
 	spinlock_t lock;
 
-#ifdef SMC_USE_PXA_DMA
+#ifdef CONFIG_ARCH_PXA
 	/* DMA needs the physical address of the chip */
 	u_long physaddr;
 	struct device *device;
@@ -510,7 +510,7 @@ struct smc_local {
 #define SMC_16BIT(p)	((p)->cfg.flags & SMC91X_USE_16BIT)
 #define SMC_32BIT(p)	((p)->cfg.flags & SMC91X_USE_32BIT)
 
-#ifdef SMC_USE_PXA_DMA
+#ifdef CONFIG_ARCH_PXA
 /*
  * Let's use the DMA engine on the XScale PXA2xx for RX packets. This is
  * always happening in irq context so no need to worry about races.  TX is
@@ -604,7 +604,7 @@ smc_pxa_dma_irq(int dma, void *dummy)
 {
 	DCSR(dma) = 0;
 }
-#endif  /* SMC_USE_PXA_DMA */
+#endif  /* CONFIG_ARCH_PXA */
 
 
 /*
diff --git a/include/linux/smc91x.h b/include/linux/smc91x.h
index 0dea9459a8e4..3827b922ba1f 100644
--- a/include/linux/smc91x.h
+++ b/include/linux/smc91x.h
@@ -14,6 +14,8 @@
 #define SMC91X_IO_SHIFT_3	(3 << 4)
 #define SMC91X_IO_SHIFT(x)	(((x) >> 4) & 0x3)
 
+#define SMC91X_USE_DMA		(1 << 6)
+
 struct smc91x_platdata {
 	unsigned long flags;
 };
-- 
cgit v1.2.3


From 5b5d0e704880addfd979c262e6441f126708539c Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Sun, 13 Jul 2008 11:51:40 +0100
Subject: dccp: Upgrade NDP count from 3 to 6 bytes

RFC 4340, 7.7 specifies up to 6 bytes for the NDP Count option, whereas the code
is currently limited to up to 3 bytes. This seems to be a relict of an earlier
draft version and is brought up to date by the patch.

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
---
 include/linux/dccp.h                |  6 ++----
 net/dccp/ccids/ccid3.c              |  2 +-
 net/dccp/ccids/lib/packet_history.c |  6 +++---
 net/dccp/ccids/lib/packet_history.h |  6 +++---
 net/dccp/options.c                  | 14 ++++++++------
 5 files changed, 17 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index aa0737019e37..6080449fbec9 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -364,8 +364,6 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb)
 /* FIXME: for now we're default to 1 but it should really be 0 */
 #define DCCPF_INITIAL_SEND_NDP_COUNT		1
 
-#define DCCP_NDP_LIMIT 0xFFFFFF
-
 /**
   * struct dccp_minisock - Minimal DCCP connection representation
   *
@@ -437,7 +435,7 @@ extern int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
 			      struct sk_buff *skb);
 
 struct dccp_options_received {
-	u32	dccpor_ndp; /* only 24 bits */
+	u64	dccpor_ndp:48;
 	u32	dccpor_timestamp;
 	u32	dccpor_timestamp_echo;
 	u32	dccpor_elapsed_time;
@@ -533,7 +531,7 @@ struct dccp_sock {
 	__u16				dccps_r_ack_ratio;
 	__u16				dccps_pcslen;
 	__u16				dccps_pcrlen;
-	unsigned long			dccps_ndp_count;
+	__u64				dccps_ndp_count:48;
 	unsigned long			dccps_rate_last;
 	struct dccp_minisock		dccps_minisock;
 	struct dccp_ackvec		*dccps_hc_rx_ackvec;
diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c
index a1929f33d703..523db262c18f 100644
--- a/net/dccp/ccids/ccid3.c
+++ b/net/dccp/ccids/ccid3.c
@@ -794,7 +794,7 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct ccid3_hc_rx_sock *hcrx = ccid3_hc_rx_sk(sk);
 	enum ccid3_fback_type do_feedback = CCID3_FBACK_NONE;
-	const u32 ndp = dccp_sk(sk)->dccps_options_received.dccpor_ndp;
+	const u64 ndp = dccp_sk(sk)->dccps_options_received.dccpor_ndp;
 	const bool is_data_packet = dccp_data_packet(skb);
 
 	if (unlikely(hcrx->ccid3hcrx_state == TFRC_RSTATE_NO_DATA)) {
diff --git a/net/dccp/ccids/lib/packet_history.c b/net/dccp/ccids/lib/packet_history.c
index 8b5c41ec7ee1..712930564c38 100644
--- a/net/dccp/ccids/lib/packet_history.c
+++ b/net/dccp/ccids/lib/packet_history.c
@@ -153,7 +153,7 @@ void tfrc_rx_packet_history_exit(void)
 
 static inline void tfrc_rx_hist_entry_from_skb(struct tfrc_rx_hist_entry *entry,
 					       const struct sk_buff *skb,
-					       const u32 ndp)
+					       const u64 ndp)
 {
 	const struct dccp_hdr *dh = dccp_hdr(skb);
 
@@ -166,7 +166,7 @@ static inline void tfrc_rx_hist_entry_from_skb(struct tfrc_rx_hist_entry *entry,
 
 void tfrc_rx_hist_add_packet(struct tfrc_rx_hist *h,
 			     const struct sk_buff *skb,
-			     const u32 ndp)
+			     const u64 ndp)
 {
 	struct tfrc_rx_hist_entry *entry = tfrc_rx_hist_last_rcv(h);
 
@@ -356,7 +356,7 @@ static void __three_after_loss(struct tfrc_rx_hist *h)
  */
 int tfrc_rx_handle_loss(struct tfrc_rx_hist *h,
 			struct tfrc_loss_hist *lh,
-			struct sk_buff *skb, u32 ndp,
+			struct sk_buff *skb, const u64 ndp,
 			u32 (*calc_first_li)(struct sock *), struct sock *sk)
 {
 	int is_new_loss = 0;
diff --git a/net/dccp/ccids/lib/packet_history.h b/net/dccp/ccids/lib/packet_history.h
index c7eeda49cb20..6976156cda3c 100644
--- a/net/dccp/ccids/lib/packet_history.h
+++ b/net/dccp/ccids/lib/packet_history.h
@@ -64,7 +64,7 @@ struct tfrc_rx_hist_entry {
 	u64		 tfrchrx_seqno:48,
 			 tfrchrx_ccval:4,
 			 tfrchrx_type:4;
-	u32		 tfrchrx_ndp; /* In fact it is from 8 to 24 bits */
+	u64		 tfrchrx_ndp:48;
 	ktime_t		 tfrchrx_tstamp;
 };
 
@@ -145,14 +145,14 @@ static inline int tfrc_rx_hist_new_loss_indicated(struct tfrc_rx_hist *h,
 }
 
 extern void tfrc_rx_hist_add_packet(struct tfrc_rx_hist *h,
-				    const struct sk_buff *skb, const u32 ndp);
+				    const struct sk_buff *skb, const u64 ndp);
 
 extern int tfrc_rx_hist_duplicate(struct tfrc_rx_hist *h, struct sk_buff *skb);
 
 struct tfrc_loss_hist;
 extern int  tfrc_rx_handle_loss(struct tfrc_rx_hist *h,
 				struct tfrc_loss_hist *lh,
-				struct sk_buff *skb, u32 ndp,
+				struct sk_buff *skb, const u64 ndp,
 				u32 (*first_li)(struct sock *sk),
 				struct sock *sk);
 extern u32 tfrc_rx_hist_sample_rtt(struct tfrc_rx_hist *h,
diff --git a/net/dccp/options.c b/net/dccp/options.c
index 43bc24e761d0..dc7c158a2f4b 100644
--- a/net/dccp/options.c
+++ b/net/dccp/options.c
@@ -124,12 +124,12 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
 				mandatory = 1;
 			break;
 		case DCCPO_NDP_COUNT:
-			if (len > 3)
+			if (len > 6)
 				goto out_invalid_option;
 
 			opt_recv->dccpor_ndp = dccp_decode_value_var(value, len);
-			dccp_pr_debug("%s rx opt: NDP count=%d\n", dccp_role(sk),
-				      opt_recv->dccpor_ndp);
+			dccp_pr_debug("%s opt: NDP count=%llu\n", dccp_role(sk),
+				      (unsigned long long)opt_recv->dccpor_ndp);
 			break;
 		case DCCPO_CHANGE_L:
 			/* fall through */
@@ -307,9 +307,11 @@ static void dccp_encode_value_var(const u32 value, unsigned char *to,
 		*to++ = (value & 0xFF);
 }
 
-static inline int dccp_ndp_len(const int ndp)
+static inline u8 dccp_ndp_len(const u64 ndp)
 {
-	return likely(ndp <= 0xFF) ? 1 : ndp <= 0xFFFF ? 2 : 3;
+	if (likely(ndp <= 0xFF))
+		return 1;
+	return likely(ndp <= USHORT_MAX) ? 2 : (ndp <= UINT_MAX ? 4 : 6);
 }
 
 int dccp_insert_option(struct sock *sk, struct sk_buff *skb,
@@ -336,7 +338,7 @@ EXPORT_SYMBOL_GPL(dccp_insert_option);
 static int dccp_insert_option_ndp(struct sock *sk, struct sk_buff *skb)
 {
 	struct dccp_sock *dp = dccp_sk(sk);
-	int ndp = dp->dccps_ndp_count;
+	u64 ndp = dp->dccps_ndp_count;
 
 	if (dccp_non_data_packet(skb))
 		++dp->dccps_ndp_count;
-- 
cgit v1.2.3


From afc1246f917c664b0df98b3c22fa62db74d2ca33 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Fri, 11 Jul 2008 17:20:49 -0700
Subject: file lock: reorder struct file_lock to save space on 64 bit builds

Reduce sizeof struct file_lock by 8 on 64 bit builds allowing +1 objects
per slab in the file_lock_cache

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/fs.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7c1080826832..87f89bd0f6ee 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -917,12 +917,12 @@ struct file_lock {
 	struct list_head fl_link;	/* doubly linked list of all locks */
 	struct list_head fl_block;	/* circular list of blocked processes */
 	fl_owner_t fl_owner;
+	unsigned char fl_flags;
+	unsigned char fl_type;
 	unsigned int fl_pid;
 	struct pid *fl_nspid;
 	wait_queue_head_t fl_wait;
 	struct file *fl_file;
-	unsigned char fl_flags;
-	unsigned char fl_type;
 	loff_t fl_start;
 	loff_t fl_end;
 
-- 
cgit v1.2.3


From 006ebb40d3d65338bd74abb03b945f8d60e362bd Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 19 May 2008 08:32:49 -0400
Subject: Security: split proc ptrace checking into read vs. attach

Enable security modules to distinguish reading of process state via
proc from full ptrace access by renaming ptrace_may_attach to
ptrace_may_access and adding a mode argument indicating whether only
read access or full attach access is requested.  This allows security
modules to permit access to reading process state without granting
full ptrace access.  The base DAC/capability checking remains unchanged.

Read access to /proc/pid/mem continues to apply a full ptrace attach
check since check_mem_permission() already requires the current task
to already be ptracing the target.  The other ptrace checks within
proc for elements like environ, maps, and fds are changed to pass the
read mode instead of attach.

In the SELinux case, we model such reading of process state as a
reading of a proc file labeled with the target process' label.  This
enables SELinux policy to permit such reading of process state without
permitting control or manipulation of the target process, as there are
a number of cases where programs probe for such information via proc
but do not need to be able to control the target (e.g. procps,
lsof, PolicyKit, ConsoleKit).  At present we have to choose between
allowing full ptrace in policy (more permissive than required/desired)
or breaking functionality (or in some cases just silencing the denials
via dontaudit rules but this can hide genuine attacks).

This version of the patch incorporates comments from Casey Schaufler
(change/replace existing ptrace_may_attach interface, pass access
mode), and Chris Wright (provide greater consistency in the checking).

Note that like their predecessors __ptrace_may_attach and
ptrace_may_attach, the __ptrace_may_access and ptrace_may_access
interfaces use different return value conventions from each other (0
or -errno vs. 1 or 0).  I retained this difference to avoid any
changes to the caller logic but made the difference clearer by
changing the latter interface to return a bool rather than an int and
by adding a comment about it to ptrace.h for any future callers.

Signed-off-by:  Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: James Morris <jmorris@namei.org>
---
 fs/proc/base.c             |  9 +++++----
 fs/proc/task_mmu.c         |  6 +++---
 fs/proc/task_nommu.c       |  2 +-
 include/linux/ptrace.h     |  8 ++++++--
 include/linux/security.h   | 16 +++++++++++-----
 kernel/ptrace.c            | 15 ++++++++-------
 security/commoncap.c       |  3 ++-
 security/dummy.c           |  3 ++-
 security/security.c        |  5 +++--
 security/selinux/hooks.c   | 13 +++++++++++--
 security/smack/smack_lsm.c |  5 +++--
 11 files changed, 55 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/fs/proc/base.c b/fs/proc/base.c
index 3b455371e7ff..58c3e6a8e15e 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -233,7 +233,7 @@ static int check_mem_permission(struct task_struct *task)
 	 */
 	if (task->parent == current && (task->ptrace & PT_PTRACED) &&
 	    task_is_stopped_or_traced(task) &&
-	    ptrace_may_attach(task))
+	    ptrace_may_access(task, PTRACE_MODE_ATTACH))
 		return 0;
 
 	/*
@@ -251,7 +251,8 @@ struct mm_struct *mm_for_maps(struct task_struct *task)
 	task_lock(task);
 	if (task->mm != mm)
 		goto out;
-	if (task->mm != current->mm && __ptrace_may_attach(task) < 0)
+	if (task->mm != current->mm &&
+	    __ptrace_may_access(task, PTRACE_MODE_READ) < 0)
 		goto out;
 	task_unlock(task);
 	return mm;
@@ -518,7 +519,7 @@ static int proc_fd_access_allowed(struct inode *inode)
 	 */
 	task = get_proc_task(inode);
 	if (task) {
-		allowed = ptrace_may_attach(task);
+		allowed = ptrace_may_access(task, PTRACE_MODE_READ);
 		put_task_struct(task);
 	}
 	return allowed;
@@ -904,7 +905,7 @@ static ssize_t environ_read(struct file *file, char __user *buf,
 	if (!task)
 		goto out_no_task;
 
-	if (!ptrace_may_attach(task))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ))
 		goto out;
 
 	ret = -ENOMEM;
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index c492449f3b45..164bd9f9ede3 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -210,7 +210,7 @@ static int show_map(struct seq_file *m, void *v)
 	dev_t dev = 0;
 	int len;
 
-	if (maps_protect && !ptrace_may_attach(task))
+	if (maps_protect && !ptrace_may_access(task, PTRACE_MODE_READ))
 		return -EACCES;
 
 	if (file) {
@@ -646,7 +646,7 @@ static ssize_t pagemap_read(struct file *file, char __user *buf,
 		goto out;
 
 	ret = -EACCES;
-	if (!ptrace_may_attach(task))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ))
 		goto out_task;
 
 	ret = -EINVAL;
@@ -747,7 +747,7 @@ static int show_numa_map_checked(struct seq_file *m, void *v)
 	struct proc_maps_private *priv = m->private;
 	struct task_struct *task = priv->task;
 
-	if (maps_protect && !ptrace_may_attach(task))
+	if (maps_protect && !ptrace_may_access(task, PTRACE_MODE_READ))
 		return -EACCES;
 
 	return show_numa_map(m, v);
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 4b4f9cc2f186..5d84e7121df8 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -113,7 +113,7 @@ static int show_map(struct seq_file *m, void *_vml)
 	struct proc_maps_private *priv = m->private;
 	struct task_struct *task = priv->task;
 
-	if (maps_protect && !ptrace_may_attach(task))
+	if (maps_protect && !ptrace_may_access(task, PTRACE_MODE_READ))
 		return -EACCES;
 
 	return nommu_vma_show(m, vml->vma);
diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index f98501ba557e..c6f5f9dd0cee 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -95,8 +95,12 @@ extern void __ptrace_link(struct task_struct *child,
 			  struct task_struct *new_parent);
 extern void __ptrace_unlink(struct task_struct *child);
 extern void ptrace_untrace(struct task_struct *child);
-extern int ptrace_may_attach(struct task_struct *task);
-extern int __ptrace_may_attach(struct task_struct *task);
+#define PTRACE_MODE_READ   1
+#define PTRACE_MODE_ATTACH 2
+/* Returns 0 on success, -errno on denial. */
+extern int __ptrace_may_access(struct task_struct *task, unsigned int mode);
+/* Returns true on success, false on denial. */
+extern bool ptrace_may_access(struct task_struct *task, unsigned int mode);
 
 static inline int ptrace_reparented(struct task_struct *child)
 {
diff --git a/include/linux/security.h b/include/linux/security.h
index 50737c70e78e..62bd80cb7f87 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -46,7 +46,8 @@ struct audit_krule;
  */
 extern int cap_capable(struct task_struct *tsk, int cap);
 extern int cap_settime(struct timespec *ts, struct timezone *tz);
-extern int cap_ptrace(struct task_struct *parent, struct task_struct *child);
+extern int cap_ptrace(struct task_struct *parent, struct task_struct *child,
+		      unsigned int mode);
 extern int cap_capget(struct task_struct *target, kernel_cap_t *effective, kernel_cap_t *inheritable, kernel_cap_t *permitted);
 extern int cap_capset_check(struct task_struct *target, kernel_cap_t *effective, kernel_cap_t *inheritable, kernel_cap_t *permitted);
 extern void cap_capset_set(struct task_struct *target, kernel_cap_t *effective, kernel_cap_t *inheritable, kernel_cap_t *permitted);
@@ -1170,6 +1171,7 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	attributes would be changed by the execve.
  *	@parent contains the task_struct structure for parent process.
  *	@child contains the task_struct structure for child process.
+ *	@mode contains the PTRACE_MODE flags indicating the form of access.
  *	Return 0 if permission is granted.
  * @capget:
  *	Get the @effective, @inheritable, and @permitted capability sets for
@@ -1295,7 +1297,8 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
 struct security_operations {
 	char name[SECURITY_NAME_MAX + 1];
 
-	int (*ptrace) (struct task_struct *parent, struct task_struct *child);
+	int (*ptrace) (struct task_struct *parent, struct task_struct *child,
+		       unsigned int mode);
 	int (*capget) (struct task_struct *target,
 		       kernel_cap_t *effective,
 		       kernel_cap_t *inheritable, kernel_cap_t *permitted);
@@ -1573,7 +1576,8 @@ extern struct dentry *securityfs_create_dir(const char *name, struct dentry *par
 extern void securityfs_remove(struct dentry *dentry);
 
 /* Security operations */
-int security_ptrace(struct task_struct *parent, struct task_struct *child);
+int security_ptrace(struct task_struct *parent, struct task_struct *child,
+		    unsigned int mode);
 int security_capget(struct task_struct *target,
 		    kernel_cap_t *effective,
 		    kernel_cap_t *inheritable,
@@ -1755,9 +1759,11 @@ static inline int security_init(void)
 	return 0;
 }
 
-static inline int security_ptrace(struct task_struct *parent, struct task_struct *child)
+static inline int security_ptrace(struct task_struct *parent,
+				  struct task_struct *child,
+				  unsigned int mode)
 {
-	return cap_ptrace(parent, child);
+	return cap_ptrace(parent, child, mode);
 }
 
 static inline int security_capget(struct task_struct *target,
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index 6c19e94fd0a5..e337390fce01 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -121,7 +121,7 @@ int ptrace_check_attach(struct task_struct *child, int kill)
 	return ret;
 }
 
-int __ptrace_may_attach(struct task_struct *task)
+int __ptrace_may_access(struct task_struct *task, unsigned int mode)
 {
 	/* May we inspect the given task?
 	 * This check is used both for attaching with ptrace
@@ -148,16 +148,16 @@ int __ptrace_may_attach(struct task_struct *task)
 	if (!dumpable && !capable(CAP_SYS_PTRACE))
 		return -EPERM;
 
-	return security_ptrace(current, task);
+	return security_ptrace(current, task, mode);
 }
 
-int ptrace_may_attach(struct task_struct *task)
+bool ptrace_may_access(struct task_struct *task, unsigned int mode)
 {
 	int err;
 	task_lock(task);
-	err = __ptrace_may_attach(task);
+	err = __ptrace_may_access(task, mode);
 	task_unlock(task);
-	return !err;
+	return (!err ? true : false);
 }
 
 int ptrace_attach(struct task_struct *task)
@@ -195,7 +195,7 @@ repeat:
 	/* the same process cannot be attached many times */
 	if (task->ptrace & PT_PTRACED)
 		goto bad;
-	retval = __ptrace_may_attach(task);
+	retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH);
 	if (retval)
 		goto bad;
 
@@ -494,7 +494,8 @@ int ptrace_traceme(void)
 	 */
 	task_lock(current);
 	if (!(current->ptrace & PT_PTRACED)) {
-		ret = security_ptrace(current->parent, current);
+		ret = security_ptrace(current->parent, current,
+				      PTRACE_MODE_ATTACH);
 		/*
 		 * Set the ptrace bit in the process ptrace flags.
 		 */
diff --git a/security/commoncap.c b/security/commoncap.c
index 33d343308413..0b6537a3672d 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -63,7 +63,8 @@ int cap_settime(struct timespec *ts, struct timezone *tz)
 	return 0;
 }
 
-int cap_ptrace (struct task_struct *parent, struct task_struct *child)
+int cap_ptrace (struct task_struct *parent, struct task_struct *child,
+		unsigned int mode)
 {
 	/* Derived from arch/i386/kernel/ptrace.c:sys_ptrace. */
 	if (!cap_issubset(child->cap_permitted, parent->cap_permitted) &&
diff --git a/security/dummy.c b/security/dummy.c
index b8916883b77f..1db712d99dc7 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -30,7 +30,8 @@
 #include <linux/prctl.h>
 #include <linux/securebits.h>
 
-static int dummy_ptrace (struct task_struct *parent, struct task_struct *child)
+static int dummy_ptrace (struct task_struct *parent, struct task_struct *child,
+			 unsigned int mode)
 {
 	return 0;
 }
diff --git a/security/security.c b/security/security.c
index 59838a99b80e..c4507ce2a5a0 100644
--- a/security/security.c
+++ b/security/security.c
@@ -161,9 +161,10 @@ int mod_reg_security(const char *name, struct security_operations *ops)
 
 /* Security operations */
 
-int security_ptrace(struct task_struct *parent, struct task_struct *child)
+int security_ptrace(struct task_struct *parent, struct task_struct *child,
+		    unsigned int mode)
 {
-	return security_ops->ptrace(parent, child);
+	return security_ops->ptrace(parent, child, mode);
 }
 
 int security_capget(struct task_struct *target,
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index eca70f42e678..4be156334b22 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -1686,14 +1686,23 @@ static inline u32 file_to_av(struct file *file)
 
 /* Hook functions begin here. */
 
-static int selinux_ptrace(struct task_struct *parent, struct task_struct *child)
+static int selinux_ptrace(struct task_struct *parent,
+			  struct task_struct *child,
+			  unsigned int mode)
 {
 	int rc;
 
-	rc = secondary_ops->ptrace(parent, child);
+	rc = secondary_ops->ptrace(parent, child, mode);
 	if (rc)
 		return rc;
 
+	if (mode == PTRACE_MODE_READ) {
+		struct task_security_struct *tsec = parent->security;
+		struct task_security_struct *csec = child->security;
+		return avc_has_perm(tsec->sid, csec->sid,
+				    SECCLASS_FILE, FILE__READ, NULL);
+	}
+
 	return task_has_perm(parent, child, PROCESS__PTRACE);
 }
 
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 4a09293efa00..3c7150b3493d 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -95,11 +95,12 @@ struct inode_smack *new_inode_smack(char *smack)
  *
  * Do the capability checks, and require read and write.
  */
-static int smack_ptrace(struct task_struct *ptp, struct task_struct *ctp)
+static int smack_ptrace(struct task_struct *ptp, struct task_struct *ctp,
+			unsigned int mode)
 {
 	int rc;
 
-	rc = cap_ptrace(ptp, ctp);
+	rc = cap_ptrace(ptp, ctp, mode);
 	if (rc != 0)
 		return rc;
 
-- 
cgit v1.2.3


From 2069f457848f846cb31149c9aa29b330a6b66d1b Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Fri, 4 Jul 2008 09:47:13 +1000
Subject: LSM/SELinux: show LSM mount options in /proc/mounts

This patch causes SELinux mount options to show up in /proc/mounts.  As
with other code in the area seq_put errors are ignored.  Other LSM's
will not have their mount options displayed until they fill in their own
security_sb_show_options() function.

Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: James Morris <jmorris@namei.org>
---
 fs/namespace.c           | 14 +++++++++---
 include/linux/security.h |  9 ++++++++
 security/dummy.c         |  6 ++++++
 security/security.c      |  5 +++++
 security/selinux/hooks.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++-
 5 files changed, 85 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/fs/namespace.c b/fs/namespace.c
index 4fc302c2a0e0..4f6f7635b59c 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -750,7 +750,7 @@ struct proc_fs_info {
 	const char *str;
 };
 
-static void show_sb_opts(struct seq_file *m, struct super_block *sb)
+static int show_sb_opts(struct seq_file *m, struct super_block *sb)
 {
 	static const struct proc_fs_info fs_info[] = {
 		{ MS_SYNCHRONOUS, ",sync" },
@@ -764,6 +764,8 @@ static void show_sb_opts(struct seq_file *m, struct super_block *sb)
 		if (sb->s_flags & fs_infop->flag)
 			seq_puts(m, fs_infop->str);
 	}
+
+	return security_sb_show_options(m, sb);
 }
 
 static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt)
@@ -806,11 +808,14 @@ static int show_vfsmnt(struct seq_file *m, void *v)
 	seq_putc(m, ' ');
 	show_type(m, mnt->mnt_sb);
 	seq_puts(m, __mnt_is_readonly(mnt) ? " ro" : " rw");
-	show_sb_opts(m, mnt->mnt_sb);
+	err = show_sb_opts(m, mnt->mnt_sb);
+	if (err)
+		goto out;
 	show_mnt_opts(m, mnt);
 	if (mnt->mnt_sb->s_op->show_options)
 		err = mnt->mnt_sb->s_op->show_options(m, mnt);
 	seq_puts(m, " 0 0\n");
+out:
 	return err;
 }
 
@@ -865,10 +870,13 @@ static int show_mountinfo(struct seq_file *m, void *v)
 	seq_putc(m, ' ');
 	mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none");
 	seq_puts(m, sb->s_flags & MS_RDONLY ? " ro" : " rw");
-	show_sb_opts(m, sb);
+	err = show_sb_opts(m, sb);
+	if (err)
+		goto out;
 	if (sb->s_op->show_options)
 		err = sb->s_op->show_options(m, mnt);
 	seq_putc(m, '\n');
+out:
 	return err;
 }
 
diff --git a/include/linux/security.h b/include/linux/security.h
index 62bd80cb7f87..c8ad8ec684b4 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -80,6 +80,7 @@ struct xfrm_selector;
 struct xfrm_policy;
 struct xfrm_state;
 struct xfrm_user_sec_ctx;
+struct seq_file;
 
 extern int cap_netlink_send(struct sock *sk, struct sk_buff *skb);
 extern int cap_netlink_recv(struct sk_buff *skb, int cap);
@@ -1331,6 +1332,7 @@ struct security_operations {
 	void (*sb_free_security) (struct super_block *sb);
 	int (*sb_copy_data) (char *orig, char *copy);
 	int (*sb_kern_mount) (struct super_block *sb, void *data);
+	int (*sb_show_options) (struct seq_file *m, struct super_block *sb);
 	int (*sb_statfs) (struct dentry *dentry);
 	int (*sb_mount) (char *dev_name, struct path *path,
 			 char *type, unsigned long flags, void *data);
@@ -1610,6 +1612,7 @@ int security_sb_alloc(struct super_block *sb);
 void security_sb_free(struct super_block *sb);
 int security_sb_copy_data(char *orig, char *copy);
 int security_sb_kern_mount(struct super_block *sb, void *data);
+int security_sb_show_options(struct seq_file *m, struct super_block *sb);
 int security_sb_statfs(struct dentry *dentry);
 int security_sb_mount(char *dev_name, struct path *path,
 		      char *type, unsigned long flags, void *data);
@@ -1887,6 +1890,12 @@ static inline int security_sb_kern_mount(struct super_block *sb, void *data)
 	return 0;
 }
 
+static inline int security_sb_show_options(struct seq_file *m,
+					   struct super_block *sb)
+{
+	return 0;
+}
+
 static inline int security_sb_statfs(struct dentry *dentry)
 {
 	return 0;
diff --git a/security/dummy.c b/security/dummy.c
index 1db712d99dc7..c155f08e9dd8 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -194,6 +194,11 @@ static int dummy_sb_kern_mount (struct super_block *sb, void *data)
 	return 0;
 }
 
+static int dummy_sb_show_options(struct seq_file *m, struct super_block *sb)
+{
+	return 0;
+}
+
 static int dummy_sb_statfs (struct dentry *dentry)
 {
 	return 0;
@@ -1088,6 +1093,7 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, sb_free_security);
 	set_to_dummy_if_null(ops, sb_copy_data);
 	set_to_dummy_if_null(ops, sb_kern_mount);
+	set_to_dummy_if_null(ops, sb_show_options);
 	set_to_dummy_if_null(ops, sb_statfs);
 	set_to_dummy_if_null(ops, sb_mount);
 	set_to_dummy_if_null(ops, sb_check_sb);
diff --git a/security/security.c b/security/security.c
index 2c0a5876b939..de74fdccde26 100644
--- a/security/security.c
+++ b/security/security.c
@@ -292,6 +292,11 @@ int security_sb_kern_mount(struct super_block *sb, void *data)
 	return security_ops->sb_kern_mount(sb, data);
 }
 
+int security_sb_show_options(struct seq_file *m, struct super_block *sb)
+{
+	return security_ops->sb_show_options(m, sb);
+}
+
 int security_sb_statfs(struct dentry *dentry)
 {
 	return security_ops->sb_statfs(dentry);
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 85f74f665765..33dee83fdd2f 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -9,7 +9,8 @@
  *	      James Morris <jmorris@redhat.com>
  *
  *  Copyright (C) 2001,2002 Networks Associates Technology, Inc.
- *  Copyright (C) 2003 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ *  Copyright (C) 2003-2008 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ *					   Eric Paris <eparis@redhat.com>
  *  Copyright (C) 2004-2005 Trusted Computer Solutions, Inc.
  *			    <dgoeddel@trustedcs.com>
  *  Copyright (C) 2006, 2007 Hewlett-Packard Development Company, L.P.
@@ -970,6 +971,57 @@ out_err:
 	return rc;
 }
 
+void selinux_write_opts(struct seq_file *m, struct security_mnt_opts *opts)
+{
+	int i;
+	char *prefix;
+
+	for (i = 0; i < opts->num_mnt_opts; i++) {
+		char *has_comma = strchr(opts->mnt_opts[i], ',');
+
+		switch (opts->mnt_opts_flags[i]) {
+		case CONTEXT_MNT:
+			prefix = CONTEXT_STR;
+			break;
+		case FSCONTEXT_MNT:
+			prefix = FSCONTEXT_STR;
+			break;
+		case ROOTCONTEXT_MNT:
+			prefix = ROOTCONTEXT_STR;
+			break;
+		case DEFCONTEXT_MNT:
+			prefix = DEFCONTEXT_STR;
+			break;
+		default:
+			BUG();
+		};
+		/* we need a comma before each option */
+		seq_putc(m, ',');
+		seq_puts(m, prefix);
+		if (has_comma)
+			seq_putc(m, '\"');
+		seq_puts(m, opts->mnt_opts[i]);
+		if (has_comma)
+			seq_putc(m, '\"');
+	}
+}
+
+static int selinux_sb_show_options(struct seq_file *m, struct super_block *sb)
+{
+	struct security_mnt_opts opts;
+	int rc;
+
+	rc = selinux_get_mnt_opts(sb, &opts);
+	if (rc)
+		return rc;
+
+	selinux_write_opts(m, &opts);
+
+	security_free_mnt_opts(&opts);
+
+	return rc;
+}
+
 static inline u16 inode_mode_to_security_class(umode_t mode)
 {
 	switch (mode & S_IFMT) {
@@ -5365,6 +5417,7 @@ static struct security_operations selinux_ops = {
 	.sb_free_security =		selinux_sb_free_security,
 	.sb_copy_data =			selinux_sb_copy_data,
 	.sb_kern_mount =		selinux_sb_kern_mount,
+	.sb_show_options =		selinux_sb_show_options,
 	.sb_statfs =			selinux_sb_statfs,
 	.sb_mount =			selinux_mount,
 	.sb_umount =			selinux_umount,
-- 
cgit v1.2.3


From b478a9f9889c81e88077d1495daadee64c0af541 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Thu, 3 Jul 2008 20:56:04 +0200
Subject: security: remove unused sb_get_mnt_opts hook

The sb_get_mnt_opts() hook is unused, and is superseded by the
sb_show_options() hook.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Acked-by: James Morris <jmorris@namei.org>
---
 include/linux/security.h | 14 --------------
 security/dummy.c         |  8 --------
 security/security.c      |  6 ------
 security/selinux/hooks.c |  1 -
 4 files changed, 29 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/security.h b/include/linux/security.h
index c8ad8ec684b4..43c6357568a3 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -291,10 +291,6 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	Update module state after a successful pivot.
  *	@old_path contains the path for the old root.
  *	@new_path contains the path for the new root.
- * @sb_get_mnt_opts:
- *	Get the security relevant mount options used for a superblock
- *	@sb the superblock to get security mount options from
- *	@opts binary data structure containing all lsm mount data
  * @sb_set_mnt_opts:
  *	Set the security relevant mount options used for a superblock
  *	@sb the superblock to set security mount options for
@@ -1348,8 +1344,6 @@ struct security_operations {
 			     struct path *new_path);
 	void (*sb_post_pivotroot) (struct path *old_path,
 				   struct path *new_path);
-	int (*sb_get_mnt_opts) (const struct super_block *sb,
-				struct security_mnt_opts *opts);
 	int (*sb_set_mnt_opts) (struct super_block *sb,
 				struct security_mnt_opts *opts);
 	void (*sb_clone_mnt_opts) (const struct super_block *oldsb,
@@ -1624,8 +1618,6 @@ void security_sb_post_remount(struct vfsmount *mnt, unsigned long flags, void *d
 void security_sb_post_addmount(struct vfsmount *mnt, struct path *mountpoint);
 int security_sb_pivotroot(struct path *old_path, struct path *new_path);
 void security_sb_post_pivotroot(struct path *old_path, struct path *new_path);
-int security_sb_get_mnt_opts(const struct super_block *sb,
-				struct security_mnt_opts *opts);
 int security_sb_set_mnt_opts(struct super_block *sb, struct security_mnt_opts *opts);
 void security_sb_clone_mnt_opts(const struct super_block *oldsb,
 				struct super_block *newsb);
@@ -1942,12 +1934,6 @@ static inline int security_sb_pivotroot(struct path *old_path,
 static inline void security_sb_post_pivotroot(struct path *old_path,
 					      struct path *new_path)
 { }
-static inline int security_sb_get_mnt_opts(const struct super_block *sb,
-					   struct security_mnt_opts *opts)
-{
-	security_init_mnt_opts(opts);
-	return 0;
-}
 
 static inline int security_sb_set_mnt_opts(struct super_block *sb,
 					   struct security_mnt_opts *opts)
diff --git a/security/dummy.c b/security/dummy.c
index c155f08e9dd8..793856691641 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -252,13 +252,6 @@ static void dummy_sb_post_pivotroot (struct path *old_path, struct path *new_pat
 	return;
 }
 
-static int dummy_sb_get_mnt_opts(const struct super_block *sb,
-				 struct security_mnt_opts *opts)
-{
-	security_init_mnt_opts(opts);
-	return 0;
-}
-
 static int dummy_sb_set_mnt_opts(struct super_block *sb,
 				 struct security_mnt_opts *opts)
 {
@@ -1104,7 +1097,6 @@ void security_fixup_ops (struct security_operations *ops)
 	set_to_dummy_if_null(ops, sb_post_addmount);
 	set_to_dummy_if_null(ops, sb_pivotroot);
 	set_to_dummy_if_null(ops, sb_post_pivotroot);
-	set_to_dummy_if_null(ops, sb_get_mnt_opts);
 	set_to_dummy_if_null(ops, sb_set_mnt_opts);
 	set_to_dummy_if_null(ops, sb_clone_mnt_opts);
 	set_to_dummy_if_null(ops, sb_parse_opts_str);
diff --git a/security/security.c b/security/security.c
index de74fdccde26..28b2860c1129 100644
--- a/security/security.c
+++ b/security/security.c
@@ -348,12 +348,6 @@ void security_sb_post_pivotroot(struct path *old_path, struct path *new_path)
 	security_ops->sb_post_pivotroot(old_path, new_path);
 }
 
-int security_sb_get_mnt_opts(const struct super_block *sb,
-				struct security_mnt_opts *opts)
-{
-	return security_ops->sb_get_mnt_opts(sb, opts);
-}
-
 int security_sb_set_mnt_opts(struct super_block *sb,
 				struct security_mnt_opts *opts)
 {
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 33dee83fdd2f..745a69e74e38 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -5421,7 +5421,6 @@ static struct security_operations selinux_ops = {
 	.sb_statfs =			selinux_sb_statfs,
 	.sb_mount =			selinux_mount,
 	.sb_umount =			selinux_umount,
-	.sb_get_mnt_opts =		selinux_get_mnt_opts,
 	.sb_set_mnt_opts =		selinux_set_mnt_opts,
 	.sb_clone_mnt_opts =		selinux_sb_clone_mnt_opts,
 	.sb_parse_opts_str = 		selinux_parse_opts_str,
-- 
cgit v1.2.3


From 6f0f0fd496333777d53daff21a4e3b28c4d03a6d Mon Sep 17 00:00:00 2001
From: James Morris <jmorris@namei.org>
Date: Thu, 10 Jul 2008 17:02:07 +0900
Subject: security: remove register_security hook

The register security hook is no longer required, as the capability
module is always registered.  LSMs wishing to stack capability as
a secondary module should do so explicitly.

Signed-off-by: James Morris <jmorris@namei.org>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/security.h   | 10 ----------
 security/capability.c      |  7 -------
 security/root_plug.c       |  9 ---------
 security/security.c        | 29 -----------------------------
 security/selinux/hooks.c   | 32 +++++---------------------------
 security/smack/smack_lsm.c | 23 -----------------------
 6 files changed, 5 insertions(+), 105 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/security.h b/include/linux/security.h
index 43c6357568a3..31c8851ec5d0 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1239,11 +1239,6 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@pages contains the number of pages.
  *	Return 0 if permission is granted.
  *
- * @register_security:
- *	allow module stacking.
- *	@name contains the name of the security module being stacked.
- *	@ops contains a pointer to the struct security_operations of the module to stack.
- *
  * @secid_to_secctx:
  *	Convert secid to security context.
  *	@secid contains the security ID.
@@ -1471,10 +1466,6 @@ struct security_operations {
 	int (*netlink_send) (struct sock *sk, struct sk_buff *skb);
 	int (*netlink_recv) (struct sk_buff *skb, int cap);
 
-	/* allow module stacking */
-	int (*register_security) (const char *name,
-				  struct security_operations *ops);
-
 	void (*d_instantiate) (struct dentry *dentry, struct inode *inode);
 
 	int (*getprocattr) (struct task_struct *p, char *name, char **value);
@@ -1564,7 +1555,6 @@ struct security_operations {
 extern int security_init(void);
 extern int security_module_enable(struct security_operations *ops);
 extern int register_security(struct security_operations *ops);
-extern int mod_reg_security(const char *name, struct security_operations *ops);
 extern struct dentry *securityfs_create_file(const char *name, mode_t mode,
 					     struct dentry *parent, void *data,
 					     const struct file_operations *fops);
diff --git a/security/capability.c b/security/capability.c
index 6e0671c82018..5b01c0b02422 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -721,12 +721,6 @@ static int cap_xfrm_decode_session(struct sk_buff *skb, u32 *fl, int ckall)
 }
 
 #endif /* CONFIG_SECURITY_NETWORK_XFRM */
-static int cap_register_security(const char *name,
-				 struct security_operations *ops)
-{
-	return -EINVAL;
-}
-
 static void cap_d_instantiate(struct dentry *dentry, struct inode *inode)
 {
 }
@@ -940,7 +934,6 @@ void security_fixup_ops(struct security_operations *ops)
 	set_to_cap_if_null(ops, sem_semop);
 	set_to_cap_if_null(ops, netlink_send);
 	set_to_cap_if_null(ops, netlink_recv);
-	set_to_cap_if_null(ops, register_security);
 	set_to_cap_if_null(ops, d_instantiate);
 	set_to_cap_if_null(ops, getprocattr);
 	set_to_cap_if_null(ops, setprocattr);
diff --git a/security/root_plug.c b/security/root_plug.c
index a41cf42a4fa0..be0ebec2580b 100644
--- a/security/root_plug.c
+++ b/security/root_plug.c
@@ -28,9 +28,6 @@
 #include <linux/usb.h>
 #include <linux/moduleparam.h>
 
-/* flag to keep track of how we were registered */
-static int secondary;
-
 /* default is a generic type of usb to serial converter */
 static int vendor_id = 0x0557;
 static int product_id = 0x2008;
@@ -97,13 +94,7 @@ static int __init rootplug_init (void)
 	if (register_security (&rootplug_security_ops)) {
 		printk (KERN_INFO 
 			"Failure registering Root Plug module with the kernel\n");
-		/* try registering with primary module */
-		if (mod_reg_security (MY_NAME, &rootplug_security_ops)) {
-			printk (KERN_INFO "Failure registering Root Plug "
-				" module with primary security module.\n");
 			return -EINVAL;
-		}
-		secondary = 1;
 	}
 	printk (KERN_INFO "Root Plug module initialized, "
 		"vendor_id = %4.4x, product id = %4.4x\n", vendor_id, product_id);
diff --git a/security/security.c b/security/security.c
index 30b0278de394..59f23b5918b3 100644
--- a/security/security.c
+++ b/security/security.c
@@ -125,35 +125,6 @@ int register_security(struct security_operations *ops)
 	return 0;
 }
 
-/**
- * mod_reg_security - allows security modules to be "stacked"
- * @name: a pointer to a string with the name of the security_options to be registered
- * @ops: a pointer to the struct security_options that is to be registered
- *
- * This function allows security modules to be stacked if the currently loaded
- * security module allows this to happen.  It passes the @name and @ops to the
- * register_security function of the currently loaded security module.
- *
- * The return value depends on the currently loaded security module, with 0 as
- * success.
- */
-int mod_reg_security(const char *name, struct security_operations *ops)
-{
-	if (verify(ops)) {
-		printk(KERN_INFO "%s could not verify "
-		       "security operations.\n", __func__);
-		return -EINVAL;
-	}
-
-	if (ops == security_ops) {
-		printk(KERN_INFO "%s security operations "
-		       "already registered.\n", __func__);
-		return -EINVAL;
-	}
-
-	return security_ops->register_security(name, ops);
-}
-
 /* Security operations */
 
 int security_ptrace(struct task_struct *parent, struct task_struct *child,
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 745a69e74e38..91200feb3f9c 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -126,13 +126,11 @@ __setup("selinux=", selinux_enabled_setup);
 int selinux_enabled = 1;
 #endif
 
-/* Original (dummy) security module. */
-static struct security_operations *original_ops;
 
-/* Minimal support for a secondary security module,
-   just to allow the use of the dummy or capability modules.
-   The owlsm module can alternatively be used as a secondary
-   module as long as CONFIG_OWLSM_FD is not enabled. */
+/*
+ * Minimal support for a secondary security module,
+ * just to allow the use of the capability module.
+ */
 static struct security_operations *secondary_ops;
 
 /* Lists of inode and superblock security structures initialized
@@ -5115,24 +5113,6 @@ static void selinux_ipc_getsecid(struct kern_ipc_perm *ipcp, u32 *secid)
 	*secid = isec->sid;
 }
 
-/* module stacking operations */
-static int selinux_register_security(const char *name, struct security_operations *ops)
-{
-	if (secondary_ops != original_ops) {
-		printk(KERN_ERR "%s:  There is already a secondary security "
-		       "module registered.\n", __func__);
-		return -EINVAL;
-	}
-
-	secondary_ops = ops;
-
-	printk(KERN_INFO "%s:  Registering secondary module %s\n",
-	       __func__,
-	       name);
-
-	return 0;
-}
-
 static void selinux_d_instantiate(struct dentry *dentry, struct inode *inode)
 {
 	if (inode)
@@ -5517,8 +5497,6 @@ static struct security_operations selinux_ops = {
 	.sem_semctl =			selinux_sem_semctl,
 	.sem_semop =			selinux_sem_semop,
 
-	.register_security =		selinux_register_security,
-
 	.d_instantiate =		selinux_d_instantiate,
 
 	.getprocattr =			selinux_getprocattr,
@@ -5612,7 +5590,7 @@ static __init int selinux_init(void)
 					    0, SLAB_PANIC, NULL);
 	avc_init();
 
-	original_ops = secondary_ops = security_ops;
+	secondary_ops = security_ops;
 	if (!secondary_ops)
 		panic("SELinux: No initial security operations\n");
 	if (register_security(&selinux_ops))
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 3c7150b3493d..ee5a51cbc5eb 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -1822,27 +1822,6 @@ static void smack_ipc_getsecid(struct kern_ipc_perm *ipp, u32 *secid)
 	*secid = smack_to_secid(smack);
 }
 
-/* module stacking operations */
-
-/**
- * smack_register_security - stack capability module
- * @name: module name
- * @ops: module operations - ignored
- *
- * Allow the capability module to register.
- */
-static int smack_register_security(const char *name,
-				   struct security_operations *ops)
-{
-	if (strcmp(name, "capability") != 0)
-		return -EINVAL;
-
-	printk(KERN_INFO "%s:  Registering secondary module %s\n",
-	       __func__, name);
-
-	return 0;
-}
-
 /**
  * smack_d_instantiate - Make sure the blob is correct on an inode
  * @opt_dentry: unused
@@ -2673,8 +2652,6 @@ struct security_operations smack_ops = {
 	.netlink_send =			cap_netlink_send,
 	.netlink_recv = 		cap_netlink_recv,
 
-	.register_security = 		smack_register_security,
-
 	.d_instantiate = 		smack_d_instantiate,
 
 	.getprocattr = 			smack_getprocattr,
-- 
cgit v1.2.3


From 7e9db9eaefdb8798730790214ff1b7746006ec98 Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Mon, 14 Jul 2008 09:58:44 +0200
Subject: [S390] cio: Introduce modalias for css bus.

Add modalias and subchannel type attributes for all subchannels.
I/O subchannel specific attributes are now created in
io_subchannel_probe(). modalias and subchannel type are also
added to the uevent for the css bus. Also make the css modalias
known.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 Documentation/ABI/testing/sysfs-bus-css | 35 +++++++++++++++++
 drivers/s390/cio/cio.h                  |  1 +
 drivers/s390/cio/css.c                  | 69 ++++++++++++++++++++++++++++++---
 drivers/s390/cio/css.h                  |  2 -
 drivers/s390/cio/device.c               | 47 ++++++++++++++--------
 include/linux/mod_devicetable.h         |  9 +++++
 scripts/mod/file2alias.c                | 12 ++++++
 7 files changed, 151 insertions(+), 24 deletions(-)
 create mode 100644 Documentation/ABI/testing/sysfs-bus-css

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-bus-css b/Documentation/ABI/testing/sysfs-bus-css
new file mode 100644
index 000000000000..b585ec258a08
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-bus-css
@@ -0,0 +1,35 @@
+What:		/sys/bus/css/devices/.../type
+Date:		March 2008
+Contact:	Cornelia Huck <cornelia.huck@de.ibm.com>
+		linux-s390@vger.kernel.org
+Description:	Contains the subchannel type, as reported by the hardware.
+		This attribute is present for all subchannel types.
+
+What:		/sys/bus/css/devices/.../modalias
+Date:		March 2008
+Contact:	Cornelia Huck <cornelia.huck@de.ibm.com>
+		linux-s390@vger.kernel.org
+Description:	Contains the module alias as reported with uevents.
+		It is of the format css:t<type> and present for all
+		subchannel types.
+
+What:		/sys/bus/css/drivers/io_subchannel/.../chpids
+Date:		December 2002
+Contact:	Cornelia Huck <cornelia.huck@de.ibm.com>
+		linux-s390@vger.kernel.org
+Description:	Contains the ids of the channel paths used by this
+		subchannel, as reported by the channel subsystem
+		during subchannel recognition.
+		Note: This is an I/O-subchannel specific attribute.
+Users:		s390-tools, HAL
+
+What:		/sys/bus/css/drivers/io_subchannel/.../pimpampom
+Date:		December 2002
+Contact:	Cornelia Huck <cornelia.huck@de.ibm.com>
+		linux-s390@vger.kernel.org
+Description:	Contains the PIM/PAM/POM values, as reported by the
+		channel subsystem when last queried by the common I/O
+		layer (this implies that this attribute is not neccessarily
+		in sync with the values current in the channel subsystem).
+		Note: This is an I/O-subchannel specific attribute.
+Users:		s390-tools, HAL
diff --git a/drivers/s390/cio/cio.h b/drivers/s390/cio/cio.h
index 6e933aebe013..4062748e8346 100644
--- a/drivers/s390/cio/cio.h
+++ b/drivers/s390/cio/cio.h
@@ -3,6 +3,7 @@
 
 #include <linux/mutex.h>
 #include <linux/device.h>
+#include <linux/mod_devicetable.h>
 #include <asm/chpid.h>
 #include "chsc.h"
 #include "schid.h"
diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c
index b7f4b52c5a9a..53e7496dc90c 100644
--- a/drivers/s390/cio/css.c
+++ b/drivers/s390/cio/css.c
@@ -2,8 +2,7 @@
  *  drivers/s390/cio/css.c
  *  driver for channel subsystem
  *
- *    Copyright (C) 2002 IBM Deutschland Entwicklung GmbH,
- *			 IBM Corporation
+ *    Copyright IBM Corp. 2002,2008
  *    Author(s): Arnd Bergmann (arndb@de.ibm.com)
  *		 Cornelia Huck (cornelia.huck@de.ibm.com)
  */
@@ -210,6 +209,41 @@ void css_update_ssd_info(struct subchannel *sch)
 	}
 }
 
+static ssize_t type_show(struct device *dev, struct device_attribute *attr,
+			 char *buf)
+{
+	struct subchannel *sch = to_subchannel(dev);
+
+	return sprintf(buf, "%01x\n", sch->st);
+}
+
+static DEVICE_ATTR(type, 0444, type_show, NULL);
+
+static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	struct subchannel *sch = to_subchannel(dev);
+
+	return sprintf(buf, "css:t%01X\n", sch->st);
+}
+
+static DEVICE_ATTR(modalias, 0444, modalias_show, NULL);
+
+static struct attribute *subch_attrs[] = {
+	&dev_attr_type.attr,
+	&dev_attr_modalias.attr,
+	NULL,
+};
+
+static struct attribute_group subch_attr_group = {
+	.attrs = subch_attrs,
+};
+
+static struct attribute_group *default_subch_attr_groups[] = {
+	&subch_attr_group,
+	NULL,
+};
+
 static int css_register_subchannel(struct subchannel *sch)
 {
 	int ret;
@@ -218,16 +252,17 @@ static int css_register_subchannel(struct subchannel *sch)
 	sch->dev.parent = &channel_subsystems[0]->device;
 	sch->dev.bus = &css_bus_type;
 	sch->dev.release = &css_subchannel_release;
-	sch->dev.groups = subch_attr_groups;
+	sch->dev.groups = default_subch_attr_groups;
 	/*
 	 * We don't want to generate uevents for I/O subchannels that don't
 	 * have a working ccw device behind them since they will be
 	 * unregistered before they can be used anyway, so we delay the add
 	 * uevent until after device recognition was successful.
+	 * Note that we suppress the uevent for all subchannel types;
+	 * the subchannel driver can decide itself when it wants to inform
+	 * userspace of its existence.
 	 */
-	if (!cio_is_console(sch->schid))
-		/* Console is special, no need to suppress. */
-		sch->dev.uevent_suppress = 1;
+	sch->dev.uevent_suppress = 1;
 	css_update_ssd_info(sch);
 	/* make it known to the system */
 	ret = css_sch_device_register(sch);
@@ -236,6 +271,15 @@ static int css_register_subchannel(struct subchannel *sch)
 			      sch->schid.ssid, sch->schid.sch_no, ret);
 		return ret;
 	}
+	if (!sch->driver) {
+		/*
+		 * No driver matched. Generate the uevent now so that
+		 * a fitting driver module may be loaded based on the
+		 * modalias.
+		 */
+		sch->dev.uevent_suppress = 0;
+		kobject_uevent(&sch->dev.kobj, KOBJ_ADD);
+	}
 	return ret;
 }
 
@@ -926,12 +970,25 @@ static void css_shutdown(struct device *dev)
 		sch->driver->shutdown(sch);
 }
 
+static int css_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct subchannel *sch = to_subchannel(dev);
+	int ret;
+
+	ret = add_uevent_var(env, "ST=%01X", sch->st);
+	if (ret)
+		return ret;
+	ret = add_uevent_var(env, "MODALIAS=css:t%01X", sch->st);
+	return ret;
+}
+
 struct bus_type css_bus_type = {
 	.name     = "css",
 	.match    = css_bus_match,
 	.probe    = css_probe,
 	.remove   = css_remove,
 	.shutdown = css_shutdown,
+	.uevent   = css_uevent,
 };
 
 /**
diff --git a/drivers/s390/cio/css.h b/drivers/s390/cio/css.h
index bfe0ada43f2c..e0fc7b499784 100644
--- a/drivers/s390/cio/css.h
+++ b/drivers/s390/cio/css.h
@@ -143,6 +143,4 @@ int css_sch_is_valid(struct schib *);
 
 extern struct workqueue_struct *slow_path_wq;
 void css_wait_for_slow_path(void);
-
-extern struct attribute_group *subch_attr_groups[];
 #endif
diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c
index 0ed5a81260bc..23b129fd4d8d 100644
--- a/drivers/s390/cio/device.c
+++ b/drivers/s390/cio/device.c
@@ -585,19 +585,14 @@ static DEVICE_ATTR(modalias, 0444, modalias_show, NULL);
 static DEVICE_ATTR(online, 0644, online_show, online_store);
 static DEVICE_ATTR(availability, 0444, available_show, NULL);
 
-static struct attribute * subch_attrs[] = {
+static struct attribute *io_subchannel_attrs[] = {
 	&dev_attr_chpids.attr,
 	&dev_attr_pimpampom.attr,
 	NULL,
 };
 
-static struct attribute_group subch_attr_group = {
-	.attrs = subch_attrs,
-};
-
-struct attribute_group *subch_attr_groups[] = {
-	&subch_attr_group,
-	NULL,
+static struct attribute_group io_subchannel_attr_group = {
+	.attrs = io_subchannel_attrs,
 };
 
 static struct attribute * ccwdev_attrs[] = {
@@ -1157,11 +1152,21 @@ static int io_subchannel_probe(struct subchannel *sch)
 
 	cdev = sch_get_cdev(sch);
 	if (cdev) {
+		rc = sysfs_create_group(&sch->dev.kobj,
+					&io_subchannel_attr_group);
+		if (rc)
+			CIO_MSG_EVENT(0, "Failed to create io subchannel "
+				      "attributes for subchannel "
+				      "0.%x.%04x (rc=%d)\n",
+				      sch->schid.ssid, sch->schid.sch_no, rc);
 		/*
 		 * This subchannel already has an associated ccw_device.
-		 * Register it and exit. This happens for all early
-		 * device, e.g. the console.
+		 * Throw the delayed uevent for the subchannel, register
+		 * the ccw_device and exit. This happens for all early
+		 * devices, e.g. the console.
 		 */
+		sch->dev.uevent_suppress = 0;
+		kobject_uevent(&sch->dev.kobj, KOBJ_ADD);
 		cdev->dev.groups = ccwdev_attr_groups;
 		device_initialize(&cdev->dev);
 		ccw_device_register(cdev);
@@ -1184,11 +1189,17 @@ static int io_subchannel_probe(struct subchannel *sch)
 	 */
 	dev_id.devno = sch->schib.pmcw.dev;
 	dev_id.ssid = sch->schid.ssid;
+	rc = sysfs_create_group(&sch->dev.kobj,
+				&io_subchannel_attr_group);
+	if (rc)
+		return rc;
 	/* Allocate I/O subchannel private data. */
 	sch->private = kzalloc(sizeof(struct io_subchannel_private),
 			       GFP_KERNEL | GFP_DMA);
-	if (!sch->private)
-		return -ENOMEM;
+	if (!sch->private) {
+		rc = -ENOMEM;
+		goto out_err;
+	}
 	cdev = get_disc_ccwdev_by_dev_id(&dev_id, NULL);
 	if (!cdev)
 		cdev = get_orphaned_ccwdev_by_dev_id(to_css(sch->dev.parent),
@@ -1207,8 +1218,8 @@ static int io_subchannel_probe(struct subchannel *sch)
 	}
 	cdev = io_subchannel_create_ccwdev(sch);
 	if (IS_ERR(cdev)) {
-		kfree(sch->private);
-		return PTR_ERR(cdev);
+		rc = PTR_ERR(cdev);
+		goto out_err;
 	}
 	rc = io_subchannel_recog(cdev, sch);
 	if (rc) {
@@ -1217,9 +1228,12 @@ static int io_subchannel_probe(struct subchannel *sch)
 		spin_unlock_irqrestore(sch->lock, flags);
 		if (cdev->dev.release)
 			cdev->dev.release(&cdev->dev);
-		kfree(sch->private);
+		goto out_err;
 	}
-
+	return 0;
+out_err:
+	kfree(sch->private);
+	sysfs_remove_group(&sch->dev.kobj, &io_subchannel_attr_group);
 	return rc;
 }
 
@@ -1240,6 +1254,7 @@ io_subchannel_remove (struct subchannel *sch)
 	ccw_device_unregister(cdev);
 	put_device(&cdev->dev);
 	kfree(sch->private);
+	sysfs_remove_group(&sch->dev.kobj, &io_subchannel_attr_group);
 	return 0;
 }
 
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index 69b2342d5ebb..1fd03e732e07 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -159,6 +159,15 @@ struct ap_device_id {
 
 #define AP_DEVICE_ID_MATCH_DEVICE_TYPE		0x01
 
+/* s390 css bus devices (subchannels) */
+struct css_device_id {
+	__u8 type; /* subchannel type */
+	__u8 pad1;
+	__u16 pad2;
+	__u32 pad3;
+	kernel_ulong_t driver_data;
+};
+
 #define ACPI_ID_LEN	16 /* only 9 bytes needed here, 16 bytes are used */
 			   /* to workaround crosscompile issues */
 
diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c
index cea4a790e1e9..37d5c363fbcd 100644
--- a/scripts/mod/file2alias.c
+++ b/scripts/mod/file2alias.c
@@ -304,6 +304,14 @@ static int do_ap_entry(const char *filename,
 	return 1;
 }
 
+/* looks like: "css:tN" */
+static int do_css_entry(const char *filename,
+			struct css_device_id *id, char *alias)
+{
+	sprintf(alias, "css:t%01X", id->type);
+	return 1;
+}
+
 /* Looks like: "serio:tyNprNidNexN" */
 static int do_serio_entry(const char *filename,
 			  struct serio_device_id *id, char *alias)
@@ -680,6 +688,10 @@ void handle_moddevtable(struct module *mod, struct elf_info *info,
 		do_table(symval, sym->st_size,
 			 sizeof(struct ap_device_id), "ap",
 			 do_ap_entry, mod);
+	else if (sym_is(symname, "__mod_css_device_table"))
+		do_table(symval, sym->st_size,
+			 sizeof(struct css_device_id), "css",
+			 do_css_entry, mod);
 	else if (sym_is(symname, "__mod_serio_device_table"))
 		do_table(symval, sym->st_size,
 			 sizeof(struct serio_device_id), "serio",
-- 
cgit v1.2.3


From f08adc008d84f6b03d377ede951e29ed169e76e2 Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Mon, 14 Jul 2008 09:59:03 +0200
Subject: [S390] css: Use css_device_id for bus matching.

css_device_id exists, so use it for determining the right driver
(and add a match_flags which is always 1 for valid types).

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 drivers/s390/cio/css.c          | 15 ++++++---------
 drivers/s390/cio/css.h          |  2 +-
 drivers/s390/cio/device.c       |  8 +++++++-
 include/linux/mod_devicetable.h |  2 +-
 4 files changed, 15 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c
index 45ba07c0a286..4e2f2bbf4ba5 100644
--- a/drivers/s390/cio/css.c
+++ b/drivers/s390/cio/css.c
@@ -850,19 +850,16 @@ int sch_is_pseudo_sch(struct subchannel *sch)
 	return sch == to_css(sch->dev.parent)->pseudo_subchannel;
 }
 
-/*
- * find a driver for a subchannel. They identify by the subchannel
- * type with the exception that the console subchannel driver has its own
- * subchannel type although the device is an i/o subchannel
- */
-static int
-css_bus_match (struct device *dev, struct device_driver *drv)
+static int css_bus_match(struct device *dev, struct device_driver *drv)
 {
 	struct subchannel *sch = to_subchannel(dev);
 	struct css_driver *driver = to_cssdriver(drv);
+	struct css_device_id *id;
 
-	if (sch->st == driver->subchannel_type)
-		return 1;
+	for (id = driver->subchannel_type; id->match_flags; id++) {
+		if (sch->st == id->type)
+			return 1;
+	}
 
 	return 0;
 }
diff --git a/drivers/s390/cio/css.h b/drivers/s390/cio/css.h
index 38bf9ddb8412..58020bf41ed0 100644
--- a/drivers/s390/cio/css.h
+++ b/drivers/s390/cio/css.h
@@ -75,7 +75,7 @@ struct chp_link;
  */
 struct css_driver {
 	struct module *owner;
-	unsigned int subchannel_type;
+	struct css_device_id *subchannel_type;
 	struct device_driver drv;
 	void (*irq)(struct subchannel *);
 	int (*chp_event)(struct subchannel *, struct chp_link *, int);
diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c
index 522d47afc950..c904cb84d75e 100644
--- a/drivers/s390/cio/device.c
+++ b/drivers/s390/cio/device.c
@@ -131,9 +131,15 @@ static int io_subchannel_sch_event(struct subchannel *, int);
 static int io_subchannel_chp_event(struct subchannel *, struct chp_link *,
 				   int);
 
+static struct css_device_id io_subchannel_ids[] = {
+	{ .match_flags = 0x1, .type = SUBCHANNEL_TYPE_IO, },
+	{ /* end of list */ },
+};
+MODULE_DEVICE_TABLE(css, io_subchannel_ids);
+
 static struct css_driver io_subchannel_driver = {
 	.owner = THIS_MODULE,
-	.subchannel_type = SUBCHANNEL_TYPE_IO,
+	.subchannel_type = io_subchannel_ids,
 	.name = "io_subchannel",
 	.irq = io_subchannel_irq,
 	.sch_event = io_subchannel_sch_event,
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index 1fd03e732e07..c4db5827963d 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -161,8 +161,8 @@ struct ap_device_id {
 
 /* s390 css bus devices (subchannels) */
 struct css_device_id {
+	__u8 match_flags;
 	__u8 type; /* subchannel type */
-	__u8 pad1;
 	__u16 pad2;
 	__u32 pad3;
 	kernel_ulong_t driver_data;
-- 
cgit v1.2.3


From 2f3804edf971d2080243d2b4552bfd61ddfbf969 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Wed, 2 Jul 2008 01:36:15 -0500
Subject: powerpc/85xx: Add support for MPC8536DS

Add support for the MPC8536 process and MPC8536DS reference board.  The
MPC8536 is an e500v2 based SoC which eTSEC, USB, SATA, PCI, and PCIe.

The USB and SATA IP blocks are similiar to those on the PQ2 Pro SoCs and
thus use the same drivers.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/boot/dts/mpc8536ds.dts       |  420 ++++++++
 arch/powerpc/configs/mpc8536_ds_defconfig | 1637 +++++++++++++++++++++++++++++
 arch/powerpc/platforms/85xx/Kconfig       |    6 +
 arch/powerpc/platforms/85xx/Makefile      |    1 +
 arch/powerpc/platforms/85xx/mpc8536_ds.c  |  125 +++
 arch/powerpc/sysdev/fsl_pci.c             |    2 +
 include/linux/pci_ids.h                   |    2 +
 7 files changed, 2193 insertions(+)
 create mode 100644 arch/powerpc/boot/dts/mpc8536ds.dts
 create mode 100644 arch/powerpc/configs/mpc8536_ds_defconfig
 create mode 100644 arch/powerpc/platforms/85xx/mpc8536_ds.c

(limited to 'include/linux')

diff --git a/arch/powerpc/boot/dts/mpc8536ds.dts b/arch/powerpc/boot/dts/mpc8536ds.dts
new file mode 100644
index 000000000000..98ad27a2ddee
--- /dev/null
+++ b/arch/powerpc/boot/dts/mpc8536ds.dts
@@ -0,0 +1,420 @@
+/*
+ * MPC8536 DS Device Tree Source
+ *
+ * Copyright 2008 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+/dts-v1/;
+
+/ {
+	model = "fsl,mpc8536ds";
+	compatible = "fsl,mpc8536ds";
+	#address-cells = <1>;
+	#size-cells = <1>;
+
+	aliases {
+		ethernet0 = &enet0;
+		ethernet1 = &enet1;
+		serial0 = &serial0;
+		serial1 = &serial1;
+		pci0 = &pci0;
+		pci1 = &pci1;
+		pci2 = &pci2;
+		pci3 = &pci3;
+	};
+
+	cpus {
+		#cpus = <1>;
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		PowerPC,8536@0 {
+			device_type = "cpu";
+			reg = <0>;
+			next-level-cache = <&L2>;
+		};
+	};
+
+	memory {
+		device_type = "memory";
+		reg = <00000000 00000000>;	// Filled by U-Boot
+	};
+
+	soc@ffe00000 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		device_type = "soc";
+		ranges = <0x0 0xffe00000 0x100000>;
+		reg = <0xffe00000 0x1000>;
+		bus-frequency = <0>;		// Filled out by uboot.
+
+		memory-controller@2000 {
+			compatible = "fsl,mpc8536-memory-controller";
+			reg = <0x2000 0x1000>;
+			interrupt-parent = <&mpic>;
+			interrupts = <18 0x2>;
+		};
+
+		L2: l2-cache-controller@20000 {
+			compatible = "fsl,mpc8536-l2-cache-controller";
+			reg = <0x20000 0x1000>;
+			interrupt-parent = <&mpic>;
+			interrupts = <16 0x2>;
+		};
+
+		i2c@3000 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			cell-index = <0>;
+			compatible = "fsl-i2c";
+			reg = <0x3000 0x100>;
+			interrupts = <43 0x2>;
+			interrupt-parent = <&mpic>;
+			dfsrr;
+		};
+
+		i2c@3100 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			cell-index = <1>;
+			compatible = "fsl-i2c";
+			reg = <0x3100 0x100>;
+			interrupts = <43 0x2>;
+			interrupt-parent = <&mpic>;
+			dfsrr;
+			rtc@68 {
+				compatible = "dallas,ds3232";
+				reg = <0x68>;
+			};
+		};
+
+		dma@21300 {
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "fsl,mpc8536-dma", "fsl,eloplus-dma";
+			reg = <0x21300 4>;
+			ranges = <0 0x21100 0x200>;
+			cell-index = <0>;
+			dma-channel@0 {
+				compatible = "fsl,mpc8536-dma-channel",
+					     "fsl,eloplus-dma-channel";
+				reg = <0x0 0x80>;
+				cell-index = <0>;
+				interrupt-parent = <&mpic>;
+				interrupts = <14 0x2>;
+			};
+			dma-channel@80 {
+				compatible = "fsl,mpc8536-dma-channel",
+					     "fsl,eloplus-dma-channel";
+				reg = <0x80 0x80>;
+				cell-index = <1>;
+				interrupt-parent = <&mpic>;
+				interrupts = <15 0x2>;
+			};
+			dma-channel@100 {
+				compatible = "fsl,mpc8536-dma-channel",
+					     "fsl,eloplus-dma-channel";
+				reg = <0x100 0x80>;
+				cell-index = <2>;
+				interrupt-parent = <&mpic>;
+				interrupts = <16 0x2>;
+			};
+			dma-channel@180 {
+				compatible = "fsl,mpc8536-dma-channel",
+					     "fsl,eloplus-dma-channel";
+				reg = <0x180 0x80>;
+				cell-index = <3>;
+				interrupt-parent = <&mpic>;
+				interrupts = <17 0x2>;
+			};
+		};
+
+		mdio@24520 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			compatible = "fsl,gianfar-mdio";
+			reg = <0x24520 0x20>;
+
+			phy0: ethernet-phy@0 {
+				interrupt-parent = <&mpic>;
+				interrupts = <10 0x1>;
+				reg = <0>;
+				device_type = "ethernet-phy";
+			};
+			phy1: ethernet-phy@1 {
+				interrupt-parent = <&mpic>;
+				interrupts = <10 0x1>;
+				reg = <1>;
+				device_type = "ethernet-phy";
+			};
+		};
+
+		usb@22000 {
+			compatible = "fsl,mpc8536-usb2-mph", "fsl-usb2-mph";
+			reg = <0x22000 0x1000>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+			interrupt-parent = <&mpic>;
+			interrupts = <28 0x2>;
+			phy_type = "ulpi";
+		};
+
+		usb@23000 {
+			compatible = "fsl,mpc8536-usb2-mph", "fsl-usb2-mph";
+			reg = <0x23000 0x1000>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+			interrupt-parent = <&mpic>;
+			interrupts = <46 0x2>;
+			phy_type = "ulpi";
+		};
+
+		enet0: ethernet@24000 {
+			cell-index = <0>;
+			device_type = "network";
+			model = "TSEC";
+			compatible = "gianfar";
+			reg = <0x24000 0x1000>;
+			local-mac-address = [ 00 00 00 00 00 00 ];
+			interrupts = <29 2 30 2 34 2>;
+			interrupt-parent = <&mpic>;
+			phy-handle = <&phy1>;
+			phy-connection-type = "rgmii-id";
+		};
+
+		enet1: ethernet@26000 {
+			cell-index = <1>;
+			device_type = "network";
+			model = "TSEC";
+			compatible = "gianfar";
+			reg = <0x26000 0x1000>;
+			local-mac-address = [ 00 00 00 00 00 00 ];
+			interrupts = <31 2 32 2 33 2>;
+			interrupt-parent = <&mpic>;
+			phy-handle = <&phy0>;
+			phy-connection-type = "rgmii-id";
+		};
+
+		usb@2b000 {
+			compatible = "fsl,mpc8536-usb2-dr", "fsl-usb2-dr";
+			reg = <0x2b000 0x1000>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+			interrupt-parent = <&mpic>;
+			interrupts = <60 0x2>;
+			dr_mode = "peripheral";
+			phy_type = "ulpi";
+		};
+
+		serial0: serial@4500 {
+			cell-index = <0>;
+			device_type = "serial";
+			compatible = "ns16550";
+			reg = <0x4500 0x100>;
+			clock-frequency = <0>;
+			interrupts = <42 0x2>;
+			interrupt-parent = <&mpic>;
+		};
+
+		serial1: serial@4600 {
+			cell-index = <1>;
+			device_type = "serial";
+			compatible = "ns16550";
+			reg = <0x4600 0x100>;
+			clock-frequency = <0>;
+			interrupts = <42 0x2>;
+			interrupt-parent = <&mpic>;
+		};
+
+		sata@18000 {
+			compatible = "fsl,mpc8536-sata", "fsl,pq-sata";
+			reg = <0x18000 0x1000>;
+			cell-index = <1>;
+			interrupts = <74 0x2>;
+			interrupt-parent = <&mpic>;
+		};
+
+		sata@19000 {
+			compatible = "fsl,mpc8536-sata", "fsl,pq-sata";
+			reg = <0x19000 0x1000>;
+			cell-index = <2>;
+			interrupts = <41 0x2>;
+			interrupt-parent = <&mpic>;
+		};
+
+		global-utilities@e0000 {	//global utilities block
+			compatible = "fsl,mpc8548-guts";
+			reg = <0xe0000 0x1000>;
+			fsl,has-rstcr;
+		};
+
+		mpic: pic@40000 {
+			clock-frequency = <0>;
+			interrupt-controller;
+			#address-cells = <0>;
+			#interrupt-cells = <2>;
+			reg = <0x40000 0x40000>;
+			compatible = "chrp,open-pic";
+			device_type = "open-pic";
+			big-endian;
+		};
+
+		msi@41600 {
+			compatible = "fsl,mpc8536-msi", "fsl,mpic-msi";
+			reg = <0x41600 0x80>;
+			msi-available-ranges = <0 0x100>;
+			interrupts = <
+				0xe0 0
+				0xe1 0
+				0xe2 0
+				0xe3 0
+				0xe4 0
+				0xe5 0
+				0xe6 0
+				0xe7 0>;
+			interrupt-parent = <&mpic>;
+		};
+	};
+
+	pci0: pci@ffe08000 {
+		cell-index = <0>;
+		compatible = "fsl,mpc8540-pci";
+		device_type = "pci";
+		interrupt-map-mask = <0xf800 0x0 0x0 0x7>;
+		interrupt-map = <
+
+			/* IDSEL 0x11 J17 Slot 1 */
+			0x8800 0 0 1 &mpic 1 1
+			0x8800 0 0 2 &mpic 2 1
+			0x8800 0 0 3 &mpic 3 1
+			0x8800 0 0 4 &mpic 4 1>;
+
+		interrupt-parent = <&mpic>;
+		interrupts = <24 0x2>;
+		bus-range = <0 0xff>;
+		ranges = <0x02000000 0 0x80000000 0x80000000 0 0x10000000
+			  0x01000000 0 0x00000000 0xffc00000 0 0x00010000>;
+		clock-frequency = <66666666>;
+		#interrupt-cells = <1>;
+		#size-cells = <2>;
+		#address-cells = <3>;
+		reg = <0xffe08000 0x1000>;
+	};
+
+	pci1: pcie@ffe09000 {
+		cell-index = <1>;
+		compatible = "fsl,mpc8548-pcie";
+		device_type = "pci";
+		#interrupt-cells = <1>;
+		#size-cells = <2>;
+		#address-cells = <3>;
+		reg = <0xffe09000 0x1000>;
+		bus-range = <0 0xff>;
+		ranges = <0x02000000 0 0x98000000 0x98000000 0 0x08000000
+			  0x01000000 0 0x00000000 0xffc20000 0 0x00010000>;
+		clock-frequency = <33333333>;
+		interrupt-parent = <&mpic>;
+		interrupts = <25 0x2>;
+		interrupt-map-mask = <0xf800 0 0 7>;
+		interrupt-map = <
+			/* IDSEL 0x0 */
+			0000 0 0 1 &mpic 4 1
+			0000 0 0 2 &mpic 5 1
+			0000 0 0 3 &mpic 6 1
+			0000 0 0 4 &mpic 7 1
+			>;
+		pcie@0 {
+			reg = <0 0 0 0 0>;
+			#size-cells = <2>;
+			#address-cells = <3>;
+			device_type = "pci";
+			ranges = <0x02000000 0 0x98000000
+				  0x02000000 0 0x98000000
+				  0 0x08000000
+
+				  0x01000000 0 0x00000000
+				  0x01000000 0 0x00000000
+				  0 0x00010000>;
+		};
+	};
+
+	pci2: pcie@ffe0a000 {
+		cell-index = <2>;
+		compatible = "fsl,mpc8548-pcie";
+		device_type = "pci";
+		#interrupt-cells = <1>;
+		#size-cells = <2>;
+		#address-cells = <3>;
+		reg = <0xffe0a000 0x1000>;
+		bus-range = <0 0xff>;
+		ranges = <0x02000000 0 0x90000000 0x90000000 0 0x08000000
+			  0x01000000 0 0x00000000 0xffc10000 0 0x00010000>;
+		clock-frequency = <33333333>;
+		interrupt-parent = <&mpic>;
+		interrupts = <26 0x2>;
+		interrupt-map-mask = <0xf800 0 0 7>;
+		interrupt-map = <
+			/* IDSEL 0x0 */
+			0000 0 0 1 &mpic 0 1
+			0000 0 0 2 &mpic 1 1
+			0000 0 0 3 &mpic 2 1
+			0000 0 0 4 &mpic 3 1
+			>;
+		pcie@0 {
+			reg = <0 0 0 0 0>;
+			#size-cells = <2>;
+			#address-cells = <3>;
+			device_type = "pci";
+			ranges = <0x02000000 0 0x90000000
+				  0x02000000 0 0x90000000
+				  0 0x08000000
+
+				  0x01000000 0 0x00000000
+				  0x01000000 0 0x00000000
+				  0 0x00010000>;
+		};
+	};
+
+	pci3: pcie@ffe0b000 {
+		cell-index = <3>;
+		compatible = "fsl,mpc8548-pcie";
+		device_type = "pci";
+		#interrupt-cells = <1>;
+		#size-cells = <2>;
+		#address-cells = <3>;
+		reg = <0xffe0b000 0x1000>;
+		bus-range = <0 0xff>;
+		ranges = <0x02000000 0 0xa0000000 0xa0000000 0 0x20000000
+			  0x01000000 0 0x00000000 0xffc30000 0 0x00010000>;
+		clock-frequency = <33333333>;
+		interrupt-parent = <&mpic>;
+		interrupts = <27 0x2>;
+		interrupt-map-mask = <0xf800 0 0 7>;
+		interrupt-map = <
+			/* IDSEL 0x0 */
+			0000 0 0 1 &mpic 8 1
+			0000 0 0 2 &mpic 9 1
+			0000 0 0 3 &mpic 10 1
+			0000 0 0 4 &mpic 11 1
+			>;
+
+		pcie@0 {
+			reg = <0 0 0 0 0>;
+			#size-cells = <2>;
+			#address-cells = <3>;
+			device_type = "pci";
+			ranges = <0x02000000 0 0xa0000000
+				  0x02000000 0 0xa0000000
+				  0 0x20000000
+
+				  0x01000000 0 0x00000000
+				  0x01000000 0 0x00000000
+				  0 0x00100000>;
+		};
+	};
+};
diff --git a/arch/powerpc/configs/mpc8536_ds_defconfig b/arch/powerpc/configs/mpc8536_ds_defconfig
new file mode 100644
index 000000000000..f1e2931de5da
--- /dev/null
+++ b/arch/powerpc/configs/mpc8536_ds_defconfig
@@ -0,0 +1,1637 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.26-rc8
+# Wed Jul  2 01:34:26 2008
+#
+# CONFIG_PPC64 is not set
+
+#
+# Processor support
+#
+# CONFIG_6xx is not set
+CONFIG_PPC_85xx=y
+# CONFIG_PPC_8xx is not set
+# CONFIG_40x is not set
+# CONFIG_44x is not set
+# CONFIG_E200 is not set
+CONFIG_E500=y
+# CONFIG_PPC_E500MC is not set
+CONFIG_BOOKE=y
+CONFIG_FSL_BOOKE=y
+CONFIG_FSL_EMB_PERFMON=y
+# CONFIG_PHYS_64BIT is not set
+CONFIG_SPE=y
+# CONFIG_PPC_MM_SLICES is not set
+CONFIG_PPC32=y
+CONFIG_WORD_SIZE=32
+CONFIG_PPC_MERGE=y
+CONFIG_MMU=y
+CONFIG_GENERIC_CMOS_UPDATE=y
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_TIME_VSYSCALL=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+CONFIG_GENERIC_HARDIRQS=y
+# CONFIG_HAVE_SETUP_PER_CPU_AREA is not set
+CONFIG_IRQ_PER_CPU=y
+CONFIG_STACKTRACE_SUPPORT=y
+CONFIG_LOCKDEP_SUPPORT=y
+CONFIG_RWSEM_XCHGADD_ALGORITHM=y
+CONFIG_ARCH_HAS_ILOG2_U32=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+# CONFIG_ARCH_NO_VIRT_TO_BUS is not set
+CONFIG_PPC=y
+CONFIG_EARLY_PRINTK=y
+CONFIG_GENERIC_NVRAM=y
+CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
+CONFIG_ARCH_MAY_HAVE_PC_FDC=y
+CONFIG_PPC_OF=y
+CONFIG_OF=y
+CONFIG_PPC_UDBG_16550=y
+# CONFIG_GENERIC_TBSYNC is not set
+CONFIG_AUDIT_ARCH=y
+CONFIG_GENERIC_BUG=y
+CONFIG_DEFAULT_UIMAGE=y
+# CONFIG_PPC_DCR_NATIVE is not set
+# CONFIG_PPC_DCR_MMIO is not set
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+CONFIG_SYSVIPC_SYSCTL=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_BSD_PROCESS_ACCT=y
+# CONFIG_BSD_PROCESS_ACCT_V3 is not set
+# CONFIG_TASKSTATS is not set
+CONFIG_AUDIT=y
+# CONFIG_AUDITSYSCALL is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+CONFIG_GROUP_SCHED=y
+# CONFIG_FAIR_GROUP_SCHED is not set
+# CONFIG_RT_GROUP_SCHED is not set
+CONFIG_USER_SCHED=y
+# CONFIG_CGROUP_SCHED is not set
+CONFIG_SYSFS_DEPRECATED=y
+CONFIG_SYSFS_DEPRECATED_V2=y
+# CONFIG_RELAY is not set
+# CONFIG_NAMESPACES is not set
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+CONFIG_SYSCTL_SYSCALL=y
+CONFIG_SYSCTL_SYSCALL_CHECK=y
+CONFIG_KALLSYMS=y
+CONFIG_KALLSYMS_ALL=y
+CONFIG_KALLSYMS_EXTRA_PASS=y
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_COMPAT_BRK=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_ANON_INODES=y
+CONFIG_EPOLL=y
+CONFIG_SIGNALFD=y
+CONFIG_TIMERFD=y
+CONFIG_EVENTFD=y
+CONFIG_SHMEM=y
+CONFIG_VM_EVENT_COUNTERS=y
+CONFIG_SLUB_DEBUG=y
+# CONFIG_SLAB is not set
+CONFIG_SLUB=y
+# CONFIG_SLOB is not set
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+CONFIG_HAVE_OPROFILE=y
+# CONFIG_KPROBES is not set
+CONFIG_HAVE_KPROBES=y
+CONFIG_HAVE_KRETPROBES=y
+# CONFIG_HAVE_DMA_ATTRS is not set
+CONFIG_PROC_PAGE_MONITOR=y
+CONFIG_SLABINFO=y
+CONFIG_RT_MUTEXES=y
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+# CONFIG_MODULE_FORCE_LOAD is not set
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+CONFIG_MODVERSIONS=y
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+CONFIG_BLOCK=y
+CONFIG_LBD=y
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+CONFIG_DEFAULT_CFQ=y
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="cfq"
+CONFIG_CLASSIC_RCU=y
+
+#
+# Platform support
+#
+# CONFIG_PPC_MPC512x is not set
+# CONFIG_PPC_MPC5121 is not set
+# CONFIG_PPC_CELL is not set
+# CONFIG_PPC_CELL_NATIVE is not set
+# CONFIG_PQ2ADS is not set
+CONFIG_MPC85xx=y
+# CONFIG_MPC8540_ADS is not set
+# CONFIG_MPC8560_ADS is not set
+# CONFIG_MPC85xx_CDS is not set
+# CONFIG_MPC85xx_MDS is not set
+CONFIG_MPC8536_DS=y
+# CONFIG_MPC85xx_DS is not set
+# CONFIG_KSI8560 is not set
+# CONFIG_STX_GP3 is not set
+# CONFIG_TQM8540 is not set
+# CONFIG_TQM8541 is not set
+# CONFIG_TQM8548 is not set
+# CONFIG_TQM8555 is not set
+# CONFIG_TQM8560 is not set
+# CONFIG_SBC8548 is not set
+# CONFIG_SBC8560 is not set
+# CONFIG_IPIC is not set
+CONFIG_MPIC=y
+# CONFIG_MPIC_WEIRD is not set
+# CONFIG_PPC_I8259 is not set
+# CONFIG_PPC_RTAS is not set
+# CONFIG_MMIO_NVRAM is not set
+# CONFIG_PPC_MPC106 is not set
+# CONFIG_PPC_970_NAP is not set
+# CONFIG_PPC_INDIRECT_IO is not set
+# CONFIG_GENERIC_IOMAP is not set
+# CONFIG_CPU_FREQ is not set
+# CONFIG_CPM2 is not set
+# CONFIG_FSL_ULI1575 is not set
+
+#
+# Kernel options
+#
+CONFIG_HIGHMEM=y
+CONFIG_TICK_ONESHOT=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_300 is not set
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+# CONFIG_SCHED_HRTICK is not set
+CONFIG_PREEMPT_NONE=y
+# CONFIG_PREEMPT_VOLUNTARY is not set
+# CONFIG_PREEMPT is not set
+CONFIG_BINFMT_ELF=y
+CONFIG_BINFMT_MISC=m
+CONFIG_MATH_EMULATION=y
+# CONFIG_IOMMU_HELPER is not set
+CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y
+CONFIG_ARCH_HAS_WALK_MEMORY=y
+CONFIG_ARCH_ENABLE_MEMORY_HOTREMOVE=y
+CONFIG_ARCH_FLATMEM_ENABLE=y
+CONFIG_ARCH_POPULATES_NODE_MAP=y
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_PAGEFLAGS_EXTENDED=y
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_BOUNCE=y
+CONFIG_VIRT_TO_BUS=y
+CONFIG_FORCE_MAX_ZONEORDER=11
+CONFIG_PROC_DEVICETREE=y
+# CONFIG_CMDLINE_BOOL is not set
+# CONFIG_PM is not set
+CONFIG_SECCOMP=y
+CONFIG_ISA_DMA_API=y
+
+#
+# Bus options
+#
+CONFIG_ZONE_DMA=y
+CONFIG_PPC_INDIRECT_PCI=y
+CONFIG_FSL_SOC=y
+CONFIG_FSL_PCI=y
+CONFIG_PCI=y
+CONFIG_PCI_DOMAINS=y
+CONFIG_PCI_SYSCALL=y
+# CONFIG_PCIEPORTBUS is not set
+CONFIG_ARCH_SUPPORTS_MSI=y
+# CONFIG_PCI_MSI is not set
+CONFIG_PCI_LEGACY=y
+# CONFIG_PCI_DEBUG is not set
+# CONFIG_PCCARD is not set
+# CONFIG_HOTPLUG_PCI is not set
+# CONFIG_HAS_RAPIDIO is not set
+
+#
+# Advanced setup
+#
+# CONFIG_ADVANCED_OPTIONS is not set
+
+#
+# Default settings for advanced configuration options are used
+#
+CONFIG_LOWMEM_SIZE=0x30000000
+CONFIG_PAGE_OFFSET=0xc0000000
+CONFIG_KERNEL_START=0xc0000000
+CONFIG_PHYSICAL_START=0x00000000
+CONFIG_PHYSICAL_ALIGN=0x10000000
+CONFIG_TASK_SIZE=0xc0000000
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+CONFIG_XFRM=y
+CONFIG_XFRM_USER=y
+# CONFIG_XFRM_SUB_POLICY is not set
+# CONFIG_XFRM_MIGRATE is not set
+# CONFIG_XFRM_STATISTICS is not set
+CONFIG_NET_KEY=m
+# CONFIG_NET_KEY_MIGRATE is not set
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IP_ROUTE_MULTIPATH=y
+CONFIG_IP_ROUTE_VERBOSE=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+CONFIG_NET_IPIP=y
+CONFIG_NET_IPGRE=y
+CONFIG_NET_IPGRE_BROADCAST=y
+CONFIG_IP_MROUTE=y
+CONFIG_IP_PIMSM_V1=y
+CONFIG_IP_PIMSM_V2=y
+CONFIG_ARPD=y
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+CONFIG_INET_TUNNEL=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+CONFIG_IPV6=y
+# CONFIG_IPV6_PRIVACY is not set
+# CONFIG_IPV6_ROUTER_PREF is not set
+# CONFIG_IPV6_OPTIMISTIC_DAD is not set
+# CONFIG_INET6_AH is not set
+# CONFIG_INET6_ESP is not set
+# CONFIG_INET6_IPCOMP is not set
+# CONFIG_IPV6_MIP6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+CONFIG_INET6_XFRM_MODE_TRANSPORT=y
+CONFIG_INET6_XFRM_MODE_TUNNEL=y
+CONFIG_INET6_XFRM_MODE_BEET=y
+# CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set
+CONFIG_IPV6_SIT=y
+CONFIG_IPV6_NDISC_NODETYPE=y
+# CONFIG_IPV6_TUNNEL is not set
+# CONFIG_IPV6_MULTIPLE_TABLES is not set
+# CONFIG_IPV6_MROUTE is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+# CONFIG_IP_DCCP is not set
+CONFIG_IP_SCTP=m
+# CONFIG_SCTP_DBG_MSG is not set
+# CONFIG_SCTP_DBG_OBJCNT is not set
+# CONFIG_SCTP_HMAC_NONE is not set
+# CONFIG_SCTP_HMAC_SHA1 is not set
+CONFIG_SCTP_HMAC_MD5=y
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_CAN is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_AF_RXRPC is not set
+CONFIG_FIB_RULES=y
+
+#
+# Wireless
+#
+# CONFIG_CFG80211 is not set
+# CONFIG_WIRELESS_EXT is not set
+# CONFIG_MAC80211 is not set
+# CONFIG_IEEE80211 is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+# CONFIG_DEBUG_DRIVER is not set
+# CONFIG_DEBUG_DEVRES is not set
+# CONFIG_SYS_HYPERVISOR is not set
+# CONFIG_CONNECTOR is not set
+# CONFIG_MTD is not set
+CONFIG_OF_DEVICE=y
+CONFIG_OF_I2C=y
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_FD is not set
+# CONFIG_BLK_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+CONFIG_BLK_DEV_NBD=y
+# CONFIG_BLK_DEV_SX8 is not set
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=524288
+# CONFIG_BLK_DEV_XIP is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+CONFIG_MISC_DEVICES=y
+# CONFIG_PHANTOM is not set
+# CONFIG_EEPROM_93CX6 is not set
+# CONFIG_SGI_IOC4 is not set
+# CONFIG_TIFM_CORE is not set
+# CONFIG_ENCLOSURE_SERVICES is not set
+CONFIG_HAVE_IDE=y
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=y
+CONFIG_SCSI_DMA=y
+# CONFIG_SCSI_TGT is not set
+# CONFIG_SCSI_NETLINK is not set
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=y
+CONFIG_CHR_DEV_ST=y
+# CONFIG_CHR_DEV_OSST is not set
+CONFIG_BLK_DEV_SR=y
+# CONFIG_BLK_DEV_SR_VENDOR is not set
+CONFIG_CHR_DEV_SG=y
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+CONFIG_SCSI_MULTI_LUN=y
+# CONFIG_SCSI_CONSTANTS is not set
+CONFIG_SCSI_LOGGING=y
+# CONFIG_SCSI_SCAN_ASYNC is not set
+CONFIG_SCSI_WAIT_SCAN=m
+
+#
+# SCSI Transports
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_LIBSAS is not set
+# CONFIG_SCSI_SRP_ATTRS is not set
+CONFIG_SCSI_LOWLEVEL=y
+# CONFIG_ISCSI_TCP is not set
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_AIC94XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_SCSI_ADVANSYS is not set
+# CONFIG_SCSI_ARCMSR is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_HPTIOP is not set
+# CONFIG_SCSI_BUSLOGIC is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_EATA is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_GDTH is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_MVSAS is not set
+# CONFIG_SCSI_STEX is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+# CONFIG_SCSI_QLA_FC is not set
+# CONFIG_SCSI_QLA_ISCSI is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+# CONFIG_SCSI_SRP is not set
+CONFIG_ATA=y
+# CONFIG_ATA_NONSTANDARD is not set
+CONFIG_SATA_PMP=y
+# CONFIG_SATA_AHCI is not set
+CONFIG_SATA_SIL24=y
+CONFIG_SATA_FSL=y
+CONFIG_ATA_SFF=y
+# CONFIG_SATA_SVW is not set
+# CONFIG_ATA_PIIX is not set
+# CONFIG_SATA_MV is not set
+# CONFIG_SATA_NV is not set
+# CONFIG_PDC_ADMA is not set
+# CONFIG_SATA_QSTOR is not set
+# CONFIG_SATA_PROMISE is not set
+# CONFIG_SATA_SX4 is not set
+CONFIG_SATA_SIL=y
+# CONFIG_SATA_SIS is not set
+# CONFIG_SATA_ULI is not set
+# CONFIG_SATA_VIA is not set
+# CONFIG_SATA_VITESSE is not set
+# CONFIG_SATA_INIC162X is not set
+# CONFIG_PATA_ALI is not set
+# CONFIG_PATA_AMD is not set
+# CONFIG_PATA_ARTOP is not set
+# CONFIG_PATA_ATIIXP is not set
+# CONFIG_PATA_CMD640_PCI is not set
+# CONFIG_PATA_CMD64X is not set
+# CONFIG_PATA_CS5520 is not set
+# CONFIG_PATA_CS5530 is not set
+# CONFIG_PATA_CYPRESS is not set
+# CONFIG_PATA_EFAR is not set
+# CONFIG_ATA_GENERIC is not set
+# CONFIG_PATA_HPT366 is not set
+# CONFIG_PATA_HPT37X is not set
+# CONFIG_PATA_HPT3X2N is not set
+# CONFIG_PATA_HPT3X3 is not set
+# CONFIG_PATA_IT821X is not set
+# CONFIG_PATA_IT8213 is not set
+# CONFIG_PATA_JMICRON is not set
+# CONFIG_PATA_TRIFLEX is not set
+# CONFIG_PATA_MARVELL is not set
+# CONFIG_PATA_MPIIX is not set
+# CONFIG_PATA_OLDPIIX is not set
+# CONFIG_PATA_NETCELL is not set
+# CONFIG_PATA_NINJA32 is not set
+# CONFIG_PATA_NS87410 is not set
+# CONFIG_PATA_NS87415 is not set
+# CONFIG_PATA_OPTI is not set
+# CONFIG_PATA_OPTIDMA is not set
+# CONFIG_PATA_PDC_OLD is not set
+# CONFIG_PATA_RADISYS is not set
+# CONFIG_PATA_RZ1000 is not set
+# CONFIG_PATA_SC1200 is not set
+# CONFIG_PATA_SERVERWORKS is not set
+# CONFIG_PATA_PDC2027X is not set
+# CONFIG_PATA_SIL680 is not set
+# CONFIG_PATA_SIS is not set
+# CONFIG_PATA_VIA is not set
+# CONFIG_PATA_WINBOND is not set
+# CONFIG_PATA_PLATFORM is not set
+# CONFIG_PATA_SCH is not set
+# CONFIG_MD is not set
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# Enable only one of the two stacks, unless you know what you are doing
+#
+# CONFIG_FIREWIRE is not set
+# CONFIG_IEEE1394 is not set
+# CONFIG_I2O is not set
+# CONFIG_MACINTOSH_DRIVERS is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+CONFIG_DUMMY=y
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+# CONFIG_ARCNET is not set
+CONFIG_PHYLIB=y
+
+#
+# MII PHY device drivers
+#
+# CONFIG_MARVELL_PHY is not set
+# CONFIG_DAVICOM_PHY is not set
+# CONFIG_QSEMI_PHY is not set
+# CONFIG_LXT_PHY is not set
+# CONFIG_CICADA_PHY is not set
+CONFIG_VITESSE_PHY=y
+# CONFIG_SMSC_PHY is not set
+# CONFIG_BROADCOM_PHY is not set
+# CONFIG_ICPLUS_PHY is not set
+# CONFIG_REALTEK_PHY is not set
+# CONFIG_FIXED_PHY is not set
+# CONFIG_MDIO_BITBANG is not set
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+# CONFIG_CASSINI is not set
+# CONFIG_NET_VENDOR_3COM is not set
+# CONFIG_NET_TULIP is not set
+# CONFIG_HP100 is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_NET_PCI is not set
+# CONFIG_B44 is not set
+CONFIG_NETDEV_1000=y
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+# CONFIG_E1000 is not set
+# CONFIG_E1000E is not set
+# CONFIG_E1000E_ENABLED is not set
+# CONFIG_IP1000 is not set
+# CONFIG_IGB is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+# CONFIG_R8169 is not set
+# CONFIG_SIS190 is not set
+CONFIG_SKGE=y
+# CONFIG_SKGE_DEBUG is not set
+CONFIG_SKY2=y
+# CONFIG_SKY2_DEBUG is not set
+# CONFIG_VIA_VELOCITY is not set
+# CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
+CONFIG_GIANFAR=y
+CONFIG_GFAR_NAPI=y
+# CONFIG_QLA3XXX is not set
+# CONFIG_ATL1 is not set
+CONFIG_NETDEV_10000=y
+# CONFIG_CHELSIO_T1 is not set
+# CONFIG_CHELSIO_T3 is not set
+# CONFIG_IXGBE is not set
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+# CONFIG_MYRI10GE is not set
+# CONFIG_NETXEN_NIC is not set
+# CONFIG_NIU is not set
+# CONFIG_MLX4_CORE is not set
+# CONFIG_TEHUTI is not set
+# CONFIG_BNX2X is not set
+# CONFIG_SFC is not set
+# CONFIG_TR is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+# CONFIG_IWLWIFI_LEDS is not set
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+# CONFIG_INPUT_FF_MEMLESS is not set
+# CONFIG_INPUT_POLLDEV is not set
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TABLET is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+CONFIG_SERIO_I8042=y
+CONFIG_SERIO_SERPORT=y
+# CONFIG_SERIO_PCIPS2 is not set
+CONFIG_SERIO_LIBPS2=y
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_VT_HW_CONSOLE_BINDING is not set
+CONFIG_DEVKMEM=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+# CONFIG_NOZOMI is not set
+
+#
+# Serial drivers
+#
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_SERIAL_8250_PCI=y
+CONFIG_SERIAL_8250_NR_UARTS=2
+CONFIG_SERIAL_8250_RUNTIME_UARTS=2
+CONFIG_SERIAL_8250_EXTENDED=y
+CONFIG_SERIAL_8250_MANY_PORTS=y
+CONFIG_SERIAL_8250_SHARE_IRQ=y
+CONFIG_SERIAL_8250_DETECT_IRQ=y
+CONFIG_SERIAL_8250_RSA=y
+
+#
+# Non-8250 serial port support
+#
+# CONFIG_SERIAL_UARTLITE is not set
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+# CONFIG_SERIAL_OF_PLATFORM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_IPMI_HANDLER is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_NVRAM is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+CONFIG_DEVPORT=y
+CONFIG_I2C=y
+CONFIG_I2C_BOARDINFO=y
+# CONFIG_I2C_CHARDEV is not set
+
+#
+# I2C Hardware Bus support
+#
+# CONFIG_I2C_ALI1535 is not set
+# CONFIG_I2C_ALI1563 is not set
+# CONFIG_I2C_ALI15X3 is not set
+# CONFIG_I2C_AMD756 is not set
+# CONFIG_I2C_AMD8111 is not set
+# CONFIG_I2C_I801 is not set
+# CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
+CONFIG_I2C_MPC=y
+# CONFIG_I2C_NFORCE2 is not set
+# CONFIG_I2C_OCORES is not set
+# CONFIG_I2C_PARPORT_LIGHT is not set
+# CONFIG_I2C_PROSAVAGE is not set
+# CONFIG_I2C_SAVAGE4 is not set
+# CONFIG_I2C_SIMTEC is not set
+# CONFIG_I2C_SIS5595 is not set
+# CONFIG_I2C_SIS630 is not set
+# CONFIG_I2C_SIS96X is not set
+# CONFIG_I2C_TAOS_EVM is not set
+# CONFIG_I2C_STUB is not set
+# CONFIG_I2C_TINY_USB is not set
+# CONFIG_I2C_VIA is not set
+# CONFIG_I2C_VIAPRO is not set
+# CONFIG_I2C_VOODOO3 is not set
+# CONFIG_I2C_PCA_PLATFORM is not set
+
+#
+# Miscellaneous I2C Chip support
+#
+# CONFIG_DS1682 is not set
+CONFIG_SENSORS_EEPROM=y
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_PCF8575 is not set
+# CONFIG_SENSORS_PCF8591 is not set
+# CONFIG_SENSORS_MAX6875 is not set
+# CONFIG_SENSORS_TSL2550 is not set
+# CONFIG_I2C_DEBUG_CORE is not set
+# CONFIG_I2C_DEBUG_ALGO is not set
+# CONFIG_I2C_DEBUG_BUS is not set
+# CONFIG_I2C_DEBUG_CHIP is not set
+# CONFIG_SPI is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+# CONFIG_HWMON is not set
+CONFIG_THERMAL=y
+# CONFIG_WATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+# CONFIG_HTC_PASIC3 is not set
+
+#
+# Multimedia devices
+#
+
+#
+# Multimedia core support
+#
+# CONFIG_VIDEO_DEV is not set
+CONFIG_DVB_CORE=m
+CONFIG_VIDEO_MEDIA=m
+
+#
+# Multimedia drivers
+#
+# CONFIG_MEDIA_ATTACH is not set
+CONFIG_MEDIA_TUNER=m
+# CONFIG_MEDIA_TUNER_CUSTOMIZE is not set
+CONFIG_MEDIA_TUNER_SIMPLE=m
+CONFIG_MEDIA_TUNER_TDA8290=m
+CONFIG_MEDIA_TUNER_TDA9887=m
+CONFIG_MEDIA_TUNER_TEA5761=m
+CONFIG_MEDIA_TUNER_TEA5767=m
+CONFIG_MEDIA_TUNER_MT20XX=m
+CONFIG_MEDIA_TUNER_XC2028=m
+CONFIG_MEDIA_TUNER_XC5000=m
+CONFIG_DVB_CAPTURE_DRIVERS=y
+
+#
+# Supported SAA7146 based PCI Adapters
+#
+# CONFIG_TTPCI_EEPROM is not set
+# CONFIG_DVB_BUDGET_CORE is not set
+
+#
+# Supported USB Adapters
+#
+# CONFIG_DVB_USB is not set
+# CONFIG_DVB_TTUSB_BUDGET is not set
+# CONFIG_DVB_TTUSB_DEC is not set
+# CONFIG_DVB_CINERGYT2 is not set
+
+#
+# Supported FlexCopII (B2C2) Adapters
+#
+# CONFIG_DVB_B2C2_FLEXCOP is not set
+
+#
+# Supported BT878 Adapters
+#
+
+#
+# Supported Pluto2 Adapters
+#
+# CONFIG_DVB_PLUTO2 is not set
+
+#
+# Supported DVB Frontends
+#
+
+#
+# Customise DVB Frontends
+#
+# CONFIG_DVB_FE_CUSTOMISE is not set
+
+#
+# DVB-S (satellite) frontends
+#
+# CONFIG_DVB_CX24110 is not set
+# CONFIG_DVB_CX24123 is not set
+# CONFIG_DVB_MT312 is not set
+# CONFIG_DVB_S5H1420 is not set
+# CONFIG_DVB_STV0299 is not set
+# CONFIG_DVB_TDA8083 is not set
+# CONFIG_DVB_TDA10086 is not set
+# CONFIG_DVB_VES1X93 is not set
+# CONFIG_DVB_TUNER_ITD1000 is not set
+# CONFIG_DVB_TDA826X is not set
+# CONFIG_DVB_TUA6100 is not set
+
+#
+# DVB-T (terrestrial) frontends
+#
+# CONFIG_DVB_SP8870 is not set
+# CONFIG_DVB_SP887X is not set
+# CONFIG_DVB_CX22700 is not set
+# CONFIG_DVB_CX22702 is not set
+# CONFIG_DVB_L64781 is not set
+# CONFIG_DVB_TDA1004X is not set
+# CONFIG_DVB_NXT6000 is not set
+# CONFIG_DVB_MT352 is not set
+# CONFIG_DVB_ZL10353 is not set
+# CONFIG_DVB_DIB3000MB is not set
+# CONFIG_DVB_DIB3000MC is not set
+# CONFIG_DVB_DIB7000M is not set
+# CONFIG_DVB_DIB7000P is not set
+# CONFIG_DVB_TDA10048 is not set
+
+#
+# DVB-C (cable) frontends
+#
+# CONFIG_DVB_VES1820 is not set
+# CONFIG_DVB_TDA10021 is not set
+# CONFIG_DVB_TDA10023 is not set
+# CONFIG_DVB_STV0297 is not set
+
+#
+# ATSC (North American/Korean Terrestrial/Cable DTV) frontends
+#
+# CONFIG_DVB_NXT200X is not set
+# CONFIG_DVB_OR51211 is not set
+# CONFIG_DVB_OR51132 is not set
+# CONFIG_DVB_BCM3510 is not set
+# CONFIG_DVB_LGDT330X is not set
+# CONFIG_DVB_S5H1409 is not set
+# CONFIG_DVB_AU8522 is not set
+# CONFIG_DVB_S5H1411 is not set
+
+#
+# Digital terrestrial only tuners/PLL
+#
+# CONFIG_DVB_PLL is not set
+# CONFIG_DVB_TUNER_DIB0070 is not set
+
+#
+# SEC control devices for DVB-S
+#
+# CONFIG_DVB_LNBP21 is not set
+# CONFIG_DVB_ISL6405 is not set
+# CONFIG_DVB_ISL6421 is not set
+CONFIG_DAB=y
+# CONFIG_USB_DABUSB is not set
+
+#
+# Graphics support
+#
+# CONFIG_AGP is not set
+# CONFIG_DRM is not set
+# CONFIG_VGASTATE is not set
+CONFIG_VIDEO_OUTPUT_CONTROL=y
+# CONFIG_FB is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Console display driver support
+#
+CONFIG_VGA_CONSOLE=y
+# CONFIG_VGACON_SOFT_SCROLLBACK is not set
+CONFIG_DUMMY_CONSOLE=y
+
+#
+# Sound
+#
+CONFIG_SOUND=y
+
+#
+# Advanced Linux Sound Architecture
+#
+CONFIG_SND=y
+CONFIG_SND_TIMER=y
+CONFIG_SND_PCM=y
+# CONFIG_SND_SEQUENCER is not set
+# CONFIG_SND_MIXER_OSS is not set
+# CONFIG_SND_PCM_OSS is not set
+# CONFIG_SND_DYNAMIC_MINORS is not set
+CONFIG_SND_SUPPORT_OLD_API=y
+CONFIG_SND_VERBOSE_PROCFS=y
+# CONFIG_SND_VERBOSE_PRINTK is not set
+# CONFIG_SND_DEBUG is not set
+
+#
+# Generic devices
+#
+CONFIG_SND_AC97_CODEC=y
+# CONFIG_SND_DUMMY is not set
+# CONFIG_SND_MTPAV is not set
+# CONFIG_SND_SERIAL_U16550 is not set
+# CONFIG_SND_MPU401 is not set
+
+#
+# PCI devices
+#
+# CONFIG_SND_AD1889 is not set
+# CONFIG_SND_ALS300 is not set
+# CONFIG_SND_ALS4000 is not set
+# CONFIG_SND_ALI5451 is not set
+# CONFIG_SND_ATIIXP is not set
+# CONFIG_SND_ATIIXP_MODEM is not set
+# CONFIG_SND_AU8810 is not set
+# CONFIG_SND_AU8820 is not set
+# CONFIG_SND_AU8830 is not set
+# CONFIG_SND_AW2 is not set
+# CONFIG_SND_AZT3328 is not set
+# CONFIG_SND_BT87X is not set
+# CONFIG_SND_CA0106 is not set
+# CONFIG_SND_CMIPCI is not set
+# CONFIG_SND_OXYGEN is not set
+# CONFIG_SND_CS4281 is not set
+# CONFIG_SND_CS46XX is not set
+# CONFIG_SND_CS5530 is not set
+# CONFIG_SND_DARLA20 is not set
+# CONFIG_SND_GINA20 is not set
+# CONFIG_SND_LAYLA20 is not set
+# CONFIG_SND_DARLA24 is not set
+# CONFIG_SND_GINA24 is not set
+# CONFIG_SND_LAYLA24 is not set
+# CONFIG_SND_MONA is not set
+# CONFIG_SND_MIA is not set
+# CONFIG_SND_ECHO3G is not set
+# CONFIG_SND_INDIGO is not set
+# CONFIG_SND_INDIGOIO is not set
+# CONFIG_SND_INDIGODJ is not set
+# CONFIG_SND_EMU10K1 is not set
+# CONFIG_SND_EMU10K1X is not set
+# CONFIG_SND_ENS1370 is not set
+# CONFIG_SND_ENS1371 is not set
+# CONFIG_SND_ES1938 is not set
+# CONFIG_SND_ES1968 is not set
+# CONFIG_SND_FM801 is not set
+# CONFIG_SND_HDA_INTEL is not set
+# CONFIG_SND_HDSP is not set
+# CONFIG_SND_HDSPM is not set
+# CONFIG_SND_HIFIER is not set
+# CONFIG_SND_ICE1712 is not set
+# CONFIG_SND_ICE1724 is not set
+CONFIG_SND_INTEL8X0=y
+# CONFIG_SND_INTEL8X0M is not set
+# CONFIG_SND_KORG1212 is not set
+# CONFIG_SND_MAESTRO3 is not set
+# CONFIG_SND_MIXART is not set
+# CONFIG_SND_NM256 is not set
+# CONFIG_SND_PCXHR is not set
+# CONFIG_SND_RIPTIDE is not set
+# CONFIG_SND_RME32 is not set
+# CONFIG_SND_RME96 is not set
+# CONFIG_SND_RME9652 is not set
+# CONFIG_SND_SONICVIBES is not set
+# CONFIG_SND_TRIDENT is not set
+# CONFIG_SND_VIA82XX is not set
+# CONFIG_SND_VIA82XX_MODEM is not set
+# CONFIG_SND_VIRTUOSO is not set
+# CONFIG_SND_VX222 is not set
+# CONFIG_SND_YMFPCI is not set
+# CONFIG_SND_AC97_POWER_SAVE is not set
+
+#
+# ALSA PowerMac devices
+#
+
+#
+# ALSA PowerPC devices
+#
+
+#
+# USB devices
+#
+# CONFIG_SND_USB_AUDIO is not set
+# CONFIG_SND_USB_USX2Y is not set
+# CONFIG_SND_USB_CAIAQ is not set
+
+#
+# System on Chip audio support
+#
+# CONFIG_SND_SOC is not set
+
+#
+# ALSA SoC audio for Freescale SOCs
+#
+
+#
+# SoC Audio for the Texas Instruments OMAP
+#
+
+#
+# Open Sound System
+#
+# CONFIG_SOUND_PRIME is not set
+CONFIG_AC97_BUS=y
+CONFIG_HID_SUPPORT=y
+CONFIG_HID=y
+# CONFIG_HID_DEBUG is not set
+# CONFIG_HIDRAW is not set
+
+#
+# USB Input Devices
+#
+CONFIG_USB_HID=y
+# CONFIG_USB_HIDINPUT_POWERBOOK is not set
+# CONFIG_HID_FF is not set
+# CONFIG_USB_HIDDEV is not set
+CONFIG_USB_SUPPORT=y
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+CONFIG_USB_ARCH_HAS_EHCI=y
+CONFIG_USB=y
+# CONFIG_USB_DEBUG is not set
+# CONFIG_USB_ANNOUNCE_NEW_DEVICES is not set
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+CONFIG_USB_DEVICE_CLASS=y
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+# CONFIG_USB_OTG_WHITELIST is not set
+# CONFIG_USB_OTG_BLACKLIST_HUB is not set
+
+#
+# USB Host Controller Drivers
+#
+# CONFIG_USB_C67X00_HCD is not set
+CONFIG_USB_EHCI_HCD=y
+CONFIG_USB_EHCI_ROOT_HUB_TT=y
+# CONFIG_USB_EHCI_TT_NEWSCHED is not set
+CONFIG_USB_EHCI_FSL=y
+CONFIG_USB_EHCI_HCD_PPC_OF=y
+# CONFIG_USB_ISP116X_HCD is not set
+# CONFIG_USB_ISP1760_HCD is not set
+CONFIG_USB_OHCI_HCD=y
+CONFIG_USB_OHCI_HCD_PPC_OF=y
+CONFIG_USB_OHCI_HCD_PPC_OF_BE=y
+CONFIG_USB_OHCI_HCD_PPC_OF_LE=y
+CONFIG_USB_OHCI_HCD_PCI=y
+CONFIG_USB_OHCI_BIG_ENDIAN_DESC=y
+CONFIG_USB_OHCI_BIG_ENDIAN_MMIO=y
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
+# CONFIG_USB_UHCI_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+# CONFIG_USB_R8A66597_HCD is not set
+
+#
+# USB Device Class drivers
+#
+# CONFIG_USB_ACM is not set
+# CONFIG_USB_PRINTER is not set
+# CONFIG_USB_WDM is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# may also be needed; see USB_STORAGE Help for more information
+#
+CONFIG_USB_STORAGE=y
+# CONFIG_USB_STORAGE_DEBUG is not set
+# CONFIG_USB_STORAGE_DATAFAB is not set
+# CONFIG_USB_STORAGE_FREECOM is not set
+# CONFIG_USB_STORAGE_ISD200 is not set
+# CONFIG_USB_STORAGE_DPCM is not set
+# CONFIG_USB_STORAGE_USBAT is not set
+# CONFIG_USB_STORAGE_SDDR09 is not set
+# CONFIG_USB_STORAGE_SDDR55 is not set
+# CONFIG_USB_STORAGE_JUMPSHOT is not set
+# CONFIG_USB_STORAGE_ALAUDA is not set
+# CONFIG_USB_STORAGE_ONETOUCH is not set
+# CONFIG_USB_STORAGE_KARMA is not set
+# CONFIG_USB_STORAGE_CYPRESS_ATACB is not set
+# CONFIG_USB_LIBUSUAL is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+# CONFIG_USB_SERIAL is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_ADUTUX is not set
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_BERRY_CHARGE is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGET is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_FTDI_ELAN is not set
+# CONFIG_USB_APPLEDISPLAY is not set
+# CONFIG_USB_SISUSBVGA is not set
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TRANCEVIBRATOR is not set
+# CONFIG_USB_IOWARRIOR is not set
+# CONFIG_USB_TEST is not set
+# CONFIG_USB_ISIGHTFW is not set
+# CONFIG_USB_GADGET is not set
+# CONFIG_MMC is not set
+# CONFIG_MEMSTICK is not set
+# CONFIG_NEW_LEDS is not set
+# CONFIG_ACCESSIBILITY is not set
+# CONFIG_INFINIBAND is not set
+# CONFIG_EDAC is not set
+CONFIG_RTC_LIB=y
+CONFIG_RTC_CLASS=y
+CONFIG_RTC_HCTOSYS=y
+CONFIG_RTC_HCTOSYS_DEVICE="rtc0"
+# CONFIG_RTC_DEBUG is not set
+
+#
+# RTC interfaces
+#
+CONFIG_RTC_INTF_SYSFS=y
+CONFIG_RTC_INTF_PROC=y
+CONFIG_RTC_INTF_DEV=y
+# CONFIG_RTC_INTF_DEV_UIE_EMUL is not set
+# CONFIG_RTC_DRV_TEST is not set
+
+#
+# I2C RTC drivers
+#
+# CONFIG_RTC_DRV_DS1307 is not set
+# CONFIG_RTC_DRV_DS1374 is not set
+# CONFIG_RTC_DRV_DS1672 is not set
+# CONFIG_RTC_DRV_MAX6900 is not set
+# CONFIG_RTC_DRV_RS5C372 is not set
+# CONFIG_RTC_DRV_ISL1208 is not set
+# CONFIG_RTC_DRV_X1205 is not set
+# CONFIG_RTC_DRV_PCF8563 is not set
+# CONFIG_RTC_DRV_PCF8583 is not set
+# CONFIG_RTC_DRV_M41T80 is not set
+# CONFIG_RTC_DRV_S35390A is not set
+# CONFIG_RTC_DRV_FM3130 is not set
+
+#
+# SPI RTC drivers
+#
+
+#
+# Platform RTC drivers
+#
+CONFIG_RTC_DRV_CMOS=y
+# CONFIG_RTC_DRV_DS1511 is not set
+# CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_DS1742 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
+# CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
+# CONFIG_RTC_DRV_V3020 is not set
+
+#
+# on-CPU RTC drivers
+#
+# CONFIG_RTC_DRV_PPC is not set
+CONFIG_DMADEVICES=y
+
+#
+# DMA Devices
+#
+CONFIG_FSL_DMA=y
+CONFIG_DMA_ENGINE=y
+
+#
+# DMA Clients
+#
+# CONFIG_NET_DMA is not set
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+# CONFIG_EXT3_FS_POSIX_ACL is not set
+# CONFIG_EXT3_FS_SECURITY is not set
+# CONFIG_EXT4DEV_FS is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+CONFIG_DNOTIFY=y
+CONFIG_INOTIFY=y
+CONFIG_INOTIFY_USER=y
+# CONFIG_QUOTA is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+CONFIG_ISO9660_FS=m
+CONFIG_JOLIET=y
+CONFIG_ZISOFS=y
+CONFIG_UDF_FS=m
+CONFIG_UDF_NLS=y
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=y
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=y
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+CONFIG_NTFS_FS=y
+# CONFIG_NTFS_DEBUG is not set
+# CONFIG_NTFS_RW is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_TMPFS_POSIX_ACL is not set
+# CONFIG_HUGETLB_PAGE is not set
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+CONFIG_ADFS_FS=m
+# CONFIG_ADFS_FS_RW is not set
+CONFIG_AFFS_FS=m
+CONFIG_HFS_FS=m
+CONFIG_HFSPLUS_FS=m
+CONFIG_BEFS_FS=m
+# CONFIG_BEFS_DEBUG is not set
+CONFIG_BFS_FS=m
+CONFIG_EFS_FS=m
+CONFIG_CRAMFS=y
+CONFIG_VXFS_FS=m
+# CONFIG_MINIX_FS is not set
+CONFIG_HPFS_FS=m
+CONFIG_QNX4FS_FS=m
+# CONFIG_ROMFS_FS is not set
+CONFIG_SYSV_FS=m
+CONFIG_UFS_FS=m
+# CONFIG_UFS_FS_WRITE is not set
+# CONFIG_UFS_DEBUG is not set
+CONFIG_NETWORK_FILESYSTEMS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+CONFIG_NFS_V4=y
+CONFIG_NFSD=y
+# CONFIG_NFSD_V3 is not set
+# CONFIG_NFSD_V4 is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+CONFIG_SUNRPC_GSS=y
+# CONFIG_SUNRPC_BIND34 is not set
+CONFIG_RPCSEC_GSS_KRB5=y
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_ACORN_PARTITION is not set
+# CONFIG_OSF_PARTITION is not set
+# CONFIG_AMIGA_PARTITION is not set
+# CONFIG_ATARI_PARTITION is not set
+CONFIG_MAC_PARTITION=y
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_BSD_DISKLABEL is not set
+# CONFIG_MINIX_SUBPARTITION is not set
+# CONFIG_SOLARIS_X86_PARTITION is not set
+# CONFIG_UNIXWARE_DISKLABEL is not set
+# CONFIG_LDM_PARTITION is not set
+# CONFIG_SGI_PARTITION is not set
+# CONFIG_ULTRIX_PARTITION is not set
+# CONFIG_SUN_PARTITION is not set
+# CONFIG_KARMA_PARTITION is not set
+# CONFIG_EFI_PARTITION is not set
+# CONFIG_SYSV68_PARTITION is not set
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-1"
+# CONFIG_NLS_CODEPAGE_437 is not set
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+# CONFIG_NLS_ISO8859_1 is not set
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+CONFIG_NLS_UTF8=m
+# CONFIG_DLM is not set
+
+#
+# Library routines
+#
+CONFIG_BITREVERSE=y
+# CONFIG_GENERIC_FIND_FIRST_BIT is not set
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC_ITU_T=m
+CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
+CONFIG_LIBCRC32C=m
+CONFIG_ZLIB_INFLATE=y
+CONFIG_PLIST=y
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_IOPORT=y
+CONFIG_HAS_DMA=y
+CONFIG_HAVE_LMB=y
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_FRAME_WARN=1024
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_HEADERS_CHECK is not set
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_DEBUG_SHIRQ is not set
+CONFIG_DETECT_SOFTLOCKUP=y
+CONFIG_SCHED_DEBUG=y
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_TIMER_STATS is not set
+# CONFIG_DEBUG_OBJECTS is not set
+# CONFIG_SLUB_DEBUG_ON is not set
+# CONFIG_SLUB_STATS is not set
+# CONFIG_DEBUG_RT_MUTEXES is not set
+# CONFIG_RT_MUTEX_TESTER is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_MUTEXES is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
+# CONFIG_DEBUG_KOBJECT is not set
+# CONFIG_DEBUG_HIGHMEM is not set
+# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_INFO=y
+# CONFIG_DEBUG_VM is not set
+# CONFIG_DEBUG_WRITECOUNT is not set
+# CONFIG_DEBUG_LIST is not set
+# CONFIG_DEBUG_SG is not set
+# CONFIG_BOOT_PRINTK_DELAY is not set
+# CONFIG_RCU_TORTURE_TEST is not set
+# CONFIG_BACKTRACE_SELF_TEST is not set
+# CONFIG_FAULT_INJECTION is not set
+# CONFIG_SAMPLES is not set
+# CONFIG_DEBUG_STACKOVERFLOW is not set
+# CONFIG_DEBUG_STACK_USAGE is not set
+# CONFIG_DEBUG_PAGEALLOC is not set
+# CONFIG_DEBUGGER is not set
+# CONFIG_CODE_PATCHING_SELFTEST is not set
+# CONFIG_FTR_FIXUP_SELFTEST is not set
+# CONFIG_IRQSTACKS is not set
+# CONFIG_VIRQ_DEBUG is not set
+# CONFIG_BDI_SWITCH is not set
+# CONFIG_PPC_EARLY_DEBUG is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+CONFIG_CRYPTO=y
+
+#
+# Crypto core or helper
+#
+CONFIG_CRYPTO_ALGAPI=y
+CONFIG_CRYPTO_BLKCIPHER=y
+CONFIG_CRYPTO_HASH=y
+CONFIG_CRYPTO_MANAGER=y
+# CONFIG_CRYPTO_GF128MUL is not set
+# CONFIG_CRYPTO_NULL is not set
+# CONFIG_CRYPTO_CRYPTD is not set
+# CONFIG_CRYPTO_AUTHENC is not set
+# CONFIG_CRYPTO_TEST is not set
+
+#
+# Authenticated Encryption with Associated Data
+#
+# CONFIG_CRYPTO_CCM is not set
+# CONFIG_CRYPTO_GCM is not set
+# CONFIG_CRYPTO_SEQIV is not set
+
+#
+# Block modes
+#
+CONFIG_CRYPTO_CBC=y
+# CONFIG_CRYPTO_CTR is not set
+# CONFIG_CRYPTO_CTS is not set
+# CONFIG_CRYPTO_ECB is not set
+# CONFIG_CRYPTO_LRW is not set
+CONFIG_CRYPTO_PCBC=m
+# CONFIG_CRYPTO_XTS is not set
+
+#
+# Hash modes
+#
+CONFIG_CRYPTO_HMAC=y
+# CONFIG_CRYPTO_XCBC is not set
+
+#
+# Digest
+#
+# CONFIG_CRYPTO_CRC32C is not set
+# CONFIG_CRYPTO_MD4 is not set
+CONFIG_CRYPTO_MD5=y
+# CONFIG_CRYPTO_MICHAEL_MIC is not set
+CONFIG_CRYPTO_SHA1=m
+# CONFIG_CRYPTO_SHA256 is not set
+# CONFIG_CRYPTO_SHA512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
+# CONFIG_CRYPTO_WP512 is not set
+
+#
+# Ciphers
+#
+# CONFIG_CRYPTO_AES is not set
+# CONFIG_CRYPTO_ANUBIS is not set
+# CONFIG_CRYPTO_ARC4 is not set
+# CONFIG_CRYPTO_BLOWFISH is not set
+# CONFIG_CRYPTO_CAMELLIA is not set
+# CONFIG_CRYPTO_CAST5 is not set
+# CONFIG_CRYPTO_CAST6 is not set
+CONFIG_CRYPTO_DES=y
+# CONFIG_CRYPTO_FCRYPT is not set
+# CONFIG_CRYPTO_KHAZAD is not set
+# CONFIG_CRYPTO_SALSA20 is not set
+# CONFIG_CRYPTO_SEED is not set
+# CONFIG_CRYPTO_SERPENT is not set
+# CONFIG_CRYPTO_TEA is not set
+# CONFIG_CRYPTO_TWOFISH is not set
+
+#
+# Compression
+#
+# CONFIG_CRYPTO_DEFLATE is not set
+# CONFIG_CRYPTO_LZO is not set
+CONFIG_CRYPTO_HW=y
+# CONFIG_CRYPTO_DEV_HIFN_795X is not set
+# CONFIG_PPC_CLOCK is not set
+# CONFIG_VIRTUALIZATION is not set
diff --git a/arch/powerpc/platforms/85xx/Kconfig b/arch/powerpc/platforms/85xx/Kconfig
index 9cb8e29987a3..cebea5cadbc1 100644
--- a/arch/powerpc/platforms/85xx/Kconfig
+++ b/arch/powerpc/platforms/85xx/Kconfig
@@ -38,6 +38,12 @@ config MPC85xx_MDS
 	help
 	  This option enables support for the MPC85xx MDS board
 
+config MPC8536_DS
+	bool "Freescale MPC8536 DS"
+	select DEFAULT_UIMAGE
+	help
+	  This option enables support for the MPC8536 DS board
+
 config MPC85xx_DS
 	bool "Freescale MPC85xx DS"
 	select PPC_I8259
diff --git a/arch/powerpc/platforms/85xx/Makefile b/arch/powerpc/platforms/85xx/Makefile
index 6cea185f62b2..cb3054e1001d 100644
--- a/arch/powerpc/platforms/85xx/Makefile
+++ b/arch/powerpc/platforms/85xx/Makefile
@@ -4,6 +4,7 @@
 obj-$(CONFIG_MPC8540_ADS) += mpc85xx_ads.o
 obj-$(CONFIG_MPC8560_ADS) += mpc85xx_ads.o
 obj-$(CONFIG_MPC85xx_CDS) += mpc85xx_cds.o
+obj-$(CONFIG_MPC8536_DS)  += mpc8536_ds.o
 obj-$(CONFIG_MPC85xx_DS)  += mpc85xx_ds.o
 obj-$(CONFIG_MPC85xx_MDS) += mpc85xx_mds.o
 obj-$(CONFIG_STX_GP3)	  += stx_gp3.o
diff --git a/arch/powerpc/platforms/85xx/mpc8536_ds.c b/arch/powerpc/platforms/85xx/mpc8536_ds.c
new file mode 100644
index 000000000000..71d7c038961f
--- /dev/null
+++ b/arch/powerpc/platforms/85xx/mpc8536_ds.c
@@ -0,0 +1,125 @@
+/*
+ * MPC8536 DS Board Setup
+ *
+ * Copyright 2008 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/kdev_t.h>
+#include <linux/delay.h>
+#include <linux/seq_file.h>
+#include <linux/interrupt.h>
+#include <linux/of_platform.h>
+
+#include <asm/system.h>
+#include <asm/time.h>
+#include <asm/machdep.h>
+#include <asm/pci-bridge.h>
+#include <mm/mmu_decl.h>
+#include <asm/prom.h>
+#include <asm/udbg.h>
+#include <asm/mpic.h>
+
+#include <sysdev/fsl_soc.h>
+#include <sysdev/fsl_pci.h>
+
+void __init mpc8536_ds_pic_init(void)
+{
+	struct mpic *mpic;
+	struct resource r;
+	struct device_node *np;
+
+	np = of_find_node_by_type(np, "open-pic");
+	if (np == NULL) {
+		printk(KERN_ERR "Could not find open-pic node\n");
+		return;
+	}
+
+	if (of_address_to_resource(np, 0, &r)) {
+		printk(KERN_ERR "Failed to map mpic register space\n");
+		of_node_put(np);
+		return;
+	}
+
+	mpic = mpic_alloc(np, r.start,
+			  MPIC_PRIMARY | MPIC_WANTS_RESET |
+			  MPIC_BIG_ENDIAN | MPIC_BROKEN_FRR_NIRQS,
+			0, 256, " OpenPIC  ");
+	BUG_ON(mpic == NULL);
+	of_node_put(np);
+
+	mpic_init(mpic);
+}
+
+/*
+ * Setup the architecture
+ */
+static void __init mpc8536_ds_setup_arch(void)
+{
+#ifdef CONFIG_PCI
+	struct device_node *np;
+#endif
+
+	if (ppc_md.progress)
+		ppc_md.progress("mpc8536_ds_setup_arch()", 0);
+
+#ifdef CONFIG_PCI
+	for_each_node_by_type(np, "pci") {
+		if (of_device_is_compatible(np, "fsl,mpc8540-pci") ||
+		    of_device_is_compatible(np, "fsl,mpc8548-pcie")) {
+			struct resource rsrc;
+			of_address_to_resource(np, 0, &rsrc);
+			if ((rsrc.start & 0xfffff) == 0x8000)
+				fsl_add_bridge(np, 1);
+			else
+				fsl_add_bridge(np, 0);
+		}
+	}
+
+#endif
+
+	printk("MPC8536 DS board from Freescale Semiconductor\n");
+}
+
+static struct of_device_id __initdata mpc8536_ds_ids[] = {
+	{ .type = "soc", },
+	{ .compatible = "soc", },
+	{},
+};
+
+static int __init mpc8536_ds_publish_devices(void)
+{
+	return of_platform_bus_probe(NULL, mpc8536_ds_ids, NULL);
+}
+machine_device_initcall(mpc8536_ds, mpc8536_ds_publish_devices);
+
+/*
+ * Called very early, device-tree isn't unflattened
+ */
+static int __init mpc8536_ds_probe(void)
+{
+	unsigned long root = of_get_flat_dt_root();
+
+	return of_flat_dt_is_compatible(root, "fsl,mpc8536ds");
+}
+
+define_machine(mpc8536_ds) {
+	.name			= "MPC8536 DS",
+	.probe			= mpc8536_ds_probe,
+	.setup_arch		= mpc8536_ds_setup_arch,
+	.init_IRQ		= mpc8536_ds_pic_init,
+#ifdef CONFIG_PCI
+	.pcibios_fixup_bus	= fsl_pcibios_fixup_bus,
+#endif
+	.get_irq		= mpic_get_irq,
+	.restart		= fsl_rstcr_restart,
+	.calibrate_decr		= generic_calibrate_decr,
+	.progress		= udbg_progress,
+};
diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c
index 489ca5a397b1..87b0aa13ab48 100644
--- a/arch/powerpc/sysdev/fsl_pci.c
+++ b/arch/powerpc/sysdev/fsl_pci.c
@@ -243,6 +243,8 @@ DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8544E, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8544, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8572E, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8572, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8536E, quirk_fsl_pcie_header);
+DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8536, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8641, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8641D, quirk_fsl_pcie_header);
 DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_MPC8610, quirk_fsl_pcie_header);
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 65953822c9cb..1cf4084b51e8 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2171,6 +2171,8 @@
 #define PCI_DEVICE_ID_MPC8544		0x0033
 #define PCI_DEVICE_ID_MPC8572E		0x0040
 #define PCI_DEVICE_ID_MPC8572		0x0041
+#define PCI_DEVICE_ID_MPC8536E		0x0050
+#define PCI_DEVICE_ID_MPC8536		0x0051
 #define PCI_DEVICE_ID_MPC8641		0x7010
 #define PCI_DEVICE_ID_MPC8641D		0x7011
 #define PCI_DEVICE_ID_MPC8610		0x7018
-- 
cgit v1.2.3


From 4ee6afd34409d296782a5b667d7991b1050e910a Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 7 May 2008 21:01:30 +0300
Subject: VFS: export sync_sb_inodes

This patch exports the 'sync_sb_inodes()' which is needed for
UBIFS because it has to force write-back from time to time.
Namely, the UBIFS budgeting subsystem forces write-back when
its pessimistic callculations show that there is no free
space on the media.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 fs/fs-writeback.c  | 11 +++++++++--
 include/linux/fs.h |  2 ++
 2 files changed, 11 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 16519fe1399c..25adfc3c693a 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -439,8 +439,8 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
  * on the writer throttling path, and we get decent balancing between many
  * throttled threads: we don't want them all piling up on inode_sync_wait.
  */
-static void
-sync_sb_inodes(struct super_block *sb, struct writeback_control *wbc)
+void generic_sync_sb_inodes(struct super_block *sb,
+				struct writeback_control *wbc)
 {
 	const unsigned long start = jiffies;	/* livelock avoidance */
 
@@ -526,6 +526,13 @@ sync_sb_inodes(struct super_block *sb, struct writeback_control *wbc)
 	spin_unlock(&inode_lock);
 	return;		/* Leave any unwritten inodes on s_io */
 }
+EXPORT_SYMBOL_GPL(generic_sync_sb_inodes);
+
+static void sync_sb_inodes(struct super_block *sb,
+				struct writeback_control *wbc)
+{
+	generic_sync_sb_inodes(sb, wbc);
+}
 
 /*
  * Start writeback of dirty pagecache data against all unlocked inodes.
diff --git a/include/linux/fs.h b/include/linux/fs.h
index d8e2762ed14d..f9d2aab47eda 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1729,6 +1729,8 @@ static inline void invalidate_remote_inode(struct inode *inode)
 extern int invalidate_inode_pages2(struct address_space *mapping);
 extern int invalidate_inode_pages2_range(struct address_space *mapping,
 					 pgoff_t start, pgoff_t end);
+extern void generic_sync_sb_inodes(struct super_block *sb,
+				struct writeback_control *wbc);
 extern int write_inode_now(struct inode *, int);
 extern int filemap_fdatawrite(struct address_space *);
 extern int filemap_flush(struct address_space *);
-- 
cgit v1.2.3


From 9c0c7a429a0cf02c2ac1998d5cf4c26f6be5c989 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Wed, 9 Jul 2008 15:48:45 +0200
Subject: ssb: Include dma-mapping.h

ssb.h implements DMA mapping functions, so it should
include dma-mapping.h. This fixes compile failures on certain architectures.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/ssb/ssb.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 0fe5a0ded3ea..4bf8cade9dbc 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -7,6 +7,7 @@
 #include <linux/spinlock.h>
 #include <linux/pci.h>
 #include <linux/mod_devicetable.h>
+#include <linux/dma-mapping.h>
 
 #include <linux/ssb/ssb_regs.h>
 
-- 
cgit v1.2.3


From 341c2c958ec7bdd9f54733a8b0b432fe76842a82 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 20 May 2008 02:17:51 +0900
Subject: libata: consistently use msecs for time durations

libata has been using mix of jiffies and msecs for time druations.
This is getting confusing.  As writing sub HZ values in jiffies is
PITA and msecs_to_jiffies() can't be used as initializer, unify unit
for all time durations to msecs.  So, durations are in msecs and
deadlines are in jiffies.  ata_deadline() is added to compute deadline
from a start time and duration in msecs.

While at it, drop now superflous _msec suffix from arguments and
rename @timeout to @deadline if it represents a fixed point in time
rather than duration.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 44 +++++++++++++++++++++-----------------------
 drivers/ata/libata-eh.c   | 33 +++++++++++++++++----------------
 drivers/ata/libata-pmp.c  |  3 ++-
 drivers/ata/libata-sff.c  | 15 ++++++++-------
 drivers/ata/pata_bf54x.c  |  6 +++---
 drivers/ata/pata_scc.c    |  2 +-
 include/linux/libata.h    | 26 ++++++++++++++++----------
 7 files changed, 68 insertions(+), 61 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 303fc0d2b978..c5c3b1b516e1 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -54,7 +54,6 @@
 #include <linux/completion.h>
 #include <linux/suspend.h>
 #include <linux/workqueue.h>
-#include <linux/jiffies.h>
 #include <linux/scatterlist.h>
 #include <linux/io.h>
 #include <scsi/scsi.h>
@@ -145,7 +144,7 @@ static int libata_dma_mask = ATA_DMA_MASK_ATA|ATA_DMA_MASK_ATAPI|ATA_DMA_MASK_CF
 module_param_named(dma, libata_dma_mask, int, 0444);
 MODULE_PARM_DESC(dma, "DMA enable/disable (0x1==ATA, 0x2==ATAPI, 0x4==CF)");
 
-static int ata_probe_timeout = ATA_TMOUT_INTERNAL / HZ;
+static int ata_probe_timeout = ATA_TMOUT_INTERNAL / 1000;
 module_param(ata_probe_timeout, int, 0444);
 MODULE_PARM_DESC(ata_probe_timeout, "Set ATA probing timeout (seconds)");
 
@@ -1533,7 +1532,7 @@ unsigned long ata_id_xfermask(const u16 *id)
  *	@ap: The ata_port to queue port_task for
  *	@fn: workqueue function to be scheduled
  *	@data: data for @fn to use
- *	@delay: delay time for workqueue function
+ *	@delay: delay time in msecs for workqueue function
  *
  *	Schedule @fn(@data) for execution after @delay jiffies using
  *	port_task.  There is one port_task per port and it's the
@@ -1552,7 +1551,7 @@ void ata_pio_queue_task(struct ata_port *ap, void *data, unsigned long delay)
 	ap->port_task_data = data;
 
 	/* may fail if ata_port_flush_task() in progress */
-	queue_delayed_work(ata_wq, &ap->port_task, delay);
+	queue_delayed_work(ata_wq, &ap->port_task, msecs_to_jiffies(delay));
 }
 
 /**
@@ -1685,7 +1684,7 @@ unsigned ata_exec_internal_sg(struct ata_device *dev,
 	spin_unlock_irqrestore(ap->lock, flags);
 
 	if (!timeout)
-		timeout = ata_probe_timeout * 1000 / HZ;
+		timeout = ata_probe_timeout * 1000;
 
 	rc = wait_for_completion_timeout(&wait, msecs_to_jiffies(timeout));
 
@@ -3319,7 +3318,7 @@ int ata_wait_ready(struct ata_link *link, unsigned long deadline,
 		   int (*check_ready)(struct ata_link *link))
 {
 	unsigned long start = jiffies;
-	unsigned long nodev_deadline = start + ATA_TMOUT_FF_WAIT;
+	unsigned long nodev_deadline = ata_deadline(start, ATA_TMOUT_FF_WAIT);
 	int warned = 0;
 
 	if (time_after(nodev_deadline, deadline))
@@ -3387,7 +3386,7 @@ int ata_wait_ready(struct ata_link *link, unsigned long deadline,
 int ata_wait_after_reset(struct ata_link *link, unsigned long deadline,
 				int (*check_ready)(struct ata_link *link))
 {
-	msleep(ATA_WAIT_AFTER_RESET_MSECS);
+	msleep(ATA_WAIT_AFTER_RESET);
 
 	return ata_wait_ready(link, deadline, check_ready);
 }
@@ -3417,13 +3416,13 @@ int ata_wait_after_reset(struct ata_link *link, unsigned long deadline,
 int sata_link_debounce(struct ata_link *link, const unsigned long *params,
 		       unsigned long deadline)
 {
-	unsigned long interval_msec = params[0];
-	unsigned long duration = msecs_to_jiffies(params[1]);
+	unsigned long interval = params[0];
+	unsigned long duration = params[1];
 	unsigned long last_jiffies, t;
 	u32 last, cur;
 	int rc;
 
-	t = jiffies + msecs_to_jiffies(params[2]);
+	t = ata_deadline(jiffies, params[2]);
 	if (time_before(t, deadline))
 		deadline = t;
 
@@ -3435,7 +3434,7 @@ int sata_link_debounce(struct ata_link *link, const unsigned long *params,
 	last_jiffies = jiffies;
 
 	while (1) {
-		msleep(interval_msec);
+		msleep(interval);
 		if ((rc = sata_scr_read(link, SCR_STATUS, &cur)))
 			return rc;
 		cur &= 0xf;
@@ -3444,7 +3443,8 @@ int sata_link_debounce(struct ata_link *link, const unsigned long *params,
 		if (cur == last) {
 			if (cur == 1 && time_before(jiffies, deadline))
 				continue;
-			if (time_after(jiffies, last_jiffies + duration))
+			if (time_after(jiffies,
+				       ata_deadline(last_jiffies, duration)))
 				return 0;
 			continue;
 		}
@@ -3636,7 +3636,8 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 		if (check_ready) {
 			unsigned long pmp_deadline;
 
-			pmp_deadline = jiffies + ATA_TMOUT_PMP_SRST_WAIT;
+			pmp_deadline = ata_deadline(jiffies,
+						    ATA_TMOUT_PMP_SRST_WAIT);
 			if (time_after(pmp_deadline, deadline))
 				pmp_deadline = deadline;
 			ata_wait_ready(link, pmp_deadline, check_ready);
@@ -6073,8 +6074,6 @@ static void __init ata_parse_force_param(void)
 
 static int __init ata_init(void)
 {
-	ata_probe_timeout *= HZ;
-
 	ata_parse_force_param();
 
 	ata_wq = create_workqueue("ata");
@@ -6127,8 +6126,8 @@ int ata_ratelimit(void)
  *	@reg: IO-mapped register
  *	@mask: Mask to apply to read register value
  *	@val: Wait condition
- *	@interval_msec: polling interval in milliseconds
- *	@timeout_msec: timeout in milliseconds
+ *	@interval: polling interval in milliseconds
+ *	@timeout: timeout in milliseconds
  *
  *	Waiting for some bits of register to change is a common
  *	operation for ATA controllers.  This function reads 32bit LE
@@ -6146,10 +6145,9 @@ int ata_ratelimit(void)
  *	The final register value.
  */
 u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
-		      unsigned long interval_msec,
-		      unsigned long timeout_msec)
+		      unsigned long interval, unsigned long timeout)
 {
-	unsigned long timeout;
+	unsigned long deadline;
 	u32 tmp;
 
 	tmp = ioread32(reg);
@@ -6158,10 +6156,10 @@ u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
 	 * preceding writes reach the controller before starting to
 	 * eat away the timeout.
 	 */
-	timeout = jiffies + (timeout_msec * HZ) / 1000;
+	deadline = ata_deadline(jiffies, timeout);
 
-	while ((tmp & mask) == val && time_before(jiffies, timeout)) {
-		msleep(interval_msec);
+	while ((tmp & mask) == val && time_before(jiffies, deadline)) {
+		msleep(interval);
 		tmp = ioread32(reg);
 	}
 
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 7894d83ea1eb..08dd07f10008 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -66,15 +66,14 @@ enum {
 	ATA_ECAT_DUBIOUS_TOUT_HSM	= 6,
 	ATA_ECAT_DUBIOUS_UNK_DEV	= 7,
 	ATA_ECAT_NR			= 8,
-};
 
-/* Waiting in ->prereset can never be reliable.  It's sometimes nice
- * to wait there but it can't be depended upon; otherwise, we wouldn't
- * be resetting.  Just give it enough time for most drives to spin up.
- */
-enum {
-	ATA_EH_PRERESET_TIMEOUT		= 10 * HZ,
-	ATA_EH_FASTDRAIN_INTERVAL	= 3 * HZ,
+	/* Waiting in ->prereset can never be reliable.  It's
+	 * sometimes nice to wait there but it can't be depended upon;
+	 * otherwise, we wouldn't be resetting.  Just give it enough
+	 * time for most drives to spin up.
+	 */
+	ATA_EH_PRERESET_TIMEOUT		= 10000,
+	ATA_EH_FASTDRAIN_INTERVAL	=  3000,
 };
 
 /* The following table determines how we sequence resets.  Each entry
@@ -84,10 +83,10 @@ enum {
  * are mostly for error handling, hotplug and retarded devices.
  */
 static const unsigned long ata_eh_reset_timeouts[] = {
-	10 * HZ,	/* most drives spin up by 10sec */
-	10 * HZ,	/* > 99% working drives spin up before 20sec */
-	35 * HZ,	/* give > 30 secs of idleness for retarded devices */
-	5 * HZ,		/* and sweet one last chance */
+	10000,	/* most drives spin up by 10sec */
+	10000,	/* > 99% working drives spin up before 20sec */
+	35000,	/* give > 30 secs of idleness for retarded devices */
+	 5000,	/* and sweet one last chance */
 	/* > 1 min has elapsed, give up */
 };
 
@@ -641,7 +640,7 @@ void ata_eh_fastdrain_timerfn(unsigned long arg)
 		/* some qcs have finished, give it another chance */
 		ap->fastdrain_cnt = cnt;
 		ap->fastdrain_timer.expires =
-			jiffies + ATA_EH_FASTDRAIN_INTERVAL;
+			ata_deadline(jiffies, ATA_EH_FASTDRAIN_INTERVAL);
 		add_timer(&ap->fastdrain_timer);
 	}
 
@@ -681,7 +680,8 @@ static void ata_eh_set_pending(struct ata_port *ap, int fastdrain)
 
 	/* activate fast drain */
 	ap->fastdrain_cnt = cnt;
-	ap->fastdrain_timer.expires = jiffies + ATA_EH_FASTDRAIN_INTERVAL;
+	ap->fastdrain_timer.expires =
+		ata_deadline(jiffies, ATA_EH_FASTDRAIN_INTERVAL);
 	add_timer(&ap->fastdrain_timer);
 }
 
@@ -2125,7 +2125,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	}
 
 	if (prereset) {
-		rc = prereset(link, jiffies + ATA_EH_PRERESET_TIMEOUT);
+		rc = prereset(link,
+			      ata_deadline(jiffies, ATA_EH_PRERESET_TIMEOUT));
 		if (rc) {
 			if (rc == -ENOENT) {
 				ata_link_printk(link, KERN_DEBUG,
@@ -2160,7 +2161,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	if (ata_is_host_link(link))
 		ata_eh_freeze_port(ap);
 
-	deadline = jiffies + ata_eh_reset_timeouts[try++];
+	deadline = ata_deadline(jiffies, ata_eh_reset_timeouts[try++]);
 
 	if (reset) {
 		if (verbose)
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 7daf4c0f6216..63691d77ac43 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -785,7 +785,8 @@ static int sata_pmp_eh_handle_disabled_links(struct ata_port *ap)
 		 * SError.N working.
 		 */
 		sata_link_hardreset(link, sata_deb_timing_normal,
-				jiffies + ATA_TMOUT_INTERNAL_QUICK, NULL, NULL);
+				ata_deadline(jiffies, ATA_TMOUT_INTERNAL_QUICK),
+				NULL, NULL);
 
 		/* unconditionally clear SError.N */
 		rc = sata_scr_write(link, SCR_ERROR, SERR_PHYRDY_CHG);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index c0908c225483..304fdc6f1dc2 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -345,8 +345,8 @@ void ata_sff_dma_pause(struct ata_port *ap)
 /**
  *	ata_sff_busy_sleep - sleep until BSY clears, or timeout
  *	@ap: port containing status register to be polled
- *	@tmout_pat: impatience timeout
- *	@tmout: overall timeout
+ *	@tmout_pat: impatience timeout in msecs
+ *	@tmout: overall timeout in msecs
  *
  *	Sleep until ATA Status register bit BSY clears,
  *	or a timeout occurs.
@@ -365,7 +365,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 
 	status = ata_sff_busy_wait(ap, ATA_BUSY, 300);
 	timer_start = jiffies;
-	timeout = timer_start + tmout_pat;
+	timeout = ata_deadline(timer_start, tmout_pat);
 	while (status != 0xff && (status & ATA_BUSY) &&
 	       time_before(jiffies, timeout)) {
 		msleep(50);
@@ -377,7 +377,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 				"port is slow to respond, please be patient "
 				"(Status 0x%x)\n", status);
 
-	timeout = timer_start + tmout;
+	timeout = ata_deadline(timer_start, tmout);
 	while (status != 0xff && (status & ATA_BUSY) &&
 	       time_before(jiffies, timeout)) {
 		msleep(50);
@@ -390,7 +390,7 @@ int ata_sff_busy_sleep(struct ata_port *ap,
 	if (status & ATA_BUSY) {
 		ata_port_printk(ap, KERN_ERR, "port failed to respond "
 				"(%lu secs, Status 0x%x)\n",
-				tmout / HZ, status);
+				DIV_ROUND_UP(tmout, 1000), status);
 		return -EBUSY;
 	}
 
@@ -1888,7 +1888,7 @@ int ata_sff_wait_after_reset(struct ata_link *link, unsigned int devmask,
 	unsigned int dev1 = devmask & (1 << 1);
 	int rc, ret = 0;
 
-	msleep(ATA_WAIT_AFTER_RESET_MSECS);
+	msleep(ATA_WAIT_AFTER_RESET);
 
 	/* always check readiness of the master device */
 	rc = ata_sff_wait_ready(link, deadline);
@@ -2371,7 +2371,8 @@ void ata_bus_reset(struct ata_port *ap)
 
 	/* issue bus reset */
 	if (ap->flags & ATA_FLAG_SRST) {
-		rc = ata_bus_softreset(ap, devmask, jiffies + 40 * HZ);
+		rc = ata_bus_softreset(ap, devmask,
+				       ata_deadline(jiffies, 40000));
 		if (rc && rc != -ENODEV)
 			goto err_out;
 	}
diff --git a/drivers/ata/pata_bf54x.c b/drivers/ata/pata_bf54x.c
index 55516103626a..d3932901a3b3 100644
--- a/drivers/ata/pata_bf54x.c
+++ b/drivers/ata/pata_bf54x.c
@@ -1011,7 +1011,7 @@ static void bfin_bus_post_reset(struct ata_port *ap, unsigned int devmask)
 	void __iomem *base = (void __iomem *)ap->ioaddr.ctl_addr;
 	unsigned int dev0 = devmask & (1 << 0);
 	unsigned int dev1 = devmask & (1 << 1);
-	unsigned long timeout;
+	unsigned long deadline;
 
 	/* if device 0 was found in ata_devchk, wait for its
 	 * BSY bit to clear
@@ -1022,7 +1022,7 @@ static void bfin_bus_post_reset(struct ata_port *ap, unsigned int devmask)
 	/* if device 1 was found in ata_devchk, wait for
 	 * register access, then wait for BSY to clear
 	 */
-	timeout = jiffies + ATA_TMOUT_BOOT;
+	deadline = ata_deadline(jiffies, ATA_TMOUT_BOOT);
 	while (dev1) {
 		u8 nsect, lbal;
 
@@ -1031,7 +1031,7 @@ static void bfin_bus_post_reset(struct ata_port *ap, unsigned int devmask)
 		lbal = read_atapi_register(base, ATA_REG_LBAL);
 		if ((nsect == 1) && (lbal == 1))
 			break;
-		if (time_after(jiffies, timeout)) {
+		if (time_after(jiffies, deadline)) {
 			dev1 = 0;
 			break;
 		}
diff --git a/drivers/ata/pata_scc.c b/drivers/ata/pata_scc.c
index bbf5aa345e68..16673d168573 100644
--- a/drivers/ata/pata_scc.c
+++ b/drivers/ata/pata_scc.c
@@ -696,7 +696,7 @@ static void scc_bmdma_stop (struct ata_queued_cmd *qc)
 
 		if (reg & INTSTS_BMSINT) {
 			unsigned int classes;
-			unsigned long deadline = jiffies + ATA_TMOUT_BOOT;
+			unsigned long deadline = ata_deadline(jiffies, ATA_TMOUT_BOOT);
 			printk(KERN_WARNING "%s: Internal Bus Error\n", DRV_NAME);
 			out_be32(bmid_base + SCC_DMA_INTST, INTSTS_BMSINT);
 			/* TBD: SW reset */
diff --git a/include/linux/libata.h b/include/linux/libata.h
index e57e5d08312d..94110b652b30 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -27,6 +27,7 @@
 #define __LINUX_LIBATA_H__
 
 #include <linux/delay.h>
+#include <linux/jiffies.h>
 #include <linux/interrupt.h>
 #include <linux/dma-mapping.h>
 #include <linux/scatterlist.h>
@@ -115,7 +116,7 @@ enum {
 	/* tag ATA_MAX_QUEUE - 1 is reserved for internal commands */
 	ATA_MAX_QUEUE		= 32,
 	ATA_TAG_INTERNAL	= ATA_MAX_QUEUE - 1,
-	ATA_SHORT_PAUSE		= (HZ >> 6) + 1,
+	ATA_SHORT_PAUSE		= 16,
 
 	ATAPI_MAX_DRAIN		= 16 << 10,
 
@@ -234,17 +235,17 @@ enum {
 	/* bits 24:31 of host->flags are reserved for LLD specific flags */
 
 	/* various lengths of time */
-	ATA_TMOUT_BOOT		= 30 * HZ,	/* heuristic */
-	ATA_TMOUT_BOOT_QUICK	= 7 * HZ,	/* heuristic */
-	ATA_TMOUT_INTERNAL	= 30 * HZ,
-	ATA_TMOUT_INTERNAL_QUICK = 5 * HZ,
+	ATA_TMOUT_BOOT		= 30000,	/* heuristic */
+	ATA_TMOUT_BOOT_QUICK	=  7000,	/* heuristic */
+	ATA_TMOUT_INTERNAL	= 30000,
+	ATA_TMOUT_INTERNAL_QUICK = 5000,
 
 	/* FIXME: GoVault needs 2s but we can't afford that without
 	 * parallel probing.  800ms is enough for iVDR disk
 	 * HHD424020F7SV00.  Increase to 2secs when parallel probing
 	 * is in place.
 	 */
-	ATA_TMOUT_FF_WAIT	= 4 * HZ / 5,
+	ATA_TMOUT_FF_WAIT	=  800,
 
 	/* Spec mandates to wait for ">= 2ms" before checking status
 	 * after reset.  We wait 150ms, because that was the magic
@@ -256,14 +257,14 @@ enum {
 	 *
 	 * Old drivers/ide uses the 2mS rule and then waits for ready.
 	 */
-	ATA_WAIT_AFTER_RESET_MSECS = 150,
+	ATA_WAIT_AFTER_RESET	=  150,
 
 	/* If PMP is supported, we have to do follow-up SRST.  As some
 	 * PMPs don't send D2H Reg FIS after hardreset, LLDs are
 	 * advised to wait only for the following duration before
 	 * doing SRST.
 	 */
-	ATA_TMOUT_PMP_SRST_WAIT	= 1 * HZ,
+	ATA_TMOUT_PMP_SRST_WAIT	= 1000,
 
 	/* ATA bus states */
 	BUS_UNKNOWN		= 0,
@@ -895,8 +896,7 @@ extern void ata_host_resume(struct ata_host *host);
 #endif
 extern int ata_ratelimit(void);
 extern u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val,
-			     unsigned long interval_msec,
-			     unsigned long timeout_msec);
+			     unsigned long interval, unsigned long timeout);
 extern int atapi_cmd_type(u8 opcode);
 extern void ata_tf_to_fis(const struct ata_taskfile *tf,
 			  u8 pmp, int is_cmd, u8 *fis);
@@ -1389,6 +1389,12 @@ static inline int ata_check_ready(u8 status)
 	return 0;
 }
 
+static inline unsigned long ata_deadline(unsigned long from_jiffies,
+					 unsigned long timeout_msecs)
+{
+	return from_jiffies + msecs_to_jiffies(timeout_msecs);
+}
+
 
 /**************************************************************************
  * PMP - drivers/ata/libata-pmp.c
-- 
cgit v1.2.3


From 0a2c0f56159999e20015241d3b8fa89b1ab14309 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 20 May 2008 02:17:52 +0900
Subject: libata: improve EH retry delay handling

EH retries were delayed by 5 seconds to ensure that resets don't occur
back-to-back.  However, this 5 second delay is superflous or excessive
in many cases.  For example, after IDENTIFY times out, there's no
reason to wait five more seconds before retrying.

This patch adds ehc->last_reset timestamp and record the timestamp for
the last reset trial or success and uses it to space resets by
ATA_EH_RESET_COOL_DOWN which is 5 secs and removes unconditional 5 sec
sleeps.

As this change makes inter-try waits often shorter and they're
redundant in nature, this patch also removes the "retrying..."
messages.

While at it, convert explicit rounding up division to DIV_ROUND_UP().

This change speeds up EH in many cases w/o sacrificing robustness.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-eh.c  | 38 ++++++++++++++++++++------------------
 drivers/ata/libata-pmp.c | 10 ----------
 include/linux/libata.h   |  2 ++
 3 files changed, 22 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 08dd07f10008..5b5ae631ed03 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -67,6 +67,9 @@ enum {
 	ATA_ECAT_DUBIOUS_UNK_DEV	= 7,
 	ATA_ECAT_NR			= 8,
 
+	/* always put at least this amount of time between resets */
+	ATA_EH_RESET_COOL_DOWN		=  5000,
+
 	/* Waiting in ->prereset can never be reliable.  It's
 	 * sometimes nice to wait there but it can't be depended upon;
 	 * otherwise, we wouldn't be resetting.  Just give it enough
@@ -485,6 +488,9 @@ void ata_scsi_error(struct Scsi_Host *host)
 				if (ata_ncq_enabled(dev))
 					ehc->saved_ncq_enabled |= 1 << devno;
 			}
+
+			/* set last reset timestamp to some time in the past */
+			ehc->last_reset = jiffies - 60 * HZ;
 		}
 
 		ap->pflags |= ATA_PFLAG_EH_IN_PROGRESS;
@@ -2088,11 +2094,17 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	/*
 	 * Prepare to reset
 	 */
+	now = jiffies;
+	deadline = ata_deadline(ehc->last_reset, ATA_EH_RESET_COOL_DOWN);
+	if (time_before(now, deadline))
+		schedule_timeout_uninterruptible(deadline - now);
+
 	spin_lock_irqsave(ap->lock, flags);
 	ap->pflags |= ATA_PFLAG_RESETTING;
 	spin_unlock_irqrestore(ap->lock, flags);
 
 	ata_eh_about_to_do(link, NULL, ATA_EH_RESET);
+	ehc->last_reset = jiffies;
 
 	ata_link_for_each_dev(dev, link) {
 		/* If we issue an SRST then an ATA drive (not ATAPI)
@@ -2158,6 +2170,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	/*
 	 * Perform reset
 	 */
+	ehc->last_reset = jiffies;
 	if (ata_is_host_link(link))
 		ata_eh_freeze_port(ap);
 
@@ -2278,6 +2291,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 
 	/* reset successful, schedule revalidation */
 	ata_eh_done(link, NULL, ATA_EH_RESET);
+	ehc->last_reset = jiffies;
 	ehc->i.action |= ATA_EH_REVALIDATE;
 
 	rc = 0;
@@ -2304,9 +2318,9 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	if (time_before(now, deadline)) {
 		unsigned long delta = deadline - now;
 
-		ata_link_printk(link, KERN_WARNING, "reset failed "
-				"(errno=%d), retrying in %u secs\n",
-				rc, (jiffies_to_msecs(delta) + 999) / 1000);
+		ata_link_printk(link, KERN_WARNING,
+			"reset failed (errno=%d), retrying in %u secs\n",
+			rc, DIV_ROUND_UP(jiffies_to_msecs(delta), 1000));
 
 		while (delta)
 			delta = schedule_timeout_uninterruptible(delta);
@@ -2623,7 +2637,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 {
 	struct ata_link *link;
 	struct ata_device *dev;
-	int nr_failed_devs, nr_disabled_devs;
+	int nr_failed_devs;
 	int rc;
 	unsigned long flags;
 
@@ -2666,7 +2680,6 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
  retry:
 	rc = 0;
 	nr_failed_devs = 0;
-	nr_disabled_devs = 0;
 
 	/* if UNLOADING, finish immediately */
 	if (ap->pflags & ATA_PFLAG_UNLOADING)
@@ -2733,8 +2746,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 
 dev_fail:
 		nr_failed_devs++;
-		if (ata_eh_handle_dev_fail(dev, rc))
-			nr_disabled_devs++;
+		ata_eh_handle_dev_fail(dev, rc);
 
 		if (ap->pflags & ATA_PFLAG_FROZEN) {
 			/* PMP reset requires working host port.
@@ -2746,18 +2758,8 @@ dev_fail:
 		}
 	}
 
-	if (nr_failed_devs) {
-		if (nr_failed_devs != nr_disabled_devs) {
-			ata_port_printk(ap, KERN_WARNING, "failed to recover "
-					"some devices, retrying in 5 secs\n");
-			ssleep(5);
-		} else {
-			/* no device left to recover, repeat fast */
-			msleep(500);
-		}
-
+	if (nr_failed_devs)
 		goto retry;
-	}
 
  out:
 	if (rc && r_failed_link)
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 63691d77ac43..b65db309c181 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -727,19 +727,12 @@ static int sata_pmp_eh_recover_pmp(struct ata_port *ap,
 		}
 
 		if (tries) {
-			int sleep = ehc->i.flags & ATA_EHI_DID_RESET;
-
 			/* consecutive revalidation failures? speed down */
 			if (reval_failed)
 				sata_down_spd_limit(link);
 			else
 				reval_failed = 1;
 
-			ata_dev_printk(dev, KERN_WARNING,
-				       "retrying reset%s\n",
-				       sleep ? " in 5 secs" : "");
-			if (sleep)
-				ssleep(5);
 			ehc->i.action |= ATA_EH_RESET;
 			goto retry;
 		} else {
@@ -991,10 +984,7 @@ static int sata_pmp_eh_recover(struct ata_port *ap)
 		goto retry;
 
 	if (--pmp_tries) {
-		ata_port_printk(ap, KERN_WARNING,
-				"failed to recover PMP, retrying in 5 secs\n");
 		pmp_ehc->i.action |= ATA_EH_RESET;
-		ssleep(5);
 		goto retry;
 	}
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 94110b652b30..9058c2a325a9 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -602,6 +602,8 @@ struct ata_eh_context {
 	unsigned int		did_probe_mask;
 	unsigned int		saved_ncq_enabled;
 	u8			saved_xfer_mode[ATA_MAX_DEVICES];
+	/* timestamp for the last reset attempt or success */
+	unsigned long		last_reset;
 };
 
 struct ata_acpi_drive
-- 
cgit v1.2.3


From 87fbc5a060faf2394bee88a93519f9b9d434727c Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 20 May 2008 02:17:54 +0900
Subject: libata: improve EH internal command timeout handling

ATA_TMOUT_INTERNAL which was 30secs were used for all internal
commands which is way too long when something goes wrong.  This patch
implements command type based stepped timeouts.  Different command
types can use different timeouts and each command type can use
different timeout values after timeouts.

ie. the initial timeout is set to a value which should cover most of
the cases but not too long so that run away cases don't delay things
too much.  After the first try times out, the second try can use
longer timeout and if that one times out too, it can go for full 30sec
timeout.

IDENTIFYs use 5s - 10s - 30s timeout and all other commands use 5s -
10s timeouts.

This patch significantly cuts down the needed time to handle failure
cases while still allowing libata to work with nut job devices through
retries.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c |  16 ++++--
 drivers/ata/libata-eh.c   | 121 +++++++++++++++++++++++++++++++++++++++++++++-
 drivers/ata/libata.h      |   2 +
 include/linux/libata.h    |   8 ++-
 4 files changed, 142 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index c5c3b1b516e1..9bef1a84fe3f 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -144,7 +144,7 @@ static int libata_dma_mask = ATA_DMA_MASK_ATA|ATA_DMA_MASK_ATAPI|ATA_DMA_MASK_CF
 module_param_named(dma, libata_dma_mask, int, 0444);
 MODULE_PARM_DESC(dma, "DMA enable/disable (0x1==ATA, 0x2==ATAPI, 0x4==CF)");
 
-static int ata_probe_timeout = ATA_TMOUT_INTERNAL / 1000;
+static int ata_probe_timeout;
 module_param(ata_probe_timeout, int, 0444);
 MODULE_PARM_DESC(ata_probe_timeout, "Set ATA probing timeout (seconds)");
 
@@ -1611,6 +1611,7 @@ unsigned ata_exec_internal_sg(struct ata_device *dev,
 	struct ata_link *link = dev->link;
 	struct ata_port *ap = link->ap;
 	u8 command = tf->command;
+	int auto_timeout = 0;
 	struct ata_queued_cmd *qc;
 	unsigned int tag, preempted_tag;
 	u32 preempted_sactive, preempted_qc_active;
@@ -1683,8 +1684,14 @@ unsigned ata_exec_internal_sg(struct ata_device *dev,
 
 	spin_unlock_irqrestore(ap->lock, flags);
 
-	if (!timeout)
-		timeout = ata_probe_timeout * 1000;
+	if (!timeout) {
+		if (ata_probe_timeout)
+			timeout = ata_probe_timeout * 1000;
+		else {
+			timeout = ata_internal_cmd_timeout(dev, command);
+			auto_timeout = 1;
+		}
+	}
 
 	rc = wait_for_completion_timeout(&wait, msecs_to_jiffies(timeout));
 
@@ -1760,6 +1767,9 @@ unsigned ata_exec_internal_sg(struct ata_device *dev,
 
 	spin_unlock_irqrestore(ap->lock, flags);
 
+	if ((err_mask & AC_ERR_TIMEOUT) && auto_timeout)
+		ata_internal_cmd_timed_out(dev, command);
+
 	return err_mask;
 }
 
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 83d1451fa714..d5f03a6e3334 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -67,6 +67,8 @@ enum {
 	ATA_ECAT_DUBIOUS_UNK_DEV	= 7,
 	ATA_ECAT_NR			= 8,
 
+	ATA_EH_CMD_DFL_TIMEOUT		=  5000,
+
 	/* always put at least this amount of time between resets */
 	ATA_EH_RESET_COOL_DOWN		=  5000,
 
@@ -93,6 +95,53 @@ static const unsigned long ata_eh_reset_timeouts[] = {
 	ULONG_MAX, /* > 1 min has elapsed, give up */
 };
 
+static const unsigned long ata_eh_identify_timeouts[] = {
+	 5000,	/* covers > 99% of successes and not too boring on failures */
+	10000,  /* combined time till here is enough even for media access */
+	30000,	/* for true idiots */
+	ULONG_MAX,
+};
+
+static const unsigned long ata_eh_other_timeouts[] = {
+	 5000,	/* same rationale as identify timeout */
+	10000,	/* ditto */
+	/* but no merciful 30sec for other commands, it just isn't worth it */
+	ULONG_MAX,
+};
+
+struct ata_eh_cmd_timeout_ent {
+	const u8		*commands;
+	const unsigned long	*timeouts;
+};
+
+/* The following table determines timeouts to use for EH internal
+ * commands.  Each table entry is a command class and matches the
+ * commands the entry applies to and the timeout table to use.
+ *
+ * On the retry after a command timed out, the next timeout value from
+ * the table is used.  If the table doesn't contain further entries,
+ * the last value is used.
+ *
+ * ehc->cmd_timeout_idx keeps track of which timeout to use per
+ * command class, so if SET_FEATURES times out on the first try, the
+ * next try will use the second timeout value only for that class.
+ */
+#define CMDS(cmds...)	(const u8 []){ cmds, 0 }
+static const struct ata_eh_cmd_timeout_ent
+ata_eh_cmd_timeout_table[ATA_EH_CMD_TIMEOUT_TABLE_SIZE] = {
+	{ .commands = CMDS(ATA_CMD_ID_ATA, ATA_CMD_ID_ATAPI),
+	  .timeouts = ata_eh_identify_timeouts, },
+	{ .commands = CMDS(ATA_CMD_READ_NATIVE_MAX, ATA_CMD_READ_NATIVE_MAX_EXT),
+	  .timeouts = ata_eh_other_timeouts, },
+	{ .commands = CMDS(ATA_CMD_SET_MAX, ATA_CMD_SET_MAX_EXT),
+	  .timeouts = ata_eh_other_timeouts, },
+	{ .commands = CMDS(ATA_CMD_SET_FEATURES),
+	  .timeouts = ata_eh_other_timeouts, },
+	{ .commands = CMDS(ATA_CMD_INIT_DEV_PARAMS),
+	  .timeouts = ata_eh_other_timeouts, },
+};
+#undef CMDS
+
 static void __ata_port_freeze(struct ata_port *ap);
 #ifdef CONFIG_PM
 static void ata_eh_handle_port_suspend(struct ata_port *ap);
@@ -238,6 +287,73 @@ void ata_port_pbar_desc(struct ata_port *ap, int bar, ssize_t offset,
 
 #endif /* CONFIG_PCI */
 
+static int ata_lookup_timeout_table(u8 cmd)
+{
+	int i;
+
+	for (i = 0; i < ATA_EH_CMD_TIMEOUT_TABLE_SIZE; i++) {
+		const u8 *cur;
+
+		for (cur = ata_eh_cmd_timeout_table[i].commands; *cur; cur++)
+			if (*cur == cmd)
+				return i;
+	}
+
+	return -1;
+}
+
+/**
+ *	ata_internal_cmd_timeout - determine timeout for an internal command
+ *	@dev: target device
+ *	@cmd: internal command to be issued
+ *
+ *	Determine timeout for internal command @cmd for @dev.
+ *
+ *	LOCKING:
+ *	EH context.
+ *
+ *	RETURNS:
+ *	Determined timeout.
+ */
+unsigned long ata_internal_cmd_timeout(struct ata_device *dev, u8 cmd)
+{
+	struct ata_eh_context *ehc = &dev->link->eh_context;
+	int ent = ata_lookup_timeout_table(cmd);
+	int idx;
+
+	if (ent < 0)
+		return ATA_EH_CMD_DFL_TIMEOUT;
+
+	idx = ehc->cmd_timeout_idx[dev->devno][ent];
+	return ata_eh_cmd_timeout_table[ent].timeouts[idx];
+}
+
+/**
+ *	ata_internal_cmd_timed_out - notification for internal command timeout
+ *	@dev: target device
+ *	@cmd: internal command which timed out
+ *
+ *	Notify EH that internal command @cmd for @dev timed out.  This
+ *	function should be called only for commands whose timeouts are
+ *	determined using ata_internal_cmd_timeout().
+ *
+ *	LOCKING:
+ *	EH context.
+ */
+void ata_internal_cmd_timed_out(struct ata_device *dev, u8 cmd)
+{
+	struct ata_eh_context *ehc = &dev->link->eh_context;
+	int ent = ata_lookup_timeout_table(cmd);
+	int idx;
+
+	if (ent < 0)
+		return;
+
+	idx = ehc->cmd_timeout_idx[dev->devno][ent];
+	if (ata_eh_cmd_timeout_table[ent].timeouts[idx + 1] != ULONG_MAX)
+		ehc->cmd_timeout_idx[dev->devno][ent]++;
+}
+
 static void ata_ering_record(struct ata_ering *ering, unsigned int eflags,
 			     unsigned int err_mask)
 {
@@ -2600,8 +2716,11 @@ static int ata_eh_handle_dev_fail(struct ata_device *dev, int err)
 			ata_eh_detach_dev(dev);
 
 		/* schedule probe if necessary */
-		if (ata_eh_schedule_probe(dev))
+		if (ata_eh_schedule_probe(dev)) {
 			ehc->tries[dev->devno] = ATA_EH_DEV_TRIES;
+			memset(ehc->cmd_timeout_idx[dev->devno], 0,
+			       sizeof(ehc->cmd_timeout_idx[dev->devno]));
+		}
 
 		return 1;
 	} else {
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index 1cf803adbc95..f6f9c28ec7f8 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -151,6 +151,8 @@ extern void ata_scsi_dev_rescan(struct work_struct *work);
 extern int ata_bus_probe(struct ata_port *ap);
 
 /* libata-eh.c */
+extern unsigned long ata_internal_cmd_timeout(struct ata_device *dev, u8 cmd);
+extern void ata_internal_cmd_timed_out(struct ata_device *dev, u8 cmd);
 extern enum scsi_eh_timer_return ata_scsi_timed_out(struct scsi_cmnd *cmd);
 extern void ata_scsi_error(struct Scsi_Host *host);
 extern void ata_port_wait_eh(struct ata_port *ap);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 9058c2a325a9..035f8e1cd0ac 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -237,7 +237,6 @@ enum {
 	/* various lengths of time */
 	ATA_TMOUT_BOOT		= 30000,	/* heuristic */
 	ATA_TMOUT_BOOT_QUICK	=  7000,	/* heuristic */
-	ATA_TMOUT_INTERNAL	= 30000,
 	ATA_TMOUT_INTERNAL_QUICK = 5000,
 
 	/* FIXME: GoVault needs 2s but we can't afford that without
@@ -341,6 +340,11 @@ enum {
 
 	SATA_PMP_RW_TIMEOUT	= 3000,		/* PMP read/write timeout */
 
+	/* This should match the actual table size of
+	 * ata_eh_cmd_timeout_table in libata-eh.c.
+	 */
+	ATA_EH_CMD_TIMEOUT_TABLE_SIZE = 5,
+
 	/* Horkage types. May be set by libata or controller on drives
 	   (some horkage may be drive/controller pair dependant */
 
@@ -598,6 +602,8 @@ struct ata_eh_info {
 struct ata_eh_context {
 	struct ata_eh_info	i;
 	int			tries[ATA_MAX_DEVICES];
+	int			cmd_timeout_idx[ATA_MAX_DEVICES]
+					       [ATA_EH_CMD_TIMEOUT_TABLE_SIZE];
 	unsigned int		classes[ATA_MAX_DEVICES];
 	unsigned int		did_probe_mask;
 	unsigned int		saved_ncq_enabled;
-- 
cgit v1.2.3


From 18f7ba4c2f4be6b37d925931f04d6cc28d88d1ee Mon Sep 17 00:00:00 2001
From: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Date: Tue, 3 Jun 2008 10:33:55 -0700
Subject: libata/ahci: enclosure management support

Add Enclosure Management support to libata and ahci.

Signed-off-by:  Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/ahci.c        | 321 +++++++++++++++++++++++++++++++++++++++++++++-
 drivers/ata/libata-scsi.c |  79 ++++++++++++
 include/linux/libata.h    |  21 +++
 3 files changed, 419 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 5e6468a7ca4b..65d4e968feb4 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -56,6 +56,12 @@ MODULE_PARM_DESC(skip_host_reset, "skip global host reset (0=don't skip, 1=skip)
 static int ahci_enable_alpm(struct ata_port *ap,
 		enum link_pm policy);
 static void ahci_disable_alpm(struct ata_port *ap);
+static ssize_t ahci_led_show(struct ata_port *ap, char *buf);
+static ssize_t ahci_led_store(struct ata_port *ap, const char *buf,
+			      size_t size);
+static ssize_t ahci_transmit_led_message(struct ata_port *ap, u32 state,
+					ssize_t size);
+#define MAX_SLOTS 8
 
 enum {
 	AHCI_PCI_BAR		= 5,
@@ -98,6 +104,8 @@ enum {
 	HOST_IRQ_STAT		= 0x08, /* interrupt status */
 	HOST_PORTS_IMPL		= 0x0c, /* bitmap of implemented ports */
 	HOST_VERSION		= 0x10, /* AHCI spec. version compliancy */
+	HOST_EM_LOC		= 0x1c, /* Enclosure Management location */
+	HOST_EM_CTL		= 0x20, /* Enclosure Management Control */
 
 	/* HOST_CTL bits */
 	HOST_RESET		= (1 << 0),  /* reset controller; self-clear */
@@ -105,6 +113,7 @@ enum {
 	HOST_AHCI_EN		= (1 << 31), /* AHCI enabled */
 
 	/* HOST_CAP bits */
+	HOST_CAP_EMS		= (1 << 6),  /* Enclosure Management support */
 	HOST_CAP_SSC		= (1 << 14), /* Slumber capable */
 	HOST_CAP_PMP		= (1 << 17), /* Port Multiplier support */
 	HOST_CAP_CLO		= (1 << 24), /* Command List Override support */
@@ -202,6 +211,11 @@ enum {
 					  ATA_FLAG_IPM,
 
 	ICH_MAP				= 0x90, /* ICH MAP register */
+
+	/* em_ctl bits */
+	EM_CTL_RST			= (1 << 9), /* Reset */
+	EM_CTL_TM			= (1 << 8), /* Transmit Message */
+	EM_CTL_ALHD			= (1 << 26), /* Activity LED */
 };
 
 struct ahci_cmd_hdr {
@@ -219,12 +233,21 @@ struct ahci_sg {
 	__le32			flags_size;
 };
 
+struct ahci_em_priv {
+	enum sw_activity blink_policy;
+	struct timer_list timer;
+	unsigned long saved_activity;
+	unsigned long activity;
+	unsigned long led_state;
+};
+
 struct ahci_host_priv {
 	unsigned int		flags;		/* AHCI_HFLAG_* */
 	u32			cap;		/* cap to use */
 	u32			port_map;	/* port map to use */
 	u32			saved_cap;	/* saved initial cap */
 	u32			saved_port_map;	/* saved initial port_map */
+	u32 			em_loc; /* enclosure management location */
 };
 
 struct ahci_port_priv {
@@ -240,6 +263,8 @@ struct ahci_port_priv {
 	unsigned int		ncq_saw_dmas:1;
 	unsigned int		ncq_saw_sdb:1;
 	u32 			intr_mask;	/* interrupts to enable */
+	struct ahci_em_priv	em_priv[MAX_SLOTS];/* enclosure management info
+					 	 * per PM slot */
 };
 
 static int ahci_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
@@ -277,9 +302,20 @@ static int ahci_port_suspend(struct ata_port *ap, pm_message_t mesg);
 static int ahci_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg);
 static int ahci_pci_device_resume(struct pci_dev *pdev);
 #endif
+static ssize_t ahci_activity_show(struct ata_device *dev, char *buf);
+static ssize_t ahci_activity_store(struct ata_device *dev,
+				   enum sw_activity val);
+static void ahci_init_sw_activity(struct ata_link *link);
 
 static struct device_attribute *ahci_shost_attrs[] = {
 	&dev_attr_link_power_management_policy,
+	&dev_attr_em_message_type,
+	&dev_attr_em_message,
+	NULL
+};
+
+static struct device_attribute *ahci_sdev_attrs[] = {
+	&dev_attr_sw_activity,
 	NULL
 };
 
@@ -289,6 +325,7 @@ static struct scsi_host_template ahci_sht = {
 	.sg_tablesize		= AHCI_MAX_SG,
 	.dma_boundary		= AHCI_DMA_BOUNDARY,
 	.shost_attrs		= ahci_shost_attrs,
+	.sdev_attrs		= ahci_sdev_attrs,
 };
 
 static struct ata_port_operations ahci_ops = {
@@ -316,6 +353,10 @@ static struct ata_port_operations ahci_ops = {
 
 	.enable_pm		= ahci_enable_alpm,
 	.disable_pm		= ahci_disable_alpm,
+	.em_show		= ahci_led_show,
+	.em_store		= ahci_led_store,
+	.sw_activity_show	= ahci_activity_show,
+	.sw_activity_store	= ahci_activity_store,
 #ifdef CONFIG_PM
 	.port_suspend		= ahci_port_suspend,
 	.port_resume		= ahci_port_resume,
@@ -561,6 +602,11 @@ static struct pci_driver ahci_pci_driver = {
 #endif
 };
 
+static int ahci_em_messages = 1;
+module_param(ahci_em_messages, int, 0444);
+/* add other LED protocol types when they become supported */
+MODULE_PARM_DESC(ahci_em_messages,
+	"Set AHCI Enclosure Management Message type (0 = disabled, 1 = LED");
 
 static inline int ahci_nr_ports(u32 cap)
 {
@@ -1031,11 +1077,28 @@ static void ahci_power_down(struct ata_port *ap)
 
 static void ahci_start_port(struct ata_port *ap)
 {
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ata_link *link;
+	struct ahci_em_priv *emp;
+
 	/* enable FIS reception */
 	ahci_start_fis_rx(ap);
 
 	/* enable DMA */
 	ahci_start_engine(ap);
+
+	/* turn on LEDs */
+	if (ap->flags & ATA_FLAG_EM) {
+		ata_port_for_each_link(link, ap) {
+			emp = &pp->em_priv[link->pmp];
+			ahci_transmit_led_message(ap, emp->led_state, 4);
+		}
+	}
+
+	if (ap->flags & ATA_FLAG_SW_ACTIVITY)
+		ata_port_for_each_link(link, ap)
+			ahci_init_sw_activity(link);
+
 }
 
 static int ahci_deinit_port(struct ata_port *ap, const char **emsg)
@@ -1116,6 +1179,230 @@ static int ahci_reset_controller(struct ata_host *host)
 	return 0;
 }
 
+static void ahci_sw_activity(struct ata_link *link)
+{
+	struct ata_port *ap = link->ap;
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ahci_em_priv *emp = &pp->em_priv[link->pmp];
+
+	if (!(link->flags & ATA_LFLAG_SW_ACTIVITY))
+		return;
+
+	emp->activity++;
+	if (!timer_pending(&emp->timer))
+		mod_timer(&emp->timer, jiffies + msecs_to_jiffies(10));
+}
+
+static void ahci_sw_activity_blink(unsigned long arg)
+{
+	struct ata_link *link = (struct ata_link *)arg;
+	struct ata_port *ap = link->ap;
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ahci_em_priv *emp = &pp->em_priv[link->pmp];
+	unsigned long led_message = emp->led_state;
+	u32 activity_led_state;
+
+	led_message &= 0xffff0000;
+	led_message |= ap->port_no | (link->pmp << 8);
+
+	/* check to see if we've had activity.  If so,
+	 * toggle state of LED and reset timer.  If not,
+	 * turn LED to desired idle state.
+	 */
+	if (emp->saved_activity != emp->activity) {
+		emp->saved_activity = emp->activity;
+		/* get the current LED state */
+		activity_led_state = led_message & 0x00010000;
+
+		if (activity_led_state)
+			activity_led_state = 0;
+		else
+			activity_led_state = 1;
+
+		/* clear old state */
+		led_message &= 0xfff8ffff;
+
+		/* toggle state */
+		led_message |= (activity_led_state << 16);
+		mod_timer(&emp->timer, jiffies + msecs_to_jiffies(100));
+	} else {
+		/* switch to idle */
+		led_message &= 0xfff8ffff;
+		if (emp->blink_policy == BLINK_OFF)
+			led_message |= (1 << 16);
+	}
+	ahci_transmit_led_message(ap, led_message, 4);
+}
+
+static void ahci_init_sw_activity(struct ata_link *link)
+{
+	struct ata_port *ap = link->ap;
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ahci_em_priv *emp = &pp->em_priv[link->pmp];
+
+	/* init activity stats, setup timer */
+	emp->saved_activity = emp->activity = 0;
+	setup_timer(&emp->timer, ahci_sw_activity_blink, (unsigned long)link);
+
+	/* check our blink policy and set flag for link if it's enabled */
+	if (emp->blink_policy)
+		link->flags |= ATA_LFLAG_SW_ACTIVITY;
+}
+
+static int ahci_reset_em(struct ata_host *host)
+{
+	void __iomem *mmio = host->iomap[AHCI_PCI_BAR];
+	u32 em_ctl;
+
+	em_ctl = readl(mmio + HOST_EM_CTL);
+	if ((em_ctl & EM_CTL_TM) || (em_ctl & EM_CTL_RST))
+		return -EINVAL;
+
+	writel(em_ctl | EM_CTL_RST, mmio + HOST_EM_CTL);
+	return 0;
+}
+
+static ssize_t ahci_transmit_led_message(struct ata_port *ap, u32 state,
+					ssize_t size)
+{
+	struct ahci_host_priv *hpriv = ap->host->private_data;
+	struct ahci_port_priv *pp = ap->private_data;
+	void __iomem *mmio = ap->host->iomap[AHCI_PCI_BAR];
+	u32 em_ctl;
+	u32 message[] = {0, 0};
+	unsigned int flags;
+	int pmp;
+	struct ahci_em_priv *emp;
+
+	/* get the slot number from the message */
+	pmp = (state & 0x0000ff00) >> 8;
+	if (pmp < MAX_SLOTS)
+		emp = &pp->em_priv[pmp];
+	else
+		return -EINVAL;
+
+	spin_lock_irqsave(ap->lock, flags);
+
+	/*
+	 * if we are still busy transmitting a previous message,
+	 * do not allow
+	 */
+	em_ctl = readl(mmio + HOST_EM_CTL);
+	if (em_ctl & EM_CTL_TM) {
+		spin_unlock_irqrestore(ap->lock, flags);
+		return -EINVAL;
+	}
+
+	/*
+	 * create message header - this is all zero except for
+	 * the message size, which is 4 bytes.
+	 */
+	message[0] |= (4 << 8);
+
+	/* ignore 0:4 of byte zero, fill in port info yourself */
+	message[1] = ((state & 0xfffffff0) | ap->port_no);
+
+	/* write message to EM_LOC */
+	writel(message[0], mmio + hpriv->em_loc);
+	writel(message[1], mmio + hpriv->em_loc+4);
+
+	/* save off new led state for port/slot */
+	emp->led_state = message[1];
+
+	/*
+	 * tell hardware to transmit the message
+	 */
+	writel(em_ctl | EM_CTL_TM, mmio + HOST_EM_CTL);
+
+	spin_unlock_irqrestore(ap->lock, flags);
+	return size;
+}
+
+static ssize_t ahci_led_show(struct ata_port *ap, char *buf)
+{
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ata_link *link;
+	struct ahci_em_priv *emp;
+	int rc = 0;
+
+	ata_port_for_each_link(link, ap) {
+		emp = &pp->em_priv[link->pmp];
+		rc += sprintf(buf, "%lx\n", emp->led_state);
+	}
+	return rc;
+}
+
+static ssize_t ahci_led_store(struct ata_port *ap, const char *buf,
+				size_t size)
+{
+	int state;
+	int pmp;
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ahci_em_priv *emp;
+
+	state = simple_strtoul(buf, NULL, 0);
+
+	/* get the slot number from the message */
+	pmp = (state & 0x0000ff00) >> 8;
+	if (pmp < MAX_SLOTS)
+		emp = &pp->em_priv[pmp];
+	else
+		return -EINVAL;
+
+	/* mask off the activity bits if we are in sw_activity
+	 * mode, user should turn off sw_activity before setting
+	 * activity led through em_message
+	 */
+	if (emp->blink_policy)
+		state &= 0xfff8ffff;
+
+	return ahci_transmit_led_message(ap, state, size);
+}
+
+static ssize_t ahci_activity_store(struct ata_device *dev, enum sw_activity val)
+{
+	struct ata_link *link = dev->link;
+	struct ata_port *ap = link->ap;
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ahci_em_priv *emp = &pp->em_priv[link->pmp];
+	u32 port_led_state = emp->led_state;
+
+	/* save the desired Activity LED behavior */
+	if (val == OFF) {
+		/* clear LFLAG */
+		link->flags &= ~(ATA_LFLAG_SW_ACTIVITY);
+
+		/* set the LED to OFF */
+		port_led_state &= 0xfff80000;
+		port_led_state |= (ap->port_no | (link->pmp << 8));
+		ahci_transmit_led_message(ap, port_led_state, 4);
+	} else {
+		link->flags |= ATA_LFLAG_SW_ACTIVITY;
+		if (val == BLINK_OFF) {
+			/* set LED to ON for idle */
+			port_led_state &= 0xfff80000;
+			port_led_state |= (ap->port_no | (link->pmp << 8));
+			port_led_state |= 0x00010000; /* check this */
+			ahci_transmit_led_message(ap, port_led_state, 4);
+		}
+	}
+	emp->blink_policy = val;
+	return 0;
+}
+
+static ssize_t ahci_activity_show(struct ata_device *dev, char *buf)
+{
+	struct ata_link *link = dev->link;
+	struct ata_port *ap = link->ap;
+	struct ahci_port_priv *pp = ap->private_data;
+	struct ahci_em_priv *emp = &pp->em_priv[link->pmp];
+
+	/* display the saved value of activity behavior for this
+	 * disk.
+	 */
+	return sprintf(buf, "%d\n", emp->blink_policy);
+}
+
 static void ahci_port_init(struct pci_dev *pdev, struct ata_port *ap,
 			   int port_no, void __iomem *mmio,
 			   void __iomem *port_mmio)
@@ -1848,6 +2135,8 @@ static unsigned int ahci_qc_issue(struct ata_queued_cmd *qc)
 	writel(1 << qc->tag, port_mmio + PORT_CMD_ISSUE);
 	readl(port_mmio + PORT_CMD_ISSUE);	/* flush */
 
+	ahci_sw_activity(qc->dev->link);
+
 	return 0;
 }
 
@@ -2154,7 +2443,8 @@ static void ahci_print_info(struct ata_host *host)
 	dev_printk(KERN_INFO, &pdev->dev,
 		"flags: "
 		"%s%s%s%s%s%s%s"
-		"%s%s%s%s%s%s%s\n"
+		"%s%s%s%s%s%s%s"
+		"%s\n"
 		,
 
 		cap & (1 << 31) ? "64bit " : "",
@@ -2171,7 +2461,8 @@ static void ahci_print_info(struct ata_host *host)
 		cap & (1 << 17) ? "pmp " : "",
 		cap & (1 << 15) ? "pio " : "",
 		cap & (1 << 14) ? "slum " : "",
-		cap & (1 << 13) ? "part " : ""
+		cap & (1 << 13) ? "part " : "",
+		cap & (1 << 6) ? "ems ": ""
 		);
 }
 
@@ -2291,6 +2582,24 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (hpriv->cap & HOST_CAP_PMP)
 		pi.flags |= ATA_FLAG_PMP;
 
+	if (ahci_em_messages && (hpriv->cap & HOST_CAP_EMS)) {
+		u8 messages;
+		void __iomem *mmio = pcim_iomap_table(pdev)[AHCI_PCI_BAR];
+		u32 em_loc = readl(mmio + HOST_EM_LOC);
+		u32 em_ctl = readl(mmio + HOST_EM_CTL);
+
+		messages = (em_ctl & 0x000f0000) >> 16;
+
+		/* we only support LED message type right now */
+		if ((messages & 0x01) && (ahci_em_messages == 1)) {
+			/* store em_loc */
+			hpriv->em_loc = ((em_loc >> 16) * 4);
+			pi.flags |= ATA_FLAG_EM;
+			if (!(em_ctl & EM_CTL_ALHD))
+				pi.flags |= ATA_FLAG_SW_ACTIVITY;
+		}
+	}
+
 	/* CAP.NP sometimes indicate the index of the last enabled
 	 * port, at other times, that of the last possible port, so
 	 * determining the maximum port number requires looking at
@@ -2304,6 +2613,9 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	host->iomap = pcim_iomap_table(pdev);
 	host->private_data = hpriv;
 
+	if (pi.flags & ATA_FLAG_EM)
+		ahci_reset_em(host);
+
 	for (i = 0; i < host->n_ports; i++) {
 		struct ata_port *ap = host->ports[i];
 
@@ -2314,6 +2626,11 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		/* set initial link pm policy */
 		ap->pm_policy = NOT_AVAILABLE;
 
+		/* set enclosure management message type */
+		if (ap->flags & ATA_FLAG_EM)
+			ap->em_message_type = ahci_em_messages;
+
+
 		/* disabled/not-implemented port */
 		if (!(hpriv->port_map & (1 << i)))
 			ap->ops = &ata_dummy_port_ops;
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 57a43649a461..b578b11caa7b 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -190,6 +190,85 @@ static void ata_scsi_set_sense(struct scsi_cmnd *cmd, u8 sk, u8 asc, u8 ascq)
 	scsi_build_sense_buffer(0, cmd->sense_buffer, sk, asc, ascq);
 }
 
+static ssize_t
+ata_scsi_em_message_store(struct device *dev, struct device_attribute *attr,
+			  const char *buf, size_t count)
+{
+	struct Scsi_Host *shost = class_to_shost(dev);
+	struct ata_port *ap = ata_shost_to_port(shost);
+	if (ap->ops->em_store && (ap->flags & ATA_FLAG_EM))
+		return ap->ops->em_store(ap, buf, count);
+	return -EINVAL;
+}
+
+static ssize_t
+ata_scsi_em_message_show(struct device *dev, struct device_attribute *attr,
+			 char *buf)
+{
+	struct Scsi_Host *shost = class_to_shost(dev);
+	struct ata_port *ap = ata_shost_to_port(shost);
+
+	if (ap->ops->em_show && (ap->flags & ATA_FLAG_EM))
+		return ap->ops->em_show(ap, buf);
+	return -EINVAL;
+}
+DEVICE_ATTR(em_message, S_IRUGO | S_IWUGO,
+		ata_scsi_em_message_show, ata_scsi_em_message_store);
+EXPORT_SYMBOL_GPL(dev_attr_em_message);
+
+static ssize_t
+ata_scsi_em_message_type_show(struct device *dev, struct device_attribute *attr,
+			      char *buf)
+{
+	struct Scsi_Host *shost = class_to_shost(dev);
+	struct ata_port *ap = ata_shost_to_port(shost);
+
+	return snprintf(buf, 23, "%d\n", ap->em_message_type);
+}
+DEVICE_ATTR(em_message_type, S_IRUGO,
+		  ata_scsi_em_message_type_show, NULL);
+EXPORT_SYMBOL_GPL(dev_attr_em_message_type);
+
+static ssize_t
+ata_scsi_activity_show(struct device *dev, struct device_attribute *attr,
+		char *buf)
+{
+	struct scsi_device *sdev = to_scsi_device(dev);
+	struct ata_port *ap = ata_shost_to_port(sdev->host);
+	struct ata_device *atadev = ata_scsi_find_dev(ap, sdev);
+
+	if (ap->ops->sw_activity_show && (ap->flags & ATA_FLAG_SW_ACTIVITY))
+		return ap->ops->sw_activity_show(atadev, buf);
+	return -EINVAL;
+}
+
+static ssize_t
+ata_scsi_activity_store(struct device *dev, struct device_attribute *attr,
+	const char *buf, size_t count)
+{
+	struct scsi_device *sdev = to_scsi_device(dev);
+	struct ata_port *ap = ata_shost_to_port(sdev->host);
+	struct ata_device *atadev = ata_scsi_find_dev(ap, sdev);
+	enum sw_activity val;
+	int rc;
+
+	if (ap->ops->sw_activity_store && (ap->flags & ATA_FLAG_SW_ACTIVITY)) {
+		val = simple_strtoul(buf, NULL, 0);
+		switch (val) {
+		case OFF: case BLINK_ON: case BLINK_OFF:
+			rc = ap->ops->sw_activity_store(atadev, val);
+			if (!rc)
+				return count;
+			else
+				return rc;
+		}
+	}
+	return -EINVAL;
+}
+DEVICE_ATTR(sw_activity, S_IWUGO | S_IRUGO, ata_scsi_activity_show,
+			ata_scsi_activity_store);
+EXPORT_SYMBOL_GPL(dev_attr_sw_activity);
+
 static void ata_scsi_invalid_field(struct scsi_cmnd *cmd,
 				   void (*done)(struct scsi_cmnd *))
 {
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 035f8e1cd0ac..5b247b8a6b3b 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -169,6 +169,7 @@ enum {
 	ATA_LFLAG_ASSUME_CLASS	= ATA_LFLAG_ASSUME_ATA | ATA_LFLAG_ASSUME_SEMB,
 	ATA_LFLAG_NO_RETRY	= (1 << 5), /* don't retry this link */
 	ATA_LFLAG_DISABLED	= (1 << 6), /* link is disabled */
+	ATA_LFLAG_SW_ACTIVITY	= (1 << 7), /* keep activity stats */
 
 	/* struct ata_port flags */
 	ATA_FLAG_SLAVE_POSS	= (1 << 0), /* host supports slave dev */
@@ -191,6 +192,10 @@ enum {
 	ATA_FLAG_AN		= (1 << 18), /* controller supports AN */
 	ATA_FLAG_PMP		= (1 << 19), /* controller supports PMP */
 	ATA_FLAG_IPM		= (1 << 20), /* driver can handle IPM */
+	ATA_FLAG_EM		= (1 << 21), /* driver supports enclosure
+					      * management */
+	ATA_FLAG_SW_ACTIVITY	= (1 << 22), /* driver supports sw activity
+					      * led */
 
 	/* The following flag belongs to ap->pflags but is kept in
 	 * ap->flags because it's referenced in many LLDs and will be
@@ -446,6 +451,15 @@ enum link_pm {
 	MEDIUM_POWER,
 };
 extern struct device_attribute dev_attr_link_power_management_policy;
+extern struct device_attribute dev_attr_em_message_type;
+extern struct device_attribute dev_attr_em_message;
+extern struct device_attribute dev_attr_sw_activity;
+
+enum sw_activity {
+	OFF,
+	BLINK_ON,
+	BLINK_OFF,
+};
 
 #ifdef CONFIG_ATA_SFF
 struct ata_ioports {
@@ -701,6 +715,7 @@ struct ata_port {
 	struct timer_list	fastdrain_timer;
 	unsigned long		fastdrain_cnt;
 
+	int			em_message_type;
 	void			*private_data;
 
 #ifdef CONFIG_ATA_ACPI
@@ -792,6 +807,12 @@ struct ata_port_operations {
 	u8   (*bmdma_status)(struct ata_port *ap);
 #endif /* CONFIG_ATA_SFF */
 
+	ssize_t (*em_show)(struct ata_port *ap, char *buf);
+	ssize_t (*em_store)(struct ata_port *ap, const char *message,
+			    size_t size);
+	ssize_t (*sw_activity_show)(struct ata_device *dev, char *buf);
+	ssize_t (*sw_activity_store)(struct ata_device *dev,
+				     enum sw_activity val);
 	/*
 	 * Obsolete
 	 */
-- 
cgit v1.2.3


From 20a9b6e7c303f2a6f9afe17c0997bc9a3c734442 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Mon, 14 Jul 2008 22:38:22 +0200
Subject: i2c: Remove 3 deprecated bus drivers

This patch contains the scheduled removal of i2c-i810, i2c-prosavage
and i2c-savage4.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 Documentation/feature-removal-schedule.txt |   7 -
 Documentation/i2c/busses/i2c-i810          |  47 -----
 Documentation/i2c/busses/i2c-prosavage     |  23 --
 Documentation/i2c/busses/i2c-savage4       |  26 ---
 drivers/i2c/busses/Kconfig                 |  52 -----
 drivers/i2c/busses/Makefile                |   3 -
 drivers/i2c/busses/i2c-i810.c              | 260 -----------------------
 drivers/i2c/busses/i2c-prosavage.c         | 325 -----------------------------
 drivers/i2c/busses/i2c-savage4.c           | 185 ----------------
 include/linux/i2c-id.h                     |   1 -
 10 files changed, 929 deletions(-)
 delete mode 100644 Documentation/i2c/busses/i2c-i810
 delete mode 100644 Documentation/i2c/busses/i2c-prosavage
 delete mode 100644 Documentation/i2c/busses/i2c-savage4
 delete mode 100644 drivers/i2c/busses/i2c-i810.c
 delete mode 100644 drivers/i2c/busses/i2c-prosavage.c
 delete mode 100644 drivers/i2c/busses/i2c-savage4.c

(limited to 'include/linux')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 46ece3fba6f9..65a1482457a8 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -222,13 +222,6 @@ Who:	Thomas Gleixner <tglx@linutronix.de>
 
 ---------------------------
 
-What:	i2c-i810, i2c-prosavage and i2c-savage4
-When:	May 2008
-Why:	These drivers are superseded by i810fb, intelfb and savagefb.
-Who:	Jean Delvare <khali@linux-fr.org>
-
----------------------------
-
 What (Why):
 	- include/linux/netfilter_ipv4/ipt_TOS.h ipt_tos.h header files
 	  (superseded by xt_TOS/xt_tos target & match)
diff --git a/Documentation/i2c/busses/i2c-i810 b/Documentation/i2c/busses/i2c-i810
deleted file mode 100644
index 778210ee1583..000000000000
--- a/Documentation/i2c/busses/i2c-i810
+++ /dev/null
@@ -1,47 +0,0 @@
-Kernel driver i2c-i810
-
-Supported adapters:
-  * Intel 82810, 82810-DC100, 82810E, and 82815 (GMCH)
-  * Intel 82845G (GMCH)
-
-Authors: 
-	Frodo Looijaard <frodol@dds.nl>, 
-	Philip Edelbrock <phil@netroedge.com>,
-        Kyösti Mälkki <kmalkki@cc.hut.fi>,
-	Ralph Metzler <rjkm@thp.uni-koeln.de>,
-	Mark D. Studebaker <mdsxyz123@yahoo.com>
-
-Main contact: Mark Studebaker <mdsxyz123@yahoo.com>
-
-Description 
------------ 
-
-WARNING: If you have an '810' or '815' motherboard, your standard I2C
-temperature sensors are most likely on the 801's I2C bus. You want the
-i2c-i801 driver for those, not this driver.
-
-Now for the i2c-i810...
-
-The GMCH chip contains two I2C interfaces.
-
-The first interface is used for DDC (Data Display Channel) which is a
-serial channel through the VGA monitor connector to a DDC-compliant
-monitor. This interface is defined by the Video Electronics Standards
-Association (VESA). The standards are available for purchase at
-http://www.vesa.org .
-
-The second interface is a general-purpose I2C bus. It may be connected to a
-TV-out chip such as the BT869 or possibly to a digital flat-panel display.
-
-Features
--------- 
-
-Both busses use the i2c-algo-bit driver for 'bit banging'
-and support for specific transactions is provided by i2c-algo-bit.
-
-Issues
-------
-
-If you enable bus testing in i2c-algo-bit (insmod i2c-algo-bit bit_test=1),
-the test may fail; if so, the i2c-i810 driver won't be inserted. However,
-we think this has been fixed.
diff --git a/Documentation/i2c/busses/i2c-prosavage b/Documentation/i2c/busses/i2c-prosavage
deleted file mode 100644
index 703687902511..000000000000
--- a/Documentation/i2c/busses/i2c-prosavage
+++ /dev/null
@@ -1,23 +0,0 @@
-Kernel driver i2c-prosavage
-
-Supported adapters:
-	
-	S3/VIA KM266/VT8375 aka ProSavage8 
-	S3/VIA KM133/VT8365 aka Savage4 
-
-Author: Henk Vergonet <henk@god.dyndns.org>
-
-Description
------------
-
-The Savage4 chips contain two I2C interfaces (aka a I2C 'master' or
-'host'). 
-
-The first interface is used for DDC (Data Display Channel) which is a
-serial channel through the VGA monitor connector to a DDC-compliant
-monitor. This interface is defined by the Video Electronics Standards
-Association (VESA). The standards are available for purchase at
-http://www.vesa.org . The second interface is a general-purpose I2C bus.
-
-Usefull for gaining access to the TV Encoder chips.
-
diff --git a/Documentation/i2c/busses/i2c-savage4 b/Documentation/i2c/busses/i2c-savage4
deleted file mode 100644
index 6ecceab618d3..000000000000
--- a/Documentation/i2c/busses/i2c-savage4
+++ /dev/null
@@ -1,26 +0,0 @@
-Kernel driver i2c-savage4
-
-Supported adapters:
-  * Savage4
-  * Savage2000
-
-Authors: 
-	Alexander Wold <awold@bigfoot.com>,
-	Mark D. Studebaker <mdsxyz123@yahoo.com> 
-
-Description
------------
-
-The Savage4 chips contain two I2C interfaces (aka a I2C 'master'
-or 'host'). 
-
-The first interface is used for DDC (Data Display Channel) which is a
-serial channel through the VGA monitor connector to a DDC-compliant
-monitor. This interface is defined by the Video Electronics Standards
-Association (VESA). The standards are available for purchase at
-http://www.vesa.org . The DDC bus is not yet supported because its register
-is not directly memory-mapped.
-
-The second interface is a general-purpose I2C bus. This is the only
-interface supported by the driver at the moment.
-
diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 00d76e13588f..b7cce9211838 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -186,26 +186,6 @@ config I2C_I801
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-i801.
 
-config I2C_I810
-	tristate "Intel 810/815 (DEPRECATED)"
-	default n
-	depends on PCI
-	select I2C_ALGOBIT
-	help
-	  If you say yes to this option, support will be included for the Intel
-	  810/815 family of mainboard I2C interfaces.  Specifically, the
-	  following versions of the chipset are supported:
-	    i810AA
-	    i810AB
-	    i810E
-	    i815
-	    i845G
-
-	  This driver is deprecated in favor of the i810fb and intelfb drivers.
-
-	  This driver can also be built as a module.  If so, the module
-	  will be called i2c-i810.
-
 config I2C_PXA
 	tristate "Intel PXA2XX I2C adapter (EXPERIMENTAL)"
 	depends on EXPERIMENTAL && ARCH_PXA
@@ -402,24 +382,6 @@ config I2C_PASEMI
 	help
 	  Supports the PA Semi PWRficient on-chip SMBus interfaces.
 
-config I2C_PROSAVAGE
-	tristate "S3/VIA (Pro)Savage (DEPRECATED)"
-	default n
-	depends on PCI
-	select I2C_ALGOBIT
-	help
-	  If you say yes to this option, support will be included for the
-	  I2C bus and DDC bus of the S3VIA embedded Savage4 and ProSavage8
-	  graphics processors.
-	  chipsets supported:
-	    S3/VIA KM266/VT8375 aka ProSavage8
-	    S3/VIA KM133/VT8365 aka Savage4
-
-	  This driver is deprecated in favor of the savagefb driver.
-
-	  This support is also available as a module.  If so, the module
-	  will be called i2c-prosavage.
-
 config I2C_S3C2410
 	tristate "S3C2410 I2C Driver"
 	depends on ARCH_S3C2410
@@ -427,20 +389,6 @@ config I2C_S3C2410
 	  Say Y here to include support for I2C controller in the
 	  Samsung S3C2410 based System-on-Chip devices.
 
-config I2C_SAVAGE4
-	tristate "S3 Savage 4 (DEPRECATED)"
-	default n
-	depends on PCI
-	select I2C_ALGOBIT
-	help
-	  If you say yes to this option, support will be included for the
-	  S3 Savage 4 I2C interface.
-
-	  This driver is deprecated in favor of the savagefb driver.
-
-	  This driver can also be built as a module.  If so, the module
-	  will be called i2c-savage4.
-
 config I2C_SIBYTE
 	tristate "SiByte SMBus interface"
 	depends on SIBYTE_SB1xxx_SOC
diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile
index 8b0a8c257905..81bb407d24cc 100644
--- a/drivers/i2c/busses/Makefile
+++ b/drivers/i2c/busses/Makefile
@@ -16,7 +16,6 @@ obj-$(CONFIG_I2C_ELEKTOR)	+= i2c-elektor.o
 obj-$(CONFIG_I2C_GPIO)		+= i2c-gpio.o
 obj-$(CONFIG_I2C_HYDRA)		+= i2c-hydra.o
 obj-$(CONFIG_I2C_I801)		+= i2c-i801.o
-obj-$(CONFIG_I2C_I810)		+= i2c-i810.o
 obj-$(CONFIG_I2C_IBM_IIC)	+= i2c-ibm_iic.o
 obj-$(CONFIG_I2C_IOP3XX)	+= i2c-iop3xx.o
 obj-$(CONFIG_I2C_IXP2000)	+= i2c-ixp2000.o
@@ -35,10 +34,8 @@ obj-$(CONFIG_I2C_PCA_PLATFORM)	+= i2c-pca-platform.o
 obj-$(CONFIG_I2C_PIIX4)		+= i2c-piix4.o
 obj-$(CONFIG_I2C_PMCMSP)	+= i2c-pmcmsp.o
 obj-$(CONFIG_I2C_PNX)		+= i2c-pnx.o
-obj-$(CONFIG_I2C_PROSAVAGE)	+= i2c-prosavage.o
 obj-$(CONFIG_I2C_PXA)		+= i2c-pxa.o
 obj-$(CONFIG_I2C_S3C2410)	+= i2c-s3c2410.o
-obj-$(CONFIG_I2C_SAVAGE4)	+= i2c-savage4.o
 obj-$(CONFIG_I2C_SH7760)	+= i2c-sh7760.o
 obj-$(CONFIG_I2C_SH_MOBILE)	+= i2c-sh_mobile.o
 obj-$(CONFIG_I2C_SIBYTE)	+= i2c-sibyte.o
diff --git a/drivers/i2c/busses/i2c-i810.c b/drivers/i2c/busses/i2c-i810.c
deleted file mode 100644
index 42e8d94c276f..000000000000
--- a/drivers/i2c/busses/i2c-i810.c
+++ /dev/null
@@ -1,260 +0,0 @@
-/*
-    i2c-i810.c - Part of lm_sensors, Linux kernel modules for hardware
-              monitoring
-    Copyright (c) 1998, 1999, 2000  Frodo Looijaard <frodol@dds.nl>,
-    Philip Edelbrock <phil@netroedge.com>,
-    Ralph Metzler <rjkm@thp.uni-koeln.de>, and
-    Mark D. Studebaker <mdsxyz123@yahoo.com>
-    
-    Based on code written by Ralph Metzler <rjkm@thp.uni-koeln.de> and
-    Simon Vogl
-
-    This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-*/
-/*
-   This interfaces to the I810/I815 to provide access to
-   the DDC Bus and the I2C Bus.
-
-   SUPPORTED DEVICES	PCI ID
-   i810AA		7121           
-   i810AB		7123           
-   i810E		7125           
-   i815			1132           
-   i845G		2562
-*/
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/pci.h>
-#include <linux/i2c.h>
-#include <linux/i2c-algo-bit.h>
-#include <asm/io.h>
-
-/* GPIO register locations */
-#define I810_IOCONTROL_OFFSET	0x5000
-#define I810_HVSYNC		0x00	/* not used */
-#define I810_GPIOA		0x10
-#define I810_GPIOB		0x14
-
-/* bit locations in the registers */
-#define SCL_DIR_MASK		0x0001
-#define SCL_DIR			0x0002
-#define SCL_VAL_MASK		0x0004
-#define SCL_VAL_OUT		0x0008
-#define SCL_VAL_IN		0x0010
-#define SDA_DIR_MASK		0x0100
-#define SDA_DIR			0x0200
-#define SDA_VAL_MASK		0x0400
-#define SDA_VAL_OUT		0x0800
-#define SDA_VAL_IN		0x1000
-
-/* initialization states */
-#define INIT1			0x1
-#define INIT2			0x2
-#define INIT3			0x4
-
-/* delays */
-#define CYCLE_DELAY		10
-#define TIMEOUT			(HZ / 2)
-
-static void __iomem *ioaddr;
-
-/* The i810 GPIO registers have individual masks for each bit
-   so we never have to read before writing. Nice. */
-
-static void bit_i810i2c_setscl(void *data, int val)
-{
-	writel((val ? SCL_VAL_OUT : 0) | SCL_DIR | SCL_DIR_MASK | SCL_VAL_MASK,
-	     ioaddr + I810_GPIOB);
-	readl(ioaddr + I810_GPIOB);	/* flush posted write */
-}
-
-static void bit_i810i2c_setsda(void *data, int val)
-{
- 	writel((val ? SDA_VAL_OUT : 0) | SDA_DIR | SDA_DIR_MASK | SDA_VAL_MASK,
-	     ioaddr + I810_GPIOB);
-	readl(ioaddr + I810_GPIOB);	/* flush posted write */
-}
-
-/* The GPIO pins are open drain, so the pins could always remain outputs.
-   However, some chip versions don't latch the inputs unless they
-   are set as inputs.
-   We rely on the i2c-algo-bit routines to set the pins high before
-   reading the input from other chips. Following guidance in the 815
-   prog. ref. guide, we do a "dummy write" of 0 to the register before
-   reading which forces the input value to be latched. We presume this
-   applies to the 810 as well; shouldn't hurt anyway. This is necessary to get
-   i2c_algo_bit bit_test=1 to pass. */
-
-static int bit_i810i2c_getscl(void *data)
-{
-	writel(SCL_DIR_MASK, ioaddr + I810_GPIOB);
-	writel(0, ioaddr + I810_GPIOB);
-	return (0 != (readl(ioaddr + I810_GPIOB) & SCL_VAL_IN));
-}
-
-static int bit_i810i2c_getsda(void *data)
-{
-	writel(SDA_DIR_MASK, ioaddr + I810_GPIOB);
-	writel(0, ioaddr + I810_GPIOB);
-	return (0 != (readl(ioaddr + I810_GPIOB) & SDA_VAL_IN));
-}
-
-static void bit_i810ddc_setscl(void *data, int val)
-{
-	writel((val ? SCL_VAL_OUT : 0) | SCL_DIR | SCL_DIR_MASK | SCL_VAL_MASK,
-	     ioaddr + I810_GPIOA);
-	readl(ioaddr + I810_GPIOA);	/* flush posted write */
-}
-
-static void bit_i810ddc_setsda(void *data, int val)
-{
- 	writel((val ? SDA_VAL_OUT : 0) | SDA_DIR | SDA_DIR_MASK | SDA_VAL_MASK,
-	     ioaddr + I810_GPIOA);
-	readl(ioaddr + I810_GPIOA);	/* flush posted write */
-}
-
-static int bit_i810ddc_getscl(void *data)
-{
-	writel(SCL_DIR_MASK, ioaddr + I810_GPIOA);
-	writel(0, ioaddr + I810_GPIOA);
-	return (0 != (readl(ioaddr + I810_GPIOA) & SCL_VAL_IN));
-}
-
-static int bit_i810ddc_getsda(void *data)
-{
-	writel(SDA_DIR_MASK, ioaddr + I810_GPIOA);
-	writel(0, ioaddr + I810_GPIOA);
-	return (0 != (readl(ioaddr + I810_GPIOA) & SDA_VAL_IN));
-}
-
-static int config_i810(struct pci_dev *dev)
-{
-	unsigned long cadr;
-
-	/* map I810 memory */
-	cadr = dev->resource[1].start;
-	cadr += I810_IOCONTROL_OFFSET;
-	cadr &= PCI_BASE_ADDRESS_MEM_MASK;
-	ioaddr = ioremap_nocache(cadr, 0x1000);
-	if (ioaddr) {
-		bit_i810i2c_setscl(NULL, 1);
-		bit_i810i2c_setsda(NULL, 1);
-		bit_i810ddc_setscl(NULL, 1);
-		bit_i810ddc_setsda(NULL, 1);
-		return 0;
-	}
-	return -ENODEV;
-}
-
-static struct i2c_algo_bit_data i810_i2c_bit_data = {
-	.setsda		= bit_i810i2c_setsda,
-	.setscl		= bit_i810i2c_setscl,
-	.getsda		= bit_i810i2c_getsda,
-	.getscl		= bit_i810i2c_getscl,
-	.udelay		= CYCLE_DELAY,
-	.timeout	= TIMEOUT,
-};
-
-static struct i2c_adapter i810_i2c_adapter = {
-	.owner		= THIS_MODULE,
-	.id		= I2C_HW_B_I810,
-	.name		= "I810/I815 I2C Adapter",
-	.algo_data	= &i810_i2c_bit_data,
-};
-
-static struct i2c_algo_bit_data i810_ddc_bit_data = {
-	.setsda		= bit_i810ddc_setsda,
-	.setscl		= bit_i810ddc_setscl,
-	.getsda		= bit_i810ddc_getsda,
-	.getscl		= bit_i810ddc_getscl,
-	.udelay		= CYCLE_DELAY,
-	.timeout	= TIMEOUT,
-};
-
-static struct i2c_adapter i810_ddc_adapter = {
-	.owner		= THIS_MODULE,
-	.id		= I2C_HW_B_I810,
-	.name		= "I810/I815 DDC Adapter",
-	.algo_data	= &i810_ddc_bit_data,
-};
-
-static struct pci_device_id i810_ids[] __devinitdata = {
-	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82810_IG1) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82810_IG3) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82810E_IG) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82815_CGC) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82845G_IG) },
-	{ 0, },
-};
-
-MODULE_DEVICE_TABLE (pci, i810_ids);
-
-static int __devinit i810_probe(struct pci_dev *dev, const struct pci_device_id *id)
-{
-	int retval;
-
-	retval = config_i810(dev);
-	if (retval)
-		return retval;
-	dev_info(&dev->dev, "i810/i815 i2c device found.\n");
-
-	/* set up the sysfs linkage to our parent device */
-	i810_i2c_adapter.dev.parent = &dev->dev;
-	i810_ddc_adapter.dev.parent = &dev->dev;
-
-	retval = i2c_bit_add_bus(&i810_i2c_adapter);
-	if (retval)
-		return retval;
-	retval = i2c_bit_add_bus(&i810_ddc_adapter);
-	if (retval)
-		i2c_del_adapter(&i810_i2c_adapter);
-	return retval;
-}
-
-static void __devexit i810_remove(struct pci_dev *dev)
-{
-	i2c_del_adapter(&i810_ddc_adapter);
-	i2c_del_adapter(&i810_i2c_adapter);
-	iounmap(ioaddr);
-}
-
-static struct pci_driver i810_driver = {
-	.name		= "i810_smbus",
-	.id_table	= i810_ids,
-	.probe		= i810_probe,
-	.remove		= __devexit_p(i810_remove),
-};
-
-static int __init i2c_i810_init(void)
-{
-	return pci_register_driver(&i810_driver);
-}
-
-static void __exit i2c_i810_exit(void)
-{
-	pci_unregister_driver(&i810_driver);
-}
-
-MODULE_AUTHOR("Frodo Looijaard <frodol@dds.nl>, "
-		"Philip Edelbrock <phil@netroedge.com>, "
-		"Ralph Metzler <rjkm@thp.uni-koeln.de>, "
-		"and Mark D. Studebaker <mdsxyz123@yahoo.com>");
-MODULE_DESCRIPTION("I810/I815 I2C/DDC driver");
-MODULE_LICENSE("GPL");
-
-module_init(i2c_i810_init);
-module_exit(i2c_i810_exit);
diff --git a/drivers/i2c/busses/i2c-prosavage.c b/drivers/i2c/busses/i2c-prosavage.c
deleted file mode 100644
index 07c1f1e27df1..000000000000
--- a/drivers/i2c/busses/i2c-prosavage.c
+++ /dev/null
@@ -1,325 +0,0 @@
-/*
- *    kernel/busses/i2c-prosavage.c
- *
- *    i2c bus driver for S3/VIA 8365/8375 graphics processor.
- *    Copyright (c) 2003 Henk Vergonet <henk@god.dyndns.org>
- *    Based on code written by:
- *	Frodo Looijaard <frodol@dds.nl>,
- *	Philip Edelbrock <phil@netroedge.com>,
- *	Ralph Metzler <rjkm@thp.uni-koeln.de>, and
- *	Mark D. Studebaker <mdsxyz123@yahoo.com>
- *	Simon Vogl
- *	and others
- *
- *    Please read the lm_sensors documentation for details on use.
- *
- *    This program is free software; you can redistribute it and/or modify
- *    it under the terms of the GNU General Public License as published by
- *    the Free Software Foundation; either version 2 of the License, or
- *    (at your option) any later version.
- *
- *    This program is distributed in the hope that it will be useful,
- *    but WITHOUT ANY WARRANTY; without even the implied warranty of
- *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *    GNU General Public License for more details.
- *
- *    You should have received a copy of the GNU General Public License
- *    along with this program; if not, write to the Free Software
- *    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- */
-/*  18-05-2003 HVE - created
- *  14-06-2003 HVE - adapted for lm_sensors2
- *  17-06-2003 HVE - linux 2.5.xx compatible
- *  18-06-2003 HVE - codingstyle
- *  21-06-2003 HVE - compatibility lm_sensors2 and linux 2.5.xx
- *		     codingstyle, mmio enabled
- *
- *  This driver interfaces to the I2C bus of the VIA north bridge embedded
- *  ProSavage4/8 devices. Usefull for gaining access to the TV Encoder chips.
- *
- *  Graphics cores:
- *   S3/VIA KM266/VT8375 aka ProSavage8
- *   S3/VIA KM133/VT8365 aka Savage4
- *
- *  Two serial busses are implemented:
- *   SERIAL1 - I2C serial communications interface
- *   SERIAL2 - DDC2 monitor communications interface
- *
- *  Tested on a FX41 mainboard, see http://www.shuttle.com
- * 
- *
- *  TODO:
- *  - integration with prosavage framebuffer device
- *    (Additional documentation needed :(
- */
-
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/pci.h>
-#include <linux/i2c.h>
-#include <linux/i2c-algo-bit.h>
-#include <asm/io.h>
-
-/*
- * driver configuration
- */
-#define MAX_BUSSES	2
-
-struct s_i2c_bus {
-	void __iomem *mmvga;
-	int	i2c_reg;
-	int	adap_ok;
-	struct i2c_adapter		adap;
-	struct i2c_algo_bit_data	algo;
-};
-
-struct s_i2c_chip {
-	void __iomem *mmio;
-	struct s_i2c_bus	i2c_bus[MAX_BUSSES];
-};
-
-
-/*
- * i2c configuration
- */
-#define CYCLE_DELAY	10
-#define TIMEOUT		(HZ / 2)
-
-
-/* 
- * S3/VIA 8365/8375 registers
- */
-#define VGA_CR_IX	0x3d4
-#define VGA_CR_DATA	0x3d5
-
-#define CR_SERIAL1	0xa0	/* I2C serial communications interface */
-#define MM_SERIAL1	0xff20
-#define CR_SERIAL2	0xb1	/* DDC2 monitor communications interface */
-
-/* based on vt8365 documentation */
-#define I2C_ENAB	0x10
-#define I2C_SCL_OUT	0x01
-#define I2C_SDA_OUT	0x02
-#define I2C_SCL_IN	0x04
-#define I2C_SDA_IN	0x08
-
-#define SET_CR_IX(p, val)	writeb((val), (p)->mmvga + VGA_CR_IX)
-#define SET_CR_DATA(p, val)	writeb((val), (p)->mmvga + VGA_CR_DATA)
-#define GET_CR_DATA(p)		readb((p)->mmvga + VGA_CR_DATA)
-
-
-/*
- * Serial bus line handling
- *
- * serial communications register as parameter in private data
- *
- * TODO: locks with other code sections accessing video registers?
- */
-static void bit_s3via_setscl(void *bus, int val)
-{
-	struct s_i2c_bus *p = (struct s_i2c_bus *)bus;
-	unsigned int r;
-
-	SET_CR_IX(p, p->i2c_reg);
-	r = GET_CR_DATA(p);
-	r |= I2C_ENAB;
-	if (val) {
-		r |= I2C_SCL_OUT;
-	} else {
-		r &= ~I2C_SCL_OUT;
-	}
-	SET_CR_DATA(p, r);
-}
-
-static void bit_s3via_setsda(void *bus, int val)
-{
-	struct s_i2c_bus *p = (struct s_i2c_bus *)bus;
-	unsigned int r;
-	
-	SET_CR_IX(p, p->i2c_reg);
-	r = GET_CR_DATA(p);
-	r |= I2C_ENAB;
-	if (val) {
-		r |= I2C_SDA_OUT;
-	} else {
-		r &= ~I2C_SDA_OUT;
-	}
-	SET_CR_DATA(p, r);
-}
-
-static int bit_s3via_getscl(void *bus)
-{
-	struct s_i2c_bus *p = (struct s_i2c_bus *)bus;
-
-	SET_CR_IX(p, p->i2c_reg);
-	return (0 != (GET_CR_DATA(p) & I2C_SCL_IN));
-}
-
-static int bit_s3via_getsda(void *bus)
-{
-	struct s_i2c_bus *p = (struct s_i2c_bus *)bus;
-
-	SET_CR_IX(p, p->i2c_reg);
-	return (0 != (GET_CR_DATA(p) & I2C_SDA_IN));
-}
-
-
-/*
- * adapter initialisation
- */
-static int i2c_register_bus(struct pci_dev *dev, struct s_i2c_bus *p, void __iomem *mmvga, u32 i2c_reg)
-{
-	int ret;
-	p->adap.owner	  = THIS_MODULE;
-	p->adap.id	  = I2C_HW_B_S3VIA;
-	p->adap.algo_data = &p->algo;
-	p->adap.dev.parent = &dev->dev;
-	p->algo.setsda	  = bit_s3via_setsda;
-	p->algo.setscl	  = bit_s3via_setscl;
-	p->algo.getsda	  = bit_s3via_getsda;
-	p->algo.getscl	  = bit_s3via_getscl;
-	p->algo.udelay	  = CYCLE_DELAY;
-	p->algo.timeout	  = TIMEOUT;
-	p->algo.data	  = p;
-	p->mmvga	  = mmvga;
-	p->i2c_reg	  = i2c_reg;
-    
-	ret = i2c_bit_add_bus(&p->adap);
-	if (ret) {
-		return ret;
-	}
-
-	p->adap_ok = 1;
-	return 0;
-}
-
-
-/*
- * Cleanup stuff
- */
-static void prosavage_remove(struct pci_dev *dev)
-{
-	struct s_i2c_chip *chip;
-	int i, ret;
-
-	chip = (struct s_i2c_chip *)pci_get_drvdata(dev);
-
-	if (!chip) {
-		return;
-	}
-	for (i = MAX_BUSSES - 1; i >= 0; i--) {
-		if (chip->i2c_bus[i].adap_ok == 0)
-			continue;
-
-		ret = i2c_del_adapter(&chip->i2c_bus[i].adap);
-	        if (ret) {
-			dev_err(&dev->dev, "%s not removed\n",
-				chip->i2c_bus[i].adap.name);
-		}
-	}
-	if (chip->mmio) {
-		iounmap(chip->mmio);
-	}
-	kfree(chip);
-}
-
-
-/*
- * Detect chip and initialize it
- */
-static int __devinit prosavage_probe(struct pci_dev *dev, const struct pci_device_id *id)
-{
-	int ret;
-	unsigned long base, len;
-	struct s_i2c_chip *chip;
-	struct s_i2c_bus  *bus;
-
-	pci_set_drvdata(dev, kzalloc(sizeof(struct s_i2c_chip), GFP_KERNEL));
-	chip = (struct s_i2c_chip *)pci_get_drvdata(dev);
-	if (chip == NULL) {
-		return -ENOMEM;
-	}
-
-	base = dev->resource[0].start & PCI_BASE_ADDRESS_MEM_MASK;
-	len  = dev->resource[0].end - base + 1;
-	chip->mmio = ioremap_nocache(base, len);
-
-	if (chip->mmio == NULL) {
-		dev_err(&dev->dev, "ioremap failed\n");
-		prosavage_remove(dev);
-		return -ENODEV;
-	}
-
-
-	/*
-	 * Chip initialisation
-	 */
-	/* Unlock Extended IO Space ??? */
-
-
-	/*
-	 * i2c bus registration
-	 */
-	bus = &chip->i2c_bus[0];
-	snprintf(bus->adap.name, sizeof(bus->adap.name),
-		"ProSavage I2C bus at %02x:%02x.%x",
-		dev->bus->number, PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
-	ret = i2c_register_bus(dev, bus, chip->mmio + 0x8000, CR_SERIAL1);
-	if (ret) {
-		goto err_adap;
-	}
-	/*
-	 * ddc bus registration
-	 */
-	bus = &chip->i2c_bus[1];
-	snprintf(bus->adap.name, sizeof(bus->adap.name),
-		"ProSavage DDC bus at %02x:%02x.%x",
-		dev->bus->number, PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
-	ret = i2c_register_bus(dev, bus, chip->mmio + 0x8000, CR_SERIAL2);
-	if (ret) {
-		goto err_adap;
-	}
-	return 0;
-err_adap:
-	dev_err(&dev->dev, "%s failed\n", bus->adap.name);
-	prosavage_remove(dev);
-	return ret;
-}
-
-
-/*
- * Data for PCI driver interface
- */
-static struct pci_device_id prosavage_pci_tbl[] = {
-	{ PCI_DEVICE(PCI_VENDOR_ID_S3, PCI_DEVICE_ID_S3_SAVAGE4) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_S3, PCI_DEVICE_ID_S3_PROSAVAGE8) },
-	{ 0, },
-};
-
-MODULE_DEVICE_TABLE (pci, prosavage_pci_tbl);
-
-static struct pci_driver prosavage_driver = {
-	.name		=	"prosavage_smbus",
-	.id_table	=	prosavage_pci_tbl,
-	.probe		=	prosavage_probe,
-	.remove		=	prosavage_remove,
-};
-
-static int __init i2c_prosavage_init(void)
-{
-	return pci_register_driver(&prosavage_driver);
-}
-
-static void __exit i2c_prosavage_exit(void)
-{
-	pci_unregister_driver(&prosavage_driver);
-}
-
-MODULE_DEVICE_TABLE(pci, prosavage_pci_tbl);
-MODULE_AUTHOR("Henk Vergonet");
-MODULE_DESCRIPTION("ProSavage VIA 8365/8375 smbus driver");
-MODULE_LICENSE("GPL");
-
-module_init (i2c_prosavage_init);
-module_exit (i2c_prosavage_exit);
diff --git a/drivers/i2c/busses/i2c-savage4.c b/drivers/i2c/busses/i2c-savage4.c
deleted file mode 100644
index 8adf4abaa035..000000000000
--- a/drivers/i2c/busses/i2c-savage4.c
+++ /dev/null
@@ -1,185 +0,0 @@
-/*
-    i2c-savage4.c - Part of lm_sensors, Linux kernel modules for hardware
-              monitoring
-    Copyright (C) 1998-2003  The LM Sensors Team
-    Alexander Wold <awold@bigfoot.com>
-    Mark D. Studebaker <mdsxyz123@yahoo.com>
-    
-    Based on i2c-voodoo3.c.
-
-    This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-*/
-
-/* This interfaces to the I2C bus of the Savage4 to gain access to
-   the BT869 and possibly other I2C devices. The DDC bus is not
-   yet supported because its register is not memory-mapped.
-*/
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/pci.h>
-#include <linux/i2c.h>
-#include <linux/i2c-algo-bit.h>
-#include <asm/io.h>
-
-/* device IDs */
-#define PCI_CHIP_SAVAGE4	0x8A22
-#define PCI_CHIP_SAVAGE2000	0x9102
-
-#define REG			0xff20	/* Serial Port 1 Register */
-
-/* bit locations in the register */
-#define I2C_ENAB		0x00000020
-#define I2C_SCL_OUT		0x00000001
-#define I2C_SDA_OUT		0x00000002
-#define I2C_SCL_IN		0x00000008
-#define I2C_SDA_IN		0x00000010
-
-/* delays */
-#define CYCLE_DELAY		10
-#define TIMEOUT			(HZ / 2)
-
-
-static void __iomem *ioaddr;
-
-/* The sav GPIO registers don't have individual masks for each bit
-   so we always have to read before writing. */
-
-static void bit_savi2c_setscl(void *data, int val)
-{
-	unsigned int r;
-	r = readl(ioaddr + REG);
-	if(val)
-		r |= I2C_SCL_OUT;
-	else
-		r &= ~I2C_SCL_OUT;
-	writel(r, ioaddr + REG);
-	readl(ioaddr + REG);	/* flush posted write */
-}
-
-static void bit_savi2c_setsda(void *data, int val)
-{
-	unsigned int r;
-	r = readl(ioaddr + REG);
-	if(val)
-		r |= I2C_SDA_OUT;
-	else
-		r &= ~I2C_SDA_OUT;
-	writel(r, ioaddr + REG);
-	readl(ioaddr + REG);	/* flush posted write */
-}
-
-/* The GPIO pins are open drain, so the pins always remain outputs.
-   We rely on the i2c-algo-bit routines to set the pins high before
-   reading the input from other chips. */
-
-static int bit_savi2c_getscl(void *data)
-{
-	return (0 != (readl(ioaddr + REG) & I2C_SCL_IN));
-}
-
-static int bit_savi2c_getsda(void *data)
-{
-	return (0 != (readl(ioaddr + REG) & I2C_SDA_IN));
-}
-
-/* Configures the chip */
-
-static int config_s4(struct pci_dev *dev)
-{
-	unsigned long cadr;
-
-	/* map memory */
-	cadr = dev->resource[0].start;
-	cadr &= PCI_BASE_ADDRESS_MEM_MASK;
-	ioaddr = ioremap_nocache(cadr, 0x0080000);
-	if (ioaddr) {
-		/* writel(0x8160, ioaddr + REG2); */
-		writel(0x00000020, ioaddr + REG);
-		dev_info(&dev->dev, "Using Savage4 at %p\n", ioaddr);
-		return 0;
-	}
-	return -ENODEV;
-}
-
-static struct i2c_algo_bit_data sav_i2c_bit_data = {
-	.setsda		= bit_savi2c_setsda,
-	.setscl		= bit_savi2c_setscl,
-	.getsda		= bit_savi2c_getsda,
-	.getscl		= bit_savi2c_getscl,
-	.udelay		= CYCLE_DELAY,
-	.timeout	= TIMEOUT
-};
-
-static struct i2c_adapter savage4_i2c_adapter = {
-	.owner		= THIS_MODULE,
-	.id		= I2C_HW_B_SAVAGE,
-	.name		= "I2C Savage4 adapter",
-	.algo_data	= &sav_i2c_bit_data,
-};
-
-static struct pci_device_id savage4_ids[] __devinitdata = {
-	{ PCI_DEVICE(PCI_VENDOR_ID_S3, PCI_CHIP_SAVAGE4) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_S3, PCI_CHIP_SAVAGE2000) },
-	{ 0, }
-};
-
-MODULE_DEVICE_TABLE (pci, savage4_ids);
-
-static int __devinit savage4_probe(struct pci_dev *dev, const struct pci_device_id *id)
-{
-	int retval;
-
-	retval = config_s4(dev);
-	if (retval)
-		return retval;
-
-	/* set up the sysfs linkage to our parent device */
-	savage4_i2c_adapter.dev.parent = &dev->dev;
-
-	return i2c_bit_add_bus(&savage4_i2c_adapter);
-}
-
-static void __devexit savage4_remove(struct pci_dev *dev)
-{
-	i2c_del_adapter(&savage4_i2c_adapter);
-	iounmap(ioaddr);
-}
-
-static struct pci_driver savage4_driver = {
-	.name		= "savage4_smbus",
-	.id_table	= savage4_ids,
-	.probe		= savage4_probe,
-	.remove		= __devexit_p(savage4_remove),
-};
-
-static int __init i2c_savage4_init(void)
-{
-	return pci_register_driver(&savage4_driver);
-}
-
-static void __exit i2c_savage4_exit(void)
-{
-	pci_unregister_driver(&savage4_driver);
-}
-
-MODULE_AUTHOR("Alexander Wold <awold@bigfoot.com> "
-		"and Mark D. Studebaker <mdsxyz123@yahoo.com>");
-MODULE_DESCRIPTION("Savage4 I2C/SMBus driver");
-MODULE_LICENSE("GPL");
-
-module_init(i2c_savage4_init);
-module_exit(i2c_savage4_exit);
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 580acc93903e..988e566d3ed5 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -111,7 +111,6 @@
 #define I2C_HW_B_RIVA		0x010010 /* Riva based graphics cards */
 #define I2C_HW_B_IOC		0x010011 /* IOC bit-wiggling */
 #define I2C_HW_B_IXP2000	0x010016 /* GPIO on IXP2000 systems */
-#define I2C_HW_B_S3VIA		0x010018 /* S3Via ProSavage adapter */
 #define I2C_HW_B_ZR36067	0x010019 /* Zoran-36057/36067 based boards */
 #define I2C_HW_B_PCILYNX	0x01001a /* TI PCILynx I2C adapter */
 #define I2C_HW_B_CX2388x	0x01001b /* connexant 2388x based tv cards */
-- 
cgit v1.2.3


From 67c2e66571c383404a5acd08189194da660da942 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 14 Jul 2008 22:38:23 +0200
Subject: i2c: Delete unused function i2c_smbus_write_quick

Function i2c_smbus_write_quick has no users left, so we can delete it.

Also update the list of these helper functions which are gone but
could be added back if needed.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 Documentation/i2c/smbus-protocol  |  4 ++--
 Documentation/i2c/writing-clients | 14 +++++++-------
 drivers/i2c/i2c-core.c            |  7 -------
 include/linux/i2c.h               |  1 -
 4 files changed, 9 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/i2c/smbus-protocol b/Documentation/i2c/smbus-protocol
index 03f08fb491cc..24bfb65da17d 100644
--- a/Documentation/i2c/smbus-protocol
+++ b/Documentation/i2c/smbus-protocol
@@ -42,8 +42,8 @@ Count (8 bits): A data byte containing the length of a block operation.
 [..]: Data sent by I2C device, as opposed to data sent by the host adapter.
 
 
-SMBus Quick Command:  i2c_smbus_write_quick()
-=============================================
+SMBus Quick Command
+===================
 
 This sends a single bit to the device, at the place of the Rd/Wr bit.
 
diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients
index ba5d1971f35f..63722d3c9cdf 100644
--- a/Documentation/i2c/writing-clients
+++ b/Documentation/i2c/writing-clients
@@ -569,7 +569,6 @@ SMBus communication
   in terms of it. Never use this function directly!
 
 
-  extern s32 i2c_smbus_write_quick(struct i2c_client * client, u8 value);
   extern s32 i2c_smbus_read_byte(struct i2c_client * client);
   extern s32 i2c_smbus_write_byte(struct i2c_client * client, u8 value);
   extern s32 i2c_smbus_read_byte_data(struct i2c_client * client, u8 command);
@@ -578,20 +577,21 @@ SMBus communication
   extern s32 i2c_smbus_read_word_data(struct i2c_client * client, u8 command);
   extern s32 i2c_smbus_write_word_data(struct i2c_client * client,
                                        u8 command, u16 value);
+  extern s32 i2c_smbus_read_block_data(struct i2c_client * client,
+                                       u8 command, u8 *values);
   extern s32 i2c_smbus_write_block_data(struct i2c_client * client,
                                         u8 command, u8 length,
                                         u8 *values);
   extern s32 i2c_smbus_read_i2c_block_data(struct i2c_client * client,
                                            u8 command, u8 length, u8 *values);
-
-These ones were removed in Linux 2.6.10 because they had no users, but could
-be added back later if needed:
-
-  extern s32 i2c_smbus_read_block_data(struct i2c_client * client,
-                                       u8 command, u8 *values);
   extern s32 i2c_smbus_write_i2c_block_data(struct i2c_client * client,
                                             u8 command, u8 length,
                                             u8 *values);
+
+These ones were removed from i2c-core because they had no users, but could
+be added back later if needed:
+
+  extern s32 i2c_smbus_write_quick(struct i2c_client * client, u8 value);
   extern s32 i2c_smbus_process_call(struct i2c_client * client,
                                     u8 command, u16 value);
   extern s32 i2c_smbus_block_process_call(struct i2c_client *client,
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 937f1dcbf3d7..3695a4a1ab77 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -1303,13 +1303,6 @@ static int i2c_smbus_check_pec(u8 cpec, struct i2c_msg *msg)
 	return 0;
 }
 
-s32 i2c_smbus_write_quick(struct i2c_client *client, u8 value)
-{
-	return i2c_smbus_xfer(client->adapter,client->addr,client->flags,
-	                      value,0,I2C_SMBUS_QUICK,NULL);
-}
-EXPORT_SYMBOL(i2c_smbus_write_quick);
-
 s32 i2c_smbus_read_byte(struct i2c_client *client)
 {
 	union i2c_smbus_data data;
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 8dc730132192..b3695f353f79 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -71,7 +71,6 @@ extern s32 i2c_smbus_xfer (struct i2c_adapter * adapter, u16 addr,
 /* Now follow the 'nice' access routines. These also document the calling
    conventions of smbus_access. */
 
-extern s32 i2c_smbus_write_quick(struct i2c_client * client, u8 value);
 extern s32 i2c_smbus_read_byte(struct i2c_client * client);
 extern s32 i2c_smbus_write_byte(struct i2c_client * client, u8 value);
 extern s32 i2c_smbus_read_byte_data(struct i2c_client * client, u8 command);
-- 
cgit v1.2.3


From ae7193f7fa3e1735ab70807eb6e35a2a6575623f Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 14 Jul 2008 22:38:24 +0200
Subject: i2c: Update stray references to smbus_access

That function is actually named i2c_smbus_xfer.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 include/linux/i2c.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index b3695f353f79..7c36d5188d39 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -69,7 +69,7 @@ extern s32 i2c_smbus_xfer (struct i2c_adapter * adapter, u16 addr,
                            union i2c_smbus_data * data);
 
 /* Now follow the 'nice' access routines. These also document the calling
-   conventions of smbus_access. */
+   conventions of i2c_smbus_xfer. */
 
 extern s32 i2c_smbus_read_byte(struct i2c_client * client);
 extern s32 i2c_smbus_write_byte(struct i2c_client * client, u8 value);
@@ -536,7 +536,7 @@ union i2c_smbus_data {
 	                       /* and one more for user-space compatibility */
 };
 
-/* smbus_access read or write markers */
+/* i2c_smbus_xfer read or write markers */
 #define I2C_SMBUS_READ	1
 #define I2C_SMBUS_WRITE	0
 
-- 
cgit v1.2.3


From c1b6b4f2342d073698dfc2547240c35045a1d00e Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 14 Jul 2008 22:38:28 +0200
Subject: i2c: Let framebuffer drivers set their I2C bus class to DDC

Let framebuffer drivers set their I2C bus class to DDC. Once this is
done, we will be able to tell the eeprom driver to only probe for
EDID EEPROMs on these buses.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/video/fb_ddc.c              |  1 +
 drivers/video/intelfb/intelfb_i2c.c | 12 +++++++-----
 drivers/video/matrox/i2c-matroxfb.c | 20 +++++++++++++++-----
 include/linux/i2c.h                 |  2 +-
 4 files changed, 24 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/video/fb_ddc.c b/drivers/video/fb_ddc.c
index a0df63289b5f..0cf96eb8a60f 100644
--- a/drivers/video/fb_ddc.c
+++ b/drivers/video/fb_ddc.c
@@ -106,6 +106,7 @@ unsigned char *fb_ddc_read(struct i2c_adapter *adapter)
 	algo_data->setsda(algo_data->data, 1);
 	algo_data->setscl(algo_data->data, 1);
 
+	adapter->class |= I2C_CLASS_DDC;
 	return edid;
 }
 
diff --git a/drivers/video/intelfb/intelfb_i2c.c b/drivers/video/intelfb/intelfb_i2c.c
index ca95f09d8b43..fcf9fadbf572 100644
--- a/drivers/video/intelfb/intelfb_i2c.c
+++ b/drivers/video/intelfb/intelfb_i2c.c
@@ -100,7 +100,8 @@ static int intelfb_gpio_getsda(void *data)
 
 static int intelfb_setup_i2c_bus(struct intelfb_info *dinfo,
 				 struct intelfb_i2c_chan *chan,
-				 const u32 reg, const char *name)
+				 const u32 reg, const char *name,
+				 int class)
 {
 	int rc;
 
@@ -108,6 +109,7 @@ static int intelfb_setup_i2c_bus(struct intelfb_info *dinfo,
 	chan->reg			= reg;
 	snprintf(chan->adapter.name, sizeof(chan->adapter.name),
 		 "intelfb %s", name);
+	chan->adapter.class		= class;
 	chan->adapter.owner		= THIS_MODULE;
 	chan->adapter.id		= I2C_HW_B_INTELFB;
 	chan->adapter.algo_data		= &chan->algo;
@@ -145,7 +147,7 @@ void intelfb_create_i2c_busses(struct intelfb_info *dinfo)
 
 	/* setup the DDC bus for analog output */
 	intelfb_setup_i2c_bus(dinfo, &dinfo->output[i].ddc_bus, GPIOA,
-			      "CRTDDC_A");
+			      "CRTDDC_A", I2C_CLASS_DDC);
 	i++;
 
 	/* need to add the output busses for each device
@@ -159,9 +161,9 @@ void intelfb_create_i2c_busses(struct intelfb_info *dinfo)
 	case INTEL_865G:
 		dinfo->output[i].type = INTELFB_OUTPUT_DVO;
 		intelfb_setup_i2c_bus(dinfo, &dinfo->output[i].ddc_bus,
-				      GPIOD, "DVODDC_D");
+				      GPIOD, "DVODDC_D", I2C_CLASS_DDC);
 		intelfb_setup_i2c_bus(dinfo, &dinfo->output[i].i2c_bus,
-				      GPIOE, "DVOI2C_E");
+				      GPIOE, "DVOI2C_E", 0);
 		i++;
 		break;
 	case INTEL_915G:
@@ -174,7 +176,7 @@ void intelfb_create_i2c_busses(struct intelfb_info *dinfo)
 		/* SDVO ports have a single control bus - 2 devices */
 		dinfo->output[i].type = INTELFB_OUTPUT_SDVO;
 		intelfb_setup_i2c_bus(dinfo, &dinfo->output[i].i2c_bus,
-				      GPIOE, "SDVOCTRL_E");
+				      GPIOE, "SDVOCTRL_E", 0);
 		/* TODO: initialize the SDVO */
 		/* I830SDVOInit(pScrn, i, DVOB); */
 		i++;
diff --git a/drivers/video/matrox/i2c-matroxfb.c b/drivers/video/matrox/i2c-matroxfb.c
index 4baab7be58de..75ee5a12e549 100644
--- a/drivers/video/matrox/i2c-matroxfb.c
+++ b/drivers/video/matrox/i2c-matroxfb.c
@@ -104,7 +104,9 @@ static struct i2c_algo_bit_data matrox_i2c_algo_template =
 };
 
 static int i2c_bus_reg(struct i2c_bit_adapter* b, struct matrox_fb_info* minfo, 
-		unsigned int data, unsigned int clock, const char* name) {
+		unsigned int data, unsigned int clock, const char *name,
+		int class)
+{
 	int err;
 
 	b->minfo = minfo;
@@ -114,6 +116,7 @@ static int i2c_bus_reg(struct i2c_bit_adapter* b, struct matrox_fb_info* minfo,
 	snprintf(b->adapter.name, sizeof(b->adapter.name), name,
 		minfo->fbcon.node);
 	i2c_set_adapdata(&b->adapter, b);
+	b->adapter.class = class;
 	b->adapter.algo_data = &b->bac;
 	b->adapter.dev.parent = &ACCESS_FBINFO(pcidev)->dev;
 	b->bac = matrox_i2c_algo_template;
@@ -159,22 +162,29 @@ static void* i2c_matroxfb_probe(struct matrox_fb_info* minfo) {
 	switch (ACCESS_FBINFO(chip)) {
 		case MGA_2064:
 		case MGA_2164:
-			err = i2c_bus_reg(&m2info->ddc1, minfo, DDC1B_DATA, DDC1B_CLK, "DDC:fb%u #0");
+			err = i2c_bus_reg(&m2info->ddc1, minfo,
+					  DDC1B_DATA, DDC1B_CLK,
+					  "DDC:fb%u #0", I2C_CLASS_DDC);
 			break;
 		default:
-			err = i2c_bus_reg(&m2info->ddc1, minfo, DDC1_DATA, DDC1_CLK, "DDC:fb%u #0");
+			err = i2c_bus_reg(&m2info->ddc1, minfo,
+					  DDC1_DATA, DDC1_CLK,
+					  "DDC:fb%u #0", I2C_CLASS_DDC);
 			break;
 	}
 	if (err)
 		goto fail_ddc1;
 	if (ACCESS_FBINFO(devflags.dualhead)) {
-		err = i2c_bus_reg(&m2info->ddc2, minfo, DDC2_DATA, DDC2_CLK, "DDC:fb%u #1");
+		err = i2c_bus_reg(&m2info->ddc2, minfo,
+				  DDC2_DATA, DDC2_CLK,
+				  "DDC:fb%u #1", I2C_CLASS_DDC);
 		if (err == -ENODEV) {
 			printk(KERN_INFO "i2c-matroxfb: VGA->TV plug detected, DDC unavailable.\n");
 		} else if (err)
 			printk(KERN_INFO "i2c-matroxfb: Could not register secondary output i2c bus. Continuing anyway.\n");
 		/* Register maven bus even on G450/G550 */
-		err = i2c_bus_reg(&m2info->maven, minfo, MAT_DATA, MAT_CLK, "MAVEN:fb%u");
+		err = i2c_bus_reg(&m2info->maven, minfo,
+				  MAT_DATA, MAT_CLK, "MAVEN:fb%u", 0);
 		if (err)
 			printk(KERN_INFO "i2c-matroxfb: Could not register Maven i2c bus. Continuing anyway.\n");
 	}
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 7c36d5188d39..145797fe6a31 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -349,7 +349,7 @@ static inline void i2c_set_adapdata (struct i2c_adapter *dev, void *data)
 #define I2C_CLASS_HWMON		(1<<0)	/* lm_sensors, ... */
 #define I2C_CLASS_TV_ANALOG	(1<<1)	/* bttv + friends */
 #define I2C_CLASS_TV_DIGITAL	(1<<2)	/* dvb cards */
-#define I2C_CLASS_DDC		(1<<3)	/* i2c-matroxfb ? */
+#define I2C_CLASS_DDC		(1<<3)	/* DDC bus on graphics adapters */
 #define I2C_CLASS_CAM_ANALOG	(1<<4)	/* camera with analog CCD */
 #define I2C_CLASS_CAM_DIGITAL	(1<<5)	/* most webcams */
 #define I2C_CLASS_SOUND		(1<<6)	/* sound devices */
-- 
cgit v1.2.3


From 3401b2fff38fbb8b73ea6bcc69a8370ae5d2a7a0 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 14 Jul 2008 22:38:29 +0200
Subject: i2c: Let bus drivers add SPD to their class

Let general purpose I2C/SMBus bus drivers add SPD to their class. Once
this is done, we will be able to tell the eeprom driver to only probe
for SPD EEPROMs and similar on these buses.

Note that I took a conservative approach here, adding I2C_CLASS_SPD to
many drivers that have no idea whether they can host SPD EEPROMs or not.
This is to make sure that the eeprom driver doesn't stop probing buses
where SPD EEPROMs or equivalent live.

So, bus driver maintainers and users should feel free to remove the SPD
class from drivers those buses never have SPD EEPROMs or they don't
want the eeprom driver to bind to them. Likewise, feel free to add the
SPD class to any bus driver I might have missed.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/busses/i2c-ali1535.c | 2 +-
 drivers/i2c/busses/i2c-ali1563.c | 2 +-
 drivers/i2c/busses/i2c-ali15x3.c | 2 +-
 drivers/i2c/busses/i2c-amd756.c  | 2 +-
 drivers/i2c/busses/i2c-amd8111.c | 2 +-
 drivers/i2c/busses/i2c-cpm.c     | 2 +-
 drivers/i2c/busses/i2c-elektor.c | 2 +-
 drivers/i2c/busses/i2c-gpio.c    | 2 +-
 drivers/i2c/busses/i2c-i801.c    | 2 +-
 drivers/i2c/busses/i2c-ibm_iic.c | 4 ++--
 drivers/i2c/busses/i2c-iop3xx.c  | 2 +-
 drivers/i2c/busses/i2c-isch.c    | 2 +-
 drivers/i2c/busses/i2c-mpc.c     | 2 +-
 drivers/i2c/busses/i2c-mv64xxx.c | 2 +-
 drivers/i2c/busses/i2c-nforce2.c | 2 +-
 drivers/i2c/busses/i2c-ocores.c  | 2 +-
 drivers/i2c/busses/i2c-pasemi.c  | 2 +-
 drivers/i2c/busses/i2c-piix4.c   | 2 +-
 drivers/i2c/busses/i2c-pmcmsp.c  | 2 +-
 drivers/i2c/busses/i2c-s3c2410.c | 2 +-
 drivers/i2c/busses/i2c-sibyte.c  | 4 ++--
 drivers/i2c/busses/i2c-sis5595.c | 2 +-
 drivers/i2c/busses/i2c-sis630.c  | 2 +-
 drivers/i2c/busses/i2c-sis96x.c  | 2 +-
 drivers/i2c/busses/i2c-stub.c    | 2 +-
 drivers/i2c/busses/i2c-via.c     | 2 +-
 drivers/i2c/busses/i2c-viapro.c  | 2 +-
 drivers/i2c/busses/scx200_acb.c  | 2 +-
 include/linux/i2c.h              | 1 +
 29 files changed, 31 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/i2c/busses/i2c-ali1535.c b/drivers/i2c/busses/i2c-ali1535.c
index 704436cdec8e..8d1d90ab3a90 100644
--- a/drivers/i2c/busses/i2c-ali1535.c
+++ b/drivers/i2c/busses/i2c-ali1535.c
@@ -473,7 +473,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter ali1535_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_ALI1535,
-	.class          = I2C_CLASS_HWMON,
+	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-ali1563.c b/drivers/i2c/busses/i2c-ali1563.c
index da5a382eee93..4b55ae19db8d 100644
--- a/drivers/i2c/busses/i2c-ali1563.c
+++ b/drivers/i2c/busses/i2c-ali1563.c
@@ -382,7 +382,7 @@ static const struct i2c_algorithm ali1563_algorithm = {
 static struct i2c_adapter ali1563_adapter = {
 	.owner	= THIS_MODULE,
 	.id	= I2C_HW_SMBUS_ALI1563,
-	.class	= I2C_CLASS_HWMON,
+	.class	= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo	= &ali1563_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-ali15x3.c b/drivers/i2c/busses/i2c-ali15x3.c
index 7b029b147a8e..e922c3950fcd 100644
--- a/drivers/i2c/busses/i2c-ali15x3.c
+++ b/drivers/i2c/busses/i2c-ali15x3.c
@@ -471,7 +471,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter ali15x3_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_ALI15X3,
-	.class          = I2C_CLASS_HWMON,
+	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-amd756.c b/drivers/i2c/busses/i2c-amd756.c
index f0baea62067d..bd4f6380fabe 100644
--- a/drivers/i2c/busses/i2c-amd756.c
+++ b/drivers/i2c/busses/i2c-amd756.c
@@ -301,7 +301,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 struct i2c_adapter amd756_smbus = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_AMD756,
-	.class          = I2C_CLASS_HWMON,
+	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-amd8111.c b/drivers/i2c/busses/i2c-amd8111.c
index a4f687915de1..0e18fe846010 100644
--- a/drivers/i2c/busses/i2c-amd8111.c
+++ b/drivers/i2c/busses/i2c-amd8111.c
@@ -383,7 +383,7 @@ static int __devinit amd8111_probe(struct pci_dev *dev,
 	snprintf(smbus->adapter.name, sizeof(smbus->adapter.name),
 		"SMBus2 AMD8111 adapter at %04x", smbus->base);
 	smbus->adapter.id = I2C_HW_SMBUS_AMD8111;
-	smbus->adapter.class = I2C_CLASS_HWMON;
+	smbus->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	smbus->adapter.algo = &smbus_algorithm;
 	smbus->adapter.algo_data = smbus;
 
diff --git a/drivers/i2c/busses/i2c-cpm.c b/drivers/i2c/busses/i2c-cpm.c
index 53af744a91c1..8164de1f4d72 100644
--- a/drivers/i2c/busses/i2c-cpm.c
+++ b/drivers/i2c/busses/i2c-cpm.c
@@ -423,7 +423,7 @@ static const struct i2c_adapter cpm_ops = {
 	.owner		= THIS_MODULE,
 	.name		= "i2c-cpm",
 	.algo		= &cpm_i2c_algo,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 };
 
 static int __devinit cpm_i2c_setup(struct cpm_i2c *cpm)
diff --git a/drivers/i2c/busses/i2c-elektor.c b/drivers/i2c/busses/i2c-elektor.c
index b7a9977b025f..c251cf21a62b 100644
--- a/drivers/i2c/busses/i2c-elektor.c
+++ b/drivers/i2c/busses/i2c-elektor.c
@@ -202,7 +202,7 @@ static struct i2c_algo_pcf_data pcf_isa_data = {
 
 static struct i2c_adapter pcf_isa_ops = {
 	.owner		= THIS_MODULE,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.id		= I2C_HW_P_ELEK,
 	.algo_data	= &pcf_isa_data,
 	.name		= "i2c-elektor",
diff --git a/drivers/i2c/busses/i2c-gpio.c b/drivers/i2c/busses/i2c-gpio.c
index 7c1b762aa681..79b455a1f090 100644
--- a/drivers/i2c/busses/i2c-gpio.c
+++ b/drivers/i2c/busses/i2c-gpio.c
@@ -140,7 +140,7 @@ static int __init i2c_gpio_probe(struct platform_device *pdev)
 	adap->owner = THIS_MODULE;
 	snprintf(adap->name, sizeof(adap->name), "i2c-gpio%d", pdev->id);
 	adap->algo_data = bit_data;
-	adap->class = I2C_CLASS_HWMON;
+	adap->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	adap->dev.parent = &pdev->dev;
 
 	/*
diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c
index 213119211e58..9717ffe12921 100644
--- a/drivers/i2c/busses/i2c-i801.c
+++ b/drivers/i2c/busses/i2c-i801.c
@@ -573,7 +573,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter i801_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_I801,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-ibm_iic.c b/drivers/i2c/busses/i2c-ibm_iic.c
index 85dbf34382e1..6f7bfdec3c69 100644
--- a/drivers/i2c/busses/i2c-ibm_iic.c
+++ b/drivers/i2c/busses/i2c-ibm_iic.c
@@ -740,7 +740,7 @@ static int __devinit iic_probe(struct ocp_device *ocp){
 	strcpy(adap->name, "IBM IIC");
 	i2c_set_adapdata(adap, dev);
 	adap->id = I2C_HW_OCP;
-	adap->class = I2C_CLASS_HWMON;
+	adap->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	adap->algo = &iic_algo;
 	adap->client_register = NULL;
 	adap->client_unregister = NULL;
@@ -934,7 +934,7 @@ static int __devinit iic_probe(struct of_device *ofdev,
 	strlcpy(adap->name, "IBM IIC", sizeof(adap->name));
 	i2c_set_adapdata(adap, dev);
 	adap->id = I2C_HW_OCP;
-	adap->class = I2C_CLASS_HWMON;
+	adap->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	adap->algo = &iic_algo;
 	adap->timeout = 1;
 	adap->nr = dev->idx;
diff --git a/drivers/i2c/busses/i2c-iop3xx.c b/drivers/i2c/busses/i2c-iop3xx.c
index 39884e797594..fc2714ac0c0f 100644
--- a/drivers/i2c/busses/i2c-iop3xx.c
+++ b/drivers/i2c/busses/i2c-iop3xx.c
@@ -482,7 +482,7 @@ iop3xx_i2c_probe(struct platform_device *pdev)
 	memcpy(new_adapter->name, pdev->name, strlen(pdev->name));
 	new_adapter->id = I2C_HW_IOP3XX;
 	new_adapter->owner = THIS_MODULE;
-	new_adapter->class = I2C_CLASS_HWMON;
+	new_adapter->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	new_adapter->dev.parent = &pdev->dev;
 	new_adapter->nr = pdev->id;
 
diff --git a/drivers/i2c/busses/i2c-isch.c b/drivers/i2c/busses/i2c-isch.c
index c9cd46b22692..8d648911a7f5 100644
--- a/drivers/i2c/busses/i2c-isch.c
+++ b/drivers/i2c/busses/i2c-isch.c
@@ -251,7 +251,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter sch_adapter = {
 	.owner		= THIS_MODULE,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-mpc.c b/drivers/i2c/busses/i2c-mpc.c
index a076129de7e8..10b9342a36c2 100644
--- a/drivers/i2c/busses/i2c-mpc.c
+++ b/drivers/i2c/busses/i2c-mpc.c
@@ -311,7 +311,7 @@ static struct i2c_adapter mpc_ops = {
 	.name = "MPC adapter",
 	.id = I2C_HW_MPC107,
 	.algo = &mpc_algo,
-	.class = I2C_CLASS_HWMON,
+	.class = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.timeout = 1,
 };
 
diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c
index 036e6a883e67..9e8118d2fe64 100644
--- a/drivers/i2c/busses/i2c-mv64xxx.c
+++ b/drivers/i2c/busses/i2c-mv64xxx.c
@@ -530,7 +530,7 @@ mv64xxx_i2c_probe(struct platform_device *pd)
 	drv_data->adapter.id = I2C_HW_MV64XXX;
 	drv_data->adapter.algo = &mv64xxx_i2c_algo;
 	drv_data->adapter.owner = THIS_MODULE;
-	drv_data->adapter.class = I2C_CLASS_HWMON;
+	drv_data->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	drv_data->adapter.timeout = pdata->timeout;
 	drv_data->adapter.nr = pd->id;
 	platform_set_drvdata(pd, drv_data);
diff --git a/drivers/i2c/busses/i2c-nforce2.c b/drivers/i2c/busses/i2c-nforce2.c
index 081fdf3393f4..2654f20d3a62 100644
--- a/drivers/i2c/busses/i2c-nforce2.c
+++ b/drivers/i2c/busses/i2c-nforce2.c
@@ -350,7 +350,7 @@ static int __devinit nforce2_probe_smb (struct pci_dev *dev, int bar,
 	}
 	smbus->adapter.owner = THIS_MODULE;
 	smbus->adapter.id = I2C_HW_SMBUS_NFORCE2;
-	smbus->adapter.class = I2C_CLASS_HWMON;
+	smbus->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	smbus->adapter.algo = &smbus_algorithm;
 	smbus->adapter.algo_data = smbus;
 	smbus->adapter.dev.parent = &dev->dev;
diff --git a/drivers/i2c/busses/i2c-ocores.c b/drivers/i2c/busses/i2c-ocores.c
index f145692cbb76..51ca79bf6480 100644
--- a/drivers/i2c/busses/i2c-ocores.c
+++ b/drivers/i2c/busses/i2c-ocores.c
@@ -205,7 +205,7 @@ static const struct i2c_algorithm ocores_algorithm = {
 static struct i2c_adapter ocores_adapter = {
 	.owner		= THIS_MODULE,
 	.name		= "i2c-ocores",
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &ocores_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-pasemi.c b/drivers/i2c/busses/i2c-pasemi.c
index 1603c81e39d4..adf0fbb902f0 100644
--- a/drivers/i2c/busses/i2c-pasemi.c
+++ b/drivers/i2c/busses/i2c-pasemi.c
@@ -365,7 +365,7 @@ static int __devinit pasemi_smb_probe(struct pci_dev *dev,
 	smbus->adapter.owner = THIS_MODULE;
 	snprintf(smbus->adapter.name, sizeof(smbus->adapter.name),
 		 "PA Semi SMBus adapter at 0x%lx", smbus->base);
-	smbus->adapter.class = I2C_CLASS_HWMON;
+	smbus->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	smbus->adapter.algo = &smbus_algorithm;
 	smbus->adapter.algo_data = smbus;
 	smbus->adapter.nr = PCI_FUNC(dev->devfn);
diff --git a/drivers/i2c/busses/i2c-piix4.c b/drivers/i2c/busses/i2c-piix4.c
index 2bde47509e1a..85d69f3e624f 100644
--- a/drivers/i2c/busses/i2c-piix4.c
+++ b/drivers/i2c/busses/i2c-piix4.c
@@ -402,7 +402,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter piix4_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_PIIX4,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-pmcmsp.c b/drivers/i2c/busses/i2c-pmcmsp.c
index 63b3e2c11cff..dcf2045b5222 100644
--- a/drivers/i2c/busses/i2c-pmcmsp.c
+++ b/drivers/i2c/busses/i2c-pmcmsp.c
@@ -622,7 +622,7 @@ static struct i2c_algorithm pmcmsptwi_algo = {
 
 static struct i2c_adapter pmcmsptwi_adapter = {
 	.owner		= THIS_MODULE,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &pmcmsptwi_algo,
 	.name		= DRV_NAME,
 };
diff --git a/drivers/i2c/busses/i2c-s3c2410.c b/drivers/i2c/busses/i2c-s3c2410.c
index 9e8c875437be..007390ad9810 100644
--- a/drivers/i2c/busses/i2c-s3c2410.c
+++ b/drivers/i2c/busses/i2c-s3c2410.c
@@ -590,7 +590,7 @@ static struct s3c24xx_i2c s3c24xx_i2c = {
 		.owner			= THIS_MODULE,
 		.algo			= &s3c24xx_i2c_algorithm,
 		.retries		= 2,
-		.class			= I2C_CLASS_HWMON,
+		.class			= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	},
 };
 
diff --git a/drivers/i2c/busses/i2c-sibyte.c b/drivers/i2c/busses/i2c-sibyte.c
index 114634da6c6e..ac8822e7a5b4 100644
--- a/drivers/i2c/busses/i2c-sibyte.c
+++ b/drivers/i2c/busses/i2c-sibyte.c
@@ -156,7 +156,7 @@ static struct i2c_adapter sibyte_board_adapter[2] = {
 	{
 		.owner		= THIS_MODULE,
 		.id		= I2C_HW_SIBYTE,
-		.class		= I2C_CLASS_HWMON,
+		.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 		.algo		= NULL,
 		.algo_data	= &sibyte_board_data[0],
 		.name		= "SiByte SMBus 0",
@@ -164,7 +164,7 @@ static struct i2c_adapter sibyte_board_adapter[2] = {
 	{
 		.owner		= THIS_MODULE,
 		.id		= I2C_HW_SIBYTE,
-		.class		= I2C_CLASS_HWMON,
+		.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 		.algo		= NULL,
 		.algo_data	= &sibyte_board_data[1],
 		.name		= "SiByte SMBus 1",
diff --git a/drivers/i2c/busses/i2c-sis5595.c b/drivers/i2c/busses/i2c-sis5595.c
index 328441bb5470..f76944b384f5 100644
--- a/drivers/i2c/busses/i2c-sis5595.c
+++ b/drivers/i2c/busses/i2c-sis5595.c
@@ -362,7 +362,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter sis5595_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_SIS5595,
-	.class          = I2C_CLASS_HWMON,
+	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-sis630.c b/drivers/i2c/busses/i2c-sis630.c
index d7e6ff3e0187..eb2b2181fed7 100644
--- a/drivers/i2c/busses/i2c-sis630.c
+++ b/drivers/i2c/busses/i2c-sis630.c
@@ -462,7 +462,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter sis630_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_SIS630,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-sis96x.c b/drivers/i2c/busses/i2c-sis96x.c
index cde8e5880368..413e9e477723 100644
--- a/drivers/i2c/busses/i2c-sis96x.c
+++ b/drivers/i2c/busses/i2c-sis96x.c
@@ -244,7 +244,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter sis96x_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_SIS96X,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/i2c-stub.c b/drivers/i2c/busses/i2c-stub.c
index e37ccd80f77a..1b7b2af94036 100644
--- a/drivers/i2c/busses/i2c-stub.c
+++ b/drivers/i2c/busses/i2c-stub.c
@@ -140,7 +140,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter stub_adapter = {
 	.owner		= THIS_MODULE,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 	.name		= "SMBus stub driver",
 };
diff --git a/drivers/i2c/busses/i2c-via.c b/drivers/i2c/busses/i2c-via.c
index 61716f6b14dc..6517f8a6d911 100644
--- a/drivers/i2c/busses/i2c-via.c
+++ b/drivers/i2c/busses/i2c-via.c
@@ -87,7 +87,7 @@ static struct i2c_algo_bit_data bit_data = {
 static struct i2c_adapter vt586b_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_B_VIA,
-	.class          = I2C_CLASS_HWMON,
+	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.name		= "VIA i2c",
 	.algo_data	= &bit_data,
 };
diff --git a/drivers/i2c/busses/i2c-viapro.c b/drivers/i2c/busses/i2c-viapro.c
index c611905df009..7957ce515891 100644
--- a/drivers/i2c/busses/i2c-viapro.c
+++ b/drivers/i2c/busses/i2c-viapro.c
@@ -311,7 +311,7 @@ static const struct i2c_algorithm smbus_algorithm = {
 static struct i2c_adapter vt596_adapter = {
 	.owner		= THIS_MODULE,
 	.id		= I2C_HW_SMBUS_VIA2,
-	.class		= I2C_CLASS_HWMON,
+	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
 
diff --git a/drivers/i2c/busses/scx200_acb.c b/drivers/i2c/busses/scx200_acb.c
index 61abe0f33255..ed794b145a11 100644
--- a/drivers/i2c/busses/scx200_acb.c
+++ b/drivers/i2c/busses/scx200_acb.c
@@ -442,7 +442,7 @@ static __init struct scx200_acb_iface *scx200_create_iface(const char *text,
 	adapter->owner = THIS_MODULE;
 	adapter->id = I2C_HW_SMBUS_SCX200;
 	adapter->algo = &scx200_acb_algorithm;
-	adapter->class = I2C_CLASS_HWMON;
+	adapter->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	adapter->dev.parent = dev;
 
 	mutex_init(&iface->mutex);
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 145797fe6a31..839d0ea3dca3 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -353,6 +353,7 @@ static inline void i2c_set_adapdata (struct i2c_adapter *dev, void *data)
 #define I2C_CLASS_CAM_ANALOG	(1<<4)	/* camera with analog CCD */
 #define I2C_CLASS_CAM_DIGITAL	(1<<5)	/* most webcams */
 #define I2C_CLASS_SOUND		(1<<6)	/* sound devices */
+#define I2C_CLASS_SPD		(1<<7)	/* SPD EEPROMs and similar */
 #define I2C_CLASS_ALL		(UINT_MAX) /* all of the above */
 
 /* i2c_client_address_data is the struct for holding default client
-- 
cgit v1.2.3


From 0573d11b2bbd0e4774f33f4c1959c1939c055e96 Mon Sep 17 00:00:00 2001
From: Eric Brower <ebrower@gmail.com>
Date: Mon, 14 Jul 2008 22:38:31 +0200
Subject: i2c-algo-pcf: Multi-master lost-arbitration improvement

Improve lost-arbitration handling of PCF8584.  This is necessary for
support of a currently out-of-kernel driver for Sun Microsystems E250
environmental management; perhaps others.

Signed-off-by: Eric Brower <ebrower@gmail.com>
Acked-by: Dan Smolik <marvin@mydatex.cz>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/algos/i2c-algo-pcf.c | 48 ++++++++++++++++++++++++++--------------
 include/linux/i2c-algo-pcf.h     |  6 +++++
 2 files changed, 37 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/i2c/algos/i2c-algo-pcf.c b/drivers/i2c/algos/i2c-algo-pcf.c
index 8907b0191677..1e328d19cd6d 100644
--- a/drivers/i2c/algos/i2c-algo-pcf.c
+++ b/drivers/i2c/algos/i2c-algo-pcf.c
@@ -78,6 +78,36 @@ static void i2c_stop(struct i2c_algo_pcf_data *adap)
 	set_pcf(adap, 1, I2C_PCF_STOP);
 }
 
+static void handle_lab(struct i2c_algo_pcf_data *adap, const int *status)
+{
+	DEB2(printk(KERN_INFO
+		"i2c-algo-pcf.o: lost arbitration (CSR 0x%02x)\n",
+		 *status));
+
+	/* Cleanup from LAB -- reset and enable ESO.
+	 * This resets the PCF8584; since we've lost the bus, no
+	 * further attempts should be made by callers to clean up
+	 * (no i2c_stop() etc.)
+	 */
+	set_pcf(adap, 1, I2C_PCF_PIN);
+	set_pcf(adap, 1, I2C_PCF_ESO);
+
+	/* We pause for a time period sufficient for any running
+	 * I2C transaction to complete -- the arbitration logic won't
+	 * work properly until the next START is seen.
+	 * It is assumed the bus driver or client has set a proper value.
+	 *
+	 * REVISIT: should probably use msleep instead of mdelay if we
+	 * know we can sleep.
+	 */
+	if (adap->lab_mdelay)
+		mdelay(adap->lab_mdelay);
+
+	DEB2(printk(KERN_INFO
+		"i2c-algo-pcf.o: reset LAB condition (CSR 0x%02x)\n",
+		get_pcf(adap, 1)));
+}
+
 static int wait_for_bb(struct i2c_algo_pcf_data *adap) {
 
 	int timeout = DEF_TIMEOUT;
@@ -109,23 +139,7 @@ static int wait_for_pin(struct i2c_algo_pcf_data *adap, int *status) {
 		*status = get_pcf(adap, 1);
 	}
 	if (*status & I2C_PCF_LAB) {
-		DEB2(printk(KERN_INFO 
-			"i2c-algo-pcf.o: lost arbitration (CSR 0x%02x)\n",
-			 *status));
-		/* Cleanup from LAB-- reset and enable ESO.
-		 * This resets the PCF8584; since we've lost the bus, no
-		 * further attempts should be made by callers to clean up 
-		 * (no i2c_stop() etc.)
-		 */
-		set_pcf(adap, 1, I2C_PCF_PIN);
-		set_pcf(adap, 1, I2C_PCF_ESO);
-		/* TODO: we should pause for a time period sufficient for any
-		 * running I2C transaction to complete-- the arbitration
-		 * logic won't work properly until the next START is seen.
-		 */
-		DEB2(printk(KERN_INFO 
-			"i2c-algo-pcf.o: reset LAB condition (CSR 0x%02x)\n", 
-			get_pcf(adap,1)));
+		handle_lab(adap, status);
 		return(-EINTR);
 	}
 #endif
diff --git a/include/linux/i2c-algo-pcf.h b/include/linux/i2c-algo-pcf.h
index 77afbb60fd11..74fb6f889a77 100644
--- a/include/linux/i2c-algo-pcf.h
+++ b/include/linux/i2c-algo-pcf.h
@@ -36,6 +36,12 @@ struct i2c_algo_pcf_data {
 	/* local settings */
 	int udelay;
 	int timeout;
+
+	/* Multi-master lost arbitration back-off delay (msecs)
+	 * This should be set by the bus adapter or knowledgable client
+	 * if bus is multi-mastered, else zero
+	 */
+	unsigned long lab_mdelay;
 };
 
 int i2c_pcf_add_bus(struct i2c_adapter *);
-- 
cgit v1.2.3


From e3e7fc3c401a5d53f0599a357b3cf65d6a4f52e3 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 14 Jul 2008 22:38:31 +0200
Subject: i2c-algo-pcf: Drop unused struct members

Struct members udelay and timeout aren't used anywhere, so drop them.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Eric Brower <ebrower@gmail.com>
---
 drivers/i2c/busses/i2c-elektor.c | 2 --
 include/linux/i2c-algo-pcf.h     | 4 ----
 2 files changed, 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/i2c/busses/i2c-elektor.c b/drivers/i2c/busses/i2c-elektor.c
index c251cf21a62b..7f38c01fb3a0 100644
--- a/drivers/i2c/busses/i2c-elektor.c
+++ b/drivers/i2c/busses/i2c-elektor.c
@@ -196,8 +196,6 @@ static struct i2c_algo_pcf_data pcf_isa_data = {
 	.getown	    = pcf_isa_getown,
 	.getclock   = pcf_isa_getclock,
 	.waitforpin = pcf_isa_waitforpin,
-	.udelay	    = 10,
-	.timeout    = 100,
 };
 
 static struct i2c_adapter pcf_isa_ops = {
diff --git a/include/linux/i2c-algo-pcf.h b/include/linux/i2c-algo-pcf.h
index 74fb6f889a77..0177d280f733 100644
--- a/include/linux/i2c-algo-pcf.h
+++ b/include/linux/i2c-algo-pcf.h
@@ -33,10 +33,6 @@ struct i2c_algo_pcf_data {
 	int  (*getclock) (void *data);
 	void (*waitforpin) (void);
 
-	/* local settings */
-	int udelay;
-	int timeout;
-
 	/* Multi-master lost arbitration back-off delay (msecs)
 	 * This should be set by the bus adapter or knowledgable client
 	 * if bus is multi-mastered, else zero
-- 
cgit v1.2.3


From f6a7110520037ba786f17b53790c6eb8a3d4ef55 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 14 Jul 2008 22:38:34 +0200
Subject: i2c-dev: Delete empty detach_client callback

Implementing detach_client is optional, so there is no point in
an empty implementation.

Likewise, i2c driver IDs are optional, and we don't need one.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/i2c-dev.c  | 7 -------
 include/linux/i2c-id.h | 2 --
 2 files changed, 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index e96d98696782..50df53640c78 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -548,19 +548,12 @@ static int i2cdev_detach_adapter(struct i2c_adapter *adap)
 	return 0;
 }
 
-static int i2cdev_detach_client(struct i2c_client *client)
-{
-	return 0;
-}
-
 static struct i2c_driver i2cdev_driver = {
 	.driver = {
 		.name	= "dev_driver",
 	},
-	.id		= I2C_DRIVERID_I2CDEV,
 	.attach_adapter	= i2cdev_attach_adapter,
 	.detach_adapter	= i2cdev_detach_adapter,
-	.detach_client	= i2cdev_detach_client,
 };
 
 /* ------------------------------------------------------------------------- */
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 988e566d3ed5..ef13b7c66df3 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -91,8 +91,6 @@
 #define I2C_DRIVERID_M52790 	95      /* Mitsubishi M52790SP/FP AV switch */
 #define I2C_DRIVERID_CS5345	96	/* cs5345 audio processor	*/
 
-#define I2C_DRIVERID_I2CDEV	900
-
 #define I2C_DRIVERID_OV7670 1048	/* Omnivision 7670 camera */
 
 /*
-- 
cgit v1.2.3


From e9ca9eb9d7fc7bf3dc3cec5ba7edb089c4625f7b Mon Sep 17 00:00:00 2001
From: Jon Smirl <jonsmirl@gmail.com>
Date: Mon, 14 Jul 2008 22:38:35 +0200
Subject: i2c: Export the i2c_bus_type symbol

Export the root of the i2c bus so that PowerPC device tree code can
iterate over devices on the i2c bus.

Signed-off-by: Jon Smirl <jonsmirl@gmail.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/i2c-core.c | 3 ++-
 include/linux/i2c.h    | 2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index d6cc58abf3ff..e45bb2838f42 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -201,7 +201,7 @@ static struct device_attribute i2c_dev_attrs[] = {
 	{ },
 };
 
-static struct bus_type i2c_bus_type = {
+struct bus_type i2c_bus_type = {
 	.name		= "i2c",
 	.dev_attrs	= i2c_dev_attrs,
 	.match		= i2c_device_match,
@@ -212,6 +212,7 @@ static struct bus_type i2c_bus_type = {
 	.suspend	= i2c_device_suspend,
 	.resume		= i2c_device_resume,
 };
+EXPORT_SYMBOL_GPL(i2c_bus_type);
 
 
 /**
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 839d0ea3dca3..50cbab4b62b0 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -35,6 +35,8 @@
 #include <linux/sched.h>	/* for completion */
 #include <linux/mutex.h>
 
+extern struct bus_type i2c_bus_type;
+
 /* --- General options ------------------------------------------------	*/
 
 struct i2c_msg;
-- 
cgit v1.2.3


From 2b7a5056a0a7ff17d5d2004c29c852a92a6bd632 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <w.sang@pengutronix.de>
Date: Mon, 14 Jul 2008 22:38:35 +0200
Subject: i2c: New-style EEPROM driver using device IDs

Add a new-style driver for most I2C EEPROMs, giving sysfs read/write
access to their data. Tested with various chips and clock rates.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/i2c/chips/Kconfig  |  26 ++
 drivers/i2c/chips/Makefile |   1 +
 drivers/i2c/chips/at24.c   | 583 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/i2c/at24.h   |  28 +++
 4 files changed, 638 insertions(+)
 create mode 100644 drivers/i2c/chips/at24.c
 create mode 100644 include/linux/i2c/at24.h

(limited to 'include/linux')

diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig
index 6326468d5f0b..50e0a4653741 100644
--- a/drivers/i2c/chips/Kconfig
+++ b/drivers/i2c/chips/Kconfig
@@ -14,6 +14,32 @@ config DS1682
 	  This driver can also be built as a module.  If so, the module
 	  will be called ds1682.
 
+config AT24
+	tristate "EEPROMs from most vendors"
+	depends on SYSFS && EXPERIMENTAL
+	help
+	  Enable this driver to get read/write support to most I2C EEPROMs,
+	  after you configure the driver to know about each EEPROM on
+	  your target board.  Use these generic chip names, instead of
+	  vendor-specific ones like at24c64 or 24lc02:
+
+	     24c00, 24c01, 24c02, spd (readonly 24c02), 24c04, 24c08,
+	     24c16, 24c32, 24c64, 24c128, 24c256, 24c512, 24c1024
+
+	  Unless you like data loss puzzles, always be sure that any chip
+	  you configure as a 24c32 (32 kbit) or larger is NOT really a
+	  24c16 (16 kbit) or smaller, and vice versa. Marking the chip
+	  as read-only won't help recover from this. Also, if your chip
+	  has any software write-protect mechanism you may want to review the
+	  code to make sure this driver won't turn it on by accident.
+
+	  If you use this with an SMBus adapter instead of an I2C adapter,
+	  full functionality is not available.  Only smaller devices are
+	  supported (24c16 and below, max 4 kByte).
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called at24.
+
 config SENSORS_EEPROM
 	tristate "EEPROM reader"
 	depends on EXPERIMENTAL
diff --git a/drivers/i2c/chips/Makefile b/drivers/i2c/chips/Makefile
index e47aca0ca5ae..39e3e69ed125 100644
--- a/drivers/i2c/chips/Makefile
+++ b/drivers/i2c/chips/Makefile
@@ -10,6 +10,7 @@
 #
 
 obj-$(CONFIG_DS1682)		+= ds1682.o
+obj-$(CONFIG_AT24)		+= at24.o
 obj-$(CONFIG_SENSORS_EEPROM)	+= eeprom.o
 obj-$(CONFIG_SENSORS_MAX6875)	+= max6875.o
 obj-$(CONFIG_SENSORS_PCA9539)	+= pca9539.o
diff --git a/drivers/i2c/chips/at24.c b/drivers/i2c/chips/at24.c
new file mode 100644
index 000000000000..e764c94f3e3d
--- /dev/null
+++ b/drivers/i2c/chips/at24.c
@@ -0,0 +1,583 @@
+/*
+ * at24.c - handle most I2C EEPROMs
+ *
+ * Copyright (C) 2005-2007 David Brownell
+ * Copyright (C) 2008 Wolfram Sang, Pengutronix
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/mutex.h>
+#include <linux/sysfs.h>
+#include <linux/mod_devicetable.h>
+#include <linux/log2.h>
+#include <linux/bitops.h>
+#include <linux/jiffies.h>
+#include <linux/i2c.h>
+#include <linux/i2c/at24.h>
+
+/*
+ * I2C EEPROMs from most vendors are inexpensive and mostly interchangeable.
+ * Differences between different vendor product lines (like Atmel AT24C or
+ * MicroChip 24LC, etc) won't much matter for typical read/write access.
+ * There are also I2C RAM chips, likewise interchangeable. One example
+ * would be the PCF8570, which acts like a 24c02 EEPROM (256 bytes).
+ *
+ * However, misconfiguration can lose data. "Set 16-bit memory address"
+ * to a part with 8-bit addressing will overwrite data. Writing with too
+ * big a page size also loses data. And it's not safe to assume that the
+ * conventional addresses 0x50..0x57 only hold eeproms; a PCF8563 RTC
+ * uses 0x51, for just one example.
+ *
+ * Accordingly, explicit board-specific configuration data should be used
+ * in almost all cases. (One partial exception is an SMBus used to access
+ * "SPD" data for DRAM sticks. Those only use 24c02 EEPROMs.)
+ *
+ * So this driver uses "new style" I2C driver binding, expecting to be
+ * told what devices exist. That may be in arch/X/mach-Y/board-Z.c or
+ * similar kernel-resident tables; or, configuration data coming from
+ * a bootloader.
+ *
+ * Other than binding model, current differences from "eeprom" driver are
+ * that this one handles write access and isn't restricted to 24c02 devices.
+ * It also handles larger devices (32 kbit and up) with two-byte addresses,
+ * which won't work on pure SMBus systems.
+ */
+
+struct at24_data {
+	struct at24_platform_data chip;
+	bool use_smbus;
+
+	/*
+	 * Lock protects against activities from other Linux tasks,
+	 * but not from changes by other I2C masters.
+	 */
+	struct mutex lock;
+	struct bin_attribute bin;
+
+	u8 *writebuf;
+	unsigned write_max;
+	unsigned num_addresses;
+
+	/*
+	 * Some chips tie up multiple I2C addresses; dummy devices reserve
+	 * them for us, and we'll use them with SMBus calls.
+	 */
+	struct i2c_client *client[];
+};
+
+/*
+ * This parameter is to help this driver avoid blocking other drivers out
+ * of I2C for potentially troublesome amounts of time. With a 100 kHz I2C
+ * clock, one 256 byte read takes about 1/43 second which is excessive;
+ * but the 1/170 second it takes at 400 kHz may be quite reasonable; and
+ * at 1 MHz (Fm+) a 1/430 second delay could easily be invisible.
+ *
+ * This value is forced to be a power of two so that writes align on pages.
+ */
+static unsigned io_limit = 128;
+module_param(io_limit, uint, 0);
+MODULE_PARM_DESC(io_limit, "Maximum bytes per I/O (default 128)");
+
+/*
+ * Specs often allow 5 msec for a page write, sometimes 20 msec;
+ * it's important to recover from write timeouts.
+ */
+static unsigned write_timeout = 25;
+module_param(write_timeout, uint, 0);
+MODULE_PARM_DESC(write_timeout, "Time (in ms) to try writes (default 25)");
+
+#define AT24_SIZE_BYTELEN 5
+#define AT24_SIZE_FLAGS 8
+
+#define AT24_BITMASK(x) (BIT(x) - 1)
+
+/* create non-zero magic value for given eeprom parameters */
+#define AT24_DEVICE_MAGIC(_len, _flags) 		\
+	((1 << AT24_SIZE_FLAGS | (_flags)) 		\
+	    << AT24_SIZE_BYTELEN | ilog2(_len))
+
+static const struct i2c_device_id at24_ids[] = {
+	/* needs 8 addresses as A0-A2 are ignored */
+	{ "24c00", AT24_DEVICE_MAGIC(128 / 8, AT24_FLAG_TAKE8ADDR) },
+	/* old variants can't be handled with this generic entry! */
+	{ "24c01", AT24_DEVICE_MAGIC(1024 / 8, 0) },
+	{ "24c02", AT24_DEVICE_MAGIC(2048 / 8, 0) },
+	/* spd is a 24c02 in memory DIMMs */
+	{ "spd", AT24_DEVICE_MAGIC(2048 / 8,
+		AT24_FLAG_READONLY | AT24_FLAG_IRUGO) },
+	{ "24c04", AT24_DEVICE_MAGIC(4096 / 8, 0) },
+	/* 24rf08 quirk is handled at i2c-core */
+	{ "24c08", AT24_DEVICE_MAGIC(8192 / 8, 0) },
+	{ "24c16", AT24_DEVICE_MAGIC(16384 / 8, 0) },
+	{ "24c32", AT24_DEVICE_MAGIC(32768 / 8, AT24_FLAG_ADDR16) },
+	{ "24c64", AT24_DEVICE_MAGIC(65536 / 8, AT24_FLAG_ADDR16) },
+	{ "24c128", AT24_DEVICE_MAGIC(131072 / 8, AT24_FLAG_ADDR16) },
+	{ "24c256", AT24_DEVICE_MAGIC(262144 / 8, AT24_FLAG_ADDR16) },
+	{ "24c512", AT24_DEVICE_MAGIC(524288 / 8, AT24_FLAG_ADDR16) },
+	{ "24c1024", AT24_DEVICE_MAGIC(1048576 / 8, AT24_FLAG_ADDR16) },
+	{ "at24", 0 },
+	{ /* END OF LIST */ }
+};
+MODULE_DEVICE_TABLE(i2c, at24_ids);
+
+/*-------------------------------------------------------------------------*/
+
+/*
+ * This routine supports chips which consume multiple I2C addresses. It
+ * computes the addressing information to be used for a given r/w request.
+ * Assumes that sanity checks for offset happened at sysfs-layer.
+ */
+static struct i2c_client *at24_translate_offset(struct at24_data *at24,
+		unsigned *offset)
+{
+	unsigned i;
+
+	if (at24->chip.flags & AT24_FLAG_ADDR16) {
+		i = *offset >> 16;
+		*offset &= 0xffff;
+	} else {
+		i = *offset >> 8;
+		*offset &= 0xff;
+	}
+
+	return at24->client[i];
+}
+
+static ssize_t at24_eeprom_read(struct at24_data *at24, char *buf,
+		unsigned offset, size_t count)
+{
+	struct i2c_msg msg[2];
+	u8 msgbuf[2];
+	struct i2c_client *client;
+	int status, i;
+
+	memset(msg, 0, sizeof(msg));
+
+	/*
+	 * REVISIT some multi-address chips don't rollover page reads to
+	 * the next slave address, so we may need to truncate the count.
+	 * Those chips might need another quirk flag.
+	 *
+	 * If the real hardware used four adjacent 24c02 chips and that
+	 * were misconfigured as one 24c08, that would be a similar effect:
+	 * one "eeprom" file not four, but larger reads would fail when
+	 * they crossed certain pages.
+	 */
+
+	/*
+	 * Slave address and byte offset derive from the offset. Always
+	 * set the byte address; on a multi-master board, another master
+	 * may have changed the chip's "current" address pointer.
+	 */
+	client = at24_translate_offset(at24, &offset);
+
+	if (count > io_limit)
+		count = io_limit;
+
+	/* Smaller eeproms can work given some SMBus extension calls */
+	if (at24->use_smbus) {
+		if (count > I2C_SMBUS_BLOCK_MAX)
+			count = I2C_SMBUS_BLOCK_MAX;
+		status = i2c_smbus_read_i2c_block_data(client, offset,
+				count, buf);
+		dev_dbg(&client->dev, "smbus read %zd@%d --> %d\n",
+				count, offset, status);
+		return (status < 0) ? -EIO : status;
+	}
+
+	/*
+	 * When we have a better choice than SMBus calls, use a combined
+	 * I2C message. Write address; then read up to io_limit data bytes.
+	 * Note that read page rollover helps us here (unlike writes).
+	 * msgbuf is u8 and will cast to our needs.
+	 */
+	i = 0;
+	if (at24->chip.flags & AT24_FLAG_ADDR16)
+		msgbuf[i++] = offset >> 8;
+	msgbuf[i++] = offset;
+
+	msg[0].addr = client->addr;
+	msg[0].buf = msgbuf;
+	msg[0].len = i;
+
+	msg[1].addr = client->addr;
+	msg[1].flags = I2C_M_RD;
+	msg[1].buf = buf;
+	msg[1].len = count;
+
+	status = i2c_transfer(client->adapter, msg, 2);
+	dev_dbg(&client->dev, "i2c read %zd@%d --> %d\n",
+			count, offset, status);
+
+	if (status == 2)
+		return count;
+	else if (status >= 0)
+		return -EIO;
+	else
+		return status;
+}
+
+static ssize_t at24_bin_read(struct kobject *kobj, struct bin_attribute *attr,
+		char *buf, loff_t off, size_t count)
+{
+	struct at24_data *at24;
+	ssize_t retval = 0;
+
+	at24 = dev_get_drvdata(container_of(kobj, struct device, kobj));
+
+	if (unlikely(!count))
+		return count;
+
+	/*
+	 * Read data from chip, protecting against concurrent updates
+	 * from this host, but not from other I2C masters.
+	 */
+	mutex_lock(&at24->lock);
+
+	while (count) {
+		ssize_t	status;
+
+		status = at24_eeprom_read(at24, buf, off, count);
+		if (status <= 0) {
+			if (retval == 0)
+				retval = status;
+			break;
+		}
+		buf += status;
+		off += status;
+		count -= status;
+		retval += status;
+	}
+
+	mutex_unlock(&at24->lock);
+
+	return retval;
+}
+
+
+/*
+ * REVISIT: export at24_bin{read,write}() to let other kernel code use
+ * eeprom data. For example, it might hold a board's Ethernet address, or
+ * board-specific calibration data generated on the manufacturing floor.
+ */
+
+
+/*
+ * Note that if the hardware write-protect pin is pulled high, the whole
+ * chip is normally write protected. But there are plenty of product
+ * variants here, including OTP fuses and partial chip protect.
+ *
+ * We only use page mode writes; the alternative is sloooow. This routine
+ * writes at most one page.
+ */
+static ssize_t at24_eeprom_write(struct at24_data *at24, char *buf,
+		unsigned offset, size_t count)
+{
+	struct i2c_client *client;
+	struct i2c_msg msg;
+	ssize_t status;
+	unsigned long timeout, write_time;
+	unsigned next_page;
+
+	/* Get corresponding I2C address and adjust offset */
+	client = at24_translate_offset(at24, &offset);
+
+	/* write_max is at most a page */
+	if (count > at24->write_max)
+		count = at24->write_max;
+
+	/* Never roll over backwards, to the start of this page */
+	next_page = roundup(offset + 1, at24->chip.page_size);
+	if (offset + count > next_page)
+		count = next_page - offset;
+
+	/* If we'll use I2C calls for I/O, set up the message */
+	if (!at24->use_smbus) {
+		int i = 0;
+
+		msg.addr = client->addr;
+		msg.flags = 0;
+
+		/* msg.buf is u8 and casts will mask the values */
+		msg.buf = at24->writebuf;
+		if (at24->chip.flags & AT24_FLAG_ADDR16)
+			msg.buf[i++] = offset >> 8;
+
+		msg.buf[i++] = offset;
+		memcpy(&msg.buf[i], buf, count);
+		msg.len = i + count;
+	}
+
+	/*
+	 * Writes fail if the previous one didn't complete yet. We may
+	 * loop a few times until this one succeeds, waiting at least
+	 * long enough for one entire page write to work.
+	 */
+	timeout = jiffies + msecs_to_jiffies(write_timeout);
+	do {
+		write_time = jiffies;
+		if (at24->use_smbus) {
+			status = i2c_smbus_write_i2c_block_data(client,
+					offset, count, buf);
+			if (status == 0)
+				status = count;
+		} else {
+			status = i2c_transfer(client->adapter, &msg, 1);
+			if (status == 1)
+				status = count;
+		}
+		dev_dbg(&client->dev, "write %zd@%d --> %zd (%ld)\n",
+				count, offset, status, jiffies);
+
+		if (status == count)
+			return count;
+
+		/* REVISIT: at HZ=100, this is sloooow */
+		msleep(1);
+	} while (time_before(write_time, timeout));
+
+	return -ETIMEDOUT;
+}
+
+static ssize_t at24_bin_write(struct kobject *kobj, struct bin_attribute *attr,
+		char *buf, loff_t off, size_t count)
+{
+	struct at24_data *at24;
+	ssize_t retval = 0;
+
+	at24 = dev_get_drvdata(container_of(kobj, struct device, kobj));
+
+	if (unlikely(!count))
+		return count;
+
+	/*
+	 * Write data to chip, protecting against concurrent updates
+	 * from this host, but not from other I2C masters.
+	 */
+	mutex_lock(&at24->lock);
+
+	while (count) {
+		ssize_t	status;
+
+		status = at24_eeprom_write(at24, buf, off, count);
+		if (status <= 0) {
+			if (retval == 0)
+				retval = status;
+			break;
+		}
+		buf += status;
+		off += status;
+		count -= status;
+		retval += status;
+	}
+
+	mutex_unlock(&at24->lock);
+
+	return retval;
+}
+
+/*-------------------------------------------------------------------------*/
+
+static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id)
+{
+	struct at24_platform_data chip;
+	bool writable;
+	bool use_smbus = false;
+	struct at24_data *at24;
+	int err;
+	unsigned i, num_addresses;
+	kernel_ulong_t magic;
+
+	if (client->dev.platform_data) {
+		chip = *(struct at24_platform_data *)client->dev.platform_data;
+	} else {
+		if (!id->driver_data) {
+			err = -ENODEV;
+			goto err_out;
+		}
+		magic = id->driver_data;
+		chip.byte_len = BIT(magic & AT24_BITMASK(AT24_SIZE_BYTELEN));
+		magic >>= AT24_SIZE_BYTELEN;
+		chip.flags = magic & AT24_BITMASK(AT24_SIZE_FLAGS);
+		/*
+		 * This is slow, but we can't know all eeproms, so we better
+		 * play safe. Specifying custom eeprom-types via platform_data
+		 * is recommended anyhow.
+		 */
+		chip.page_size = 1;
+	}
+
+	if (!is_power_of_2(chip.byte_len))
+		dev_warn(&client->dev,
+			"byte_len looks suspicious (no power of 2)!\n");
+	if (!is_power_of_2(chip.page_size))
+		dev_warn(&client->dev,
+			"page_size looks suspicious (no power of 2)!\n");
+
+	/* Use I2C operations unless we're stuck with SMBus extensions. */
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) {
+		if (chip.flags & AT24_FLAG_ADDR16) {
+			err = -EPFNOSUPPORT;
+			goto err_out;
+		}
+		if (!i2c_check_functionality(client->adapter,
+				I2C_FUNC_SMBUS_READ_I2C_BLOCK)) {
+			err = -EPFNOSUPPORT;
+			goto err_out;
+		}
+		use_smbus = true;
+	}
+
+	if (chip.flags & AT24_FLAG_TAKE8ADDR)
+		num_addresses = 8;
+	else
+		num_addresses =	DIV_ROUND_UP(chip.byte_len,
+			(chip.flags & AT24_FLAG_ADDR16) ? 65536 : 256);
+
+	at24 = kzalloc(sizeof(struct at24_data) +
+		num_addresses * sizeof(struct i2c_client *), GFP_KERNEL);
+	if (!at24) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	mutex_init(&at24->lock);
+	at24->use_smbus = use_smbus;
+	at24->chip = chip;
+	at24->num_addresses = num_addresses;
+
+	/*
+	 * Export the EEPROM bytes through sysfs, since that's convenient.
+	 * By default, only root should see the data (maybe passwords etc)
+	 */
+	at24->bin.attr.name = "eeprom";
+	at24->bin.attr.mode = chip.flags & AT24_FLAG_IRUGO ? S_IRUGO : S_IRUSR;
+	at24->bin.attr.owner = THIS_MODULE;
+	at24->bin.read = at24_bin_read;
+	at24->bin.size = chip.byte_len;
+
+	writable = !(chip.flags & AT24_FLAG_READONLY);
+	if (writable) {
+		if (!use_smbus || i2c_check_functionality(client->adapter,
+				I2C_FUNC_SMBUS_WRITE_I2C_BLOCK)) {
+
+			unsigned write_max = chip.page_size;
+
+			at24->bin.write = at24_bin_write;
+			at24->bin.attr.mode |= S_IWUSR;
+
+			if (write_max > io_limit)
+				write_max = io_limit;
+			if (use_smbus && write_max > I2C_SMBUS_BLOCK_MAX)
+				write_max = I2C_SMBUS_BLOCK_MAX;
+			at24->write_max = write_max;
+
+			/* buffer (data + address at the beginning) */
+			at24->writebuf = kmalloc(write_max + 2, GFP_KERNEL);
+			if (!at24->writebuf) {
+				err = -ENOMEM;
+				goto err_struct;
+			}
+		} else {
+			dev_warn(&client->dev,
+				"cannot write due to controller restrictions.");
+		}
+	}
+
+	at24->client[0] = client;
+
+	/* use dummy devices for multiple-address chips */
+	for (i = 1; i < num_addresses; i++) {
+		at24->client[i] = i2c_new_dummy(client->adapter,
+					client->addr + i);
+		if (!at24->client[i]) {
+			dev_err(&client->dev, "address 0x%02x unavailable\n",
+					client->addr + i);
+			err = -EADDRINUSE;
+			goto err_clients;
+		}
+	}
+
+	err = sysfs_create_bin_file(&client->dev.kobj, &at24->bin);
+	if (err)
+		goto err_clients;
+
+	i2c_set_clientdata(client, at24);
+
+	dev_info(&client->dev, "%Zd byte %s EEPROM %s\n",
+		at24->bin.size, client->name,
+		writable ? "(writable)" : "(read-only)");
+	dev_dbg(&client->dev,
+		"page_size %d, num_addresses %d, write_max %d%s\n",
+		chip.page_size, num_addresses,
+		at24->write_max,
+		use_smbus ? ", use_smbus" : "");
+
+	return 0;
+
+err_clients:
+	for (i = 1; i < num_addresses; i++)
+		if (at24->client[i])
+			i2c_unregister_device(at24->client[i]);
+
+	kfree(at24->writebuf);
+err_struct:
+	kfree(at24);
+err_out:
+	dev_dbg(&client->dev, "probe error %d\n", err);
+	return err;
+}
+
+static int __devexit at24_remove(struct i2c_client *client)
+{
+	struct at24_data *at24;
+	int i;
+
+	at24 = i2c_get_clientdata(client);
+	sysfs_remove_bin_file(&client->dev.kobj, &at24->bin);
+
+	for (i = 1; i < at24->num_addresses; i++)
+		i2c_unregister_device(at24->client[i]);
+
+	kfree(at24->writebuf);
+	kfree(at24);
+	i2c_set_clientdata(client, NULL);
+	return 0;
+}
+
+/*-------------------------------------------------------------------------*/
+
+static struct i2c_driver at24_driver = {
+	.driver = {
+		.name = "at24",
+		.owner = THIS_MODULE,
+	},
+	.probe = at24_probe,
+	.remove = __devexit_p(at24_remove),
+	.id_table = at24_ids,
+};
+
+static int __init at24_init(void)
+{
+	io_limit = rounddown_pow_of_two(io_limit);
+	return i2c_add_driver(&at24_driver);
+}
+module_init(at24_init);
+
+static void __exit at24_exit(void)
+{
+	i2c_del_driver(&at24_driver);
+}
+module_exit(at24_exit);
+
+MODULE_DESCRIPTION("Driver for most I2C EEPROMs");
+MODULE_AUTHOR("David Brownell and Wolfram Sang");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/i2c/at24.h b/include/linux/i2c/at24.h
new file mode 100644
index 000000000000..f6edd522a929
--- /dev/null
+++ b/include/linux/i2c/at24.h
@@ -0,0 +1,28 @@
+#ifndef _LINUX_AT24_H
+#define _LINUX_AT24_H
+
+#include <linux/types.h>
+
+/*
+ * As seen through Linux I2C, differences between the most common types of I2C
+ * memory include:
+ * - How much memory is available (usually specified in bit)?
+ * - What write page size does it support?
+ * - Special flags (16 bit addresses, read_only, world readable...)?
+ *
+ * If you set up a custom eeprom type, please double-check the parameters.
+ * Especially page_size needs extra care, as you risk data loss if your value
+ * is bigger than what the chip actually supports!
+ */
+
+struct at24_platform_data {
+	u32		byte_len;		/* size (sum of all addr) */
+	u16		page_size;		/* for writes */
+	u8		flags;
+#define AT24_FLAG_ADDR16	0x80	/* address pointer is 16 bit */
+#define AT24_FLAG_READONLY	0x40	/* sysfs-entry will be read-only */
+#define AT24_FLAG_IRUGO		0x20	/* sysfs-entry will be world-readable */
+#define AT24_FLAG_TAKE8ADDR	0x10	/* take always 8 addresses (24c00) */
+};
+
+#endif /* _LINUX_AT24_H */
-- 
cgit v1.2.3


From 4735c98f8447acb1c8977e2b8024640f7bf36dd6 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 14 Jul 2008 22:38:36 +0200
Subject: i2c: Add detection capability to new-style drivers

Add a mechanism to let new-style i2c drivers optionally autodetect
devices they would support on selected buses and ask i2c-core to
instantiate them. This is a replacement for legacy i2c drivers, much
cleaner.

Where drivers had to implement both a legacy i2c_driver and a
new-style i2c_driver so far, this mechanism makes it possible to get
rid of the legacy i2c_driver and implement both enumerated and
detected device support with just one (new-style) i2c_driver.

Here is a quick conversion guide for these drivers, step by step:

* Delete the legacy driver definition, registration and removal.
  Delete the attach_adapter and detach_client methods of the legacy
  driver.

* Change the prototype of the legacy detect function from
    static int foo_detect(struct i2c_adapter *adapter, int address, int kind);
  to
    static int foo_detect(struct i2c_client *client, int kind,
    			  struct i2c_board_info *info);

* Set the new-style driver detect callback to this new function, and
  set its address_data to &addr_data (addr_data is generally provided
  by I2C_CLIENT_INSMOD.)

* Add the appropriate class to the new-style driver. This is
  typically the class the legacy attach_adapter method was checking
  for. Class checking is now mandatory (done by i2c-core.) See
  <linux/i2c.h> for the list of available classes.

* Remove the i2c_client allocation and freeing from the detect
  function. A pre-allocated client is now handed to you by i2c-core,
  and is freed automatically.

* Make the detect function fill the type field of the i2c_board_info
  structure it was passed as a parameter, and return 0, on success. If
  the detection fails, return -ENODEV.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 Documentation/i2c/writing-clients |  29 +++++
 drivers/i2c/i2c-core.c            | 223 ++++++++++++++++++++++++++++++++++++--
 include/linux/i2c.h               |  36 +++++-
 3 files changed, 272 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients
index 63722d3c9cdf..6b61b3a2e90b 100644
--- a/Documentation/i2c/writing-clients
+++ b/Documentation/i2c/writing-clients
@@ -44,6 +44,10 @@ static struct i2c_driver foo_driver = {
 	.id_table	= foo_ids,
 	.probe		= foo_probe,
 	.remove		= foo_remove,
+	/* if device autodetection is needed: */
+	.class		= I2C_CLASS_SOMETHING,
+	.detect		= foo_detect,
+	.address_data	= &addr_data,
 
 	/* else, driver uses "legacy" binding model: */
 	.attach_adapter	= foo_attach_adapter,
@@ -217,6 +221,31 @@ in the I2C bus driver. You may want to save the returned i2c_client
 reference for later use.
 
 
+Device Detection (Standard driver model)
+----------------------------------------
+
+Sometimes you do not know in advance which I2C devices are connected to
+a given I2C bus.  This is for example the case of hardware monitoring
+devices on a PC's SMBus.  In that case, you may want to let your driver
+detect supported devices automatically.  This is how the legacy model
+was working, and is now available as an extension to the standard
+driver model (so that we can finally get rid of the legacy model.)
+
+You simply have to define a detect callback which will attempt to
+identify supported devices (returning 0 for supported ones and -ENODEV
+for unsupported ones), a list of addresses to probe, and a device type
+(or class) so that only I2C buses which may have that type of device
+connected (and not otherwise enumerated) will be probed.  The i2c
+core will then call you back as needed and will instantiate a device
+for you for every successful detection.
+
+Note that this mechanism is purely optional and not suitable for all
+devices.  You need some reliable way to identify the supported devices
+(typically using device-specific, dedicated identification registers),
+otherwise misdetections are likely to occur and things can get wrong
+quickly.
+
+
 Device Deletion (Standard driver model)
 ---------------------------------------
 
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 5e249d758828..0a79f7661017 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -42,7 +42,9 @@
 static DEFINE_MUTEX(core_lock);
 static DEFINE_IDR(i2c_adapter_idr);
 
-#define is_newstyle_driver(d) ((d)->probe || (d)->remove)
+#define is_newstyle_driver(d) ((d)->probe || (d)->remove || (d)->detect)
+
+static int i2c_detect(struct i2c_adapter *adapter, struct i2c_driver *driver);
 
 /* ------------------------------------------------------------------------- */
 
@@ -418,6 +420,10 @@ static int i2c_do_add_adapter(struct device_driver *d, void *data)
 	struct i2c_driver *driver = to_i2c_driver(d);
 	struct i2c_adapter *adap = data;
 
+	/* Detect supported devices on that bus, and instantiate them */
+	i2c_detect(adap, driver);
+
+	/* Let legacy drivers scan this bus for matching devices */
 	if (driver->attach_adapter) {
 		/* We ignore the return code; if it fails, too bad */
 		driver->attach_adapter(adap);
@@ -457,7 +463,7 @@ static int i2c_register_adapter(struct i2c_adapter *adap)
 	if (adap->nr < __i2c_first_dynamic_bus_num)
 		i2c_scan_static_board_info(adap);
 
-	/* let legacy drivers scan this bus for matching devices */
+	/* Notify drivers */
 	dummy = bus_for_each_drv(&i2c_bus_type, NULL, adap,
 				 i2c_do_add_adapter);
 
@@ -563,8 +569,19 @@ static int i2c_do_del_adapter(struct device_driver *d, void *data)
 {
 	struct i2c_driver *driver = to_i2c_driver(d);
 	struct i2c_adapter *adapter = data;
+	struct i2c_client *client, *_n;
 	int res;
 
+	/* Remove the devices we created ourselves */
+	list_for_each_entry_safe(client, _n, &driver->clients, detected) {
+		if (client->adapter == adapter) {
+			dev_dbg(&adapter->dev, "Removing %s at 0x%x\n",
+				client->name, client->addr);
+			list_del(&client->detected);
+			i2c_unregister_device(client);
+		}
+	}
+
 	if (!driver->detach_adapter)
 		return 0;
 	res = driver->detach_adapter(adapter);
@@ -651,7 +668,11 @@ static int __attach_adapter(struct device *dev, void *data)
 	struct i2c_adapter *adapter = to_i2c_adapter(dev);
 	struct i2c_driver *driver = data;
 
-	driver->attach_adapter(adapter);
+	i2c_detect(adapter, driver);
+
+	/* Legacy drivers scan i2c busses directly */
+	if (driver->attach_adapter)
+		driver->attach_adapter(adapter);
 
 	return 0;
 }
@@ -695,10 +716,9 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver)
 
 	pr_debug("i2c-core: driver [%s] registered\n", driver->driver.name);
 
-	/* legacy drivers scan i2c busses directly */
-	if (driver->attach_adapter)
-		class_for_each_device(&i2c_adapter_class, driver,
-				      __attach_adapter);
+	INIT_LIST_HEAD(&driver->clients);
+	/* Walk the adapters that are already present */
+	class_for_each_device(&i2c_adapter_class, driver, __attach_adapter);
 
 	mutex_unlock(&core_lock);
 	return 0;
@@ -709,6 +729,17 @@ static int __detach_adapter(struct device *dev, void *data)
 {
 	struct i2c_adapter *adapter = to_i2c_adapter(dev);
 	struct i2c_driver *driver = data;
+	struct i2c_client *client, *_n;
+
+	list_for_each_entry_safe(client, _n, &driver->clients, detected) {
+		dev_dbg(&adapter->dev, "Removing %s at 0x%x\n",
+			client->name, client->addr);
+		list_del(&client->detected);
+		i2c_unregister_device(client);
+	}
+
+	if (is_newstyle_driver(driver))
+		return 0;
 
 	/* Have a look at each adapter, if clients of this driver are still
 	 * attached. If so, detach them to be able to kill the driver
@@ -747,10 +778,7 @@ void i2c_del_driver(struct i2c_driver *driver)
 {
 	mutex_lock(&core_lock);
 
-	/* legacy driver? */
-	if (!is_newstyle_driver(driver))
-		class_for_each_device(&i2c_adapter_class, driver,
-				      __detach_adapter);
+	class_for_each_device(&i2c_adapter_class, driver, __detach_adapter);
 
 	driver_unregister(&driver->driver);
 	pr_debug("i2c-core: driver [%s] unregistered\n", driver->driver.name);
@@ -1205,6 +1233,179 @@ int i2c_probe(struct i2c_adapter *adapter,
 }
 EXPORT_SYMBOL(i2c_probe);
 
+/* Separate detection function for new-style drivers */
+static int i2c_detect_address(struct i2c_client *temp_client, int kind,
+			      struct i2c_driver *driver)
+{
+	struct i2c_board_info info;
+	struct i2c_adapter *adapter = temp_client->adapter;
+	int addr = temp_client->addr;
+	int err;
+
+	/* Make sure the address is valid */
+	if (addr < 0x03 || addr > 0x77) {
+		dev_warn(&adapter->dev, "Invalid probe address 0x%02x\n",
+			 addr);
+		return -EINVAL;
+	}
+
+	/* Skip if already in use */
+	if (i2c_check_addr(adapter, addr))
+		return 0;
+
+	/* Make sure there is something at this address, unless forced */
+	if (kind < 0) {
+		if (i2c_smbus_xfer(adapter, addr, 0, 0, 0,
+				   I2C_SMBUS_QUICK, NULL) < 0)
+			return 0;
+
+		/* prevent 24RF08 corruption */
+		if ((addr & ~0x0f) == 0x50)
+			i2c_smbus_xfer(adapter, addr, 0, 0, 0,
+				       I2C_SMBUS_QUICK, NULL);
+	}
+
+	/* Finally call the custom detection function */
+	memset(&info, 0, sizeof(struct i2c_board_info));
+	info.addr = addr;
+	err = driver->detect(temp_client, kind, &info);
+	if (err) {
+		/* -ENODEV is returned if the detection fails. We catch it
+		   here as this isn't an error. */
+		return err == -ENODEV ? 0 : err;
+	}
+
+	/* Consistency check */
+	if (info.type[0] == '\0') {
+		dev_err(&adapter->dev, "%s detection function provided "
+			"no name for 0x%x\n", driver->driver.name,
+			addr);
+	} else {
+		struct i2c_client *client;
+
+		/* Detection succeeded, instantiate the device */
+		dev_dbg(&adapter->dev, "Creating %s at 0x%02x\n",
+			info.type, info.addr);
+		client = i2c_new_device(adapter, &info);
+		if (client)
+			list_add_tail(&client->detected, &driver->clients);
+		else
+			dev_err(&adapter->dev, "Failed creating %s at 0x%02x\n",
+				info.type, info.addr);
+	}
+	return 0;
+}
+
+static int i2c_detect(struct i2c_adapter *adapter, struct i2c_driver *driver)
+{
+	const struct i2c_client_address_data *address_data;
+	struct i2c_client *temp_client;
+	int i, err = 0;
+	int adap_id = i2c_adapter_id(adapter);
+
+	address_data = driver->address_data;
+	if (!driver->detect || !address_data)
+		return 0;
+
+	/* Set up a temporary client to help detect callback */
+	temp_client = kzalloc(sizeof(struct i2c_client), GFP_KERNEL);
+	if (!temp_client)
+		return -ENOMEM;
+	temp_client->adapter = adapter;
+
+	/* Force entries are done first, and are not affected by ignore
+	   entries */
+	if (address_data->forces) {
+		const unsigned short * const *forces = address_data->forces;
+		int kind;
+
+		for (kind = 0; forces[kind]; kind++) {
+			for (i = 0; forces[kind][i] != I2C_CLIENT_END;
+			     i += 2) {
+				if (forces[kind][i] == adap_id
+				 || forces[kind][i] == ANY_I2C_BUS) {
+					dev_dbg(&adapter->dev, "found force "
+						"parameter for adapter %d, "
+						"addr 0x%02x, kind %d\n",
+						adap_id, forces[kind][i + 1],
+						kind);
+					temp_client->addr = forces[kind][i + 1];
+					err = i2c_detect_address(temp_client,
+						kind, driver);
+					if (err)
+						goto exit_free;
+				}
+			}
+		}
+	}
+
+	/* Stop here if we can't use SMBUS_QUICK */
+	if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_QUICK)) {
+		if (address_data->probe[0] == I2C_CLIENT_END
+		 && address_data->normal_i2c[0] == I2C_CLIENT_END)
+			goto exit_free;
+
+		dev_warn(&adapter->dev, "SMBus Quick command not supported, "
+			 "can't probe for chips\n");
+		err = -EOPNOTSUPP;
+		goto exit_free;
+	}
+
+	/* Stop here if the classes do not match */
+	if (!(adapter->class & driver->class))
+		goto exit_free;
+
+	/* Probe entries are done second, and are not affected by ignore
+	   entries either */
+	for (i = 0; address_data->probe[i] != I2C_CLIENT_END; i += 2) {
+		if (address_data->probe[i] == adap_id
+		 || address_data->probe[i] == ANY_I2C_BUS) {
+			dev_dbg(&adapter->dev, "found probe parameter for "
+				"adapter %d, addr 0x%02x\n", adap_id,
+				address_data->probe[i + 1]);
+			temp_client->addr = address_data->probe[i + 1];
+			err = i2c_detect_address(temp_client, -1, driver);
+			if (err)
+				goto exit_free;
+		}
+	}
+
+	/* Normal entries are done last, unless shadowed by an ignore entry */
+	for (i = 0; address_data->normal_i2c[i] != I2C_CLIENT_END; i += 1) {
+		int j, ignore;
+
+		ignore = 0;
+		for (j = 0; address_data->ignore[j] != I2C_CLIENT_END;
+		     j += 2) {
+			if ((address_data->ignore[j] == adap_id ||
+			     address_data->ignore[j] == ANY_I2C_BUS)
+			 && address_data->ignore[j + 1]
+			    == address_data->normal_i2c[i]) {
+				dev_dbg(&adapter->dev, "found ignore "
+					"parameter for adapter %d, "
+					"addr 0x%02x\n", adap_id,
+					address_data->ignore[j + 1]);
+				ignore = 1;
+				break;
+			}
+		}
+		if (ignore)
+			continue;
+
+		dev_dbg(&adapter->dev, "found normal entry for adapter %d, "
+			"addr 0x%02x\n", adap_id,
+			address_data->normal_i2c[i]);
+		temp_client->addr = address_data->normal_i2c[i];
+		err = i2c_detect_address(temp_client, -1, driver);
+		if (err)
+			goto exit_free;
+	}
+
+ exit_free:
+	kfree(temp_client);
+	return err;
+}
+
 struct i2c_client *
 i2c_new_probed_device(struct i2c_adapter *adap,
 		      struct i2c_board_info *info,
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 50cbab4b62b0..08be0d21864c 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -45,6 +45,7 @@ struct i2c_adapter;
 struct i2c_client;
 struct i2c_driver;
 union i2c_smbus_data;
+struct i2c_board_info;
 
 /*
  * The master routines are the ones normally used to transmit data to devices
@@ -94,15 +95,33 @@ extern s32 i2c_smbus_write_i2c_block_data(struct i2c_client * client,
 					  u8 command, u8 length,
 					  const u8 *values);
 
-/*
- * A driver is capable of handling one or more physical devices present on
- * I2C adapters. This information is used to inform the driver of adapter
- * events.
+/**
+ * struct i2c_driver - represent an I2C device driver
+ * @class: What kind of i2c device we instantiate (for detect)
+ * @detect: Callback for device detection
+ * @address_data: The I2C addresses to probe, ignore or force (for detect)
+ * @clients: List of detected clients we created (for i2c-core use only)
  *
  * The driver.owner field should be set to the module owner of this driver.
  * The driver.name field should be set to the name of this driver.
+ *
+ * For automatic device detection, both @detect and @address_data must
+ * be defined. @class should also be set, otherwise only devices forced
+ * with module parameters will be created. The detect function must
+ * fill at least the name field of the i2c_board_info structure it is
+ * handed upon successful detection, and possibly also the flags field.
+ *
+ * If @detect is missing, the driver will still work fine for enumerated
+ * devices. Detected devices simply won't be supported. This is expected
+ * for the many I2C/SMBus devices which can't be detected reliably, and
+ * the ones which can always be enumerated in practice.
+ *
+ * The i2c_client structure which is handed to the @detect callback is
+ * not a real i2c_client. It is initialized just enough so that you can
+ * call i2c_smbus_read_byte_data and friends on it. Don't do anything
+ * else with it. In particular, calling dev_dbg and friends on it is
+ * not allowed.
  */
-
 struct i2c_driver {
 	int id;
 	unsigned int class;
@@ -142,6 +161,11 @@ struct i2c_driver {
 
 	struct device_driver driver;
 	const struct i2c_device_id *id_table;
+
+	/* Device detection callback for automatic device creation */
+	int (*detect)(struct i2c_client *, int kind, struct i2c_board_info *);
+	const struct i2c_client_address_data *address_data;
+	struct list_head clients;
 };
 #define to_i2c_driver(d) container_of(d, struct i2c_driver, driver)
 
@@ -157,6 +181,7 @@ struct i2c_driver {
  * @dev: Driver model device node for the slave.
  * @irq: indicates the IRQ generated by this device (if any)
  * @list: list of active/busy clients (DEPRECATED)
+ * @detected: member of an i2c_driver.clients list
  * @released: used to synchronize client releases & detaches and references
  *
  * An i2c_client identifies a single device (i.e. chip) connected to an
@@ -174,6 +199,7 @@ struct i2c_client {
 	struct device dev;		/* the device structure		*/
 	int irq;			/* irq issued by device		*/
 	struct list_head list;		/* DEPRECATED */
+	struct list_head detected;
 	struct completion released;
 };
 #define to_i2c_client(d) container_of(d, struct i2c_client, dev)
-- 
cgit v1.2.3


From 11c3b79218390a139f2d474ee1e983a672d5839a Mon Sep 17 00:00:00 2001
From: Joel Becker <joel.becker@oracle.com>
Date: Thu, 12 Jun 2008 14:00:18 -0700
Subject: configfs: Allow ->make_item() and ->make_group() to return detailed
 errors.

The configfs operations ->make_item() and ->make_group() currently
return a new item/group.  A return of NULL signifies an error.  Because
of this, -ENOMEM is the only return code bubbled up the stack.

Multiple folks have requested the ability to return specific error codes
when these operations fail.  This patch adds that ability by changing the
->make_item/group() ops to return an int.

Also updated are the in-kernel users of configfs.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 Documentation/filesystems/configfs/configfs.txt    | 10 +++--
 .../filesystems/configfs/configfs_example.c        | 14 ++++---
 drivers/net/netconsole.c                           | 10 +++--
 fs/configfs/dir.c                                  | 13 +++----
 fs/dlm/config.c                                    | 45 ++++++++++++++--------
 fs/ocfs2/cluster/heartbeat.c                       | 17 ++++----
 fs/ocfs2/cluster/nodemanager.c                     | 45 ++++++++++++++--------
 include/linux/configfs.h                           |  4 +-
 8 files changed, 94 insertions(+), 64 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/configfs/configfs.txt b/Documentation/filesystems/configfs/configfs.txt
index 44c97e6accb2..15838d706ea2 100644
--- a/Documentation/filesystems/configfs/configfs.txt
+++ b/Documentation/filesystems/configfs/configfs.txt
@@ -233,10 +233,12 @@ accomplished via the group operations specified on the group's
 config_item_type.
 
 	struct configfs_group_operations {
-		struct config_item *(*make_item)(struct config_group *group,
-						 const char *name);
-		struct config_group *(*make_group)(struct config_group *group,
-						   const char *name);
+		int (*make_item)(struct config_group *group,
+				 const char *name,
+				 struct config_item **new_item);
+		int (*make_group)(struct config_group *group,
+				  const char *name,
+				  struct config_group **new_group);
 		int (*commit_item)(struct config_item *item);
 		void (*disconnect_notify)(struct config_group *group,
 					  struct config_item *item);
diff --git a/Documentation/filesystems/configfs/configfs_example.c b/Documentation/filesystems/configfs/configfs_example.c
index 25151fd5c2c6..0b422acd470c 100644
--- a/Documentation/filesystems/configfs/configfs_example.c
+++ b/Documentation/filesystems/configfs/configfs_example.c
@@ -273,13 +273,13 @@ static inline struct simple_children *to_simple_children(struct config_item *ite
 	return item ? container_of(to_config_group(item), struct simple_children, group) : NULL;
 }
 
-static struct config_item *simple_children_make_item(struct config_group *group, const char *name)
+static int simple_children_make_item(struct config_group *group, const char *name, struct config_item **new_item)
 {
 	struct simple_child *simple_child;
 
 	simple_child = kzalloc(sizeof(struct simple_child), GFP_KERNEL);
 	if (!simple_child)
-		return NULL;
+		return -ENOMEM;
 
 
 	config_item_init_type_name(&simple_child->item, name,
@@ -287,7 +287,8 @@ static struct config_item *simple_children_make_item(struct config_group *group,
 
 	simple_child->storeme = 0;
 
-	return &simple_child->item;
+	*new_item = &simple_child->item;
+	return 0;
 }
 
 static struct configfs_attribute simple_children_attr_description = {
@@ -359,20 +360,21 @@ static struct configfs_subsystem simple_children_subsys = {
  * children of its own.
  */
 
-static struct config_group *group_children_make_group(struct config_group *group, const char *name)
+static int group_children_make_group(struct config_group *group, const char *name, struct config_group **new_group)
 {
 	struct simple_children *simple_children;
 
 	simple_children = kzalloc(sizeof(struct simple_children),
 				  GFP_KERNEL);
 	if (!simple_children)
-		return NULL;
+		return -ENOMEM;
 
 
 	config_group_init_type_name(&simple_children->group, name,
 				    &simple_children_type);
 
-	return &simple_children->group;
+	*new_group = &simple_children->group;
+	return 0;
 }
 
 static struct configfs_attribute group_children_attr_description = {
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 665341e43055..387a13395015 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -585,8 +585,9 @@ static struct config_item_type netconsole_target_type = {
  * Group operations and type for netconsole_subsys.
  */
 
-static struct config_item *make_netconsole_target(struct config_group *group,
-						  const char *name)
+static int make_netconsole_target(struct config_group *group,
+				  const char *name,
+				  struct config_item **new_item)
 {
 	unsigned long flags;
 	struct netconsole_target *nt;
@@ -598,7 +599,7 @@ static struct config_item *make_netconsole_target(struct config_group *group,
 	nt = kzalloc(sizeof(*nt), GFP_KERNEL);
 	if (!nt) {
 		printk(KERN_ERR "netconsole: failed to allocate memory\n");
-		return NULL;
+		return -ENOMEM;
 	}
 
 	nt->np.name = "netconsole";
@@ -615,7 +616,8 @@ static struct config_item *make_netconsole_target(struct config_group *group,
 	list_add(&nt->list, &target_list);
 	spin_unlock_irqrestore(&target_list_lock, flags);
 
-	return &nt->item;
+	*new_item = &nt->item;
+	return 0;
 }
 
 static void drop_netconsole_target(struct config_group *group,
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 614e382a6049..0e64312a084c 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -1073,25 +1073,24 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	group = NULL;
 	item = NULL;
 	if (type->ct_group_ops->make_group) {
-		group = type->ct_group_ops->make_group(to_config_group(parent_item), name);
-		if (group) {
+		ret = type->ct_group_ops->make_group(to_config_group(parent_item), name, &group);
+		if (!ret) {
 			link_group(to_config_group(parent_item), group);
 			item = &group->cg_item;
 		}
 	} else {
-		item = type->ct_group_ops->make_item(to_config_group(parent_item), name);
-		if (item)
+		ret = type->ct_group_ops->make_item(to_config_group(parent_item), name, &item);
+		if (!ret)
 			link_obj(parent_item, item);
 	}
 	mutex_unlock(&subsys->su_mutex);
 
 	kfree(name);
-	if (!item) {
+	if (ret) {
 		/*
-		 * If item == NULL, then link_obj() was never called.
+		 * If ret != 0, then link_obj() was never called.
 		 * There are no extra references to clean up.
 		 */
-		ret = -ENOMEM;
 		goto out_put;
 	}
 
diff --git a/fs/dlm/config.c b/fs/dlm/config.c
index eac23bd288b2..492d8caaaf25 100644
--- a/fs/dlm/config.c
+++ b/fs/dlm/config.c
@@ -41,16 +41,20 @@ struct comm;
 struct nodes;
 struct node;
 
-static struct config_group *make_cluster(struct config_group *, const char *);
+static int make_cluster(struct config_group *, const char *,
+			struct config_group **);
 static void drop_cluster(struct config_group *, struct config_item *);
 static void release_cluster(struct config_item *);
-static struct config_group *make_space(struct config_group *, const char *);
+static int make_space(struct config_group *, const char *,
+		      struct config_group **);
 static void drop_space(struct config_group *, struct config_item *);
 static void release_space(struct config_item *);
-static struct config_item *make_comm(struct config_group *, const char *);
+static int make_comm(struct config_group *, const char *,
+		     struct config_item **);
 static void drop_comm(struct config_group *, struct config_item *);
 static void release_comm(struct config_item *);
-static struct config_item *make_node(struct config_group *, const char *);
+static int make_node(struct config_group *, const char *,
+		     struct config_item **);
 static void drop_node(struct config_group *, struct config_item *);
 static void release_node(struct config_item *);
 
@@ -392,8 +396,8 @@ static struct node *to_node(struct config_item *i)
 	return i ? container_of(i, struct node, item) : NULL;
 }
 
-static struct config_group *make_cluster(struct config_group *g,
-					 const char *name)
+static int make_cluster(struct config_group *g, const char *name,
+			struct config_group **new_g)
 {
 	struct cluster *cl = NULL;
 	struct spaces *sps = NULL;
@@ -431,14 +435,15 @@ static struct config_group *make_cluster(struct config_group *g,
 
 	space_list = &sps->ss_group;
 	comm_list = &cms->cs_group;
-	return &cl->group;
+	*new_g = &cl->group;
+	return 0;
 
  fail:
 	kfree(cl);
 	kfree(gps);
 	kfree(sps);
 	kfree(cms);
-	return NULL;
+	return -ENOMEM;
 }
 
 static void drop_cluster(struct config_group *g, struct config_item *i)
@@ -466,7 +471,8 @@ static void release_cluster(struct config_item *i)
 	kfree(cl);
 }
 
-static struct config_group *make_space(struct config_group *g, const char *name)
+static int make_space(struct config_group *g, const char *name,
+		      struct config_group **new_g)
 {
 	struct space *sp = NULL;
 	struct nodes *nds = NULL;
@@ -489,13 +495,14 @@ static struct config_group *make_space(struct config_group *g, const char *name)
 	INIT_LIST_HEAD(&sp->members);
 	mutex_init(&sp->members_lock);
 	sp->members_count = 0;
-	return &sp->group;
+	*new_g = &sp->group;
+	return 0;
 
  fail:
 	kfree(sp);
 	kfree(gps);
 	kfree(nds);
-	return NULL;
+	return -ENOMEM;
 }
 
 static void drop_space(struct config_group *g, struct config_item *i)
@@ -522,19 +529,21 @@ static void release_space(struct config_item *i)
 	kfree(sp);
 }
 
-static struct config_item *make_comm(struct config_group *g, const char *name)
+static int make_comm(struct config_group *g, const char *name,
+		     struct config_item **new_i)
 {
 	struct comm *cm;
 
 	cm = kzalloc(sizeof(struct comm), GFP_KERNEL);
 	if (!cm)
-		return NULL;
+		return -ENOMEM;
 
 	config_item_init_type_name(&cm->item, name, &comm_type);
 	cm->nodeid = -1;
 	cm->local = 0;
 	cm->addr_count = 0;
-	return &cm->item;
+	*new_i = &cm->item;
+	return 0;
 }
 
 static void drop_comm(struct config_group *g, struct config_item *i)
@@ -554,14 +563,15 @@ static void release_comm(struct config_item *i)
 	kfree(cm);
 }
 
-static struct config_item *make_node(struct config_group *g, const char *name)
+static int make_node(struct config_group *g, const char *name,
+		     struct config_item **new_i)
 {
 	struct space *sp = to_space(g->cg_item.ci_parent);
 	struct node *nd;
 
 	nd = kzalloc(sizeof(struct node), GFP_KERNEL);
 	if (!nd)
-		return NULL;
+		return -ENOMEM;
 
 	config_item_init_type_name(&nd->item, name, &node_type);
 	nd->nodeid = -1;
@@ -573,7 +583,8 @@ static struct config_item *make_node(struct config_group *g, const char *name)
 	sp->members_count++;
 	mutex_unlock(&sp->members_lock);
 
-	return &nd->item;
+	*new_i = &nd->item;
+	return 0;
 }
 
 static void drop_node(struct config_group *g, struct config_item *i)
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index f02ccb34604d..443d108211ab 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -1489,25 +1489,28 @@ static struct o2hb_heartbeat_group *to_o2hb_heartbeat_group(struct config_group
 		: NULL;
 }
 
-static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *group,
-							  const char *name)
+static int o2hb_heartbeat_group_make_item(struct config_group *group,
+					  const char *name,
+					  struct config_item **new_item)
 {
 	struct o2hb_region *reg = NULL;
-	struct config_item *ret = NULL;
+	int ret = 0;
 
 	reg = kzalloc(sizeof(struct o2hb_region), GFP_KERNEL);
-	if (reg == NULL)
-		goto out; /* ENOMEM */
+	if (reg == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	config_item_init_type_name(&reg->hr_item, name, &o2hb_region_type);
 
-	ret = &reg->hr_item;
+	*new_item = &reg->hr_item;
 
 	spin_lock(&o2hb_live_lock);
 	list_add_tail(&reg->hr_all_item, &o2hb_all_regions);
 	spin_unlock(&o2hb_live_lock);
 out:
-	if (ret == NULL)
+	if (ret)
 		kfree(reg);
 
 	return ret;
diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c
index cfdb08b484ed..b364b7052e46 100644
--- a/fs/ocfs2/cluster/nodemanager.c
+++ b/fs/ocfs2/cluster/nodemanager.c
@@ -644,27 +644,32 @@ out:
 	return ret;
 }
 
-static struct config_item *o2nm_node_group_make_item(struct config_group *group,
-						     const char *name)
+static int o2nm_node_group_make_item(struct config_group *group,
+				     const char *name,
+				     struct config_item **new_item)
 {
 	struct o2nm_node *node = NULL;
-	struct config_item *ret = NULL;
+	int ret = 0;
 
-	if (strlen(name) > O2NM_MAX_NAME_LEN)
-		goto out; /* ENAMETOOLONG */
+	if (strlen(name) > O2NM_MAX_NAME_LEN) {
+		ret = -ENAMETOOLONG;
+		goto out;
+	}
 
 	node = kzalloc(sizeof(struct o2nm_node), GFP_KERNEL);
-	if (node == NULL)
-		goto out; /* ENOMEM */
+	if (node == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	strcpy(node->nd_name, name); /* use item.ci_namebuf instead? */
 	config_item_init_type_name(&node->nd_item, name, &o2nm_node_type);
 	spin_lock_init(&node->nd_lock);
 
-	ret = &node->nd_item;
+	*new_item = &node->nd_item;
 
 out:
-	if (ret == NULL)
+	if (ret)
 		kfree(node);
 
 	return ret;
@@ -751,25 +756,31 @@ static struct o2nm_cluster_group *to_o2nm_cluster_group(struct config_group *gro
 }
 #endif
 
-static struct config_group *o2nm_cluster_group_make_group(struct config_group *group,
-							  const char *name)
+static int o2nm_cluster_group_make_group(struct config_group *group,
+					 const char *name,
+					 struct config_group **new_group)
 {
 	struct o2nm_cluster *cluster = NULL;
 	struct o2nm_node_group *ns = NULL;
-	struct config_group *o2hb_group = NULL, *ret = NULL;
+	struct config_group *o2hb_group = NULL;
 	void *defs = NULL;
+	int ret = 0;
 
 	/* this runs under the parent dir's i_mutex; there can be only
 	 * one caller in here at a time */
-	if (o2nm_single_cluster)
-		goto out; /* ENOSPC */
+	if (o2nm_single_cluster) {
+		ret = -ENOSPC;
+		goto out;
+	}
 
 	cluster = kzalloc(sizeof(struct o2nm_cluster), GFP_KERNEL);
 	ns = kzalloc(sizeof(struct o2nm_node_group), GFP_KERNEL);
 	defs = kcalloc(3, sizeof(struct config_group *), GFP_KERNEL);
 	o2hb_group = o2hb_alloc_hb_set();
-	if (cluster == NULL || ns == NULL || o2hb_group == NULL || defs == NULL)
+	if (cluster == NULL || ns == NULL || o2hb_group == NULL || defs == NULL) {
+		ret = -ENOMEM;
 		goto out;
+	}
 
 	config_group_init_type_name(&cluster->cl_group, name,
 				    &o2nm_cluster_type);
@@ -786,11 +797,11 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g
 	cluster->cl_idle_timeout_ms    = O2NET_IDLE_TIMEOUT_MS_DEFAULT;
 	cluster->cl_keepalive_delay_ms = O2NET_KEEPALIVE_DELAY_MS_DEFAULT;
 
-	ret = &cluster->cl_group;
+	*new_group = &cluster->cl_group;
 	o2nm_single_cluster = cluster;
 
 out:
-	if (ret == NULL) {
+	if (ret) {
 		kfree(cluster);
 		kfree(ns);
 		o2hb_free_hb_set(o2hb_group);
diff --git a/include/linux/configfs.h b/include/linux/configfs.h
index 3ae65b1bf90f..0488f937634a 100644
--- a/include/linux/configfs.h
+++ b/include/linux/configfs.h
@@ -165,8 +165,8 @@ struct configfs_item_operations {
 };
 
 struct configfs_group_operations {
-	struct config_item *(*make_item)(struct config_group *group, const char *name);
-	struct config_group *(*make_group)(struct config_group *group, const char *name);
+	int (*make_item)(struct config_group *group, const char *name, struct config_item **new_item);
+	int (*make_group)(struct config_group *group, const char *name, struct config_group **new_group);
 	int (*commit_item)(struct config_item *item);
 	void (*disconnect_notify)(struct config_group *group, struct config_item *item);
 	void (*drop_item)(struct config_group *group, struct config_item *item);
-- 
cgit v1.2.3


From c300bd2fb583afb6d68804afd38bc90b31310d95 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 10 Jul 2008 02:16:44 +0200
Subject: PCI: include linux/pm_wakeup.h for device_set_wakeup_capable

drivers/pci/pci.c needs pm_wakeup.h since it uses device_set_wakup_capable().
The latter also needs to be stubbed out for !CONFIG_PM.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c         | 1 +
 include/linux/pm_wakeup.h | 2 ++
 2 files changed, 3 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index ace518116cc5..44a46c92b721 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -17,6 +17,7 @@
 #include <linux/string.h>
 #include <linux/log2.h>
 #include <linux/pci-aspm.h>
+#include <linux/pm_wakeup.h>
 #include <asm/dma.h>	/* isa_dma_bridge_buggy */
 #include "pci.h"
 
diff --git a/include/linux/pm_wakeup.h b/include/linux/pm_wakeup.h
index 3af0c8d05cdc..0aae7776185e 100644
--- a/include/linux/pm_wakeup.h
+++ b/include/linux/pm_wakeup.h
@@ -63,6 +63,8 @@ static inline void device_init_wakeup(struct device *dev, int val)
 	dev->power.can_wakeup = !!val;
 }
 
+static inline void device_set_wakeup_capable(struct device *dev, int val) { }
+
 static inline int device_can_wakeup(struct device *dev)
 {
 	return dev->power.can_wakeup;
-- 
cgit v1.2.3


From 72d9794f444734af56ef12833b496326643e2964 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 20:36:32 -0700
Subject: net-sched: cls_flow: add perturbation support

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_cls.h |  1 +
 net/sched/cls_flow.c    | 52 ++++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 44 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h
index 99efbed81fa2..7cf7824df778 100644
--- a/include/linux/pkt_cls.h
+++ b/include/linux/pkt_cls.h
@@ -374,6 +374,7 @@ enum
 	TCA_FLOW_ACT,
 	TCA_FLOW_POLICE,
 	TCA_FLOW_EMATCHES,
+	TCA_FLOW_PERTURB,
 	__TCA_FLOW_MAX
 };
 
diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c
index 971b867e0484..8f63a1a94014 100644
--- a/net/sched/cls_flow.c
+++ b/net/sched/cls_flow.c
@@ -36,6 +36,8 @@ struct flow_filter {
 	struct list_head	list;
 	struct tcf_exts		exts;
 	struct tcf_ematch_tree	ematches;
+	struct timer_list	perturb_timer;
+	u32			perturb_period;
 	u32			handle;
 
 	u32			nkeys;
@@ -47,11 +49,9 @@ struct flow_filter {
 	u32			addend;
 	u32			divisor;
 	u32			baseclass;
+	u32			hashrnd;
 };
 
-static u32 flow_hashrnd __read_mostly;
-static int flow_hashrnd_initted __read_mostly;
-
 static const struct tcf_ext_map flow_ext_map = {
 	.action	= TCA_FLOW_ACT,
 	.police	= TCA_FLOW_POLICE,
@@ -348,7 +348,7 @@ static int flow_classify(struct sk_buff *skb, struct tcf_proto *tp,
 		}
 
 		if (f->mode == FLOW_MODE_HASH)
-			classid = jhash2(keys, f->nkeys, flow_hashrnd);
+			classid = jhash2(keys, f->nkeys, f->hashrnd);
 		else {
 			classid = keys[0];
 			classid = (classid & f->mask) ^ f->xor;
@@ -369,6 +369,15 @@ static int flow_classify(struct sk_buff *skb, struct tcf_proto *tp,
 	return -1;
 }
 
+static void flow_perturbation(unsigned long arg)
+{
+	struct flow_filter *f = (struct flow_filter *)arg;
+
+	get_random_bytes(&f->hashrnd, 4);
+	if (f->perturb_period)
+		mod_timer(&f->perturb_timer, jiffies + f->perturb_period);
+}
+
 static const struct nla_policy flow_policy[TCA_FLOW_MAX + 1] = {
 	[TCA_FLOW_KEYS]		= { .type = NLA_U32 },
 	[TCA_FLOW_MODE]		= { .type = NLA_U32 },
@@ -381,6 +390,7 @@ static const struct nla_policy flow_policy[TCA_FLOW_MAX + 1] = {
 	[TCA_FLOW_ACT]		= { .type = NLA_NESTED },
 	[TCA_FLOW_POLICE]	= { .type = NLA_NESTED },
 	[TCA_FLOW_EMATCHES]	= { .type = NLA_NESTED },
+	[TCA_FLOW_PERTURB]	= { .type = NLA_U32 },
 };
 
 static int flow_change(struct tcf_proto *tp, unsigned long base,
@@ -394,6 +404,7 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 	struct tcf_exts e;
 	struct tcf_ematch_tree t;
 	unsigned int nkeys = 0;
+	unsigned int perturb_period = 0;
 	u32 baseclass = 0;
 	u32 keymask = 0;
 	u32 mode;
@@ -442,6 +453,14 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 			mode = nla_get_u32(tb[TCA_FLOW_MODE]);
 		if (mode != FLOW_MODE_HASH && nkeys > 1)
 			goto err2;
+
+		if (mode == FLOW_MODE_HASH)
+			perturb_period = f->perturb_period;
+		if (tb[TCA_FLOW_PERTURB]) {
+			if (mode != FLOW_MODE_HASH)
+				goto err2;
+			perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ;
+		}
 	} else {
 		err = -EINVAL;
 		if (!handle)
@@ -455,6 +474,12 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 		if (mode != FLOW_MODE_HASH && nkeys > 1)
 			goto err2;
 
+		if (tb[TCA_FLOW_PERTURB]) {
+			if (mode != FLOW_MODE_HASH)
+				goto err2;
+			perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ;
+		}
+
 		if (TC_H_MAJ(baseclass) == 0)
 			baseclass = TC_H_MAKE(tp->q->handle, baseclass);
 		if (TC_H_MIN(baseclass) == 0)
@@ -467,6 +492,11 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 
 		f->handle = handle;
 		f->mask	  = ~0U;
+
+		get_random_bytes(&f->hashrnd, 4);
+		f->perturb_timer.function = flow_perturbation;
+		f->perturb_timer.data = (unsigned long)f;
+		init_timer_deferrable(&f->perturb_timer);
 	}
 
 	tcf_exts_change(tp, &f->exts, &e);
@@ -495,6 +525,11 @@ static int flow_change(struct tcf_proto *tp, unsigned long base,
 	if (baseclass)
 		f->baseclass = baseclass;
 
+	f->perturb_period = perturb_period;
+	del_timer(&f->perturb_timer);
+	if (perturb_period)
+		mod_timer(&f->perturb_timer, jiffies + perturb_period);
+
 	if (*arg == 0)
 		list_add_tail(&f->list, &head->filters);
 
@@ -512,6 +547,7 @@ err1:
 
 static void flow_destroy_filter(struct tcf_proto *tp, struct flow_filter *f)
 {
+	del_timer_sync(&f->perturb_timer);
 	tcf_exts_destroy(tp, &f->exts);
 	tcf_em_tree_destroy(tp, &f->ematches);
 	kfree(f);
@@ -532,11 +568,6 @@ static int flow_init(struct tcf_proto *tp)
 {
 	struct flow_head *head;
 
-	if (!flow_hashrnd_initted) {
-		get_random_bytes(&flow_hashrnd, 4);
-		flow_hashrnd_initted = 1;
-	}
-
 	head = kzalloc(sizeof(*head), GFP_KERNEL);
 	if (head == NULL)
 		return -ENOBUFS;
@@ -605,6 +636,9 @@ static int flow_dump(struct tcf_proto *tp, unsigned long fh,
 	if (f->baseclass)
 		NLA_PUT_U32(skb, TCA_FLOW_BASECLASS, f->baseclass);
 
+	if (f->perturb_period)
+		NLA_PUT_U32(skb, TCA_FLOW_PERTURB, f->perturb_period / HZ);
+
 	if (tcf_exts_dump(skb, &f->exts, &flow_ext_map) < 0)
 		goto nla_put_failure;
 #ifdef CONFIG_NET_EMATCH
-- 
cgit v1.2.3


From f271b2cc78f09c93ccd00a2056d3237134bf994c Mon Sep 17 00:00:00 2001
From: Max Krasnyansky <maxk@qualcomm.com>
Date: Mon, 14 Jul 2008 22:18:19 -0700
Subject: tun: Fix/rewrite packet filtering logic

Please see the following thread to get some context on this
	http://marc.info/?l=linux-netdev&m=121564433018903&w=2

Basically the issue is that current multi-cast filtering stuff in
the TUN/TAP driver is seriously broken.
Original patch went in without proper review and ACK. It was broken and
confusing to start with and subsequent patches broke it completely.
To give you an idea of what's broken here are some of the issues:

- Very confusing comments throughout the code that imply that the
character device is a network interface in its own right, and that packets
are passed between the two nics. Which is completely wrong.

- Wrong set of ioctls is used for setting up filters. They look like
shortcuts for manipulating state of the tun/tap network interface but
in reality manipulate the state of the TX filter.

- ioctls that were originally used for setting address of the the TX filter
got "fixed" and now set the address of the network interface itself. Which
made filter totaly useless.

- Filtering is done too late. Instead of filtering early on, to avoid
unnecessary wakeups, filtering is done in the read() call.

The list goes on and on :)

So the patch cleans all that up. It introduces simple and clean interface for
setting up TX filters (TUNSETTXFILTER + tun_filter spec) and does filtering
before enqueuing the packets.

TX filtering is useful in the scenarios where TAP is part of a bridge, in
which case it gets all broadcast, multicast and potentially other packets when
the bridge is learning. So for example Ethernet tunnelling app may want to
setup TX filters to avoid tunnelling multicast traffic. QEMU and other
hypervisors can push RX filtering that is currently done in the guest into the
host context therefore saving wakeups and unnecessary data transfer.

Signed-off-by: Max Krasnyansky <maxk@qualcomm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tun.c      | 316 ++++++++++++++++++++++++-------------------------
 include/linux/if_tun.h |  24 +++-
 2 files changed, 174 insertions(+), 166 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 2693f883ecda..901551c8ca0e 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -18,15 +18,11 @@
 /*
  *  Changes:
  *
- *  Brian Braunstein <linuxkernel@bristyle.com> 2007/03/23
- *    Fixed hw address handling.  Now net_device.dev_addr is kept consistent
- *    with tun.dev_addr when the address is set by this module.
- *
  *  Mike Kershaw <dragorn@kismetwireless.net> 2005/08/14
  *    Add TUNSETLINK ioctl to set the link encapsulation
  *
  *  Mark Smith <markzzzsmith@yahoo.com.au>
- *   Use random_ether_addr() for tap MAC address.
+ *    Use random_ether_addr() for tap MAC address.
  *
  *  Harald Roelle <harald.roelle@ifi.lmu.de>  2004/04/20
  *    Fixes in packet dropping, queue length setting and queue wakeup.
@@ -83,9 +79,16 @@ static int debug;
 #define DBG1( a... )
 #endif
 
+#define FLT_EXACT_COUNT 8
+struct tap_filter {
+	unsigned int    count;    /* Number of addrs. Zero means disabled */
+	u32             mask[2];  /* Mask of the hashed addrs */
+	unsigned char	addr[FLT_EXACT_COUNT][ETH_ALEN];
+};
+
 struct tun_struct {
 	struct list_head        list;
-	unsigned long 		flags;
+	unsigned int 		flags;
 	int			attached;
 	uid_t			owner;
 	gid_t			group;
@@ -94,19 +97,119 @@ struct tun_struct {
 	struct sk_buff_head	readq;
 
 	struct net_device	*dev;
+	struct fasync_struct	*fasync;
 
-	struct fasync_struct    *fasync;
-
-	unsigned long if_flags;
-	u8 dev_addr[ETH_ALEN];
-	u32 chr_filter[2];
-	u32 net_filter[2];
+	struct tap_filter       txflt;
 
 #ifdef TUN_DEBUG
 	int debug;
 #endif
 };
 
+/* TAP filterting */
+static void addr_hash_set(u32 *mask, const u8 *addr)
+{
+	int n = ether_crc(ETH_ALEN, addr) >> 26;
+	mask[n >> 5] |= (1 << (n & 31));
+}
+
+static unsigned int addr_hash_test(const u32 *mask, const u8 *addr)
+{
+	int n = ether_crc(ETH_ALEN, addr) >> 26;
+	return mask[n >> 5] & (1 << (n & 31));
+}
+
+static int update_filter(struct tap_filter *filter, void __user *arg)
+{
+	struct { u8 u[ETH_ALEN]; } *addr;
+	struct tun_filter uf;
+	int err, alen, n, nexact;
+
+	if (copy_from_user(&uf, arg, sizeof(uf)))
+		return -EFAULT;
+
+	if (!uf.count) {
+		/* Disabled */
+		filter->count = 0;
+		return 0;
+	}
+
+	alen = ETH_ALEN * uf.count;
+	addr = kmalloc(alen, GFP_KERNEL);
+	if (!addr)
+		return -ENOMEM;
+
+	if (copy_from_user(addr, arg + sizeof(uf), alen)) {
+		err = -EFAULT;
+		goto done;
+	}
+
+	/* The filter is updated without holding any locks. Which is
+	 * perfectly safe. We disable it first and in the worst
+	 * case we'll accept a few undesired packets. */
+	filter->count = 0;
+	wmb();
+
+	/* Use first set of addresses as an exact filter */
+	for (n = 0; n < uf.count && n < FLT_EXACT_COUNT; n++)
+		memcpy(filter->addr[n], addr[n].u, ETH_ALEN);
+
+	nexact = n;
+
+	/* The rest is hashed */
+	memset(filter->mask, 0, sizeof(filter->mask));
+	for (; n < uf.count; n++)
+		addr_hash_set(filter->mask, addr[n].u);
+
+	/* For ALLMULTI just set the mask to all ones.
+	 * This overrides the mask populated above. */
+	if ((uf.flags & TUN_FLT_ALLMULTI))
+		memset(filter->mask, ~0, sizeof(filter->mask));
+
+	/* Now enable the filter */
+	wmb();
+	filter->count = nexact;
+
+	/* Return the number of exact filters */
+	err = nexact;
+
+done:
+	kfree(addr);
+	return err;
+}
+
+/* Returns: 0 - drop, !=0 - accept */
+static int run_filter(struct tap_filter *filter, const struct sk_buff *skb)
+{
+	/* Cannot use eth_hdr(skb) here because skb_mac_hdr() is incorrect
+	 * at this point. */
+	struct ethhdr *eh = (struct ethhdr *) skb->data;
+	int i;
+
+	/* Exact match */
+	for (i = 0; i < filter->count; i++)
+		if (!compare_ether_addr(eh->h_dest, filter->addr[i]))
+			return 1;
+
+	/* Inexact match (multicast only) */
+	if (is_multicast_ether_addr(eh->h_dest))
+		return addr_hash_test(filter->mask, eh->h_dest);
+
+	return 0;
+}
+
+/*
+ * Checks whether the packet is accepted or not.
+ * Returns: 0 - drop, !=0 - accept
+ */
+static int check_filter(struct tap_filter *filter, const struct sk_buff *skb)
+{
+	if (!filter->count)
+		return 1;
+
+	return run_filter(filter, skb);
+}
+
 /* Network device part of the driver */
 
 static unsigned int tun_net_id;
@@ -141,7 +244,12 @@ static int tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (!tun->attached)
 		goto drop;
 
-	/* Packet dropping */
+	/* Drop if the filter does not like it.
+	 * This is a noop if the filter is disabled.
+	 * Filter can be enabled only for the TAP devices. */
+	if (!check_filter(&tun->txflt, skb))
+		goto drop;
+
 	if (skb_queue_len(&tun->readq) >= dev->tx_queue_len) {
 		if (!(tun->flags & TUN_ONE_QUEUE)) {
 			/* Normal queueing mode. */
@@ -158,7 +266,7 @@ static int tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 	}
 
-	/* Queue packet */
+	/* Enqueue packet */
 	skb_queue_tail(&tun->readq, skb);
 	dev->trans_start = jiffies;
 
@@ -174,41 +282,14 @@ drop:
 	return 0;
 }
 
-/** Add the specified Ethernet address to this multicast filter. */
-static void
-add_multi(u32* filter, const u8* addr)
-{
-	int bit_nr = ether_crc(ETH_ALEN, addr) >> 26;
-	filter[bit_nr >> 5] |= 1 << (bit_nr & 31);
-}
-
-/** Remove the specified Ethernet addres from this multicast filter. */
-static void
-del_multi(u32* filter, const u8* addr)
+static void tun_net_mclist(struct net_device *dev)
 {
-	int bit_nr = ether_crc(ETH_ALEN, addr) >> 26;
-	filter[bit_nr >> 5] &= ~(1 << (bit_nr & 31));
-}
-
-/** Update the list of multicast groups to which the network device belongs.
- * This list is used to filter packets being sent from the character device to
- * the network device. */
-static void
-tun_net_mclist(struct net_device *dev)
-{
-	struct tun_struct *tun = netdev_priv(dev);
-	const struct dev_mc_list *mclist;
-	int i;
-	DECLARE_MAC_BUF(mac);
-	DBG(KERN_DEBUG "%s: tun_net_mclist: mc_count %d\n",
-			dev->name, dev->mc_count);
-	memset(tun->chr_filter, 0, sizeof tun->chr_filter);
-	for (i = 0, mclist = dev->mc_list; i < dev->mc_count && mclist != NULL;
-			i++, mclist = mclist->next) {
-		add_multi(tun->net_filter, mclist->dmi_addr);
-		DBG(KERN_DEBUG "%s: tun_net_mclist: %s\n",
-		    dev->name, print_mac(mac, mclist->dmi_addr));
-	}
+	/*
+	 * This callback is supposed to deal with mc filter in
+	 * _rx_ path and has nothing to do with the _tx_ path.
+	 * In rx path we always accept everything userspace gives us.
+	 */
+	return;
 }
 
 #define MIN_MTU 68
@@ -244,13 +325,11 @@ static void tun_net_init(struct net_device *dev)
 
 	case TUN_TAP_DEV:
 		/* Ethernet TAP Device */
-		dev->set_multicast_list = tun_net_mclist;
-
 		ether_setup(dev);
-		dev->change_mtu = tun_net_change_mtu;
+		dev->change_mtu         = tun_net_change_mtu;
+		dev->set_multicast_list = tun_net_mclist;
 
-		/* random address already created for us by tun_set_iff, use it */
-		memcpy(dev->dev_addr, tun->dev_addr, min(sizeof(tun->dev_addr), sizeof(dev->dev_addr)) );
+		random_ether_addr(dev->dev_addr);
 
 		dev->tx_queue_len = TUN_READQ_SIZE;  /* We prefer our own queue length */
 		break;
@@ -486,7 +565,6 @@ static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
 	DECLARE_WAITQUEUE(wait, current);
 	struct sk_buff *skb;
 	ssize_t len, ret = 0;
-	DECLARE_MAC_BUF(mac);
 
 	if (!tun)
 		return -EBADFD;
@@ -499,10 +577,6 @@ static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
 
 	add_wait_queue(&tun->read_wait, &wait);
 	while (len) {
-		const u8 ones[ ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
-		u8 addr[ ETH_ALEN];
-		int bit_nr;
-
 		current->state = TASK_INTERRUPTIBLE;
 
 		/* Read frames from the queue */
@@ -522,36 +596,9 @@ static ssize_t tun_chr_aio_read(struct kiocb *iocb, const struct iovec *iv,
 		}
 		netif_wake_queue(tun->dev);
 
-		/** Decide whether to accept this packet. This code is designed to
-		 * behave identically to an Ethernet interface. Accept the packet if
-		 * - we are promiscuous.
-		 * - the packet is addressed to us.
-		 * - the packet is broadcast.
-		 * - the packet is multicast and
-		 *   - we are multicast promiscous.
-		 *   - we belong to the multicast group.
-		 */
-		skb_copy_from_linear_data(skb, addr, min_t(size_t, sizeof addr,
-								   skb->len));
-		bit_nr = ether_crc(sizeof addr, addr) >> 26;
-		if ((tun->if_flags & IFF_PROMISC) ||
-				memcmp(addr, tun->dev_addr, sizeof addr) == 0 ||
-				memcmp(addr, ones, sizeof addr) == 0 ||
-				(((addr[0] == 1 && addr[1] == 0 && addr[2] == 0x5e) ||
-				  (addr[0] == 0x33 && addr[1] == 0x33)) &&
-				 ((tun->if_flags & IFF_ALLMULTI) ||
-				  (tun->chr_filter[bit_nr >> 5] & (1 << (bit_nr & 31)))))) {
-			DBG(KERN_DEBUG "%s: tun_chr_readv: accepted: %s\n",
-					tun->dev->name, print_mac(mac, addr));
-			ret = tun_put_user(tun, skb, (struct iovec *) iv, len);
-			kfree_skb(skb);
-			break;
-		} else {
-			DBG(KERN_DEBUG "%s: tun_chr_readv: rejected: %s\n",
-					tun->dev->name, print_mac(mac, addr));
-			kfree_skb(skb);
-			continue;
-		}
+		ret = tun_put_user(tun, skb, (struct iovec *) iv, len);
+		kfree_skb(skb);
+		break;
 	}
 
 	current->state = TASK_RUNNING;
@@ -647,12 +694,7 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
 		tun = netdev_priv(dev);
 		tun->dev = dev;
 		tun->flags = flags;
-		/* Be promiscuous by default to maintain previous behaviour. */
-		tun->if_flags = IFF_PROMISC;
-		/* Generate random Ethernet address. */
-		*(__be16 *)tun->dev_addr = htons(0x00FF);
-		get_random_bytes(tun->dev_addr + sizeof(u16), 4);
-		memset(tun->chr_filter, 0, sizeof tun->chr_filter);
+		tun->txflt.count = 0;
 
 		tun_net_init(dev);
 
@@ -751,6 +793,7 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 	struct tun_struct *tun = file->private_data;
 	void __user* argp = (void __user*)arg;
 	struct ifreq ifr;
+	int ret;
 	DECLARE_MAC_BUF(mac);
 
 	if (cmd == TUNSETIFF || _IOC_TYPE(cmd) == 0x89)
@@ -826,9 +869,6 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		break;
 
 	case TUNSETLINK:
-	{
-		int ret;
-
 		/* Only allow setting the type when the interface is down */
 		rtnl_lock();
 		if (tun->dev->flags & IFF_UP) {
@@ -842,94 +882,44 @@ static int tun_chr_ioctl(struct inode *inode, struct file *file,
 		}
 		rtnl_unlock();
 		return ret;
-	}
 
 #ifdef TUN_DEBUG
 	case TUNSETDEBUG:
 		tun->debug = arg;
 		break;
 #endif
-
 	case TUNSETOFFLOAD:
-	{
-		int ret;
 		rtnl_lock();
 		ret = set_offload(tun->dev, arg);
 		rtnl_unlock();
 		return ret;
-	}
 
-	case SIOCGIFFLAGS:
-		ifr.ifr_flags = tun->if_flags;
-		if (copy_to_user( argp, &ifr, sizeof ifr))
-			return -EFAULT;
-		return 0;
-
-	case SIOCSIFFLAGS:
-		/** Set the character device's interface flags. Currently only
-		 * IFF_PROMISC and IFF_ALLMULTI are used. */
-		tun->if_flags = ifr.ifr_flags;
-		DBG(KERN_INFO "%s: interface flags 0x%lx\n",
-				tun->dev->name, tun->if_flags);
-		return 0;
+	case TUNSETTXFILTER:
+		/* Can be set only for TAPs */
+		if ((tun->flags & TUN_TYPE_MASK) != TUN_TAP_DEV)
+			return -EINVAL;
+		rtnl_lock();
+		ret = update_filter(&tun->txflt, (void *) __user arg);
+		rtnl_unlock();
+		return ret;
 
 	case SIOCGIFHWADDR:
-		/* Note: the actual net device's address may be different */
-		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev_addr,
-				min(sizeof ifr.ifr_hwaddr.sa_data, sizeof tun->dev_addr));
-		if (copy_to_user( argp, &ifr, sizeof ifr))
+		/* Get hw addres */
+		memcpy(ifr.ifr_hwaddr.sa_data, tun->dev->dev_addr, ETH_ALEN);
+		ifr.ifr_hwaddr.sa_family = tun->dev->type;
+		if (copy_to_user(argp, &ifr, sizeof ifr))
 			return -EFAULT;
 		return 0;
 
 	case SIOCSIFHWADDR:
-	{
-		/* try to set the actual net device's hw address */
-		int ret;
+		/* Set hw address */
+		DBG(KERN_DEBUG "%s: set hw address: %s\n",
+			tun->dev->name, print_mac(mac, ifr.ifr_hwaddr.sa_data));
 
 		rtnl_lock();
 		ret = dev_set_mac_address(tun->dev, &ifr.ifr_hwaddr);
 		rtnl_unlock();
-
-		if (ret == 0) {
-			/** Set the character device's hardware address. This is used when
-			 * filtering packets being sent from the network device to the character
-			 * device. */
-			memcpy(tun->dev_addr, ifr.ifr_hwaddr.sa_data,
-					min(sizeof ifr.ifr_hwaddr.sa_data, sizeof tun->dev_addr));
-			DBG(KERN_DEBUG "%s: set hardware address: %x:%x:%x:%x:%x:%x\n",
-					tun->dev->name,
-					tun->dev_addr[0], tun->dev_addr[1], tun->dev_addr[2],
-					tun->dev_addr[3], tun->dev_addr[4], tun->dev_addr[5]);
-		}
-
-		return  ret;
-	}
-
-	case SIOCADDMULTI:
-		/** Add the specified group to the character device's multicast filter
-		 * list. */
-		rtnl_lock();
-		netif_tx_lock_bh(tun->dev);
-		add_multi(tun->chr_filter, ifr.ifr_hwaddr.sa_data);
-		netif_tx_unlock_bh(tun->dev);
-		rtnl_unlock();
-
-		DBG(KERN_DEBUG "%s: add multi: %s\n",
-		    tun->dev->name, print_mac(mac, ifr.ifr_hwaddr.sa_data));
-		return 0;
-
-	case SIOCDELMULTI:
-		/** Remove the specified group from the character device's multicast
-		 * filter list. */
-		rtnl_lock();
-		netif_tx_lock_bh(tun->dev);
-		del_multi(tun->chr_filter, ifr.ifr_hwaddr.sa_data);
-		netif_tx_unlock_bh(tun->dev);
-		rtnl_unlock();
-
-		DBG(KERN_DEBUG "%s: del multi: %s\n",
-		    tun->dev->name, print_mac(mac, ifr.ifr_hwaddr.sa_data));
-		return 0;
+		return ret;
 
 	default:
 		return -EINVAL;
diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h
index 563fae542da6..4c6307ad9fdb 100644
--- a/include/linux/if_tun.h
+++ b/include/linux/if_tun.h
@@ -17,6 +17,7 @@
 #define __IF_TUN_H
 
 #include <linux/types.h>
+#include <linux/if_ether.h>
 
 /* Read queue size */
 #define TUN_READQ_SIZE	500
@@ -42,7 +43,8 @@
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
 #define TUNGETFEATURES _IOR('T', 207, unsigned int)
-#define TUNSETOFFLOAD _IOW('T', 208, unsigned int)
+#define TUNSETOFFLOAD  _IOW('T', 208, unsigned int)
+#define TUNSETTXFILTER _IOW('T', 209, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001
@@ -57,10 +59,26 @@
 #define TUN_F_TSO6	0x04	/* I can handle TSO for IPv6 packets */
 #define TUN_F_TSO_ECN	0x08	/* I can handle TSO with ECN bits. */
 
+/* Protocol info prepended to the packets (when IFF_NO_PI is not set) */
+#define TUN_PKT_STRIP	0x0001
 struct tun_pi {
-	unsigned short flags;
+	__u16  flags;
 	__be16 proto;
 };
-#define TUN_PKT_STRIP	0x0001
+
+/*
+ * Filter spec (used for SETXXFILTER ioctls)
+ * This stuff is applicable only to the TAP (Ethernet) devices.
+ * If the count is zero the filter is disabled and the driver accepts
+ * all packets (promisc mode).
+ * If the filter is enabled in order to accept broadcast packets
+ * broadcast addr must be explicitly included in the addr list.
+ */
+#define TUN_FLT_ALLMULTI 0x0001 /* Accept all multicast packets */
+struct tun_filter {
+	__u16  flags; /* TUN_FLT_ flags see above */
+	__u16  count; /* Number of addresses */
+	__u8   addr[0][ETH_ALEN];
+};
 
 #endif /* __IF_TUN_H */
-- 
cgit v1.2.3


From 6aa895b047720f71ec4eb11452f7c3ce8426941f Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:49:06 -0700
Subject: vlan: Don't store VLAN tag in cb

Use a real skb member to store the skb to avoid clashes with qdiscs,
which are allowed to use the cb area themselves. As currently only real
devices that consume the skb set the NETIF_F_HW_VLAN_TX flag, no explicit
invalidation is neccessary.

The new member fills a hole on 64 bit, the skb layout changes from:

        __u32                      mark;                 /*   172     4 */
        sk_buff_data_t             transport_header;     /*   176     4 */
        sk_buff_data_t             network_header;       /*   180     4 */
        sk_buff_data_t             mac_header;           /*   184     4 */
        sk_buff_data_t             tail;                 /*   188     4 */
        /* --- cacheline 3 boundary (192 bytes) --- */
        sk_buff_data_t             end;                  /*   192     4 */

        /* XXX 4 bytes hole, try to pack */

to

        __u32                      mark;                 /*   172     4 */
        __u16                      vlan_tci;             /*   176     2 */

        /* XXX 2 bytes hole, try to pack */

        sk_buff_data_t             transport_header;     /*   180     4 */
        sk_buff_data_t             network_header;       /*   184     4 */

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h | 31 +++++++------------------------
 include/linux/skbuff.h  |  3 +++
 net/core/skbuff.c       |  3 +++
 3 files changed, 13 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 93f5d9b0e9f9..9e7b49b8062d 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -105,17 +105,8 @@ static inline void vlan_group_set_device(struct vlan_group *vg,
 	array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN] = dev;
 }
 
-/* VLAN tx hw acceleration helpers. */
-struct vlan_skb_tx_cookie {
-	u32	magic;
-	u32	vlan_tag;
-};
-
-#define VLAN_TX_COOKIE_MAGIC	0x564c414e	/* "VLAN" in ascii. */
-#define VLAN_TX_SKB_CB(__skb)	((struct vlan_skb_tx_cookie *)&((__skb)->cb[0]))
-#define vlan_tx_tag_present(__skb) \
-	(VLAN_TX_SKB_CB(__skb)->magic == VLAN_TX_COOKIE_MAGIC)
-#define vlan_tx_tag_get(__skb)	(VLAN_TX_SKB_CB(__skb)->vlan_tag)
+#define vlan_tx_tag_present(__skb)	((__skb)->vlan_tci)
+#define vlan_tx_tag_get(__skb)		((__skb)->vlan_tci)
 
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
 extern struct net_device *vlan_dev_real_dev(const struct net_device *dev);
@@ -210,17 +201,12 @@ static inline struct sk_buff *__vlan_put_tag(struct sk_buff *skb, u16 vlan_tci)
  * @skb: skbuff to tag
  * @vlan_tci: VLAN TCI to insert
  *
- * Puts the VLAN TCI in @skb->cb[] and lets the device do the rest
+ * Puts the VLAN TCI in @skb->vlan_tci and lets the device do the rest
  */
 static inline struct sk_buff *__vlan_hwaccel_put_tag(struct sk_buff *skb,
 						     u16 vlan_tci)
 {
-	struct vlan_skb_tx_cookie *cookie;
-
-	cookie = VLAN_TX_SKB_CB(skb);
-	cookie->magic = VLAN_TX_COOKIE_MAGIC;
-	cookie->vlan_tag = vlan_tci;
-
+	skb->vlan_tci = vlan_tci;
 	return skb;
 }
 
@@ -267,16 +253,13 @@ static inline int __vlan_get_tag(const struct sk_buff *skb, u16 *vlan_tci)
  * @skb: skbuff to query
  * @vlan_tci: buffer to store vlaue
  *
- * Returns error if @skb->cb[] is not set correctly
+ * Returns error if @skb->vlan_tci is not set correctly
  */
 static inline int __vlan_hwaccel_get_tag(const struct sk_buff *skb,
 					 u16 *vlan_tci)
 {
-	struct vlan_skb_tx_cookie *cookie;
-
-	cookie = VLAN_TX_SKB_CB(skb);
-	if (cookie->magic == VLAN_TX_COOKIE_MAGIC) {
-		*vlan_tci = cookie->vlan_tag;
+	if (vlan_tx_tag_present(skb)) {
+		*vlan_tci = skb->vlan_tci;
 		return 0;
 	} else {
 		*vlan_tci = 0;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 8f10e3d08fd9..7ea44f6621f2 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -246,6 +246,7 @@ typedef unsigned char *sk_buff_data_t;
  *	@dma_cookie: a cookie to one of several possible DMA operations
  *		done by skb DMA functions
  *	@secmark: security marking
+ *	@vlan_tci: vlan tag control information
  */
 
 struct sk_buff {
@@ -326,6 +327,8 @@ struct sk_buff {
 
 	__u32			mark;
 
+	__u16			vlan_tci;
+
 	sk_buff_data_t		transport_header;
 	sk_buff_data_t		network_header;
 	sk_buff_data_t		mac_header;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 7c571560e9d2..50a853f7cd8e 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -459,6 +459,8 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
 	new->tc_verd		= old->tc_verd;
 #endif
 #endif
+	new->vlan_tci		= old->vlan_tci;
+
 	skb_copy_secmark(new, old);
 }
 
@@ -2286,6 +2288,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features)
 		skb_copy_queue_mapping(nskb, skb);
 		nskb->priority = skb->priority;
 		nskb->protocol = skb->protocol;
+		nskb->vlan_tci = skb->vlan_tci;
 		nskb->dst = dst_clone(skb->dst);
 		memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
 		nskb->pkt_type = skb->pkt_type;
-- 
cgit v1.2.3


From bc1d0411b804ad190cdadabac48a10067f17b9e6 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:49:30 -0700
Subject: vlan: deliver packets received with VLAN acceleration to network taps

When VLAN header stripping is used, packets currently bypass packet
sockets (and other network taps) completely. For locally existing
VLANs, they appear directly on the VLAN device, for unknown VLANs
they are silently dropped.

Add a new function netif_nit_deliver() to deliver incoming packets
to all network interface taps and use it in __vlan_hwaccel_rx() to
make VLAN packets visible on the underlying device.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 +
 net/8021q/vlan_core.c     |  4 ++++
 net/core/dev.c            | 27 +++++++++++++++++++++++++++
 3 files changed, 32 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b54ec16dfbda..ba5c4639ea91 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1165,6 +1165,7 @@ extern int		netif_rx(struct sk_buff *skb);
 extern int		netif_rx_ni(struct sk_buff *skb);
 #define HAVE_NETIF_RECEIVE_SKB 1
 extern int		netif_receive_skb(struct sk_buff *skb);
+extern void		netif_nit_deliver(struct sk_buff *skb);
 extern int		dev_valid_name(const char *name);
 extern int		dev_ioctl(struct net *net, unsigned int cmd, void __user *);
 extern int		dev_ethtool(struct net *net, struct ifreq *);
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 68df12d3664b..916061f681b6 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -14,6 +14,9 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_DROP;
 	}
 
+	skb->vlan_tci = vlan_tci;
+	netif_nit_deliver(skb);
+
 	skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
 	if (skb->dev == NULL) {
 		dev_kfree_skb_any(skb);
@@ -22,6 +25,7 @@ int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		return NET_RX_SUCCESS;
 	}
 	skb->dev->last_rx = jiffies;
+	skb->vlan_tci = 0;
 
 	stats = &skb->dev->stats;
 	stats->rx_packets++;
diff --git a/net/core/dev.c b/net/core/dev.c
index a29a359b15d1..feaab4898a5b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2068,6 +2068,33 @@ out:
 }
 #endif
 
+/*
+ * 	netif_nit_deliver - deliver received packets to network taps
+ * 	@skb: buffer
+ *
+ * 	This function is used to deliver incoming packets to network
+ * 	taps. It should be used when the normal netif_receive_skb path
+ * 	is bypassed, for example because of VLAN acceleration.
+ */
+void netif_nit_deliver(struct sk_buff *skb)
+{
+	struct packet_type *ptype;
+
+	if (list_empty(&ptype_all))
+		return;
+
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	skb->mac_len = skb->network_header - skb->mac_header;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(ptype, &ptype_all, list) {
+		if (!ptype->dev || ptype->dev == skb->dev)
+			deliver_skb(skb, ptype, skb->dev);
+	}
+	rcu_read_unlock();
+}
+
 /**
  *	netif_receive_skb - process receive buffer from network
  *	@skb: buffer to process
-- 
cgit v1.2.3


From bbd6ef87c544d88c30e4b762b1b61ef267a7d279 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:50:15 -0700
Subject: packet: support extensible, 64 bit clean mmaped ring structure

The tpacket_hdr is not 64 bit clean due to use of an unsigned long
and can't be extended because the following struct sockaddr_ll needs
to be at a fixed offset.

Add support for a version 2 tpacket protocol that removes these
limitations.

Userspace can query the header size through a new getsockopt option
and change the protocol version through a setsockopt option. The
changes needed to switch to the new protocol version are:

1. replace struct tpacket_hdr by struct tpacket2_hdr
2. query header len and save
3. set protocol version to 2
 - set up ring as usual
4. for getting the sockaddr_ll, use (void *)hdr + TPACKET_ALIGN(hdrlen)
   instead of (void *)hdr + TPACKET_ALIGN(sizeof(struct tpacket_hdr))

Steps 2 and 4 can be omitted if the struct sockaddr_ll isn't needed.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_packet.h |  21 ++++++
 net/packet/af_packet.c    | 179 +++++++++++++++++++++++++++++++++++++---------
 2 files changed, 167 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h
index ad09609227ff..d4d3c82448f5 100644
--- a/include/linux/if_packet.h
+++ b/include/linux/if_packet.h
@@ -43,6 +43,8 @@ struct sockaddr_ll
 #define PACKET_COPY_THRESH		7
 #define PACKET_AUXDATA			8
 #define PACKET_ORIGDEV			9
+#define PACKET_VERSION			10
+#define PACKET_HDRLEN			11
 
 struct tpacket_stats
 {
@@ -79,6 +81,25 @@ struct tpacket_hdr
 #define TPACKET_ALIGN(x)	(((x)+TPACKET_ALIGNMENT-1)&~(TPACKET_ALIGNMENT-1))
 #define TPACKET_HDRLEN		(TPACKET_ALIGN(sizeof(struct tpacket_hdr)) + sizeof(struct sockaddr_ll))
 
+struct tpacket2_hdr
+{
+	__u32		tp_status;
+	__u32		tp_len;
+	__u32		tp_snaplen;
+	__u16		tp_mac;
+	__u16		tp_net;
+	__u32		tp_sec;
+	__u32		tp_nsec;
+};
+
+#define TPACKET2_HDRLEN		(TPACKET_ALIGN(sizeof(struct tpacket2_hdr)) + sizeof(struct sockaddr_ll))
+
+enum tpacket_versions
+{
+	TPACKET_V1,
+	TPACKET_V2,
+};
+
 /*
    Frame structure:
 
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 9f2269166687..4f059775d48f 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -186,6 +186,8 @@ struct packet_sock {
 	unsigned int            pg_vec_order;
 	unsigned int		pg_vec_pages;
 	unsigned int		pg_vec_len;
+	enum tpacket_versions	tp_version;
+	unsigned int		tp_hdrlen;
 #endif
 };
 
@@ -201,14 +203,52 @@ struct packet_skb_cb {
 
 #ifdef CONFIG_PACKET_MMAP
 
-static inline struct tpacket_hdr *packet_lookup_frame(struct packet_sock *po, unsigned int position)
+static void *packet_lookup_frame(struct packet_sock *po, unsigned int position,
+				 int status)
 {
 	unsigned int pg_vec_pos, frame_offset;
+	union {
+		struct tpacket_hdr *h1;
+		struct tpacket2_hdr *h2;
+		void *raw;
+	} h;
 
 	pg_vec_pos = position / po->frames_per_block;
 	frame_offset = position % po->frames_per_block;
 
-	return (struct tpacket_hdr *)(po->pg_vec[pg_vec_pos] + (frame_offset * po->frame_size));
+	h.raw = po->pg_vec[pg_vec_pos] + (frame_offset * po->frame_size);
+	switch (po->tp_version) {
+	case TPACKET_V1:
+		if (status != h.h1->tp_status ? TP_STATUS_USER :
+						TP_STATUS_KERNEL)
+			return NULL;
+		break;
+	case TPACKET_V2:
+		if (status != h.h2->tp_status ? TP_STATUS_USER :
+						TP_STATUS_KERNEL)
+			return NULL;
+		break;
+	}
+	return h.raw;
+}
+
+static void __packet_set_status(struct packet_sock *po, void *frame, int status)
+{
+	union {
+		struct tpacket_hdr *h1;
+		struct tpacket2_hdr *h2;
+		void *raw;
+	} h;
+
+	h.raw = frame;
+	switch (po->tp_version) {
+	case TPACKET_V1:
+		h.h1->tp_status = status;
+		break;
+	case TPACKET_V2:
+		h.h2->tp_status = status;
+		break;
+	}
 }
 #endif
 
@@ -551,14 +591,19 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	struct sock *sk;
 	struct packet_sock *po;
 	struct sockaddr_ll *sll;
-	struct tpacket_hdr *h;
+	union {
+		struct tpacket_hdr *h1;
+		struct tpacket2_hdr *h2;
+		void *raw;
+	} h;
 	u8 * skb_head = skb->data;
 	int skb_len = skb->len;
 	unsigned int snaplen, res;
 	unsigned long status = TP_STATUS_LOSING|TP_STATUS_USER;
-	unsigned short macoff, netoff;
+	unsigned short macoff, netoff, hdrlen;
 	struct sk_buff *copy_skb = NULL;
 	struct timeval tv;
+	struct timespec ts;
 
 	if (skb->pkt_type == PACKET_LOOPBACK)
 		goto drop;
@@ -590,10 +635,11 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 		snaplen = res;
 
 	if (sk->sk_type == SOCK_DGRAM) {
-		macoff = netoff = TPACKET_ALIGN(TPACKET_HDRLEN) + 16;
+		macoff = netoff = TPACKET_ALIGN(po->tp_hdrlen) + 16;
 	} else {
 		unsigned maclen = skb_network_offset(skb);
-		netoff = TPACKET_ALIGN(TPACKET_HDRLEN + (maclen < 16 ? 16 : maclen));
+		netoff = TPACKET_ALIGN(po->tp_hdrlen +
+				       (maclen < 16 ? 16 : maclen));
 		macoff = netoff - maclen;
 	}
 
@@ -616,9 +662,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	}
 
 	spin_lock(&sk->sk_receive_queue.lock);
-	h = packet_lookup_frame(po, po->head);
-
-	if (h->tp_status)
+	h.raw = packet_lookup_frame(po, po->head, TP_STATUS_KERNEL);
+	if (!h.raw)
 		goto ring_is_full;
 	po->head = po->head != po->frame_max ? po->head+1 : 0;
 	po->stats.tp_packets++;
@@ -630,20 +675,40 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 		status &= ~TP_STATUS_LOSING;
 	spin_unlock(&sk->sk_receive_queue.lock);
 
-	skb_copy_bits(skb, 0, (u8*)h + macoff, snaplen);
+	skb_copy_bits(skb, 0, h.raw + macoff, snaplen);
 
-	h->tp_len = skb->len;
-	h->tp_snaplen = snaplen;
-	h->tp_mac = macoff;
-	h->tp_net = netoff;
-	if (skb->tstamp.tv64)
-		tv = ktime_to_timeval(skb->tstamp);
-	else
-		do_gettimeofday(&tv);
-	h->tp_sec = tv.tv_sec;
-	h->tp_usec = tv.tv_usec;
+	switch (po->tp_version) {
+	case TPACKET_V1:
+		h.h1->tp_len = skb->len;
+		h.h1->tp_snaplen = snaplen;
+		h.h1->tp_mac = macoff;
+		h.h1->tp_net = netoff;
+		if (skb->tstamp.tv64)
+			tv = ktime_to_timeval(skb->tstamp);
+		else
+			do_gettimeofday(&tv);
+		h.h1->tp_sec = tv.tv_sec;
+		h.h1->tp_usec = tv.tv_usec;
+		hdrlen = sizeof(*h.h1);
+		break;
+	case TPACKET_V2:
+		h.h2->tp_len = skb->len;
+		h.h2->tp_snaplen = snaplen;
+		h.h2->tp_mac = macoff;
+		h.h2->tp_net = netoff;
+		if (skb->tstamp.tv64)
+			ts = ktime_to_timespec(skb->tstamp);
+		else
+			getnstimeofday(&ts);
+		h.h2->tp_sec = ts.tv_sec;
+		h.h2->tp_nsec = ts.tv_nsec;
+		hdrlen = sizeof(*h.h2);
+		break;
+	default:
+		BUG();
+	}
 
-	sll = (struct sockaddr_ll*)((u8*)h + TPACKET_ALIGN(sizeof(*h)));
+	sll = h.raw + TPACKET_ALIGN(hdrlen);
 	sll->sll_halen = dev_parse_header(skb, sll->sll_addr);
 	sll->sll_family = AF_PACKET;
 	sll->sll_hatype = dev->type;
@@ -654,14 +719,14 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 	else
 		sll->sll_ifindex = dev->ifindex;
 
-	h->tp_status = status;
+	__packet_set_status(po, h.raw, status);
 	smp_mb();
 
 	{
 		struct page *p_start, *p_end;
-		u8 *h_end = (u8 *)h + macoff + snaplen - 1;
+		u8 *h_end = h.raw + macoff + snaplen - 1;
 
-		p_start = virt_to_page(h);
+		p_start = virt_to_page(h.raw);
 		p_end = virt_to_page(h_end);
 		while (p_start <= p_end) {
 			flush_dcache_page(p_start);
@@ -1362,6 +1427,25 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv
 		pkt_sk(sk)->copy_thresh = val;
 		return 0;
 	}
+	case PACKET_VERSION:
+	{
+		int val;
+
+		if (optlen != sizeof(val))
+			return -EINVAL;
+		if (po->pg_vec)
+			return -EBUSY;
+		if (copy_from_user(&val, optval, sizeof(val)))
+			return -EFAULT;
+		switch (val) {
+		case TPACKET_V1:
+		case TPACKET_V2:
+			po->tp_version = val;
+			return 0;
+		default:
+			return -EINVAL;
+		}
+	}
 #endif
 	case PACKET_AUXDATA:
 	{
@@ -1437,6 +1521,31 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 
 		data = &val;
 		break;
+#ifdef CONFIG_PACKET_MMAP
+	case PACKET_VERSION:
+		if (len > sizeof(int))
+			len = sizeof(int);
+		val = po->tp_version;
+		data = &val;
+		break;
+	case PACKET_HDRLEN:
+		if (len > sizeof(int))
+			len = sizeof(int);
+		if (copy_from_user(&val, optval, len))
+			return -EFAULT;
+		switch (val) {
+		case TPACKET_V1:
+			val = sizeof(struct tpacket_hdr);
+			break;
+		case TPACKET_V2:
+			val = sizeof(struct tpacket2_hdr);
+			break;
+		default:
+			return -EINVAL;
+		}
+		data = &val;
+		break;
+#endif
 	default:
 		return -ENOPROTOOPT;
 	}
@@ -1570,11 +1679,8 @@ static unsigned int packet_poll(struct file * file, struct socket *sock,
 	spin_lock_bh(&sk->sk_receive_queue.lock);
 	if (po->pg_vec) {
 		unsigned last = po->head ? po->head-1 : po->frame_max;
-		struct tpacket_hdr *h;
-
-		h = packet_lookup_frame(po, last);
 
-		if (h->tp_status)
+		if (packet_lookup_frame(po, last, TP_STATUS_USER))
 			mask |= POLLIN | POLLRDNORM;
 	}
 	spin_unlock_bh(&sk->sk_receive_queue.lock);
@@ -1669,11 +1775,20 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing
 		if (unlikely(po->pg_vec))
 			return -EBUSY;
 
+		switch (po->tp_version) {
+		case TPACKET_V1:
+			po->tp_hdrlen = TPACKET_HDRLEN;
+			break;
+		case TPACKET_V2:
+			po->tp_hdrlen = TPACKET2_HDRLEN;
+			break;
+		}
+
 		if (unlikely((int)req->tp_block_size <= 0))
 			return -EINVAL;
 		if (unlikely(req->tp_block_size & (PAGE_SIZE - 1)))
 			return -EINVAL;
-		if (unlikely(req->tp_frame_size < TPACKET_HDRLEN))
+		if (unlikely(req->tp_frame_size < po->tp_hdrlen))
 			return -EINVAL;
 		if (unlikely(req->tp_frame_size & (TPACKET_ALIGNMENT - 1)))
 			return -EINVAL;
@@ -1692,13 +1807,11 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing
 			goto out;
 
 		for (i = 0; i < req->tp_block_nr; i++) {
-			char *ptr = pg_vec[i];
-			struct tpacket_hdr *header;
+			void *ptr = pg_vec[i];
 			int k;
 
 			for (k = 0; k < po->frames_per_block; k++) {
-				header = (struct tpacket_hdr *) ptr;
-				header->tp_status = TP_STATUS_KERNEL;
+				__packet_set_status(po, ptr, TP_STATUS_KERNEL);
 				ptr += req->tp_frame_size;
 			}
 		}
-- 
cgit v1.2.3


From 393e52e33c6c26ec7db290dab803bac1bed962d4 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 14 Jul 2008 22:50:39 -0700
Subject: packet: deliver VLAN TCI to userspace

Store the VLAN tag in the auxillary data/tpacket2_hdr so userspace can
properly deal with hardware VLAN tagging/stripping.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_packet.h | 2 ++
 net/packet/af_packet.c    | 2 ++
 2 files changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h
index d4d3c82448f5..a630295b255f 100644
--- a/include/linux/if_packet.h
+++ b/include/linux/if_packet.h
@@ -59,6 +59,7 @@ struct tpacket_auxdata
 	__u32		tp_snaplen;
 	__u16		tp_mac;
 	__u16		tp_net;
+	__u16		tp_vlan_tci;
 };
 
 struct tpacket_hdr
@@ -90,6 +91,7 @@ struct tpacket2_hdr
 	__u16		tp_net;
 	__u32		tp_sec;
 	__u32		tp_nsec;
+	__u16		tp_vlan_tci;
 };
 
 #define TPACKET2_HDRLEN		(TPACKET_ALIGN(sizeof(struct tpacket2_hdr)) + sizeof(struct sockaddr_ll))
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 4f059775d48f..db792e02a37f 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -702,6 +702,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 			getnstimeofday(&ts);
 		h.h2->tp_sec = ts.tv_sec;
 		h.h2->tp_nsec = ts.tv_nsec;
+		h.h2->tp_vlan_tci = skb->vlan_tci;
 		hdrlen = sizeof(*h.h2);
 		break;
 	default:
@@ -1172,6 +1173,7 @@ static int packet_recvmsg(struct kiocb *iocb, struct socket *sock,
 		aux.tp_snaplen = skb->len;
 		aux.tp_mac = 0;
 		aux.tp_net = skb_network_offset(skb);
+		aux.tp_vlan_tci = skb->vlan_tci;
 
 		put_cmsg(msg, SOL_PACKET, PACKET_AUXDATA, sizeof(aux), &aux);
 	}
-- 
cgit v1.2.3


From 521e575b9a7324a0bca762622139f69582a042bf Mon Sep 17 00:00:00 2001
From: Ron Livne <ronli@voltaire.com>
Date: Mon, 14 Jul 2008 23:48:48 -0700
Subject: IB/mlx4: Add support for blocking multicast loopback packets

Add support for handling the IB_QP_CREATE_MULTICAST_BLOCK_LOOPBACK
flag by using the per-multicast group loopback blocking feature of
mlx4 hardware.

Signed-off-by: Ron Livne <ronli@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/main.c    |  7 +++++--
 drivers/infiniband/hw/mlx4/mlx4_ib.h |  3 ++-
 drivers/infiniband/hw/mlx4/qp.c      | 21 ++++++++++++++++++---
 drivers/net/mlx4/mcg.c               | 17 +++++++++++++----
 include/linux/mlx4/device.h          |  3 ++-
 5 files changed, 40 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 4d61e32866c6..bcf50648fa18 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -90,7 +90,8 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
 	props->device_cap_flags    = IB_DEVICE_CHANGE_PHY_PORT |
 		IB_DEVICE_PORT_ACTIVE_EVENT		|
 		IB_DEVICE_SYS_IMAGE_GUID		|
-		IB_DEVICE_RC_RNR_NAK_GEN;
+		IB_DEVICE_RC_RNR_NAK_GEN		|
+		IB_DEVICE_BLOCK_MULTICAST_LOOPBACK;
 	if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_BAD_PKEY_CNTR)
 		props->device_cap_flags |= IB_DEVICE_BAD_PKEY_CNTR;
 	if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_BAD_QKEY_CNTR)
@@ -437,7 +438,9 @@ static int mlx4_ib_dealloc_pd(struct ib_pd *pd)
 static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
 {
 	return mlx4_multicast_attach(to_mdev(ibqp->device)->dev,
-				     &to_mqp(ibqp)->mqp, gid->raw);
+				     &to_mqp(ibqp)->mqp, gid->raw,
+				     !!(to_mqp(ibqp)->flags &
+					MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK));
 }
 
 static int mlx4_ib_mcg_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index 5cf994794d25..c4cf5b69eefa 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -101,7 +101,8 @@ struct mlx4_ib_wq {
 };
 
 enum mlx4_ib_qp_flags {
-	MLX4_IB_QP_LSO		= 1 << 0
+	MLX4_IB_QP_LSO				= 1 << 0,
+	MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK	= 1 << 1,
 };
 
 struct mlx4_ib_qp {
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 44bbd6c2e315..91590e7fba0c 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -511,6 +511,9 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 	} else {
 		qp->sq_no_prefetch = 0;
 
+		if (init_attr->create_flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK)
+			qp->flags |= MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK;
+
 		if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO)
 			qp->flags |= MLX4_IB_QP_LSO;
 
@@ -684,10 +687,15 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd,
 	struct mlx4_ib_qp *qp;
 	int err;
 
-	/* We only support LSO, and only for kernel UD QPs. */
-	if (init_attr->create_flags & ~IB_QP_CREATE_IPOIB_UD_LSO)
+	/*
+	 * We only support LSO and multicast loopback blocking, and
+	 * only for kernel UD QPs.
+	 */
+	if (init_attr->create_flags & ~(IB_QP_CREATE_IPOIB_UD_LSO |
+					IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK))
 		return ERR_PTR(-EINVAL);
-	if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO &&
+
+	if (init_attr->create_flags &&
 	    (pd->uobject || init_attr->qp_type != IB_QPT_UD))
 		return ERR_PTR(-EINVAL);
 
@@ -1844,6 +1852,13 @@ done:
 
 	qp_init_attr->cap	     = qp_attr->cap;
 
+	qp_init_attr->create_flags = 0;
+	if (qp->flags & MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK)
+		qp_init_attr->create_flags |= IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK;
+
+	if (qp->flags & MLX4_IB_QP_LSO)
+		qp_init_attr->create_flags |= IB_QP_CREATE_IPOIB_UD_LSO;
+
 out:
 	mutex_unlock(&qp->mutex);
 	return err;
diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c
index 57f7f1f0d4ec..b4b57870ddfd 100644
--- a/drivers/net/mlx4/mcg.c
+++ b/drivers/net/mlx4/mcg.c
@@ -38,6 +38,9 @@
 
 #include "mlx4.h"
 
+#define MGM_QPN_MASK       0x00FFFFFF
+#define MGM_BLCK_LB_BIT    30
+
 struct mlx4_mgm {
 	__be32			next_gid_index;
 	__be32			members_count;
@@ -153,7 +156,8 @@ static int find_mgm(struct mlx4_dev *dev,
 	return err;
 }
 
-int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16])
+int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  int block_mcast_loopback)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_cmd_mailbox *mailbox;
@@ -202,13 +206,18 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16])
 	}
 
 	for (i = 0; i < members_count; ++i)
-		if (mgm->qp[i] == cpu_to_be32(qp->qpn)) {
+		if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qp->qpn) {
 			mlx4_dbg(dev, "QP %06x already a member of MGM\n", qp->qpn);
 			err = 0;
 			goto out;
 		}
 
-	mgm->qp[members_count++] = cpu_to_be32(qp->qpn);
+	if (block_mcast_loopback)
+		mgm->qp[members_count++] = cpu_to_be32((qp->qpn & MGM_QPN_MASK) |
+						       (1 << MGM_BLCK_LB_BIT));
+	else
+		mgm->qp[members_count++] = cpu_to_be32(qp->qpn & MGM_QPN_MASK);
+
 	mgm->members_count       = cpu_to_be32(members_count);
 
 	err = mlx4_WRITE_MCG(dev, index, mailbox);
@@ -283,7 +292,7 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16])
 
 	members_count = be32_to_cpu(mgm->members_count);
 	for (loc = -1, i = 0; i < members_count; ++i)
-		if (mgm->qp[i] == cpu_to_be32(qp->qpn))
+		if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qp->qpn)
 			loc = i;
 
 	if (loc == -1) {
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index a744383d16e9..81b3dd5206e0 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -398,7 +398,8 @@ int mlx4_srq_query(struct mlx4_dev *dev, struct mlx4_srq *srq, int *limit_waterm
 int mlx4_INIT_PORT(struct mlx4_dev *dev, int port);
 int mlx4_CLOSE_PORT(struct mlx4_dev *dev, int port);
 
-int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]);
+int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  int block_mcast_loopback);
 int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]);
 
 int mlx4_map_phys_fmr(struct mlx4_dev *dev, struct mlx4_fmr *fmr, u64 *page_list,
-- 
cgit v1.2.3


From f1f28aa3510ddb84c966bac65611bb866c77a092 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 00:08:33 -0700
Subject: netdev: Add addr_list_lock to struct net_device.

This will be used to protect the per-device unicast and multicast
address lists, as well as the callbacks into the drivers which
configure such state such as ->set_rx_mode() and ->set_multicast_list().

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 1 +
 net/core/dev.c            | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index ba5c4639ea91..fd0365219181 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -609,6 +609,7 @@ struct net_device
 	unsigned char		addr_len;	/* hardware address length	*/
 	unsigned short          dev_id;		/* for shared network cards */
 
+	spinlock_t		addr_list_lock;
 	struct dev_addr_list	*uc_list;	/* Secondary unicast mac addresses */
 	int			uc_count;	/* Number of installed ucasts	*/
 	int			uc_promisc;
diff --git a/net/core/dev.c b/net/core/dev.c
index feaab4898a5b..d933d1bfa6fa 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3836,6 +3836,7 @@ int register_netdevice(struct net_device *dev)
 	BUG_ON(!dev_net(dev));
 	net = dev_net(dev);
 
+	spin_lock_init(&dev->addr_list_lock);
 	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
-- 
cgit v1.2.3


From e308a5d806c852f56590ffdd3834d0df0cbed8d7 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 00:13:44 -0700
Subject: netdev: Add netdev->addr_list_lock protection.

Add netif_addr_{lock,unlock}{,_bh}() helpers.

Use them to protect operations that operate on or read
the network device unicast and multicast address lists.

Also use them in cases where the code simply wants to
block calls into the driver's ->set_rx_mode() and
->set_multicast_list() methods.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c |  2 ++
 drivers/media/dvb/dvb-core/dvb_net.c           |  2 ++
 drivers/net/bonding/bond_main.c                |  8 ++++++++
 drivers/net/forcedeth.c                        | 16 ++++++++++++++++
 drivers/net/hamradio/6pack.c                   |  2 ++
 drivers/net/hamradio/mkiss.c                   |  2 ++
 drivers/net/ibm_newemac/core.c                 |  4 ++++
 drivers/net/sfc/efx.c                          |  2 ++
 drivers/net/wireless/libertas/main.c           |  2 ++
 include/linux/netdevice.h                      | 20 ++++++++++++++++++++
 net/core/dev.c                                 | 14 ++++++++++++++
 net/core/dev_mcast.c                           | 12 ++++++++++++
 net/mac80211/main.c                            |  4 ++++
 net/mac80211/mlme.c                            |  4 ++++
 14 files changed, 94 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 3f663fb852c1..261ab7150431 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -775,6 +775,7 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 
 	local_irq_save(flags);
 	netif_tx_lock(dev);
+	netif_addr_lock(dev);
 	spin_lock(&priv->lock);
 
 	/*
@@ -851,6 +852,7 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 	}
 
 	spin_unlock(&priv->lock);
+	netif_addr_unlock(dev);
 	netif_tx_unlock(dev);
 	local_irq_restore(flags);
 
diff --git a/drivers/media/dvb/dvb-core/dvb_net.c b/drivers/media/dvb/dvb-core/dvb_net.c
index c2334aef4143..809d18c663bc 100644
--- a/drivers/media/dvb/dvb-core/dvb_net.c
+++ b/drivers/media/dvb/dvb-core/dvb_net.c
@@ -1134,6 +1134,7 @@ static void wq_set_multicast_list (struct work_struct *work)
 	dvb_net_feed_stop(dev);
 	priv->rx_mode = RX_MODE_UNI;
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 
 	if (dev->flags & IFF_PROMISC) {
 		dprintk("%s: promiscuous mode\n", dev->name);
@@ -1158,6 +1159,7 @@ static void wq_set_multicast_list (struct work_struct *work)
 		}
 	}
 
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	dvb_net_feed_start(dev);
 }
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 8ae7ff313218..ea71abd6f728 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1568,10 +1568,12 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 		}
 
 		netif_tx_lock_bh(bond_dev);
+		netif_addr_lock(bond_dev);
 		/* upload master's mc_list to new slave */
 		for (dmi = bond_dev->mc_list; dmi; dmi = dmi->next) {
 			dev_mc_add (slave_dev, dmi->dmi_addr, dmi->dmi_addrlen, 0);
 		}
+		netif_addr_unlock(bond_dev);
 		netif_tx_unlock_bh(bond_dev);
 	}
 
@@ -1937,7 +1939,9 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 
 		/* flush master's mc_list from slave */
 		netif_tx_lock_bh(bond_dev);
+		netif_addr_lock(bond_dev);
 		bond_mc_list_flush(bond_dev, slave_dev);
+		netif_addr_unlock(bond_dev);
 		netif_tx_unlock_bh(bond_dev);
 	}
 
@@ -2060,7 +2064,9 @@ static int bond_release_all(struct net_device *bond_dev)
 
 			/* flush master's mc_list from slave */
 			netif_tx_lock_bh(bond_dev);
+			netif_addr_lock(bond_dev);
 			bond_mc_list_flush(bond_dev, slave_dev);
+			netif_addr_unlock(bond_dev);
 			netif_tx_unlock_bh(bond_dev);
 		}
 
@@ -4674,7 +4680,9 @@ static void bond_free_all(void)
 
 		bond_work_cancel_all(bond);
 		netif_tx_lock_bh(bond_dev);
+		netif_addr_lock(bond_dev);
 		bond_mc_list_destroy(bond);
+		netif_addr_unlock(bond_dev);
 		netif_tx_unlock_bh(bond_dev);
 		/* Release the bonded slaves */
 		bond_release_all(bond_dev);
diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c
index 786d668c612e..4ed89fa9ae46 100644
--- a/drivers/net/forcedeth.c
+++ b/drivers/net/forcedeth.c
@@ -2831,6 +2831,7 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu)
 		 */
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock(&np->lock);
 		/* stop engines */
 		nv_stop_rxtx(dev);
@@ -2855,6 +2856,7 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu)
 		/* restart rx engine */
 		nv_start_rxtx(dev);
 		spin_unlock(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 		nv_enable_irq(dev);
 	}
@@ -2891,6 +2893,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr)
 
 	if (netif_running(dev)) {
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock_irq(&np->lock);
 
 		/* stop rx engine */
@@ -2902,6 +2905,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr)
 		/* restart rx engine */
 		nv_start_rx(dev);
 		spin_unlock_irq(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 	} else {
 		nv_copy_mac_to_hw(dev);
@@ -3971,6 +3975,7 @@ static void nv_do_nic_poll(unsigned long data)
 		printk(KERN_INFO "forcedeth: MAC in recoverable error state\n");
 		if (netif_running(dev)) {
 			netif_tx_lock_bh(dev);
+			netif_addr_lock(dev);
 			spin_lock(&np->lock);
 			/* stop engines */
 			nv_stop_rxtx(dev);
@@ -3995,6 +4000,7 @@ static void nv_do_nic_poll(unsigned long data)
 			/* restart rx engine */
 			nv_start_rxtx(dev);
 			spin_unlock(&np->lock);
+			netif_addr_unlock(dev);
 			netif_tx_unlock_bh(dev);
 		}
 	}
@@ -4202,6 +4208,7 @@ static int nv_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd)
 
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		/* with plain spinlock lockdep complains */
 		spin_lock_irqsave(&np->lock, flags);
 		/* stop engines */
@@ -4215,6 +4222,7 @@ static int nv_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd)
 		 */
 		nv_stop_rxtx(dev);
 		spin_unlock_irqrestore(&np->lock, flags);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 	}
 
@@ -4360,10 +4368,12 @@ static int nv_nway_reset(struct net_device *dev)
 		if (netif_running(dev)) {
 			nv_disable_irq(dev);
 			netif_tx_lock_bh(dev);
+			netif_addr_lock(dev);
 			spin_lock(&np->lock);
 			/* stop engines */
 			nv_stop_rxtx(dev);
 			spin_unlock(&np->lock);
+			netif_addr_unlock(dev);
 			netif_tx_unlock_bh(dev);
 			printk(KERN_INFO "%s: link down.\n", dev->name);
 		}
@@ -4471,6 +4481,7 @@ static int nv_set_ringparam(struct net_device *dev, struct ethtool_ringparam* ri
 	if (netif_running(dev)) {
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock(&np->lock);
 		/* stop engines */
 		nv_stop_rxtx(dev);
@@ -4519,6 +4530,7 @@ static int nv_set_ringparam(struct net_device *dev, struct ethtool_ringparam* ri
 		/* restart engines */
 		nv_start_rxtx(dev);
 		spin_unlock(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 		nv_enable_irq(dev);
 	}
@@ -4556,10 +4568,12 @@ static int nv_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam*
 	if (netif_running(dev)) {
 		nv_disable_irq(dev);
 		netif_tx_lock_bh(dev);
+		netif_addr_lock(dev);
 		spin_lock(&np->lock);
 		/* stop engines */
 		nv_stop_rxtx(dev);
 		spin_unlock(&np->lock);
+		netif_addr_unlock(dev);
 		netif_tx_unlock_bh(dev);
 	}
 
@@ -4946,6 +4960,7 @@ static void nv_self_test(struct net_device *dev, struct ethtool_test *test, u64
 			napi_disable(&np->napi);
 #endif
 			netif_tx_lock_bh(dev);
+			netif_addr_lock(dev);
 			spin_lock_irq(&np->lock);
 			nv_disable_hw_interrupts(dev, np->irqmask);
 			if (!(np->msi_flags & NV_MSI_X_ENABLED)) {
@@ -4959,6 +4974,7 @@ static void nv_self_test(struct net_device *dev, struct ethtool_test *test, u64
 			/* drain rx queue */
 			nv_drain_rxtx(dev);
 			spin_unlock_irq(&np->lock);
+			netif_addr_unlock(dev);
 			netif_tx_unlock_bh(dev);
 		}
 
diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c
index 06ad9f302b5a..ffc937f5d15d 100644
--- a/drivers/net/hamradio/6pack.c
+++ b/drivers/net/hamradio/6pack.c
@@ -300,7 +300,9 @@ static int sp_set_mac_address(struct net_device *dev, void *addr)
 	struct sockaddr_ax25 *sa = addr;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 
 	return 0;
diff --git a/drivers/net/hamradio/mkiss.c b/drivers/net/hamradio/mkiss.c
index 65166035aca0..b8740e6a5cec 100644
--- a/drivers/net/hamradio/mkiss.c
+++ b/drivers/net/hamradio/mkiss.c
@@ -356,7 +356,9 @@ static int ax_set_mac_address(struct net_device *dev, void *addr)
 	struct sockaddr_ax25 *sa = addr;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 
 	return 0;
diff --git a/drivers/net/ibm_newemac/core.c b/drivers/net/ibm_newemac/core.c
index babc79ad490b..9ca57d365599 100644
--- a/drivers/net/ibm_newemac/core.c
+++ b/drivers/net/ibm_newemac/core.c
@@ -295,7 +295,9 @@ static void emac_rx_disable(struct emac_instance *dev)
 static inline void emac_netif_stop(struct emac_instance *dev)
 {
 	netif_tx_lock_bh(dev->ndev);
+	netif_addr_lock(dev->ndev);
 	dev->no_mcast = 1;
+	netif_addr_unlock(dev->ndev);
 	netif_tx_unlock_bh(dev->ndev);
 	dev->ndev->trans_start = jiffies;	/* prevent tx timeout */
 	mal_poll_disable(dev->mal, &dev->commac);
@@ -305,9 +307,11 @@ static inline void emac_netif_stop(struct emac_instance *dev)
 static inline void emac_netif_start(struct emac_instance *dev)
 {
 	netif_tx_lock_bh(dev->ndev);
+	netif_addr_lock(dev->ndev);
 	dev->no_mcast = 0;
 	if (dev->mcast_pending && netif_running(dev->ndev))
 		__emac_set_multicast_list(dev);
+	netif_addr_unlock(dev->ndev);
 	netif_tx_unlock_bh(dev->ndev);
 
 	netif_wake_queue(dev->ndev);
diff --git a/drivers/net/sfc/efx.c b/drivers/net/sfc/efx.c
index 74265d8553b8..e1257e556e48 100644
--- a/drivers/net/sfc/efx.c
+++ b/drivers/net/sfc/efx.c
@@ -697,6 +697,8 @@ static void efx_stop_port(struct efx_nic *efx)
 	/* Serialise against efx_set_multicast_list() */
 	if (efx_dev_registered(efx)) {
 		netif_tx_lock_bh(efx->net_dev);
+		netif_addr_lock(efx->net_dev);
+		netif_addr_unlock(efx->net_dev);
 		netif_tx_unlock_bh(efx->net_dev);
 	}
 }
diff --git a/drivers/net/wireless/libertas/main.c b/drivers/net/wireless/libertas/main.c
index abd6d9ed8f4b..42e9b2771eab 100644
--- a/drivers/net/wireless/libertas/main.c
+++ b/drivers/net/wireless/libertas/main.c
@@ -594,6 +594,7 @@ static int lbs_add_mcast_addrs(struct cmd_ds_mac_multicast_adr *cmd,
 		return nr_addrs;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	for (mc_list = dev->mc_list; mc_list; mc_list = mc_list->next) {
 		if (mac_in_list(cmd->maclist, nr_addrs, mc_list->dmi_addr)) {
 			lbs_deb_net("mcast address %s:%s skipped\n", dev->name,
@@ -608,6 +609,7 @@ static int lbs_add_mcast_addrs(struct cmd_ds_mac_multicast_adr *cmd,
 			    print_mac(mac, mc_list->dmi_addr));
 		i++;
 	}
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	if (mc_list)
 		return -EOVERFLOW;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fd0365219181..570cf7affa72 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1498,6 +1498,26 @@ static inline void netif_tx_disable(struct net_device *dev)
 	netif_tx_unlock_bh(dev);
 }
 
+static inline void netif_addr_lock(struct net_device *dev)
+{
+	spin_lock(&dev->addr_list_lock);
+}
+
+static inline void netif_addr_lock_bh(struct net_device *dev)
+{
+	spin_lock_bh(&dev->addr_list_lock);
+}
+
+static inline void netif_addr_unlock(struct net_device *dev)
+{
+	spin_unlock(&dev->addr_list_lock);
+}
+
+static inline void netif_addr_unlock_bh(struct net_device *dev)
+{
+	spin_unlock_bh(&dev->addr_list_lock);
+}
+
 /* These functions live elsewhere (drivers/net/net_init.c, but related) */
 
 extern void		ether_setup(struct net_device *dev);
diff --git a/net/core/dev.c b/net/core/dev.c
index d933d1bfa6fa..ef1502d71f25 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2982,7 +2982,9 @@ void __dev_set_rx_mode(struct net_device *dev)
 void dev_set_rx_mode(struct net_device *dev)
 {
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 }
 
@@ -3062,9 +3064,11 @@ int dev_unicast_delete(struct net_device *dev, void *addr, int alen)
 	ASSERT_RTNL();
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_delete(&dev->uc_list, &dev->uc_count, addr, alen, 0);
 	if (!err)
 		__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -3088,9 +3092,11 @@ int dev_unicast_add(struct net_device *dev, void *addr, int alen)
 	ASSERT_RTNL();
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_add(&dev->uc_list, &dev->uc_count, addr, alen, 0);
 	if (!err)
 		__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -3159,10 +3165,12 @@ int dev_unicast_sync(struct net_device *to, struct net_device *from)
 	int err = 0;
 
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 	err = __dev_addr_sync(&to->uc_list, &to->uc_count,
 			      &from->uc_list, &from->uc_count);
 	if (!err)
 		__dev_set_rx_mode(to);
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
 	return err;
 }
@@ -3180,13 +3188,17 @@ EXPORT_SYMBOL(dev_unicast_sync);
 void dev_unicast_unsync(struct net_device *to, struct net_device *from)
 {
 	netif_tx_lock_bh(from);
+	netif_addr_lock(from);
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 
 	__dev_addr_unsync(&to->uc_list, &to->uc_count,
 			  &from->uc_list, &from->uc_count);
 	__dev_set_rx_mode(to);
 
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
+	netif_addr_unlock(from);
 	netif_tx_unlock_bh(from);
 }
 EXPORT_SYMBOL(dev_unicast_unsync);
@@ -3208,6 +3220,7 @@ static void __dev_addr_discard(struct dev_addr_list **list)
 static void dev_addr_discard(struct net_device *dev)
 {
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 
 	__dev_addr_discard(&dev->uc_list);
 	dev->uc_count = 0;
@@ -3215,6 +3228,7 @@ static void dev_addr_discard(struct net_device *dev)
 	__dev_addr_discard(&dev->mc_list);
 	dev->mc_count = 0;
 
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 }
 
diff --git a/net/core/dev_mcast.c b/net/core/dev_mcast.c
index f8a3455f4493..b6b2a129971a 100644
--- a/net/core/dev_mcast.c
+++ b/net/core/dev_mcast.c
@@ -73,6 +73,7 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl)
 	int err;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_delete(&dev->mc_list, &dev->mc_count,
 				addr, alen, glbl);
 	if (!err) {
@@ -83,6 +84,7 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl)
 
 		__dev_set_rx_mode(dev);
 	}
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -96,9 +98,11 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl)
 	int err;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	err = __dev_addr_add(&dev->mc_list, &dev->mc_count, addr, alen, glbl);
 	if (!err)
 		__dev_set_rx_mode(dev);
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return err;
 }
@@ -120,10 +124,12 @@ int dev_mc_sync(struct net_device *to, struct net_device *from)
 	int err = 0;
 
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 	err = __dev_addr_sync(&to->mc_list, &to->mc_count,
 			      &from->mc_list, &from->mc_count);
 	if (!err)
 		__dev_set_rx_mode(to);
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
 
 	return err;
@@ -144,13 +150,17 @@ EXPORT_SYMBOL(dev_mc_sync);
 void dev_mc_unsync(struct net_device *to, struct net_device *from)
 {
 	netif_tx_lock_bh(from);
+	netif_addr_lock(from);
 	netif_tx_lock_bh(to);
+	netif_addr_lock(to);
 
 	__dev_addr_unsync(&to->mc_list, &to->mc_count,
 			  &from->mc_list, &from->mc_count);
 	__dev_set_rx_mode(to);
 
+	netif_addr_unlock(to);
 	netif_tx_unlock_bh(to);
+	netif_addr_unlock(from);
 	netif_tx_unlock_bh(from);
 }
 EXPORT_SYMBOL(dev_mc_unsync);
@@ -165,6 +175,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v)
 		return 0;
 
 	netif_tx_lock_bh(dev);
+	netif_addr_lock(dev);
 	for (m = dev->mc_list; m; m = m->next) {
 		int i;
 
@@ -176,6 +187,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v)
 
 		seq_putc(seq, '\n');
 	}
+	netif_addr_unlock(dev);
 	netif_tx_unlock_bh(dev);
 	return 0;
 }
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 36859e794928..095b7d928d64 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -292,7 +292,9 @@ static int ieee80211_open(struct net_device *dev)
 			local->fif_other_bss++;
 
 		netif_tx_lock_bh(local->mdev);
+		netif_addr_lock(local->mdev);
 		ieee80211_configure_filter(local);
+		netif_addr_unlock(local->mdev);
 		netif_tx_unlock_bh(local->mdev);
 		break;
 	case IEEE80211_IF_TYPE_STA:
@@ -491,7 +493,9 @@ static int ieee80211_stop(struct net_device *dev)
 			local->fif_other_bss--;
 
 		netif_tx_lock_bh(local->mdev);
+		netif_addr_lock(local->mdev);
 		ieee80211_configure_filter(local);
+		netif_addr_unlock(local->mdev);
 		netif_tx_unlock_bh(local->mdev);
 		break;
 	case IEEE80211_IF_TYPE_MESH_POINT:
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 8f51375317dd..1232ba25e1e9 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3869,6 +3869,7 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw)
 
 
 	netif_tx_lock_bh(local->mdev);
+	netif_addr_lock(local->mdev);
 	local->filter_flags &= ~FIF_BCN_PRBRESP_PROMISC;
 	local->ops->configure_filter(local_to_hw(local),
 				     FIF_BCN_PRBRESP_PROMISC,
@@ -3876,6 +3877,7 @@ void ieee80211_scan_completed(struct ieee80211_hw *hw)
 				     local->mdev->mc_count,
 				     local->mdev->mc_list);
 
+	netif_addr_unlock(local->mdev);
 	netif_tx_unlock_bh(local->mdev);
 
 	rcu_read_lock();
@@ -4063,12 +4065,14 @@ static int ieee80211_sta_start_scan(struct net_device *dev,
 	local->scan_dev = dev;
 
 	netif_tx_lock_bh(local->mdev);
+	netif_addr_lock(local->mdev);
 	local->filter_flags |= FIF_BCN_PRBRESP_PROMISC;
 	local->ops->configure_filter(local_to_hw(local),
 				     FIF_BCN_PRBRESP_PROMISC,
 				     &local->filter_flags,
 				     local->mdev->mc_count,
 				     local->mdev->mc_list);
+	netif_addr_unlock(local->mdev);
 	netif_tx_unlock_bh(local->mdev);
 
 	/* TODO: start scan as soon as all nullfunc frames are ACKed */
-- 
cgit v1.2.3


From 4489428ab5a49a6f443d9aa17f1d891417787d7b Mon Sep 17 00:00:00 2001
From: Pierre Ossman <drzeus@drzeus.cx>
Date: Fri, 4 Apr 2008 19:36:59 +0200
Subject: sdhci: support JMicron secondary interface

JMicron chips sometimes have two interfaces to work around limitations
in Microsoft's sdhci driver. This patch allows us to use either interface.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/sdhci-pci.c | 127 ++++++++++++++++++++++++++++++++++++++++++-
 include/linux/pci_ids.h      |   1 +
 2 files changed, 126 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index ef77ed1bd114..5dcb4958e47b 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -39,12 +39,18 @@
 #define MAX_SLOTS			8
 
 struct sdhci_pci_chip;
+struct sdhci_pci_slot;
 
 struct sdhci_pci_fixes {
 	unsigned int		quirks;
 
 	int			(*probe)(struct sdhci_pci_chip*);
 
+	int			(*probe_slot)(struct sdhci_pci_slot*);
+	void			(*remove_slot)(struct sdhci_pci_slot*);
+
+	int			(*suspend)(struct sdhci_pci_chip*,
+					pm_message_t);
 	int			(*resume)(struct sdhci_pci_chip*);
 };
 
@@ -132,6 +138,38 @@ static int jmicron_probe(struct sdhci_pci_chip *chip)
 {
 	int ret;
 
+	/*
+	 * JMicron chips can have two interfaces to the same hardware
+	 * in order to work around limitations in Microsoft's driver.
+	 * We need to make sure we only bind to one of them.
+	 *
+	 * This code assumes two things:
+	 *
+	 * 1. The PCI code adds subfunctions in order.
+	 *
+	 * 2. The MMC interface has a lower subfunction number
+	 *    than the SD interface.
+	 */
+	if (chip->pdev->device == PCI_DEVICE_ID_JMICRON_JMB38X_SD) {
+		struct pci_dev *sd_dev;
+
+		sd_dev = NULL;
+		while ((sd_dev = pci_get_device(PCI_VENDOR_ID_JMICRON,
+			PCI_DEVICE_ID_JMICRON_JMB38X_MMC, sd_dev)) != NULL) {
+			if ((PCI_SLOT(chip->pdev->devfn) ==
+				PCI_SLOT(sd_dev->devfn)) &&
+				(chip->pdev->bus == sd_dev->bus))
+				break;
+		}
+
+		if (sd_dev) {
+			pci_dev_put(sd_dev);
+			dev_info(&chip->pdev->dev, "Refusing to bind to "
+				"secondary interface.\n");
+			return -ENODEV;
+		}
+	}
+
 	/*
 	 * JMicron chips need a bit of a nudge to enable the power
 	 * output pins.
@@ -145,9 +183,58 @@ static int jmicron_probe(struct sdhci_pci_chip *chip)
 	return 0;
 }
 
+static void jmicron_enable_mmc(struct sdhci_host *host, int on)
+{
+	u8 scratch;
+
+	scratch = readb(host->ioaddr + 0xC0);
+
+	if (on)
+		scratch |= 0x01;
+	else
+		scratch &= ~0x01;
+
+	writeb(scratch, host->ioaddr + 0xC0);
+}
+
+static int jmicron_probe_slot(struct sdhci_pci_slot *slot)
+{
+	/*
+	 * The secondary interface requires a bit set to get the
+	 * interrupts.
+	 */
+	if (slot->chip->pdev->device == PCI_DEVICE_ID_JMICRON_JMB38X_MMC)
+		jmicron_enable_mmc(slot->host, 1);
+
+	return 0;
+}
+
+static void jmicron_remove_slot(struct sdhci_pci_slot *slot)
+{
+	if (slot->chip->pdev->device == PCI_DEVICE_ID_JMICRON_JMB38X_MMC)
+		jmicron_enable_mmc(slot->host, 0);
+}
+
+static int jmicron_suspend(struct sdhci_pci_chip *chip, pm_message_t state)
+{
+	int i;
+
+	if (chip->pdev->device == PCI_DEVICE_ID_JMICRON_JMB38X_MMC) {
+		for (i = 0;i < chip->num_slots;i++)
+			jmicron_enable_mmc(chip->slots[i]->host, 0);
+	}
+
+	return 0;
+}
+
 static int jmicron_resume(struct sdhci_pci_chip *chip)
 {
-	int ret;
+	int ret, i;
+
+	if (chip->pdev->device == PCI_DEVICE_ID_JMICRON_JMB38X_MMC) {
+		for (i = 0;i < chip->num_slots;i++)
+			jmicron_enable_mmc(chip->slots[i]->host, 1);
+	}
 
 	ret = jmicron_pmos(chip, 1);
 	if (ret) {
@@ -165,6 +252,10 @@ static const struct sdhci_pci_fixes sdhci_jmicron = {
 
 	.probe		= jmicron_probe,
 
+	.probe_slot	= jmicron_probe_slot,
+	.remove_slot	= jmicron_remove_slot,
+
+	.suspend	= jmicron_suspend,
 	.resume		= jmicron_resume,
 };
 
@@ -225,6 +316,14 @@ static const struct pci_device_id pci_ids[] __devinitdata = {
 		.driver_data	= (kernel_ulong_t)&sdhci_jmicron,
 	},
 
+	{
+		.vendor		= PCI_VENDOR_ID_JMICRON,
+		.device		= PCI_DEVICE_ID_JMICRON_JMB38X_MMC,
+		.subvendor	= PCI_ANY_ID,
+		.subdevice	= PCI_ANY_ID,
+		.driver_data	= (kernel_ulong_t)&sdhci_jmicron,
+	},
+
 	{	/* Generic SD host controller */
 		PCI_DEVICE_CLASS((PCI_CLASS_SYSTEM_SDHCI << 8), 0xFFFF00)
 	},
@@ -301,6 +400,15 @@ static int sdhci_pci_suspend (struct pci_dev *pdev, pm_message_t state)
 		}
 	}
 
+	if (chip->fixes && chip->fixes->suspend) {
+		ret = chip->fixes->suspend(chip, state);
+		if (ret) {
+			for (i = chip->num_slots - 1;i >= 0;i--)
+				sdhci_resume_host(chip->slots[i]->host);
+			return ret;
+		}
+	}
+
 	pci_save_state(pdev);
 	pci_enable_wake(pdev, pci_choose_state(pdev, state), 0);
 	pci_disable_device(pdev);
@@ -418,12 +526,22 @@ static struct sdhci_pci_slot * __devinit sdhci_pci_probe_slot(
 		goto release;
 	}
 
+	if (chip->fixes && chip->fixes->probe_slot) {
+		ret = chip->fixes->probe_slot(slot);
+		if (ret)
+			goto unmap;
+	}
+
 	ret = sdhci_add_host(host);
 	if (ret)
-		goto unmap;
+		goto remove;
 
 	return slot;
 
+remove:
+	if (chip->fixes && chip->fixes->remove_slot)
+		chip->fixes->remove_slot(slot);
+
 unmap:
 	iounmap(host->ioaddr);
 
@@ -437,7 +555,12 @@ release:
 static void sdhci_pci_remove_slot(struct sdhci_pci_slot *slot)
 {
 	sdhci_remove_host(slot->host);
+
+	if (slot->chip->fixes && slot->chip->fixes->remove_slot)
+		slot->chip->fixes->remove_slot(slot);
+
 	pci_release_region(slot->chip->pdev, slot->pci_bar);
+
 	sdhci_free_host(slot->host);
 }
 
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 65953822c9cb..30153473bc37 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2188,6 +2188,7 @@
 #define PCI_DEVICE_ID_JMICRON_JMB366	0x2366
 #define PCI_DEVICE_ID_JMICRON_JMB368	0x2368
 #define PCI_DEVICE_ID_JMICRON_JMB38X_SD	0x2381
+#define PCI_DEVICE_ID_JMICRON_JMB38X_MMC 0x2382
 #define PCI_DEVICE_ID_JMICRON_JMB38X_MS	0x2383
 
 #define PCI_VENDOR_ID_KORENIX		0x1982
-- 
cgit v1.2.3


From 150a55683b6b0ccb66aae75a10a3a514340c7c03 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 20 May 2008 00:57:27 +0300
Subject: include/linux/mmc/mmc.h: remove CVS tags

This patch removes a CVS tag that wasn't updated for a long time.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 include/linux/mmc/mmc.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/mmc/mmc.h b/include/linux/mmc/mmc.h
index 4236fbf0b6fb..14b81f3e5232 100644
--- a/include/linux/mmc/mmc.h
+++ b/include/linux/mmc/mmc.h
@@ -16,7 +16,6 @@
  * Based strongly on code by:
  *
  * Author: Yong-iL Joh <tolkien@mizi.com>
- * Date  : $Date: 2002/06/18 12:37:30 $
  *
  * Author:  Andrew Christian
  *          15 May 2002
-- 
cgit v1.2.3


From 28f52482b41edc88cdf575aa6ed414c6e116ce10 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Tue, 17 Jun 2008 18:17:15 +0400
Subject: mmc: add support for card-detection polling

Some hosts (and boards that use mmc_spi) do not use interrupts on the CD
line, so they can't trigger mmc_detect_change. We want to poll the card
and see if there was a change. 1 second poll interval seems resonable.

This patch also implements .get_cd() host operation, that could be used
by the hosts that are able to report card-detect status without need to
talk MMC.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/core.c  | 12 +++++++++---
 include/linux/mmc/host.h | 11 +++++++++++
 2 files changed, 20 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 01ced4c5a61d..ede5d1e2e20d 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -638,6 +638,9 @@ void mmc_rescan(struct work_struct *work)
 		 */
 		mmc_bus_put(host);
 
+		if (host->ops->get_cd && host->ops->get_cd(host) == 0)
+			goto out;
+
 		mmc_claim_host(host);
 
 		mmc_power_up(host);
@@ -652,7 +655,7 @@ void mmc_rescan(struct work_struct *work)
 		if (!err) {
 			if (mmc_attach_sdio(host, ocr))
 				mmc_power_off(host);
-			return;
+			goto out;
 		}
 
 		/*
@@ -662,7 +665,7 @@ void mmc_rescan(struct work_struct *work)
 		if (!err) {
 			if (mmc_attach_sd(host, ocr))
 				mmc_power_off(host);
-			return;
+			goto out;
 		}
 
 		/*
@@ -672,7 +675,7 @@ void mmc_rescan(struct work_struct *work)
 		if (!err) {
 			if (mmc_attach_mmc(host, ocr))
 				mmc_power_off(host);
-			return;
+			goto out;
 		}
 
 		mmc_release_host(host);
@@ -683,6 +686,9 @@ void mmc_rescan(struct work_struct *work)
 
 		mmc_bus_put(host);
 	}
+out:
+	if (host->caps & MMC_CAP_NEEDS_POLL)
+		mmc_schedule_delayed_work(&host->detect, HZ);
 }
 
 void mmc_start_host(struct mmc_host *host)
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 7ab962fa1d73..6188e19d2331 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -51,8 +51,18 @@ struct mmc_ios {
 
 struct mmc_host_ops {
 	void	(*request)(struct mmc_host *host, struct mmc_request *req);
+	/*
+	 * Avoid calling these three functions too often or in a "fast path",
+	 * since underlaying controller might implement them in an expensive
+	 * and/or slow way.
+	 *
+	 * Also note that these functions might sleep, so don't call them
+	 * in the atomic contexts!
+	 */
 	void	(*set_ios)(struct mmc_host *host, struct mmc_ios *ios);
 	int	(*get_ro)(struct mmc_host *host);
+	int	(*get_cd)(struct mmc_host *host);
+
 	void	(*enable_sdio_irq)(struct mmc_host *host, int enable);
 };
 
@@ -94,6 +104,7 @@ struct mmc_host {
 #define MMC_CAP_SD_HIGHSPEED	(1 << 3)	/* Can do SD high-speed timing */
 #define MMC_CAP_SDIO_IRQ	(1 << 4)	/* Can signal pending SDIO IRQs */
 #define MMC_CAP_SPI		(1 << 5)	/* Talks only SPI protocols */
+#define MMC_CAP_NEEDS_POLL	(1 << 6)	/* Needs polling for card-detection */
 
 	/* host specific block data */
 	unsigned int		max_seg_size;	/* see blk_queue_max_segment_size */
-- 
cgit v1.2.3


From 619ef4b42128709de4d89d209b2c874f560deecd Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Tue, 17 Jun 2008 18:17:21 +0400
Subject: mmc_spi: add support for card-detection polling

This patch adds new platform data variable "caps", so platforms
could pass theirs capabilities into MMC core (for example, platforms
without interrupt on the CD line will most probably want to pass
MMC_CAP_NEEDS_POLL).

New platform get_cd() callback provided to optimize polling.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/mmc_spi.c  | 19 +++++++++++++++++--
 include/linux/spi/mmc_spi.h |  9 +++++++++
 2 files changed, 26 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 35508584ac2a..547eb857b1b3 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -1131,11 +1131,20 @@ static int mmc_spi_get_ro(struct mmc_host *mmc)
 	return 0;
 }
 
+static int mmc_spi_get_cd(struct mmc_host *mmc)
+{
+	struct mmc_spi_host *host = mmc_priv(mmc);
+
+	if (host->pdata && host->pdata->get_cd)
+		return !!host->pdata->get_cd(mmc->parent);
+	return -ENOSYS;
+}
 
 static const struct mmc_host_ops mmc_spi_ops = {
 	.request	= mmc_spi_request,
 	.set_ios	= mmc_spi_set_ios,
 	.get_ro		= mmc_spi_get_ro,
+	.get_cd		= mmc_spi_get_cd,
 };
 
 
@@ -1319,17 +1328,23 @@ static int mmc_spi_probe(struct spi_device *spi)
 			goto fail_glue_init;
 	}
 
+	/* pass platform capabilities, if any */
+	if (host->pdata)
+		mmc->caps |= host->pdata->caps;
+
 	status = mmc_add_host(mmc);
 	if (status != 0)
 		goto fail_add_host;
 
-	dev_info(&spi->dev, "SD/MMC host %s%s%s%s\n",
+	dev_info(&spi->dev, "SD/MMC host %s%s%s%s%s\n",
 			mmc->class_dev.bus_id,
 			host->dma_dev ? "" : ", no DMA",
 			(host->pdata && host->pdata->get_ro)
 				? "" : ", no WP",
 			(host->pdata && host->pdata->setpower)
-				? "" : ", no poweroff");
+				? "" : ", no poweroff",
+			(mmc->caps & MMC_CAP_NEEDS_POLL)
+				? ", cd polling" : "");
 	return 0;
 
 fail_add_host:
diff --git a/include/linux/spi/mmc_spi.h b/include/linux/spi/mmc_spi.h
index d5ca78b93a3b..a3626aedaec9 100644
--- a/include/linux/spi/mmc_spi.h
+++ b/include/linux/spi/mmc_spi.h
@@ -23,6 +23,15 @@ struct mmc_spi_platform_data {
 	/* sense switch on sd cards */
 	int (*get_ro)(struct device *);
 
+	/*
+	 * If board does not use CD interrupts, driver can optimize polling
+	 * using this function.
+	 */
+	int (*get_cd)(struct device *);
+
+	/* Capabilities to pass into mmc core (e.g. MMC_CAP_NEEDS_POLL). */
+	unsigned long caps;
+
 	/* how long to debounce card detect, in msecs */
 	u16 detect_delay;
 
-- 
cgit v1.2.3


From 08f80bb5196517a0dfe50dc7c10f234c0ff2f0e8 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Tue, 17 Jun 2008 18:17:39 +0400
Subject: mmc: change .get_ro() callback semantics

Now get_ro() callback must return 0/1 values for its logical states, and
negative errno values in case of error. If particular host instance doesn't
support RO/WP switch, it should return -ENOSYS.

This patch changes some hosts in two ways:

1. Now functions should be smart to not return negative values in
   "RO asserted" case (particularly gpio_ calls could return negative
   values for the outermost GPIOs).

   Also, board code usually passes get_ro() callbacks that directly return
   gpioreg & bit result, so at91_mci, imxmmc, pxamci and mmc_spi's get_ro()
   handlers need take special care when returning platform's values to the
   mmc core.

2. In case of host instance didn't implement get_ro() callback, it should
   really return -ENOSYS and let the mmc core decide what to do about it
   (mmc core thinks the same way as the hosts, so it isn't functional
   change).

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/sd.c       |  4 ++--
 drivers/mmc/host/at91_mci.c | 18 +++++++-----------
 drivers/mmc/host/imxmmc.c   |  9 ++++++---
 drivers/mmc/host/mmc_spi.c  |  9 ++++++---
 drivers/mmc/host/pxamci.c   |  9 ++++++---
 drivers/mmc/host/wbsd.c     |  2 +-
 include/linux/mmc/host.h    | 12 ++++++++++++
 7 files changed, 40 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c
index 7ef3b15c5e3d..b122eb9ea453 100644
--- a/drivers/mmc/core/sd.c
+++ b/drivers/mmc/core/sd.c
@@ -494,13 +494,13 @@ static int mmc_sd_init_card(struct mmc_host *host, u32 ocr,
 	 * Check if read-only switch is active.
 	 */
 	if (!oldcard) {
-		if (!host->ops->get_ro) {
+		if (!host->ops->get_ro || host->ops->get_ro(host) < 0) {
 			printk(KERN_WARNING "%s: host does not "
 				"support reading read-only "
 				"switch. assuming write-enable.\n",
 				mmc_hostname(host));
 		} else {
-			if (host->ops->get_ro(host))
+			if (host->ops->get_ro(host) > 0)
 				mmc_card_set_readonly(card);
 		}
 	}
diff --git a/drivers/mmc/host/at91_mci.c b/drivers/mmc/host/at91_mci.c
index 8979ad330a4d..b9d4ed6b29b1 100644
--- a/drivers/mmc/host/at91_mci.c
+++ b/drivers/mmc/host/at91_mci.c
@@ -793,19 +793,15 @@ static irqreturn_t at91_mmc_det_irq(int irq, void *_host)
 
 static int at91_mci_get_ro(struct mmc_host *mmc)
 {
-	int read_only = 0;
 	struct at91mci_host *host = mmc_priv(mmc);
 
-	if (host->board->wp_pin) {
-		read_only = gpio_get_value(host->board->wp_pin);
-		printk(KERN_WARNING "%s: card is %s\n", mmc_hostname(mmc),
-				(read_only ? "read-only" : "read-write") );
-	}
-	else {
-		printk(KERN_WARNING "%s: host does not support reading read-only "
-				"switch.  Assuming write-enable.\n", mmc_hostname(mmc));
-	}
-	return read_only;
+	if (host->board->wp_pin)
+		return !!gpio_get_value(host->board->wp_pin);
+	/*
+	 * Board doesn't support read only detection; let the mmc core
+	 * decide what to do.
+	 */
+	return -ENOSYS;
 }
 
 static const struct mmc_host_ops at91_mci_ops = {
diff --git a/drivers/mmc/host/imxmmc.c b/drivers/mmc/host/imxmmc.c
index eed211b2ac70..5e880c0f1349 100644
--- a/drivers/mmc/host/imxmmc.c
+++ b/drivers/mmc/host/imxmmc.c
@@ -892,9 +892,12 @@ static int imxmci_get_ro(struct mmc_host *mmc)
 	struct imxmci_host *host = mmc_priv(mmc);
 
 	if (host->pdata && host->pdata->get_ro)
-		return host->pdata->get_ro(mmc_dev(mmc));
-	/* Host doesn't support read only detection so assume writeable */
-	return 0;
+		return !!host->pdata->get_ro(mmc_dev(mmc));
+	/*
+	 * Board doesn't support read only detection; let the mmc core
+	 * decide what to do.
+	 */
+	return -ENOSYS;
 }
 
 
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 547eb857b1b3..4e82f64a96bc 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -1126,9 +1126,12 @@ static int mmc_spi_get_ro(struct mmc_host *mmc)
 	struct mmc_spi_host *host = mmc_priv(mmc);
 
 	if (host->pdata && host->pdata->get_ro)
-		return host->pdata->get_ro(mmc->parent);
-	/* board doesn't support read only detection; assume writeable */
-	return 0;
+		return !!host->pdata->get_ro(mmc->parent);
+	/*
+	 * Board doesn't support read only detection; let the mmc core
+	 * decide what to do.
+	 */
+	return -ENOSYS;
 }
 
 static int mmc_spi_get_cd(struct mmc_host *mmc)
diff --git a/drivers/mmc/host/pxamci.c b/drivers/mmc/host/pxamci.c
index d89475d36988..d39f59738866 100644
--- a/drivers/mmc/host/pxamci.c
+++ b/drivers/mmc/host/pxamci.c
@@ -374,9 +374,12 @@ static int pxamci_get_ro(struct mmc_host *mmc)
 	struct pxamci_host *host = mmc_priv(mmc);
 
 	if (host->pdata && host->pdata->get_ro)
-		return host->pdata->get_ro(mmc_dev(mmc));
-	/* Host doesn't support read only detection so assume writeable */
-	return 0;
+		return !!host->pdata->get_ro(mmc_dev(mmc));
+	/*
+	 * Board doesn't support read only detection; let the mmc core
+	 * decide what to do.
+	 */
+	return -ENOSYS;
 }
 
 static void pxamci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
diff --git a/drivers/mmc/host/wbsd.c b/drivers/mmc/host/wbsd.c
index c303e7f57ab4..67e5a9b80f53 100644
--- a/drivers/mmc/host/wbsd.c
+++ b/drivers/mmc/host/wbsd.c
@@ -939,7 +939,7 @@ static int wbsd_get_ro(struct mmc_host *mmc)
 
 	spin_unlock_bh(&host->lock);
 
-	return csr & WBSD_WRPT;
+	return !!(csr & WBSD_WRPT);
 }
 
 static const struct mmc_host_ops wbsd_ops = {
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 6188e19d2331..753b7231b887 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -58,6 +58,18 @@ struct mmc_host_ops {
 	 *
 	 * Also note that these functions might sleep, so don't call them
 	 * in the atomic contexts!
+	 *
+	 * Return values for the get_ro callback should be:
+	 *   0 for a read/write card
+	 *   1 for a read-only card
+	 *   -ENOSYS when not supported (equal to NULL callback)
+	 *   or a negative errno value when something bad happened
+	 *
+	 * Return values for the get_ro callback should be:
+	 *   0 for a absent card
+	 *   1 for a present card
+	 *   -ENOSYS when not supported (equal to NULL callback)
+	 *   or a negative errno value when something bad happened
 	 */
 	void	(*set_ios)(struct mmc_host *host, struct mmc_ios *ios);
 	int	(*get_ro)(struct mmc_host *host);
-- 
cgit v1.2.3


From ad3868b2ec96ec14a1549c9e33f5f9a2a3c6ab15 Mon Sep 17 00:00:00 2001
From: Pierre Ossman <drzeus@drzeus.cx>
Date: Sat, 28 Jun 2008 12:52:45 +0200
Subject: mmc,sdio: helper function for transfer padding

There are a lot of crappy controllers out there that cannot handle
all the request sizes that the MMC/SD/SDIO specifications require.
In case the card driver can pad the data to overcome the problems,
this commit adds a helper that calculates how much that padding
should be.

A corresponding helper is also added for SDIO, but it can also deal
with all the complexities of splitting up a large transfer efficiently.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/core.c                 | 29 +++++++++-
 drivers/mmc/core/sdio_io.c              | 99 ++++++++++++++++++++++++++++++++-
 drivers/net/wireless/libertas/if_sdio.c | 20 +++----
 include/linux/mmc/core.h                |  1 +
 include/linux/mmc/sdio_func.h           |  4 +-
 5 files changed, 137 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index ede5d1e2e20d..3ee5b8c3b5ce 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -3,7 +3,7 @@
  *
  *  Copyright (C) 2003-2004 Russell King, All Rights Reserved.
  *  SD support Copyright (C) 2004 Ian Molton, All Rights Reserved.
- *  Copyright (C) 2005-2007 Pierre Ossman, All Rights Reserved.
+ *  Copyright (C) 2005-2008 Pierre Ossman, All Rights Reserved.
  *  MMCv4 support Copyright (C) 2006 Philip Langdale, All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or modify
@@ -294,6 +294,33 @@ void mmc_set_data_timeout(struct mmc_data *data, const struct mmc_card *card)
 }
 EXPORT_SYMBOL(mmc_set_data_timeout);
 
+/**
+ *	mmc_align_data_size - pads a transfer size to a more optimal value
+ *	@card: the MMC card associated with the data transfer
+ *	@sz: original transfer size
+ *
+ *	Pads the original data size with a number of extra bytes in
+ *	order to avoid controller bugs and/or performance hits
+ *	(e.g. some controllers revert to PIO for certain sizes).
+ *
+ *	Returns the improved size, which might be unmodified.
+ *
+ *	Note that this function is only relevant when issuing a
+ *	single scatter gather entry.
+ */
+unsigned int mmc_align_data_size(struct mmc_card *card, unsigned int sz)
+{
+	/*
+	 * FIXME: We don't have a system for the controller to tell
+	 * the core about its problems yet, so for now we just 32-bit
+	 * align the size.
+	 */
+	sz = ((sz + 3) / 4) * 4;
+
+	return sz;
+}
+EXPORT_SYMBOL(mmc_align_data_size);
+
 /**
  *	__mmc_claim_host - exclusively claim a host
  *	@host: mmc host to claim
diff --git a/drivers/mmc/core/sdio_io.c b/drivers/mmc/core/sdio_io.c
index 625b92ce9cef..6ee7861fceae 100644
--- a/drivers/mmc/core/sdio_io.c
+++ b/drivers/mmc/core/sdio_io.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/mmc/core/sdio_io.c
  *
- *  Copyright 2007 Pierre Ossman
+ *  Copyright 2007-2008 Pierre Ossman
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -189,6 +189,103 @@ int sdio_set_block_size(struct sdio_func *func, unsigned blksz)
 
 EXPORT_SYMBOL_GPL(sdio_set_block_size);
 
+/**
+ *	sdio_align_size - pads a transfer size to a more optimal value
+ *	@func: SDIO function
+ *	@sz: original transfer size
+ *
+ *	Pads the original data size with a number of extra bytes in
+ *	order to avoid controller bugs and/or performance hits
+ *	(e.g. some controllers revert to PIO for certain sizes).
+ *
+ *	If possible, it will also adjust the size so that it can be
+ *	handled in just a single request.
+ *
+ *	Returns the improved size, which might be unmodified.
+ */
+unsigned int sdio_align_size(struct sdio_func *func, unsigned int sz)
+{
+	unsigned int orig_sz;
+	unsigned int blk_sz, byte_sz;
+	unsigned chunk_sz;
+
+	orig_sz = sz;
+
+	/*
+	 * Do a first check with the controller, in case it
+	 * wants to increase the size up to a point where it
+	 * might need more than one block.
+	 */
+	sz = mmc_align_data_size(func->card, sz);
+
+	/*
+	 * If we can still do this with just a byte transfer, then
+	 * we're done.
+	 */
+	if ((sz <= func->cur_blksize) && (sz <= 512))
+		return sz;
+
+	if (func->card->cccr.multi_block) {
+		/*
+		 * Check if the transfer is already block aligned
+		 */
+		if ((sz % func->cur_blksize) == 0)
+			return sz;
+
+		/*
+		 * Realign it so that it can be done with one request,
+		 * and recheck if the controller still likes it.
+		 */
+		blk_sz = ((sz + func->cur_blksize - 1) /
+			func->cur_blksize) * func->cur_blksize;
+		blk_sz = mmc_align_data_size(func->card, blk_sz);
+
+		/*
+		 * This value is only good if it is still just
+		 * one request.
+		 */
+		if ((blk_sz % func->cur_blksize) == 0)
+			return blk_sz;
+
+		/*
+		 * We failed to do one request, but at least try to
+		 * pad the remainder properly.
+		 */
+		byte_sz = mmc_align_data_size(func->card,
+				sz % func->cur_blksize);
+		if ((byte_sz <= func->cur_blksize) && (byte_sz <= 512)) {
+			blk_sz = sz / func->cur_blksize;
+			return blk_sz * func->cur_blksize + byte_sz;
+		}
+	} else {
+		/*
+		 * We need multiple requests, so first check that the
+		 * controller can handle the chunk size;
+		 */
+		chunk_sz = mmc_align_data_size(func->card,
+				min(func->cur_blksize, 512u));
+		if (chunk_sz == min(func->cur_blksize, 512u)) {
+			/*
+			 * Fix up the size of the remainder (if any)
+			 */
+			byte_sz = orig_sz % chunk_sz;
+			if (byte_sz) {
+				byte_sz = mmc_align_data_size(func->card,
+						byte_sz);
+			}
+
+			return (orig_sz / chunk_sz) * chunk_sz + byte_sz;
+		}
+	}
+
+	/*
+	 * The controller is simply incapable of transferring the size
+	 * we want in decent manner, so just return the original size.
+	 */
+	return orig_sz;
+}
+EXPORT_SYMBOL_GPL(sdio_align_size);
+
 /* Split an arbitrarily sized data transfer into several
  * IO_RW_EXTENDED commands. */
 static int sdio_io_rw_ext_helper(struct sdio_func *func, int write,
diff --git a/drivers/net/wireless/libertas/if_sdio.c b/drivers/net/wireless/libertas/if_sdio.c
index 3dd537be87d8..b54e2ea8346b 100644
--- a/drivers/net/wireless/libertas/if_sdio.c
+++ b/drivers/net/wireless/libertas/if_sdio.c
@@ -1,7 +1,7 @@
 /*
  *  linux/drivers/net/wireless/libertas/if_sdio.c
  *
- *  Copyright 2007 Pierre Ossman
+ *  Copyright 2007-2008 Pierre Ossman
  *
  * Inspired by if_cs.c, Copyright 2007 Holger Schurig
  *
@@ -266,13 +266,10 @@ static int if_sdio_card_to_host(struct if_sdio_card *card)
 
 	/*
 	 * The transfer must be in one transaction or the firmware
-	 * goes suicidal.
+	 * goes suicidal. There's no way to guarantee that for all
+	 * controllers, but we can at least try.
 	 */
-	chunk = size;
-	if ((chunk > card->func->cur_blksize) || (chunk > 512)) {
-		chunk = (chunk + card->func->cur_blksize - 1) /
-			card->func->cur_blksize * card->func->cur_blksize;
-	}
+	chunk = sdio_align_size(card->func, size);
 
 	ret = sdio_readsb(card->func, card->buffer, card->ioport, chunk);
 	if (ret)
@@ -696,13 +693,10 @@ static int if_sdio_host_to_card(struct lbs_private *priv,
 
 	/*
 	 * The transfer must be in one transaction or the firmware
-	 * goes suicidal.
+	 * goes suicidal. There's no way to guarantee that for all
+	 * controllers, but we can at least try.
 	 */
-	size = nb + 4;
-	if ((size > card->func->cur_blksize) || (size > 512)) {
-		size = (size + card->func->cur_blksize - 1) /
-			card->func->cur_blksize * card->func->cur_blksize;
-	}
+	size = sdio_align_size(card->func, nb + 4);
 
 	packet = kzalloc(sizeof(struct if_sdio_packet) + size,
 			GFP_ATOMIC);
diff --git a/include/linux/mmc/core.h b/include/linux/mmc/core.h
index d0c3abed74c2..143cebf0586f 100644
--- a/include/linux/mmc/core.h
+++ b/include/linux/mmc/core.h
@@ -135,6 +135,7 @@ extern int mmc_wait_for_app_cmd(struct mmc_host *, struct mmc_card *,
 	struct mmc_command *, int);
 
 extern void mmc_set_data_timeout(struct mmc_data *, const struct mmc_card *);
+extern unsigned int mmc_align_data_size(struct mmc_card *, unsigned int);
 
 extern int __mmc_claim_host(struct mmc_host *host, atomic_t *abort);
 extern void mmc_release_host(struct mmc_host *host);
diff --git a/include/linux/mmc/sdio_func.h b/include/linux/mmc/sdio_func.h
index b050f4d7b41f..f57f22b3be88 100644
--- a/include/linux/mmc/sdio_func.h
+++ b/include/linux/mmc/sdio_func.h
@@ -1,7 +1,7 @@
 /*
  *  include/linux/mmc/sdio_func.h
  *
- *  Copyright 2007 Pierre Ossman
+ *  Copyright 2007-2008 Pierre Ossman
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -120,6 +120,8 @@ extern int sdio_set_block_size(struct sdio_func *func, unsigned blksz);
 extern int sdio_claim_irq(struct sdio_func *func, sdio_irq_handler_t *handler);
 extern int sdio_release_irq(struct sdio_func *func);
 
+extern unsigned int sdio_align_size(struct sdio_func *func, unsigned int sz);
+
 extern unsigned char sdio_readb(struct sdio_func *func,
 	unsigned int addr, int *err_ret);
 extern unsigned short sdio_readw(struct sdio_func *func,
-- 
cgit v1.2.3


From 6d37333163025b46afbcad434ec9a5f2e88e7254 Mon Sep 17 00:00:00 2001
From: Tomas Winkler <tomas.winkler@intel.com>
Date: Mon, 30 Jun 2008 10:50:24 +0300
Subject: mmc: fix sdio_io sparse errors

This patch fixes sdio_io sparse errors.
This fix changes signature of API functions,
changing
unsigned char -> u8
unsigned short -> u16
unsigned long -> u32 - this was probably a bug in 64 bit platforms

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/sdio_io.c    | 42 ++++++++++++++++--------------------------
 include/linux/mmc/sdio_func.h | 15 ++++++---------
 2 files changed, 22 insertions(+), 35 deletions(-)
 mode change 100644 => 100755 drivers/mmc/core/sdio_io.c
 mode change 100644 => 100755 include/linux/mmc/sdio_func.h

(limited to 'include/linux')

diff --git a/drivers/mmc/core/sdio_io.c b/drivers/mmc/core/sdio_io.c
old mode 100644
new mode 100755
index cc42a41ff6ab..3ccf6919877c
--- a/drivers/mmc/core/sdio_io.c
+++ b/drivers/mmc/core/sdio_io.c
@@ -167,10 +167,8 @@ int sdio_set_block_size(struct sdio_func *func, unsigned blksz)
 		return -EINVAL;
 
 	if (blksz == 0) {
-		blksz = min(min(
-			func->max_blksize,
-			func->card->host->max_blk_size),
-			512u);
+		blksz = min(func->max_blksize, func->card->host->max_blk_size);
+		blksz = min(blksz, 512u);
 	}
 
 	ret = mmc_io_rw_direct(func->card, 1, 0,
@@ -311,10 +309,9 @@ static int sdio_io_rw_ext_helper(struct sdio_func *func, int write,
 		/* Blocks per command is limited by host count, host transfer
 		 * size (we only use a single sg entry) and the maximum for
 		 * IO_RW_EXTENDED of 511 blocks. */
-		max_blocks = min(min(
-			func->card->host->max_blk_count,
-			func->card->host->max_seg_size / func->cur_blksize),
-			511u);
+		max_blocks = min(func->card->host->max_blk_count,
+			func->card->host->max_seg_size / func->cur_blksize);
+		max_blocks = min(max_blocks, 511u);
 
 		while (remainder > func->cur_blksize) {
 			unsigned blocks;
@@ -364,11 +361,10 @@ static int sdio_io_rw_ext_helper(struct sdio_func *func, int write,
  *	function. If there is a problem reading the address, 0xff
  *	is returned and @err_ret will contain the error code.
  */
-unsigned char sdio_readb(struct sdio_func *func, unsigned int addr,
-	int *err_ret)
+u8 sdio_readb(struct sdio_func *func, unsigned int addr, int *err_ret)
 {
 	int ret;
-	unsigned char val;
+	u8 val;
 
 	BUG_ON(!func);
 
@@ -397,8 +393,7 @@ EXPORT_SYMBOL_GPL(sdio_readb);
  *	function. @err_ret will contain the status of the actual
  *	transfer.
  */
-void sdio_writeb(struct sdio_func *func, unsigned char b, unsigned int addr,
-	int *err_ret)
+void sdio_writeb(struct sdio_func *func, u8 b, unsigned int addr, int *err_ret)
 {
 	int ret;
 
@@ -459,7 +454,6 @@ int sdio_readsb(struct sdio_func *func, void *dst, unsigned int addr,
 {
 	return sdio_io_rw_ext_helper(func, 0, addr, 0, dst, count);
 }
-
 EXPORT_SYMBOL_GPL(sdio_readsb);
 
 /**
@@ -489,8 +483,7 @@ EXPORT_SYMBOL_GPL(sdio_writesb);
  *	function. If there is a problem reading the address, 0xffff
  *	is returned and @err_ret will contain the error code.
  */
-unsigned short sdio_readw(struct sdio_func *func, unsigned int addr,
-	int *err_ret)
+u16 sdio_readw(struct sdio_func *func, unsigned int addr, int *err_ret)
 {
 	int ret;
 
@@ -504,7 +497,7 @@ unsigned short sdio_readw(struct sdio_func *func, unsigned int addr,
 		return 0xFFFF;
 	}
 
-	return le16_to_cpu(*(u16*)func->tmpbuf);
+	return le16_to_cpup((__le16 *)func->tmpbuf);
 }
 EXPORT_SYMBOL_GPL(sdio_readw);
 
@@ -519,12 +512,11 @@ EXPORT_SYMBOL_GPL(sdio_readw);
  *	function. @err_ret will contain the status of the actual
  *	transfer.
  */
-void sdio_writew(struct sdio_func *func, unsigned short b, unsigned int addr,
-	int *err_ret)
+void sdio_writew(struct sdio_func *func, u16 b, unsigned int addr, int *err_ret)
 {
 	int ret;
 
-	*(u16*)func->tmpbuf = cpu_to_le16(b);
+	*(__le16 *)func->tmpbuf = cpu_to_le16(b);
 
 	ret = sdio_memcpy_toio(func, addr, func->tmpbuf, 2);
 	if (err_ret)
@@ -543,8 +535,7 @@ EXPORT_SYMBOL_GPL(sdio_writew);
  *	0xffffffff is returned and @err_ret will contain the error
  *	code.
  */
-unsigned long sdio_readl(struct sdio_func *func, unsigned int addr,
-	int *err_ret)
+u32 sdio_readl(struct sdio_func *func, unsigned int addr, int *err_ret)
 {
 	int ret;
 
@@ -558,7 +549,7 @@ unsigned long sdio_readl(struct sdio_func *func, unsigned int addr,
 		return 0xFFFFFFFF;
 	}
 
-	return le32_to_cpu(*(u32*)func->tmpbuf);
+	return le32_to_cpup((__le32 *)func->tmpbuf);
 }
 EXPORT_SYMBOL_GPL(sdio_readl);
 
@@ -573,12 +564,11 @@ EXPORT_SYMBOL_GPL(sdio_readl);
  *	function. @err_ret will contain the status of the actual
  *	transfer.
  */
-void sdio_writel(struct sdio_func *func, unsigned long b, unsigned int addr,
-	int *err_ret)
+void sdio_writel(struct sdio_func *func, u32 b, unsigned int addr, int *err_ret)
 {
 	int ret;
 
-	*(u32*)func->tmpbuf = cpu_to_le32(b);
+	*(__le32 *)func->tmpbuf = cpu_to_le32(b);
 
 	ret = sdio_memcpy_toio(func, addr, func->tmpbuf, 4);
 	if (err_ret)
diff --git a/include/linux/mmc/sdio_func.h b/include/linux/mmc/sdio_func.h
old mode 100644
new mode 100755
index f57f22b3be88..28fb0a33acf8
--- a/include/linux/mmc/sdio_func.h
+++ b/include/linux/mmc/sdio_func.h
@@ -122,23 +122,20 @@ extern int sdio_release_irq(struct sdio_func *func);
 
 extern unsigned int sdio_align_size(struct sdio_func *func, unsigned int sz);
 
-extern unsigned char sdio_readb(struct sdio_func *func,
-	unsigned int addr, int *err_ret);
-extern unsigned short sdio_readw(struct sdio_func *func,
-	unsigned int addr, int *err_ret);
-extern unsigned long sdio_readl(struct sdio_func *func,
-	unsigned int addr, int *err_ret);
+extern u8 sdio_readb(struct sdio_func *func, unsigned int addr, int *err_ret);
+extern u16 sdio_readw(struct sdio_func *func, unsigned int addr, int *err_ret);
+extern u32 sdio_readl(struct sdio_func *func, unsigned int addr, int *err_ret);
 
 extern int sdio_memcpy_fromio(struct sdio_func *func, void *dst,
 	unsigned int addr, int count);
 extern int sdio_readsb(struct sdio_func *func, void *dst,
 	unsigned int addr, int count);
 
-extern void sdio_writeb(struct sdio_func *func, unsigned char b,
+extern void sdio_writeb(struct sdio_func *func, u8 b,
 	unsigned int addr, int *err_ret);
-extern void sdio_writew(struct sdio_func *func, unsigned short b,
+extern void sdio_writew(struct sdio_func *func, u16 b,
 	unsigned int addr, int *err_ret);
-extern void sdio_writel(struct sdio_func *func, unsigned long b,
+extern void sdio_writel(struct sdio_func *func, u32 b,
 	unsigned int addr, int *err_ret);
 
 extern int sdio_memcpy_toio(struct sdio_func *func, unsigned int addr,
-- 
cgit v1.2.3


From 23af60398af2f5033e2f53665538a09f498dbc03 Mon Sep 17 00:00:00 2001
From: Pierre Ossman <drzeus@drzeus.cx>
Date: Sun, 6 Jul 2008 01:10:27 +0200
Subject: mmc: remove multiwrite capability

Relax requirements on host controllers and only require that they do not
report a transfer count than is larger than the actual one (i.e. a lower
value is okay). This is how many other parts of the kernel behaves so
upper layers should already be prepared to handle that scenario. This
gives us a performance boost on MMC cards.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/card/block.c     | 47 ++++++++++++++++++--------------------------
 drivers/mmc/host/at91_mci.c  |  2 +-
 drivers/mmc/host/atmel-mci.c |  2 +-
 drivers/mmc/host/mmc_spi.c   |  5 +----
 drivers/mmc/host/mmci.c      |  1 -
 drivers/mmc/host/omap.c      |  2 +-
 drivers/mmc/host/tifm_sd.c   |  2 +-
 drivers/mmc/host/wbsd.c      |  2 +-
 include/linux/mmc/host.h     | 11 +++++------
 9 files changed, 30 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index 4b0f8220f153..66e5a5487c20 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -237,17 +237,6 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 		if (brq.data.blocks > card->host->max_blk_count)
 			brq.data.blocks = card->host->max_blk_count;
 
-		/*
-		 * If the host doesn't support multiple block writes, force
-		 * block writes to single block. SD cards are excepted from
-		 * this rule as they support querying the number of
-		 * successfully written sectors.
-		 */
-		if (rq_data_dir(req) != READ &&
-		    !(card->host->caps & MMC_CAP_MULTIWRITE) &&
-		    !mmc_card_sd(card))
-			brq.data.blocks = 1;
-
 		if (brq.data.blocks > 1) {
 			/* SPI multiblock writes terminate using a special
 			 * token, not a STOP_TRANSMISSION request.
@@ -367,30 +356,32 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
  	 * mark the known good sectors as ok.
  	 *
 	 * If the card is not SD, we can still ok written sectors
-	 * if the controller can do proper error reporting.
+	 * as reported by the controller (which might be less than
+	 * the real number of written sectors, but never more).
 	 *
 	 * For reads we just fail the entire chunk as that should
 	 * be safe in all cases.
 	 */
- 	if (rq_data_dir(req) != READ && mmc_card_sd(card)) {
-		u32 blocks;
-		unsigned int bytes;
-
-		blocks = mmc_sd_num_wr_blocks(card);
-		if (blocks != (u32)-1) {
-			if (card->csd.write_partial)
-				bytes = blocks << md->block_bits;
-			else
-				bytes = blocks << 9;
+	if (rq_data_dir(req) != READ) {
+		if (mmc_card_sd(card)) {
+			u32 blocks;
+			unsigned int bytes;
+
+			blocks = mmc_sd_num_wr_blocks(card);
+			if (blocks != (u32)-1) {
+				if (card->csd.write_partial)
+					bytes = blocks << md->block_bits;
+				else
+					bytes = blocks << 9;
+				spin_lock_irq(&md->lock);
+				ret = __blk_end_request(req, 0, bytes);
+				spin_unlock_irq(&md->lock);
+			}
+		} else {
 			spin_lock_irq(&md->lock);
-			ret = __blk_end_request(req, 0, bytes);
+			ret = __blk_end_request(req, 0, brq.data.bytes_xfered);
 			spin_unlock_irq(&md->lock);
 		}
-	} else if (rq_data_dir(req) != READ &&
-		   (card->host->caps & MMC_CAP_MULTIWRITE)) {
-		spin_lock_irq(&md->lock);
-		ret = __blk_end_request(req, 0, brq.data.bytes_xfered);
-		spin_unlock_irq(&md->lock);
 	}
 
 	mmc_release_host(card->host);
diff --git a/drivers/mmc/host/at91_mci.c b/drivers/mmc/host/at91_mci.c
index 16df235fcc2e..f15e2064305c 100644
--- a/drivers/mmc/host/at91_mci.c
+++ b/drivers/mmc/host/at91_mci.c
@@ -995,7 +995,7 @@ static int __init at91_mci_probe(struct platform_device *pdev)
 	mmc->f_min = 375000;
 	mmc->f_max = 25000000;
 	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-	mmc->caps = MMC_CAP_MULTIWRITE | MMC_CAP_SDIO_IRQ;
+	mmc->caps = MMC_CAP_SDIO_IRQ;
 
 	mmc->max_blk_size = 4095;
 	mmc->max_blk_count = mmc->max_req_size;
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 25d5324ab7e2..cce873c5a149 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -849,7 +849,7 @@ static int __init atmci_probe(struct platform_device *pdev)
 	mmc->f_min = (host->bus_hz + 511) / 512;
 	mmc->f_max = host->bus_hz / 2;
 	mmc->ocr_avail	= MMC_VDD_32_33 | MMC_VDD_33_34;
-	mmc->caps |= MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE;
+	mmc->caps |= MMC_CAP_4_BIT_DATA;
 
 	mmc->max_hw_segs = 64;
 	mmc->max_phys_segs = 64;
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 4e82f64a96bc..41cc63360e43 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -1252,10 +1252,7 @@ static int mmc_spi_probe(struct spi_device *spi)
 	mmc->ops = &mmc_spi_ops;
 	mmc->max_blk_size = MMC_SPI_BLOCKSIZE;
 
-	/* As long as we keep track of the number of successfully
-	 * transmitted blocks, we're good for multiwrite.
-	 */
-	mmc->caps = MMC_CAP_SPI | MMC_CAP_MULTIWRITE;
+	mmc->caps = MMC_CAP_SPI;
 
 	/* SPI doesn't need the lowspeed device identification thing for
 	 * MMC or SD cards, since it never comes up in open drain mode.
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index da5fecad74d9..696cf3647ceb 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -535,7 +535,6 @@ static int mmci_probe(struct amba_device *dev, void *id)
 	mmc->f_min = (host->mclk + 511) / 512;
 	mmc->f_max = min(host->mclk, fmax);
 	mmc->ocr_avail = plat->ocr_mask;
-	mmc->caps = MMC_CAP_MULTIWRITE;
 
 	/*
 	 * We can do SGIO
diff --git a/drivers/mmc/host/omap.c b/drivers/mmc/host/omap.c
index 549517c35675..dbc26eb6a89e 100644
--- a/drivers/mmc/host/omap.c
+++ b/drivers/mmc/host/omap.c
@@ -1317,7 +1317,7 @@ static int __init mmc_omap_new_slot(struct mmc_omap_host *host, int id)
 
 	host->slots[id] = slot;
 
-	mmc->caps = MMC_CAP_MULTIWRITE;
+	mmc->caps = 0;
 	if (host->pdata->conf.wire4)
 		mmc->caps |= MMC_CAP_4_BIT_DATA;
 
diff --git a/drivers/mmc/host/tifm_sd.c b/drivers/mmc/host/tifm_sd.c
index 1c14a186f000..13844843e8de 100644
--- a/drivers/mmc/host/tifm_sd.c
+++ b/drivers/mmc/host/tifm_sd.c
@@ -973,7 +973,7 @@ static int tifm_sd_probe(struct tifm_dev *sock)
 
 	mmc->ops = &tifm_sd_ops;
 	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE;
+	mmc->caps = MMC_CAP_4_BIT_DATA;
 	mmc->f_min = 20000000 / 60;
 	mmc->f_max = 24000000;
 
diff --git a/drivers/mmc/host/wbsd.c b/drivers/mmc/host/wbsd.c
index f7dcd8ec0d7f..adda37952032 100644
--- a/drivers/mmc/host/wbsd.c
+++ b/drivers/mmc/host/wbsd.c
@@ -1219,7 +1219,7 @@ static int __devinit wbsd_alloc_mmc(struct device *dev)
 	mmc->f_min = 375000;
 	mmc->f_max = 24000000;
 	mmc->ocr_avail = MMC_VDD_32_33 | MMC_VDD_33_34;
-	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_MULTIWRITE;
+	mmc->caps = MMC_CAP_4_BIT_DATA;
 
 	spin_lock_init(&host->lock);
 
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 753b7231b887..10a2080086ca 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -111,12 +111,11 @@ struct mmc_host {
 	unsigned long		caps;		/* Host capabilities */
 
 #define MMC_CAP_4_BIT_DATA	(1 << 0)	/* Can the host do 4 bit transfers */
-#define MMC_CAP_MULTIWRITE	(1 << 1)	/* Can accurately report bytes sent to card on error */
-#define MMC_CAP_MMC_HIGHSPEED	(1 << 2)	/* Can do MMC high-speed timing */
-#define MMC_CAP_SD_HIGHSPEED	(1 << 3)	/* Can do SD high-speed timing */
-#define MMC_CAP_SDIO_IRQ	(1 << 4)	/* Can signal pending SDIO IRQs */
-#define MMC_CAP_SPI		(1 << 5)	/* Talks only SPI protocols */
-#define MMC_CAP_NEEDS_POLL	(1 << 6)	/* Needs polling for card-detection */
+#define MMC_CAP_MMC_HIGHSPEED	(1 << 1)	/* Can do MMC high-speed timing */
+#define MMC_CAP_SD_HIGHSPEED	(1 << 2)	/* Can do SD high-speed timing */
+#define MMC_CAP_SDIO_IRQ	(1 << 3)	/* Can signal pending SDIO IRQs */
+#define MMC_CAP_SPI		(1 << 4)	/* Talks only SPI protocols */
+#define MMC_CAP_NEEDS_POLL	(1 << 5)	/* Needs polling for card-detection */
 
 	/* host specific block data */
 	unsigned int		max_seg_size;	/* see blk_queue_max_segment_size */
-- 
cgit v1.2.3


From 62a7573ee9f31d4fdb330b3e68ebf6efaba1d57c Mon Sep 17 00:00:00 2001
From: Benzi Zbit <benzi.zbit@intel.com>
Date: Thu, 10 Jul 2008 02:41:43 +0300
Subject: sdio: fix the use of hard coded timeout value.

This adds reading and using of enable_timeout from the CIS

Signed-off-by: Benzi Zbit <benzi.zbit@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/sdio_cis.c   | 6 ++++++
 drivers/mmc/core/sdio_io.c    | 6 +-----
 include/linux/mmc/sdio_func.h | 2 ++
 3 files changed, 9 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/core/sdio_cis.c b/drivers/mmc/core/sdio_cis.c
index d5e51b1c7b3f..956bd7677502 100644
--- a/drivers/mmc/core/sdio_cis.c
+++ b/drivers/mmc/core/sdio_cis.c
@@ -129,6 +129,12 @@ static int cistpl_funce_func(struct sdio_func *func,
 	/* TPLFE_MAX_BLK_SIZE */
 	func->max_blksize = buf[12] | (buf[13] << 8);
 
+	/* TPLFE_ENABLE_TIMEOUT_VAL, present in ver 1.1 and above */
+	if (vsn > SDIO_SDIO_REV_1_00)
+		func->enable_timeout = (buf[28] | (buf[29] << 8)) * 10;
+	else
+		func->enable_timeout = jiffies_to_msecs(HZ);
+
 	return 0;
 }
 
diff --git a/drivers/mmc/core/sdio_io.c b/drivers/mmc/core/sdio_io.c
index 3ccf6919877c..0888df64581f 100755
--- a/drivers/mmc/core/sdio_io.c
+++ b/drivers/mmc/core/sdio_io.c
@@ -76,11 +76,7 @@ int sdio_enable_func(struct sdio_func *func)
 	if (ret)
 		goto err;
 
-	/*
-	 * FIXME: This should timeout based on information in the CIS,
-	 * but we don't have card to parse that yet.
-	 */
-	timeout = jiffies + HZ;
+	timeout = jiffies + msecs_to_jiffies(func->enable_timeout);
 
 	while (1) {
 		ret = mmc_io_rw_direct(func->card, 0, 0, SDIO_CCCR_IORx, 0, &reg);
diff --git a/include/linux/mmc/sdio_func.h b/include/linux/mmc/sdio_func.h
index 28fb0a33acf8..07bee4a0d457 100755
--- a/include/linux/mmc/sdio_func.h
+++ b/include/linux/mmc/sdio_func.h
@@ -46,6 +46,8 @@ struct sdio_func {
 	unsigned		max_blksize;	/* maximum block size */
 	unsigned		cur_blksize;	/* current block size */
 
+	unsigned		enable_timeout;	/* max enable timeout in msec */
+
 	unsigned int		state;		/* function state */
 #define SDIO_STATE_PRESENT	(1<<0)		/* present in sysfs */
 
-- 
cgit v1.2.3


From 124cafc5eb973e748c4ce3dc1caad29274e64613 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Tue, 15 Jul 2008 21:21:44 +0200
Subject: ide: remove ide_init_drive_cmd

ide_init_drive_cmd just calls blk_rq_init. This converts the users of
ide_init_drive_cmd to use blk_rq_init directly and removes
ide_init_drive_cmd.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c     |  2 +-
 drivers/ide/ide-floppy.c |  2 +-
 drivers/ide/ide-io.c     | 17 -----------------
 drivers/scsi/ide-scsi.c  |  4 ++--
 include/linux/ide.h      |  2 --
 5 files changed, 4 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 792a3cf73d6e..7917cd576446 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -193,7 +193,7 @@ void ide_cd_init_rq(ide_drive_t *drive, struct request *rq)
 {
 	struct cdrom_info *cd = drive->driver_data;
 
-	ide_init_drive_cmd(rq);
+	blk_rq_init(NULL, rq);
 	rq->cmd_type = REQ_TYPE_ATA_PC;
 	rq->rq_disk = cd->disk;
 }
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index b10e9a813cde..9161cd92a842 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -286,7 +286,7 @@ static void idefloppy_queue_pc_head(ide_drive_t *drive, struct ide_atapi_pc *pc,
 {
 	struct ide_floppy_obj *floppy = drive->driver_data;
 
-	ide_init_drive_cmd(rq);
+	blk_rq_init(NULL, rq);
 	rq->buffer = (char *) pc;
 	rq->cmd_type = REQ_TYPE_SPECIAL;
 	rq->cmd_flags |= REQ_PREEMPT;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 29f5cc863f6e..d8b4d9f81ae2 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1538,23 +1538,6 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-/**
- *	ide_init_drive_cmd	-	initialize a drive command request
- *	@rq: request object
- *
- *	Initialize a request before we fill it in and send it down to
- *	ide_do_drive_cmd. Commands must be set up by this function. Right
- *	now it doesn't do a lot, but if that changes abusers will have a
- *	nasty surprise.
- */
-
-void ide_init_drive_cmd (struct request *rq)
-{
-	blk_rq_init(NULL, rq);
-}
-
-EXPORT_SYMBOL(ide_init_drive_cmd);
-
 /**
  *	ide_do_drive_cmd	-	issue IDE special command
  *	@drive: device to issue command
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 89ecf0132191..da261806d62a 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -228,7 +228,7 @@ static int idescsi_check_condition(ide_drive_t *drive,
 		kfree(pc);
 		return -ENOMEM;
 	}
-	ide_init_drive_cmd(rq);
+	blk_rq_init(NULL, rq);
 	rq->special = (char *) pc;
 	pc->rq = rq;
 	pc->buf = buf;
@@ -786,7 +786,7 @@ static int idescsi_queue (struct scsi_cmnd *cmd,
 		}
 	}
 
-	ide_init_drive_cmd (rq);
+	blk_rq_init(NULL, rq);
 	rq->special = (char *) pc;
 	rq->cmd_type = REQ_TYPE_SPECIAL;
 	spin_unlock_irq(host->host_lock);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index eddb6daadf4a..3261c6691759 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -857,8 +857,6 @@ int ide_wait_stat(ide_startstop_t *, ide_drive_t *, u8, u8, unsigned long);
 
 extern ide_startstop_t ide_do_reset (ide_drive_t *);
 
-extern void ide_init_drive_cmd (struct request *rq);
-
 /*
  * "action" parameter type for ide_do_drive_cmd() below.
  */
-- 
cgit v1.2.3


From 681a561b7ec7fdcd8f35b68e44ac6d6c70aecc04 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Tue, 15 Jul 2008 21:21:45 +0200
Subject: block: unexport blk_end_sync_rq

All the users of blk_end_sync_rq has gone (they are converted to use
blk_execute_rq). This unexports blk_end_sync_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 block/blk-exec.c       | 3 +--
 include/linux/blkdev.h | 1 -
 2 files changed, 1 insertion(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-exec.c b/block/blk-exec.c
index 4f52f2792059..9bceff7674f2 100644
--- a/block/blk-exec.c
+++ b/block/blk-exec.c
@@ -18,7 +18,7 @@
  * @rq: request to complete
  * @error: end io status of the request
  */
-void blk_end_sync_rq(struct request *rq, int error)
+static void blk_end_sync_rq(struct request *rq, int error)
 {
 	struct completion *waiting = rq->end_io_data;
 
@@ -31,7 +31,6 @@ void blk_end_sync_rq(struct request *rq, int error)
 	 */
 	complete(waiting);
 }
-EXPORT_SYMBOL(blk_end_sync_rq);
 
 /**
  * blk_execute_rq_nowait - insert a request into queue for execution
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d2a1b71e93c3..1171abd7eb17 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -623,7 +623,6 @@ extern void generic_make_request(struct bio *bio);
 extern void blk_rq_init(struct request_queue *q, struct request *rq);
 extern void blk_put_request(struct request *);
 extern void __blk_put_request(struct request_queue *, struct request *);
-extern void blk_end_sync_rq(struct request *rq, int error);
 extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
 extern void blk_insert_request(struct request_queue *, struct request *, int, void *);
 extern void blk_requeue_request(struct request_queue *, struct request *);
-- 
cgit v1.2.3


From 30e5ee4d1a651a0c66e86c6612c003034bd20ba2 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:46 +0200
Subject: ide: remove obsoleted "idebus=" kernel parameter

* Remove obsoleted "idebus=" kernel parameter.

* Remove no longer needed ide_system_bus_speed() and system_bus_clock()
  (together with idebus_parameter and system_bus_speed variables).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide.c            | 71 +-------------------------------------------
 drivers/ide/legacy/ali14xx.c |  2 +-
 drivers/ide/legacy/ht6560b.c |  2 +-
 drivers/ide/legacy/qd65xx.c  |  4 +--
 drivers/ide/pci/aec62xx.c    |  2 +-
 drivers/ide/pci/alim15x3.c   |  2 +-
 drivers/ide/pci/amd74xx.c    |  2 +-
 drivers/ide/pci/cmd640.c     |  8 ++---
 drivers/ide/pci/cmd64x.c     |  4 +--
 drivers/ide/pci/cy82c693.c   |  2 +-
 drivers/ide/pci/via82cxxx.c  |  2 +-
 include/linux/ide.h          |  2 --
 12 files changed, 15 insertions(+), 88 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 8823df1b8716..f65be738b16a 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -86,9 +86,6 @@ static const u8 ide_hwif_to_major[] = { IDE0_MAJOR, IDE1_MAJOR,
 					IDE6_MAJOR, IDE7_MAJOR,
 					IDE8_MAJOR, IDE9_MAJOR };
 
-static int idebus_parameter;	/* holds the "idebus=" parameter */
-static int system_bus_speed;	/* holds what we think is VESA/PCI bus speed */
-
 DEFINE_MUTEX(ide_cfg_mtx);
  __cacheline_aligned_in_smp DEFINE_SPINLOCK(ide_lock);
 
@@ -189,38 +186,6 @@ static void __init init_ide_data (void)
 	}
 }
 
-/**
- *	ide_system_bus_speed	-	guess bus speed
- *
- *	ide_system_bus_speed() returns what we think is the system VESA/PCI
- *	bus speed (in MHz). This is used for calculating interface PIO timings.
- *	The default is 40 for known PCI systems, 50 otherwise.
- *	The "idebus=xx" parameter can be used to override this value.
- *	The actual value to be used is computed/displayed the first time
- *	through. Drivers should only use this as a last resort.
- *
- *	Returns a guessed speed in MHz.
- */
-
-static int ide_system_bus_speed(void)
-{
-#ifdef CONFIG_PCI
-	static struct pci_device_id pci_default[] = {
-		{ PCI_DEVICE(PCI_ANY_ID, PCI_ANY_ID) },
-		{ }
-	};
-#else
-#define pci_default 0
-#endif /* CONFIG_PCI */
-
-	/* user supplied value */
-	if (idebus_parameter)
-		return idebus_parameter;
-
-	/* safe default value for PCI or VESA and PCI*/
-	return pci_dev_present(pci_default) ? 33 : 50;
-}
-
 void ide_remove_port_from_hwgroup(ide_hwif_t *hwif)
 {
 	ide_hwgroup_t *hwgroup = hwif->hwgroup;
@@ -540,20 +505,6 @@ static int set_unmaskirq(ide_drive_t *drive, int arg)
 	return 0;
 }
 
-/**
- *	system_bus_clock	-	clock guess
- *
- *	External version of the bus clock guess used by very old IDE drivers
- *	for things like VLB timings. Should not be used.
- */
-
-int system_bus_clock (void)
-{
-	return system_bus_speed;
-}
-
-EXPORT_SYMBOL(system_bus_clock);
-
 static int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 {
 	ide_drive_t *drive = dev->driver_data;
@@ -851,7 +802,7 @@ static int __init ide_setup(char *s)
 	if (strncmp(s,"hd",2) == 0 && s[2] == '=')	/* hd= is for hd.c   */
 		return 0;				/* driver and not us */
 
-	if (strncmp(s,"ide",3) && strncmp(s,"idebus",6) && strncmp(s,"hd",2))
+	if (strncmp(s, "ide", 3) && strncmp(s, "hd", 2))
 		return 0;
 
 	printk(KERN_INFO "ide_setup: %s", s);
@@ -951,21 +902,6 @@ static int __init ide_setup(char *s)
 		}
 	}
 
-	if (s[0] != 'i' || s[1] != 'd' || s[2] != 'e')
-		goto bad_option;
-	/*
-	 * Look for bus speed option:  "idebus="
-	 */
-	if (s[3] == 'b' && s[4] == 'u' && s[5] == 's') {
-		if (match_parm(&s[6], NULL, vals, 1) != 1)
-			goto bad_option;
-		if (vals[0] >= 20 && vals[0] <= 66) {
-			idebus_parameter = vals[0];
-		} else
-			printk(" -- BAD BUS SPEED! Expected value from 20 to 66");
-		goto obsolete_option;
-	}
-
 bad_option:
 	printk(" -- BAD OPTION\n");
 	return 1;
@@ -1287,11 +1223,6 @@ static int __init ide_init(void)
 	int ret;
 
 	printk(KERN_INFO "Uniform Multi-Platform E-IDE driver\n");
-	system_bus_speed = ide_system_bus_speed();
-
-	printk(KERN_INFO "ide: Assuming %dMHz system bus speed "
-			 "for PIO modes%s\n", system_bus_speed,
-			idebus_parameter ? "" : "; override with idebus=xx");
 
 	ret = bus_register(&ide_bus_type);
 	if (ret < 0) {
diff --git a/drivers/ide/legacy/ali14xx.c b/drivers/ide/legacy/ali14xx.c
index 90c65cf97448..052125fafcfa 100644
--- a/drivers/ide/legacy/ali14xx.c
+++ b/drivers/ide/legacy/ali14xx.c
@@ -116,7 +116,7 @@ static void ali14xx_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	int time1, time2;
 	u8 param1, param2, param3, param4;
 	unsigned long flags;
-	int bus_speed = ide_vlb_clk ? ide_vlb_clk : system_bus_clock();
+	int bus_speed = ide_vlb_clk ? ide_vlb_clk : 50;
 
 	/* calculate timing, according to PIO mode */
 	time1 = ide_pio_cycle_time(drive, pio);
diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c
index 4fe516df9f74..dd6dfb32e853 100644
--- a/drivers/ide/legacy/ht6560b.c
+++ b/drivers/ide/legacy/ht6560b.c
@@ -212,7 +212,7 @@ static u8 ht_pio2timings(ide_drive_t *drive, const u8 pio)
 {
 	int active_time, recovery_time;
 	int active_cycles, recovery_cycles;
-	int bus_speed = ide_vlb_clk ? ide_vlb_clk : system_bus_clock();
+	int bus_speed = ide_vlb_clk ? ide_vlb_clk : 50;
 
         if (pio) {
 		unsigned int cycle_time;
diff --git a/drivers/ide/legacy/qd65xx.c b/drivers/ide/legacy/qd65xx.c
index 6424af154325..51dba82f8812 100644
--- a/drivers/ide/legacy/qd65xx.c
+++ b/drivers/ide/legacy/qd65xx.c
@@ -110,7 +110,7 @@ static void qd65xx_select(ide_drive_t *drive)
 
 static u8 qd6500_compute_timing (ide_hwif_t *hwif, int active_time, int recovery_time)
 {
-	int clk = ide_vlb_clk ? ide_vlb_clk : system_bus_clock();
+	int clk = ide_vlb_clk ? ide_vlb_clk : 50;
 	u8 act_cyc, rec_cyc;
 
 	if (clk <= 33) {
@@ -132,7 +132,7 @@ static u8 qd6500_compute_timing (ide_hwif_t *hwif, int active_time, int recovery
 
 static u8 qd6580_compute_timing (int active_time, int recovery_time)
 {
-	int clk = ide_vlb_clk ? ide_vlb_clk : system_bus_clock();
+	int clk = ide_vlb_clk ? ide_vlb_clk : 50;
 	u8 act_cyc, rec_cyc;
 
 	act_cyc = 17 - IDE_IN(active_time   * clk / 1000 + 1, 2, 17);
diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index 7f46c224b7c4..ae7a4329a581 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -140,7 +140,7 @@ static void aec_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static unsigned int __devinit init_chipset_aec62xx(struct pci_dev *dev, const char *name)
 {
-	int bus_speed = ide_pci_clk ? ide_pci_clk : system_bus_clock();
+	int bus_speed = ide_pci_clk ? ide_pci_clk : 33;
 
 	if (bus_speed <= 33)
 		pci_set_drvdata(dev, (void *) aec6xxx_33_base);
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index f2129d5e07f2..f2de00adf147 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -72,7 +72,7 @@ static void ali_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	int s_time, a_time, c_time;
 	u8 s_clc, a_clc, r_clc;
 	unsigned long flags;
-	int bus_speed = ide_pci_clk ? ide_pci_clk : system_bus_clock();
+	int bus_speed = ide_pci_clk ? ide_pci_clk : 33;
 	int port = hwif->channel ? 0x5c : 0x58;
 	int portFIFO = hwif->channel ? 0x55 : 0x54;
 	u8 cd_dma_fifo = 0;
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index a373101747b6..ad222206a429 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -179,7 +179,7 @@ static unsigned int __devinit init_chipset_amd74xx(struct pci_dev *dev,
  * Determine the system bus clock.
  */
 
-	amd_clock = (ide_pci_clk ? ide_pci_clk : system_bus_clock()) * 1000;
+	amd_clock = (ide_pci_clk ? ide_pci_clk : 33) * 1000;
 
 	switch (amd_clock) {
 		case 33000: amd_clock = 33333; break;
diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c
index b38a1980dcd5..cd1ba14984ab 100644
--- a/drivers/ide/pci/cmd640.c
+++ b/drivers/ide/pci/cmd640.c
@@ -525,12 +525,10 @@ static void cmd640_set_mode(ide_drive_t *drive, unsigned int index,
 	u8 setup_count, active_count, recovery_count, recovery_count2, cycle_count;
 	int bus_speed;
 
-	if (cmd640_vlb && ide_vlb_clk)
-		bus_speed = ide_vlb_clk;
-	else if (!cmd640_vlb && ide_pci_clk)
-		bus_speed = ide_pci_clk;
+	if (cmd640_vlb)
+		bus_speed = ide_vlb_clk ? ide_vlb_clk : 50;
 	else
-		bus_speed = system_bus_clock();
+		bus_speed = ide_pci_clk ? ide_pci_clk : 33;
 
 	if (pio_mode > 5)
 		pio_mode = 5;
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index 08674711d089..ca4774aa27ee 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -69,7 +69,7 @@ static u8 quantize_timing(int timing, int quant)
 static void program_cycle_times (ide_drive_t *drive, int cycle_time, int active_time)
 {
 	struct pci_dev *dev = to_pci_dev(drive->hwif->dev);
-	int clock_time = 1000 / (ide_pci_clk ? ide_pci_clk : system_bus_clock());
+	int clock_time = 1000 / (ide_pci_clk ? ide_pci_clk : 33);
 	u8  cycle_count, active_count, recovery_count, drwtim;
 	static const u8 recovery_values[] =
 		{15, 15, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 0};
@@ -128,7 +128,7 @@ static void cmd64x_tune_pio(ide_drive_t *drive, const u8 pio)
 			    ide_pio_timings[pio].active_time);
 
 	setup_count = quantize_timing(ide_pio_timings[pio].setup_time,
-			1000 / (ide_pci_clk ? ide_pci_clk : system_bus_clock()));
+			1000 / (ide_pci_clk ? ide_pci_clk : 33));
 
 	/*
 	 * The primary channel has individual address setup timing registers
diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c
index 77cc22c2ad45..8c534afcb6c8 100644
--- a/drivers/ide/pci/cy82c693.c
+++ b/drivers/ide/pci/cy82c693.c
@@ -134,7 +134,7 @@ static int calc_clk(int time, int bus_speed)
 static void compute_clocks(u8 pio, pio_clocks_t *p_pclk)
 {
 	int clk1, clk2;
-	int bus_speed = ide_pci_clk ? ide_pci_clk : system_bus_clock();
+	int bus_speed = ide_pci_clk ? ide_pci_clk : 33;
 
 	/* we don't check against CY82C693's min and max speed,
 	 * so you can play with the idebus=xx parameter
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index e8c2570003ff..3ed9728abd24 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -340,7 +340,7 @@ static unsigned int __devinit init_chipset_via82cxxx(struct pci_dev *dev, const
 	 * Determine system bus clock.
 	 */
 
-	via_clock = (ide_pci_clk ? ide_pci_clk : system_bus_clock()) * 1000;
+	via_clock = (ide_pci_clk ? ide_pci_clk : 33) * 1000;
 
 	switch (via_clock) {
 		case 33000: via_clock = 33333; break;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 3261c6691759..dad535659249 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -994,8 +994,6 @@ int ide_taskfile_ioctl(ide_drive_t *, unsigned int, unsigned long);
 int ide_cmd_ioctl(ide_drive_t *, unsigned int, unsigned long);
 int ide_task_ioctl(ide_drive_t *, unsigned int, unsigned long);
 
-extern int system_bus_clock(void);
-
 extern int ide_driveid_update(ide_drive_t *);
 extern int ide_config_drive_speed(ide_drive_t *, u8);
 extern u8 eighty_ninty_three (ide_drive_t *);
-- 
cgit v1.2.3


From 931ee0dc5c69e8113233d21942681ab8fecde7f9 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:47 +0200
Subject: ide: remove obsoleted "ide=" kernel parameters

* Remove obsoleted "ide=" kernel parameters.

* Remove no longer needed:
  - ide_setup()
  - parse_options()
  - __setup("", ...)
  - module_param(options, ...)

* Use module_{init,exit}() for MODULE=y case and remove MODULE ifdef.

* Make ide_*acpi* and ide_doubler variables static.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-acpi.c     |  6 +--
 drivers/ide/ide-dma.c      |  2 +-
 drivers/ide/ide.c          | 91 +++-------------------------------------------
 drivers/ide/legacy/gayle.c |  4 +-
 include/linux/ide.h        |  4 --
 5 files changed, 10 insertions(+), 97 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index 9d3601fa5680..6f704628c27d 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -60,15 +60,15 @@ struct ide_acpi_hwif_link {
 #define DEBPRINT(fmt, args...)	do {} while (0)
 #endif	/* DEBUGGING */
 
-int ide_noacpi;
+static int ide_noacpi;
 module_param_named(noacpi, ide_noacpi, bool, 0);
 MODULE_PARM_DESC(noacpi, "disable IDE ACPI support");
 
-int ide_acpigtf;
+static int ide_acpigtf;
 module_param_named(acpigtf, ide_acpigtf, bool, 0);
 MODULE_PARM_DESC(acpigtf, "enable IDE ACPI _GTF support");
 
-int ide_acpionboot;
+static int ide_acpionboot;
 module_param_named(acpionboot, ide_acpionboot, bool, 0);
 MODULE_PARM_DESC(acpionboot, "call IDE ACPI methods on boot");
 
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 653b1ade13d3..174f4704614c 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -692,7 +692,7 @@ static int ide_tune_dma(ide_drive_t *drive)
 	ide_hwif_t *hwif = drive->hwif;
 	u8 speed;
 
-	if (noautodma || drive->nodma || (drive->id->capability & 1) == 0)
+	if (drive->nodma || (drive->id->capability & 1) == 0)
 		return 0;
 
 	/* consult the list of known "bad" drives */
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index e8c88ff2f6b6..1defba3eefe7 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -87,9 +87,9 @@ static const u8 ide_hwif_to_major[] = { IDE0_MAJOR, IDE1_MAJOR,
 					IDE8_MAJOR, IDE9_MAJOR };
 
 DEFINE_MUTEX(ide_cfg_mtx);
- __cacheline_aligned_in_smp DEFINE_SPINLOCK(ide_lock);
 
-int noautodma = 0;
+__cacheline_aligned_in_smp DEFINE_SPINLOCK(ide_lock);
+EXPORT_SYMBOL(ide_lock);
 
 ide_hwif_t ide_hwifs[MAX_HWIFS];	/* master data repository */
 
@@ -698,59 +698,6 @@ set_val:
 
 EXPORT_SYMBOL(generic_ide_ioctl);
 
-/*
- * ide_setup() gets called VERY EARLY during initialization,
- * to handle kernel "command line" strings beginning with "ide".
- *
- * Remember to update Documentation/ide/ide.txt if you change something here.
- */
-static int __init ide_setup(char *s)
-{
-	printk(KERN_INFO "ide_setup: %s", s);
-
-#ifdef CONFIG_BLK_DEV_IDEDOUBLER
-	if (!strcmp(s, "ide=doubler")) {
-		extern int ide_doubler;
-
-		printk(" : Enabled support for IDE doublers\n");
-		ide_doubler = 1;
-		goto obsolete_option;
-	}
-#endif /* CONFIG_BLK_DEV_IDEDOUBLER */
-
-	if (!strcmp(s, "ide=nodma")) {
-		printk(" : Prevented DMA\n");
-		noautodma = 1;
-		goto obsolete_option;
-	}
-
-#ifdef CONFIG_BLK_DEV_IDEACPI
-	if (!strcmp(s, "ide=noacpi")) {
-		//printk(" : Disable IDE ACPI support.\n");
-		ide_noacpi = 1;
-		goto obsolete_option;
-	}
-	if (!strcmp(s, "ide=acpigtf")) {
-		//printk(" : Enable IDE ACPI _GTF support.\n");
-		ide_acpigtf = 1;
-		goto obsolete_option;
-	}
-	if (!strcmp(s, "ide=acpionboot")) {
-		//printk(" : Call IDE ACPI methods on boot.\n");
-		ide_acpionboot = 1;
-		goto obsolete_option;
-	}
-#endif /* CONFIG_BLK_DEV_IDEACPI */
-
-	printk(" -- BAD OPTION\n");
-	return 1;
-obsolete_option:
-	printk(" -- OBSOLETE OPTION, WILL BE REMOVED SOON!\n");
-	return 1;
-}
-
-EXPORT_SYMBOL(ide_lock);
-
 static int ide_bus_match(struct device *dev, struct device_driver *drv)
 {
 	return 1;
@@ -1087,32 +1034,7 @@ out_port_class:
 	return ret;
 }
 
-#ifdef MODULE
-static char *options = NULL;
-module_param(options, charp, 0);
-MODULE_LICENSE("GPL");
-
-static void __init parse_options (char *line)
-{
-	char *next = line;
-
-	if (line == NULL || !*line)
-		return;
-	while ((line = next) != NULL) {
- 		if ((next = strchr(line,' ')) != NULL)
-			*next++ = 0;
-		if (!ide_setup(line))
-			printk (KERN_INFO "Unknown option '%s'\n", line);
-	}
-}
-
-int __init init_module (void)
-{
-	parse_options(options);
-	return ide_init();
-}
-
-void __exit cleanup_module (void)
+static void __exit ide_exit(void)
 {
 	proc_ide_destroy();
 
@@ -1121,10 +1043,7 @@ void __exit cleanup_module (void)
 	bus_unregister(&ide_bus_type);
 }
 
-#else /* !MODULE */
-
-__setup("", ide_setup);
-
 module_init(ide_init);
+module_exit(ide_exit);
 
-#endif /* MODULE */
+MODULE_LICENSE("GPL");
diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index fed7d812761c..b78941680c32 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -64,9 +64,7 @@
 #define GAYLE_HAS_CONTROL_REG	(!ide_doubler)
 #define GAYLE_IDEREG_SIZE	(ide_doubler ? 0x1000 : 0x2000)
 
-int ide_doubler = 0;	/* support IDE doublers? */
-EXPORT_SYMBOL_GPL(ide_doubler);
-
+static int ide_doubler;
 module_param_named(doubler, ide_doubler, bool, 0);
 MODULE_PARM_DESC(doubler, "enable support for IDE doublers");
 #endif /* CONFIG_BLK_DEV_IDEDOUBLER */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index dad535659249..0fa1812d0438 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -813,10 +813,6 @@ int generic_ide_ioctl(ide_drive_t *, struct file *, struct block_device *, unsig
 #ifndef _IDE_C
 extern	ide_hwif_t	ide_hwifs[];		/* master data repository */
 #endif
-extern int ide_noacpi;
-extern int ide_acpigtf;
-extern int ide_acpionboot;
-extern int noautodma;
 
 extern int ide_vlb_clk;
 extern int ide_pci_clk;
-- 
cgit v1.2.3


From 9a410e79b552bacb4481f85618aa7333b7776ed7 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:48 +0200
Subject: ide: remove IDE_TFLAG_NO_SELECT_MASK taskfile flag

Always call SELECT_MASK(..., 0) in ide_tf_load() (needs to be done
to match ide_set_irq(..., 1)) and then remove IDE_TFLAG_NO_SELECT_MASK
taskfile flag.

This change should only affect hpt366 and icside host drivers since
->maskproc(..., 0) for sgiioc4 is equivalent to ide_set_irq(..., 1).

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c     | 4 ++--
 drivers/ide/ide-disk.c   | 3 +--
 drivers/ide/ide-floppy.c | 3 +--
 drivers/ide/ide-iops.c   | 4 +---
 drivers/ide/ide-tape.c   | 3 +--
 drivers/scsi/ide-scsi.c  | 2 +-
 include/linux/ide.h      | 1 -
 7 files changed, 7 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index ac542ffffa49..0fbc2d8d0d53 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -530,8 +530,8 @@ static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive,
 		info->dma = !hwif->dma_ops->dma_setup(drive);
 
 	/* set up the controller registers */
-	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL |
-			   IDE_TFLAG_NO_SELECT_MASK, xferlen, info->dma);
+	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL,
+			   xferlen, info->dma);
 
 	if (info->cd_flags & IDE_CD_FLAG_DRQ_INTERRUPT) {
 		/* waiting for CDB interrupt, not DMA yet. */
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index c5f22ef8ed24..5f49a4ae9dd8 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -198,8 +198,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 	}
 
 	memset(&task, 0, sizeof(task));
-	task.tf_flags = IDE_TFLAG_NO_SELECT_MASK;  /* FIXME? */
-	task.tf_flags |= (IDE_TFLAG_TF | IDE_TFLAG_DEVICE);
+	task.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 
 	if (drive->select.b.lba) {
 		if (lba48) {
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 9161cd92a842..1852008d9ee4 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -667,8 +667,7 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 	if ((pc->flags & PC_FLAG_DMA_RECOMMENDED) && drive->using_dma)
 		dma = !hwif->dma_ops->dma_setup(drive);
 
-	ide_pktcmd_tf_load(drive, IDE_TFLAG_NO_SELECT_MASK |
-			   IDE_TFLAG_OUT_DEVICE, bcount, dma);
+	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma);
 
 	if (dma) {
 		/* Begin DMA, if necessary */
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 2de4c8f581eb..9f9916fe6c2f 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -121,9 +121,7 @@ static void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 		HIHI = 0xFF;
 
 	ide_set_irq(drive, 1);
-
-	if ((task->tf_flags & IDE_TFLAG_NO_SELECT_MASK) == 0)
-		SELECT_MASK(drive, 0);
+	SELECT_MASK(drive, 0);
 
 	if (task->tf_flags & IDE_TFLAG_OUT_DATA) {
 		u16 data = (tf->hob_data << 8) | tf->data;
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index a5f0b774527b..cc7991c7c252 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1046,8 +1046,7 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 	if ((pc->flags & PC_FLAG_DMA_RECOMMENDED) && drive->using_dma)
 		dma_ok = !hwif->dma_ops->dma_setup(drive);
 
-	ide_pktcmd_tf_load(drive, IDE_TFLAG_NO_SELECT_MASK |
-			   IDE_TFLAG_OUT_DEVICE, bcount, dma_ok);
+	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma_ok);
 
 	if (dma_ok)
 		/* Will begin DMA later */
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index da261806d62a..3222aa589dbf 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -564,7 +564,7 @@ static ide_startstop_t idescsi_issue_pc(ide_drive_t *drive,
 		hwif->sg_mapped = 0;
 	}
 
-	ide_pktcmd_tf_load(drive, IDE_TFLAG_NO_SELECT_MASK, bcount, dma);
+	ide_pktcmd_tf_load(drive, 0, bcount, dma);
 
 	if (dma)
 		pc->flags |= PC_FLAG_DMA_OK;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 0fa1812d0438..d4a910cdb907 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -869,7 +869,6 @@ extern void ide_end_drive_cmd(ide_drive_t *, u8, u8);
 
 enum {
 	IDE_TFLAG_LBA48			= (1 << 0),
-	IDE_TFLAG_NO_SELECT_MASK	= (1 << 1),
 	IDE_TFLAG_FLAGGED		= (1 << 2),
 	IDE_TFLAG_OUT_DATA		= (1 << 3),
 	IDE_TFLAG_OUT_HOB_FEATURE	= (1 << 4),
-- 
cgit v1.2.3


From ed4af48fd660176680da905817f6e40d51436e4c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:48 +0200
Subject: ide: move IRQ unmasking out from ->tf_load method

Move IRQ unmasking out from ->tf_load method to its users.

There should be no functional changes caused by this patch
(SELECT_MASK() is NOP except for hpt366, icside and sgiioc4).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/h8300/ide-h8300.c | 2 --
 drivers/ide/ide-io.c          | 2 ++
 drivers/ide/ide-iops.c        | 5 +----
 drivers/ide/ide-taskfile.c    | 2 ++
 drivers/ide/pci/scc_pata.c    | 2 --
 include/linux/ide.h           | 1 +
 6 files changed, 6 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index ecf53bb0d2aa..d5afc28eaae7 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -52,8 +52,6 @@ static void h8300_tf_load(ide_drive_t *drive, ide_task_t *task)
 	if (task->tf_flags & IDE_TFLAG_FLAGGED)
 		HIHI = 0xFF;
 
-	ide_set_irq(drive, 1);
-
 	if (task->tf_flags & IDE_TFLAG_OUT_DATA)
 		mm_outw((tf->hob_data << 8) | tf->data, io_ports->data_addr);
 
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index c22a337ced4e..2083cc08b2ce 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1579,6 +1579,8 @@ void ide_pktcmd_tf_load(ide_drive_t *drive, u32 tf_flags, u16 bcount, u8 dma)
 	task.tf.lbah    = (bcount >> 8) & 0xff;
 
 	ide_tf_dump(drive->name, &task.tf);
+	ide_set_irq(drive, 1);
+	SELECT_MASK(drive, 0);
 	drive->hwif->tf_load(drive, &task);
 }
 
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 9f9916fe6c2f..491980aab866 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -95,7 +95,7 @@ void SELECT_DRIVE (ide_drive_t *drive)
 	hwif->OUTB(drive->select.all, hwif->io_ports.device_addr);
 }
 
-static void SELECT_MASK(ide_drive_t *drive, int mask)
+void SELECT_MASK(ide_drive_t *drive, int mask)
 {
 	const struct ide_port_ops *port_ops = drive->hwif->port_ops;
 
@@ -120,9 +120,6 @@ static void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 	if (task->tf_flags & IDE_TFLAG_FLAGGED)
 		HIHI = 0xFF;
 
-	ide_set_irq(drive, 1);
-	SELECT_MASK(drive, 0);
-
 	if (task->tf_flags & IDE_TFLAG_OUT_DATA) {
 		u16 data = (tf->hob_data << 8) | tf->data;
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index b6a1c4b51129..6a17ab54f801 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -109,6 +109,8 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 
 	if ((task->tf_flags & IDE_TFLAG_DMA_PIO_FALLBACK) == 0) {
 		ide_tf_dump(drive->name, tf);
+		ide_set_irq(drive, 1);
+		SELECT_MASK(drive, 0);
 		hwif->tf_load(drive, task);
 	}
 
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 910fb00deb71..37e8cfcabb4b 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -662,8 +662,6 @@ static void scc_tf_load(ide_drive_t *drive, ide_task_t *task)
 	if (task->tf_flags & IDE_TFLAG_FLAGGED)
 		HIHI = 0xFF;
 
-	ide_set_irq(drive, 1);
-
 	if (task->tf_flags & IDE_TFLAG_OUT_DATA)
 		out_be32((void *)io_ports->data_addr,
 			 (tf->hob_data << 8) | tf->data);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index d4a910cdb907..56d0bc2dffee 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -973,6 +973,7 @@ typedef struct ide_task_s {
 void ide_tf_dump(const char *, struct ide_taskfile *);
 
 extern void SELECT_DRIVE(ide_drive_t *);
+void SELECT_MASK(ide_drive_t *, int);
 
 extern int drive_is_ready(ide_drive_t *);
 
-- 
cgit v1.2.3


From 135721446144af005109c25eeacca4fdddcd9a66 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:49 +0200
Subject: ide: remove ->mmio flag from ide_hwif_t

Since scc_pata host driver no longer uses IDE PCI layer / ide_dma_setup()
and all other ->mmio users set also IDE_HFLAG_MMIO host flag we can safely
remove ->mmio flag.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/palm_bk3710.c |  1 -
 drivers/ide/ide-dma.c         |  2 +-
 drivers/ide/pci/scc_pata.c    |  1 -
 drivers/ide/pci/siimage.c     | 25 +++++++++++++------------
 drivers/ide/setup-pci.c       |  4 ++--
 include/linux/ide.h           |  1 -
 6 files changed, 16 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 74a05dc6d1e6..3839f5722985 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -405,7 +405,6 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	ide_init_port_data(hwif, i);
 	ide_init_port_hw(hwif, &hw);
 
-	hwif->mmio = 1;
 	default_hwif_mmiops(hwif);
 
 	idx[0] = i;
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 174f4704614c..7ee44f86bc54 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -463,7 +463,7 @@ int ide_dma_setup(ide_drive_t *drive)
 	}
 
 	/* PRD table */
-	if (hwif->mmio)
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		writel(hwif->dmatable_dma,
 		       (void __iomem *)(hwif->dma_base + ATA_DMA_TABLE_OFS));
 	else
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 37e8cfcabb4b..133053c7a480 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -793,7 +793,6 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 
 	hwif->dma_base = dma_base;
 	hwif->config_data = ports->ctl;
-	hwif->mmio = 1;
 }
 
 /**
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 0006b9e58567..b75e9bb390a7 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -94,7 +94,7 @@ static unsigned long siimage_selreg(ide_hwif_t *hwif, int r)
 	unsigned long base = (unsigned long)hwif->hwif_data;
 
 	base += 0xA0 + r;
-	if (hwif->mmio)
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		base += hwif->channel << 6;
 	else
 		base += hwif->channel << 4;
@@ -117,7 +117,7 @@ static inline unsigned long siimage_seldev(ide_drive_t *drive, int r)
 	unsigned long base	= (unsigned long)hwif->hwif_data;
 
 	base += 0xA0 + r;
-	if (hwif->mmio)
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		base += hwif->channel << 6;
 	else
 		base += hwif->channel << 4;
@@ -190,7 +190,9 @@ static u8 sil_pata_udma_filter(ide_drive_t *drive)
 	unsigned long base	= (unsigned long)hwif->hwif_data;
 	u8 scsc, mask		= 0;
 
-	scsc = sil_ioread8(dev, base + (hwif->mmio ? 0x4A : 0x8A));
+	base += (hwif->host_flags & IDE_HFLAG_MMIO) ? 0x4A : 0x8A;
+
+	scsc = sil_ioread8(dev, base);
 
 	switch (scsc & 0x30) {
 	case 0x10:	/* 133 */
@@ -238,8 +240,9 @@ static void sil_set_pio_mode(ide_drive_t *drive, u8 pio)
 	unsigned long tfaddr	= siimage_selreg(hwif,	0x02);
 	unsigned long base	= (unsigned long)hwif->hwif_data;
 	u8 tf_pio		= pio;
-	u8 addr_mask		= hwif->channel ? (hwif->mmio ? 0xF4 : 0x84)
-						: (hwif->mmio ? 0xB4 : 0x80);
+	u8 mmio			= (hwif->host_flags & IDE_HFLAG_MMIO) ? 1 : 0;
+	u8 addr_mask		= hwif->channel ? (mmio ? 0xF4 : 0x84)
+						: (mmio ? 0xB4 : 0x80);
 	u8 mode			= 0;
 	u8 unit			= drive->select.b.unit;
 
@@ -290,13 +293,13 @@ static void sil_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	u16 ultra = 0, multi	= 0;
 	u8 mode = 0, unit	= drive->select.b.unit;
 	unsigned long base	= (unsigned long)hwif->hwif_data;
-	u8 scsc = 0, addr_mask	= hwif->channel ?
-					(hwif->mmio ? 0xF4 : 0x84) :
-					(hwif->mmio ? 0xB4 : 0x80);
+	u8 mmio			= (hwif->host_flags & IDE_HFLAG_MMIO) ? 1 : 0;
+	u8 scsc = 0, addr_mask	= hwif->channel ? (mmio ? 0xF4 : 0x84)
+						: (mmio ? 0xB4 : 0x80);
 	unsigned long ma	= siimage_seldev(drive, 0x08);
 	unsigned long ua	= siimage_seldev(drive, 0x0C);
 
-	scsc  = sil_ioread8 (dev, base + (hwif->mmio ? 0x4A : 0x8A));
+	scsc  = sil_ioread8 (dev, base + (mmio ? 0x4A : 0x8A));
 	mode  = sil_ioread8 (dev, base + addr_mask);
 	multi = sil_ioread16(dev, ma);
 	ultra = sil_ioread16(dev, ua);
@@ -391,7 +394,7 @@ static int siimage_mmio_dma_test_irq(ide_drive_t *drive)
 
 static int siimage_dma_test_irq(ide_drive_t *drive)
 {
-	if (drive->hwif->mmio)
+	if (drive->hwif->host_flags & IDE_HFLAG_MMIO)
 		return siimage_mmio_dma_test_irq(drive);
 	else
 		return siimage_io_dma_test_irq(drive);
@@ -640,8 +643,6 @@ static void __devinit init_mmio_iops_siimage(ide_hwif_t *hwif)
 	hwif->irq = dev->irq;
 
 	hwif->dma_base = (unsigned long)addr + (ch ? 0x08 : 0x00);
-
-	hwif->mmio = 1;
 }
 
 static int is_dev_seagate_sata(ide_drive_t *drive)
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 5171601fb255..abcfb1739d4d 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -87,7 +87,7 @@ unsigned long ide_pci_dma_base(ide_hwif_t *hwif, const struct ide_port_info *d)
 	unsigned long dma_base = 0;
 	u8 dma_stat = 0;
 
-	if (hwif->mmio)
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		return hwif->dma_base;
 
 	if (hwif->mate && hwif->mate->dma_base) {
@@ -374,7 +374,7 @@ int ide_hwif_setup_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
 		if (base == 0 || ide_pci_set_master(dev, d->name) < 0)
 			return -1;
 
-		if (hwif->mmio)
+		if (hwif->host_flags & IDE_HFLAG_MMIO)
 			printk(KERN_INFO "    %s: MMIO-DMA\n", hwif->name);
 		else
 			printk(KERN_INFO "    %s: BM-DMA at 0x%04lx-0x%04lx\n",
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 56d0bc2dffee..b01b102be4de 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -532,7 +532,6 @@ typedef struct hwif_s {
 	unsigned	serialized : 1;	/* serialized all channel operation */
 	unsigned	sharing_irq: 1;	/* 1 = sharing irq with another hwif */
 	unsigned	sg_mapped  : 1;	/* sg_table and sg_nents are ready */
-	unsigned	mmio       : 1; /* host uses MMIO */
 
 	struct device		gendev;
 	struct device		*portdev;
-- 
cgit v1.2.3


From f8c4bd0ab2b8783c0f080957781e9f70bee48eaa Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:49 +0200
Subject: ide: pass 'hwif *' instead of 'drive *' to ->OUTBSYNC method

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       |  2 +-
 drivers/ide/ide-iops.c     | 18 +++++++++---------
 drivers/ide/ide-probe.c    |  6 +++---
 drivers/ide/ide-taskfile.c |  2 +-
 drivers/ide/pci/scc_pata.c |  5 +----
 drivers/ide/ppc/pmac.c     |  6 +++---
 drivers/scsi/ide-scsi.c    |  2 +-
 include/linux/ide.h        |  2 +-
 8 files changed, 20 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 2083cc08b2ce..c28fcdf0ee9e 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -437,7 +437,7 @@ static ide_startstop_t ide_atapi_error(ide_drive_t *drive, struct request *rq, u
 
 	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		hwif->OUTBSYNC(drive, WIN_IDLEIMMEDIATE,
+		hwif->OUTBSYNC(hwif, WIN_IDLEIMMEDIATE,
 			       hwif->io_ports.command_addr);
 
 	if (rq->errors >= ERROR_MAX) {
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 491980aab866..4c32cf0b623c 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -42,7 +42,7 @@ static void ide_outb (u8 val, unsigned long port)
 	outb(val, port);
 }
 
-static void ide_outbsync (ide_drive_t *drive, u8 addr, unsigned long port)
+static void ide_outbsync(ide_hwif_t *hwif, u8 addr, unsigned long port)
 {
 	outb(addr, port);
 }
@@ -68,7 +68,7 @@ static void ide_mm_outb (u8 value, unsigned long port)
 	writeb(value, (void __iomem *) port);
 }
 
-static void ide_mm_outbsync (ide_drive_t *drive, u8 value, unsigned long port)
+static void ide_mm_outbsync(ide_hwif_t *hwif, u8 value, unsigned long port)
 {
 	writeb(value, (void __iomem *) port);
 }
@@ -686,7 +686,7 @@ int ide_driveid_update(ide_drive_t *drive)
 	SELECT_MASK(drive, 1);
 	ide_set_irq(drive, 0);
 	msleep(50);
-	hwif->OUTBSYNC(drive, WIN_IDENTIFY, hwif->io_ports.command_addr);
+	hwif->OUTBSYNC(hwif, WIN_IDENTIFY, hwif->io_ports.command_addr);
 	timeout = jiffies + WAIT_WORSTCASE;
 	do {
 		if (time_after(jiffies, timeout)) {
@@ -773,7 +773,7 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 	ide_set_irq(drive, 0);
 	hwif->OUTB(speed, io_ports->nsect_addr);
 	hwif->OUTB(SETFEATURES_XFER, io_ports->feature_addr);
-	hwif->OUTBSYNC(drive, WIN_SETFEATURES, io_ports->command_addr);
+	hwif->OUTBSYNC(hwif, WIN_SETFEATURES, io_ports->command_addr);
 	if (drive->quirk_list == 2)
 		ide_set_irq(drive, 1);
 
@@ -881,7 +881,7 @@ void ide_execute_command(ide_drive_t *drive, u8 cmd, ide_handler_t *handler,
 
 	spin_lock_irqsave(&ide_lock, flags);
 	__ide_set_handler(drive, handler, timeout, expiry);
-	hwif->OUTBSYNC(drive, cmd, hwif->io_ports.command_addr);
+	hwif->OUTBSYNC(hwif, cmd, hwif->io_ports.command_addr);
 	/*
 	 * Drive takes 400nS to respond, we must avoid the IRQ being
 	 * serviced before that.
@@ -899,7 +899,7 @@ void ide_execute_pkt_cmd(ide_drive_t *drive)
 	unsigned long flags;
 
 	spin_lock_irqsave(&ide_lock, flags);
-	hwif->OUTBSYNC(drive, WIN_PACKETCMD, hwif->io_ports.command_addr);
+	hwif->OUTBSYNC(hwif, WIN_PACKETCMD, hwif->io_ports.command_addr);
 	ndelay(400);
 	spin_unlock_irqrestore(&ide_lock, flags);
 }
@@ -1094,7 +1094,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		pre_reset(drive);
 		SELECT_DRIVE(drive);
 		udelay (20);
-		hwif->OUTBSYNC(drive, WIN_SRST, io_ports->command_addr);
+		hwif->OUTBSYNC(hwif, WIN_SRST, io_ports->command_addr);
 		ndelay(400);
 		hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
 		hwgroup->polling = 1;
@@ -1125,14 +1125,14 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * recover from reset very quickly, saving us the first 50ms wait time.
 	 */
 	/* set SRST and nIEN */
-	hwif->OUTBSYNC(drive, drive->ctl|6, io_ports->ctl_addr);
+	hwif->OUTBSYNC(hwif, drive->ctl | 6, io_ports->ctl_addr);
 	/* more than enough time */
 	udelay(10);
 	if (drive->quirk_list == 2)
 		ctl = drive->ctl;	/* clear SRST and nIEN */
 	else
 		ctl = drive->ctl | 2;	/* clear SRST, leave nIEN */
-	hwif->OUTBSYNC(drive, ctl, io_ports->ctl_addr);
+	hwif->OUTBSYNC(hwif, ctl, io_ports->ctl_addr);
 	/* more than enough time */
 	udelay(10);
 	hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 12513c45d700..b010633eb5ba 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -293,7 +293,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		hwif->OUTB(0, io_ports->feature_addr);
 
 	/* ask drive for ID */
-	hwif->OUTBSYNC(drive, cmd, io_ports->command_addr);
+	hwif->OUTBSYNC(hwif, cmd, hwif->io_ports.command_addr);
 
 	timeout = ((cmd == WIN_IDENTIFY) ? WAIT_WORSTCASE : WAIT_PIDENTIFY) / 2;
 	timeout += jiffies;
@@ -480,7 +480,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 			msleep(50);
 			SELECT_DRIVE(drive);
 			msleep(50);
-			hwif->OUTBSYNC(drive, WIN_SRST, io_ports->command_addr);
+			hwif->OUTBSYNC(hwif, WIN_SRST, io_ports->command_addr);
 			(void)ide_busy_sleep(hwif);
 			rc = try_to_identify(drive, cmd);
 		}
@@ -516,7 +516,7 @@ static void enable_nest (ide_drive_t *drive)
 	printk("%s: enabling %s -- ", hwif->name, drive->id->model);
 	SELECT_DRIVE(drive);
 	msleep(50);
-	hwif->OUTBSYNC(drive, EXABYTE_ENABLE_NEST, hwif->io_ports.command_addr);
+	hwif->OUTBSYNC(hwif, EXABYTE_ENABLE_NEST, hwif->io_ports.command_addr);
 
 	if (ide_busy_sleep(hwif)) {
 		printk(KERN_CONT "failed (timeout)\n");
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 6a17ab54f801..cf55a48a7dd2 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -117,7 +117,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	switch (task->data_phase) {
 	case TASKFILE_MULTI_OUT:
 	case TASKFILE_OUT:
-		hwif->OUTBSYNC(drive, tf->command, hwif->io_ports.command_addr);
+		hwif->OUTBSYNC(hwif, tf->command, hwif->io_ports.command_addr);
 		ndelay(400);	/* FIXME */
 		return pre_task_out_intr(drive, task->rq);
 	case TASKFILE_MULTI_IN:
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 133053c7a480..32eb0877fce2 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -148,11 +148,8 @@ static void scc_ide_outb(u8 addr, unsigned long port)
 	out_be32((void*)port, addr);
 }
 
-static void
-scc_ide_outbsync(ide_drive_t * drive, u8 addr, unsigned long port)
+static void scc_ide_outbsync(ide_hwif_t *hwif, u8 addr, unsigned long port)
 {
-	ide_hwif_t *hwif = HWIF(drive);
-
 	out_be32((void*)port, addr);
 	eieio();
 	in_be32((void*)(hwif->dma_base + 0x01c));
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index ba2d58727964..dcb2c466bb97 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -480,13 +480,13 @@ pmac_ide_do_update_timings(ide_drive_t *drive)
 		pmac_ide_selectproc(drive);
 }
 
-static void
-pmac_outbsync(ide_drive_t *drive, u8 value, unsigned long port)
+static void pmac_outbsync(ide_hwif_t *hwif, u8 value, unsigned long port)
 {
 	u32 tmp;
 	
 	writeb(value, (void __iomem *) port);
-	tmp = readl(PMAC_IDE_REG(IDE_TIMING_CONFIG));
+	tmp = readl((void __iomem *)(hwif->io_ports.data_addr
+				     + IDE_TIMING_CONFIG));
 }
 
 /*
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 3222aa589dbf..d7fd5e550a25 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -257,7 +257,7 @@ idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 
 	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		hwif->OUTBSYNC(drive, WIN_IDLEIMMEDIATE,
+		hwif->OUTBSYNC(hwif, WIN_IDLEIMMEDIATE,
 			       hwif->io_ports.command_addr);
 
 	rq->errors++;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index b01b102be4de..1c3431469649 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -493,7 +493,7 @@ typedef struct hwif_s {
 	void (*ide_dma_clear_irq)(ide_drive_t *drive);
 
 	void (*OUTB)(u8 addr, unsigned long port);
-	void (*OUTBSYNC)(ide_drive_t *drive, u8 addr, unsigned long port);
+	void (*OUTBSYNC)(struct hwif_s *hwif, u8 addr, unsigned long port);
 
 	u8  (*INB)(unsigned long port);
 
-- 
cgit v1.2.3


From 0fd04dcc2ebb6ec9088c24b368b0ce1f42a98ef5 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:50 +0200
Subject: ide: use ->OUTBSYNC in ide_set_irq()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index 1c3431469649..4d1c9714f1d9 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1340,7 +1340,8 @@ static inline void ide_set_irq(ide_drive_t *drive, int on)
 {
 	ide_hwif_t *hwif = drive->hwif;
 
-	hwif->OUTB(drive->ctl | (on ? 0 : 2), hwif->io_ports.ctl_addr);
+	hwif->OUTBSYNC(hwif, drive->ctl | (on ? 0 : 2),
+		       hwif->io_ports.ctl_addr);
 }
 
 static inline u8 ide_read_status(ide_drive_t *drive)
-- 
cgit v1.2.3


From ff07488346702f554aaeb6aae982540aa0302373 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:50 +0200
Subject: ide: remove drive->ctl

Remove drive->ctl (it is always equal to 0x08 after init time).

While at it:

* Use ATA_DEVCTL_OBS define.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/h8300/ide-h8300.c |  4 ++--
 drivers/ide/ide-iops.c        | 10 +++++-----
 drivers/ide/ide-probe.c       |  2 +-
 drivers/ide/ide.c             |  1 -
 drivers/ide/pci/hpt366.c      |  3 +--
 drivers/ide/pci/ns87415.c     |  4 ++--
 drivers/ide/pci/scc_pata.c    |  4 ++--
 drivers/ide/pci/sgiioc4.c     |  2 +-
 include/linux/ide.h           |  3 +--
 9 files changed, 15 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index d5afc28eaae7..ae37ee58bae2 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -96,7 +96,7 @@ static void h8300_tf_read(ide_drive_t *drive, ide_task_t *task)
 	}
 
 	/* be sure we're looking at the low order bits */
-	outb(drive->ctl & ~0x80, io_ports->ctl_addr);
+	outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = inb(io_ports->nsect_addr);
@@ -110,7 +110,7 @@ static void h8300_tf_read(ide_drive_t *drive, ide_task_t *task)
 		tf->device = inb(io_ports->device_addr);
 
 	if (task->tf_flags & IDE_TFLAG_LBA48) {
-		outb(drive->ctl | 0x80, io_ports->ctl_addr);
+		outb(ATA_DEVCTL_OBS | 0x80, io_ports->ctl_addr);
 
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_FEATURE)
 			tf->hob_feature = inb(io_ports->feature_addr);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 4c32cf0b623c..80ad4f234f3f 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -186,7 +186,7 @@ static void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
 	}
 
 	/* be sure we're looking at the low order bits */
-	tf_outb(drive->ctl & ~0x80, io_ports->ctl_addr);
+	tf_outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = tf_inb(io_ports->nsect_addr);
@@ -200,7 +200,7 @@ static void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
 		tf->device = tf_inb(io_ports->device_addr);
 
 	if (task->tf_flags & IDE_TFLAG_LBA48) {
-		tf_outb(drive->ctl | 0x80, io_ports->ctl_addr);
+		tf_outb(ATA_DEVCTL_OBS | 0x80, io_ports->ctl_addr);
 
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_FEATURE)
 			tf->hob_feature = tf_inb(io_ports->feature_addr);
@@ -1125,13 +1125,13 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * recover from reset very quickly, saving us the first 50ms wait time.
 	 */
 	/* set SRST and nIEN */
-	hwif->OUTBSYNC(hwif, drive->ctl | 6, io_ports->ctl_addr);
+	hwif->OUTBSYNC(hwif, ATA_DEVCTL_OBS | 6, io_ports->ctl_addr);
 	/* more than enough time */
 	udelay(10);
 	if (drive->quirk_list == 2)
-		ctl = drive->ctl;	/* clear SRST and nIEN */
+		ctl = ATA_DEVCTL_OBS;		/* clear SRST and nIEN */
 	else
-		ctl = drive->ctl | 2;	/* clear SRST, leave nIEN */
+		ctl = ATA_DEVCTL_OBS | 2;	/* clear SRST, leave nIEN */
 	hwif->OUTBSYNC(hwif, ctl, io_ports->ctl_addr);
 	/* more than enough time */
 	udelay(10);
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 809362b13c99..d21e51a02c3e 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1065,7 +1065,7 @@ static int init_irq (ide_hwif_t *hwif)
 
 		if (io_ports->ctl_addr)
 			/* clear nIEN */
-			hwif->OUTBSYNC(hwif, 0x08, io_ports->ctl_addr);
+			hwif->OUTBSYNC(hwif, ATA_DEVCTL_OBS, io_ports->ctl_addr);
 
 		if (request_irq(hwif->irq,&ide_intr,sa,hwif->name,hwgroup))
 	       		goto out_unlink;
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 1defba3eefe7..2b8453510e09 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -136,7 +136,6 @@ static void ide_port_init_devices_data(ide_hwif_t *hwif)
 		drive->media			= ide_disk;
 		drive->select.all		= (unit<<4)|0xa0;
 		drive->hwif			= hwif;
-		drive->ctl			= 0x08;
 		drive->ready_stat		= READY_STAT;
 		drive->bad_wstat		= BAD_W_STAT;
 		drive->special.b.recalibrate	= 1;
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index c929dadaaaff..397c6cbe953c 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -759,8 +759,7 @@ static void hpt3xx_maskproc(ide_drive_t *drive, int mask)
 				enable_irq (hwif->irq);
 		}
 	} else
-		outb(mask ? (drive->ctl | 2) : (drive->ctl & ~2),
-		     hwif->io_ports.ctl_addr);
+		outb(ATA_DEVCTL_OBS | (mask ? 2 : 0), hwif->io_ports.ctl_addr);
 }
 
 /*
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index a7a41bb82778..45ba71a7182f 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -76,7 +76,7 @@ static void superio_tf_read(ide_drive_t *drive, ide_task_t *task)
 	}
 
 	/* be sure we're looking at the low order bits */
-	outb(drive->ctl & ~0x80, io_ports->ctl_addr);
+	outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = inb(io_ports->nsect_addr);
@@ -90,7 +90,7 @@ static void superio_tf_read(ide_drive_t *drive, ide_task_t *task)
 		tf->device = superio_ide_inb(io_ports->device_addr);
 
 	if (task->tf_flags & IDE_TFLAG_LBA48) {
-		outb(drive->ctl | 0x80, io_ports->ctl_addr);
+		outb(ATA_DEVCTL_OBS | 0x80, io_ports->ctl_addr);
 
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_FEATURE)
 			tf->hob_feature = inb(io_ports->feature_addr);
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 32eb0877fce2..1584ebb6a185 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -703,7 +703,7 @@ static void scc_tf_read(ide_drive_t *drive, ide_task_t *task)
 	}
 
 	/* be sure we're looking at the low order bits */
-	scc_ide_outb(drive->ctl & ~0x80, io_ports->ctl_addr);
+	scc_ide_outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = scc_ide_inb(io_ports->nsect_addr);
@@ -717,7 +717,7 @@ static void scc_tf_read(ide_drive_t *drive, ide_task_t *task)
 		tf->device = scc_ide_inb(io_ports->device_addr);
 
 	if (task->tf_flags & IDE_TFLAG_LBA48) {
-		scc_ide_outb(drive->ctl | 0x80, io_ports->ctl_addr);
+		scc_ide_outb(ATA_DEVCTL_OBS | 0x80, io_ports->ctl_addr);
 
 		if (task->tf_flags & IDE_TFLAG_IN_HOB_FEATURE)
 			tf->hob_feature = scc_ide_inb(io_ports->feature_addr);
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index c1b667c53f2b..24513e3dcd6b 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -111,7 +111,7 @@ sgiioc4_init_hwif_ports(hw_regs_t * hw, unsigned long data_port,
 static void
 sgiioc4_maskproc(ide_drive_t * drive, int mask)
 {
-	writeb(mask ? (drive->ctl | 2) : (drive->ctl & ~2),
+	writeb(ATA_DEVCTL_OBS | (mask ? 2 : 0),
 	       (void __iomem *)drive->hwif->io_ports.ctl_addr);
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 4d1c9714f1d9..d8c86f0362c4 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -364,7 +364,6 @@ typedef struct ide_drive_s {
         u8	wcache;		/* status of write cache */
 	u8	acoustic;	/* acoustic management */
 	u8	media;		/* disk, cdrom, tape, floppy, ... */
-	u8	ctl;		/* "normal" value for Control register */
 	u8	ready_stat;	/* min status value for drive ready */
 	u8	mult_count;	/* current multiple sector setting */
 	u8	mult_req;	/* requested multiple sector setting */
@@ -1340,7 +1339,7 @@ static inline void ide_set_irq(ide_drive_t *drive, int on)
 {
 	ide_hwif_t *hwif = drive->hwif;
 
-	hwif->OUTBSYNC(hwif, drive->ctl | (on ? 0 : 2),
+	hwif->OUTBSYNC(hwif, ATA_DEVCTL_OBS | (on ? 0 : 2),
 		       hwif->io_ports.ctl_addr);
 }
 
-- 
cgit v1.2.3


From 63f5abb0959337db0d5bece9cefba03cdcadec51 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Tue, 15 Jul 2008 21:21:51 +0200
Subject: ide: remove action argument in ide_do_drive_cmd

ide_do_drive_cmd is called only with ide_preempt action argument. So
we can remove the action argument in ide_do_drive_cmd and ide_action_t
typedef.

This patch also includes two minor cleanups: 1) ide_do_drive_cmd
always succeeds so we don't need the return value; 2) the callers use
blk_rq_init before ide_do_drive_cmd so there is no need to initialize
rq->errors.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c     |  2 +-
 drivers/ide/ide-floppy.c |  2 +-
 drivers/ide/ide-io.c     | 40 +++++++++-------------------------------
 drivers/ide/ide-tape.c   |  2 +-
 drivers/scsi/ide-scsi.c  |  3 ++-
 include/linux/ide.h      | 12 +-----------
 6 files changed, 15 insertions(+), 46 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 0fbc2d8d0d53..043129c422fe 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -213,7 +213,7 @@ static void cdrom_queue_request_sense(ide_drive_t *drive, void *sense,
 	/* NOTE! Save the failed command in "rq->buffer" */
 	rq->buffer = (void *) failed_command;
 
-	(void) ide_do_drive_cmd(drive, rq, ide_preempt);
+	ide_do_drive_cmd(drive, rq);
 }
 
 static void cdrom_end_request(ide_drive_t *drive, int uptodate)
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 1852008d9ee4..53209a473937 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -291,7 +291,7 @@ static void idefloppy_queue_pc_head(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	rq->cmd_type = REQ_TYPE_SPECIAL;
 	rq->cmd_flags |= REQ_PREEMPT;
 	rq->rq_disk = floppy->disk;
-	(void) ide_do_drive_cmd(drive, rq, ide_preempt);
+	ide_do_drive_cmd(drive, rq);
 }
 
 static struct ide_atapi_pc *idefloppy_next_pc_storage(ide_drive_t *drive)
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index c28fcdf0ee9e..28057747c1f8 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -1520,49 +1520,27 @@ irqreturn_t ide_intr (int irq, void *dev_id)
  *	ide_do_drive_cmd	-	issue IDE special command
  *	@drive: device to issue command
  *	@rq: request to issue
- *	@action: action for processing
  *
  *	This function issues a special IDE device request
  *	onto the request queue.
  *
- *	If action is ide_wait, then the rq is queued at the end of the
- *	request queue, and the function sleeps until it has been processed.
- *	This is for use when invoked from an ioctl handler.
- *
- *	If action is ide_preempt, then the rq is queued at the head of
- *	the request queue, displacing the currently-being-processed
- *	request and this function returns immediately without waiting
- *	for the new rq to be completed.  This is VERY DANGEROUS, and is
- *	intended for careful use by the ATAPI tape/cdrom driver code.
- *
- *	If action is ide_end, then the rq is queued at the end of the
- *	request queue, and the function returns immediately without waiting
- *	for the new rq to be completed. This is again intended for careful
- *	use by the ATAPI tape/cdrom driver code.
+ *	the rq is queued at the head of the request queue, displacing
+ *	the currently-being-processed request and this function
+ *	returns immediately without waiting for the new rq to be
+ *	completed.  This is VERY DANGEROUS, and is intended for
+ *	careful use by the ATAPI tape/cdrom driver code.
  */
- 
-int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t action)
+
+void ide_do_drive_cmd(ide_drive_t *drive, struct request *rq)
 {
 	unsigned long flags;
 	ide_hwgroup_t *hwgroup = HWGROUP(drive);
-	int where = ELEVATOR_INSERT_BACK;
-
-	rq->errors = 0;
-
-	if (action == ide_preempt)
-		where = ELEVATOR_INSERT_FRONT;
 
 	spin_lock_irqsave(&ide_lock, flags);
-	if (action == ide_preempt)
-		hwgroup->rq = NULL;
-	__elv_add_request(drive->queue, rq, where, 1);
+	hwgroup->rq = NULL;
+	__elv_add_request(drive->queue, rq, ELEVATOR_INSERT_FRONT, 1);
 	__generic_unplug_device(drive->queue);
-	/* the queue is stopped so it won't be plugged+unplugged */
-	if (blk_pm_resume_request(rq))
-		do_ide_request(drive->queue);
 	spin_unlock_irqrestore(&ide_lock, flags);
-
-	return 0;
 }
 
 EXPORT_SYMBOL(ide_do_drive_cmd);
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index cc7991c7c252..a562df820777 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -691,7 +691,7 @@ static void idetape_queue_pc_head(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	rq->cmd_flags |= REQ_PREEMPT;
 	rq->buffer = (char *) pc;
 	rq->rq_disk = tape->disk;
-	(void) ide_do_drive_cmd(drive, rq, ide_preempt);
+	ide_do_drive_cmd(drive, rq);
 }
 
 /*
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 58e30efe7a74..569ffde6d047 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -245,7 +245,8 @@ static int idescsi_check_condition(ide_drive_t *drive,
 		ide_scsi_hex_dump(pc->c, 6);
 	}
 	rq->rq_disk = scsi->disk;
-	return ide_do_drive_cmd(drive, rq, ide_preempt);
+	ide_do_drive_cmd(drive, rq);
+	return 0;
 }
 
 static int idescsi_end_request(ide_drive_t *, int, int);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index d8c86f0362c4..04267dc1edf2 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -851,17 +851,7 @@ int ide_wait_stat(ide_startstop_t *, ide_drive_t *, u8, u8, unsigned long);
 
 extern ide_startstop_t ide_do_reset (ide_drive_t *);
 
-/*
- * "action" parameter type for ide_do_drive_cmd() below.
- */
-typedef enum {
-	ide_wait,	/* insert rq at end of list, and wait for it */
-	ide_preempt,	/* insert rq in front of current request */
-	ide_head_wait,	/* insert rq in front of current request and wait for it */
-	ide_end		/* insert rq at end of list, but don't wait for it */
-} ide_action_t;
-
-extern int ide_do_drive_cmd(ide_drive_t *, struct request *, ide_action_t);
+extern void ide_do_drive_cmd(ide_drive_t *, struct request *);
 
 extern void ide_end_drive_cmd(ide_drive_t *, u8, u8);
 
-- 
cgit v1.2.3


From 92f5daff2b8439fa4c57c57f47823ffc459c3bd9 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:55 +0200
Subject: ide-tape: make pc->idetape_callback void

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-tape.c | 13 +++++++------
 include/linux/ide.h    |  2 +-
 2 files changed, 8 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index d387aaf0eb39..88d26efdf844 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -619,7 +619,7 @@ static int idetape_end_request(ide_drive_t *drive, int uptodate, int nr_sects)
 	return 0;
 }
 
-static ide_startstop_t ide_tape_callback(ide_drive_t *drive)
+static void ide_tape_callback(ide_drive_t *drive)
 {
 	idetape_tape_t *tape = drive->driver_data;
 	struct ide_atapi_pc *pc = tape->pc;
@@ -675,8 +675,6 @@ static ide_startstop_t ide_tape_callback(ide_drive_t *drive)
 	}
 
 	idetape_end_request(drive, uptodate, 0);
-
-	return ide_stopped;
 }
 
 static void idetape_init_pc(struct ide_atapi_pc *pc)
@@ -843,7 +841,8 @@ static ide_startstop_t idetape_pc_intr(ide_drive_t *drive)
 		if (tape->failed_pc == pc)
 			tape->failed_pc = NULL;
 		/* Command finished - Call the callback function */
-		return pc->idetape_callback(drive);
+		pc->idetape_callback(drive);
+		return ide_stopped;
 	}
 
 	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
@@ -1035,7 +1034,8 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 			pc->error = IDETAPE_ERROR_GENERAL;
 		}
 		tape->failed_pc = NULL;
-		return pc->idetape_callback(drive);
+		pc->idetape_callback(drive);
+		return ide_stopped;
 	}
 	debug_log(DBG_SENSE, "Retry #%d, cmd = %02X\n", pc->retries, pc->c[0]);
 
@@ -1120,7 +1120,8 @@ static ide_startstop_t idetape_media_access_finished(ide_drive_t *drive)
 		pc->error = IDETAPE_ERROR_GENERAL;
 		tape->failed_pc = NULL;
 	}
-	return pc->idetape_callback(drive);
+	pc->idetape_callback(drive);
+	return ide_stopped;
 }
 
 static void idetape_create_read_cmd(idetape_tape_t *tape,
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 04267dc1edf2..8936b21a7030 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -641,7 +641,7 @@ struct ide_atapi_pc {
 	 */
 	u8 pc_buf[256];
 	void (*idefloppy_callback) (ide_drive_t *);
-	ide_startstop_t (*idetape_callback) (ide_drive_t *);
+	void (*idetape_callback) (ide_drive_t *);
 
 	/* idetape only */
 	struct idetape_bh *bh;
-- 
cgit v1.2.3


From 1b06e92aa03018e4b3ba281e03a7711d9b71a998 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:56 +0200
Subject: ide-{floppy,tape}: merge pc->idefloppy_callback and
 pc->idetape_callback

Merge pc->idefloppy_callback and pc->idetape_callback into pc->callback.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 6 +++---
 drivers/ide/ide-tape.c   | 8 ++++----
 include/linux/ide.h      | 4 ++--
 3 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 2058a6f3f331..a9f3127a74ed 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -354,7 +354,7 @@ static void idefloppy_init_pc(struct ide_atapi_pc *pc)
 	pc->req_xfer = 0;
 	pc->buf = pc->pc_buf;
 	pc->buf_size = IDEFLOPPY_PC_BUFFER_SIZE;
-	pc->idefloppy_callback = &ide_floppy_callback;
+	pc->callback = ide_floppy_callback;
 }
 
 static void idefloppy_create_request_sense_cmd(struct ide_atapi_pc *pc)
@@ -438,7 +438,7 @@ static ide_startstop_t idefloppy_pc_intr(ide_drive_t *drive)
 		if (floppy->failed_pc == pc)
 			floppy->failed_pc = NULL;
 		/* Command finished - Call the callback function */
-		pc->idefloppy_callback(drive);
+		pc->callback(drive);
 		return ide_stopped;
 	}
 
@@ -612,7 +612,7 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 		pc->error = IDEFLOPPY_ERROR_GENERAL;
 
 		floppy->failed_pc = NULL;
-		pc->idefloppy_callback(drive);
+		pc->callback(drive);
 		return ide_stopped;
 	}
 
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 88d26efdf844..ce9b6d327528 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -687,7 +687,7 @@ static void idetape_init_pc(struct ide_atapi_pc *pc)
 	pc->buf_size = IDETAPE_PC_BUFFER_SIZE;
 	pc->bh = NULL;
 	pc->b_data = NULL;
-	pc->idetape_callback = ide_tape_callback;
+	pc->callback = ide_tape_callback;
 }
 
 static void idetape_create_request_sense_cmd(struct ide_atapi_pc *pc)
@@ -841,7 +841,7 @@ static ide_startstop_t idetape_pc_intr(ide_drive_t *drive)
 		if (tape->failed_pc == pc)
 			tape->failed_pc = NULL;
 		/* Command finished - Call the callback function */
-		pc->idetape_callback(drive);
+		pc->callback(drive);
 		return ide_stopped;
 	}
 
@@ -1034,7 +1034,7 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 			pc->error = IDETAPE_ERROR_GENERAL;
 		}
 		tape->failed_pc = NULL;
-		pc->idetape_callback(drive);
+		pc->callback(drive);
 		return ide_stopped;
 	}
 	debug_log(DBG_SENSE, "Retry #%d, cmd = %02X\n", pc->retries, pc->c[0]);
@@ -1120,7 +1120,7 @@ static ide_startstop_t idetape_media_access_finished(ide_drive_t *drive)
 		pc->error = IDETAPE_ERROR_GENERAL;
 		tape->failed_pc = NULL;
 	}
-	pc->idetape_callback(drive);
+	pc->callback(drive);
 	return ide_stopped;
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 8936b21a7030..f079456adfdb 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -640,8 +640,8 @@ struct ide_atapi_pc {
 	 * to change/removal later.
 	 */
 	u8 pc_buf[256];
-	void (*idefloppy_callback) (ide_drive_t *);
-	void (*idetape_callback) (ide_drive_t *);
+
+	void (*callback)(ide_drive_t *);
 
 	/* idetape only */
 	struct idetape_bh *bh;
-- 
cgit v1.2.3


From 5e3310958204912f3f00be2592c945fbc37db6ae Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:56 +0200
Subject: ide-{floppy,tape}: PC_FLAG_DMA_RECOMMENDED -> PC_FLAG_DMA_OK

* Use PC_FLAG_DMA_OK flag instead of PC_FLAG_DMA_RECOMMENDED one.

* Remove no longer used PC_FLAG_DMA_RECOMMENDED flag.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 6 +++---
 drivers/ide/ide-tape.c   | 6 +++---
 include/linux/ide.h      | 9 ++++-----
 3 files changed, 10 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index a9f3127a74ed..dbefe35c1396 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -630,7 +630,7 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 	}
 	dma = 0;
 
-	if ((pc->flags & PC_FLAG_DMA_RECOMMENDED) && drive->using_dma)
+	if ((pc->flags & PC_FLAG_DMA_OK) && drive->using_dma)
 		dma = !hwif->dma_ops->dma_setup(drive);
 
 	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma);
@@ -755,7 +755,7 @@ static void idefloppy_create_rw_cmd(idefloppy_floppy_t *floppy,
 		pc->flags |= PC_FLAG_WRITING;
 	pc->buf = NULL;
 	pc->req_xfer = pc->buf_size = blocks * floppy->block_size;
-	pc->flags |= PC_FLAG_DMA_RECOMMENDED;
+	pc->flags |= PC_FLAG_DMA_OK;
 }
 
 static void idefloppy_blockpc_cmd(idefloppy_floppy_t *floppy,
@@ -769,7 +769,7 @@ static void idefloppy_blockpc_cmd(idefloppy_floppy_t *floppy,
 		pc->flags |= PC_FLAG_WRITING;
 	pc->buf = rq->data;
 	if (rq->bio)
-		pc->flags |= PC_FLAG_DMA_RECOMMENDED;
+		pc->flags |= PC_FLAG_DMA_OK;
 	/*
 	 * possibly problematic, doesn't look like ide-floppy correctly
 	 * handled scattered requests if dma fails...
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index ce9b6d327528..e8a5852fa2d4 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1050,7 +1050,7 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 		pc->flags &= ~PC_FLAG_DMA_ERROR;
 		ide_dma_off(drive);
 	}
-	if ((pc->flags & PC_FLAG_DMA_RECOMMENDED) && drive->using_dma)
+	if ((pc->flags & PC_FLAG_DMA_OK) && drive->using_dma)
 		dma_ok = !hwif->dma_ops->dma_setup(drive);
 
 	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma_ok);
@@ -1138,7 +1138,7 @@ static void idetape_create_read_cmd(idetape_tape_t *tape,
 	pc->buf_size = length * tape->blk_size;
 	pc->req_xfer = pc->buf_size;
 	if (pc->req_xfer == tape->buffer_size)
-		pc->flags |= PC_FLAG_DMA_RECOMMENDED;
+		pc->flags |= PC_FLAG_DMA_OK;
 }
 
 static void idetape_create_write_cmd(idetape_tape_t *tape,
@@ -1157,7 +1157,7 @@ static void idetape_create_write_cmd(idetape_tape_t *tape,
 	pc->buf_size = length * tape->blk_size;
 	pc->req_xfer = pc->buf_size;
 	if (pc->req_xfer == tape->buffer_size)
-		pc->flags |= PC_FLAG_DMA_RECOMMENDED;
+		pc->flags |= PC_FLAG_DMA_OK;
 }
 
 static ide_startstop_t idetape_do_request(ide_drive_t *drive,
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f079456adfdb..63cee2947f60 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -602,12 +602,11 @@ enum {
 	PC_FLAG_SUPPRESS_ERROR		= (1 << 1),
 	PC_FLAG_WAIT_FOR_DSC		= (1 << 2),
 	PC_FLAG_DMA_OK			= (1 << 3),
-	PC_FLAG_DMA_RECOMMENDED		= (1 << 4),
-	PC_FLAG_DMA_IN_PROGRESS		= (1 << 5),
-	PC_FLAG_DMA_ERROR		= (1 << 6),
-	PC_FLAG_WRITING			= (1 << 7),
+	PC_FLAG_DMA_IN_PROGRESS		= (1 << 4),
+	PC_FLAG_DMA_ERROR		= (1 << 5),
+	PC_FLAG_WRITING			= (1 << 6),
 	/* command timed out */
-	PC_FLAG_TIMEDOUT		= (1 << 8),
+	PC_FLAG_TIMEDOUT		= (1 << 7),
 };
 
 struct ide_atapi_pc {
-- 
cgit v1.2.3


From 5d41893c0f9caf94b449eada0279a08c86f0212e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:57 +0200
Subject: ide: add PC_FLAG_ZIP_DRIVE pc flag

Add PC_FLAG_ZIP_DRIVE pc flag, set it in idefloppy_do_request()
and check for it (instead of checking for IDEFLOPPY_FLAG_ZIP_DRIVE)
in idefloppy_transfer_pc().  This is a preparation for adding
generic ide_transfer_pc() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c | 8 ++++++--
 include/linux/ide.h      | 1 +
 2 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 1df6a3143596..cff90c4b217e 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -559,7 +559,7 @@ static ide_startstop_t idefloppy_transfer_pc1(ide_drive_t *drive)
 	 * 40 and 50msec work well. idefloppy_pc_intr will not be actually
 	 * used until after the packet is moved in about 50 msec.
 	 */
-	if (floppy->flags & IDEFLOPPY_FLAG_ZIP_DRIVE) {
+	if (pc->flags & PC_FLAG_ZIP_DRIVE) {
 		timeout = floppy->ticks;
 		expiry = &idefloppy_transfer_pc2;
 	} else {
@@ -575,7 +575,7 @@ static ide_startstop_t idefloppy_transfer_pc1(ide_drive_t *drive)
 		hwif->dma_ops->dma_start(drive);
 	}
 
-	if ((floppy->flags & IDEFLOPPY_FLAG_ZIP_DRIVE) == 0)
+	if ((pc->flags & PC_FLAG_ZIP_DRIVE) == 0)
 		/* Send the actual packet */
 		hwif->output_data(drive, NULL, floppy->pc->c, 12);
 
@@ -826,7 +826,11 @@ static ide_startstop_t idefloppy_do_request(ide_drive_t *drive,
 		return ide_stopped;
 	}
 
+	if (floppy->flags & IDEFLOPPY_FLAG_ZIP_DRIVE)
+		pc->flags |= PC_FLAG_ZIP_DRIVE;
+
 	pc->rq = rq;
+
 	return idefloppy_issue_pc(drive, pc);
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 63cee2947f60..89feaea9e20b 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -607,6 +607,7 @@ enum {
 	PC_FLAG_WRITING			= (1 << 6),
 	/* command timed out */
 	PC_FLAG_TIMEDOUT		= (1 << 7),
+	PC_FLAG_ZIP_DRIVE		= (1 << 8),
 };
 
 struct ide_atapi_pc {
-- 
cgit v1.2.3


From 594c16d8dd54cd7b1c5ef1ec3ac0f6bf34301dad Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:58 +0200
Subject: ide: add ide_transfer_pc() helper

* Add ide-atapi.c file for generic ATAPI support together with
  CONFIG_IDE_ATAPI config option.

* Add generic ide_transfer_pc() helper to ide-atapi.c and then
  convert ide-{floppy,tape,scsi} device drivers to use it.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Kconfig      |  6 +++++
 drivers/ide/Makefile     |  1 +
 drivers/ide/ide-atapi.c  | 70 ++++++++++++++++++++++++++++++++++++++++++++++++
 drivers/ide/ide-floppy.c | 28 +------------------
 drivers/ide/ide-tape.c   | 56 ++------------------------------------
 drivers/scsi/ide-scsi.c  | 30 ++-------------------
 include/linux/ide.h      |  3 +++
 7 files changed, 85 insertions(+), 109 deletions(-)
 create mode 100644 drivers/ide/ide-atapi.c

(limited to 'include/linux')

diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 1607536ff5fb..cf707c8f08d4 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -98,6 +98,9 @@ if BLK_DEV_IDE
 
 comment "Please see Documentation/ide/ide.txt for help/info on IDE drives"
 
+config IDE_ATAPI
+	bool
+
 config BLK_DEV_IDE_SATA
 	bool "Support for SATA (deprecated; conflicts with libata SATA driver)"
 	default n
@@ -201,6 +204,7 @@ config BLK_DEV_IDECD_VERBOSE_ERRORS
 
 config BLK_DEV_IDETAPE
 	tristate "Include IDE/ATAPI TAPE support"
+	select IDE_ATAPI
 	help
 	  If you have an IDE tape drive using the ATAPI protocol, say Y.
 	  ATAPI is a newer protocol used by IDE tape and CD-ROM drives,
@@ -223,6 +227,7 @@ config BLK_DEV_IDETAPE
 
 config BLK_DEV_IDEFLOPPY
 	tristate "Include IDE/ATAPI FLOPPY support"
+	select IDE_ATAPI
 	---help---
 	  If you have an IDE floppy drive which uses the ATAPI protocol,
 	  answer Y.  ATAPI is a newer protocol used by IDE CD-ROM/tape/floppy
@@ -246,6 +251,7 @@ config BLK_DEV_IDEFLOPPY
 config BLK_DEV_IDESCSI
 	tristate "SCSI emulation support"
 	depends on SCSI
+	select IDE_ATAPI
 	---help---
 	  WARNING: ide-scsi is no longer needed for cd writing applications!
 	  The 2.6 kernel supports direct writing to ide-cd, which eliminates
diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index f94b679b611e..a2b3f84d710d 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -14,6 +14,7 @@ EXTRA_CFLAGS				+= -Idrivers/ide
 ide-core-y += ide.o ide-io.o ide-iops.o ide-lib.o ide-probe.o ide-taskfile.o
 
 # core IDE code
+ide-core-$(CONFIG_IDE_ATAPI)		+= ide-atapi.o
 ide-core-$(CONFIG_BLK_DEV_IDEPCI)	+= setup-pci.o
 ide-core-$(CONFIG_BLK_DEV_IDEDMA)	+= ide-dma.o
 ide-core-$(CONFIG_IDE_PROC_FS)		+= ide-proc.o
diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
new file mode 100644
index 000000000000..25939bc60402
--- /dev/null
+++ b/drivers/ide/ide-atapi.c
@@ -0,0 +1,70 @@
+/*
+ * ATAPI support.
+ */
+
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/ide.h>
+
+static u8 ide_wait_ireason(ide_drive_t *drive, u8 ireason)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	int retries = 100;
+
+	while (retries-- && ((ireason & CD) == 0 || (ireason & IO))) {
+		printk(KERN_ERR "%s: (IO,CoD != (0,1) while issuing "
+				"a packet command, retrying\n", drive->name);
+		udelay(100);
+		ireason = hwif->INB(hwif->io_ports.nsect_addr);
+		if (retries == 0) {
+			printk(KERN_ERR "%s: (IO,CoD != (0,1) while issuing "
+					"a packet command, ignoring\n",
+					drive->name);
+			ireason |= CD;
+			ireason &= ~IO;
+		}
+	}
+
+	return ireason;
+}
+
+ide_startstop_t ide_transfer_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
+				ide_handler_t *handler, unsigned int timeout,
+				ide_expiry_t *expiry)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	ide_startstop_t startstop;
+	u8 ireason;
+
+	if (ide_wait_stat(&startstop, drive, DRQ_STAT, BUSY_STAT, WAIT_READY)) {
+		printk(KERN_ERR "%s: Strange, packet command initiated yet "
+				"DRQ isn't asserted\n", drive->name);
+		return startstop;
+	}
+
+	ireason = hwif->INB(hwif->io_ports.nsect_addr);
+	if (drive->media == ide_tape && !drive->scsi)
+		ireason = ide_wait_ireason(drive, ireason);
+
+	if ((ireason & CD) == 0 || (ireason & IO)) {
+		printk(KERN_ERR "%s: (IO,CoD) != (0,1) while issuing "
+				"a packet command\n", drive->name);
+		return ide_do_reset(drive);
+	}
+
+	/* Set the interrupt routine */
+	ide_set_handler(drive, handler, timeout, expiry);
+
+	/* Begin DMA, if necessary */
+	if (pc->flags & PC_FLAG_DMA_OK) {
+		pc->flags |= PC_FLAG_DMA_IN_PROGRESS;
+		hwif->dma_ops->dma_start(drive);
+	}
+
+	/* Send the actual packet */
+	if ((pc->flags & PC_FLAG_ZIP_DRIVE) == 0)
+		hwif->output_data(drive, NULL, pc->c, 12);
+
+	return ide_started;
+}
+EXPORT_SYMBOL_GPL(ide_transfer_pc);
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index a7c138dc324c..e7a1025c03c4 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -532,25 +532,11 @@ static int idefloppy_transfer_pc2(ide_drive_t *drive)
 
 static ide_startstop_t idefloppy_transfer_pc1(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = drive->hwif;
 	idefloppy_floppy_t *floppy = drive->driver_data;
 	struct ide_atapi_pc *pc = floppy->pc;
 	ide_expiry_t *expiry;
 	unsigned int timeout;
-	ide_startstop_t startstop;
-	u8 ireason;
 
-	if (ide_wait_stat(&startstop, drive, DRQ_STAT, BUSY_STAT, WAIT_READY)) {
-		printk(KERN_ERR "%s: Strange, packet command initiated yet "
-				"DRQ isn't asserted\n", drive->name);
-		return startstop;
-	}
-	ireason = hwif->INB(hwif->io_ports.nsect_addr);
-	if ((ireason & CD) == 0 || (ireason & IO)) {
-		printk(KERN_ERR "%s: (IO,CoD) != (0,1) while issuing "
-				"a packet command\n", drive->name);
-		return ide_do_reset(drive);
-	}
 	/*
 	 * The following delay solves a problem with ATAPI Zip 100 drives
 	 * where the Busy flag was apparently being deasserted before the
@@ -567,19 +553,7 @@ static ide_startstop_t idefloppy_transfer_pc1(ide_drive_t *drive)
 		expiry = NULL;
 	}
 
-	ide_set_handler(drive, &idefloppy_pc_intr, timeout, expiry);
-
-	/* Begin DMA, if necessary */
-	if (pc->flags & PC_FLAG_DMA_OK) {
-		pc->flags |= PC_FLAG_DMA_IN_PROGRESS;
-		hwif->dma_ops->dma_start(drive);
-	}
-
-	if ((pc->flags & PC_FLAG_ZIP_DRIVE) == 0)
-		/* Send the actual packet */
-		hwif->output_data(drive, NULL, floppy->pc->c, 12);
-
-	return ide_started;
+	return ide_transfer_pc(drive, pc, idefloppy_pc_intr, timeout, expiry);
 }
 
 static void ide_floppy_report_error(idefloppy_floppy_t *floppy,
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 2a362138f973..5adc2c9ae418 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -947,64 +947,12 @@ static ide_startstop_t idetape_pc_intr(ide_drive_t *drive)
  * again, the callback function will be called and then we will handle the next
  * request.
  */
-
-static u8 ide_tape_wait_ireason(ide_drive_t *drive, u8 ireason)
-{
-	ide_hwif_t *hwif = drive->hwif;
-	int retries = 100;
-
-	while (retries-- && ((ireason & CD) == 0 || (ireason & IO))) {
-		printk(KERN_ERR "%s: (IO,CoD != (0,1) while issuing "
-				"a packet command, retrying\n", drive->name);
-		udelay(100);
-		ireason = hwif->INB(hwif->io_ports.nsect_addr);
-		if (retries == 0) {
-			printk(KERN_ERR "%s: (IO,CoD != (0,1) while issuing "
-					"a packet command, ignoring\n",
-					drive->name);
-			ireason |= CD;
-			ireason &= ~IO;
-		}
-	}
-
-	return ireason;
-}
-
 static ide_startstop_t idetape_transfer_pc(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
-	struct ide_atapi_pc *pc = tape->pc;
-	ide_startstop_t startstop;
-	u8 ireason;
-
-	if (ide_wait_stat(&startstop, drive, DRQ_STAT, BUSY_STAT, WAIT_READY)) {
-		printk(KERN_ERR "%s: Strange, packet command initiated yet "
-				"DRQ isn't asserted\n", drive->name);
-		return startstop;
-	}
-
-	ireason = hwif->INB(hwif->io_ports.nsect_addr);
-	ireason = ide_tape_wait_ireason(drive, ireason);
 
-	if ((ireason & CD) == 0 || (ireason & IO)) {
-		printk(KERN_ERR "%s: (IO,CoD) != (0,1) while issuing "
-				"a packet command\n", drive->name);
-		return ide_do_reset(drive);
-	}
-	/* Set the interrupt routine */
-	ide_set_handler(drive, &idetape_pc_intr, IDETAPE_WAIT_CMD, NULL);
-
-	/* Begin DMA, if necessary */
-	if (pc->flags & PC_FLAG_DMA_OK) {
-		pc->flags |= PC_FLAG_DMA_IN_PROGRESS;
-		hwif->dma_ops->dma_start(drive);
-	}
-
-	/* Send the actual packet */
-	hwif->output_data(drive, NULL, pc->c, 12);
-
-	return ide_started;
+	return ide_transfer_pc(drive, tape->pc, idetape_pc_intr,
+			       IDETAPE_WAIT_CMD, NULL);
 }
 
 static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index c9fdf60c9dcf..d41348f2245e 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -453,36 +453,10 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 
 static ide_startstop_t idescsi_transfer_pc(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = drive->hwif;
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-	struct ide_atapi_pc *pc = scsi->pc;
-	ide_startstop_t startstop;
-	u8 ireason;
-
-	if (ide_wait_stat(&startstop,drive,DRQ_STAT,BUSY_STAT,WAIT_READY)) {
-		printk(KERN_ERR "%s: Strange, packet command initiated yet "
-				"DRQ isn't asserted\n", drive->name);
-		return startstop;
-	}
-	ireason = hwif->INB(hwif->io_ports.nsect_addr);
-	if ((ireason & CD) == 0 || (ireason & IO)) {
-		printk(KERN_ERR "%s: (IO,CoD) != (0,1) while issuing "
-				"a packet command\n", drive->name);
-		return ide_do_reset (drive);
-	}
 
-	/* Set the interrupt routine */
-	ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
-
-	if (pc->flags & PC_FLAG_DMA_OK) {
-		pc->flags |= PC_FLAG_DMA_IN_PROGRESS;
-		hwif->dma_ops->dma_start(drive);
-	}
-
-	/* Send the actual packet */
-	hwif->output_data(drive, NULL, scsi->pc->c, 12);
-
-	return ide_started;
+	return ide_transfer_pc(drive, scsi->pc, idescsi_pc_intr,
+			       get_timeout(scsi->pc), idescsi_expiry);
 }
 
 static inline int idescsi_set_direction(struct ide_atapi_pc *pc)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 89feaea9e20b..bed3c58798ae 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -967,6 +967,9 @@ extern int drive_is_ready(ide_drive_t *);
 
 void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
 
+ide_startstop_t ide_transfer_pc(ide_drive_t *, struct ide_atapi_pc *,
+				ide_handler_t *, unsigned int, ide_expiry_t *);
+
 ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
 void task_end_request(ide_drive_t *, struct request *, u8);
-- 
cgit v1.2.3


From 28c7214bd8c2bbd4873b8f1e7f58d86d3731124f Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:21:59 +0200
Subject: ide: add PC_FLAG_DRQ_INTERRUPT pc flag

Add PC_FLAG_DRQ_INTERRUPT pc flag, set it in ide*_do_request()
and check for it (instead of checking for IDE*_FLAG_DRQ_INTERRUPT)
in ide*_issue_pc().  This is a preparation for adding generic
ide_issue_pc() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-floppy.c |  5 ++++-
 drivers/ide/ide-tape.c   | 11 ++++++++---
 drivers/scsi/ide-scsi.c  |  6 +++++-
 include/linux/ide.h      |  1 +
 4 files changed, 18 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index e7a1025c03c4..13f650fa2125 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -619,7 +619,7 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 
 	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma);
 
-	if (floppy->flags & IDEFLOPPY_FLAG_DRQ_INTERRUPT) {
+	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
 		/* Issue the packet command */
 		ide_execute_command(drive, WIN_PACKETCMD,
 				&idefloppy_transfer_pc1,
@@ -800,6 +800,9 @@ static ide_startstop_t idefloppy_do_request(ide_drive_t *drive,
 		return ide_stopped;
 	}
 
+	if (floppy->flags & IDEFLOPPY_FLAG_DRQ_INTERRUPT)
+		pc->flags |= PC_FLAG_DRQ_INTERRUPT;
+
 	if (floppy->flags & IDEFLOPPY_FLAG_ZIP_DRIVE)
 		pc->flags |= PC_FLAG_ZIP_DRIVE;
 
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 5adc2c9ae418..cba18a675506 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -1020,7 +1020,7 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 
 	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma_ok);
 
-	if (test_bit(IDETAPE_FLAG_DRQ_INTERRUPT, &tape->flags)) {
+	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
 		ide_execute_command(drive, WIN_PACKETCMD, &idetape_transfer_pc,
 				    IDETAPE_WAIT_CMD, NULL);
 		return ide_started;
@@ -1143,8 +1143,10 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	}
 
 	/* Retry a failed packet command */
-	if (tape->failed_pc && tape->pc->c[0] == REQUEST_SENSE)
-		return idetape_issue_pc(drive, tape->failed_pc);
+	if (tape->failed_pc && tape->pc->c[0] == REQUEST_SENSE) {
+		pc = tape->failed_pc;
+		goto out;
+	}
 
 	if (postponed_rq != NULL)
 		if (rq != postponed_rq) {
@@ -1216,6 +1218,9 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	}
 	BUG();
 out:
+	if (test_bit(IDETAPE_FLAG_DRQ_INTERRUPT, &tape->flags))
+		pc->flags |= PC_FLAG_DRQ_INTERRUPT;
+
 	return idetape_issue_pc(drive, pc);
 }
 
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 1d261298d61a..b7c5e8391575 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -525,7 +525,7 @@ static ide_startstop_t idescsi_issue_pc(ide_drive_t *drive,
 
 	ide_pktcmd_tf_load(drive, 0, bcount, dma);
 
-	if (test_bit(IDESCSI_DRQ_INTERRUPT, &scsi->flags)) {
+	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
 		ide_execute_command(drive, WIN_PACKETCMD, &idescsi_transfer_pc,
 				    get_timeout(pc), idescsi_expiry);
 		return ide_started;
@@ -548,6 +548,10 @@ static ide_startstop_t idescsi_do_request (ide_drive_t *drive, struct request *r
 
 	if (blk_sense_request(rq) || blk_special_request(rq)) {
 		struct ide_atapi_pc *pc = (struct ide_atapi_pc *)rq->special;
+		idescsi_scsi_t *scsi = drive_to_idescsi(drive);
+
+		if (test_bit(IDESCSI_DRQ_INTERRUPT, &scsi->flags))
+			pc->flags |= PC_FLAG_DRQ_INTERRUPT;
 
 		if (drive->using_dma && !idescsi_map_sg(drive, pc))
 			pc->flags |= PC_FLAG_DMA_OK;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index bed3c58798ae..c2274ad44b2e 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -608,6 +608,7 @@ enum {
 	/* command timed out */
 	PC_FLAG_TIMEDOUT		= (1 << 7),
 	PC_FLAG_ZIP_DRIVE		= (1 << 8),
+	PC_FLAG_DRQ_INTERRUPT		= (1 << 9),
 };
 
 struct ide_atapi_pc {
-- 
cgit v1.2.3


From 6bf1641ca1c7554f0da54aaf89788731b541bacc Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:22:00 +0200
Subject: ide: add ide_issue_pc() helper

Add generic ide_issue_pc() helper to ide-atapi.c and then
convert ide-{floppy,tape,scsi} device drivers to use it.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c  | 49 ++++++++++++++++++++++++++++++++++++++++++++++++
 drivers/ide/ide-floppy.c | 35 ++--------------------------------
 drivers/ide/ide-tape.c   | 30 ++---------------------------
 drivers/scsi/ide-scsi.c  | 30 ++---------------------------
 include/linux/ide.h      |  2 ++
 5 files changed, 57 insertions(+), 89 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 25939bc60402..932a83abaf06 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -68,3 +68,52 @@ ide_startstop_t ide_transfer_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	return ide_started;
 }
 EXPORT_SYMBOL_GPL(ide_transfer_pc);
+
+ide_startstop_t ide_issue_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
+			     ide_handler_t *handler, unsigned int timeout,
+			     ide_expiry_t *expiry)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	u16 bcount;
+	u8 dma = 0;
+
+	/* We haven't transferred any data yet */
+	pc->xferred = 0;
+	pc->cur_pos = pc->buf;
+
+	/* Request to transfer the entire buffer at once */
+	if (drive->media == ide_tape && !drive->scsi)
+		bcount = pc->req_xfer;
+	else
+		bcount = min(pc->req_xfer, 63 * 1024);
+
+	if (pc->flags & PC_FLAG_DMA_ERROR) {
+		pc->flags &= ~PC_FLAG_DMA_ERROR;
+		ide_dma_off(drive);
+	}
+
+	if ((pc->flags & PC_FLAG_DMA_OK) && drive->using_dma) {
+		if (drive->scsi)
+			hwif->sg_mapped = 1;
+		dma = !hwif->dma_ops->dma_setup(drive);
+		if (drive->scsi)
+			hwif->sg_mapped = 0;
+	}
+
+	if (!dma)
+		pc->flags &= ~PC_FLAG_DMA_OK;
+
+	ide_pktcmd_tf_load(drive, drive->scsi ? 0 : IDE_TFLAG_OUT_DEVICE,
+			   bcount, dma);
+
+	/* Issue the packet command */
+	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
+		ide_execute_command(drive, WIN_PACKETCMD, handler,
+				    timeout, NULL);
+		return ide_started;
+	} else {
+		ide_execute_pkt_cmd(drive);
+		return (*handler)(drive);
+	}
+}
+EXPORT_SYMBOL_GPL(ide_issue_pc);
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 13f650fa2125..e658aafc51da 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -576,9 +576,6 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 		struct ide_atapi_pc *pc)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
-	ide_hwif_t *hwif = drive->hwif;
-	u16 bcount;
-	u8 dma;
 
 	if (floppy->failed_pc == NULL &&
 	    pc->c[0] != GPCMD_REQUEST_SENSE)
@@ -600,37 +597,9 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 	debug_log("Retry number - %d\n", pc->retries);
 
 	pc->retries++;
-	/* We haven't transferred any data yet */
-	pc->xferred = 0;
-	pc->cur_pos = pc->buf;
-	bcount = min(pc->req_xfer, 63 * 1024);
-
-	if (pc->flags & PC_FLAG_DMA_ERROR) {
-		pc->flags &= ~PC_FLAG_DMA_ERROR;
-		ide_dma_off(drive);
-	}
-	dma = 0;
 
-	if ((pc->flags & PC_FLAG_DMA_OK) && drive->using_dma)
-		dma = !hwif->dma_ops->dma_setup(drive);
-
-	if (!dma)
-		pc->flags &= ~PC_FLAG_DMA_OK;
-
-	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma);
-
-	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
-		/* Issue the packet command */
-		ide_execute_command(drive, WIN_PACKETCMD,
-				&idefloppy_transfer_pc1,
-				IDEFLOPPY_WAIT_CMD,
-				NULL);
-		return ide_started;
-	} else {
-		/* Issue the packet command */
-		ide_execute_pkt_cmd(drive);
-		return idefloppy_transfer_pc1(drive);
-	}
+	return ide_issue_pc(drive, pc, idefloppy_transfer_pc1,
+			    IDEFLOPPY_WAIT_CMD, NULL);
 }
 
 static void idefloppy_create_prevent_cmd(struct ide_atapi_pc *pc, int prevent)
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index cba18a675506..7907a1e41918 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -958,10 +958,7 @@ static ide_startstop_t idetape_transfer_pc(ide_drive_t *drive)
 static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 		struct ide_atapi_pc *pc)
 {
-	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
-	int dma_ok = 0;
-	u16 bcount;
 
 	if (tape->pc->c[0] == REQUEST_SENSE &&
 	    pc->c[0] == REQUEST_SENSE) {
@@ -1002,32 +999,9 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 	debug_log(DBG_SENSE, "Retry #%d, cmd = %02X\n", pc->retries, pc->c[0]);
 
 	pc->retries++;
-	/* We haven't transferred any data yet */
-	pc->xferred = 0;
-	pc->cur_pos = pc->buf;
-	/* Request to transfer the entire buffer at once */
-	bcount = pc->req_xfer;
-
-	if (pc->flags & PC_FLAG_DMA_ERROR) {
-		pc->flags &= ~PC_FLAG_DMA_ERROR;
-		ide_dma_off(drive);
-	}
-	if ((pc->flags & PC_FLAG_DMA_OK) && drive->using_dma)
-		dma_ok = !hwif->dma_ops->dma_setup(drive);
-
-	if (!dma_ok)
-		pc->flags &= ~PC_FLAG_DMA_OK;
-
-	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_DEVICE, bcount, dma_ok);
 
-	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
-		ide_execute_command(drive, WIN_PACKETCMD, &idetape_transfer_pc,
-				    IDETAPE_WAIT_CMD, NULL);
-		return ide_started;
-	} else {
-		ide_execute_pkt_cmd(drive);
-		return idetape_transfer_pc(drive);
-	}
+	return ide_issue_pc(drive, pc, idetape_transfer_pc,
+			    IDETAPE_WAIT_CMD, NULL);
 }
 
 /* A mode sense command is used to "sense" tape parameters. */
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index b7c5e8391575..32415466fbfe 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -502,38 +502,12 @@ static ide_startstop_t idescsi_issue_pc(ide_drive_t *drive,
 		struct ide_atapi_pc *pc)
 {
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-	ide_hwif_t *hwif = drive->hwif;
-	u16 bcount;
-	u8 dma = 0;
 
 	/* Set the current packet command */
 	scsi->pc = pc;
-	/* We haven't transferred any data yet */
-	pc->xferred = 0;
-	pc->cur_pos = pc->buf;
-	/* Request to transfer the entire buffer at once */
-	bcount = min(pc->req_xfer, 63 * 1024);
-
-	if ((pc->flags & PC_FLAG_DMA_OK) && drive->using_dma) {
-		hwif->sg_mapped = 1;
-		dma = !hwif->dma_ops->dma_setup(drive);
-		hwif->sg_mapped = 0;
-	}
-
-	if (!dma)
-		pc->flags &= ~PC_FLAG_DMA_OK;
 
-	ide_pktcmd_tf_load(drive, 0, bcount, dma);
-
-	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
-		ide_execute_command(drive, WIN_PACKETCMD, &idescsi_transfer_pc,
-				    get_timeout(pc), idescsi_expiry);
-		return ide_started;
-	} else {
-		/* Issue the packet command */
-		ide_execute_pkt_cmd(drive);
-		return idescsi_transfer_pc(drive);
-	}
+	return ide_issue_pc(drive, pc, idescsi_transfer_pc,
+			    get_timeout(pc), idescsi_expiry);
 }
 
 /*
diff --git a/include/linux/ide.h b/include/linux/ide.h
index c2274ad44b2e..fee07a7edb19 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -970,6 +970,8 @@ void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
 
 ide_startstop_t ide_transfer_pc(ide_drive_t *, struct ide_atapi_pc *,
 				ide_handler_t *, unsigned int, ide_expiry_t *);
+ide_startstop_t ide_issue_pc(ide_drive_t *, struct ide_atapi_pc *,
+			     ide_handler_t *, unsigned int, ide_expiry_t *);
 
 ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
-- 
cgit v1.2.3


From 646c0cb6c430f8d3ad3769dd1518fe664ff0ce27 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 15 Jul 2008 21:22:03 +0200
Subject: ide: add ide_pc_intr() helper

* ide-tape.c: add 'drive' argument to idetape_update_buffers().

* Add generic ide_pc_intr() helper to ide-atapi.c and then
  convert ide-{floppy,tape,scsi} device drivers to use it.

* ide-tape.c: remove no longer needed DBG_PC_INTR.

There should be no functional changes caused by this patch
(unless the debugging is explicitely compiled in).

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c  | 177 +++++++++++++++++++++++++++++++++++++++++++++++
 drivers/ide/ide-floppy.c | 128 +---------------------------------
 drivers/ide/ide-tape.c   | 132 ++---------------------------------
 drivers/scsi/ide-scsi.c  | 115 +-----------------------------
 include/linux/ide.h      |   6 ++
 5 files changed, 195 insertions(+), 363 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 932a83abaf06..2802031de670 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -5,6 +5,183 @@
 #include <linux/kernel.h>
 #include <linux/delay.h>
 #include <linux/ide.h>
+#include <scsi/scsi.h>
+
+#ifdef DEBUG
+#define debug_log(fmt, args...) \
+	printk(KERN_INFO "ide: " fmt, ## args)
+#else
+#define debug_log(fmt, args...) do {} while (0)
+#endif
+
+/* TODO: unify the code thus making some arguments go away */
+ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
+	ide_handler_t *handler, unsigned int timeout, ide_expiry_t *expiry,
+	void (*update_buffers)(ide_drive_t *, struct ide_atapi_pc *),
+	void (*retry_pc)(ide_drive_t *), void (*dsc_handle)(ide_drive_t *),
+	void (*io_buffers)(ide_drive_t *, struct ide_atapi_pc *, unsigned, int))
+{
+	ide_hwif_t *hwif = drive->hwif;
+	xfer_func_t *xferfunc;
+	unsigned int temp;
+	u16 bcount;
+	u8 stat, ireason, scsi = drive->scsi;
+
+	debug_log("Enter %s - interrupt handler\n", __func__);
+
+	if (pc->flags & PC_FLAG_TIMEDOUT) {
+		pc->callback(drive);
+		return ide_stopped;
+	}
+
+	/* Clear the interrupt */
+	stat = ide_read_status(drive);
+
+	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
+		if (hwif->dma_ops->dma_end(drive) ||
+		    (drive->media == ide_tape && !scsi && (stat & ERR_STAT))) {
+			if (drive->media == ide_floppy && !scsi)
+				printk(KERN_ERR "%s: DMA %s error\n",
+					drive->name, rq_data_dir(pc->rq)
+						     ? "write" : "read");
+			pc->flags |= PC_FLAG_DMA_ERROR;
+		} else {
+			pc->xferred = pc->req_xfer;
+			if (update_buffers)
+				update_buffers(drive, pc);
+		}
+		debug_log("%s: DMA finished\n", drive->name);
+	}
+
+	/* No more interrupts */
+	if ((stat & DRQ_STAT) == 0) {
+		debug_log("Packet command completed, %d bytes transferred\n",
+			  pc->xferred);
+
+		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
+
+		local_irq_enable_in_hardirq();
+
+		if (drive->media == ide_tape && !scsi &&
+		    (stat & ERR_STAT) && pc->c[0] == REQUEST_SENSE)
+			stat &= ~ERR_STAT;
+		if ((stat & ERR_STAT) || (pc->flags & PC_FLAG_DMA_ERROR)) {
+			/* Error detected */
+			debug_log("%s: I/O error\n", drive->name);
+
+			if (drive->media != ide_tape || scsi) {
+				pc->rq->errors++;
+				if (scsi)
+					goto cmd_finished;
+			}
+
+			if (pc->c[0] == REQUEST_SENSE) {
+				printk(KERN_ERR "%s: I/O error in request sense"
+						" command\n", drive->name);
+				return ide_do_reset(drive);
+			}
+
+			debug_log("[cmd %x]: check condition\n", pc->c[0]);
+
+			/* Retry operation */
+			retry_pc(drive);
+			/* queued, but not started */
+			return ide_stopped;
+		}
+cmd_finished:
+		pc->error = 0;
+		if ((pc->flags & PC_FLAG_WAIT_FOR_DSC) &&
+		    (stat & SEEK_STAT) == 0) {
+			dsc_handle(drive);
+			return ide_stopped;
+		}
+		/* Command finished - Call the callback function */
+		pc->callback(drive);
+		return ide_stopped;
+	}
+
+	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
+		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
+		printk(KERN_ERR "%s: The device wants to issue more interrupts "
+				"in DMA mode\n", drive->name);
+		ide_dma_off(drive);
+		return ide_do_reset(drive);
+	}
+	/* Get the number of bytes to transfer on this interrupt. */
+	bcount = (hwif->INB(hwif->io_ports.lbah_addr) << 8) |
+		  hwif->INB(hwif->io_ports.lbam_addr);
+
+	ireason = hwif->INB(hwif->io_ports.nsect_addr);
+
+	if (ireason & CD) {
+		printk(KERN_ERR "%s: CoD != 0 in %s\n", drive->name, __func__);
+		return ide_do_reset(drive);
+	}
+	if (((ireason & IO) == IO) == !!(pc->flags & PC_FLAG_WRITING)) {
+		/* Hopefully, we will never get here */
+		printk(KERN_ERR "%s: We wanted to %s, but the device wants us "
+				"to %s!\n", drive->name,
+				(ireason & IO) ? "Write" : "Read",
+				(ireason & IO) ? "Read" : "Write");
+		return ide_do_reset(drive);
+	}
+	if (!(pc->flags & PC_FLAG_WRITING)) {
+		/* Reading - Check that we have enough space */
+		temp = pc->xferred + bcount;
+		if (temp > pc->req_xfer) {
+			if (temp > pc->buf_size) {
+				printk(KERN_ERR "%s: The device wants to send "
+						"us more data than expected - "
+						"discarding data\n",
+						drive->name);
+				if (scsi)
+					temp = pc->buf_size - pc->xferred;
+				else
+					temp = 0;
+				if (temp) {
+					if (pc->sg)
+						io_buffers(drive, pc, temp, 0);
+					else
+						hwif->input_data(drive, NULL,
+							pc->cur_pos, temp);
+					printk(KERN_ERR "%s: transferred %d of "
+							"%d bytes\n",
+							drive->name,
+							temp, bcount);
+				}
+				pc->xferred += temp;
+				pc->cur_pos += temp;
+				ide_pad_transfer(drive, 0, bcount - temp);
+				ide_set_handler(drive, handler, timeout,
+						expiry);
+				return ide_started;
+			}
+			debug_log("The device wants to send us more data than "
+				  "expected - allowing transfer\n");
+		}
+		xferfunc = hwif->input_data;
+	} else
+		xferfunc = hwif->output_data;
+
+	if ((drive->media == ide_floppy && !scsi && !pc->buf) ||
+	    (drive->media == ide_tape && !scsi && pc->bh) ||
+	    (scsi && pc->sg))
+		io_buffers(drive, pc, bcount, !!(pc->flags & PC_FLAG_WRITING));
+	else
+		xferfunc(drive, NULL, pc->cur_pos, bcount);
+
+	/* Update the current position */
+	pc->xferred += bcount;
+	pc->cur_pos += bcount;
+
+	debug_log("[cmd %x] transferred %d bytes on that intr.\n",
+		  pc->c[0], bcount);
+
+	/* And set the interrupt handler again */
+	ide_set_handler(drive, handler, timeout, expiry);
+	return ide_started;
+}
+EXPORT_SYMBOL_GPL(ide_pc_intr);
 
 static u8 ide_wait_ireason(ide_drive_t *drive, u8 ireason)
 {
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 70aef97fb8bc..0f3602a5efb0 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -388,132 +388,10 @@ static void idefloppy_retry_pc(ide_drive_t *drive)
 static ide_startstop_t idefloppy_pc_intr(ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
-	ide_hwif_t *hwif = drive->hwif;
-	struct ide_atapi_pc *pc = floppy->pc;
-	struct request *rq = pc->rq;
-	xfer_func_t *xferfunc;
-	unsigned int temp;
-	int dma_error = 0;
-	u16 bcount;
-	u8 stat, ireason;
-
-	debug_log("Enter %s - interrupt handler\n", __func__);
-
-	/* Clear the interrupt */
-	stat = ide_read_status(drive);
-
-	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
-		dma_error = hwif->dma_ops->dma_end(drive);
-		if (dma_error) {
-			printk(KERN_ERR "%s: DMA %s error\n", drive->name,
-					rq_data_dir(rq) ? "write" : "read");
-			pc->flags |= PC_FLAG_DMA_ERROR;
-		} else {
-			pc->xferred = pc->req_xfer;
-			idefloppy_update_buffers(drive, pc);
-		}
-		debug_log("%s: DMA finished\n", drive->name);
-	}
-
-	/* No more interrupts */
-	if ((stat & DRQ_STAT) == 0) {
-		debug_log("Packet command completed, %d bytes transferred\n",
-				pc->xferred);
-		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
-
-		local_irq_enable_in_hardirq();
-
-		if ((stat & ERR_STAT) || (pc->flags & PC_FLAG_DMA_ERROR)) {
-			/* Error detected */
-			debug_log("%s: I/O error\n", drive->name);
-			rq->errors++;
-			if (pc->c[0] == GPCMD_REQUEST_SENSE) {
-				printk(KERN_ERR "%s: I/O error in request sense"
-						" command\n", drive->name);
-				return ide_do_reset(drive);
-			}
-
-			debug_log("[cmd %x]: check condition\n", pc->c[0]);
-
-			/* Retry operation */
-			idefloppy_retry_pc(drive);
-			/* queued, but not started */
-			return ide_stopped;
-		}
-		pc->error = 0;
-		/* Command finished - Call the callback function */
-		pc->callback(drive);
-		return ide_stopped;
-	}
-
-	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
-		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
-		printk(KERN_ERR "%s: The device wants to issue more interrupts "
-				"in DMA mode\n", drive->name);
-		ide_dma_off(drive);
-		return ide_do_reset(drive);
-	}
-
-	/* Get the number of bytes to transfer */
-	bcount = (hwif->INB(hwif->io_ports.lbah_addr) << 8) |
-		  hwif->INB(hwif->io_ports.lbam_addr);
-	/* on this interrupt */
-	ireason = hwif->INB(hwif->io_ports.nsect_addr);
-
-	if (ireason & CD) {
-		printk(KERN_ERR "%s: CoD != 0 in %s\n", drive->name, __func__);
-		return ide_do_reset(drive);
-	}
-	if (((ireason & IO) == IO) == !!(pc->flags & PC_FLAG_WRITING)) {
-		/* Hopefully, we will never get here */
-		printk(KERN_ERR "%s: We wanted to %s, but the device wants us "
-				"to %s!\n", drive->name,
-				(ireason & IO) ? "Write" : "Read",
-				(ireason & IO) ? "Read" : "Write");
-		return ide_do_reset(drive);
-	}
-	if (!(pc->flags & PC_FLAG_WRITING)) {
-		/* Reading - Check that we have enough space */
-		temp = pc->xferred + bcount;
-		if (temp > pc->req_xfer) {
-			if (temp > pc->buf_size) {
-				printk(KERN_ERR "%s: The device wants to send "
-						"us more data than expected - "
-						"discarding data\n",
-						drive->name);
-				ide_pad_transfer(drive, 0, bcount);
-
-				ide_set_handler(drive,
-						&idefloppy_pc_intr,
-						IDEFLOPPY_WAIT_CMD,
-						NULL);
-				return ide_started;
-			}
-			debug_log("The device wants to send us more data than "
-				  "expected - allowing transfer\n");
-		}
-	}
-	if (pc->flags & PC_FLAG_WRITING)
-		xferfunc = hwif->output_data;
-	else
-		xferfunc = hwif->input_data;
-
-	if (pc->buf)
-		xferfunc(drive, NULL, pc->cur_pos, bcount);
-	else
-		ide_floppy_io_buffers(drive, pc, bcount,
-				      !!(pc->flags & PC_FLAG_WRITING));
-
-	/* Update the current position */
-	pc->xferred += bcount;
-	pc->cur_pos += bcount;
-
-	debug_log("[cmd %x] transferred %d bytes on that intr.\n",
-		  pc->c[0], bcount);
 
-	/* And set the interrupt handler again */
-	ide_set_handler(drive, &idefloppy_pc_intr, IDEFLOPPY_WAIT_CMD, NULL);
-	return ide_started;
+	return ide_pc_intr(drive, floppy->pc, idefloppy_pc_intr,
+			   IDEFLOPPY_WAIT_CMD, NULL, idefloppy_update_buffers,
+			   idefloppy_retry_pc, NULL, ide_floppy_io_buffers);
 }
 
 /*
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 10f2d3336286..0afa109ec99a 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -56,8 +56,6 @@ enum {
 	DBG_PROCS =		(1 << 3),
 	/* buffer alloc info (pc_stack & rq_stack) */
 	DBG_PCRQ_STACK =	(1 << 4),
-	/* IRQ handler (always log debug info if debugging is on) */
-	DBG_PC_INTR = 		(1 << 5),
 };
 
 /* define to see debug info */
@@ -66,7 +64,7 @@ enum {
 #if IDETAPE_DEBUG_LOG
 #define debug_log(lvl, fmt, args...)			\
 {							\
-	if ((lvl & DBG_PC_INTR) || (tape->debug_mask & lvl)) \
+	if (tape->debug_mask & lvl)			\
 	printk(KERN_INFO "ide-tape: " fmt, ## args);	\
 }
 #else
@@ -441,7 +439,7 @@ static void idetape_output_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	}
 }
 
-static void idetape_update_buffers(struct ide_atapi_pc *pc)
+static void idetape_update_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc)
 {
 	struct idetape_bh *bh = pc->bh;
 	int count;
@@ -526,7 +524,7 @@ static void idetape_analyze_error(ide_drive_t *drive, u8 *sense)
 		pc->xferred = pc->req_xfer -
 			tape->blk_size *
 			get_unaligned_be32(&sense[3]);
-		idetape_update_buffers(pc);
+		idetape_update_buffers(drive, pc);
 	}
 
 	/*
@@ -800,129 +798,11 @@ static void ide_tape_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
  */
 static ide_startstop_t idetape_pc_intr(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
-	struct ide_atapi_pc *pc = tape->pc;
-	xfer_func_t *xferfunc;
-	unsigned int temp;
-	u16 bcount;
-	u8 stat, ireason;
-
-	debug_log(DBG_PC_INTR, "Enter %s - interrupt handler\n", __func__);
-
-	/* Clear the interrupt */
-	stat = ide_read_status(drive);
-
-	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
-		if (hwif->dma_ops->dma_end(drive) || (stat & ERR_STAT)) {
-			pc->flags |= PC_FLAG_DMA_ERROR;
-		} else {
-			pc->xferred = pc->req_xfer;
-			idetape_update_buffers(pc);
-		}
-		debug_log(DBG_PC_INTR, "%s: DMA finished\n", drive->name);
-	}
-
-	/* No more interrupts */
-	if ((stat & DRQ_STAT) == 0) {
-		debug_log(DBG_PC_INTR, "Packet command completed, %d bytes"
-				" transferred\n", pc->xferred);
-
-		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
-		local_irq_enable_in_hardirq();
-
-		if ((stat & ERR_STAT) && pc->c[0] == REQUEST_SENSE)
-			stat &= ~ERR_STAT;
-		if ((stat & ERR_STAT) || (pc->flags & PC_FLAG_DMA_ERROR)) {
-			/* Error detected */
-			debug_log(DBG_PC_INTR, "%s: I/O error\n", drive->name);
-
-			if (pc->c[0] == REQUEST_SENSE) {
-				printk(KERN_ERR "%s: I/O error in request sense"
-						" command\n", drive->name);
-				return ide_do_reset(drive);
-			}
-			debug_log(DBG_PC_INTR, "[cmd %x]: check condition\n",
-					pc->c[0]);
-
-			/* Retry operation */
-			idetape_retry_pc(drive);
-			return ide_stopped;
-		}
-		pc->error = 0;
-		if ((pc->flags & PC_FLAG_WAIT_FOR_DSC) &&
-		    (stat & SEEK_STAT) == 0) {
-			ide_tape_handle_dsc(drive);
-			return ide_stopped;
-		}
-		/* Command finished - Call the callback function */
-		pc->callback(drive);
-		return ide_stopped;
-	}
-
-	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
-		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
-		printk(KERN_ERR "%s: The device wants to issue more interrupts "
-				"in DMA mode\n", drive->name);
-		ide_dma_off(drive);
-		return ide_do_reset(drive);
-	}
-	/* Get the number of bytes to transfer on this interrupt. */
-	bcount = (hwif->INB(hwif->io_ports.lbah_addr) << 8) |
-		  hwif->INB(hwif->io_ports.lbam_addr);
-
-	ireason = hwif->INB(hwif->io_ports.nsect_addr);
-
-	if (ireason & CD) {
-		printk(KERN_ERR "%s: CoD != 0 in %s\n", drive->name, __func__);
-		return ide_do_reset(drive);
-	}
-	if (((ireason & IO) == IO) == !!(pc->flags & PC_FLAG_WRITING)) {
-		/* Hopefully, we will never get here */
-		printk(KERN_ERR "%s: We wanted to %s, but the device wants us "
-				"to %s!\n", drive->name,
-				(ireason & IO) ? "Write" : "Read",
-				(ireason & IO) ? "Read" : "Write");
-		return ide_do_reset(drive);
-	}
-	if (!(pc->flags & PC_FLAG_WRITING)) {
-		/* Reading - Check that we have enough space */
-		temp = pc->xferred + bcount;
-		if (temp > pc->req_xfer) {
-			if (temp > pc->buf_size) {
-				printk(KERN_ERR "%s: The device wants to send "
-						"us more data than expected - "
-						"discarding data\n",
-						drive->name);
-				ide_pad_transfer(drive, 0, bcount);
-				ide_set_handler(drive, &idetape_pc_intr,
-						IDETAPE_WAIT_CMD, NULL);
-				return ide_started;
-			}
-			debug_log(DBG_PC_INTR, "The device wants to send us more "
-				"data than expected - allowing transfer\n");
-		}
-		xferfunc = hwif->input_data;
-	} else {
-		xferfunc = hwif->output_data;
-	}
-
-	if (pc->bh)
-		ide_tape_io_buffers(drive, pc, bcount,
-				    !!(pc->flags & PC_FLAG_WRITING));
-	else
-		xferfunc(drive, NULL, pc->cur_pos, bcount);
-
-	/* Update the current position */
-	pc->xferred += bcount;
-	pc->cur_pos += bcount;
-
-	debug_log(DBG_PC_INTR, "[cmd %x] transferred %d bytes on that intr.\n",
-			pc->c[0], bcount);
 
-	/* And set the interrupt handler again */
-	ide_set_handler(drive, &idetape_pc_intr, IDETAPE_WAIT_CMD, NULL);
-	return ide_started;
+	return ide_pc_intr(drive, tape->pc, idetape_pc_intr, IDETAPE_WAIT_CMD,
+			   NULL, idetape_update_buffers, idetape_retry_pc,
+			   ide_tape_handle_dsc, ide_tape_io_buffers);
 }
 
 /*
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index ada733ca6725..683bce375c74 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -356,120 +356,11 @@ static int idescsi_expiry(ide_drive_t *drive)
 static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 {
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-	ide_hwif_t *hwif = drive->hwif;
 	struct ide_atapi_pc *pc = scsi->pc;
-	struct request *rq = pc->rq;
-	xfer_func_t *xferfunc;
-	unsigned int temp;
-	u16 bcount;
-	u8 stat, ireason;
-
-	debug_log("Enter %s - interrupt handler\n", __func__);
-
-	if (pc->flags & PC_FLAG_TIMEDOUT) {
-		pc->callback(drive);
-		return ide_stopped;
-	}
-
-	/* Clear the interrupt */
-	stat = ide_read_status(drive);
-
-	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
-		if (hwif->dma_ops->dma_end(drive))
-			pc->flags |= PC_FLAG_DMA_ERROR;
-		else
-			pc->xferred = pc->req_xfer;
-		debug_log("%s: DMA finished\n", drive->name);
-	}
-
-	if ((stat & DRQ_STAT) == 0) {
-		/* No more interrupts */
-		debug_log("Packet command completed, %d bytes transferred\n",
-			  pc->xferred);
-		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
-		local_irq_enable_in_hardirq();
-		if ((stat & ERR_STAT) || (pc->flags & PC_FLAG_DMA_ERROR)) {
-			/* Error detected */
-			debug_log("%s: I/O error\n", drive->name);
-
-			rq->errors++;
-		}
-		pc->callback(drive);
-		return ide_stopped;
-	}
-	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
-		pc->flags &= ~PC_FLAG_DMA_IN_PROGRESS;
-		printk(KERN_ERR "%s: The device wants to issue more interrupts "
-				"in DMA mode\n", drive->name);
-		ide_dma_off(drive);
-		return ide_do_reset(drive);
-	}
-	bcount = (hwif->INB(hwif->io_ports.lbah_addr) << 8) |
-		  hwif->INB(hwif->io_ports.lbam_addr);
-	ireason = hwif->INB(hwif->io_ports.nsect_addr);
-
-	if (ireason & CD) {
-		printk(KERN_ERR "%s: CoD != 0 in %s\n", drive->name, __func__);
-		return ide_do_reset (drive);
-	}
-	if (((ireason & IO) == IO) == !!(pc->flags & PC_FLAG_WRITING)) {
-		/* Hopefully, we will never get here */
-		printk(KERN_ERR "%s: We wanted to %s, but the device wants us "
-				"to %s!\n", drive->name,
-				(ireason & IO) ? "Write" : "Read",
-				(ireason & IO) ? "Read" : "Write");
-		return ide_do_reset(drive);
-	}
-	if (!(pc->flags & PC_FLAG_WRITING)) {
-		temp = pc->xferred + bcount;
-		if (temp > pc->req_xfer) {
-			if (temp > pc->buf_size) {
-				printk(KERN_ERR "%s: The device wants to send "
-						"us more data than expected - "
-						"discarding data\n",
-						drive->name);
-				temp = pc->buf_size - pc->xferred;
-				if (temp) {
-					if (pc->sg)
-						ide_scsi_io_buffers(drive, pc,
-								    temp, 0);
-					else
-						hwif->input_data(drive, NULL,
-							pc->cur_pos, temp);
-					printk(KERN_ERR "%s: transferred %d of "
-							"%d bytes\n",
-							drive->name,
-							temp, bcount);
-				}
-				pc->xferred += temp;
-				pc->cur_pos += temp;
-				ide_pad_transfer(drive, 0, bcount - temp);
-				ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
-				return ide_started;
-			}
-			debug_log("The device wants to send us more data than "
-				  "expected - allowing transfer\n");
-		}
-		xferfunc = hwif->input_data;
-	} else
-		xferfunc = hwif->output_data;
-
-	if (pc->sg)
-		ide_scsi_io_buffers(drive, pc, bcount,
-				    !!(pc->flags & PC_FLAG_WRITING));
-	else
-		xferfunc(drive, NULL, pc->cur_pos, bcount);
-
-	/* Update the current position */
-	pc->xferred += bcount;
-	pc->cur_pos += bcount;
-
-	debug_log("[cmd %x] transferred %d bytes on that intr.\n",
-		  pc->c[0], bcount);
 
-	/* And set the interrupt handler again */
-	ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
-	return ide_started;
+	return ide_pc_intr(drive, pc, idescsi_pc_intr, get_timeout(pc),
+			   idescsi_expiry, NULL, NULL, NULL,
+			   ide_scsi_io_buffers);
 }
 
 static ide_startstop_t idescsi_transfer_pc(ide_drive_t *drive)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index fee07a7edb19..ac4eeb2932ef 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -968,6 +968,12 @@ extern int drive_is_ready(ide_drive_t *);
 
 void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
 
+ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
+	ide_handler_t *handler, unsigned int timeout, ide_expiry_t *expiry,
+	void (*update_buffers)(ide_drive_t *, struct ide_atapi_pc *),
+	void (*retry_pc)(ide_drive_t *), void (*dsc_handle)(ide_drive_t *),
+	void (*io_buffers)(ide_drive_t *, struct ide_atapi_pc *, unsigned int,
+			   int));
 ide_startstop_t ide_transfer_pc(ide_drive_t *, struct ide_atapi_pc *,
 				ide_handler_t *, unsigned int, ide_expiry_t *);
 ide_startstop_t ide_issue_pc(ide_drive_t *, struct ide_atapi_pc *,
-- 
cgit v1.2.3


From 8f920d5e29f86d3425a68e1c3bc264d1f6f55112 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Tue, 15 Jul 2008 14:06:48 -0400
Subject: lockd: eliminate duplicate nlmsvc_lookup_host call from
 nlmsvc_testlock

nlmsvc_testlock calls nlmsvc_lookup_host to find a nlm_host struct. The
callers of this functions, however, call nlmsvc_retrieve_args or
nlm4svc_retrieve_args, which also return a nlm_host struct.

Change nlmsvc_testlock to take a host arg instead of calling
nlmsvc_lookup_host itself and change the callers to pass a pointer to
the nlm_host they've already found.

We take a reference to host in the place where nlmsvc_testlock()
previous did a new lookup, so the reference counting is unchanged from
before.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/svc4proc.c         |  2 +-
 fs/lockd/svclock.c          | 12 +++---------
 fs/lockd/svcproc.c          |  2 +-
 include/linux/lockd/lockd.h |  3 ++-
 4 files changed, 7 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c
index 006a832d46f2..8cfb9daa7c77 100644
--- a/fs/lockd/svc4proc.c
+++ b/fs/lockd/svc4proc.c
@@ -99,7 +99,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp,
 		return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success;
 
 	/* Now check for conflicting locks */
-	resp->status = nlmsvc_testlock(rqstp, file, &argp->lock, &resp->lock, &resp->cookie);
+	resp->status = nlmsvc_testlock(rqstp, file, host, &argp->lock, &resp->lock, &resp->cookie);
 	if (resp->status == nlm_drop_reply)
 		rc = rpc_drop_reply;
 	else
diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c
index 81aca859bfde..f40afb3a0e69 100644
--- a/fs/lockd/svclock.c
+++ b/fs/lockd/svclock.c
@@ -460,8 +460,8 @@ out:
  */
 __be32
 nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file,
-		struct nlm_lock *lock, struct nlm_lock *conflock,
-		struct nlm_cookie *cookie)
+		struct nlm_host *host, struct nlm_lock *lock,
+		struct nlm_lock *conflock, struct nlm_cookie *cookie)
 {
 	struct nlm_block 	*block = NULL;
 	int			error;
@@ -479,16 +479,10 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file,
 
 	if (block == NULL) {
 		struct file_lock *conf = kzalloc(sizeof(*conf), GFP_KERNEL);
-		struct nlm_host	*host;
 
 		if (conf == NULL)
 			return nlm_granted;
-		/* Create host handle for callback */
-		host = nlmsvc_lookup_host(rqstp, lock->caller, lock->len);
-		if (host == NULL) {
-			kfree(conf);
-			return nlm_lck_denied_nolocks;
-		}
+		nlm_get_host(host);
 		block = nlmsvc_create_block(rqstp, host, file, lock, cookie);
 		if (block == NULL) {
 			kfree(conf);
diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c
index fce3d7039625..e099f589b61b 100644
--- a/fs/lockd/svcproc.c
+++ b/fs/lockd/svcproc.c
@@ -128,7 +128,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp,
 		return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success;
 
 	/* Now check for conflicting locks */
-	resp->status = cast_status(nlmsvc_testlock(rqstp, file, &argp->lock, &resp->lock, &resp->cookie));
+	resp->status = cast_status(nlmsvc_testlock(rqstp, file, host, &argp->lock, &resp->lock, &resp->cookie));
 	if (resp->status == nlm_drop_reply)
 		rc = rpc_drop_reply;
 	else
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 102d928f7206..b27967034b53 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -203,7 +203,8 @@ __be32		  nlmsvc_lock(struct svc_rqst *, struct nlm_file *,
 					struct nlm_lock *, int, struct nlm_cookie *);
 __be32		  nlmsvc_unlock(struct nlm_file *, struct nlm_lock *);
 __be32		  nlmsvc_testlock(struct svc_rqst *, struct nlm_file *,
-			struct nlm_lock *, struct nlm_lock *, struct nlm_cookie *);
+			struct nlm_host *, struct nlm_lock *,
+			struct nlm_lock *, struct nlm_cookie *);
 __be32		  nlmsvc_cancel_blocked(struct nlm_file *, struct nlm_lock *);
 unsigned long	  nlmsvc_retry_blocked(void);
 void		  nlmsvc_traverse_blocks(struct nlm_host *, struct nlm_file *,
-- 
cgit v1.2.3


From 6cde4de80773497d8333985b135f472eda870904 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Tue, 15 Jul 2008 14:26:17 -0400
Subject: lockd: eliminate duplicate nlmsvc_lookup_host call from nlmsvc_lock

nlmsvc_lock calls nlmsvc_lookup_host to find a nlm_host struct. The
callers of this function, however, call nlmsvc_retrieve_args or
nlm4svc_retrieve_args, which also return a nlm_host struct.

Change nlmsvc_lock to take a host arg instead of calling
nlmsvc_lookup_host itself and change the callers to pass a pointer to
the nlm_host they've already found.

Since nlmsvc_testlock() now just uses the caller's reference, we no
longer need to get or release it.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/svc4proc.c         |  2 +-
 fs/lockd/svclock.c          | 10 ++--------
 fs/lockd/svcproc.c          |  2 +-
 include/linux/lockd/lockd.h |  3 ++-
 4 files changed, 6 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c
index 8cfb9daa7c77..189b2ce01da6 100644
--- a/fs/lockd/svc4proc.c
+++ b/fs/lockd/svc4proc.c
@@ -145,7 +145,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp,
 #endif
 
 	/* Now try to lock the file */
-	resp->status = nlmsvc_lock(rqstp, file, &argp->lock,
+	resp->status = nlmsvc_lock(rqstp, file, host, &argp->lock,
 					argp->block, &argp->cookie);
 	if (resp->status == nlm_drop_reply)
 		rc = rpc_drop_reply;
diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c
index f40afb3a0e69..bcf73f6e8226 100644
--- a/fs/lockd/svclock.c
+++ b/fs/lockd/svclock.c
@@ -358,10 +358,10 @@ nlmsvc_defer_lock_rqst(struct svc_rqst *rqstp, struct nlm_block *block)
  */
 __be32
 nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
-			struct nlm_lock *lock, int wait, struct nlm_cookie *cookie)
+	    struct nlm_host *host, struct nlm_lock *lock, int wait,
+	    struct nlm_cookie *cookie)
 {
 	struct nlm_block	*block = NULL;
-	struct nlm_host		*host;
 	int			error;
 	__be32			ret;
 
@@ -373,11 +373,6 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 				(long long)lock->fl.fl_end,
 				wait);
 
-	/* Create host handle for callback */
-	host = nlmsvc_lookup_host(rqstp, lock->caller, lock->len);
-	if (host == NULL)
-		return nlm_lck_denied_nolocks;
-
 	/* Lock file against concurrent access */
 	mutex_lock(&file->f_mutex);
 	/* Get existing block (in case client is busy-waiting)
@@ -450,7 +445,6 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 out:
 	mutex_unlock(&file->f_mutex);
 	nlmsvc_release_block(block);
-	nlm_release_host(host);
 	dprintk("lockd: nlmsvc_lock returned %u\n", ret);
 	return ret;
 }
diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c
index e099f589b61b..82dc9083ba67 100644
--- a/fs/lockd/svcproc.c
+++ b/fs/lockd/svcproc.c
@@ -175,7 +175,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp,
 #endif
 
 	/* Now try to lock the file */
-	resp->status = cast_status(nlmsvc_lock(rqstp, file, &argp->lock,
+	resp->status = cast_status(nlmsvc_lock(rqstp, file, host, &argp->lock,
 					       argp->block, &argp->cookie));
 	if (resp->status == nlm_drop_reply)
 		rc = rpc_drop_reply;
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index b27967034b53..f81f9dd5f147 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -200,7 +200,8 @@ typedef int	  (*nlm_host_match_fn_t)(void *cur, struct nlm_host *ref);
  * Server-side lock handling
  */
 __be32		  nlmsvc_lock(struct svc_rqst *, struct nlm_file *,
-					struct nlm_lock *, int, struct nlm_cookie *);
+			      struct nlm_host *, struct nlm_lock *, int,
+			      struct nlm_cookie *);
 __be32		  nlmsvc_unlock(struct nlm_file *, struct nlm_lock *);
 __be32		  nlmsvc_testlock(struct svc_rqst *, struct nlm_file *,
 			struct nlm_host *, struct nlm_lock *,
-- 
cgit v1.2.3


From 367c8c7bd9a2882daad6c9cb607e1db8ef781ad4 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 30 Jun 2008 18:58:14 -0400
Subject: lockd: Pass "struct sockaddr *" to new failover-by-IP function

Pass a more generic socket address type to nlmsvc_unlock_all_by_ip() to
allow for future support of IPv6.  Also provide additional sanity
checking in failover_unlock_ip() when constructing the server's IP
address.

As an added bonus, provide clean kerneldoc comments on related NLM
interfaces which were recently added.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/svcsubs.c          | 32 +++++++++++++++++++++++---------
 fs/nfsd/nfsctl.c            | 15 ++++++++++-----
 include/linux/lockd/lockd.h |  2 +-
 3 files changed, 34 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c
index d1c48b539df8..198b4e55b373 100644
--- a/fs/lockd/svcsubs.c
+++ b/fs/lockd/svcsubs.c
@@ -373,13 +373,16 @@ nlmsvc_free_host_resources(struct nlm_host *host)
 	}
 }
 
-/*
- * Remove all locks held for clients
+/**
+ * nlmsvc_invalidate_all - remove all locks held for clients
+ *
+ * Release all locks held by NFS clients.
+ *
  */
 void
 nlmsvc_invalidate_all(void)
 {
-	/* Release all locks held by NFS clients.
+	/*
 	 * Previously, the code would call
 	 * nlmsvc_free_host_resources for each client in
 	 * turn, which is about as inefficient as it gets.
@@ -396,6 +399,12 @@ nlmsvc_match_sb(void *datap, struct nlm_file *file)
 	return sb == file->f_file->f_path.mnt->mnt_sb;
 }
 
+/**
+ * nlmsvc_unlock_all_by_sb - release locks held on this file system
+ * @sb: super block
+ *
+ * Release all locks held by clients accessing this file system.
+ */
 int
 nlmsvc_unlock_all_by_sb(struct super_block *sb)
 {
@@ -409,17 +418,22 @@ EXPORT_SYMBOL_GPL(nlmsvc_unlock_all_by_sb);
 static int
 nlmsvc_match_ip(void *datap, struct nlm_host *host)
 {
-	__be32 *server_addr = datap;
-
-	return host->h_saddr.sin_addr.s_addr == *server_addr;
+	return nlm_cmp_addr(&host->h_saddr, datap);
 }
 
+/**
+ * nlmsvc_unlock_all_by_ip - release local locks by IP address
+ * @server_addr: server's IP address as seen by clients
+ *
+ * Release all locks held by clients accessing this host
+ * via the passed in IP address.
+ */
 int
-nlmsvc_unlock_all_by_ip(__be32 server_addr)
+nlmsvc_unlock_all_by_ip(struct sockaddr *server_addr)
 {
 	int ret;
-	ret = nlm_traverse_files(&server_addr, nlmsvc_match_ip, NULL);
-	return ret ? -EIO : 0;
 
+	ret = nlm_traverse_files(server_addr, nlmsvc_match_ip, NULL);
+	return ret ? -EIO : 0;
 }
 EXPORT_SYMBOL_GPL(nlmsvc_unlock_all_by_ip);
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 2c2eb8796c10..1955a2702e60 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -310,9 +310,12 @@ static ssize_t write_getfd(struct file *file, char *buf, size_t size)
 
 static ssize_t failover_unlock_ip(struct file *file, char *buf, size_t size)
 {
-	__be32 server_ip;
-	char *fo_path, c;
+	struct sockaddr_in sin = {
+		.sin_family	= AF_INET,
+	};
 	int b1, b2, b3, b4;
+	char c;
+	char *fo_path;
 
 	/* sanity check */
 	if (size == 0)
@@ -326,11 +329,13 @@ static ssize_t failover_unlock_ip(struct file *file, char *buf, size_t size)
 		return -EINVAL;
 
 	/* get ipv4 address */
-	if (sscanf(fo_path, "%u.%u.%u.%u%c", &b1, &b2, &b3, &b4, &c) != 4)
+	if (sscanf(fo_path, NIPQUAD_FMT "%c", &b1, &b2, &b3, &b4, &c) != 4)
+		return -EINVAL;
+	if (b1 > 255 || b2 > 255 || b3 > 255 || b4 > 255)
 		return -EINVAL;
-	server_ip = htonl((((((b1<<8)|b2)<<8)|b3)<<8)|b4);
+	sin.sin_addr.s_addr = htonl((b1 << 24) | (b2 << 16) | (b3 << 8) | b4);
 
-	return nlmsvc_unlock_all_by_ip(server_ip);
+	return nlmsvc_unlock_all_by_ip((struct sockaddr *)&sin);
 }
 
 static ssize_t failover_unlock_fs(struct file *file, char *buf, size_t size)
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index f81f9dd5f147..dbb87ab282e8 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -226,7 +226,7 @@ void		  nlmsvc_invalidate_all(void);
  * Cluster failover support
  */
 int           nlmsvc_unlock_all_by_sb(struct super_block *sb);
-int           nlmsvc_unlock_all_by_ip(__be32 server_addr);
+int           nlmsvc_unlock_all_by_ip(struct sockaddr *server_addr);
 
 static inline struct inode *nlmsvc_file_inode(struct nlm_file *file)
 {
-- 
cgit v1.2.3


From c2e1b09ff237c0a3687b9a804cc8bf489743cffc Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Mon, 14 Jul 2008 16:03:30 -0400
Subject: SUNRPC: Support registering IPv6 interfaces with local rpcbind daemon

Introduce a new API to register RPC services on IPv6 interfaces to allow
the NFS server and lockd to advertise on IPv6 networks.

Unlike rpcb_register(), the new rpcb_v4_register() function uses rpcbind
protocol version 4 to contact the local rpcbind daemon.  The version 4
SET/UNSET procedures allow services to register address families besides
AF_INET, register at specific network interfaces, and register transport
protocols besides UDP and TCP.  All of this functionality is exposed via
the new rpcb_v4_register() kernel API.

A user-space rpcbind daemon implementation that supports version 4 of the
rpcbind protocol is required in order to make use of this new API.

Note that rpcbind version 3 is sufficient to support the new rpcbind
facilities listed above, but most extant implementations use version 4.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h |   3 +
 net/sunrpc/rpcb_clnt.c      | 178 +++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 177 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index 764fd4c286e0..e5bfe01ee305 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -125,6 +125,9 @@ void		rpc_shutdown_client(struct rpc_clnt *);
 void		rpc_release_client(struct rpc_clnt *);
 
 int		rpcb_register(u32, u32, int, unsigned short, int *);
+int		rpcb_v4_register(const u32 program, const u32 version,
+				 const struct sockaddr *address,
+				 const char *netid, int *result);
 int		rpcb_getport_sync(struct sockaddr_in *, u32, u32, int);
 void		rpcb_getport_async(struct rpc_task *);
 
diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c
index 8b75c306e661..24db2b4d12d3 100644
--- a/net/sunrpc/rpcb_clnt.c
+++ b/net/sunrpc/rpcb_clnt.c
@@ -87,6 +87,7 @@ struct rpcbind_args {
 
 static struct rpc_procinfo rpcb_procedures2[];
 static struct rpc_procinfo rpcb_procedures3[];
+static struct rpc_procinfo rpcb_procedures4[];
 
 struct rpcb_info {
 	u32			rpc_vers;
@@ -122,6 +123,12 @@ static const struct sockaddr_in rpcb_inaddr_loopback = {
 	.sin_port		= htons(RPCBIND_PORT),
 };
 
+static const struct sockaddr_in6 rpcb_in6addr_loopback = {
+	.sin6_family		= AF_INET6,
+	.sin6_addr		= IN6ADDR_LOOPBACK_INIT,
+	.sin6_port		= htons(RPCBIND_PORT),
+};
+
 static struct rpc_clnt *rpcb_create_local(struct sockaddr *addr,
 					  size_t addrlen, u32 version)
 {
@@ -196,13 +203,38 @@ static int rpcb_register_call(struct sockaddr *addr, size_t addrlen,
  * rpcb_register - set or unset a port registration with the local rpcbind svc
  * @prog: RPC program number to bind
  * @vers: RPC version number to bind
- * @prot: transport protocol to use to make this request
+ * @prot: transport protocol to register
  * @port: port value to register
- * @okay: result code
+ * @okay: OUT: result code
+ *
+ * RPC services invoke this function to advertise their contact
+ * information via the system's rpcbind daemon.  RPC services
+ * invoke this function once for each [program, version, transport]
+ * tuple they wish to advertise.
+ *
+ * Callers may also unregister RPC services that are no longer
+ * available by setting the passed-in port to zero.  This removes
+ * all registered transports for [program, version] from the local
+ * rpcbind database.
+ *
+ * Returns zero if the registration request was dispatched
+ * successfully and a reply was received.  The rpcbind daemon's
+ * boolean result code is stored in *okay.
  *
- * port == 0 means unregister, port != 0 means register.
+ * Returns an errno value and sets *result to zero if there was
+ * some problem that prevented the rpcbind request from being
+ * dispatched, or if the rpcbind daemon did not respond within
+ * the timeout.
  *
- * This routine supports only rpcbind version 2.
+ * This function uses rpcbind protocol version 2 to contact the
+ * local rpcbind daemon.
+ *
+ * Registration works over both AF_INET and AF_INET6, and services
+ * registered via this function are advertised as available for any
+ * address.  If the local rpcbind daemon is listening on AF_INET6,
+ * services registered via this function will be advertised on
+ * IN6ADDR_ANY (ie available for all AF_INET and AF_INET6
+ * addresses).
  */
 int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay)
 {
@@ -230,6 +262,144 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay)
 					RPCBVERS_2, &msg, okay);
 }
 
+/*
+ * Fill in AF_INET family-specific arguments to register
+ */
+static int rpcb_register_netid4(struct sockaddr_in *address_to_register,
+				struct rpc_message *msg)
+{
+	struct rpcbind_args *map = msg->rpc_argp;
+	unsigned short port = ntohs(address_to_register->sin_port);
+	char buf[32];
+
+	/* Construct AF_INET universal address */
+	snprintf(buf, sizeof(buf),
+			NIPQUAD_FMT".%u.%u",
+			NIPQUAD(address_to_register->sin_addr.s_addr),
+			port >> 8, port & 0xff);
+	map->r_addr = buf;
+
+	dprintk("RPC:       %sregistering [%u, %u, %s, '%s'] with "
+		"local rpcbind\n", (port ? "" : "un"),
+			map->r_prog, map->r_vers,
+			map->r_addr, map->r_netid);
+
+	msg->rpc_proc = &rpcb_procedures4[RPCBPROC_UNSET];
+	if (port)
+		msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET];
+
+	return rpcb_register_call((struct sockaddr *)&rpcb_inaddr_loopback,
+					sizeof(rpcb_inaddr_loopback),
+					RPCBVERS_4, msg, msg->rpc_resp);
+}
+
+/*
+ * Fill in AF_INET6 family-specific arguments to register
+ */
+static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register,
+				struct rpc_message *msg)
+{
+	struct rpcbind_args *map = msg->rpc_argp;
+	unsigned short port = ntohs(address_to_register->sin6_port);
+	char buf[64];
+
+	/* Construct AF_INET6 universal address */
+	snprintf(buf, sizeof(buf),
+			NIP6_FMT".%u.%u",
+			NIP6(address_to_register->sin6_addr),
+			port >> 8, port & 0xff);
+	map->r_addr = buf;
+
+	dprintk("RPC:       %sregistering [%u, %u, %s, '%s'] with "
+		"local rpcbind\n", (port ? "" : "un"),
+			map->r_prog, map->r_vers,
+			map->r_addr, map->r_netid);
+
+	msg->rpc_proc = &rpcb_procedures4[RPCBPROC_UNSET];
+	if (port)
+		msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET];
+
+	return rpcb_register_call((struct sockaddr *)&rpcb_in6addr_loopback,
+					sizeof(rpcb_in6addr_loopback),
+					RPCBVERS_4, msg, msg->rpc_resp);
+}
+
+/**
+ * rpcb_v4_register - set or unset a port registration with the local rpcbind
+ * @program: RPC program number of service to (un)register
+ * @version: RPC version number of service to (un)register
+ * @address: address family, IP address, and port to (un)register
+ * @netid: netid of transport protocol to (un)register
+ * @result: result code from rpcbind RPC call
+ *
+ * RPC services invoke this function to advertise their contact
+ * information via the system's rpcbind daemon.  RPC services
+ * invoke this function once for each [program, version, address,
+ * netid] tuple they wish to advertise.
+ *
+ * Callers may also unregister RPC services that are no longer
+ * available by setting the port number in the passed-in address
+ * to zero.  Callers pass a netid of "" to unregister all
+ * transport netids associated with [program, version, address].
+ *
+ * Returns zero if the registration request was dispatched
+ * successfully and a reply was received.  The rpcbind daemon's
+ * result code is stored in *result.
+ *
+ * Returns an errno value and sets *result to zero if there was
+ * some problem that prevented the rpcbind request from being
+ * dispatched, or if the rpcbind daemon did not respond within
+ * the timeout.
+ *
+ * This function uses rpcbind protocol version 4 to contact the
+ * local rpcbind daemon.  The local rpcbind daemon must support
+ * version 4 of the rpcbind protocol in order for these functions
+ * to register a service successfully.
+ *
+ * Supported netids include "udp" and "tcp" for UDP and TCP over
+ * IPv4, and "udp6" and "tcp6" for UDP and TCP over IPv6,
+ * respectively.
+ *
+ * The contents of @address determine the address family and the
+ * port to be registered.  The usual practice is to pass INADDR_ANY
+ * as the raw address, but specifying a non-zero address is also
+ * supported by this API if the caller wishes to advertise an RPC
+ * service on a specific network interface.
+ *
+ * Note that passing in INADDR_ANY does not create the same service
+ * registration as IN6ADDR_ANY.  The former advertises an RPC
+ * service on any IPv4 address, but not on IPv6.  The latter
+ * advertises the service on all IPv4 and IPv6 addresses.
+ */
+int rpcb_v4_register(const u32 program, const u32 version,
+		     const struct sockaddr *address, const char *netid,
+		     int *result)
+{
+	struct rpcbind_args map = {
+		.r_prog		= program,
+		.r_vers		= version,
+		.r_netid	= netid,
+		.r_owner	= RPCB_OWNER_STRING,
+	};
+	struct rpc_message msg = {
+		.rpc_argp	= &map,
+		.rpc_resp	= result,
+	};
+
+	*result = 0;
+
+	switch (address->sa_family) {
+	case AF_INET:
+		return rpcb_register_netid4((struct sockaddr_in *)address,
+					    &msg);
+	case AF_INET6:
+		return rpcb_register_netid6((struct sockaddr_in6 *)address,
+					    &msg);
+	}
+
+	return -EAFNOSUPPORT;
+}
+
 /**
  * rpcb_getport_sync - obtain the port for an RPC service on a given host
  * @sin: address of remote peer
-- 
cgit v1.2.3


From 3be53f3f213223f50d8e29b5e1869685bf040a1e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:36 +0200
Subject: ide: move some bits from ide-timing.h to <linux/ide.h>

Move struct ide_timing and IDE_TIMING_* defines to <linux/ide.h>
from drivers/ide/ide-timing.h.

While at it:

- use u8/u16 instead of short for struct ide_timing fields

- use enum for IDE_TIMING_*

There should be no functional changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-timing.h | 23 -----------------------
 include/linux/ide.h      | 28 ++++++++++++++++++++++++++++
 2 files changed, 28 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-timing.h b/drivers/ide/ide-timing.h
index ebe884d76881..724879910ac8 100644
--- a/drivers/ide/ide-timing.h
+++ b/drivers/ide/ide-timing.h
@@ -28,18 +28,6 @@
 #include <linux/kernel.h>
 #include <linux/hdreg.h>
 
-struct ide_timing {
-	u8    mode;
-	short setup;	/* t1 */
-	short act8b;	/* t2 for 8-bit io */
-	short rec8b;	/* t2i for 8-bit io */
-	short cyc8b;	/* t0 for 8-bit io */
-	short active;	/* t2 or tD */
-	short recover;	/* t2i or tK */
-	short cycle;	/* t0 */
-	short udma;	/* t2CYCTYP/2 */
-};
-
 /*
  * PIO 0-5, MWDMA 0-2 and UDMA 0-6 timings (in nanoseconds).
  * These were taken from ATA/ATAPI-6 standard, rev 0a, except
@@ -79,17 +67,6 @@ static struct ide_timing ide_timing[] = {
 	{ 0xff }
 };
 
-#define IDE_TIMING_SETUP	0x01
-#define IDE_TIMING_ACT8B	0x02
-#define IDE_TIMING_REC8B	0x04
-#define IDE_TIMING_CYC8B	0x08
-#define IDE_TIMING_8BIT		0x0e
-#define IDE_TIMING_ACTIVE	0x10
-#define IDE_TIMING_RECOVER	0x20
-#define IDE_TIMING_CYCLE	0x40
-#define IDE_TIMING_UDMA		0x80
-#define IDE_TIMING_ALL		0xff
-
 #define ENOUGH(v,unit)		(((v)-1)/(unit)+1)
 #define EZ(v,unit)		((v)?ENOUGH(v,unit):0)
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ac4eeb2932ef..81c6ea436beb 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1271,6 +1271,34 @@ static inline int ide_dev_is_sata(struct hd_driveid *id)
 u64 ide_get_lba_addr(struct ide_taskfile *, int);
 u8 ide_dump_status(ide_drive_t *, const char *, u8);
 
+struct ide_timing {
+	u8  mode;
+	u8  setup;	/* t1 */
+	u16 act8b;	/* t2 for 8-bit io */
+	u16 rec8b;	/* t2i for 8-bit io */
+	u16 cyc8b;	/* t0 for 8-bit io */
+	u16 active;	/* t2 or tD */
+	u16 recover;	/* t2i or tK */
+	u16 cycle;	/* t0 */
+	u16 udma;	/* t2CYCTYP/2 */
+};
+
+enum {
+	IDE_TIMING_SETUP	= (1 << 0),
+	IDE_TIMING_ACT8B	= (1 << 1),
+	IDE_TIMING_REC8B	= (1 << 2),
+	IDE_TIMING_CYC8B	= (1 << 3),
+	IDE_TIMING_8BIT		= IDE_TIMING_ACT8B | IDE_TIMING_REC8B |
+				  IDE_TIMING_CYC8B,
+	IDE_TIMING_ACTIVE	= (1 << 4),
+	IDE_TIMING_RECOVER	= (1 << 5),
+	IDE_TIMING_CYCLE	= (1 << 6),
+	IDE_TIMING_UDMA		= (1 << 7),
+	IDE_TIMING_ALL		= IDE_TIMING_SETUP | IDE_TIMING_8BIT |
+				  IDE_TIMING_ACTIVE | IDE_TIMING_RECOVER |
+				  IDE_TIMING_CYCLE | IDE_TIMING_UDMA,
+};
+
 typedef struct ide_pio_timings_s {
 	int	setup_time;	/* Address setup (ns) minimum */
 	int	active_time;	/* Active pulse (ns) minimum */
-- 
cgit v1.2.3


From f06ab3402aa2d6de060442c1053ea10b24b65076 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:37 +0200
Subject: ide: convert ide-timing.h to ide-timings.c library (take 2)

* Don't include ide-timing.h in cs5535 and sis5513 host drivers
  (they don't need it currently).

* Convert ide-timing.h to ide-timings.c library and add CONFIG_IDE_TIMINGS
  config option to be selected by host drivers using the library.

While at it:

- fix ide_timing_find_mode() placement

v2:
* Add missing EXPORT_SYMBOLs. (Stephen Rothwell <sfr@canb.auug.org.au>)

There should be no functional changes caused by this patch.

Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Kconfig           |   7 ++
 drivers/ide/Makefile          |   1 +
 drivers/ide/arm/palm_bk3710.c |   2 -
 drivers/ide/ide-timing.h      | 183 ------------------------------------------
 drivers/ide/ide-timings.c     | 183 ++++++++++++++++++++++++++++++++++++++++++
 drivers/ide/pci/amd74xx.c     |   2 -
 drivers/ide/pci/cs5535.c      |   2 -
 drivers/ide/pci/sis5513.c     |   3 -
 drivers/ide/pci/via82cxxx.c   |   2 -
 drivers/ide/ppc/pmac.c        |   2 -
 include/linux/ide.h           |   5 ++
 11 files changed, 196 insertions(+), 196 deletions(-)
 delete mode 100644 drivers/ide/ide-timing.h
 create mode 100644 drivers/ide/ide-timings.c

(limited to 'include/linux')

diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index cd08dba8261b..994b6d39b559 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -98,6 +98,9 @@ if BLK_DEV_IDE
 
 comment "Please see Documentation/ide/ide.txt for help/info on IDE drives"
 
+config IDE_TIMINGS
+	bool
+
 config IDE_ATAPI
 	bool
 
@@ -469,6 +472,7 @@ config BLK_DEV_ALI15X3
 config BLK_DEV_AMD74XX
 	tristate "AMD and nVidia IDE support"
 	depends on !ARM
+	select IDE_TIMINGS
 	select BLK_DEV_IDEDMA_PCI
 	help
 	  This driver adds explicit support for AMD-7xx and AMD-8111 chips
@@ -725,6 +729,7 @@ config BLK_DEV_TRM290
 
 config BLK_DEV_VIA82CXXX
 	tristate "VIA82CXXX chipset support"
+	select IDE_TIMINGS
 	select BLK_DEV_IDEDMA_PCI
 	help
 	  This driver adds explicit support for VIA BusMastering IDE chips.
@@ -751,6 +756,7 @@ endif
 config BLK_DEV_IDE_PMAC
 	tristate "PowerMac on-board IDE support"
 	depends on PPC_PMAC && IDE=y && BLK_DEV_IDE=y
+	select IDE_TIMINGS
 	help
 	  This driver provides support for the on-board IDE controller on
 	  most of the recent Apple Power Macintoshes and PowerBooks.
@@ -912,6 +918,7 @@ config BLK_DEV_Q40IDE
 config BLK_DEV_PALMCHIP_BK3710
 	tristate "Palmchip bk3710 IDE controller support"
 	depends on ARCH_DAVINCI
+	select IDE_TIMINGS
 	select BLK_DEV_IDEDMA_SFF
 	help
 	  Say Y here if you want to support the onchip IDE controller on the
diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index a2b3f84d710d..cb1350684c9a 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -14,6 +14,7 @@ EXTRA_CFLAGS				+= -Idrivers/ide
 ide-core-y += ide.o ide-io.o ide-iops.o ide-lib.o ide-probe.o ide-taskfile.o
 
 # core IDE code
+ide-core-$(CONFIG_IDE_TIMINGS)		+= ide-timings.o
 ide-core-$(CONFIG_IDE_ATAPI)		+= ide-atapi.o
 ide-core-$(CONFIG_BLK_DEV_IDEPCI)	+= setup-pci.o
 ide-core-$(CONFIG_BLK_DEV_IDEDMA)	+= ide-dma.o
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 3839f5722985..9b8a45d2cf17 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -74,8 +74,6 @@ struct palm_bk3710_udmatiming {
 #define BK3710_IORDYTMP		0x78
 #define BK3710_IORDYTMS		0x7C
 
-#include "../ide-timing.h"
-
 static unsigned ideclk_period; /* in nanoseconds */
 
 static const struct palm_bk3710_udmatiming palm_bk3710_udmatimings[6] = {
diff --git a/drivers/ide/ide-timing.h b/drivers/ide/ide-timing.h
deleted file mode 100644
index 98e05f545450..000000000000
--- a/drivers/ide/ide-timing.h
+++ /dev/null
@@ -1,183 +0,0 @@
-#ifndef _IDE_TIMING_H
-#define _IDE_TIMING_H
-
-/*
- *  Copyright (c) 1999-2001 Vojtech Pavlik
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- *
- * Should you need to contact me, the author, you can do so either by
- * e-mail - mail your message to <vojtech@ucw.cz>, or by paper mail:
- * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic
- */
-
-#include <linux/kernel.h>
-#include <linux/hdreg.h>
-
-/*
- * PIO 0-5, MWDMA 0-2 and UDMA 0-6 timings (in nanoseconds).
- * These were taken from ATA/ATAPI-6 standard, rev 0a, except
- * for PIO 5, which is a nonstandard extension and UDMA6, which
- * is currently supported only by Maxtor drives.
- */
-
-static struct ide_timing ide_timing[] = {
-
-	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0,   0,  15 },
-	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0,   0,  20 },
-	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
-	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
-
-	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
-	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
-	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
-
-	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 120,   0 },
-	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 150,   0 },
-	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 480,   0 },
-
-	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 240,   0 },
-	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
-	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
-
-	{ XFER_PIO_5,     20,  50,  30, 100,  50,  30, 100,   0 },
-	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
-	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
-
-	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 240,   0 },
-	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 383,   0 },
-	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 600,   0 },
-
-	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 960,   0 },
-
-	{ 0xff }
-};
-
-#define ENOUGH(v, unit)		(((v) - 1) / (unit) + 1)
-#define EZ(v, unit)		((v) ? ENOUGH(v, unit) : 0)
-
-static void ide_timing_quantize(struct ide_timing *t, struct ide_timing *q,
-				int T, int UT)
-{
-	q->setup   = EZ(t->setup   * 1000,  T);
-	q->act8b   = EZ(t->act8b   * 1000,  T);
-	q->rec8b   = EZ(t->rec8b   * 1000,  T);
-	q->cyc8b   = EZ(t->cyc8b   * 1000,  T);
-	q->active  = EZ(t->active  * 1000,  T);
-	q->recover = EZ(t->recover * 1000,  T);
-	q->cycle   = EZ(t->cycle   * 1000,  T);
-	q->udma    = EZ(t->udma    * 1000, UT);
-}
-
-static void ide_timing_merge(struct ide_timing *a, struct ide_timing *b,
-			     struct ide_timing *m, unsigned int what)
-{
-	if (what & IDE_TIMING_SETUP)
-		m->setup   = max(a->setup,   b->setup);
-	if (what & IDE_TIMING_ACT8B)
-		m->act8b   = max(a->act8b,   b->act8b);
-	if (what & IDE_TIMING_REC8B)
-		m->rec8b   = max(a->rec8b,   b->rec8b);
-	if (what & IDE_TIMING_CYC8B)
-		m->cyc8b   = max(a->cyc8b,   b->cyc8b);
-	if (what & IDE_TIMING_ACTIVE)
-		m->active  = max(a->active,  b->active);
-	if (what & IDE_TIMING_RECOVER)
-		m->recover = max(a->recover, b->recover);
-	if (what & IDE_TIMING_CYCLE)
-		m->cycle   = max(a->cycle,   b->cycle);
-	if (what & IDE_TIMING_UDMA)
-		m->udma    = max(a->udma,    b->udma);
-}
-
-static struct ide_timing *ide_timing_find_mode(u8 speed)
-{
-	struct ide_timing *t;
-
-	for (t = ide_timing; t->mode != speed; t++)
-		if (t->mode == 0xff)
-			return NULL;
-	return t;
-}
-
-static int ide_timing_compute(ide_drive_t *drive, u8 speed,
-			      struct ide_timing *t, int T, int UT)
-{
-	struct hd_driveid *id = drive->id;
-	struct ide_timing *s, p;
-
-	/*
-	 * Find the mode.
-	 */
-	s = ide_timing_find_mode(speed);
-	if (s == NULL)
-		return -EINVAL;
-
-	/*
-	 * Copy the timing from the table.
-	 */
-	*t = *s;
-
-	/*
-	 * If the drive is an EIDE drive, it can tell us it needs extended
-	 * PIO/MWDMA cycle timing.
-	 */
-	if (id && id->field_valid & 2) {	/* EIDE drive */
-
-		memset(&p, 0, sizeof(p));
-
-		if (speed <= XFER_PIO_2)
-			p.cycle = p.cyc8b = id->eide_pio;
-		else if (speed <= XFER_PIO_5)
-			p.cycle = p.cyc8b = id->eide_pio_iordy;
-		else if (speed >= XFER_MW_DMA_0 && speed <= XFER_MW_DMA_2)
-			p.cycle = id->eide_dma_min;
-
-		ide_timing_merge(&p, t, t, IDE_TIMING_CYCLE | IDE_TIMING_CYC8B);
-	}
-
-	/*
-	 * Convert the timing to bus clock counts.
-	 */
-	ide_timing_quantize(t, t, T, UT);
-
-	/*
-	 * Even in DMA/UDMA modes we still use PIO access for IDENTIFY,
-	 * S.M.A.R.T and some other commands. We have to ensure that the
-	 * DMA cycle timing is slower/equal than the fastest PIO timing.
-	 */
-	if (speed >= XFER_SW_DMA_0) {
-		u8 pio = ide_get_best_pio_mode(drive, 255, 5);
-		ide_timing_compute(drive, XFER_PIO_0 + pio, &p, T, UT);
-		ide_timing_merge(&p, t, t, IDE_TIMING_ALL);
-	}
-
-	/*
-	 * Lengthen active & recovery time so that cycle time is correct.
-	 */
-	if (t->act8b + t->rec8b < t->cyc8b) {
-		t->act8b += (t->cyc8b - (t->act8b + t->rec8b)) / 2;
-		t->rec8b = t->cyc8b - t->act8b;
-	}
-
-	if (t->active + t->recover < t->cycle) {
-		t->active += (t->cycle - (t->active + t->recover)) / 2;
-		t->recover = t->cycle - t->active;
-	}
-
-	return 0;
-}
-
-#endif
diff --git a/drivers/ide/ide-timings.c b/drivers/ide/ide-timings.c
new file mode 100644
index 000000000000..ebef6d4e3f63
--- /dev/null
+++ b/drivers/ide/ide-timings.c
@@ -0,0 +1,183 @@
+/*
+ *  Copyright (c) 1999-2001 Vojtech Pavlik
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Should you need to contact me, the author, you can do so either by
+ * e-mail - mail your message to <vojtech@ucw.cz>, or by paper mail:
+ * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic
+ */
+
+#include <linux/kernel.h>
+#include <linux/hdreg.h>
+#include <linux/ide.h>
+#include <linux/module.h>
+
+/*
+ * PIO 0-5, MWDMA 0-2 and UDMA 0-6 timings (in nanoseconds).
+ * These were taken from ATA/ATAPI-6 standard, rev 0a, except
+ * for PIO 5, which is a nonstandard extension and UDMA6, which
+ * is currently supported only by Maxtor drives.
+ */
+
+static struct ide_timing ide_timing[] = {
+
+	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0,   0,  15 },
+	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0,   0,  20 },
+	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
+	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
+
+	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
+	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
+	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
+
+	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 120,   0 },
+	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 150,   0 },
+	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 480,   0 },
+
+	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 240,   0 },
+	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
+	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
+
+	{ XFER_PIO_5,     20,  50,  30, 100,  50,  30, 100,   0 },
+	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
+	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
+
+	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 240,   0 },
+	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 383,   0 },
+	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 600,   0 },
+
+	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 960,   0 },
+
+	{ 0xff }
+};
+
+struct ide_timing *ide_timing_find_mode(u8 speed)
+{
+	struct ide_timing *t;
+
+	for (t = ide_timing; t->mode != speed; t++)
+		if (t->mode == 0xff)
+			return NULL;
+	return t;
+}
+EXPORT_SYMBOL_GPL(ide_timing_find_mode);
+
+#define ENOUGH(v, unit)		(((v) - 1) / (unit) + 1)
+#define EZ(v, unit)		((v) ? ENOUGH(v, unit) : 0)
+
+static void ide_timing_quantize(struct ide_timing *t, struct ide_timing *q,
+				int T, int UT)
+{
+	q->setup   = EZ(t->setup   * 1000,  T);
+	q->act8b   = EZ(t->act8b   * 1000,  T);
+	q->rec8b   = EZ(t->rec8b   * 1000,  T);
+	q->cyc8b   = EZ(t->cyc8b   * 1000,  T);
+	q->active  = EZ(t->active  * 1000,  T);
+	q->recover = EZ(t->recover * 1000,  T);
+	q->cycle   = EZ(t->cycle   * 1000,  T);
+	q->udma    = EZ(t->udma    * 1000, UT);
+}
+
+void ide_timing_merge(struct ide_timing *a, struct ide_timing *b,
+		      struct ide_timing *m, unsigned int what)
+{
+	if (what & IDE_TIMING_SETUP)
+		m->setup   = max(a->setup,   b->setup);
+	if (what & IDE_TIMING_ACT8B)
+		m->act8b   = max(a->act8b,   b->act8b);
+	if (what & IDE_TIMING_REC8B)
+		m->rec8b   = max(a->rec8b,   b->rec8b);
+	if (what & IDE_TIMING_CYC8B)
+		m->cyc8b   = max(a->cyc8b,   b->cyc8b);
+	if (what & IDE_TIMING_ACTIVE)
+		m->active  = max(a->active,  b->active);
+	if (what & IDE_TIMING_RECOVER)
+		m->recover = max(a->recover, b->recover);
+	if (what & IDE_TIMING_CYCLE)
+		m->cycle   = max(a->cycle,   b->cycle);
+	if (what & IDE_TIMING_UDMA)
+		m->udma    = max(a->udma,    b->udma);
+}
+EXPORT_SYMBOL_GPL(ide_timing_merge);
+
+int ide_timing_compute(ide_drive_t *drive, u8 speed,
+		       struct ide_timing *t, int T, int UT)
+{
+	struct hd_driveid *id = drive->id;
+	struct ide_timing *s, p;
+
+	/*
+	 * Find the mode.
+	 */
+	s = ide_timing_find_mode(speed);
+	if (s == NULL)
+		return -EINVAL;
+
+	/*
+	 * Copy the timing from the table.
+	 */
+	*t = *s;
+
+	/*
+	 * If the drive is an EIDE drive, it can tell us it needs extended
+	 * PIO/MWDMA cycle timing.
+	 */
+	if (id && id->field_valid & 2) {	/* EIDE drive */
+
+		memset(&p, 0, sizeof(p));
+
+		if (speed <= XFER_PIO_2)
+			p.cycle = p.cyc8b = id->eide_pio;
+		else if (speed <= XFER_PIO_5)
+			p.cycle = p.cyc8b = id->eide_pio_iordy;
+		else if (speed >= XFER_MW_DMA_0 && speed <= XFER_MW_DMA_2)
+			p.cycle = id->eide_dma_min;
+
+		ide_timing_merge(&p, t, t, IDE_TIMING_CYCLE | IDE_TIMING_CYC8B);
+	}
+
+	/*
+	 * Convert the timing to bus clock counts.
+	 */
+	ide_timing_quantize(t, t, T, UT);
+
+	/*
+	 * Even in DMA/UDMA modes we still use PIO access for IDENTIFY,
+	 * S.M.A.R.T and some other commands. We have to ensure that the
+	 * DMA cycle timing is slower/equal than the fastest PIO timing.
+	 */
+	if (speed >= XFER_SW_DMA_0) {
+		u8 pio = ide_get_best_pio_mode(drive, 255, 5);
+		ide_timing_compute(drive, XFER_PIO_0 + pio, &p, T, UT);
+		ide_timing_merge(&p, t, t, IDE_TIMING_ALL);
+	}
+
+	/*
+	 * Lengthen active & recovery time so that cycle time is correct.
+	 */
+	if (t->act8b + t->rec8b < t->cyc8b) {
+		t->act8b += (t->cyc8b - (t->act8b + t->rec8b)) / 2;
+		t->rec8b = t->cyc8b - t->act8b;
+	}
+
+	if (t->active + t->recover < t->cycle) {
+		t->active += (t->cycle - (t->active + t->recover)) / 2;
+		t->recover = t->cycle - t->active;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ide_timing_compute);
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index ad222206a429..0bfcdd0e77b3 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -21,8 +21,6 @@
 #include <linux/init.h>
 #include <linux/ide.h>
 
-#include "ide-timing.h"
-
 enum {
 	AMD_IDE_CONFIG		= 0x41,
 	AMD_CABLE_DETECT	= 0x42,
diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c
index 2a2cb4911905..dc97c48623f3 100644
--- a/drivers/ide/pci/cs5535.c
+++ b/drivers/ide/pci/cs5535.c
@@ -26,8 +26,6 @@
 #include <linux/pci.h>
 #include <linux/ide.h>
 
-#include "ide-timing.h"
-
 #define MSR_ATAC_BASE		0x51300000
 #define ATAC_GLD_MSR_CAP	(MSR_ATAC_BASE+0)
 #define ATAC_GLD_MSR_CONFIG	(MSR_ATAC_BASE+0x01)
diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index e127eb25ab63..2389945ca95d 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -52,8 +52,6 @@
 #include <linux/init.h>
 #include <linux/ide.h>
 
-#include "ide-timing.h"
-
 /* registers layout and init values are chipset family dependant */
 
 #define ATA_16		0x01
@@ -616,7 +614,6 @@ MODULE_LICENSE("GPL");
 /*
  * TODO:
  *	- CLEANUP
- *	- Use drivers/ide/ide-timing.h !
  *	- More checks in the config registers (force values instead of
  *	  relying on the BIOS setting them correctly).
  *	- Further optimisations ?
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index 3ed9728abd24..e47384c70c40 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -35,8 +35,6 @@
 #include <asm/processor.h>
 #endif
 
-#include "ide-timing.h"
-
 #define VIA_IDE_ENABLE		0x40
 #define VIA_IDE_CONFIG		0x41
 #define VIA_FIFO_CONFIG		0x43
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index dcb2c466bb97..5b91d23269d3 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -48,8 +48,6 @@
 #include <asm/mediabay.h>
 #endif
 
-#include "../ide-timing.h"
-
 #undef IDE_PMAC_DEBUG
 
 #define DMA_WAIT_TIMEOUT	50
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 81c6ea436beb..057001f6b1dc 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1299,6 +1299,11 @@ enum {
 				  IDE_TIMING_CYCLE | IDE_TIMING_UDMA,
 };
 
+struct ide_timing *ide_timing_find_mode(u8);
+void ide_timing_merge(struct ide_timing *, struct ide_timing *,
+		      struct ide_timing *, unsigned int);
+int ide_timing_compute(ide_drive_t *, u8, struct ide_timing *, int, int);
+
 typedef struct ide_pio_timings_s {
 	int	setup_time;	/* Address setup (ns) minimum */
 	int	active_time;	/* Active pulse (ns) minimum */
-- 
cgit v1.2.3


From c9d6c1a2379373219bb3271bdcbdc0ab2edf349d Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:39 +0200
Subject: ide: move ide_pio_cycle_time() to ide-timings.c

All ide_pio_cycle_time() users already select CONFIG_IDE_TIMINGS
so move the function from ide-lib.c to ide-timings.c.

While at it:

- convert ide_pio_cycle_time() to use ide_timing_find_mode()

- cleanup ide_pio_cycle_time() a bit

There should be no functional changes caused by this patch.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-lib.c     | 23 -----------------------
 drivers/ide/ide-timings.c | 22 ++++++++++++++++++++++
 include/linux/ide.h       |  2 +-
 3 files changed, 23 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 7e053d217732..efa5bfa64d01 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -188,29 +188,6 @@ static int ide_scan_pio_blacklist (char *model)
 	return -1;
 }
 
-unsigned int ide_pio_cycle_time(ide_drive_t *drive, u8 pio)
-{
-	struct hd_driveid *id = drive->id;
-	int cycle_time = 0;
-
-	if (id->field_valid & 2) {
-		if (id->capability & 8)
-			cycle_time = id->eide_pio_iordy;
-		else
-			cycle_time = id->eide_pio;
-	}
-
-	/* conservative "downgrade" for all pre-ATA2 drives */
-	if (pio < 3) {
-		if (cycle_time && cycle_time < ide_pio_timings[pio].cycle_time)
-			cycle_time = 0; /* use standard timing */
-	}
-
-	return cycle_time ? cycle_time : ide_pio_timings[pio].cycle_time;
-}
-
-EXPORT_SYMBOL_GPL(ide_pio_cycle_time);
-
 /**
  *	ide_get_best_pio_mode	-	get PIO mode from drive
  *	@drive: drive to consider
diff --git a/drivers/ide/ide-timings.c b/drivers/ide/ide-timings.c
index ebef6d4e3f63..8c2f8327f487 100644
--- a/drivers/ide/ide-timings.c
+++ b/drivers/ide/ide-timings.c
@@ -1,5 +1,6 @@
 /*
  *  Copyright (c) 1999-2001 Vojtech Pavlik
+ *  Copyright (c) 2007-2008 Bartlomiej Zolnierkiewicz
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -75,6 +76,27 @@ struct ide_timing *ide_timing_find_mode(u8 speed)
 }
 EXPORT_SYMBOL_GPL(ide_timing_find_mode);
 
+u16 ide_pio_cycle_time(ide_drive_t *drive, u8 pio)
+{
+	struct hd_driveid *id = drive->id;
+	struct ide_timing *t = ide_timing_find_mode(XFER_PIO_0 + pio);
+	u16 cycle = 0;
+
+	if (id->field_valid & 2) {
+		if (id->capability & 8)
+			cycle = id->eide_pio_iordy;
+		else
+			cycle = id->eide_pio;
+
+		/* conservative "downgrade" for all pre-ATA2 drives */
+		if (pio < 3 && cycle < t->cycle)
+			cycle = 0; /* use standard timing */
+	}
+
+	return cycle ? cycle : t->cycle;
+}
+EXPORT_SYMBOL_GPL(ide_pio_cycle_time);
+
 #define ENOUGH(v, unit)		(((v) - 1) / (unit) + 1)
 #define EZ(v, unit)		((v) ? ENOUGH(v, unit) : 0)
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 057001f6b1dc..3899c761b302 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1300,6 +1300,7 @@ enum {
 };
 
 struct ide_timing *ide_timing_find_mode(u8);
+u16 ide_pio_cycle_time(ide_drive_t *, u8);
 void ide_timing_merge(struct ide_timing *, struct ide_timing *,
 		      struct ide_timing *, unsigned int);
 int ide_timing_compute(ide_drive_t *, u8, struct ide_timing *, int, int);
@@ -1311,7 +1312,6 @@ typedef struct ide_pio_timings_s {
 				/* active + recovery (+ setup for some chips) */
 } ide_pio_timings_t;
 
-unsigned int ide_pio_cycle_time(ide_drive_t *, u8);
 u8 ide_get_best_pio_mode(ide_drive_t *, u8, u8);
 extern const ide_pio_timings_t ide_pio_timings[6];
 
-- 
cgit v1.2.3


From 3e153cfb5e38ae237ff27a10a833946ac95db8a4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:39 +0200
Subject: ide: remove no longer used ide_pio_timings[]

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-lib.c | 17 -----------------
 include/linux/ide.h   |  8 --------
 2 files changed, 25 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index efa5bfa64d01..3e12f229bd5f 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -75,23 +75,6 @@ static u8 ide_rate_filter(ide_drive_t *drive, u8 speed)
 	return min(speed, mode);
 }
 
-/*
- * Standard (generic) timings for PIO modes, from ATA2 specification.
- * These timings are for access to the IDE data port register *only*.
- * Some drives may specify a mode, while also specifying a different
- * value for cycle_time (from drive identification data).
- */
-const ide_pio_timings_t ide_pio_timings[6] = {
-	{ 70,	165,	600 },	/* PIO Mode 0 */
-	{ 50,	125,	383 },	/* PIO Mode 1 */
-	{ 30,	100,	240 },	/* PIO Mode 2 */
-	{ 30,	80,	180 },	/* PIO Mode 3 with IORDY */
-	{ 25,	70,	120 },	/* PIO Mode 4 with IORDY */
-	{ 20,	50,	100 }	/* PIO Mode 5 with IORDY (nonstandard) */
-};
-
-EXPORT_SYMBOL_GPL(ide_pio_timings);
-
 /*
  * Shared data/functions for determining best PIO mode for an IDE drive.
  * Most of this stuff originally lived in cmd640.c, and changes to the
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 3899c761b302..4e44525fa5ca 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1305,15 +1305,7 @@ void ide_timing_merge(struct ide_timing *, struct ide_timing *,
 		      struct ide_timing *, unsigned int);
 int ide_timing_compute(ide_drive_t *, u8, struct ide_timing *, int, int);
 
-typedef struct ide_pio_timings_s {
-	int	setup_time;	/* Address setup (ns) minimum */
-	int	active_time;	/* Active pulse (ns) minimum */
-	int	cycle_time;	/* Cycle time (ns) minimum = */
-				/* active + recovery (+ setup for some chips) */
-} ide_pio_timings_t;
-
 u8 ide_get_best_pio_mode(ide_drive_t *, u8, u8);
-extern const ide_pio_timings_t ide_pio_timings[6];
 
 int ide_set_pio_mode(ide_drive_t *, u8);
 int ide_set_dma_mode(ide_drive_t *, u8);
-- 
cgit v1.2.3


From 9ad540937554a3779c5fe7af13aa390b1d2aeb3e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:39 +0200
Subject: ide: move PIO blacklist to ide-pio-blacklist.c

Move PIO blacklist to ide-pio-blacklist.c.

While at it:

- fix comment

- fix whitespace damage

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Makefile            |  3 +-
 drivers/ide/ide-lib.c           | 96 -----------------------------------------
 drivers/ide/ide-pio-blacklist.c | 94 ++++++++++++++++++++++++++++++++++++++++
 include/linux/ide.h             |  2 +
 4 files changed, 98 insertions(+), 97 deletions(-)
 create mode 100644 drivers/ide/ide-pio-blacklist.c

(limited to 'include/linux')

diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index cb1350684c9a..8605536ea18f 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -11,7 +11,8 @@
 
 EXTRA_CFLAGS				+= -Idrivers/ide
 
-ide-core-y += ide.o ide-io.o ide-iops.o ide-lib.o ide-probe.o ide-taskfile.o
+ide-core-y += ide.o ide-io.o ide-iops.o ide-lib.o ide-probe.o ide-taskfile.o \
+	      ide-pio-blacklist.o
 
 # core IDE code
 ide-core-$(CONFIG_IDE_TIMINGS)		+= ide-timings.o
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 3e12f229bd5f..13af72f09ec4 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -75,102 +75,6 @@ static u8 ide_rate_filter(ide_drive_t *drive, u8 speed)
 	return min(speed, mode);
 }
 
-/*
- * Shared data/functions for determining best PIO mode for an IDE drive.
- * Most of this stuff originally lived in cmd640.c, and changes to the
- * ide_pio_blacklist[] table should be made with EXTREME CAUTION to avoid
- * breaking the fragile cmd640.c support.
- */
-
-/*
- * Black list. Some drives incorrectly report their maximal PIO mode,
- * at least in respect to CMD640. Here we keep info on some known drives.
- */
-static struct ide_pio_info {
-	const char	*name;
-	int		pio;
-} ide_pio_blacklist [] = {
-	{ "Conner Peripherals 540MB - CFS540A", 3 },
-
-	{ "WDC AC2700",  3 },
-	{ "WDC AC2540",  3 },
-	{ "WDC AC2420",  3 },
-	{ "WDC AC2340",  3 },
-	{ "WDC AC2250",  0 },
-	{ "WDC AC2200",  0 },
-	{ "WDC AC21200", 4 },
-	{ "WDC AC2120",  0 },
-	{ "WDC AC2850",  3 },
-	{ "WDC AC1270",  3 },
-	{ "WDC AC1170",  1 },
-	{ "WDC AC1210",  1 },
-	{ "WDC AC280",   0 },
-	{ "WDC AC31000", 3 },
-	{ "WDC AC31200", 3 },
-
-	{ "Maxtor 7131 AT", 1 },
-	{ "Maxtor 7171 AT", 1 },
-	{ "Maxtor 7213 AT", 1 },
-	{ "Maxtor 7245 AT", 1 },
-	{ "Maxtor 7345 AT", 1 },
-	{ "Maxtor 7546 AT", 3 },
-	{ "Maxtor 7540 AV", 3 },
-
-	{ "SAMSUNG SHD-3121A", 1 },
-	{ "SAMSUNG SHD-3122A", 1 },
-	{ "SAMSUNG SHD-3172A", 1 },
-
-	{ "ST5660A",  3 },
-	{ "ST3660A",  3 },
-	{ "ST3630A",  3 },
-	{ "ST3655A",  3 },
-	{ "ST3391A",  3 },
-	{ "ST3390A",  1 },
-	{ "ST3600A",  1 },
-	{ "ST3290A",  0 },
-	{ "ST3144A",  0 },
-	{ "ST3491A",  1 },	/* reports 3, should be 1 or 2 (depending on */	
-				/* drive) according to Seagates FIND-ATA program */
-
-	{ "QUANTUM ELS127A", 0 },
-	{ "QUANTUM ELS170A", 0 },
-	{ "QUANTUM LPS240A", 0 },
-	{ "QUANTUM LPS210A", 3 },
-	{ "QUANTUM LPS270A", 3 },
-	{ "QUANTUM LPS365A", 3 },
-	{ "QUANTUM LPS540A", 3 },
-	{ "QUANTUM LIGHTNING 540A", 3 },
-	{ "QUANTUM LIGHTNING 730A", 3 },
-
-        { "QUANTUM FIREBALL_540", 3 }, /* Older Quantum Fireballs don't work */
-        { "QUANTUM FIREBALL_640", 3 }, 
-        { "QUANTUM FIREBALL_1080", 3 },
-        { "QUANTUM FIREBALL_1280", 3 },
-	{ NULL,	0 }
-};
-
-/**
- *	ide_scan_pio_blacklist 	-	check for a blacklisted drive
- *	@model: Drive model string
- *
- *	This routine searches the ide_pio_blacklist for an entry
- *	matching the start/whole of the supplied model name.
- *
- *	Returns -1 if no match found.
- *	Otherwise returns the recommended PIO mode from ide_pio_blacklist[].
- */
-
-static int ide_scan_pio_blacklist (char *model)
-{
-	struct ide_pio_info *p;
-
-	for (p = ide_pio_blacklist; p->name != NULL; p++) {
-		if (strncmp(p->name, model, strlen(p->name)) == 0)
-			return p->pio;
-	}
-	return -1;
-}
-
 /**
  *	ide_get_best_pio_mode	-	get PIO mode from drive
  *	@drive: drive to consider
diff --git a/drivers/ide/ide-pio-blacklist.c b/drivers/ide/ide-pio-blacklist.c
new file mode 100644
index 000000000000..a8c2c8f8660a
--- /dev/null
+++ b/drivers/ide/ide-pio-blacklist.c
@@ -0,0 +1,94 @@
+/*
+ * PIO blacklist.  Some drives incorrectly report their maximal PIO mode,
+ * at least in respect to CMD640.  Here we keep info on some known drives.
+ *
+ * Changes to the ide_pio_blacklist[] should be made with EXTREME CAUTION
+ * to avoid breaking the fragile cmd640.c support.
+ */
+
+#include <linux/string.h>
+
+static struct ide_pio_info {
+	const char	*name;
+	int		pio;
+} ide_pio_blacklist [] = {
+	{ "Conner Peripherals 540MB - CFS540A", 3 },
+
+	{ "WDC AC2700",  3 },
+	{ "WDC AC2540",  3 },
+	{ "WDC AC2420",  3 },
+	{ "WDC AC2340",  3 },
+	{ "WDC AC2250",  0 },
+	{ "WDC AC2200",  0 },
+	{ "WDC AC21200", 4 },
+	{ "WDC AC2120",  0 },
+	{ "WDC AC2850",  3 },
+	{ "WDC AC1270",  3 },
+	{ "WDC AC1170",  1 },
+	{ "WDC AC1210",  1 },
+	{ "WDC AC280",   0 },
+	{ "WDC AC31000", 3 },
+	{ "WDC AC31200", 3 },
+
+	{ "Maxtor 7131 AT", 1 },
+	{ "Maxtor 7171 AT", 1 },
+	{ "Maxtor 7213 AT", 1 },
+	{ "Maxtor 7245 AT", 1 },
+	{ "Maxtor 7345 AT", 1 },
+	{ "Maxtor 7546 AT", 3 },
+	{ "Maxtor 7540 AV", 3 },
+
+	{ "SAMSUNG SHD-3121A", 1 },
+	{ "SAMSUNG SHD-3122A", 1 },
+	{ "SAMSUNG SHD-3172A", 1 },
+
+	{ "ST5660A",  3 },
+	{ "ST3660A",  3 },
+	{ "ST3630A",  3 },
+	{ "ST3655A",  3 },
+	{ "ST3391A",  3 },
+	{ "ST3390A",  1 },
+	{ "ST3600A",  1 },
+	{ "ST3290A",  0 },
+	{ "ST3144A",  0 },
+	{ "ST3491A",  1 }, /* reports 3, should be 1 or 2 (depending on drive)
+			      according to Seagate's FIND-ATA program */
+
+	{ "QUANTUM ELS127A", 0 },
+	{ "QUANTUM ELS170A", 0 },
+	{ "QUANTUM LPS240A", 0 },
+	{ "QUANTUM LPS210A", 3 },
+	{ "QUANTUM LPS270A", 3 },
+	{ "QUANTUM LPS365A", 3 },
+	{ "QUANTUM LPS540A", 3 },
+	{ "QUANTUM LIGHTNING 540A", 3 },
+	{ "QUANTUM LIGHTNING 730A", 3 },
+
+	{ "QUANTUM FIREBALL_540", 3 }, /* Older Quantum Fireballs don't work */
+	{ "QUANTUM FIREBALL_640", 3 },
+	{ "QUANTUM FIREBALL_1080", 3 },
+	{ "QUANTUM FIREBALL_1280", 3 },
+	{ NULL, 0 }
+};
+
+/**
+ *	ide_scan_pio_blacklist 	-	check for a blacklisted drive
+ *	@model: Drive model string
+ *
+ *	This routine searches the ide_pio_blacklist for an entry
+ *	matching the start/whole of the supplied model name.
+ *
+ *	Returns -1 if no match found.
+ *	Otherwise returns the recommended PIO mode from ide_pio_blacklist[].
+ */
+
+int ide_scan_pio_blacklist(char *model)
+{
+	struct ide_pio_info *p;
+
+	for (p = ide_pio_blacklist; p->name != NULL; p++) {
+		if (strncmp(p->name, model, strlen(p->name)) == 0)
+			return p->pio;
+	}
+	return -1;
+}
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 4e44525fa5ca..535c439fd8f0 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1305,6 +1305,8 @@ void ide_timing_merge(struct ide_timing *, struct ide_timing *,
 		      struct ide_timing *, unsigned int);
 int ide_timing_compute(ide_drive_t *, u8, struct ide_timing *, int, int);
 
+int ide_scan_pio_blacklist(char *);
+
 u8 ide_get_best_pio_mode(ide_drive_t *, u8, u8);
 
 int ide_set_pio_mode(ide_drive_t *, u8);
-- 
cgit v1.2.3


From 63b51c6d1d63276fd320615c042f1ff5d94ebab8 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:40 +0200
Subject: ide: make ide_hwifs[] static

Move ide_hwifs[] from ide.c to ide-probe.c and make it static.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c |  2 ++
 drivers/ide/ide.c       |  2 --
 include/linux/ide.h     | 12 ------------
 3 files changed, 2 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 592d424412ea..97dda05914a1 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -39,6 +39,8 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
+static ide_hwif_t ide_hwifs[MAX_HWIFS]; /* master data repository */
+
 /**
  *	generic_id		-	add a generic drive id
  *	@drive:	drive to make an ID block for
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 32d8ee281d56..b7855a13f0ef 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -91,8 +91,6 @@ DEFINE_MUTEX(ide_cfg_mtx);
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(ide_lock);
 EXPORT_SYMBOL(ide_lock);
 
-ide_hwif_t ide_hwifs[MAX_HWIFS];	/* master data repository */
-
 static void ide_port_init_devices_data(ide_hwif_t *);
 
 /*
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 535c439fd8f0..15d5668198a6 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -801,18 +801,6 @@ struct ide_driver_s {
 
 int generic_ide_ioctl(ide_drive_t *, struct file *, struct block_device *, unsigned, unsigned long);
 
-/*
- * ide_hwifs[] is the master data structure used to keep track
- * of just about everything in ide.c.  Whenever possible, routines
- * should be using pointers to a drive (ide_drive_t *) or
- * pointers to a hwif (ide_hwif_t *), rather than indexing this
- * structure directly (the allocation/layout may change!).
- *
- */
-#ifndef _IDE_C
-extern	ide_hwif_t	ide_hwifs[];		/* master data repository */
-#endif
-
 extern int ide_vlb_clk;
 extern int ide_pci_clk;
 
-- 
cgit v1.2.3


From c56c5648a3bd15ff14c50f284b261140cd5b5472 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:40 +0200
Subject: ide: set hwif->dev in ide_init_port_hw() (take 2)

* Add 'parent' field to hw_regs_t for optional parent device pointer (needed
  by macio PMAC IDE controllers) and set hwif->dev in ide_init_port_hw().

* Update au1xxx-ide.c, sgiioc4.c, pmac.c and setup-pci.c accordingly.

v2:

* Update scc_pata.c.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide.c             | 3 ++-
 drivers/ide/mips/au1xxx-ide.c | 2 --
 drivers/ide/pci/scc_pata.c    | 1 -
 drivers/ide/pci/sgiioc4.c     | 2 --
 drivers/ide/ppc/pmac.c        | 6 ++----
 drivers/ide/setup-pci.c       | 2 --
 include/linux/ide.h           | 2 +-
 7 files changed, 5 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index b7855a13f0ef..9240888e5efd 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -297,7 +297,8 @@ void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
 	memcpy(&hwif->io_ports, &hw->io_ports, sizeof(hwif->io_ports));
 	hwif->irq = hw->irq;
 	hwif->chipset = hw->chipset;
-	hwif->gendev.parent = hw->dev;
+	hwif->dev = hw->dev;
+	hwif->gendev.parent = hw->parent ? hw->parent : hw->dev;
 	hwif->ack_intr = hw->ack_intr;
 }
 EXPORT_SYMBOL_GPL(ide_init_port_hw);
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 1a6c27b32498..08c46c36b7e7 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -600,8 +600,6 @@ static int au_ide_probe(struct device *dev)
 
 	ide_init_port_hw(hwif, &hw);
 
-	hwif->dev = dev;
-
 	/* If the user has selected DDMA assisted copies,
 	   then set up a few local I/O function entry points 
 	*/
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 1584ebb6a185..71bff3131b96 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -572,7 +572,6 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev,
 	hw.dev = &dev->dev;
 	hw.chipset = ide_pci;
 	ide_init_port_hw(hwif, &hw);
-	hwif->dev = &dev->dev;
 
 	idx[0] = hwif->index;
 
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 24513e3dcd6b..af91cc525789 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -625,8 +625,6 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	hw.dev = &dev->dev;
 	ide_init_port_hw(hwif, &hw);
 
-	hwif->dev = &dev->dev;
-
 	/* The IOC4 uses MMIO rather than Port IO. */
 	default_hwif_mmiops(hwif);
 
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index cfa103a51b31..93fb9067c043 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1148,8 +1148,6 @@ pmac_ide_macio_attach(struct macio_dev *mdev, const struct of_device_id *match)
 	base = ioremap(macio_resource_start(mdev, 0), 0x400);
 	regbase = (unsigned long) base;
 
-	hwif->dev = &mdev->bus->pdev->dev;
-
 	pmif->mdev = mdev;
 	pmif->node = mdev->ofdev.node;
 	pmif->regbase = regbase;
@@ -1171,7 +1169,8 @@ pmac_ide_macio_attach(struct macio_dev *mdev, const struct of_device_id *match)
 	memset(&hw, 0, sizeof(hw));
 	pmac_ide_init_ports(&hw, pmif->regbase);
 	hw.irq = irq;
-	hw.dev = &mdev->ofdev.dev;
+	hw.dev = &mdev->bus->pdev->dev;
+	hw.parent = &mdev->ofdev.dev;
 
 	rc = pmac_ide_setup_device(pmif, hwif, &hw);
 	if (rc != 0) {
@@ -1271,7 +1270,6 @@ pmac_ide_pci_attach(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto out_free_pmif;
 	}
 
-	hwif->dev = &pdev->dev;
 	pmif->mdev = NULL;
 	pmif->node = np;
 
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index abcfb1739d4d..3dea5a59626e 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -346,8 +346,6 @@ static ide_hwif_t *ide_hwif_configure(struct pci_dev *dev,
 
 	ide_init_port_hw(hwif, &hw);
 
-	hwif->dev = &dev->dev;
-
 	return hwif;
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 15d5668198a6..a6a2eccb6526 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -171,7 +171,7 @@ typedef struct hw_regs_s {
 	int		irq;			/* our irq number */
 	ide_ack_intr_t	*ack_intr;		/* acknowledge interrupt */
 	hwif_chipset_t  chipset;
-	struct device	*dev;
+	struct device	*dev, *parent;
 } hw_regs_t;
 
 void ide_init_port_data(struct hwif_s *, unsigned int);
-- 
cgit v1.2.3


From e6d95bd14928926d6658b5e4ace905e8b83ed27a Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 16 Jul 2008 20:33:42 +0200
Subject: ide: ->port_init_devs -> ->init_dev

Change ->port_init_devs method to take 'ide_drive_t *' as an argument
instead of 'ide_hwif_t *' and rename it to ->init_dev.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c          |  6 +++---
 drivers/ide/legacy/ht6560b.c     |  8 ++++----
 drivers/ide/legacy/ide-4drives.c | 10 ++++------
 drivers/ide/legacy/qd65xx.c      | 16 ++++++++--------
 include/linux/ide.h              |  4 ++--
 5 files changed, 21 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 97dda05914a1..c5dc18849303 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1320,10 +1320,10 @@ static void ide_port_init_devices(ide_hwif_t *hwif)
 			drive->unmask = 1;
 		if (hwif->host_flags & IDE_HFLAG_NO_UNMASK_IRQS)
 			drive->no_unmask = 1;
-	}
 
-	if (port_ops && port_ops->port_init_devs)
-		port_ops->port_init_devs(hwif);
+		if (port_ops && port_ops->init_dev)
+			port_ops->init_dev(drive);
+	}
 }
 
 static void ide_init_port(ide_hwif_t *hwif, unsigned int port,
diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c
index bd2f579946fa..7bc8fd59ea9e 100644
--- a/drivers/ide/legacy/ht6560b.c
+++ b/drivers/ide/legacy/ht6560b.c
@@ -310,16 +310,16 @@ static void ht6560b_set_pio_mode(ide_drive_t *drive, const u8 pio)
 #endif
 }
 
-static void __init ht6560b_port_init_devs(ide_hwif_t *hwif)
+static void __init ht6560b_init_dev(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	/* Setting default configurations for drives. */
 	int t = (HT_CONFIG_DEFAULT << 8) | HT_TIMING_DEFAULT;
 
 	if (hwif->channel)
 		t |= (HT_SECONDARY_IF << 8);
 
-	hwif->drives[0].drive_data = t;
-	hwif->drives[1].drive_data = t;
+	drive->drive_data = t;
 }
 
 static int probe_ht6560b;
@@ -328,7 +328,7 @@ module_param_named(probe, probe_ht6560b, bool, 0);
 MODULE_PARM_DESC(probe, "probe for HT6560B chipset");
 
 static const struct ide_port_ops ht6560b_port_ops = {
-	.port_init_devs		= ht6560b_port_init_devs,
+	.init_dev		= ht6560b_init_dev,
 	.set_pio_mode		= ht6560b_set_pio_mode,
 	.selectproc		= ht6560b_selectproc,
 };
diff --git a/drivers/ide/legacy/ide-4drives.c b/drivers/ide/legacy/ide-4drives.c
index 5cd6ce537eea..89c8ff0a4d08 100644
--- a/drivers/ide/legacy/ide-4drives.c
+++ b/drivers/ide/legacy/ide-4drives.c
@@ -11,16 +11,14 @@ static int probe_4drives;
 module_param_named(probe, probe_4drives, bool, 0);
 MODULE_PARM_DESC(probe, "probe for generic IDE chipset with 4 drives/port");
 
-static void ide_4drives_port_init_devs(ide_hwif_t *hwif)
+static void ide_4drives_init_dev(ide_drive_t *drive)
 {
-	if (hwif->channel) {
-		hwif->drives[0].select.all ^= 0x20;
-		hwif->drives[1].select.all ^= 0x20;
-	}
+	if (drive->hwif->channel)
+		drive->select.all ^= 0x20;
 }
 
 static const struct ide_port_ops ide_4drives_port_ops = {
-	.port_init_devs		= ide_4drives_port_init_devs,
+	.init_dev		= ide_4drives_init_dev,
 };
 
 static const struct ide_port_info ide_4drives_port_info = {
diff --git a/drivers/ide/legacy/qd65xx.c b/drivers/ide/legacy/qd65xx.c
index 63f6c31d16ea..2338f344ea24 100644
--- a/drivers/ide/legacy/qd65xx.c
+++ b/drivers/ide/legacy/qd65xx.c
@@ -282,17 +282,18 @@ static int __init qd_testreg(int port)
 	return (readreg != QD_TESTVAL);
 }
 
-static void __init qd6500_port_init_devs(ide_hwif_t *hwif)
+static void __init qd6500_init_dev(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	u8 base = (hwif->config_data & 0xff00) >> 8;
 	u8 config = QD_CONFIG(hwif);
 
-	hwif->drives[0].drive_data = QD6500_DEF_DATA;
-	hwif->drives[1].drive_data = QD6500_DEF_DATA;
+	drive->drive_data = QD6500_DEF_DATA;
 }
 
-static void __init qd6580_port_init_devs(ide_hwif_t *hwif)
+static void __init qd6580_init_dev(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	u16 t1, t2;
 	u8 base = (hwif->config_data & 0xff00) >> 8;
 	u8 config = QD_CONFIG(hwif);
@@ -303,18 +304,17 @@ static void __init qd6580_port_init_devs(ide_hwif_t *hwif)
 	} else
 		t2 = t1 = hwif->channel ? QD6580_DEF_DATA2 : QD6580_DEF_DATA;
 
-	hwif->drives[0].drive_data = t1;
-	hwif->drives[1].drive_data = t2;
+	drive->drive_data = drive->select.b.unit ? t2 : t1;
 }
 
 static const struct ide_port_ops qd6500_port_ops = {
-	.port_init_devs		= qd6500_port_init_devs,
+	.init_dev		= qd6500_init_dev,
 	.set_pio_mode		= qd6500_set_pio_mode,
 	.selectproc		= qd65xx_select,
 };
 
 static const struct ide_port_ops qd6580_port_ops = {
-	.port_init_devs		= qd6580_port_init_devs,
+	.init_dev		= qd6580_init_dev,
 	.set_pio_mode		= qd6580_set_pio_mode,
 	.selectproc		= qd65xx_select,
 };
diff --git a/include/linux/ide.h b/include/linux/ide.h
index a6a2eccb6526..f9cbe9350cad 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -405,8 +405,8 @@ typedef struct ide_drive_s {
 struct ide_port_info;
 
 struct ide_port_ops {
-	/* host specific initialization of devices on a port */
-	void	(*port_init_devs)(struct hwif_s *);
+	/* host specific initialization of a device */
+	void	(*init_dev)(ide_drive_t *);
 	/* routine to program host for PIO mode */
 	void	(*set_pio_mode)(ide_drive_t *, const u8);
 	/* routine to program host for DMA mode */
-- 
cgit v1.2.3


From 79e36a9f54aaf4a52eb2d9520953aa3960e99294 Mon Sep 17 00:00:00 2001
From: Elias Oltmanns <eo@nebensachen.de>
Date: Wed, 16 Jul 2008 20:33:48 +0200
Subject: IDE: Fix HDIO_DRIVE_RESET handling

Currently, the code path executing an HDIO_DRIVE_RESET ioctl is broken
in various ways.  Most importantly, it is treated as an out of band
request in an illegal way which may very likely lead to system lock ups.
Use the drive's request queue to avoid this problem (and fix a locking
issue for free along the way).

Signed-off-by: Elias Oltmanns <eo@nebensachen.de>
Cc: "Alan Cox" <alan@lxorguk.ukuu.org.uk>
Cc: "Randy Dunlap" <randy.dunlap@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c   | 23 ++++++++++++++++++++++-
 drivers/ide/ide-iops.c | 18 +++++++++++++-----
 drivers/ide/ide.c      | 41 +++++++++++++++--------------------------
 include/linux/ide.h    |  6 ++++++
 4 files changed, 56 insertions(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 28057747c1f8..2b33c129740b 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -766,6 +766,18 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
  	return ide_stopped;
 }
 
+static ide_startstop_t ide_special_rq(ide_drive_t *drive, struct request *rq)
+{
+	switch (rq->cmd[0]) {
+	case REQ_DRIVE_RESET:
+		return ide_do_reset(drive);
+	default:
+		blk_dump_rq_flags(rq, "ide_special_rq - bad request");
+		ide_end_request(drive, 0, 0);
+		return ide_stopped;
+	}
+}
+
 static void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
 {
 	struct request_pm_state *pm = rq->data;
@@ -869,7 +881,16 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 			    pm->pm_step == ide_pm_state_completed)
 				ide_complete_pm_request(drive, rq);
 			return startstop;
-		}
+		} else if (!rq->rq_disk && blk_special_request(rq))
+			/*
+			 * TODO: Once all ULDs have been modified to
+			 * check for specific op codes rather than
+			 * blindly accepting any special request, the
+			 * check for ->rq_disk above may be replaced
+			 * by a more suitable mechanism or even
+			 * dropped entirely.
+			 */
+			return ide_special_rq(drive, rq);
 
 		drv = *(ide_driver_t **)rq->rq_disk->private_data;
 		return drv->do_request(drive, rq, block);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 80ad4f234f3f..96f63eb12092 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -905,6 +905,14 @@ void ide_execute_pkt_cmd(ide_drive_t *drive)
 }
 EXPORT_SYMBOL_GPL(ide_execute_pkt_cmd);
 
+static inline void ide_complete_drive_reset(ide_drive_t *drive)
+{
+	struct request *rq = drive->hwif->hwgroup->rq;
+
+	if (rq && blk_special_request(rq) && rq->cmd[0] == REQ_DRIVE_RESET)
+		ide_end_request(drive, 1, 0);
+}
+
 /* needed below */
 static ide_startstop_t do_reset1 (ide_drive_t *, int);
 
@@ -940,7 +948,7 @@ static ide_startstop_t atapi_reset_pollfunc (ide_drive_t *drive)
 	}
 	/* done polling */
 	hwgroup->polling = 0;
-	hwgroup->resetting = 0;
+	ide_complete_drive_reset(drive);
 	return ide_stopped;
 }
 
@@ -961,7 +969,7 @@ static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 		if (port_ops->reset_poll(drive)) {
 			printk(KERN_ERR "%s: host reset_poll failure for %s.\n",
 				hwif->name, drive->name);
-			return ide_stopped;
+			goto out;
 		}
 	}
 
@@ -1004,7 +1012,8 @@ static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 		}
 	}
 	hwgroup->polling = 0;	/* done polling */
-	hwgroup->resetting = 0; /* done reset attempt */
+out:
+	ide_complete_drive_reset(drive);
 	return ide_stopped;
 }
 
@@ -1090,7 +1099,6 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 
 	/* For an ATAPI device, first try an ATAPI SRST. */
 	if (drive->media != ide_disk && !do_not_try_atapi) {
-		hwgroup->resetting = 1;
 		pre_reset(drive);
 		SELECT_DRIVE(drive);
 		udelay (20);
@@ -1112,10 +1120,10 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 
 	if (io_ports->ctl_addr == 0) {
 		spin_unlock_irqrestore(&ide_lock, flags);
+		ide_complete_drive_reset(drive);
 		return ide_stopped;
 	}
 
-	hwgroup->resetting = 1;
 	/*
 	 * Note that we also set nIEN while resetting the device,
 	 * to mask unwanted interrupts from the interface during the reset.
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 90ae00d4aaf5..1ec983b00511 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -529,6 +529,19 @@ static int generic_ide_resume(struct device *dev)
 	return err;
 }
 
+static void generic_drive_reset(ide_drive_t *drive)
+{
+	struct request *rq;
+
+	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
+	rq->cmd_type = REQ_TYPE_SPECIAL;
+	rq->cmd_len = 1;
+	rq->cmd[0] = REQ_DRIVE_RESET;
+	rq->cmd_flags |= REQ_SOFTBARRIER;
+	blk_execute_rq(drive->queue, NULL, rq, 1);
+	blk_put_request(rq);
+}
+
 int generic_ide_ioctl(ide_drive_t *drive, struct file *file, struct block_device *bdev,
 			unsigned int cmd, unsigned long arg)
 {
@@ -603,33 +616,9 @@ int generic_ide_ioctl(ide_drive_t *drive, struct file *file, struct block_device
 			if (!capable(CAP_SYS_ADMIN))
 				return -EACCES;
 
-			/*
-			 *	Abort the current command on the
-			 *	group if there is one, taking
-			 *	care not to allow anything else
-			 *	to be queued and to die on the
-			 *	spot if we miss one somehow
-			 */
-
-			spin_lock_irqsave(&ide_lock, flags);
-
-			if (HWGROUP(drive)->resetting) {
-				spin_unlock_irqrestore(&ide_lock, flags);
-				return -EBUSY;
-			}
-
-			ide_abort(drive, "drive reset");
-
-			BUG_ON(HWGROUP(drive)->handler);
-
-			/* Ensure nothing gets queued after we
-			   drop the lock. Reset will clear the busy */
-
-			HWGROUP(drive)->busy = 1;
-			spin_unlock_irqrestore(&ide_lock, flags);
-			(void) ide_do_reset(drive);
-
+			generic_drive_reset(drive);
 			return 0;
+
 		case HDIO_GET_BUSSTATE:
 			if (!capable(CAP_SYS_ADMIN))
 				return -EACCES;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f9cbe9350cad..021710cc1b1c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -138,6 +138,12 @@ struct ide_io_ports {
 #define WAIT_CMD	(10*HZ)	/* 10sec  - maximum wait for an IRQ to happen */
 #define WAIT_MIN_SLEEP	(2*HZ/100)	/* 20msec - minimum sleep time */
 
+/*
+ * Op codes for special requests to be handled by ide_special_rq().
+ * Values should be in the range of 0x20 to 0x3f.
+ */
+#define REQ_DRIVE_RESET		0x20
+
 /*
  * Check for an interrupt and acknowledge the interrupt status
  */
-- 
cgit v1.2.3


From 3ef5eb424ebf0cd981192a416358fd707a9f959b Mon Sep 17 00:00:00 2001
From: Elias Oltmanns <eo@nebensachen.de>
Date: Wed, 16 Jul 2008 20:33:48 +0200
Subject: IDE: Remove unused code

Remove some code which has been made obsolete and hasn't worked properly
before anyway.  Part of the infrastructure may be reintroduced in a
follow up patch to implement a working command aborting facility.

Signed-off-by: Elias Oltmanns <eo@nebensachen.de>
Cc: "Alan Cox" <alan@lxorguk.ukuu.org.uk>
Cc: "Randy Dunlap" <randy.dunlap@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c     |  1 -
 drivers/ide/ide-disk.c   |  1 -
 drivers/ide/ide-floppy.c |  1 -
 drivers/ide/ide-io.c     | 49 ------------------------------------------------
 drivers/ide/ide-tape.c   |  1 -
 drivers/scsi/ide-scsi.c  | 14 --------------
 include/linux/ide.h      |  7 -------
 7 files changed, 74 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 2b3c69d1f53f..6e29dd532090 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -1952,7 +1952,6 @@ static ide_driver_t ide_cdrom_driver = {
 	.do_request		= ide_cd_do_request,
 	.end_request		= ide_end_request,
 	.error			= __ide_error,
-	.abort			= __ide_abort,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idecd_proc,
 #endif
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 5f49a4ae9dd8..3a2e80237c10 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -985,7 +985,6 @@ static ide_driver_t idedisk_driver = {
 	.do_request		= ide_do_rw_disk,
 	.end_request		= ide_end_request,
 	.error			= __ide_error,
-	.abort			= __ide_abort,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idedisk_proc,
 #endif
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index c7e4433fcbcf..011d72011cc4 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -1129,7 +1129,6 @@ static ide_driver_t idefloppy_driver = {
 	.do_request		= idefloppy_do_request,
 	.end_request		= idefloppy_end_request,
 	.error			= __ide_error,
-	.abort			= __ide_abort,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idefloppy_proc,
 #endif
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 2b33c129740b..661b75a89d4d 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -504,55 +504,6 @@ ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, u8 stat)
 
 EXPORT_SYMBOL_GPL(ide_error);
 
-ide_startstop_t __ide_abort(ide_drive_t *drive, struct request *rq)
-{
-	if (drive->media != ide_disk)
-		rq->errors |= ERROR_RESET;
-
-	ide_kill_rq(drive, rq);
-
-	return ide_stopped;
-}
-
-EXPORT_SYMBOL_GPL(__ide_abort);
-
-/**
- *	ide_abort	-	abort pending IDE operations
- *	@drive: drive the error occurred on
- *	@msg: message to report
- *
- *	ide_abort kills and cleans up when we are about to do a 
- *	host initiated reset on active commands. Longer term we
- *	want handlers to have sensible abort handling themselves
- *
- *	This differs fundamentally from ide_error because in 
- *	this case the command is doing just fine when we
- *	blow it away.
- */
- 
-ide_startstop_t ide_abort(ide_drive_t *drive, const char *msg)
-{
-	struct request *rq;
-
-	if (drive == NULL || (rq = HWGROUP(drive)->rq) == NULL)
-		return ide_stopped;
-
-	/* retry only "normal" I/O: */
-	if (!blk_fs_request(rq)) {
-		rq->errors = 1;
-		ide_end_drive_cmd(drive, BUSY_STAT, 0);
-		return ide_stopped;
-	}
-
-	if (rq->rq_disk) {
-		ide_driver_t *drv;
-
-		drv = *(ide_driver_t **)rq->rq_disk->private_data;
-		return drv->abort(drive, rq);
-	} else
-		return __ide_abort(drive, rq);
-}
-
 static void ide_tf_set_specify_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 {
 	tf->nsect   = drive->sect;
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index f9cf1670e4e1..b711ab96e287 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -2591,7 +2591,6 @@ static ide_driver_t idetape_driver = {
 	.do_request		= idetape_do_request,
 	.end_request		= idetape_end_request,
 	.error			= __ide_error,
-	.abort			= __ide_abort,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idetape_proc,
 #endif
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 683bce375c74..f843c1383a4b 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -258,19 +258,6 @@ idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 	return ide_stopped;
 }
 
-static ide_startstop_t
-idescsi_atapi_abort(ide_drive_t *drive, struct request *rq)
-{
-	debug_log("%s called for %lu\n", __func__,
-		((struct ide_atapi_pc *) rq->special)->scsi_cmd->serial_number);
-
-	rq->errors |= ERROR_MAX;
-
-	idescsi_end_request(drive, 0, 0);
-
-	return ide_stopped;
-}
-
 static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs)
 {
 	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
@@ -524,7 +511,6 @@ static ide_driver_t idescsi_driver = {
 	.do_request		= idescsi_do_request,
 	.end_request		= idescsi_end_request,
 	.error                  = idescsi_atapi_error,
-	.abort                  = idescsi_atapi_abort,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idescsi_proc,
 #endif
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 021710cc1b1c..4726126f5a59 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -571,8 +571,6 @@ typedef struct hwgroup_s {
 	unsigned int sleeping	: 1;
 		/* BOOL: polling active & poll_timeout field valid */
 	unsigned int polling	: 1;
-	 	/* BOOL: in a polling reset situation. Must not trigger another reset yet */
-	unsigned int resetting  : 1;
 
 		/* current drive */
 	ide_drive_t *drive;
@@ -792,7 +790,6 @@ struct ide_driver_s {
 	ide_startstop_t	(*do_request)(ide_drive_t *, struct request *, sector_t);
 	int		(*end_request)(ide_drive_t *, int, int);
 	ide_startstop_t	(*error)(ide_drive_t *, struct request *rq, u8, u8);
-	ide_startstop_t	(*abort)(ide_drive_t *, struct request *rq);
 	struct device_driver	gen_driver;
 	int		(*probe)(ide_drive_t *);
 	void		(*remove)(ide_drive_t *);
@@ -834,10 +831,6 @@ ide_startstop_t __ide_error(ide_drive_t *, struct request *, u8, u8);
 
 ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, byte stat);
 
-ide_startstop_t __ide_abort(ide_drive_t *, struct request *);
-
-extern ide_startstop_t ide_abort(ide_drive_t *, const char *);
-
 extern void ide_fix_driveid(struct hd_driveid *);
 
 extern void ide_fixstring(u8 *, const int, const int);
-- 
cgit v1.2.3


From ebb12db51f6c13b30752fcf506baad4c617b153c Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 11 Jun 2008 22:04:29 +0200
Subject: Freezer: Introduce PF_FREEZER_NOSIG

The freezer currently attempts to distinguish kernel threads from
user space tasks by checking if their mm pointer is unset and it
does not send fake signals to kernel threads.  However, there are
kernel threads, mostly related to networking, that behave like
user space tasks and may want to be sent a fake signal to be frozen.

Introduce the new process flag PF_FREEZER_NOSIG that will be set
by default for all kernel threads and make the freezer only send
fake signals to the tasks having PF_FREEZER_NOSIG unset.  Provide
the set_freezable_with_signal() function to be called by the kernel
threads that want to be sent a fake signal for freezing.

This patch should not change the freezer's observable behavior.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 include/linux/freezer.h | 10 +++++
 include/linux/sched.h   |  1 +
 kernel/kthread.c        |  2 +-
 kernel/power/process.c  | 97 +++++++++++++++++++++----------------------------
 4 files changed, 54 insertions(+), 56 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/freezer.h b/include/linux/freezer.h
index 08934995c7ab..deddeedf3257 100644
--- a/include/linux/freezer.h
+++ b/include/linux/freezer.h
@@ -127,6 +127,15 @@ static inline void set_freezable(void)
 	current->flags &= ~PF_NOFREEZE;
 }
 
+/*
+ * Tell the freezer that the current task should be frozen by it and that it
+ * should send a fake signal to the task to freeze it.
+ */
+static inline void set_freezable_with_signal(void)
+{
+	current->flags &= ~(PF_NOFREEZE | PF_FREEZER_NOSIG);
+}
+
 /*
  * Freezer-friendly wrappers around wait_event_interruptible() and
  * wait_event_interruptible_timeout(), originally defined in <linux/wait.h>
@@ -174,6 +183,7 @@ static inline void freezer_do_not_count(void) {}
 static inline void freezer_count(void) {}
 static inline int freezer_should_skip(struct task_struct *p) { return 0; }
 static inline void set_freezable(void) {}
+static inline void set_freezable_with_signal(void) {}
 
 #define wait_event_freezable(wq, condition)				\
 		wait_event_interruptible(wq, condition)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 21349173d148..ba2f859c6e4f 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1494,6 +1494,7 @@ static inline void put_task_struct(struct task_struct *t)
 #define PF_MEMPOLICY	0x10000000	/* Non-default NUMA mempolicy */
 #define PF_MUTEX_TESTER	0x20000000	/* Thread belongs to the rt mutex tester */
 #define PF_FREEZER_SKIP	0x40000000	/* Freezer should not count it as freezeable */
+#define PF_FREEZER_NOSIG 0x80000000	/* Freezer won't send signals to it */
 
 /*
  * Only the _current_ task can read/write to tsk->flags, but other
diff --git a/kernel/kthread.c b/kernel/kthread.c
index 97747cdd37c9..ac3fb7326641 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -235,7 +235,7 @@ int kthreadd(void *unused)
 	set_user_nice(tsk, KTHREAD_NICE_LEVEL);
 	set_cpus_allowed(tsk, CPU_MASK_ALL);
 
-	current->flags |= PF_NOFREEZE;
+	current->flags |= PF_NOFREEZE | PF_FREEZER_NOSIG;
 
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
diff --git a/kernel/power/process.c b/kernel/power/process.c
index f1d0b345c9ba..5fb87652f214 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -19,9 +19,6 @@
  */
 #define TIMEOUT	(20 * HZ)
 
-#define FREEZER_KERNEL_THREADS 0
-#define FREEZER_USER_SPACE 1
-
 static inline int freezeable(struct task_struct * p)
 {
 	if ((p == current) ||
@@ -84,63 +81,53 @@ static void fake_signal_wake_up(struct task_struct *p)
 	spin_unlock_irqrestore(&p->sighand->siglock, flags);
 }
 
-static int has_mm(struct task_struct *p)
+static inline bool should_send_signal(struct task_struct *p)
 {
-	return (p->mm && !(p->flags & PF_BORROWED_MM));
+	return !(p->flags & PF_FREEZER_NOSIG);
 }
 
 /**
  *	freeze_task - send a freeze request to given task
  *	@p: task to send the request to
- *	@with_mm_only: if set, the request will only be sent if the task has its
- *		own mm
- *	Return value: 0, if @with_mm_only is set and the task has no mm of its
- *		own or the task is frozen, 1, otherwise
+ *	@sig_only: if set, the request will only be sent if the task has the
+ *		PF_FREEZER_NOSIG flag unset
+ *	Return value: 'false', if @sig_only is set and the task has
+ *		PF_FREEZER_NOSIG set or the task is frozen, 'true', otherwise
  *
- *	The freeze request is sent by seting the tasks's TIF_FREEZE flag and
+ *	The freeze request is sent by setting the tasks's TIF_FREEZE flag and
  *	either sending a fake signal to it or waking it up, depending on whether
- *	or not it has its own mm (ie. it is a user land task).  If @with_mm_only
- *	is set and the task has no mm of its own (ie. it is a kernel thread),
- *	its TIF_FREEZE flag should not be set.
- *
- *	The task_lock() is necessary to prevent races with exit_mm() or
- *	use_mm()/unuse_mm() from occuring.
+ *	or not it has PF_FREEZER_NOSIG set.  If @sig_only is set and the task
+ *	has PF_FREEZER_NOSIG set (ie. it is a typical kernel thread), its
+ *	TIF_FREEZE flag will not be set.
  */
-static int freeze_task(struct task_struct *p, int with_mm_only)
+static bool freeze_task(struct task_struct *p, bool sig_only)
 {
-	int ret = 1;
+	/*
+	 * We first check if the task is freezing and next if it has already
+	 * been frozen to avoid the race with frozen_process() which first marks
+	 * the task as frozen and next clears its TIF_FREEZE.
+	 */
+	if (!freezing(p)) {
+		rmb();
+		if (frozen(p))
+			return false;
 
-	task_lock(p);
-	if (freezing(p)) {
-		if (has_mm(p)) {
-			if (!signal_pending(p))
-				fake_signal_wake_up(p);
-		} else {
-			if (with_mm_only)
-				ret = 0;
-			else
-				wake_up_state(p, TASK_INTERRUPTIBLE);
-		}
+		if (!sig_only || should_send_signal(p))
+			set_freeze_flag(p);
+		else
+			return false;
+	}
+
+	if (should_send_signal(p)) {
+		if (!signal_pending(p))
+			fake_signal_wake_up(p);
+	} else if (sig_only) {
+		return false;
 	} else {
-		rmb();
-		if (frozen(p)) {
-			ret = 0;
-		} else {
-			if (has_mm(p)) {
-				set_freeze_flag(p);
-				fake_signal_wake_up(p);
-			} else {
-				if (with_mm_only) {
-					ret = 0;
-				} else {
-					set_freeze_flag(p);
-					wake_up_state(p, TASK_INTERRUPTIBLE);
-				}
-			}
-		}
+		wake_up_state(p, TASK_INTERRUPTIBLE);
 	}
-	task_unlock(p);
-	return ret;
+
+	return true;
 }
 
 static void cancel_freezing(struct task_struct *p)
@@ -156,7 +143,7 @@ static void cancel_freezing(struct task_struct *p)
 	}
 }
 
-static int try_to_freeze_tasks(int freeze_user_space)
+static int try_to_freeze_tasks(bool sig_only)
 {
 	struct task_struct *g, *p;
 	unsigned long end_time;
@@ -175,7 +162,7 @@ static int try_to_freeze_tasks(int freeze_user_space)
 			if (frozen(p) || !freezeable(p))
 				continue;
 
-			if (!freeze_task(p, freeze_user_space))
+			if (!freeze_task(p, sig_only))
 				continue;
 
 			/*
@@ -235,13 +222,13 @@ int freeze_processes(void)
 	int error;
 
 	printk("Freezing user space processes ... ");
-	error = try_to_freeze_tasks(FREEZER_USER_SPACE);
+	error = try_to_freeze_tasks(true);
 	if (error)
 		goto Exit;
 	printk("done.\n");
 
 	printk("Freezing remaining freezable tasks ... ");
-	error = try_to_freeze_tasks(FREEZER_KERNEL_THREADS);
+	error = try_to_freeze_tasks(false);
 	if (error)
 		goto Exit;
 	printk("done.");
@@ -251,7 +238,7 @@ int freeze_processes(void)
 	return error;
 }
 
-static void thaw_tasks(int thaw_user_space)
+static void thaw_tasks(bool nosig_only)
 {
 	struct task_struct *g, *p;
 
@@ -260,7 +247,7 @@ static void thaw_tasks(int thaw_user_space)
 		if (!freezeable(p))
 			continue;
 
-		if (!p->mm == thaw_user_space)
+		if (nosig_only && should_send_signal(p))
 			continue;
 
 		thaw_process(p);
@@ -271,8 +258,8 @@ static void thaw_tasks(int thaw_user_space)
 void thaw_processes(void)
 {
 	printk("Restarting tasks ... ");
-	thaw_tasks(FREEZER_KERNEL_THREADS);
-	thaw_tasks(FREEZER_USER_SPACE);
+	thaw_tasks(true);
+	thaw_tasks(false);
 	schedule();
 	printk("done.\n");
 }
-- 
cgit v1.2.3


From 20bfdbba7212d19613b93dcea93f26cb65af91fe Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 27 Jun 2008 16:56:56 -0600
Subject: PNP: make pnp_{port,mem,etc}_start(), et al work for invalid
 resources

Some callers use pnp_port_start() and similar functions without
making sure the resource is valid.  This patch makes us fall
back to returning the initial values if the resource is not
valid or not even present.

This mostly preserves the previous behavior, where we would just
return the initial values set by pnp_init_resource_table().  The
original 2.6.25 code didn't range-check the "bar", so it would
return garbage if the bar exceeded the table size.  This code
returns sensible values instead.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 include/linux/pnp.h | 72 ++++++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 60 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pnp.h b/include/linux/pnp.h
index 63b128d512fb..8b607aecd959 100644
--- a/include/linux/pnp.h
+++ b/include/linux/pnp.h
@@ -40,19 +40,31 @@ static inline resource_size_t pnp_resource_len(struct resource *res)
 static inline resource_size_t pnp_port_start(struct pnp_dev *dev,
 					     unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_IO, bar)->start;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_IO, bar);
+
+	if (pnp_resource_valid(res))
+		return res->start;
+	return 0;
 }
 
 static inline resource_size_t pnp_port_end(struct pnp_dev *dev,
 					   unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_IO, bar)->end;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_IO, bar);
+
+	if (pnp_resource_valid(res))
+		return res->end;
+	return 0;
 }
 
 static inline unsigned long pnp_port_flags(struct pnp_dev *dev,
 					   unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_IO, bar)->flags;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_IO, bar);
+
+	if (pnp_resource_valid(res))
+		return res->flags;
+	return IORESOURCE_IO | IORESOURCE_AUTO | IORESOURCE_UNSET;
 }
 
 static inline int pnp_port_valid(struct pnp_dev *dev, unsigned int bar)
@@ -63,25 +75,41 @@ static inline int pnp_port_valid(struct pnp_dev *dev, unsigned int bar)
 static inline resource_size_t pnp_port_len(struct pnp_dev *dev,
 					   unsigned int bar)
 {
-	return pnp_resource_len(pnp_get_resource(dev, IORESOURCE_IO, bar));
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_IO, bar);
+
+	if (pnp_resource_valid(res))
+		return pnp_resource_len(res);
+	return 0;
 }
 
 
 static inline resource_size_t pnp_mem_start(struct pnp_dev *dev,
 					    unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_MEM, bar)->start;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_MEM, bar);
+
+	if (pnp_resource_valid(res))
+		return res->start;
+	return 0;
 }
 
 static inline resource_size_t pnp_mem_end(struct pnp_dev *dev,
 					  unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_MEM, bar)->end;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_MEM, bar);
+
+	if (pnp_resource_valid(res))
+		return res->end;
+	return 0;
 }
 
 static inline unsigned long pnp_mem_flags(struct pnp_dev *dev, unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_MEM, bar)->flags;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_MEM, bar);
+
+	if (pnp_resource_valid(res))
+		return res->flags;
+	return IORESOURCE_MEM | IORESOURCE_AUTO | IORESOURCE_UNSET;
 }
 
 static inline int pnp_mem_valid(struct pnp_dev *dev, unsigned int bar)
@@ -92,18 +120,30 @@ static inline int pnp_mem_valid(struct pnp_dev *dev, unsigned int bar)
 static inline resource_size_t pnp_mem_len(struct pnp_dev *dev,
 					  unsigned int bar)
 {
-	return pnp_resource_len(pnp_get_resource(dev, IORESOURCE_MEM, bar));
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_MEM, bar);
+
+	if (pnp_resource_valid(res))
+		return pnp_resource_len(res);
+	return 0;
 }
 
 
 static inline resource_size_t pnp_irq(struct pnp_dev *dev, unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_IRQ, bar)->start;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_IRQ, bar);
+
+	if (pnp_resource_valid(res))
+		return res->start;
+	return -1;
 }
 
 static inline unsigned long pnp_irq_flags(struct pnp_dev *dev, unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_IRQ, bar)->flags;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_IRQ, bar);
+
+	if (pnp_resource_valid(res))
+		return res->flags;
+	return IORESOURCE_IRQ | IORESOURCE_AUTO | IORESOURCE_UNSET;
 }
 
 static inline int pnp_irq_valid(struct pnp_dev *dev, unsigned int bar)
@@ -114,12 +154,20 @@ static inline int pnp_irq_valid(struct pnp_dev *dev, unsigned int bar)
 
 static inline resource_size_t pnp_dma(struct pnp_dev *dev, unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_DMA, bar)->start;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_DMA, bar);
+
+	if (pnp_resource_valid(res))
+		return res->start;
+	return -1;
 }
 
 static inline unsigned long pnp_dma_flags(struct pnp_dev *dev, unsigned int bar)
 {
-	return pnp_get_resource(dev, IORESOURCE_DMA, bar)->flags;
+	struct resource *res = pnp_get_resource(dev, IORESOURCE_DMA, bar);
+
+	if (pnp_resource_valid(res))
+		return res->flags;
+	return IORESOURCE_DMA | IORESOURCE_AUTO | IORESOURCE_UNSET;
 }
 
 static inline int pnp_dma_valid(struct pnp_dev *dev, unsigned int bar)
-- 
cgit v1.2.3


From aee3ad815dd291a7193ab01da0f1a30c84d00061 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 27 Jun 2008 16:56:57 -0600
Subject: PNP: replace pnp_resource_table with dynamically allocated resources

PNP used to have a fixed-size pnp_resource_table for tracking the
resources used by a device.  This table often overflowed, so we've
had to increase the table size, which wastes memory because most
devices have very few resources.

This patch replaces the table with a linked list of resources where
the entries are allocated on demand.

This removes messages like these:

    pnpacpi: exceeded the max number of IO resources
    00:01: too many I/O port resources

References:

    http://bugzilla.kernel.org/show_bug.cgi?id=9535
    http://bugzilla.kernel.org/show_bug.cgi?id=9740
    http://lkml.org/lkml/2007/11/30/110

This patch also changes the way PNP uses the IORESOURCE_UNSET,
IORESOURCE_AUTO, and IORESOURCE_DISABLED flags.

Prior to this patch, the pnp_resource_table entries used the flags
like this:

    IORESOURCE_UNSET
	This table entry is unused and available for use.  When this flag
	is set, we shouldn't look at anything else in the resource structure.
	This flag is set when a resource table entry is initialized.

    IORESOURCE_AUTO
	This resource was assigned automatically by pnp_assign_{io,mem,etc}().

	This flag is set when a resource table entry is initialized and
	cleared whenever we discover a resource setting by reading an ISAPNP
	config register, parsing a PNPBIOS resource data stream, parsing an
	ACPI _CRS list, or interpreting a sysfs "set" command.

	Resources marked IORESOURCE_AUTO are reinitialized and marked as
	IORESOURCE_UNSET by pnp_clean_resource_table() in these cases:

	    - before we attempt to assign resources automatically,
	    - if we fail to assign resources automatically,
	    - after disabling a device

    IORESOURCE_DISABLED
	Set by pnp_assign_{io,mem,etc}() when automatic assignment fails.
	Also set by PNPBIOS and PNPACPI for:

	    - invalid IRQs or GSI registration failures
	    - invalid DMA channels
	    - I/O ports above 0x10000
	    - mem ranges with negative length

After this patch, there is no pnp_resource_table, and the resource list
entries use the flags like this:

    IORESOURCE_UNSET
	This flag is no longer used in PNP.  Instead of keeping
	IORESOURCE_UNSET entries in the resource list, we remove
	entries from the list and free them.

    IORESOURCE_AUTO
	No change in meaning: it still means the resource was assigned
	automatically by pnp_assign_{port,mem,etc}(), but these functions
	now set the bit explicitly.

	We still "clean" a device's resource list in the same places,
	but rather than reinitializing IORESOURCE_AUTO entries, we
	just remove them from the list.

	Note that IORESOURCE_AUTO entries are always at the end of the
	list, so removing them doesn't reorder other list entries.
	This is because non-IORESOURCE_AUTO entries are added by the
	ISAPNP, PNPBIOS, or PNPACPI "get resources" methods and by the
	sysfs "set" command.  In each of these cases, we completely free
	the resource list first.

    IORESOURCE_DISABLED
	In addition to the cases where we used to set this flag, ISAPNP now
	adds an IORESOURCE_DISABLED resource when it reads a configuration
	register with a "disabled" value.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/pnp/base.h             |  18 +---
 drivers/pnp/core.c             |  25 +++--
 drivers/pnp/interface.c        |  60 +++++------
 drivers/pnp/isapnp/core.c      |  12 +--
 drivers/pnp/manager.c          | 220 +++++++++++------------------------------
 drivers/pnp/pnpacpi/rsparser.c | 131 ++++++++++++++++--------
 drivers/pnp/pnpbios/rsparser.c |  95 +++++++++++++-----
 drivers/pnp/quirks.c           |   3 +-
 drivers/pnp/resource.c         |  86 ++++------------
 drivers/pnp/support.c          |  79 +++++++--------
 drivers/pnp/system.c           |   4 +-
 include/linux/pnp.h            |  20 ++--
 12 files changed, 331 insertions(+), 422 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pnp/base.h b/drivers/pnp/base.h
index c91315826da2..1667ac3ca45b 100644
--- a/drivers/pnp/base.h
+++ b/drivers/pnp/base.h
@@ -46,27 +46,15 @@ int pnp_check_dma(struct pnp_dev *dev, struct resource *res);
 char *pnp_resource_type_name(struct resource *res);
 void dbg_pnp_show_resources(struct pnp_dev *dev, char *desc);
 
-void pnp_init_resource(struct resource *res);
+void pnp_free_resources(struct pnp_dev *dev);
 int pnp_resource_type(struct resource *res);
 
-struct pnp_resource *pnp_get_pnp_resource(struct pnp_dev *dev,
-					  unsigned int type, unsigned int num);
-
-#define PNP_MAX_PORT		40
-#define PNP_MAX_MEM		24
-#define PNP_MAX_IRQ		 2
-#define PNP_MAX_DMA		 2
-
 struct pnp_resource {
+	struct list_head list;
 	struct resource res;
 };
 
-struct pnp_resource_table {
-	struct pnp_resource port[PNP_MAX_PORT];
-	struct pnp_resource mem[PNP_MAX_MEM];
-	struct pnp_resource dma[PNP_MAX_DMA];
-	struct pnp_resource irq[PNP_MAX_IRQ];
-};
+void pnp_free_resource(struct pnp_resource *pnp_res);
 
 struct pnp_resource *pnp_add_irq_resource(struct pnp_dev *dev, int irq,
 					  int flags);
diff --git a/drivers/pnp/core.c b/drivers/pnp/core.c
index 20771b7d4482..7182da92aec3 100644
--- a/drivers/pnp/core.c
+++ b/drivers/pnp/core.c
@@ -99,6 +99,21 @@ static void pnp_free_ids(struct pnp_dev *dev)
 	}
 }
 
+void pnp_free_resource(struct pnp_resource *pnp_res)
+{
+	list_del(&pnp_res->list);
+	kfree(pnp_res);
+}
+
+void pnp_free_resources(struct pnp_dev *dev)
+{
+	struct pnp_resource *pnp_res, *tmp;
+
+	list_for_each_entry_safe(pnp_res, tmp, &dev->resources, list) {
+		pnp_free_resource(pnp_res);
+	}
+}
+
 static void pnp_release_device(struct device *dmdev)
 {
 	struct pnp_dev *dev = to_pnp_dev(dmdev);
@@ -106,7 +121,7 @@ static void pnp_release_device(struct device *dmdev)
 	pnp_free_option(dev->independent);
 	pnp_free_option(dev->dependent);
 	pnp_free_ids(dev);
-	kfree(dev->res);
+	pnp_free_resources(dev);
 	kfree(dev);
 }
 
@@ -119,12 +134,7 @@ struct pnp_dev *pnp_alloc_dev(struct pnp_protocol *protocol, int id, char *pnpid
 	if (!dev)
 		return NULL;
 
-	dev->res = kzalloc(sizeof(struct pnp_resource_table), GFP_KERNEL);
-	if (!dev->res) {
-		kfree(dev);
-		return NULL;
-	}
-
+	INIT_LIST_HEAD(&dev->resources);
 	dev->protocol = protocol;
 	dev->number = id;
 	dev->dma_mask = DMA_24BIT_MASK;
@@ -140,7 +150,6 @@ struct pnp_dev *pnp_alloc_dev(struct pnp_protocol *protocol, int id, char *pnpid
 
 	dev_id = pnp_add_id(dev, pnpid);
 	if (!dev_id) {
-		kfree(dev->res);
 		kfree(dev);
 		return NULL;
 	}
diff --git a/drivers/pnp/interface.c b/drivers/pnp/interface.c
index 3f8007ab94e3..7fc86bbed88e 100644
--- a/drivers/pnp/interface.c
+++ b/drivers/pnp/interface.c
@@ -269,46 +269,38 @@ static ssize_t pnp_show_current_resources(struct device *dmdev,
 		pnp_printf(buffer, "disabled\n");
 
 	for (i = 0; (res = pnp_get_resource(dev, IORESOURCE_IO, i)); i++) {
-		if (pnp_resource_valid(res)) {
-			pnp_printf(buffer, "io");
-			if (res->flags & IORESOURCE_DISABLED)
-				pnp_printf(buffer, " disabled\n");
-			else
-				pnp_printf(buffer, " 0x%llx-0x%llx\n",
-					   (unsigned long long) res->start,
-					   (unsigned long long) res->end);
-		}
+		pnp_printf(buffer, "io");
+		if (res->flags & IORESOURCE_DISABLED)
+			pnp_printf(buffer, " disabled\n");
+		else
+			pnp_printf(buffer, " 0x%llx-0x%llx\n",
+				   (unsigned long long) res->start,
+				   (unsigned long long) res->end);
 	}
 	for (i = 0; (res = pnp_get_resource(dev, IORESOURCE_MEM, i)); i++) {
-		if (pnp_resource_valid(res)) {
-			pnp_printf(buffer, "mem");
-			if (res->flags & IORESOURCE_DISABLED)
-				pnp_printf(buffer, " disabled\n");
-			else
-				pnp_printf(buffer, " 0x%llx-0x%llx\n",
-					   (unsigned long long) res->start,
-					   (unsigned long long) res->end);
-		}
+		pnp_printf(buffer, "mem");
+		if (res->flags & IORESOURCE_DISABLED)
+			pnp_printf(buffer, " disabled\n");
+		else
+			pnp_printf(buffer, " 0x%llx-0x%llx\n",
+				   (unsigned long long) res->start,
+				   (unsigned long long) res->end);
 	}
 	for (i = 0; (res = pnp_get_resource(dev, IORESOURCE_IRQ, i)); i++) {
-		if (pnp_resource_valid(res)) {
-			pnp_printf(buffer, "irq");
-			if (res->flags & IORESOURCE_DISABLED)
-				pnp_printf(buffer, " disabled\n");
-			else
-				pnp_printf(buffer, " %lld\n",
-					   (unsigned long long) res->start);
-		}
+		pnp_printf(buffer, "irq");
+		if (res->flags & IORESOURCE_DISABLED)
+			pnp_printf(buffer, " disabled\n");
+		else
+			pnp_printf(buffer, " %lld\n",
+				   (unsigned long long) res->start);
 	}
 	for (i = 0; (res = pnp_get_resource(dev, IORESOURCE_DMA, i)); i++) {
-		if (pnp_resource_valid(res)) {
-			pnp_printf(buffer, "dma");
-			if (res->flags & IORESOURCE_DISABLED)
-				pnp_printf(buffer, " disabled\n");
-			else
-				pnp_printf(buffer, " %lld\n",
-					   (unsigned long long) res->start);
-		}
+		pnp_printf(buffer, "dma");
+		if (res->flags & IORESOURCE_DISABLED)
+			pnp_printf(buffer, " disabled\n");
+		else
+			pnp_printf(buffer, " %lld\n",
+				   (unsigned long long) res->start);
 	}
 	ret = (buffer->curr - buf);
 	kfree(buffer);
diff --git a/drivers/pnp/isapnp/core.c b/drivers/pnp/isapnp/core.c
index 752b51fbaa6c..ca4457ec403b 100644
--- a/drivers/pnp/isapnp/core.c
+++ b/drivers/pnp/isapnp/core.c
@@ -973,8 +973,7 @@ static int isapnp_set_resources(struct pnp_dev *dev)
 	dev->active = 1;
 	for (tmp = 0; tmp < ISAPNP_MAX_PORT; tmp++) {
 		res = pnp_get_resource(dev, IORESOURCE_IO, tmp);
-		if (res && pnp_resource_valid(res) &&
-		    !(res->flags & IORESOURCE_DISABLED)) {
+		if (pnp_resource_enabled(res)) {
 			dev_dbg(&dev->dev, "  set io  %d to %#llx\n",
 				tmp, (unsigned long long) res->start);
 			isapnp_write_word(ISAPNP_CFG_PORT + (tmp << 1),
@@ -983,8 +982,7 @@ static int isapnp_set_resources(struct pnp_dev *dev)
 	}
 	for (tmp = 0; tmp < ISAPNP_MAX_IRQ; tmp++) {
 		res = pnp_get_resource(dev, IORESOURCE_IRQ, tmp);
-		if (res && pnp_resource_valid(res) &&
-		    !(res->flags & IORESOURCE_DISABLED)) {
+		if (pnp_resource_enabled(res)) {
 			int irq = res->start;
 			if (irq == 2)
 				irq = 9;
@@ -994,8 +992,7 @@ static int isapnp_set_resources(struct pnp_dev *dev)
 	}
 	for (tmp = 0; tmp < ISAPNP_MAX_DMA; tmp++) {
 		res = pnp_get_resource(dev, IORESOURCE_DMA, tmp);
-		if (res && pnp_resource_valid(res) &&
-		    !(res->flags & IORESOURCE_DISABLED)) {
+		if (pnp_resource_enabled(res)) {
 			dev_dbg(&dev->dev, "  set dma %d to %lld\n",
 				tmp, (unsigned long long) res->start);
 			isapnp_write_byte(ISAPNP_CFG_DMA + tmp, res->start);
@@ -1003,8 +1000,7 @@ static int isapnp_set_resources(struct pnp_dev *dev)
 	}
 	for (tmp = 0; tmp < ISAPNP_MAX_MEM; tmp++) {
 		res = pnp_get_resource(dev, IORESOURCE_MEM, tmp);
-		if (res && pnp_resource_valid(res) &&
-		    !(res->flags & IORESOURCE_DISABLED)) {
+		if (pnp_resource_enabled(res)) {
 			dev_dbg(&dev->dev, "  set mem %d to %#llx\n",
 				tmp, (unsigned long long) res->start);
 			isapnp_write_word(ISAPNP_CFG_MEM + (tmp << 3),
diff --git a/drivers/pnp/manager.c b/drivers/pnp/manager.c
index 90bd9cb65563..165b624081ad 100644
--- a/drivers/pnp/manager.c
+++ b/drivers/pnp/manager.c
@@ -19,40 +19,30 @@ DEFINE_MUTEX(pnp_res_mutex);
 
 static int pnp_assign_port(struct pnp_dev *dev, struct pnp_port *rule, int idx)
 {
-	struct pnp_resource *pnp_res;
-	struct resource *res;
+	struct resource *res, local_res;
 
-	pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_IO, idx);
-	if (!pnp_res) {
-		dev_err(&dev->dev, "too many I/O port resources\n");
-		/* pretend we were successful so at least the manager won't try again */
-		return 1;
-	}
-
-	res = &pnp_res->res;
-
-	/* check if this resource has been manually set, if so skip */
-	if (!(res->flags & IORESOURCE_AUTO)) {
+	res = pnp_get_resource(dev, IORESOURCE_IO, idx);
+	if (res) {
 		dev_dbg(&dev->dev, "  io %d already set to %#llx-%#llx "
 			"flags %#lx\n", idx, (unsigned long long) res->start,
 			(unsigned long long) res->end, res->flags);
 		return 1;
 	}
 
-	/* set the initial values */
-	res->flags |= rule->flags | IORESOURCE_IO;
-	res->flags &= ~IORESOURCE_UNSET;
+	res = &local_res;
+	res->flags = rule->flags | IORESOURCE_AUTO;
+	res->start = 0;
+	res->end = 0;
 
 	if (!rule->size) {
 		res->flags |= IORESOURCE_DISABLED;
 		dev_dbg(&dev->dev, "  io %d disabled\n", idx);
-		return 1;	/* skip disabled resource requests */
+		goto __add;
 	}
 
 	res->start = rule->min;
 	res->end = res->start + rule->size - 1;
 
-	/* run through until pnp_check_port is happy */
 	while (!pnp_check_port(dev, res)) {
 		res->start += rule->align;
 		res->end = res->start + rule->size - 1;
@@ -61,38 +51,29 @@ static int pnp_assign_port(struct pnp_dev *dev, struct pnp_port *rule, int idx)
 			return 0;
 		}
 	}
-	dev_dbg(&dev->dev, "  assign io  %d %#llx-%#llx\n", idx,
-		(unsigned long long) res->start, (unsigned long long) res->end);
+
+__add:
+	pnp_add_io_resource(dev, res->start, res->end, res->flags);
 	return 1;
 }
 
 static int pnp_assign_mem(struct pnp_dev *dev, struct pnp_mem *rule, int idx)
 {
-	struct pnp_resource *pnp_res;
-	struct resource *res;
+	struct resource *res, local_res;
 
-	pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_MEM, idx);
-	if (!pnp_res) {
-		dev_err(&dev->dev, "too many memory resources\n");
-		/* pretend we were successful so at least the manager won't try again */
-		return 1;
-	}
-
-	res = &pnp_res->res;
-
-	/* check if this resource has been manually set, if so skip */
-	if (!(res->flags & IORESOURCE_AUTO)) {
+	res = pnp_get_resource(dev, IORESOURCE_MEM, idx);
+	if (res) {
 		dev_dbg(&dev->dev, "  mem %d already set to %#llx-%#llx "
 			"flags %#lx\n", idx, (unsigned long long) res->start,
 			(unsigned long long) res->end, res->flags);
 		return 1;
 	}
 
-	/* set the initial values */
-	res->flags |= rule->flags | IORESOURCE_MEM;
-	res->flags &= ~IORESOURCE_UNSET;
+	res = &local_res;
+	res->flags = rule->flags | IORESOURCE_AUTO;
+	res->start = 0;
+	res->end = 0;
 
-	/* convert pnp flags to standard Linux flags */
 	if (!(rule->flags & IORESOURCE_MEM_WRITEABLE))
 		res->flags |= IORESOURCE_READONLY;
 	if (rule->flags & IORESOURCE_MEM_CACHEABLE)
@@ -105,13 +86,12 @@ static int pnp_assign_mem(struct pnp_dev *dev, struct pnp_mem *rule, int idx)
 	if (!rule->size) {
 		res->flags |= IORESOURCE_DISABLED;
 		dev_dbg(&dev->dev, "  mem %d disabled\n", idx);
-		return 1;	/* skip disabled resource requests */
+		goto __add;
 	}
 
 	res->start = rule->min;
 	res->end = res->start + rule->size - 1;
 
-	/* run through until pnp_check_mem is happy */
 	while (!pnp_check_mem(dev, res)) {
 		res->start += rule->align;
 		res->end = res->start + rule->size - 1;
@@ -120,15 +100,15 @@ static int pnp_assign_mem(struct pnp_dev *dev, struct pnp_mem *rule, int idx)
 			return 0;
 		}
 	}
-	dev_dbg(&dev->dev, "  assign mem %d %#llx-%#llx\n", idx,
-		(unsigned long long) res->start, (unsigned long long) res->end);
+
+__add:
+	pnp_add_mem_resource(dev, res->start, res->end, res->flags);
 	return 1;
 }
 
 static int pnp_assign_irq(struct pnp_dev *dev, struct pnp_irq *rule, int idx)
 {
-	struct pnp_resource *pnp_res;
-	struct resource *res;
+	struct resource *res, local_res;
 	int i;
 
 	/* IRQ priority: this table is good for i386 */
@@ -136,58 +116,48 @@ static int pnp_assign_irq(struct pnp_dev *dev, struct pnp_irq *rule, int idx)
 		5, 10, 11, 12, 9, 14, 15, 7, 3, 4, 13, 0, 1, 6, 8, 2
 	};
 
-	pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_IRQ, idx);
-	if (!pnp_res) {
-		dev_err(&dev->dev, "too many IRQ resources\n");
-		/* pretend we were successful so at least the manager won't try again */
-		return 1;
-	}
-
-	res = &pnp_res->res;
-
-	/* check if this resource has been manually set, if so skip */
-	if (!(res->flags & IORESOURCE_AUTO)) {
+	res = pnp_get_resource(dev, IORESOURCE_IRQ, idx);
+	if (res) {
 		dev_dbg(&dev->dev, "  irq %d already set to %d flags %#lx\n",
 			idx, (int) res->start, res->flags);
 		return 1;
 	}
 
-	/* set the initial values */
-	res->flags |= rule->flags | IORESOURCE_IRQ;
-	res->flags &= ~IORESOURCE_UNSET;
+	res = &local_res;
+	res->flags = rule->flags | IORESOURCE_AUTO;
+	res->start = -1;
+	res->end = -1;
 
 	if (bitmap_empty(rule->map, PNP_IRQ_NR)) {
 		res->flags |= IORESOURCE_DISABLED;
 		dev_dbg(&dev->dev, "  irq %d disabled\n", idx);
-		return 1;	/* skip disabled resource requests */
+		goto __add;
 	}
 
 	/* TBD: need check for >16 IRQ */
 	res->start = find_next_bit(rule->map, PNP_IRQ_NR, 16);
 	if (res->start < PNP_IRQ_NR) {
 		res->end = res->start;
-		dev_dbg(&dev->dev, "  assign irq %d %d\n", idx,
-			(int) res->start);
-		return 1;
+		goto __add;
 	}
 	for (i = 0; i < 16; i++) {
 		if (test_bit(xtab[i], rule->map)) {
 			res->start = res->end = xtab[i];
-			if (pnp_check_irq(dev, res)) {
-				dev_dbg(&dev->dev, "  assign irq %d %d\n", idx,
-					(int) res->start);
-				return 1;
-			}
+			if (pnp_check_irq(dev, res))
+				goto __add;
 		}
 	}
 	dev_dbg(&dev->dev, "  couldn't assign irq %d\n", idx);
 	return 0;
+
+__add:
+	pnp_add_irq_resource(dev, res->start, res->flags);
+	return 1;
 }
 
 static void pnp_assign_dma(struct pnp_dev *dev, struct pnp_dma *rule, int idx)
 {
-	struct pnp_resource *pnp_res;
-	struct resource *res;
+	struct resource *res, local_res;
 	int i;
 
 	/* DMA priority: this table is good for i386 */
@@ -195,127 +165,47 @@ static void pnp_assign_dma(struct pnp_dev *dev, struct pnp_dma *rule, int idx)
 		1, 3, 5, 6, 7, 0, 2, 4
 	};
 
-	pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_DMA, idx);
-	if (!pnp_res) {
-		dev_err(&dev->dev, "too many DMA resources\n");
-		return;
-	}
-
-	res = &pnp_res->res;
-
-	/* check if this resource has been manually set, if so skip */
-	if (!(res->flags & IORESOURCE_AUTO)) {
+	res = pnp_get_resource(dev, IORESOURCE_DMA, idx);
+	if (res) {
 		dev_dbg(&dev->dev, "  dma %d already set to %d flags %#lx\n",
 			idx, (int) res->start, res->flags);
 		return;
 	}
 
-	/* set the initial values */
-	res->flags |= rule->flags | IORESOURCE_DMA;
-	res->flags &= ~IORESOURCE_UNSET;
+	res = &local_res;
+	res->flags = rule->flags | IORESOURCE_AUTO;
+	res->start = -1;
+	res->end = -1;
 
 	for (i = 0; i < 8; i++) {
 		if (rule->map & (1 << xtab[i])) {
 			res->start = res->end = xtab[i];
-			if (pnp_check_dma(dev, res)) {
-				dev_dbg(&dev->dev, "  assign dma %d %d\n", idx,
-					(int) res->start);
-				return;
-			}
+			if (pnp_check_dma(dev, res))
+				goto __add;
 		}
 	}
 #ifdef MAX_DMA_CHANNELS
 	res->start = res->end = MAX_DMA_CHANNELS;
 #endif
-	res->flags |= IORESOURCE_UNSET | IORESOURCE_DISABLED;
+	res->flags |= IORESOURCE_DISABLED;
 	dev_dbg(&dev->dev, "  disable dma %d\n", idx);
-}
 
-void pnp_init_resource(struct resource *res)
-{
-	unsigned long type;
-
-	type = res->flags & (IORESOURCE_IO  | IORESOURCE_MEM |
-			     IORESOURCE_IRQ | IORESOURCE_DMA);
-
-	res->name = NULL;
-	res->flags = type | IORESOURCE_AUTO | IORESOURCE_UNSET;
-	if (type == IORESOURCE_IRQ || type == IORESOURCE_DMA) {
-		res->start = -1;
-		res->end = -1;
-	} else {
-		res->start = 0;
-		res->end = 0;
-	}
+__add:
+	pnp_add_dma_resource(dev, res->start, res->flags);
 }
 
-/**
- * pnp_init_resources - Resets a resource table to default values.
- * @table: pointer to the desired resource table
- */
 void pnp_init_resources(struct pnp_dev *dev)
 {
-	struct resource *res;
-	int idx;
-
-	for (idx = 0; idx < PNP_MAX_IRQ; idx++) {
-		res = &dev->res->irq[idx].res;
-		res->flags = IORESOURCE_IRQ;
-		pnp_init_resource(res);
-	}
-	for (idx = 0; idx < PNP_MAX_DMA; idx++) {
-		res = &dev->res->dma[idx].res;
-		res->flags = IORESOURCE_DMA;
-		pnp_init_resource(res);
-	}
-	for (idx = 0; idx < PNP_MAX_PORT; idx++) {
-		res = &dev->res->port[idx].res;
-		res->flags = IORESOURCE_IO;
-		pnp_init_resource(res);
-	}
-	for (idx = 0; idx < PNP_MAX_MEM; idx++) {
-		res = &dev->res->mem[idx].res;
-		res->flags = IORESOURCE_MEM;
-		pnp_init_resource(res);
-	}
+	pnp_free_resources(dev);
 }
 
-/**
- * pnp_clean_resources - clears resources that were not manually set
- * @res: the resources to clean
- */
 static void pnp_clean_resource_table(struct pnp_dev *dev)
 {
-	struct resource *res;
-	int idx;
-
-	for (idx = 0; idx < PNP_MAX_IRQ; idx++) {
-		res = &dev->res->irq[idx].res;
-		if (res->flags & IORESOURCE_AUTO) {
-			res->flags = IORESOURCE_IRQ;
-			pnp_init_resource(res);
-		}
-	}
-	for (idx = 0; idx < PNP_MAX_DMA; idx++) {
-		res = &dev->res->dma[idx].res;
-		if (res->flags & IORESOURCE_AUTO) {
-			res->flags = IORESOURCE_DMA;
-			pnp_init_resource(res);
-		}
-	}
-	for (idx = 0; idx < PNP_MAX_PORT; idx++) {
-		res = &dev->res->port[idx].res;
-		if (res->flags & IORESOURCE_AUTO) {
-			res->flags = IORESOURCE_IO;
-			pnp_init_resource(res);
-		}
-	}
-	for (idx = 0; idx < PNP_MAX_MEM; idx++) {
-		res = &dev->res->mem[idx].res;
-		if (res->flags & IORESOURCE_AUTO) {
-			res->flags = IORESOURCE_MEM;
-			pnp_init_resource(res);
-		}
+	struct pnp_resource *pnp_res, *tmp;
+
+	list_for_each_entry_safe(pnp_res, tmp, &dev->resources, list) {
+		if (pnp_res->res.flags & IORESOURCE_AUTO)
+			pnp_free_resource(pnp_res);
 	}
 }
 
diff --git a/drivers/pnp/pnpacpi/rsparser.c b/drivers/pnp/pnpacpi/rsparser.c
index 46c791adb894..9a45c25b46d2 100644
--- a/drivers/pnp/pnpacpi/rsparser.c
+++ b/drivers/pnp/pnpacpi/rsparser.c
@@ -806,6 +806,13 @@ static void pnpacpi_encode_irq(struct pnp_dev *dev,
 	struct acpi_resource_irq *irq = &resource->data.irq;
 	int triggering, polarity, shareable;
 
+	if (!pnp_resource_enabled(p)) {
+		irq->interrupt_count = 0;
+		dev_dbg(&dev->dev, "  encode irq (%s)\n",
+			p ? "disabled" : "missing");
+		return;
+	}
+
 	decode_irq_flags(dev, p->flags, &triggering, &polarity, &shareable);
 	irq->triggering = triggering;
 	irq->polarity = polarity;
@@ -828,6 +835,13 @@ static void pnpacpi_encode_ext_irq(struct pnp_dev *dev,
 	struct acpi_resource_extended_irq *extended_irq = &resource->data.extended_irq;
 	int triggering, polarity, shareable;
 
+	if (!pnp_resource_enabled(p)) {
+		extended_irq->interrupt_count = 0;
+		dev_dbg(&dev->dev, "  encode extended irq (%s)\n",
+			p ? "disabled" : "missing");
+		return;
+	}
+
 	decode_irq_flags(dev, p->flags, &triggering, &polarity, &shareable);
 	extended_irq->producer_consumer = ACPI_CONSUMER;
 	extended_irq->triggering = triggering;
@@ -848,6 +862,13 @@ static void pnpacpi_encode_dma(struct pnp_dev *dev,
 {
 	struct acpi_resource_dma *dma = &resource->data.dma;
 
+	if (!pnp_resource_enabled(p)) {
+		dma->channel_count = 0;
+		dev_dbg(&dev->dev, "  encode dma (%s)\n",
+			p ? "disabled" : "missing");
+		return;
+	}
+
 	/* Note: pnp_assign_dma will copy pnp_dma->flags into p->flags */
 	switch (p->flags & IORESOURCE_DMA_SPEED_MASK) {
 	case IORESOURCE_DMA_TYPEA:
@@ -889,17 +910,21 @@ static void pnpacpi_encode_io(struct pnp_dev *dev,
 {
 	struct acpi_resource_io *io = &resource->data.io;
 
-	/* Note: pnp_assign_port will copy pnp_port->flags into p->flags */
-	io->io_decode = (p->flags & PNP_PORT_FLAG_16BITADDR) ?
-	    ACPI_DECODE_16 : ACPI_DECODE_10;
-	io->minimum = p->start;
-	io->maximum = p->end;
-	io->alignment = 0;	/* Correct? */
-	io->address_length = p->end - p->start + 1;
-
-	dev_dbg(&dev->dev, "  encode io %#llx-%#llx decode %#x\n",
-		(unsigned long long) p->start, (unsigned long long) p->end,
-		io->io_decode);
+	if (pnp_resource_enabled(p)) {
+		/* Note: pnp_assign_port copies pnp_port->flags into p->flags */
+		io->io_decode = (p->flags & PNP_PORT_FLAG_16BITADDR) ?
+		    ACPI_DECODE_16 : ACPI_DECODE_10;
+		io->minimum = p->start;
+		io->maximum = p->end;
+		io->alignment = 0;	/* Correct? */
+		io->address_length = p->end - p->start + 1;
+	} else {
+		io->minimum = 0;
+		io->address_length = 0;
+	}
+
+	dev_dbg(&dev->dev, "  encode io %#x-%#x decode %#x\n", io->minimum,
+		io->minimum + io->address_length - 1, io->io_decode);
 }
 
 static void pnpacpi_encode_fixed_io(struct pnp_dev *dev,
@@ -908,11 +933,16 @@ static void pnpacpi_encode_fixed_io(struct pnp_dev *dev,
 {
 	struct acpi_resource_fixed_io *fixed_io = &resource->data.fixed_io;
 
-	fixed_io->address = p->start;
-	fixed_io->address_length = p->end - p->start + 1;
+	if (pnp_resource_enabled(p)) {
+		fixed_io->address = p->start;
+		fixed_io->address_length = p->end - p->start + 1;
+	} else {
+		fixed_io->address = 0;
+		fixed_io->address_length = 0;
+	}
 
-	dev_dbg(&dev->dev, "  encode fixed_io %#llx-%#llx\n",
-		(unsigned long long) p->start, (unsigned long long) p->end);
+	dev_dbg(&dev->dev, "  encode fixed_io %#x-%#x\n", fixed_io->address,
+		fixed_io->address + fixed_io->address_length - 1);
 }
 
 static void pnpacpi_encode_mem24(struct pnp_dev *dev,
@@ -921,17 +951,22 @@ static void pnpacpi_encode_mem24(struct pnp_dev *dev,
 {
 	struct acpi_resource_memory24 *memory24 = &resource->data.memory24;
 
-	/* Note: pnp_assign_mem will copy pnp_mem->flags into p->flags */
-	memory24->write_protect =
-	    (p->flags & IORESOURCE_MEM_WRITEABLE) ?
-	    ACPI_READ_WRITE_MEMORY : ACPI_READ_ONLY_MEMORY;
-	memory24->minimum = p->start;
-	memory24->maximum = p->end;
-	memory24->alignment = 0;
-	memory24->address_length = p->end - p->start + 1;
-
-	dev_dbg(&dev->dev, "  encode mem24 %#llx-%#llx write_protect %#x\n",
-		(unsigned long long) p->start, (unsigned long long) p->end,
+	if (pnp_resource_enabled(p)) {
+		/* Note: pnp_assign_mem copies pnp_mem->flags into p->flags */
+		memory24->write_protect = p->flags & IORESOURCE_MEM_WRITEABLE ?
+		    ACPI_READ_WRITE_MEMORY : ACPI_READ_ONLY_MEMORY;
+		memory24->minimum = p->start;
+		memory24->maximum = p->end;
+		memory24->alignment = 0;
+		memory24->address_length = p->end - p->start + 1;
+	} else {
+		memory24->minimum = 0;
+		memory24->address_length = 0;
+	}
+
+	dev_dbg(&dev->dev, "  encode mem24 %#x-%#x write_protect %#x\n",
+		memory24->minimum,
+		memory24->minimum + memory24->address_length - 1,
 		memory24->write_protect);
 }
 
@@ -941,16 +976,21 @@ static void pnpacpi_encode_mem32(struct pnp_dev *dev,
 {
 	struct acpi_resource_memory32 *memory32 = &resource->data.memory32;
 
-	memory32->write_protect =
-	    (p->flags & IORESOURCE_MEM_WRITEABLE) ?
-	    ACPI_READ_WRITE_MEMORY : ACPI_READ_ONLY_MEMORY;
-	memory32->minimum = p->start;
-	memory32->maximum = p->end;
-	memory32->alignment = 0;
-	memory32->address_length = p->end - p->start + 1;
+	if (pnp_resource_enabled(p)) {
+		memory32->write_protect = p->flags & IORESOURCE_MEM_WRITEABLE ?
+		    ACPI_READ_WRITE_MEMORY : ACPI_READ_ONLY_MEMORY;
+		memory32->minimum = p->start;
+		memory32->maximum = p->end;
+		memory32->alignment = 0;
+		memory32->address_length = p->end - p->start + 1;
+	} else {
+		memory32->minimum = 0;
+		memory32->alignment = 0;
+	}
 
-	dev_dbg(&dev->dev, "  encode mem32 %#llx-%#llx write_protect %#x\n",
-		(unsigned long long) p->start, (unsigned long long) p->end,
+	dev_dbg(&dev->dev, "  encode mem32 %#x-%#x write_protect %#x\n",
+		memory32->minimum,
+		memory32->minimum + memory32->address_length - 1,
 		memory32->write_protect);
 }
 
@@ -960,15 +1000,20 @@ static void pnpacpi_encode_fixed_mem32(struct pnp_dev *dev,
 {
 	struct acpi_resource_fixed_memory32 *fixed_memory32 = &resource->data.fixed_memory32;
 
-	fixed_memory32->write_protect =
-	    (p->flags & IORESOURCE_MEM_WRITEABLE) ?
-	    ACPI_READ_WRITE_MEMORY : ACPI_READ_ONLY_MEMORY;
-	fixed_memory32->address = p->start;
-	fixed_memory32->address_length = p->end - p->start + 1;
+	if (pnp_resource_enabled(p)) {
+		fixed_memory32->write_protect =
+		    p->flags & IORESOURCE_MEM_WRITEABLE ?
+		    ACPI_READ_WRITE_MEMORY : ACPI_READ_ONLY_MEMORY;
+		fixed_memory32->address = p->start;
+		fixed_memory32->address_length = p->end - p->start + 1;
+	} else {
+		fixed_memory32->address = 0;
+		fixed_memory32->address_length = 0;
+	}
 
-	dev_dbg(&dev->dev, "  encode fixed_mem32 %#llx-%#llx "
-		"write_protect %#x\n",
-		(unsigned long long) p->start, (unsigned long long) p->end,
+	dev_dbg(&dev->dev, "  encode fixed_mem32 %#x-%#x write_protect %#x\n",
+		fixed_memory32->address,
+		fixed_memory32->address + fixed_memory32->address_length - 1,
 		fixed_memory32->write_protect);
 }
 
diff --git a/drivers/pnp/pnpbios/rsparser.c b/drivers/pnp/pnpbios/rsparser.c
index 5ff9a4c0447e..01f0c3dd1b08 100644
--- a/drivers/pnp/pnpbios/rsparser.c
+++ b/drivers/pnp/pnpbios/rsparser.c
@@ -526,8 +526,16 @@ len_err:
 static void pnpbios_encode_mem(struct pnp_dev *dev, unsigned char *p,
 			       struct resource *res)
 {
-	unsigned long base = res->start;
-	unsigned long len = res->end - res->start + 1;
+	unsigned long base;
+	unsigned long len;
+
+	if (pnp_resource_enabled(res)) {
+		base = res->start;
+		len = res->end - res->start + 1;
+	} else {
+		base = 0;
+		len = 0;
+	}
 
 	p[4] = (base >> 8) & 0xff;
 	p[5] = ((base >> 8) >> 8) & 0xff;
@@ -536,15 +544,22 @@ static void pnpbios_encode_mem(struct pnp_dev *dev, unsigned char *p,
 	p[10] = (len >> 8) & 0xff;
 	p[11] = ((len >> 8) >> 8) & 0xff;
 
-	dev_dbg(&dev->dev, "  encode mem %#llx-%#llx\n",
-		(unsigned long long) res->start, (unsigned long long) res->end);
+	dev_dbg(&dev->dev, "  encode mem %#lx-%#lx\n", base, base + len - 1);
 }
 
 static void pnpbios_encode_mem32(struct pnp_dev *dev, unsigned char *p,
 				 struct resource *res)
 {
-	unsigned long base = res->start;
-	unsigned long len = res->end - res->start + 1;
+	unsigned long base;
+	unsigned long len;
+
+	if (pnp_resource_enabled(res)) {
+		base = res->start;
+		len = res->end - res->start + 1;
+	} else {
+		base = 0;
+		len = 0;
+	}
 
 	p[4] = base & 0xff;
 	p[5] = (base >> 8) & 0xff;
@@ -559,15 +574,22 @@ static void pnpbios_encode_mem32(struct pnp_dev *dev, unsigned char *p,
 	p[18] = (len >> 16) & 0xff;
 	p[19] = (len >> 24) & 0xff;
 
-	dev_dbg(&dev->dev, "  encode mem32 %#llx-%#llx\n",
-		(unsigned long long) res->start, (unsigned long long) res->end);
+	dev_dbg(&dev->dev, "  encode mem32 %#lx-%#lx\n", base, base + len - 1);
 }
 
 static void pnpbios_encode_fixed_mem32(struct pnp_dev *dev, unsigned char *p,
 				       struct resource *res)
 {
-	unsigned long base = res->start;
-	unsigned long len = res->end - res->start + 1;
+	unsigned long base;
+	unsigned long len;
+
+	if (pnp_resource_enabled(res)) {
+		base = res->start;
+		len = res->end - res->start + 1;
+	} else {
+		base = 0;
+		len = 0;
+	}
 
 	p[4] = base & 0xff;
 	p[5] = (base >> 8) & 0xff;
@@ -578,40 +600,54 @@ static void pnpbios_encode_fixed_mem32(struct pnp_dev *dev, unsigned char *p,
 	p[10] = (len >> 16) & 0xff;
 	p[11] = (len >> 24) & 0xff;
 
-	dev_dbg(&dev->dev, "  encode fixed_mem32 %#llx-%#llx\n",
-		(unsigned long long) res->start, (unsigned long long) res->end);
+	dev_dbg(&dev->dev, "  encode fixed_mem32 %#lx-%#lx\n", base,
+		base + len - 1);
 }
 
 static void pnpbios_encode_irq(struct pnp_dev *dev, unsigned char *p,
 			       struct resource *res)
 {
-	unsigned long map = 0;
+	unsigned long map;
+
+	if (pnp_resource_enabled(res))
+		map = 1 << res->start;
+	else
+		map = 0;
 
-	map = 1 << res->start;
 	p[1] = map & 0xff;
 	p[2] = (map >> 8) & 0xff;
 
-	dev_dbg(&dev->dev, "  encode irq %llu\n",
-		(unsigned long long)res->start);
+	dev_dbg(&dev->dev, "  encode irq mask %#lx\n", map);
 }
 
 static void pnpbios_encode_dma(struct pnp_dev *dev, unsigned char *p,
 			       struct resource *res)
 {
-	unsigned long map = 0;
+	unsigned long map;
+
+	if (pnp_resource_enabled(res))
+		map = 1 << res->start;
+	else
+		map = 0;
 
-	map = 1 << res->start;
 	p[1] = map & 0xff;
 
-	dev_dbg(&dev->dev, "  encode dma %llu\n",
-		(unsigned long long)res->start);
+	dev_dbg(&dev->dev, "  encode dma mask %#lx\n", map);
 }
 
 static void pnpbios_encode_port(struct pnp_dev *dev, unsigned char *p,
 				struct resource *res)
 {
-	unsigned long base = res->start;
-	unsigned long len = res->end - res->start + 1;
+	unsigned long base;
+	unsigned long len;
+
+	if (pnp_resource_enabled(res)) {
+		base = res->start;
+		len = res->end - res->start + 1;
+	} else {
+		base = 0;
+		len = 0;
+	}
 
 	p[2] = base & 0xff;
 	p[3] = (base >> 8) & 0xff;
@@ -619,8 +655,7 @@ static void pnpbios_encode_port(struct pnp_dev *dev, unsigned char *p,
 	p[5] = (base >> 8) & 0xff;
 	p[7] = len & 0xff;
 
-	dev_dbg(&dev->dev, "  encode io %#llx-%#llx\n",
-		(unsigned long long) res->start, (unsigned long long) res->end);
+	dev_dbg(&dev->dev, "  encode io %#lx-%#lx\n", base, base + len - 1);
 }
 
 static void pnpbios_encode_fixed_port(struct pnp_dev *dev, unsigned char *p,
@@ -629,12 +664,20 @@ static void pnpbios_encode_fixed_port(struct pnp_dev *dev, unsigned char *p,
 	unsigned long base = res->start;
 	unsigned long len = res->end - res->start + 1;
 
+	if (pnp_resource_enabled(res)) {
+		base = res->start;
+		len = res->end - res->start + 1;
+	} else {
+		base = 0;
+		len = 0;
+	}
+
 	p[1] = base & 0xff;
 	p[2] = (base >> 8) & 0xff;
 	p[3] = len & 0xff;
 
-	dev_dbg(&dev->dev, "  encode fixed_io %#llx-%#llx\n",
-		(unsigned long long) res->start, (unsigned long long) res->end);
+	dev_dbg(&dev->dev, "  encode fixed_io %#lx-%#lx\n", base,
+		base + len - 1);
 }
 
 static unsigned char *pnpbios_encode_allocated_resource_data(struct pnp_dev
diff --git a/drivers/pnp/quirks.c b/drivers/pnp/quirks.c
index 1ff3bb585ab2..21acb54eff6d 100644
--- a/drivers/pnp/quirks.c
+++ b/drivers/pnp/quirks.c
@@ -248,8 +248,7 @@ static void quirk_system_pci_resources(struct pnp_dev *dev)
 			for (j = 0;
 			     (res = pnp_get_resource(dev, IORESOURCE_MEM, j));
 			     j++) {
-				if (res->flags & IORESOURCE_UNSET ||
-				    (res->start == 0 && res->end == 0))
+				if (res->start == 0 && res->end == 0)
 					continue;
 
 				pnp_start = res->start;
diff --git a/drivers/pnp/resource.c b/drivers/pnp/resource.c
index cce341f743d4..0797a77bd042 100644
--- a/drivers/pnp/resource.c
+++ b/drivers/pnp/resource.c
@@ -237,7 +237,7 @@ void pnp_free_option(struct pnp_option *option)
 	!((*(enda) < *(startb)) || (*(endb) < *(starta)))
 
 #define cannot_compare(flags) \
-((flags) & (IORESOURCE_UNSET | IORESOURCE_DISABLED))
+((flags) & IORESOURCE_DISABLED)
 
 int pnp_check_port(struct pnp_dev *dev, struct resource *res)
 {
@@ -505,81 +505,31 @@ int pnp_resource_type(struct resource *res)
 			     IORESOURCE_IRQ | IORESOURCE_DMA);
 }
 
-struct pnp_resource *pnp_get_pnp_resource(struct pnp_dev *dev,
-					  unsigned int type, unsigned int num)
-{
-	struct pnp_resource_table *res = dev->res;
-
-	switch (type) {
-	case IORESOURCE_IO:
-		if (num >= PNP_MAX_PORT)
-			return NULL;
-		return &res->port[num];
-	case IORESOURCE_MEM:
-		if (num >= PNP_MAX_MEM)
-			return NULL;
-		return &res->mem[num];
-	case IORESOURCE_IRQ:
-		if (num >= PNP_MAX_IRQ)
-			return NULL;
-		return &res->irq[num];
-	case IORESOURCE_DMA:
-		if (num >= PNP_MAX_DMA)
-			return NULL;
-		return &res->dma[num];
-	}
-	return NULL;
-}
-
 struct resource *pnp_get_resource(struct pnp_dev *dev,
 				  unsigned int type, unsigned int num)
 {
 	struct pnp_resource *pnp_res;
+	struct resource *res;
 
-	pnp_res = pnp_get_pnp_resource(dev, type, num);
-	if (pnp_res)
-		return &pnp_res->res;
-
+	list_for_each_entry(pnp_res, &dev->resources, list) {
+		res = &pnp_res->res;
+		if (pnp_resource_type(res) == type && num-- == 0)
+			return res;
+	}
 	return NULL;
 }
 EXPORT_SYMBOL(pnp_get_resource);
 
-static struct pnp_resource *pnp_new_resource(struct pnp_dev *dev, int type)
+static struct pnp_resource *pnp_new_resource(struct pnp_dev *dev)
 {
 	struct pnp_resource *pnp_res;
-	int i;
 
-	switch (type) {
-	case IORESOURCE_IO:
-		for (i = 0; i < PNP_MAX_PORT; i++) {
-			pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_IO, i);
-			if (pnp_res && !pnp_resource_valid(&pnp_res->res))
-				return pnp_res;
-		}
-		break;
-	case IORESOURCE_MEM:
-		for (i = 0; i < PNP_MAX_MEM; i++) {
-			pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_MEM, i);
-			if (pnp_res && !pnp_resource_valid(&pnp_res->res))
-				return pnp_res;
-		}
-		break;
-	case IORESOURCE_IRQ:
-		for (i = 0; i < PNP_MAX_IRQ; i++) {
-			pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_IRQ, i);
-			if (pnp_res && !pnp_resource_valid(&pnp_res->res))
-				return pnp_res;
-		}
-		break;
-	case IORESOURCE_DMA:
-		for (i = 0; i < PNP_MAX_DMA; i++) {
-			pnp_res = pnp_get_pnp_resource(dev, IORESOURCE_DMA, i);
-			if (pnp_res && !pnp_resource_valid(&pnp_res->res))
-				return pnp_res;
-		}
-		break;
-	}
-	return NULL;
+	pnp_res = kzalloc(sizeof(struct pnp_resource), GFP_KERNEL);
+	if (!pnp_res)
+		return NULL;
+
+	list_add_tail(&pnp_res->list, &dev->resources);
+	return pnp_res;
 }
 
 struct pnp_resource *pnp_add_irq_resource(struct pnp_dev *dev, int irq,
@@ -589,7 +539,7 @@ struct pnp_resource *pnp_add_irq_resource(struct pnp_dev *dev, int irq,
 	struct resource *res;
 	static unsigned char warned;
 
-	pnp_res = pnp_new_resource(dev, IORESOURCE_IRQ);
+	pnp_res = pnp_new_resource(dev);
 	if (!pnp_res) {
 		if (!warned) {
 			dev_err(&dev->dev, "can't add resource for IRQ %d\n",
@@ -615,7 +565,7 @@ struct pnp_resource *pnp_add_dma_resource(struct pnp_dev *dev, int dma,
 	struct resource *res;
 	static unsigned char warned;
 
-	pnp_res = pnp_new_resource(dev, IORESOURCE_DMA);
+	pnp_res = pnp_new_resource(dev);
 	if (!pnp_res) {
 		if (!warned) {
 			dev_err(&dev->dev, "can't add resource for DMA %d\n",
@@ -642,7 +592,7 @@ struct pnp_resource *pnp_add_io_resource(struct pnp_dev *dev,
 	struct resource *res;
 	static unsigned char warned;
 
-	pnp_res = pnp_new_resource(dev, IORESOURCE_IO);
+	pnp_res = pnp_new_resource(dev);
 	if (!pnp_res) {
 		if (!warned) {
 			dev_err(&dev->dev, "can't add resource for IO "
@@ -671,7 +621,7 @@ struct pnp_resource *pnp_add_mem_resource(struct pnp_dev *dev,
 	struct resource *res;
 	static unsigned char warned;
 
-	pnp_res = pnp_new_resource(dev, IORESOURCE_MEM);
+	pnp_res = pnp_new_resource(dev);
 	if (!pnp_res) {
 		if (!warned) {
 			dev_err(&dev->dev, "can't add resource for MEM "
diff --git a/drivers/pnp/support.c b/drivers/pnp/support.c
index eb07345f5cf7..1566e4a73849 100644
--- a/drivers/pnp/support.c
+++ b/drivers/pnp/support.c
@@ -16,6 +16,10 @@
  */
 int pnp_is_active(struct pnp_dev *dev)
 {
+	/*
+	 * I don't think this is very reliable because pnp_disable_dev()
+	 * only clears out auto-assigned resources.
+	 */
 	if (!pnp_port_start(dev, 0) && pnp_port_len(dev, 0) <= 1 &&
 	    !pnp_mem_start(dev, 0) && pnp_mem_len(dev, 0) <= 1 &&
 	    pnp_irq(dev, 0) == -1 && pnp_dma(dev, 0) == -1)
@@ -70,54 +74,41 @@ char *pnp_resource_type_name(struct resource *res)
 void dbg_pnp_show_resources(struct pnp_dev *dev, char *desc)
 {
 #ifdef DEBUG
+	char buf[128];
+	int len = 0;
+	struct pnp_resource *pnp_res;
 	struct resource *res;
-	int i;
 
 	dev_dbg(&dev->dev, "current resources: %s\n", desc);
+	list_for_each_entry(pnp_res, &dev->resources, list) {
+		res = &pnp_res->res;
 
-	for (i = 0; i < PNP_MAX_IRQ; i++) {
-		res = pnp_get_resource(dev, IORESOURCE_IRQ, i);
-		if (res && !(res->flags & IORESOURCE_UNSET))
-			dev_dbg(&dev->dev, "  irq %lld flags %#lx%s%s\n",
-				(unsigned long long) res->start, res->flags,
-				res->flags & IORESOURCE_DISABLED ?
-					" DISABLED" : "",
-				res->flags & IORESOURCE_AUTO ?
-					" AUTO" : "");
-	}
-	for (i = 0; i < PNP_MAX_DMA; i++) {
-		res = pnp_get_resource(dev, IORESOURCE_DMA, i);
-		if (res && !(res->flags & IORESOURCE_UNSET))
-			dev_dbg(&dev->dev, "  dma %lld flags %#lx%s%s\n",
-				(unsigned long long) res->start, res->flags,
-				res->flags & IORESOURCE_DISABLED ?
-					" DISABLED" : "",
-				res->flags & IORESOURCE_AUTO ?
-					" AUTO" : "");
-	}
-	for (i = 0; i < PNP_MAX_PORT; i++) {
-		res = pnp_get_resource(dev, IORESOURCE_IO, i);
-		if (res && !(res->flags & IORESOURCE_UNSET))
-			dev_dbg(&dev->dev, "  io  %#llx-%#llx flags %#lx"
-				"%s%s\n",
-				(unsigned long long) res->start,
-				(unsigned long long) res->end, res->flags,
-				res->flags & IORESOURCE_DISABLED ?
-					" DISABLED" : "",
-				res->flags & IORESOURCE_AUTO ?
-					" AUTO" : "");
-	}
-	for (i = 0; i < PNP_MAX_MEM; i++) {
-		res = pnp_get_resource(dev, IORESOURCE_MEM, i);
-		if (res && !(res->flags & IORESOURCE_UNSET))
-			dev_dbg(&dev->dev, "  mem %#llx-%#llx flags %#lx"
-				"%s%s\n",
-				(unsigned long long) res->start,
-				(unsigned long long) res->end, res->flags,
-				res->flags & IORESOURCE_DISABLED ?
-					" DISABLED" : "",
-				res->flags & IORESOURCE_AUTO ?
-					" AUTO" : "");
+		len += snprintf(buf + len, sizeof(buf) - len, "  %-3s ",
+				pnp_resource_type_name(res));
+
+		if (res->flags & IORESOURCE_DISABLED) {
+			dev_dbg(&dev->dev, "%sdisabled\n", buf);
+			continue;
+		}
+
+		switch (pnp_resource_type(res)) {
+		case IORESOURCE_IO:
+		case IORESOURCE_MEM:
+			len += snprintf(buf + len, sizeof(buf) - len,
+					"%#llx-%#llx flags %#lx",
+					(unsigned long long) res->start,
+					(unsigned long long) res->end,
+					res->flags);
+			break;
+		case IORESOURCE_IRQ:
+		case IORESOURCE_DMA:
+			len += snprintf(buf + len, sizeof(buf) - len,
+					"%lld flags %#lx",
+					(unsigned long long) res->start,
+					res->flags);
+			break;
+		}
+		dev_dbg(&dev->dev, "%s\n", buf);
 	}
 #endif
 }
diff --git a/drivers/pnp/system.c b/drivers/pnp/system.c
index cf4e07b01d48..764f3a310685 100644
--- a/drivers/pnp/system.c
+++ b/drivers/pnp/system.c
@@ -60,7 +60,7 @@ static void reserve_resources_of_dev(struct pnp_dev *dev)
 	int i;
 
 	for (i = 0; (res = pnp_get_resource(dev, IORESOURCE_IO, i)); i++) {
-		if (res->flags & IORESOURCE_UNSET)
+		if (res->flags & IORESOURCE_DISABLED)
 			continue;
 		if (res->start == 0)
 			continue;	/* disabled */
@@ -81,7 +81,7 @@ static void reserve_resources_of_dev(struct pnp_dev *dev)
 	}
 
 	for (i = 0; (res = pnp_get_resource(dev, IORESOURCE_MEM, i)); i++) {
-		if (res->flags & (IORESOURCE_UNSET | IORESOURCE_DISABLED))
+		if (res->flags & IORESOURCE_DISABLED)
 			continue;
 
 		reserve_range(dev, res->start, res->end, 0);
diff --git a/include/linux/pnp.h b/include/linux/pnp.h
index 8b607aecd959..dfaa567e04a8 100644
--- a/include/linux/pnp.h
+++ b/include/linux/pnp.h
@@ -15,7 +15,6 @@
 
 struct pnp_protocol;
 struct pnp_dev;
-struct pnp_resource_table;
 
 /*
  * Resource Management
@@ -24,7 +23,14 @@ struct resource *pnp_get_resource(struct pnp_dev *, unsigned int, unsigned int);
 
 static inline int pnp_resource_valid(struct resource *res)
 {
-	if (res && !(res->flags & IORESOURCE_UNSET))
+	if (res)
+		return 1;
+	return 0;
+}
+
+static inline int pnp_resource_enabled(struct resource *res)
+{
+	if (res && !(res->flags & IORESOURCE_DISABLED))
 		return 1;
 	return 0;
 }
@@ -64,7 +70,7 @@ static inline unsigned long pnp_port_flags(struct pnp_dev *dev,
 
 	if (pnp_resource_valid(res))
 		return res->flags;
-	return IORESOURCE_IO | IORESOURCE_AUTO | IORESOURCE_UNSET;
+	return IORESOURCE_IO | IORESOURCE_AUTO;
 }
 
 static inline int pnp_port_valid(struct pnp_dev *dev, unsigned int bar)
@@ -109,7 +115,7 @@ static inline unsigned long pnp_mem_flags(struct pnp_dev *dev, unsigned int bar)
 
 	if (pnp_resource_valid(res))
 		return res->flags;
-	return IORESOURCE_MEM | IORESOURCE_AUTO | IORESOURCE_UNSET;
+	return IORESOURCE_MEM | IORESOURCE_AUTO;
 }
 
 static inline int pnp_mem_valid(struct pnp_dev *dev, unsigned int bar)
@@ -143,7 +149,7 @@ static inline unsigned long pnp_irq_flags(struct pnp_dev *dev, unsigned int bar)
 
 	if (pnp_resource_valid(res))
 		return res->flags;
-	return IORESOURCE_IRQ | IORESOURCE_AUTO | IORESOURCE_UNSET;
+	return IORESOURCE_IRQ | IORESOURCE_AUTO;
 }
 
 static inline int pnp_irq_valid(struct pnp_dev *dev, unsigned int bar)
@@ -167,7 +173,7 @@ static inline unsigned long pnp_dma_flags(struct pnp_dev *dev, unsigned int bar)
 
 	if (pnp_resource_valid(res))
 		return res->flags;
-	return IORESOURCE_DMA | IORESOURCE_AUTO | IORESOURCE_UNSET;
+	return IORESOURCE_DMA | IORESOURCE_AUTO;
 }
 
 static inline int pnp_dma_valid(struct pnp_dev *dev, unsigned int bar)
@@ -296,7 +302,7 @@ struct pnp_dev {
 	int capabilities;
 	struct pnp_option *independent;
 	struct pnp_option *dependent;
-	struct pnp_resource_table *res;
+	struct list_head resources;
 
 	char name[PNP_NAME_LEN];	/* contains a human-readable name */
 	int flags;			/* used by protocols */
-- 
cgit v1.2.3


From 57fd51a8be26921b56747ddd09d1d9e01c11c9e0 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 27 Jun 2008 16:57:01 -0600
Subject: PNP: add pnp_possible_config() -- can a device could be configured
 this way?

As part of a heuristic to identify modem devices, 8250_pnp.c
checks to see whether a device can be configured at any of the
legacy COM port addresses.

This patch moves the code that traverses the PNP "possible resource
options" from 8250_pnp.c to the PNP subsystem.  This encapsulation
is important because a future patch will change the implementation
of those resource options.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/pnp/resource.c    | 62 +++++++++++++++++++++++++++++++++++++++++++++++
 drivers/serial/8250_pnp.c | 24 ++++++------------
 include/linux/pnp.h       |  5 ++++
 3 files changed, 74 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pnp/resource.c b/drivers/pnp/resource.c
index ff79aa6168cf..786fd356916d 100644
--- a/drivers/pnp/resource.c
+++ b/drivers/pnp/resource.c
@@ -624,6 +624,68 @@ struct pnp_resource *pnp_add_mem_resource(struct pnp_dev *dev,
 	return pnp_res;
 }
 
+static int pnp_possible_option(struct pnp_option *option, int type,
+			       resource_size_t start, resource_size_t size)
+{
+	struct pnp_option *tmp;
+	struct pnp_port *port;
+	struct pnp_mem *mem;
+	struct pnp_irq *irq;
+	struct pnp_dma *dma;
+
+	if (!option)
+		return 0;
+
+	for (tmp = option; tmp; tmp = tmp->next) {
+		switch (type) {
+		case IORESOURCE_IO:
+			for (port = tmp->port; port; port = port->next) {
+				if (port->min == start && port->size == size)
+					return 1;
+			}
+			break;
+		case IORESOURCE_MEM:
+			for (mem = tmp->mem; mem; mem = mem->next) {
+				if (mem->min == start && mem->size == size)
+					return 1;
+			}
+			break;
+		case IORESOURCE_IRQ:
+			for (irq = tmp->irq; irq; irq = irq->next) {
+				if (start < PNP_IRQ_NR &&
+				    test_bit(start, irq->map))
+					return 1;
+			}
+			break;
+		case IORESOURCE_DMA:
+			for (dma = tmp->dma; dma; dma = dma->next) {
+				if (dma->map & (1 << start))
+					return 1;
+			}
+			break;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * Determine whether the specified resource is a possible configuration
+ * for this device.
+ */
+int pnp_possible_config(struct pnp_dev *dev, int type, resource_size_t start,
+			resource_size_t size)
+{
+	if (pnp_possible_option(dev->independent, type, start, size))
+		return 1;
+
+	if (pnp_possible_option(dev->dependent, type, start, size))
+		return 1;
+
+	return 0;
+}
+EXPORT_SYMBOL(pnp_possible_config);
+
 /* format is: pnp_reserve_irq=irq1[,irq2] .... */
 static int __init pnp_setup_reserve_irq(char *str)
 {
diff --git a/drivers/serial/8250_pnp.c b/drivers/serial/8250_pnp.c
index 97c68d021d28..638b68649e79 100644
--- a/drivers/serial/8250_pnp.c
+++ b/drivers/serial/8250_pnp.c
@@ -383,21 +383,14 @@ static int __devinit check_name(char *name)
 	return 0;
 }
 
-static int __devinit check_resources(struct pnp_option *option)
+static int __devinit check_resources(struct pnp_dev *dev)
 {
-	struct pnp_option *tmp;
-	if (!option)
-		return 0;
+	resource_size_t base[] = {0x2f8, 0x3f8, 0x2e8, 0x3e8};
+	int i;
 
-	for (tmp = option; tmp; tmp = tmp->next) {
-		struct pnp_port *port;
-		for (port = tmp->port; port; port = port->next)
-			if ((port->size == 8) &&
-			    ((port->min == 0x2f8) ||
-			     (port->min == 0x3f8) ||
-			     (port->min == 0x2e8) ||
-			     (port->min == 0x3e8)))
-				return 1;
+	for (i = 0; i < ARRAY_SIZE(base); i++) {
+		if (pnp_possible_config(dev, IORESOURCE_IO, base[i], 8))
+			return 1;
 	}
 
 	return 0;
@@ -420,10 +413,7 @@ static int __devinit serial_pnp_guess_board(struct pnp_dev *dev, int *flags)
 		(dev->card && check_name(dev->card->name))))
 			return -ENODEV;
 
-	if (check_resources(dev->independent))
-		return 0;
-
-	if (check_resources(dev->dependent))
+	if (check_resources(dev))
 		return 0;
 
 	return -ENODEV;
diff --git a/include/linux/pnp.h b/include/linux/pnp.h
index dfaa567e04a8..e033e1b14c27 100644
--- a/include/linux/pnp.h
+++ b/include/linux/pnp.h
@@ -479,6 +479,8 @@ void pnp_unregister_card_driver(struct pnp_card_driver *drv);
 extern struct list_head pnp_cards;
 
 /* resource management */
+int pnp_possible_config(struct pnp_dev *dev, int type, resource_size_t base,
+			resource_size_t size);
 int pnp_auto_config_dev(struct pnp_dev *dev);
 int pnp_start_dev(struct pnp_dev *dev);
 int pnp_stop_dev(struct pnp_dev *dev);
@@ -506,6 +508,9 @@ static inline int pnp_register_card_driver(struct pnp_card_driver *drv) { return
 static inline void pnp_unregister_card_driver(struct pnp_card_driver *drv) { }
 
 /* resource management */
+static inline int pnp_possible_config(struct pnp_dev *dev, int type,
+				      resource_size_t base,
+				      resource_size_t size) { return 0; }
 static inline int pnp_auto_config_dev(struct pnp_dev *dev) { return -ENODEV; }
 static inline int pnp_start_dev(struct pnp_dev *dev) { return -ENODEV; }
 static inline int pnp_stop_dev(struct pnp_dev *dev) { return -ENODEV; }
-- 
cgit v1.2.3


From 08c9f262f268f7948be13bf3a5bda1d635c649b4 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 27 Jun 2008 16:57:03 -0600
Subject: PNP: define PNP-specific IORESOURCE_IO_* flags alongside IRQ, DMA,
 MEM

PNP previously defined PNP_PORT_FLAG_16BITADDR and PNP_PORT_FLAG_FIXED
in a private header file, but put those flags in struct resource.flags
fields.  Better to make them IORESOURCE_IO_* flags like the existing
IRQ, DMA, and MEM flags.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/pnp/interface.c        |  2 +-
 drivers/pnp/isapnp/core.c      |  4 ++--
 drivers/pnp/pnpacpi/rsparser.c | 10 +++++-----
 drivers/pnp/pnpbios/rsparser.c |  4 ++--
 include/linux/ioport.h         |  4 ++++
 include/linux/pnp.h            |  3 ---
 6 files changed, 14 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pnp/interface.c b/drivers/pnp/interface.c
index 239923a300cd..c172b6de6b71 100644
--- a/drivers/pnp/interface.c
+++ b/drivers/pnp/interface.c
@@ -57,7 +57,7 @@ static void pnp_print_port(pnp_info_buffer_t * buffer, char *space,
 		   "%sport 0x%x-0x%x, align 0x%x, size 0x%x, %i-bit address decoding\n",
 		   space, port->min, port->max,
 		   port->align ? (port->align - 1) : 0, port->size,
-		   port->flags & PNP_PORT_FLAG_16BITADDR ? 16 : 10);
+		   port->flags & IORESOURCE_IO_16BIT_ADDR ? 16 : 10);
 }
 
 static void pnp_print_irq(pnp_info_buffer_t * buffer, char *space,
diff --git a/drivers/pnp/isapnp/core.c b/drivers/pnp/isapnp/core.c
index ca4457ec403b..c5b92526963b 100644
--- a/drivers/pnp/isapnp/core.c
+++ b/drivers/pnp/isapnp/core.c
@@ -486,7 +486,7 @@ static void __init isapnp_parse_port_resource(struct pnp_dev *dev,
 	port->max = (tmp[4] << 8) | tmp[3];
 	port->align = tmp[5];
 	port->size = tmp[6];
-	port->flags = tmp[0] ? PNP_PORT_FLAG_16BITADDR : 0;
+	port->flags = tmp[0] ? IORESOURCE_IO_16BIT_ADDR : 0;
 	pnp_register_port_resource(dev, option, port);
 }
 
@@ -507,7 +507,7 @@ static void __init isapnp_parse_fixed_port_resource(struct pnp_dev *dev,
 	port->min = port->max = (tmp[1] << 8) | tmp[0];
 	port->size = tmp[2];
 	port->align = 0;
-	port->flags = PNP_PORT_FLAG_FIXED;
+	port->flags = IORESOURCE_IO_FIXED;
 	pnp_register_port_resource(dev, option, port);
 }
 
diff --git a/drivers/pnp/pnpacpi/rsparser.c b/drivers/pnp/pnpacpi/rsparser.c
index 595252b65205..46069e64a6b2 100644
--- a/drivers/pnp/pnpacpi/rsparser.c
+++ b/drivers/pnp/pnpacpi/rsparser.c
@@ -180,7 +180,7 @@ static void pnpacpi_parse_allocated_ioresource(struct pnp_dev *dev, u64 start,
 	u64 end = start + len - 1;
 
 	if (io_decode == ACPI_DECODE_16)
-		flags |= PNP_PORT_FLAG_16BITADDR;
+		flags |= IORESOURCE_IO_16BIT_ADDR;
 	if (len == 0 || end >= 0x10003)
 		flags |= IORESOURCE_DISABLED;
 
@@ -485,7 +485,7 @@ static __init void pnpacpi_parse_port_option(struct pnp_dev *dev,
 	port->align = io->alignment;
 	port->size = io->address_length;
 	port->flags = ACPI_DECODE_16 == io->io_decode ?
-	    PNP_PORT_FLAG_16BITADDR : 0;
+	    IORESOURCE_IO_16BIT_ADDR : 0;
 	pnp_register_port_resource(dev, option, port);
 }
 
@@ -503,7 +503,7 @@ static __init void pnpacpi_parse_fixed_port_option(struct pnp_dev *dev,
 	port->min = port->max = io->address;
 	port->size = io->address_length;
 	port->align = 0;
-	port->flags = PNP_PORT_FLAG_FIXED;
+	port->flags = IORESOURCE_IO_FIXED;
 	pnp_register_port_resource(dev, option, port);
 }
 
@@ -609,7 +609,7 @@ static __init void pnpacpi_parse_address_option(struct pnp_dev *dev,
 		port->min = port->max = p->minimum;
 		port->size = p->address_length;
 		port->align = 0;
-		port->flags = PNP_PORT_FLAG_FIXED;
+		port->flags = IORESOURCE_IO_FIXED;
 		pnp_register_port_resource(dev, option, port);
 	}
 }
@@ -946,7 +946,7 @@ static void pnpacpi_encode_io(struct pnp_dev *dev,
 
 	if (pnp_resource_enabled(p)) {
 		/* Note: pnp_assign_port copies pnp_port->flags into p->flags */
-		io->io_decode = (p->flags & PNP_PORT_FLAG_16BITADDR) ?
+		io->io_decode = (p->flags & IORESOURCE_IO_16BIT_ADDR) ?
 		    ACPI_DECODE_16 : ACPI_DECODE_10;
 		io->minimum = p->start;
 		io->maximum = p->end;
diff --git a/drivers/pnp/pnpbios/rsparser.c b/drivers/pnp/pnpbios/rsparser.c
index 01f0c3dd1b08..489fec3b7974 100644
--- a/drivers/pnp/pnpbios/rsparser.c
+++ b/drivers/pnp/pnpbios/rsparser.c
@@ -310,7 +310,7 @@ static __init void pnpbios_parse_port_option(struct pnp_dev *dev,
 	port->max = (p[5] << 8) | p[4];
 	port->align = p[6];
 	port->size = p[7];
-	port->flags = p[1] ? PNP_PORT_FLAG_16BITADDR : 0;
+	port->flags = p[1] ? IORESOURCE_IO_16BIT_ADDR : 0;
 	pnp_register_port_resource(dev, option, port);
 }
 
@@ -326,7 +326,7 @@ static __init void pnpbios_parse_fixed_port_option(struct pnp_dev *dev,
 	port->min = port->max = (p[2] << 8) | p[1];
 	port->size = p[3];
 	port->align = 0;
-	port->flags = PNP_PORT_FLAG_FIXED;
+	port->flags = IORESOURCE_IO_FIXED;
 	pnp_register_port_resource(dev, option, port);
 }
 
diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index c6801bffe76d..39db059ffb8b 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -88,6 +88,10 @@ struct resource_list {
 #define IORESOURCE_MEM_SHADOWABLE	(1<<5)	/* dup: IORESOURCE_SHADOWABLE */
 #define IORESOURCE_MEM_EXPANSIONROM	(1<<6)
 
+/* PnP I/O specific bits (IORESOURCE_BITS) */
+#define IORESOURCE_IO_16BIT_ADDR	(1<<0)
+#define IORESOURCE_IO_FIXED		(1<<1)
+
 /* PCI ROM control bits (IORESOURCE_BITS) */
 #define IORESOURCE_ROM_ENABLE		(1<<0)	/* ROM is enabled, same as PCI_ROM_ADDRESS_ENABLE */
 #define IORESOURCE_ROM_SHADOW		(1<<1)	/* ROM is copy at C000:0 */
diff --git a/include/linux/pnp.h b/include/linux/pnp.h
index e033e1b14c27..e1454dabde10 100644
--- a/include/linux/pnp.h
+++ b/include/linux/pnp.h
@@ -182,9 +182,6 @@ static inline int pnp_dma_valid(struct pnp_dev *dev, unsigned int bar)
 }
 
 
-#define PNP_PORT_FLAG_16BITADDR	(1<<0)
-#define PNP_PORT_FLAG_FIXED	(1<<1)
-
 struct pnp_port {
 	unsigned short min;	/* min base number */
 	unsigned short max;	/* max base number */
-- 
cgit v1.2.3


From a1802c42950403657d07e64558eff612d550ce16 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 27 Jun 2008 16:57:04 -0600
Subject: PNP: make resource option structures private to PNP subsystem

Nothing outside the PNP subsystem should need access to a
device's resource options, so this patch moves the option
structure declarations to a private header file.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/pnp/base.h  | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pnp.h | 48 ------------------------------------------------
 2 files changed, 48 insertions(+), 48 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pnp/base.h b/drivers/pnp/base.h
index 1667ac3ca45b..3126e4582008 100644
--- a/drivers/pnp/base.h
+++ b/drivers/pnp/base.h
@@ -19,6 +19,54 @@ void pnp_remove_card(struct pnp_card *card);
 int pnp_add_card_device(struct pnp_card *card, struct pnp_dev *dev);
 void pnp_remove_card_device(struct pnp_dev *dev);
 
+struct pnp_port {
+	unsigned short min;	/* min base number */
+	unsigned short max;	/* max base number */
+	unsigned char align;	/* align boundary */
+	unsigned char size;	/* size of range */
+	unsigned char flags;	/* port flags */
+	unsigned char pad;	/* pad */
+	struct pnp_port *next;	/* next port */
+};
+
+#define PNP_IRQ_NR 256
+struct pnp_irq {
+	DECLARE_BITMAP(map, PNP_IRQ_NR);	/* bitmask for IRQ lines */
+	unsigned char flags;	/* IRQ flags */
+	unsigned char pad;	/* pad */
+	struct pnp_irq *next;	/* next IRQ */
+};
+
+struct pnp_dma {
+	unsigned char map;	/* bitmask for DMA channels */
+	unsigned char flags;	/* DMA flags */
+	struct pnp_dma *next;	/* next port */
+};
+
+struct pnp_mem {
+	unsigned int min;	/* min base number */
+	unsigned int max;	/* max base number */
+	unsigned int align;	/* align boundary */
+	unsigned int size;	/* size of range */
+	unsigned char flags;	/* memory flags */
+	unsigned char pad;	/* pad */
+	struct pnp_mem *next;	/* next memory resource */
+};
+
+#define PNP_RES_PRIORITY_PREFERRED	0
+#define PNP_RES_PRIORITY_ACCEPTABLE	1
+#define PNP_RES_PRIORITY_FUNCTIONAL	2
+#define PNP_RES_PRIORITY_INVALID	65535
+
+struct pnp_option {
+	unsigned short priority;	/* priority */
+	struct pnp_port *port;		/* first port */
+	struct pnp_irq *irq;		/* first IRQ */
+	struct pnp_dma *dma;		/* first DMA */
+	struct pnp_mem *mem;		/* first memory resource */
+	struct pnp_option *next;	/* used to chain dependent resources */
+};
+
 struct pnp_option *pnp_build_option(int priority);
 struct pnp_option *pnp_register_independent_option(struct pnp_dev *dev);
 struct pnp_option *pnp_register_dependent_option(struct pnp_dev *dev,
diff --git a/include/linux/pnp.h b/include/linux/pnp.h
index e1454dabde10..785126ffcc11 100644
--- a/include/linux/pnp.h
+++ b/include/linux/pnp.h
@@ -182,54 +182,6 @@ static inline int pnp_dma_valid(struct pnp_dev *dev, unsigned int bar)
 }
 
 
-struct pnp_port {
-	unsigned short min;	/* min base number */
-	unsigned short max;	/* max base number */
-	unsigned char align;	/* align boundary */
-	unsigned char size;	/* size of range */
-	unsigned char flags;	/* port flags */
-	unsigned char pad;	/* pad */
-	struct pnp_port *next;	/* next port */
-};
-
-#define PNP_IRQ_NR 256
-struct pnp_irq {
-	DECLARE_BITMAP(map, PNP_IRQ_NR);	/* bitmask for IRQ lines */
-	unsigned char flags;	/* IRQ flags */
-	unsigned char pad;	/* pad */
-	struct pnp_irq *next;	/* next IRQ */
-};
-
-struct pnp_dma {
-	unsigned char map;	/* bitmask for DMA channels */
-	unsigned char flags;	/* DMA flags */
-	struct pnp_dma *next;	/* next port */
-};
-
-struct pnp_mem {
-	unsigned int min;	/* min base number */
-	unsigned int max;	/* max base number */
-	unsigned int align;	/* align boundary */
-	unsigned int size;	/* size of range */
-	unsigned char flags;	/* memory flags */
-	unsigned char pad;	/* pad */
-	struct pnp_mem *next;	/* next memory resource */
-};
-
-#define PNP_RES_PRIORITY_PREFERRED	0
-#define PNP_RES_PRIORITY_ACCEPTABLE	1
-#define PNP_RES_PRIORITY_FUNCTIONAL	2
-#define PNP_RES_PRIORITY_INVALID	65535
-
-struct pnp_option {
-	unsigned short priority;	/* priority */
-	struct pnp_port *port;		/* first port */
-	struct pnp_irq *irq;		/* first IRQ */
-	struct pnp_dma *dma;		/* first DMA */
-	struct pnp_mem *mem;		/* first memory resource */
-	struct pnp_option *next;	/* used to chain dependent resources */
-};
-
 /*
  * Device Management
  */
-- 
cgit v1.2.3


From d5ebde6ef5c2d51828f975a81d7d0e58bccfd833 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 27 Jun 2008 16:57:14 -0600
Subject: PNP: support optional IRQ resources

This patch adds an IORESOURCE_IRQ_OPTIONAL flag for use when
assigning resources to a device.  If the flag is set and we are
unable to assign an IRQ to the device, we can leave the IRQ
disabled but allow the overall resource allocation to succeed.

Some devices request an IRQ, but can run without an IRQ
(possibly with degraded performance).  This flag lets us run
the device without the IRQ instead of just leaving the
device disabled.

This is a reimplementation of this previous change by Rene
Herman <rene.herman@gmail.com>:
    http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=3b73a223661ed137c5d3d2635f954382e94f5a43

I reimplemented this for two reasons:
    - to prepare for converting all resource options into a single linked
      list, as opposed to the per-resource-type lists we have now, and
    - to preserve the order and number of resource options.

In PNPBIOS and ACPI, we configure a device by giving firmware a
list of resource assignments.  It is important that this list
has exactly the same number of resources, in the same order,
as the "template" list we got from the firmware in the first
place.

The problem of a sound card MPU401 being left disabled for want of
an IRQ was reported by Uwe Bugla <uwe.bugla@gmx.de>.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/pnp/interface.c |  2 ++
 drivers/pnp/manager.c   |  9 +++++++
 drivers/pnp/quirks.c    | 71 +++++++++++++++++++------------------------------
 include/linux/ioport.h  |  1 +
 4 files changed, 39 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pnp/interface.c b/drivers/pnp/interface.c
index b09f67de13d0..7a9fb5544b80 100644
--- a/drivers/pnp/interface.c
+++ b/drivers/pnp/interface.c
@@ -90,6 +90,8 @@ static void pnp_print_irq(pnp_info_buffer_t * buffer, char *space,
 		pnp_printf(buffer, " High-Level");
 	if (irq->flags & IORESOURCE_IRQ_LOWLEVEL)
 		pnp_printf(buffer, " Low-Level");
+	if (irq->flags & IORESOURCE_IRQ_OPTIONAL)
+		pnp_printf(buffer, " (optional)");
 	pnp_printf(buffer, "\n");
 }
 
diff --git a/drivers/pnp/manager.c b/drivers/pnp/manager.c
index 7ea9e1e28003..a20accb5ef8f 100644
--- a/drivers/pnp/manager.c
+++ b/drivers/pnp/manager.c
@@ -153,6 +153,15 @@ static int pnp_assign_irq(struct pnp_dev *dev, struct pnp_irq *rule, int idx)
 				goto __add;
 		}
 	}
+
+	if (rule->flags & IORESOURCE_IRQ_OPTIONAL) {
+		res->start = -1;
+		res->end = -1;
+		res->flags |= IORESOURCE_DISABLED;
+		dev_dbg(&dev->dev, "  irq %d disabled (optional)\n", idx);
+		goto __add;
+	}
+
 	dev_dbg(&dev->dev, "  couldn't assign irq %d\n", idx);
 	return -EBUSY;
 
diff --git a/drivers/pnp/quirks.c b/drivers/pnp/quirks.c
index 48e60171b3ba..e8515ce0d296 100644
--- a/drivers/pnp/quirks.c
+++ b/drivers/pnp/quirks.c
@@ -121,34 +121,46 @@ static struct pnp_option *quirk_isapnp_mpu_options(struct pnp_dev *dev)
 	struct pnp_option *res;
 
 	/*
-	 * Build a functional IRQ-less variant of each MPU option.
+	 * Build a functional IRQ-optional variant of each MPU option.
 	 */
 
 	for (res = dev->dependent; res; res = res->next) {
 		struct pnp_option *curr;
 		struct pnp_port *port;
-		struct pnp_port *copy;
+		struct pnp_port *copy_port;
+		struct pnp_irq *irq;
+		struct pnp_irq *copy_irq;
 
 		port = res->port;
-		if (!port || !res->irq)
+		irq = res->irq;
+		if (!port || !irq)
 			continue;
 
-		copy = pnp_alloc(sizeof *copy);
-		if (!copy)
+		copy_port = pnp_alloc(sizeof *copy_port);
+		if (!copy_port)
+			break;
+
+		copy_irq = pnp_alloc(sizeof *copy_irq);
+		if (!copy_irq) {
+			kfree(copy_port);
 			break;
+		}
+
+		*copy_port = *port;
+		copy_port->next = NULL;
 
-		copy->min = port->min;
-		copy->max = port->max;
-		copy->align = port->align;
-		copy->size = port->size;
-		copy->flags = port->flags;
+		*copy_irq = *irq;
+		copy_irq->flags |= IORESOURCE_IRQ_OPTIONAL;
+		copy_irq->next = NULL;
 
 		curr = pnp_build_option(PNP_RES_PRIORITY_FUNCTIONAL);
 		if (!curr) {
-			kfree(copy);
+			kfree(copy_port);
+			kfree(copy_irq);
 			break;
 		}
-		curr->port = copy;
+		curr->port = copy_port;
+		curr->irq = copy_irq;
 
 		if (prev)
 			prev->next = curr;
@@ -157,7 +169,7 @@ static struct pnp_option *quirk_isapnp_mpu_options(struct pnp_dev *dev)
 		prev = curr;
 	}
 	if (head)
-		dev_info(&dev->dev, "adding IRQ-less MPU options\n");
+		dev_info(&dev->dev, "adding IRQ-optional MPU options\n");
 
 	return head;
 }
@@ -167,10 +179,6 @@ static void quirk_ad1815_mpu_resources(struct pnp_dev *dev)
 	struct pnp_option *res;
 	struct pnp_irq *irq;
 
-	/*
-	 * Distribute the independent IRQ over the dependent options
-	 */
-
 	res = dev->independent;
 	if (!res)
 		return;
@@ -179,33 +187,8 @@ static void quirk_ad1815_mpu_resources(struct pnp_dev *dev)
 	if (!irq || irq->next)
 		return;
 
-	res = dev->dependent;
-	if (!res)
-		return;
-
-	while (1) {
-		struct pnp_irq *copy;
-
-		copy = pnp_alloc(sizeof *copy);
-		if (!copy)
-			break;
-
-		bitmap_copy(copy->map.bits, irq->map.bits, PNP_IRQ_NR);
-		copy->flags = irq->flags;
-
-		copy->next = res->irq; /* Yes, this is NULL */
-		res->irq = copy;
-
-		if (!res->next)
-			break;
-		res = res->next;
-	}
-	kfree(irq);
-
-	res->next = quirk_isapnp_mpu_options(dev);
-
-	res = dev->independent;
-	res->irq = NULL;
+	irq->flags |= IORESOURCE_IRQ_OPTIONAL;
+	dev_info(&dev->dev, "made independent IRQ optional\n");
 }
 
 static void quirk_isapnp_mpu_resources(struct pnp_dev *dev)
diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index 39db059ffb8b..2cd07cc29687 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -59,6 +59,7 @@ struct resource_list {
 #define IORESOURCE_IRQ_HIGHLEVEL	(1<<2)
 #define IORESOURCE_IRQ_LOWLEVEL		(1<<3)
 #define IORESOURCE_IRQ_SHAREABLE	(1<<4)
+#define IORESOURCE_IRQ_OPTIONAL 	(1<<5)
 
 /* PnP DMA specific bits (IORESOURCE_BITS) */
 #define IORESOURCE_DMA_TYPE_MASK	(3<<0)
-- 
cgit v1.2.3


From 1f32ca31e7409d37c1b25e5f81840fb184380cdf Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 27 Jun 2008 16:57:17 -0600
Subject: PNP: convert resource options to single linked list

ISAPNP, PNPBIOS, and ACPI describe the "possible resource settings" of
a device, i.e., the possibilities an OS bus driver has when it assigns
I/O port, MMIO, and other resources to the device.

PNP used to maintain this "possible resource setting" information in
one independent option structure and a list of dependent option
structures for each device.  Each of these option structures had lists
of I/O, memory, IRQ, and DMA resources, for example:

  dev
    independent options
      ind-io0  -> ind-io1  ...
      ind-mem0 -> ind-mem1 ...
      ...
    dependent option set 0
      dep0-io0  -> dep0-io1  ...
      dep0-mem0 -> dep0-mem1 ...
      ...
    dependent option set 1
      dep1-io0  -> dep1-io1  ...
      dep1-mem0 -> dep1-mem1 ...
      ...
    ...

This data structure was designed for ISAPNP, where the OS configures
device resource settings by writing directly to configuration
registers.  The OS can write the registers in arbitrary order much
like it writes PCI BARs.

However, for PNPBIOS and ACPI devices, the OS uses firmware interfaces
that perform device configuration, and it is important to pass the
desired settings to those interfaces in the correct order.  The OS
learns the correct order by using firmware interfaces that return the
"current resource settings" and "possible resource settings," but the
option structures above doesn't store the ordering information.

This patch replaces the independent and dependent lists with a single
list of options.  For example, a device might have possible resource
settings like this:

  dev
    options
      ind-io0 -> dep0-io0 -> dep1->io0 -> ind-io1 ...

All the possible settings are in the same list, in the order they
come from the firmware "possible resource settings" list.  Each entry
is tagged with an independent/dependent flag.  Dependent entries also
have a "set number" and an optional priority value.  All dependent
entries must be assigned from the same set.  For example, the OS can
use all the entries from dependent set 0, or all the entries from
dependent set 1, but it cannot mix entries from set 0 with entries
from set 1.

Prior to this patch PNP didn't keep track of the order of this list,
and it assigned all independent options first, then all dependent
ones.  Using the example above, that resulted in a "desired
configuration" list like this:

  ind->io0 -> ind->io1 -> depN-io0 ...

instead of the list the firmware expects, which looks like this:

  ind->io0 -> depN-io0 -> ind-io1 ...

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/pnp/base.h             |  93 +++++++++----
 drivers/pnp/core.c             |   4 +-
 drivers/pnp/interface.c        |  75 +++++------
 drivers/pnp/isapnp/core.c      |  72 +++++-----
 drivers/pnp/manager.c          | 145 +++++++--------------
 drivers/pnp/pnpacpi/rsparser.c |  93 ++++++-------
 drivers/pnp/pnpbios/rsparser.c |  67 +++++-----
 drivers/pnp/quirks.c           | 290 ++++++++++++++++++++++-------------------
 drivers/pnp/resource.c         | 268 +++++++++----------------------------
 drivers/pnp/support.c          |  92 +++++++++++++
 include/linux/pnp.h            |   6 +-
 11 files changed, 571 insertions(+), 634 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pnp/base.h b/drivers/pnp/base.h
index 360c6385686c..e3fa9a2d9a3d 100644
--- a/drivers/pnp/base.h
+++ b/drivers/pnp/base.h
@@ -1,3 +1,8 @@
+/*
+ * Copyright (C) 2008 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
+ */
+
 extern spinlock_t pnp_lock;
 void *pnp_alloc(long size);
 
@@ -25,8 +30,6 @@ struct pnp_port {
 	resource_size_t align;	/* align boundary */
 	resource_size_t size;	/* size of range */
 	unsigned char flags;	/* port flags */
-	unsigned char pad;	/* pad */
-	struct pnp_port *next;	/* next port */
 };
 
 #define PNP_IRQ_NR 256
@@ -35,14 +38,11 @@ typedef struct { DECLARE_BITMAP(bits, PNP_IRQ_NR); } pnp_irq_mask_t;
 struct pnp_irq {
 	pnp_irq_mask_t map;	/* bitmap for IRQ lines */
 	unsigned char flags;	/* IRQ flags */
-	unsigned char pad;	/* pad */
-	struct pnp_irq *next;	/* next IRQ */
 };
 
 struct pnp_dma {
 	unsigned char map;	/* bitmask for DMA channels */
 	unsigned char flags;	/* DMA flags */
-	struct pnp_dma *next;	/* next port */
 };
 
 struct pnp_mem {
@@ -51,44 +51,91 @@ struct pnp_mem {
 	resource_size_t align;	/* align boundary */
 	resource_size_t size;	/* size of range */
 	unsigned char flags;	/* memory flags */
-	unsigned char pad;	/* pad */
-	struct pnp_mem *next;	/* next memory resource */
 };
 
+#define PNP_OPTION_DEPENDENT		0x80000000
+#define PNP_OPTION_SET_MASK		0xffff
+#define PNP_OPTION_SET_SHIFT		12
+#define PNP_OPTION_PRIORITY_MASK	0xfff
+#define PNP_OPTION_PRIORITY_SHIFT	0
+
 #define PNP_RES_PRIORITY_PREFERRED	0
 #define PNP_RES_PRIORITY_ACCEPTABLE	1
 #define PNP_RES_PRIORITY_FUNCTIONAL	2
-#define PNP_RES_PRIORITY_INVALID	65535
+#define PNP_RES_PRIORITY_INVALID	PNP_OPTION_PRIORITY_MASK
 
 struct pnp_option {
-	unsigned short priority;	/* priority */
-	struct pnp_port *port;		/* first port */
-	struct pnp_irq *irq;		/* first IRQ */
-	struct pnp_dma *dma;		/* first DMA */
-	struct pnp_mem *mem;		/* first memory resource */
-	struct pnp_option *next;	/* used to chain dependent resources */
+	struct list_head list;
+	unsigned int flags;	/* independent/dependent, set, priority */
+
+	unsigned long type;	/* IORESOURCE_{IO,MEM,IRQ,DMA} */
+	union {
+		struct pnp_port port;
+		struct pnp_irq irq;
+		struct pnp_dma dma;
+		struct pnp_mem mem;
+	} u;
 };
 
-struct pnp_option *pnp_build_option(int priority);
-struct pnp_option *pnp_register_independent_option(struct pnp_dev *dev);
-struct pnp_option *pnp_register_dependent_option(struct pnp_dev *dev,
-						 int priority);
-int pnp_register_irq_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_irq_resource(struct pnp_dev *dev, unsigned int option_flags,
 			      pnp_irq_mask_t *map, unsigned char flags);
-int pnp_register_dma_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_dma_resource(struct pnp_dev *dev, unsigned int option_flags,
 			      unsigned char map, unsigned char flags);
-int pnp_register_port_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_port_resource(struct pnp_dev *dev, unsigned int option_flags,
 			       resource_size_t min, resource_size_t max,
 			       resource_size_t align, resource_size_t size,
 			       unsigned char flags);
-int pnp_register_mem_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_mem_resource(struct pnp_dev *dev, unsigned int option_flags,
 			      resource_size_t min, resource_size_t max,
 			      resource_size_t align, resource_size_t size,
 			      unsigned char flags);
+
+static inline int pnp_option_is_dependent(struct pnp_option *option)
+{
+	return option->flags & PNP_OPTION_DEPENDENT ? 1 : 0;
+}
+
+static inline unsigned int pnp_option_set(struct pnp_option *option)
+{
+	return (option->flags >> PNP_OPTION_SET_SHIFT) & PNP_OPTION_SET_MASK;
+}
+
+static inline unsigned int pnp_option_priority(struct pnp_option *option)
+{
+	return (option->flags >> PNP_OPTION_PRIORITY_SHIFT) &
+	    PNP_OPTION_PRIORITY_MASK;
+}
+
+static inline unsigned int pnp_new_dependent_set(struct pnp_dev *dev,
+						 int priority)
+{
+	unsigned int flags;
+
+	if (priority > PNP_RES_PRIORITY_FUNCTIONAL) {
+		dev_warn(&dev->dev, "invalid dependent option priority %d "
+			 "clipped to %d", priority,
+			 PNP_RES_PRIORITY_INVALID);
+		priority = PNP_RES_PRIORITY_INVALID;
+	}
+
+	flags = PNP_OPTION_DEPENDENT |
+	    ((dev->num_dependent_sets & PNP_OPTION_SET_MASK) <<
+		PNP_OPTION_SET_SHIFT) |
+	    ((priority & PNP_OPTION_PRIORITY_MASK) <<
+		PNP_OPTION_PRIORITY_SHIFT);
+
+	dev->num_dependent_sets++;
+
+	return flags;
+}
+
+char *pnp_option_priority_name(struct pnp_option *option);
+void dbg_pnp_show_option(struct pnp_dev *dev, struct pnp_option *option);
+
 void pnp_init_resources(struct pnp_dev *dev);
 
 void pnp_fixup_device(struct pnp_dev *dev);
-void pnp_free_option(struct pnp_option *option);
+void pnp_free_options(struct pnp_dev *dev);
 int __pnp_add_device(struct pnp_dev *dev);
 void __pnp_remove_device(struct pnp_dev *dev);
 
diff --git a/drivers/pnp/core.c b/drivers/pnp/core.c
index 7182da92aec3..a411582bcd72 100644
--- a/drivers/pnp/core.c
+++ b/drivers/pnp/core.c
@@ -118,10 +118,9 @@ static void pnp_release_device(struct device *dmdev)
 {
 	struct pnp_dev *dev = to_pnp_dev(dmdev);
 
-	pnp_free_option(dev->independent);
-	pnp_free_option(dev->dependent);
 	pnp_free_ids(dev);
 	pnp_free_resources(dev);
+	pnp_free_options(dev);
 	kfree(dev);
 }
 
@@ -135,6 +134,7 @@ struct pnp_dev *pnp_alloc_dev(struct pnp_protocol *protocol, int id, char *pnpid
 		return NULL;
 
 	INIT_LIST_HEAD(&dev->resources);
+	INIT_LIST_HEAD(&dev->options);
 	dev->protocol = protocol;
 	dev->number = id;
 	dev->dma_mask = DMA_24BIT_MASK;
diff --git a/drivers/pnp/interface.c b/drivers/pnp/interface.c
index 7a9fb5544b80..a876ecf7028c 100644
--- a/drivers/pnp/interface.c
+++ b/drivers/pnp/interface.c
@@ -3,6 +3,8 @@
  *
  * Some code, especially possible resource dumping is based on isapnp_proc.c (c) Jaroslav Kysela <perex@perex.cz>
  * Copyright 2002 Adam Belay <ambx1@neo.rr.com>
+ * Copyright (C) 2008 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
  */
 
 #include <linux/pnp.h>
@@ -184,39 +186,22 @@ static void pnp_print_mem(pnp_info_buffer_t * buffer, char *space,
 }
 
 static void pnp_print_option(pnp_info_buffer_t * buffer, char *space,
-			     struct pnp_option *option, int dep)
+			     struct pnp_option *option)
 {
-	char *s;
-	struct pnp_port *port;
-	struct pnp_irq *irq;
-	struct pnp_dma *dma;
-	struct pnp_mem *mem;
-
-	if (dep) {
-		switch (option->priority) {
-		case PNP_RES_PRIORITY_PREFERRED:
-			s = "preferred";
-			break;
-		case PNP_RES_PRIORITY_ACCEPTABLE:
-			s = "acceptable";
-			break;
-		case PNP_RES_PRIORITY_FUNCTIONAL:
-			s = "functional";
-			break;
-		default:
-			s = "invalid";
-		}
-		pnp_printf(buffer, "Dependent: %02i - Priority %s\n", dep, s);
+	switch (option->type) {
+	case IORESOURCE_IO:
+		pnp_print_port(buffer, space, &option->u.port);
+		break;
+	case IORESOURCE_MEM:
+		pnp_print_mem(buffer, space, &option->u.mem);
+		break;
+	case IORESOURCE_IRQ:
+		pnp_print_irq(buffer, space, &option->u.irq);
+		break;
+	case IORESOURCE_DMA:
+		pnp_print_dma(buffer, space, &option->u.dma);
+		break;
 	}
-
-	for (port = option->port; port; port = port->next)
-		pnp_print_port(buffer, space, port);
-	for (irq = option->irq; irq; irq = irq->next)
-		pnp_print_irq(buffer, space, irq);
-	for (dma = option->dma; dma; dma = dma->next)
-		pnp_print_dma(buffer, space, dma);
-	for (mem = option->mem; mem; mem = mem->next)
-		pnp_print_mem(buffer, space, mem);
 }
 
 static ssize_t pnp_show_options(struct device *dmdev,
@@ -224,9 +209,9 @@ static ssize_t pnp_show_options(struct device *dmdev,
 {
 	struct pnp_dev *dev = to_pnp_dev(dmdev);
 	pnp_info_buffer_t *buffer;
-	struct pnp_option *independent = dev->independent;
-	struct pnp_option *dependent = dev->dependent;
-	int ret, dep = 1;
+	struct pnp_option *option;
+	int ret, dep = 0, set = 0;
+	char *indent;
 
 	buffer = pnp_alloc(sizeof(pnp_info_buffer_t));
 	if (!buffer)
@@ -235,14 +220,24 @@ static ssize_t pnp_show_options(struct device *dmdev,
 	buffer->len = PAGE_SIZE;
 	buffer->buffer = buf;
 	buffer->curr = buffer->buffer;
-	if (independent)
-		pnp_print_option(buffer, "", independent, 0);
 
-	while (dependent) {
-		pnp_print_option(buffer, "   ", dependent, dep);
-		dependent = dependent->next;
-		dep++;
+	list_for_each_entry(option, &dev->options, list) {
+		if (pnp_option_is_dependent(option)) {
+			indent = "  ";
+			if (!dep || pnp_option_set(option) != set) {
+				set = pnp_option_set(option);
+				dep = 1;
+				pnp_printf(buffer, "Dependent: %02i - "
+					   "Priority %s\n", set,
+					   pnp_option_priority_name(option));
+			}
+		} else {
+			dep = 0;
+			indent = "";
+		}
+		pnp_print_option(buffer, indent, option);
 	}
+
 	ret = (buffer->curr - buf);
 	kfree(buffer);
 	return ret;
diff --git a/drivers/pnp/isapnp/core.c b/drivers/pnp/isapnp/core.c
index 53cc4d6133e6..101a835e8759 100644
--- a/drivers/pnp/isapnp/core.c
+++ b/drivers/pnp/isapnp/core.c
@@ -429,7 +429,7 @@ static struct pnp_dev *__init isapnp_parse_device(struct pnp_card *card,
  *  Add IRQ resource to resources list.
  */
 static void __init isapnp_parse_irq_resource(struct pnp_dev *dev,
-					     struct pnp_option *option,
+					     unsigned int option_flags,
 					     int size)
 {
 	unsigned char tmp[3];
@@ -446,27 +446,27 @@ static void __init isapnp_parse_irq_resource(struct pnp_dev *dev,
 	if (size > 2)
 		flags = tmp[2];
 
-	pnp_register_irq_resource(dev, option, &map, flags);
+	pnp_register_irq_resource(dev, option_flags, &map, flags);
 }
 
 /*
  *  Add DMA resource to resources list.
  */
 static void __init isapnp_parse_dma_resource(struct pnp_dev *dev,
-					     struct pnp_option *option,
+					     unsigned int option_flags,
 					     int size)
 {
 	unsigned char tmp[2];
 
 	isapnp_peek(tmp, size);
-	pnp_register_dma_resource(dev, option, tmp[0], tmp[1]);
+	pnp_register_dma_resource(dev, option_flags, tmp[0], tmp[1]);
 }
 
 /*
  *  Add port resource to resources list.
  */
 static void __init isapnp_parse_port_resource(struct pnp_dev *dev,
-					      struct pnp_option *option,
+					      unsigned int option_flags,
 					      int size)
 {
 	unsigned char tmp[7];
@@ -479,14 +479,15 @@ static void __init isapnp_parse_port_resource(struct pnp_dev *dev,
 	align = tmp[5];
 	len = tmp[6];
 	flags = tmp[0] ? IORESOURCE_IO_16BIT_ADDR : 0;
-	pnp_register_port_resource(dev, option, min, max, align, len, flags);
+	pnp_register_port_resource(dev, option_flags,
+				   min, max, align, len, flags);
 }
 
 /*
  *  Add fixed port resource to resources list.
  */
 static void __init isapnp_parse_fixed_port_resource(struct pnp_dev *dev,
-						    struct pnp_option *option,
+						    unsigned int option_flags,
 						    int size)
 {
 	unsigned char tmp[3];
@@ -495,7 +496,7 @@ static void __init isapnp_parse_fixed_port_resource(struct pnp_dev *dev,
 	isapnp_peek(tmp, size);
 	base = (tmp[1] << 8) | tmp[0];
 	len = tmp[2];
-	pnp_register_port_resource(dev, option, base, base, 0, len,
+	pnp_register_port_resource(dev, option_flags, base, base, 0, len,
 				   IORESOURCE_IO_FIXED);
 }
 
@@ -503,7 +504,7 @@ static void __init isapnp_parse_fixed_port_resource(struct pnp_dev *dev,
  *  Add memory resource to resources list.
  */
 static void __init isapnp_parse_mem_resource(struct pnp_dev *dev,
-					     struct pnp_option *option,
+					     unsigned int option_flags,
 					     int size)
 {
 	unsigned char tmp[9];
@@ -516,14 +517,15 @@ static void __init isapnp_parse_mem_resource(struct pnp_dev *dev,
 	align = (tmp[6] << 8) | tmp[5];
 	len = ((tmp[8] << 8) | tmp[7]) << 8;
 	flags = tmp[0];
-	pnp_register_mem_resource(dev, option, min, max, align, len, flags);
+	pnp_register_mem_resource(dev, option_flags,
+				  min, max, align, len, flags);
 }
 
 /*
  *  Add 32-bit memory resource to resources list.
  */
 static void __init isapnp_parse_mem32_resource(struct pnp_dev *dev,
-					       struct pnp_option *option,
+					       unsigned int option_flags,
 					       int size)
 {
 	unsigned char tmp[17];
@@ -536,14 +538,15 @@ static void __init isapnp_parse_mem32_resource(struct pnp_dev *dev,
 	align = (tmp[12] << 24) | (tmp[11] << 16) | (tmp[10] << 8) | tmp[9];
 	len = (tmp[16] << 24) | (tmp[15] << 16) | (tmp[14] << 8) | tmp[13];
 	flags = tmp[0];
-	pnp_register_mem_resource(dev, option, min, max, align, len, flags);
+	pnp_register_mem_resource(dev, option_flags,
+				  min, max, align, len, flags);
 }
 
 /*
  *  Add 32-bit fixed memory resource to resources list.
  */
 static void __init isapnp_parse_fixed_mem32_resource(struct pnp_dev *dev,
-						     struct pnp_option *option,
+						     unsigned int option_flags,
 						     int size)
 {
 	unsigned char tmp[9];
@@ -554,7 +557,7 @@ static void __init isapnp_parse_fixed_mem32_resource(struct pnp_dev *dev,
 	base = (tmp[4] << 24) | (tmp[3] << 16) | (tmp[2] << 8) | tmp[1];
 	len = (tmp[8] << 24) | (tmp[7] << 16) | (tmp[6] << 8) | tmp[5];
 	flags = tmp[0];
-	pnp_register_mem_resource(dev, option, base, base, 0, len, flags);
+	pnp_register_mem_resource(dev, option_flags, base, base, 0, len, flags);
 }
 
 /*
@@ -584,18 +587,14 @@ static int __init isapnp_create_device(struct pnp_card *card,
 {
 	int number = 0, skip = 0, priority, compat = 0;
 	unsigned char type, tmp[17];
-	struct pnp_option *option, *option_independent;
+	unsigned int option_flags;
 	struct pnp_dev *dev;
 	u32 eisa_id;
 	char id[8];
 
 	if ((dev = isapnp_parse_device(card, size, number++)) == NULL)
 		return 1;
-	option_independent = option = pnp_register_independent_option(dev);
-	if (!option) {
-		kfree(dev);
-		return 1;
-	}
+	option_flags = 0;
 	pnp_add_card_device(card, dev);
 
 	while (1) {
@@ -612,12 +611,7 @@ static int __init isapnp_create_device(struct pnp_card *card,
 					return 1;
 				size = 0;
 				skip = 0;
-				option = pnp_register_independent_option(dev);
-				option_independent = option;
-				if (!option) {
-					kfree(dev);
-					return 1;
-				}
+				option_flags = 0;
 				pnp_add_card_device(card, dev);
 			} else {
 				skip = 1;
@@ -638,13 +632,13 @@ static int __init isapnp_create_device(struct pnp_card *card,
 		case _STAG_IRQ:
 			if (size < 2 || size > 3)
 				goto __skip;
-			isapnp_parse_irq_resource(dev, option, size);
+			isapnp_parse_irq_resource(dev, option_flags, size);
 			size = 0;
 			break;
 		case _STAG_DMA:
 			if (size != 2)
 				goto __skip;
-			isapnp_parse_dma_resource(dev, option, size);
+			isapnp_parse_dma_resource(dev, option_flags, size);
 			size = 0;
 			break;
 		case _STAG_STARTDEP:
@@ -656,29 +650,24 @@ static int __init isapnp_create_device(struct pnp_card *card,
 				priority = tmp[0];
 				size = 0;
 			}
-			option = pnp_register_dependent_option(dev, priority);
-			if (!option)
-				return 1;
+			option_flags = pnp_new_dependent_set(dev, priority);
 			break;
 		case _STAG_ENDDEP:
 			if (size != 0)
 				goto __skip;
-			if (option_independent == option)
-				dev_warn(&dev->dev, "missing "
-					 "_STAG_STARTDEP tag\n");
-			option = option_independent;
-			dev_dbg(&dev->dev, "end dependent options\n");
+			option_flags = 0;
 			break;
 		case _STAG_IOPORT:
 			if (size != 7)
 				goto __skip;
-			isapnp_parse_port_resource(dev, option, size);
+			isapnp_parse_port_resource(dev, option_flags, size);
 			size = 0;
 			break;
 		case _STAG_FIXEDIO:
 			if (size != 3)
 				goto __skip;
-			isapnp_parse_fixed_port_resource(dev, option, size);
+			isapnp_parse_fixed_port_resource(dev, option_flags,
+							 size);
 			size = 0;
 			break;
 		case _STAG_VENDOR:
@@ -686,7 +675,7 @@ static int __init isapnp_create_device(struct pnp_card *card,
 		case _LTAG_MEMRANGE:
 			if (size != 9)
 				goto __skip;
-			isapnp_parse_mem_resource(dev, option, size);
+			isapnp_parse_mem_resource(dev, option_flags, size);
 			size = 0;
 			break;
 		case _LTAG_ANSISTR:
@@ -701,13 +690,14 @@ static int __init isapnp_create_device(struct pnp_card *card,
 		case _LTAG_MEM32RANGE:
 			if (size != 17)
 				goto __skip;
-			isapnp_parse_mem32_resource(dev, option, size);
+			isapnp_parse_mem32_resource(dev, option_flags, size);
 			size = 0;
 			break;
 		case _LTAG_FIXEDMEM32RANGE:
 			if (size != 9)
 				goto __skip;
-			isapnp_parse_fixed_mem32_resource(dev, option, size);
+			isapnp_parse_fixed_mem32_resource(dev, option_flags,
+							  size);
 			size = 0;
 			break;
 		case _STAG_END:
diff --git a/drivers/pnp/manager.c b/drivers/pnp/manager.c
index a20accb5ef8f..b526eaad3f6c 100644
--- a/drivers/pnp/manager.c
+++ b/drivers/pnp/manager.c
@@ -3,6 +3,8 @@
  *
  * based on isapnp.c resource management (c) Jaroslav Kysela <perex@perex.cz>
  * Copyright 2003 Adam Belay <ambx1@neo.rr.com>
+ * Copyright (C) 2008 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
  */
 
 #include <linux/errno.h>
@@ -228,102 +230,51 @@ static void pnp_clean_resource_table(struct pnp_dev *dev)
 /**
  * pnp_assign_resources - assigns resources to the device based on the specified dependent number
  * @dev: pointer to the desired device
- * @depnum: the dependent function number
- *
- * Only set depnum to 0 if the device does not have dependent options.
+ * @set: the dependent function number
  */
-static int pnp_assign_resources(struct pnp_dev *dev, int depnum)
+static int pnp_assign_resources(struct pnp_dev *dev, int set)
 {
-	struct pnp_port *port;
-	struct pnp_mem *mem;
-	struct pnp_irq *irq;
-	struct pnp_dma *dma;
+	struct pnp_option *option;
 	int nport = 0, nmem = 0, nirq = 0, ndma = 0;
+	int ret = 0;
 
-	dbg_pnp_show_resources(dev, "before pnp_assign_resources");
+	dev_dbg(&dev->dev, "pnp_assign_resources, try dependent set %d\n", set);
 	mutex_lock(&pnp_res_mutex);
 	pnp_clean_resource_table(dev);
-	if (dev->independent) {
-		dev_dbg(&dev->dev, "assigning independent options\n");
-		port = dev->independent->port;
-		mem = dev->independent->mem;
-		irq = dev->independent->irq;
-		dma = dev->independent->dma;
-		while (port) {
-			if (pnp_assign_port(dev, port, nport) < 0)
-				goto fail;
-			nport++;
-			port = port->next;
-		}
-		while (mem) {
-			if (pnp_assign_mem(dev, mem, nmem) < 0)
-				goto fail;
-			nmem++;
-			mem = mem->next;
-		}
-		while (irq) {
-			if (pnp_assign_irq(dev, irq, nirq) < 0)
-				goto fail;
-			nirq++;
-			irq = irq->next;
-		}
-		while (dma) {
-			if (pnp_assign_dma(dev, dma, ndma) < 0)
-				goto fail;
-			ndma++;
-			dma = dma->next;
-		}
-	}
 
-	if (depnum) {
-		struct pnp_option *dep;
-		int i;
-
-		dev_dbg(&dev->dev, "assigning dependent option %d\n", depnum);
-		for (i = 1, dep = dev->dependent; i < depnum;
-		     i++, dep = dep->next)
-			if (!dep)
-				goto fail;
-		port = dep->port;
-		mem = dep->mem;
-		irq = dep->irq;
-		dma = dep->dma;
-		while (port) {
-			if (pnp_assign_port(dev, port, nport) < 0)
-				goto fail;
-			nport++;
-			port = port->next;
-		}
-		while (mem) {
-			if (pnp_assign_mem(dev, mem, nmem) < 0)
-				goto fail;
-			nmem++;
-			mem = mem->next;
-		}
-		while (irq) {
-			if (pnp_assign_irq(dev, irq, nirq) < 0)
-				goto fail;
-			nirq++;
-			irq = irq->next;
-		}
-		while (dma) {
-			if (pnp_assign_dma(dev, dma, ndma) < 0)
-				goto fail;
-			ndma++;
-			dma = dma->next;
+	list_for_each_entry(option, &dev->options, list) {
+		if (pnp_option_is_dependent(option) &&
+		    pnp_option_set(option) != set)
+				continue;
+
+		switch (option->type) {
+		case IORESOURCE_IO:
+			ret = pnp_assign_port(dev, &option->u.port, nport++);
+			break;
+		case IORESOURCE_MEM:
+			ret = pnp_assign_mem(dev, &option->u.mem, nmem++);
+			break;
+		case IORESOURCE_IRQ:
+			ret = pnp_assign_irq(dev, &option->u.irq, nirq++);
+			break;
+		case IORESOURCE_DMA:
+			ret = pnp_assign_dma(dev, &option->u.dma, ndma++);
+			break;
+		default:
+			ret = -EINVAL;
+			break;
 		}
-	} else if (dev->dependent)
-		goto fail;
-
-	mutex_unlock(&pnp_res_mutex);
-	dbg_pnp_show_resources(dev, "after pnp_assign_resources");
-	return 1;
+		if (ret < 0)
+			break;
+	}
 
-fail:
-	pnp_clean_resource_table(dev);
 	mutex_unlock(&pnp_res_mutex);
-	dbg_pnp_show_resources(dev, "after pnp_assign_resources (failed)");
-	return 0;
+	if (ret < 0) {
+		dev_dbg(&dev->dev, "pnp_assign_resources failed (%d)\n", ret);
+		pnp_clean_resource_table(dev);
+	} else
+		dbg_pnp_show_resources(dev, "pnp_assign_resources succeeded");
+	return ret;
 }
 
 /**
@@ -332,29 +283,25 @@ fail:
  */
 int pnp_auto_config_dev(struct pnp_dev *dev)
 {
-	struct pnp_option *dep;
-	int i = 1;
+	int i, ret;
 
 	if (!pnp_can_configure(dev)) {
 		dev_dbg(&dev->dev, "configuration not supported\n");
 		return -ENODEV;
 	}
 
-	if (!dev->dependent) {
-		if (pnp_assign_resources(dev, 0))
+	ret = pnp_assign_resources(dev, 0);
+	if (ret == 0)
+		return 0;
+
+	for (i = 1; i < dev->num_dependent_sets; i++) {
+		ret = pnp_assign_resources(dev, i);
+		if (ret == 0)
 			return 0;
-	} else {
-		dep = dev->dependent;
-		do {
-			if (pnp_assign_resources(dev, i))
-				return 0;
-			dep = dep->next;
-			i++;
-		} while (dep);
 	}
 
 	dev_err(&dev->dev, "unable to assign resources\n");
-	return -EBUSY;
+	return ret;
 }
 
 /**
diff --git a/drivers/pnp/pnpacpi/rsparser.c b/drivers/pnp/pnpacpi/rsparser.c
index e114b3d2b933..c2f59f4d20bc 100644
--- a/drivers/pnp/pnpacpi/rsparser.c
+++ b/drivers/pnp/pnpacpi/rsparser.c
@@ -407,7 +407,7 @@ int pnpacpi_parse_allocated_resource(struct pnp_dev *dev)
 }
 
 static __init void pnpacpi_parse_dma_option(struct pnp_dev *dev,
-					    struct pnp_option *option,
+					    unsigned int option_flags,
 					    struct acpi_resource_dma *p)
 {
 	int i;
@@ -420,11 +420,11 @@ static __init void pnpacpi_parse_dma_option(struct pnp_dev *dev,
 		map |= 1 << p->channels[i];
 
 	flags = dma_flags(p->type, p->bus_master, p->transfer);
-	pnp_register_dma_resource(dev, option, map, flags);
+	pnp_register_dma_resource(dev, option_flags, map, flags);
 }
 
 static __init void pnpacpi_parse_irq_option(struct pnp_dev *dev,
-					    struct pnp_option *option,
+					    unsigned int option_flags,
 					    struct acpi_resource_irq *p)
 {
 	int i;
@@ -440,11 +440,11 @@ static __init void pnpacpi_parse_irq_option(struct pnp_dev *dev,
 			__set_bit(p->interrupts[i], map.bits);
 
 	flags = irq_flags(p->triggering, p->polarity, p->sharable);
-	pnp_register_irq_resource(dev, option, &map, flags);
+	pnp_register_irq_resource(dev, option_flags, &map, flags);
 }
 
 static __init void pnpacpi_parse_ext_irq_option(struct pnp_dev *dev,
-						struct pnp_option *option,
+					unsigned int option_flags,
 					struct acpi_resource_extended_irq *p)
 {
 	int i;
@@ -467,11 +467,11 @@ static __init void pnpacpi_parse_ext_irq_option(struct pnp_dev *dev,
 	}
 
 	flags = irq_flags(p->triggering, p->polarity, p->sharable);
-	pnp_register_irq_resource(dev, option, &map, flags);
+	pnp_register_irq_resource(dev, option_flags, &map, flags);
 }
 
 static __init void pnpacpi_parse_port_option(struct pnp_dev *dev,
-					     struct pnp_option *option,
+					     unsigned int option_flags,
 					     struct acpi_resource_io *io)
 {
 	unsigned char flags = 0;
@@ -481,23 +481,23 @@ static __init void pnpacpi_parse_port_option(struct pnp_dev *dev,
 
 	if (io->io_decode == ACPI_DECODE_16)
 		flags = IORESOURCE_IO_16BIT_ADDR;
-	pnp_register_port_resource(dev, option, io->minimum, io->maximum,
+	pnp_register_port_resource(dev, option_flags, io->minimum, io->maximum,
 				   io->alignment, io->address_length, flags);
 }
 
 static __init void pnpacpi_parse_fixed_port_option(struct pnp_dev *dev,
-						   struct pnp_option *option,
+					unsigned int option_flags,
 					struct acpi_resource_fixed_io *io)
 {
 	if (io->address_length == 0)
 		return;
 
-	pnp_register_port_resource(dev, option, io->address, io->address, 0,
-				   io->address_length, IORESOURCE_IO_FIXED);
+	pnp_register_port_resource(dev, option_flags, io->address, io->address,
+				   0, io->address_length, IORESOURCE_IO_FIXED);
 }
 
 static __init void pnpacpi_parse_mem24_option(struct pnp_dev *dev,
-					      struct pnp_option *option,
+					      unsigned int option_flags,
 					      struct acpi_resource_memory24 *p)
 {
 	unsigned char flags = 0;
@@ -507,12 +507,12 @@ static __init void pnpacpi_parse_mem24_option(struct pnp_dev *dev,
 
 	if (p->write_protect == ACPI_READ_WRITE_MEMORY)
 		flags = IORESOURCE_MEM_WRITEABLE;
-	pnp_register_mem_resource(dev, option, p->minimum, p->maximum,
+	pnp_register_mem_resource(dev, option_flags, p->minimum, p->maximum,
 				  p->alignment, p->address_length, flags);
 }
 
 static __init void pnpacpi_parse_mem32_option(struct pnp_dev *dev,
-					      struct pnp_option *option,
+					      unsigned int option_flags,
 					      struct acpi_resource_memory32 *p)
 {
 	unsigned char flags = 0;
@@ -522,12 +522,12 @@ static __init void pnpacpi_parse_mem32_option(struct pnp_dev *dev,
 
 	if (p->write_protect == ACPI_READ_WRITE_MEMORY)
 		flags = IORESOURCE_MEM_WRITEABLE;
-	pnp_register_mem_resource(dev, option, p->minimum, p->maximum,
+	pnp_register_mem_resource(dev, option_flags, p->minimum, p->maximum,
 				  p->alignment, p->address_length, flags);
 }
 
 static __init void pnpacpi_parse_fixed_mem32_option(struct pnp_dev *dev,
-						    struct pnp_option *option,
+					unsigned int option_flags,
 					struct acpi_resource_fixed_memory32 *p)
 {
 	unsigned char flags = 0;
@@ -537,12 +537,12 @@ static __init void pnpacpi_parse_fixed_mem32_option(struct pnp_dev *dev,
 
 	if (p->write_protect == ACPI_READ_WRITE_MEMORY)
 		flags = IORESOURCE_MEM_WRITEABLE;
-	pnp_register_mem_resource(dev, option, p->address, p->address,
+	pnp_register_mem_resource(dev, option_flags, p->address, p->address,
 				  0, p->address_length, flags);
 }
 
 static __init void pnpacpi_parse_address_option(struct pnp_dev *dev,
-						struct pnp_option *option,
+						unsigned int option_flags,
 						struct acpi_resource *r)
 {
 	struct acpi_resource_address64 addr, *p = &addr;
@@ -562,18 +562,18 @@ static __init void pnpacpi_parse_address_option(struct pnp_dev *dev,
 	if (p->resource_type == ACPI_MEMORY_RANGE) {
 		if (p->info.mem.write_protect == ACPI_READ_WRITE_MEMORY)
 			flags = IORESOURCE_MEM_WRITEABLE;
-		pnp_register_mem_resource(dev, option, p->minimum, p->minimum,
-					  0, p->address_length, flags);
+		pnp_register_mem_resource(dev, option_flags, p->minimum,
+					  p->minimum, 0, p->address_length,
+					  flags);
 	} else if (p->resource_type == ACPI_IO_RANGE)
-		pnp_register_port_resource(dev, option, p->minimum, p->minimum,
-					   0, p->address_length,
+		pnp_register_port_resource(dev, option_flags, p->minimum,
+					   p->minimum, 0, p->address_length,
 					   IORESOURCE_IO_FIXED);
 }
 
 struct acpipnp_parse_option_s {
-	struct pnp_option *option;
-	struct pnp_option *option_independent;
 	struct pnp_dev *dev;
+	unsigned int option_flags;
 };
 
 static __init acpi_status pnpacpi_option_resource(struct acpi_resource *res,
@@ -582,15 +582,15 @@ static __init acpi_status pnpacpi_option_resource(struct acpi_resource *res,
 	int priority;
 	struct acpipnp_parse_option_s *parse_data = data;
 	struct pnp_dev *dev = parse_data->dev;
-	struct pnp_option *option = parse_data->option;
+	unsigned int option_flags = parse_data->option_flags;
 
 	switch (res->type) {
 	case ACPI_RESOURCE_TYPE_IRQ:
-		pnpacpi_parse_irq_option(dev, option, &res->data.irq);
+		pnpacpi_parse_irq_option(dev, option_flags, &res->data.irq);
 		break;
 
 	case ACPI_RESOURCE_TYPE_DMA:
-		pnpacpi_parse_dma_option(dev, option, &res->data.dma);
+		pnpacpi_parse_dma_option(dev, option_flags, &res->data.dma);
 		break;
 
 	case ACPI_RESOURCE_TYPE_START_DEPENDENT:
@@ -610,31 +610,19 @@ static __init acpi_status pnpacpi_option_resource(struct acpi_resource *res,
 			priority = PNP_RES_PRIORITY_INVALID;
 			break;
 		}
-		/* TBD: Consider performance/robustness bits */
-		option = pnp_register_dependent_option(dev, priority);
-		if (!option)
-			return AE_ERROR;
-		parse_data->option = option;
+		parse_data->option_flags = pnp_new_dependent_set(dev, priority);
 		break;
 
 	case ACPI_RESOURCE_TYPE_END_DEPENDENT:
-		/*only one EndDependentFn is allowed */
-		if (!parse_data->option_independent) {
-			dev_warn(&dev->dev, "more than one EndDependentFn "
-				 "in _PRS\n");
-			return AE_ERROR;
-		}
-		parse_data->option = parse_data->option_independent;
-		parse_data->option_independent = NULL;
-		dev_dbg(&dev->dev, "end dependent options\n");
+		parse_data->option_flags = 0;
 		break;
 
 	case ACPI_RESOURCE_TYPE_IO:
-		pnpacpi_parse_port_option(dev, option, &res->data.io);
+		pnpacpi_parse_port_option(dev, option_flags, &res->data.io);
 		break;
 
 	case ACPI_RESOURCE_TYPE_FIXED_IO:
-		pnpacpi_parse_fixed_port_option(dev, option,
+		pnpacpi_parse_fixed_port_option(dev, option_flags,
 					        &res->data.fixed_io);
 		break;
 
@@ -643,29 +631,31 @@ static __init acpi_status pnpacpi_option_resource(struct acpi_resource *res,
 		break;
 
 	case ACPI_RESOURCE_TYPE_MEMORY24:
-		pnpacpi_parse_mem24_option(dev, option, &res->data.memory24);
+		pnpacpi_parse_mem24_option(dev, option_flags,
+					   &res->data.memory24);
 		break;
 
 	case ACPI_RESOURCE_TYPE_MEMORY32:
-		pnpacpi_parse_mem32_option(dev, option, &res->data.memory32);
+		pnpacpi_parse_mem32_option(dev, option_flags,
+					   &res->data.memory32);
 		break;
 
 	case ACPI_RESOURCE_TYPE_FIXED_MEMORY32:
-		pnpacpi_parse_fixed_mem32_option(dev, option,
+		pnpacpi_parse_fixed_mem32_option(dev, option_flags,
 						 &res->data.fixed_memory32);
 		break;
 
 	case ACPI_RESOURCE_TYPE_ADDRESS16:
 	case ACPI_RESOURCE_TYPE_ADDRESS32:
 	case ACPI_RESOURCE_TYPE_ADDRESS64:
-		pnpacpi_parse_address_option(dev, option, res);
+		pnpacpi_parse_address_option(dev, option_flags, res);
 		break;
 
 	case ACPI_RESOURCE_TYPE_EXTENDED_ADDRESS64:
 		break;
 
 	case ACPI_RESOURCE_TYPE_EXTENDED_IRQ:
-		pnpacpi_parse_ext_irq_option(dev, option,
+		pnpacpi_parse_ext_irq_option(dev, option_flags,
 					     &res->data.extended_irq);
 		break;
 
@@ -689,12 +679,9 @@ int __init pnpacpi_parse_resource_option_data(struct pnp_dev *dev)
 
 	dev_dbg(&dev->dev, "parse resource options\n");
 
-	parse_data.option = pnp_register_independent_option(dev);
-	if (!parse_data.option)
-		return -ENOMEM;
-
-	parse_data.option_independent = parse_data.option;
 	parse_data.dev = dev;
+	parse_data.option_flags = 0;
+
 	status = acpi_walk_resources(handle, METHOD_NAME__PRS,
 				     pnpacpi_option_resource, &parse_data);
 
diff --git a/drivers/pnp/pnpbios/rsparser.c b/drivers/pnp/pnpbios/rsparser.c
index db23ba78d39c..ca567671379e 100644
--- a/drivers/pnp/pnpbios/rsparser.c
+++ b/drivers/pnp/pnpbios/rsparser.c
@@ -216,7 +216,7 @@ len_err:
 
 static __init void pnpbios_parse_mem_option(struct pnp_dev *dev,
 					    unsigned char *p, int size,
-					    struct pnp_option *option)
+					    unsigned int option_flags)
 {
 	resource_size_t min, max, align, len;
 	unsigned char flags;
@@ -226,12 +226,13 @@ static __init void pnpbios_parse_mem_option(struct pnp_dev *dev,
 	align = (p[9] << 8) | p[8];
 	len = ((p[11] << 8) | p[10]) << 8;
 	flags = p[3];
-	pnp_register_mem_resource(dev, option, min, max, align, len, flags);
+	pnp_register_mem_resource(dev, option_flags, min, max, align, len,
+				  flags);
 }
 
 static __init void pnpbios_parse_mem32_option(struct pnp_dev *dev,
 					      unsigned char *p, int size,
-					      struct pnp_option *option)
+					      unsigned int option_flags)
 {
 	resource_size_t min, max, align, len;
 	unsigned char flags;
@@ -241,12 +242,13 @@ static __init void pnpbios_parse_mem32_option(struct pnp_dev *dev,
 	align = (p[15] << 24) | (p[14] << 16) | (p[13] << 8) | p[12];
 	len = (p[19] << 24) | (p[18] << 16) | (p[17] << 8) | p[16];
 	flags = p[3];
-	pnp_register_mem_resource(dev, option, min, max, align, len, flags);
+	pnp_register_mem_resource(dev, option_flags, min, max, align, len,
+				  flags);
 }
 
 static __init void pnpbios_parse_fixed_mem32_option(struct pnp_dev *dev,
 						    unsigned char *p, int size,
-						    struct pnp_option *option)
+						    unsigned int option_flags)
 {
 	resource_size_t base, len;
 	unsigned char flags;
@@ -254,12 +256,12 @@ static __init void pnpbios_parse_fixed_mem32_option(struct pnp_dev *dev,
 	base = (p[7] << 24) | (p[6] << 16) | (p[5] << 8) | p[4];
 	len = (p[11] << 24) | (p[10] << 16) | (p[9] << 8) | p[8];
 	flags = p[3];
-	pnp_register_mem_resource(dev, option, base, base, 0, len, flags);
+	pnp_register_mem_resource(dev, option_flags, base, base, 0, len, flags);
 }
 
 static __init void pnpbios_parse_irq_option(struct pnp_dev *dev,
 					    unsigned char *p, int size,
-					    struct pnp_option *option)
+					    unsigned int option_flags)
 {
 	unsigned long bits;
 	pnp_irq_mask_t map;
@@ -273,19 +275,19 @@ static __init void pnpbios_parse_irq_option(struct pnp_dev *dev,
 	if (size > 2)
 		flags = p[3];
 
-	pnp_register_irq_resource(dev, option, &map, flags);
+	pnp_register_irq_resource(dev, option_flags, &map, flags);
 }
 
 static __init void pnpbios_parse_dma_option(struct pnp_dev *dev,
 					    unsigned char *p, int size,
-					    struct pnp_option *option)
+					    unsigned int option_flags)
 {
-	pnp_register_dma_resource(dev, option, p[1], p[2]);
+	pnp_register_dma_resource(dev, option_flags, p[1], p[2]);
 }
 
 static __init void pnpbios_parse_port_option(struct pnp_dev *dev,
 					     unsigned char *p, int size,
-					     struct pnp_option *option)
+					     unsigned int option_flags)
 {
 	resource_size_t min, max, align, len;
 	unsigned char flags;
@@ -295,38 +297,35 @@ static __init void pnpbios_parse_port_option(struct pnp_dev *dev,
 	align = p[6];
 	len = p[7];
 	flags = p[1] ? IORESOURCE_IO_16BIT_ADDR : 0;
-	pnp_register_port_resource(dev, option, min, max, align, len, flags);
+	pnp_register_port_resource(dev, option_flags, min, max, align, len,
+				   flags);
 }
 
 static __init void pnpbios_parse_fixed_port_option(struct pnp_dev *dev,
 						   unsigned char *p, int size,
-						   struct pnp_option *option)
+						   unsigned int option_flags)
 {
 	resource_size_t base, len;
 
 	base = (p[2] << 8) | p[1];
 	len = p[3];
-	pnp_register_port_resource(dev, option, base, base, 0, len,
+	pnp_register_port_resource(dev, option_flags, base, base, 0, len,
 				   IORESOURCE_IO_FIXED);
 }
 
 static __init unsigned char *
 pnpbios_parse_resource_option_data(unsigned char *p, unsigned char *end,
-					struct pnp_dev *dev)
+				   struct pnp_dev *dev)
 {
 	unsigned int len, tag;
 	int priority;
-	struct pnp_option *option, *option_independent;
+	unsigned int option_flags;
 
 	if (!p)
 		return NULL;
 
 	dev_dbg(&dev->dev, "parse resource options\n");
-
-	option_independent = option = pnp_register_independent_option(dev);
-	if (!option)
-		return NULL;
-
+	option_flags = 0;
 	while ((char *)p < (char *)end) {
 
 		/* determine the type of tag */
@@ -343,37 +342,38 @@ pnpbios_parse_resource_option_data(unsigned char *p, unsigned char *end,
 		case LARGE_TAG_MEM:
 			if (len != 9)
 				goto len_err;
-			pnpbios_parse_mem_option(dev, p, len, option);
+			pnpbios_parse_mem_option(dev, p, len, option_flags);
 			break;
 
 		case LARGE_TAG_MEM32:
 			if (len != 17)
 				goto len_err;
-			pnpbios_parse_mem32_option(dev, p, len, option);
+			pnpbios_parse_mem32_option(dev, p, len, option_flags);
 			break;
 
 		case LARGE_TAG_FIXEDMEM32:
 			if (len != 9)
 				goto len_err;
-			pnpbios_parse_fixed_mem32_option(dev, p, len, option);
+			pnpbios_parse_fixed_mem32_option(dev, p, len,
+							 option_flags);
 			break;
 
 		case SMALL_TAG_IRQ:
 			if (len < 2 || len > 3)
 				goto len_err;
-			pnpbios_parse_irq_option(dev, p, len, option);
+			pnpbios_parse_irq_option(dev, p, len, option_flags);
 			break;
 
 		case SMALL_TAG_DMA:
 			if (len != 2)
 				goto len_err;
-			pnpbios_parse_dma_option(dev, p, len, option);
+			pnpbios_parse_dma_option(dev, p, len, option_flags);
 			break;
 
 		case SMALL_TAG_PORT:
 			if (len != 7)
 				goto len_err;
-			pnpbios_parse_port_option(dev, p, len, option);
+			pnpbios_parse_port_option(dev, p, len, option_flags);
 			break;
 
 		case SMALL_TAG_VENDOR:
@@ -383,7 +383,8 @@ pnpbios_parse_resource_option_data(unsigned char *p, unsigned char *end,
 		case SMALL_TAG_FIXEDPORT:
 			if (len != 3)
 				goto len_err;
-			pnpbios_parse_fixed_port_option(dev, p, len, option);
+			pnpbios_parse_fixed_port_option(dev, p, len,
+							option_flags);
 			break;
 
 		case SMALL_TAG_STARTDEP:
@@ -392,19 +393,13 @@ pnpbios_parse_resource_option_data(unsigned char *p, unsigned char *end,
 			priority = PNP_RES_PRIORITY_ACCEPTABLE;
 			if (len > 0)
 				priority = p[1];
-			option = pnp_register_dependent_option(dev, priority);
-			if (!option)
-				return NULL;
+			option_flags = pnp_new_dependent_set(dev, priority);
 			break;
 
 		case SMALL_TAG_ENDDEP:
 			if (len != 0)
 				goto len_err;
-			if (option_independent == option)
-				dev_warn(&dev->dev, "missing "
-					 "SMALL_TAG_STARTDEP tag\n");
-			option = option_independent;
-			dev_dbg(&dev->dev, "end dependent options\n");
+			option_flags = 0;
 			break;
 
 		case SMALL_TAG_END:
diff --git a/drivers/pnp/quirks.c b/drivers/pnp/quirks.c
index e8515ce0d296..55f55ed72dc7 100644
--- a/drivers/pnp/quirks.c
+++ b/drivers/pnp/quirks.c
@@ -5,6 +5,8 @@
  *  when building up the resource structure for the first time.
  *
  *  Copyright (c) 2000 Peter Denison <peterd@pnd-pc.demon.co.uk>
+ *  Copyright (C) 2008 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
  *
  *  Heavily based on PCI quirks handling which is
  *
@@ -20,189 +22,207 @@
 #include <linux/kallsyms.h>
 #include "base.h"
 
+static void quirk_awe32_add_ports(struct pnp_dev *dev,
+				  struct pnp_option *option,
+				  unsigned int offset)
+{
+	struct pnp_option *new_option;
+
+	new_option = kmalloc(sizeof(struct pnp_option), GFP_KERNEL);
+	if (!new_option) {
+		dev_err(&dev->dev, "couldn't add ioport region to option set "
+			"%d\n", pnp_option_set(option));
+		return;
+	}
+
+	*new_option = *option;
+	new_option->u.port.min += offset;
+	new_option->u.port.max += offset;
+	list_add(&new_option->list, &option->list);
+
+	dev_info(&dev->dev, "added ioport region %#llx-%#llx to set %d\n",
+		(unsigned long long) new_option->u.port.min,
+		(unsigned long long) new_option->u.port.max,
+		pnp_option_set(option));
+}
+
 static void quirk_awe32_resources(struct pnp_dev *dev)
 {
-	struct pnp_port *port, *port2, *port3;
-	struct pnp_option *res = dev->dependent;
+	struct pnp_option *option;
+	unsigned int set = ~0;
 
 	/*
-	 * Unfortunately the isapnp_add_port_resource is too tightly bound
-	 * into the PnP discovery sequence, and cannot be used. Link in the
-	 * two extra ports (at offset 0x400 and 0x800 from the one given) by
-	 * hand.
+	 * Add two extra ioport regions (at offset 0x400 and 0x800 from the
+	 * one given) to every dependent option set.
 	 */
-	for (; res; res = res->next) {
-		port2 = pnp_alloc(sizeof(struct pnp_port));
-		if (!port2)
-			return;
-		port3 = pnp_alloc(sizeof(struct pnp_port));
-		if (!port3) {
-			kfree(port2);
-			return;
+	list_for_each_entry(option, &dev->options, list) {
+		if (pnp_option_is_dependent(option) &&
+		    pnp_option_set(option) != set) {
+			set = pnp_option_set(option);
+			quirk_awe32_add_ports(dev, option, 0x800);
+			quirk_awe32_add_ports(dev, option, 0x400);
 		}
-		port = res->port;
-		memcpy(port2, port, sizeof(struct pnp_port));
-		memcpy(port3, port, sizeof(struct pnp_port));
-		port->next = port2;
-		port2->next = port3;
-		port2->min += 0x400;
-		port2->max += 0x400;
-		port3->min += 0x800;
-		port3->max += 0x800;
-		dev_info(&dev->dev,
-			"AWE32 quirk - added ioports 0x%lx and 0x%lx\n",
-			(unsigned long)port2->min,
-			(unsigned long)port3->min);
 	}
 }
 
 static void quirk_cmi8330_resources(struct pnp_dev *dev)
 {
-	struct pnp_option *res = dev->dependent;
-	unsigned long tmp;
-
-	for (; res; res = res->next) {
-
-		struct pnp_irq *irq;
-		struct pnp_dma *dma;
+	struct pnp_option *option;
+	struct pnp_irq *irq;
+	struct pnp_dma *dma;
 
-		for (irq = res->irq; irq; irq = irq->next) {
-			/* Valid irqs are 5, 7, 10 */
-			tmp = 0x04A0;
-			bitmap_copy(irq->map.bits, &tmp, 16);
-		}
+	list_for_each_entry(option, &dev->options, list) {
+		if (!pnp_option_is_dependent(option))
+			continue;
 
-		for (dma = res->dma; dma; dma = dma->next) {
-			/* Valid 8bit dma channels are 1,3 */
+		if (option->type == IORESOURCE_IRQ) {
+			irq = &option->u.irq;
+			bitmap_zero(irq->map.bits, PNP_IRQ_NR);
+			__set_bit(5, irq->map.bits);
+			__set_bit(7, irq->map.bits);
+			__set_bit(10, irq->map.bits);
+			dev_info(&dev->dev, "set possible IRQs in "
+				 "option set %d to 5, 7, 10\n",
+				 pnp_option_set(option));
+		} else if (option->type == IORESOURCE_DMA) {
+			dma = &option->u.dma;
 			if ((dma->flags & IORESOURCE_DMA_TYPE_MASK) ==
-			    IORESOURCE_DMA_8BIT)
-				dma->map = 0x000A;
+						IORESOURCE_DMA_8BIT &&
+			    dma->map != 0x0A) {
+				dev_info(&dev->dev, "changing possible "
+					 "DMA channel mask in option set %d "
+					 "from %#02x to 0x0A (1, 3)\n",
+					 pnp_option_set(option), dma->map);
+				dma->map = 0x0A;
+			}
 		}
 	}
-	dev_info(&dev->dev, "CMI8330 quirk - forced possible IRQs to 5, 7, 10 "
-		"and DMA channels to 1, 3\n");
 }
 
 static void quirk_sb16audio_resources(struct pnp_dev *dev)
 {
+	struct pnp_option *option;
+	unsigned int prev_option_flags = ~0, n = 0;
 	struct pnp_port *port;
-	struct pnp_option *res = dev->dependent;
-	int changed = 0;
 
 	/*
-	 * The default range on the mpu port for these devices is 0x388-0x388.
+	 * The default range on the OPL port for these devices is 0x388-0x388.
 	 * Here we increase that range so that two such cards can be
 	 * auto-configured.
 	 */
+	list_for_each_entry(option, &dev->options, list) {
+		if (prev_option_flags != option->flags) {
+			prev_option_flags = option->flags;
+			n = 0;
+		}
 
-	for (; res; res = res->next) {
-		port = res->port;
-		if (!port)
-			continue;
-		port = port->next;
-		if (!port)
-			continue;
-		port = port->next;
-		if (!port)
-			continue;
-		if (port->min != port->max)
-			continue;
-		port->max += 0x70;
-		changed = 1;
+		if (pnp_option_is_dependent(option) &&
+		    option->type == IORESOURCE_IO) {
+			n++;
+			port = &option->u.port;
+			if (n == 3 && port->min == port->max) {
+				port->max += 0x70;
+				dev_info(&dev->dev, "increased option port "
+					 "range from %#llx-%#llx to "
+					 "%#llx-%#llx\n",
+					 (unsigned long long) port->min,
+					 (unsigned long long) port->min,
+					 (unsigned long long) port->min,
+					 (unsigned long long) port->max);
+			}
+		}
 	}
-	if (changed)
-		dev_info(&dev->dev, "SB audio device quirk - increased port range\n");
 }
 
-static struct pnp_option *quirk_isapnp_mpu_options(struct pnp_dev *dev)
+static struct pnp_option *pnp_clone_dependent_set(struct pnp_dev *dev,
+						  unsigned int set)
 {
-	struct pnp_option *head = NULL;
-	struct pnp_option *prev = NULL;
-	struct pnp_option *res;
-
-	/*
-	 * Build a functional IRQ-optional variant of each MPU option.
-	 */
-
-	for (res = dev->dependent; res; res = res->next) {
-		struct pnp_option *curr;
-		struct pnp_port *port;
-		struct pnp_port *copy_port;
-		struct pnp_irq *irq;
-		struct pnp_irq *copy_irq;
-
-		port = res->port;
-		irq = res->irq;
-		if (!port || !irq)
-			continue;
+	struct pnp_option *tail = NULL, *first_new_option = NULL;
+	struct pnp_option *option, *new_option;
+	unsigned int flags;
 
-		copy_port = pnp_alloc(sizeof *copy_port);
-		if (!copy_port)
-			break;
-
-		copy_irq = pnp_alloc(sizeof *copy_irq);
-		if (!copy_irq) {
-			kfree(copy_port);
-			break;
-		}
+	list_for_each_entry(option, &dev->options, list) {
+		if (pnp_option_is_dependent(option))
+			tail = option;
+	}
+	if (!tail) {
+		dev_err(&dev->dev, "no dependent option sets\n");
+		return NULL;
+	}
 
-		*copy_port = *port;
-		copy_port->next = NULL;
+	flags = pnp_new_dependent_set(dev, PNP_RES_PRIORITY_FUNCTIONAL);
+	list_for_each_entry(option, &dev->options, list) {
+		if (pnp_option_is_dependent(option) &&
+		    pnp_option_set(option) == set) {
+			new_option = kmalloc(sizeof(struct pnp_option),
+					     GFP_KERNEL);
+			if (!new_option) {
+				dev_err(&dev->dev, "couldn't clone dependent "
+					"set %d\n", set);
+				return NULL;
+			}
 
-		*copy_irq = *irq;
-		copy_irq->flags |= IORESOURCE_IRQ_OPTIONAL;
-		copy_irq->next = NULL;
+			*new_option = *option;
+			new_option->flags = flags;
+			if (!first_new_option)
+				first_new_option = new_option;
 
-		curr = pnp_build_option(PNP_RES_PRIORITY_FUNCTIONAL);
-		if (!curr) {
-			kfree(copy_port);
-			kfree(copy_irq);
-			break;
+			list_add(&new_option->list, &tail->list);
+			tail = new_option;
 		}
-		curr->port = copy_port;
-		curr->irq = copy_irq;
-
-		if (prev)
-			prev->next = curr;
-		else
-			head = curr;
-		prev = curr;
 	}
-	if (head)
-		dev_info(&dev->dev, "adding IRQ-optional MPU options\n");
 
-	return head;
+	return first_new_option;
 }
 
-static void quirk_ad1815_mpu_resources(struct pnp_dev *dev)
+
+static void quirk_add_irq_optional_dependent_sets(struct pnp_dev *dev)
 {
-	struct pnp_option *res;
+	struct pnp_option *new_option;
+	unsigned int num_sets, i, set;
 	struct pnp_irq *irq;
 
-	res = dev->independent;
-	if (!res)
-		return;
+	num_sets = dev->num_dependent_sets;
+	for (i = 0; i < num_sets; i++) {
+		new_option = pnp_clone_dependent_set(dev, i);
+		if (!new_option)
+			return;
 
-	irq = res->irq;
-	if (!irq || irq->next)
-		return;
+		set = pnp_option_set(new_option);
+		while (new_option && pnp_option_set(new_option) == set) {
+			if (new_option->type == IORESOURCE_IRQ) {
+				irq = &new_option->u.irq;
+				irq->flags |= IORESOURCE_IRQ_OPTIONAL;
+			}
+			dbg_pnp_show_option(dev, new_option);
+			new_option = list_entry(new_option->list.next,
+						struct pnp_option, list);
+		}
 
-	irq->flags |= IORESOURCE_IRQ_OPTIONAL;
-	dev_info(&dev->dev, "made independent IRQ optional\n");
+		dev_info(&dev->dev, "added dependent option set %d (same as "
+			 "set %d except IRQ optional)\n", set, i);
+	}
 }
 
-static void quirk_isapnp_mpu_resources(struct pnp_dev *dev)
+static void quirk_ad1815_mpu_resources(struct pnp_dev *dev)
 {
-	struct pnp_option *res;
+	struct pnp_option *option;
+	struct pnp_irq *irq = NULL;
+	unsigned int independent_irqs = 0;
+
+	list_for_each_entry(option, &dev->options, list) {
+		if (option->type == IORESOURCE_IRQ &&
+		    !pnp_option_is_dependent(option)) {
+			independent_irqs++;
+			irq = &option->u.irq;
+		}
+	}
 
-	res = dev->dependent;
-	if (!res)
+	if (independent_irqs != 1)
 		return;
 
-	while (res->next)
-		res = res->next;
-
-	res->next = quirk_isapnp_mpu_options(dev);
+	irq->flags |= IORESOURCE_IRQ_OPTIONAL;
+	dev_info(&dev->dev, "made independent IRQ optional\n");
 }
 
 #include <linux/pci.h>
@@ -297,10 +317,10 @@ static struct pnp_fixup pnp_fixups[] = {
 	{"CTL0043", quirk_sb16audio_resources},
 	{"CTL0044", quirk_sb16audio_resources},
 	{"CTL0045", quirk_sb16audio_resources},
-	/* Add IRQ-less MPU options */
+	/* Add IRQ-optional MPU options */
 	{"ADS7151", quirk_ad1815_mpu_resources},
-	{"ADS7181", quirk_isapnp_mpu_resources},
-	{"AZT0002", quirk_isapnp_mpu_resources},
+	{"ADS7181", quirk_add_irq_optional_dependent_sets},
+	{"AZT0002", quirk_add_irq_optional_dependent_sets},
 	/* PnP resources that might overlap PCI BARs */
 	{"PNP0c01", quirk_system_pci_resources},
 	{"PNP0c02", quirk_system_pci_resources},
diff --git a/drivers/pnp/resource.c b/drivers/pnp/resource.c
index a795864dc695..d6388970a1a4 100644
--- a/drivers/pnp/resource.c
+++ b/drivers/pnp/resource.c
@@ -3,6 +3,8 @@
  *
  * based on isapnp.c resource management (c) Jaroslav Kysela <perex@perex.cz>
  * Copyright 2003 Adam Belay <ambx1@neo.rr.com>
+ * Copyright (C) 2008 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
  */
 
 #include <linux/module.h>
@@ -28,78 +30,36 @@ static int pnp_reserve_mem[16] = {[0 ... 15] = -1 };	/* reserve (don't use) some
  * option registration
  */
 
-struct pnp_option *pnp_build_option(int priority)
+struct pnp_option *pnp_build_option(struct pnp_dev *dev, unsigned long type,
+				    unsigned int option_flags)
 {
-	struct pnp_option *option = pnp_alloc(sizeof(struct pnp_option));
+	struct pnp_option *option;
 
+	option = kzalloc(sizeof(struct pnp_option), GFP_KERNEL);
 	if (!option)
 		return NULL;
 
-	option->priority = priority & 0xff;
-	/* make sure the priority is valid */
-	if (option->priority > PNP_RES_PRIORITY_FUNCTIONAL)
-		option->priority = PNP_RES_PRIORITY_INVALID;
-
-	return option;
-}
-
-struct pnp_option *pnp_register_independent_option(struct pnp_dev *dev)
-{
-	struct pnp_option *option;
-
-	option = pnp_build_option(PNP_RES_PRIORITY_PREFERRED);
-
-	/* this should never happen but if it does we'll try to continue */
-	if (dev->independent)
-		dev_err(&dev->dev, "independent resource already registered\n");
-	dev->independent = option;
-
-	dev_dbg(&dev->dev, "new independent option\n");
-	return option;
-}
-
-struct pnp_option *pnp_register_dependent_option(struct pnp_dev *dev,
-						 int priority)
-{
-	struct pnp_option *option;
+	option->flags = option_flags;
+	option->type = type;
 
-	option = pnp_build_option(priority);
-
-	if (dev->dependent) {
-		struct pnp_option *parent = dev->dependent;
-		while (parent->next)
-			parent = parent->next;
-		parent->next = option;
-	} else
-		dev->dependent = option;
-
-	dev_dbg(&dev->dev, "new dependent option (priority %#x)\n", priority);
+	list_add_tail(&option->list, &dev->options);
 	return option;
 }
 
-int pnp_register_irq_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_irq_resource(struct pnp_dev *dev, unsigned int option_flags,
 			      pnp_irq_mask_t *map, unsigned char flags)
 {
-	struct pnp_irq *irq, *ptr;
-#ifdef DEBUG
-	char buf[PNP_IRQ_NR];   /* hex-encoded, so this is overkill but safe */
-#endif
+	struct pnp_option *option;
+	struct pnp_irq *irq;
 
-	irq = kzalloc(sizeof(struct pnp_irq), GFP_KERNEL);
-	if (!irq)
+	option = pnp_build_option(dev, IORESOURCE_IRQ, option_flags);
+	if (!option)
 		return -ENOMEM;
 
+	irq = &option->u.irq;
 	irq->map = *map;
 	irq->flags = flags;
 
-	ptr = option->irq;
-	while (ptr && ptr->next)
-		ptr = ptr->next;
-	if (ptr)
-		ptr->next = irq;
-	else
-		option->irq = irq;
-
 #ifdef CONFIG_PCI
 	{
 		int i;
@@ -110,163 +70,81 @@ int pnp_register_irq_resource(struct pnp_dev *dev, struct pnp_option *option,
 	}
 #endif
 
-#ifdef DEBUG
-	bitmap_scnprintf(buf, sizeof(buf), irq->map.bits, PNP_IRQ_NR);
-	dev_dbg(&dev->dev, "  irq bitmask %s flags %#x\n", buf,
-		irq->flags);
-#endif
+	dbg_pnp_show_option(dev, option);
 	return 0;
 }
 
-int pnp_register_dma_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_dma_resource(struct pnp_dev *dev, unsigned int option_flags,
 			      unsigned char map, unsigned char flags)
 {
-	struct pnp_dma *dma, *ptr;
+	struct pnp_option *option;
+	struct pnp_dma *dma;
 
-	dma = kzalloc(sizeof(struct pnp_dma), GFP_KERNEL);
-	if (!dma)
+	option = pnp_build_option(dev, IORESOURCE_DMA, option_flags);
+	if (!option)
 		return -ENOMEM;
 
+	dma = &option->u.dma;
 	dma->map = map;
 	dma->flags = flags;
 
-	ptr = option->dma;
-	while (ptr && ptr->next)
-		ptr = ptr->next;
-	if (ptr)
-		ptr->next = dma;
-	else
-		option->dma = dma;
-
-	dev_dbg(&dev->dev, "  dma bitmask %#x flags %#x\n", dma->map,
-		dma->flags);
+	dbg_pnp_show_option(dev, option);
 	return 0;
 }
 
-int pnp_register_port_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_port_resource(struct pnp_dev *dev, unsigned int option_flags,
 			       resource_size_t min, resource_size_t max,
 			       resource_size_t align, resource_size_t size,
 			       unsigned char flags)
 {
-	struct pnp_port *port, *ptr;
+	struct pnp_option *option;
+	struct pnp_port *port;
 
-	port = kzalloc(sizeof(struct pnp_port), GFP_KERNEL);
-	if (!port)
+	option = pnp_build_option(dev, IORESOURCE_IO, option_flags);
+	if (!option)
 		return -ENOMEM;
 
+	port = &option->u.port;
 	port->min = min;
 	port->max = max;
 	port->align = align;
 	port->size = size;
 	port->flags = flags;
 
-	ptr = option->port;
-	while (ptr && ptr->next)
-		ptr = ptr->next;
-	if (ptr)
-		ptr->next = port;
-	else
-		option->port = port;
-
-	dev_dbg(&dev->dev, "  io  "
-		"min %#llx max %#llx align %lld size %lld flags %#x\n",
-		(unsigned long long) port->min,
-		(unsigned long long) port->max,
-		(unsigned long long) port->align,
-		(unsigned long long) port->size, port->flags);
+	dbg_pnp_show_option(dev, option);
 	return 0;
 }
 
-int pnp_register_mem_resource(struct pnp_dev *dev, struct pnp_option *option,
+int pnp_register_mem_resource(struct pnp_dev *dev, unsigned int option_flags,
 			      resource_size_t min, resource_size_t max,
 			      resource_size_t align, resource_size_t size,
 			      unsigned char flags)
 {
-	struct pnp_mem *mem, *ptr;
+	struct pnp_option *option;
+	struct pnp_mem *mem;
 
-	mem = kzalloc(sizeof(struct pnp_mem), GFP_KERNEL);
-	if (!mem)
+	option = pnp_build_option(dev, IORESOURCE_MEM, option_flags);
+	if (!option)
 		return -ENOMEM;
 
+	mem = &option->u.mem;
 	mem->min = min;
 	mem->max = max;
 	mem->align = align;
 	mem->size = size;
 	mem->flags = flags;
 
-	ptr = option->mem;
-	while (ptr && ptr->next)
-		ptr = ptr->next;
-	if (ptr)
-		ptr->next = mem;
-	else
-		option->mem = mem;
-
-	dev_dbg(&dev->dev, "  mem "
-		"min %#llx max %#llx align %lld size %lld flags %#x\n",
-		(unsigned long long) mem->min,
-		(unsigned long long) mem->max,
-		(unsigned long long) mem->align,
-		(unsigned long long) mem->size, mem->flags);
+	dbg_pnp_show_option(dev, option);
 	return 0;
 }
 
-static void pnp_free_port(struct pnp_port *port)
-{
-	struct pnp_port *next;
-
-	while (port) {
-		next = port->next;
-		kfree(port);
-		port = next;
-	}
-}
-
-static void pnp_free_irq(struct pnp_irq *irq)
-{
-	struct pnp_irq *next;
-
-	while (irq) {
-		next = irq->next;
-		kfree(irq);
-		irq = next;
-	}
-}
-
-static void pnp_free_dma(struct pnp_dma *dma)
-{
-	struct pnp_dma *next;
-
-	while (dma) {
-		next = dma->next;
-		kfree(dma);
-		dma = next;
-	}
-}
-
-static void pnp_free_mem(struct pnp_mem *mem)
+void pnp_free_options(struct pnp_dev *dev)
 {
-	struct pnp_mem *next;
-
-	while (mem) {
-		next = mem->next;
-		kfree(mem);
-		mem = next;
-	}
-}
+	struct pnp_option *option, *tmp;
 
-void pnp_free_option(struct pnp_option *option)
-{
-	struct pnp_option *next;
-
-	while (option) {
-		next = option->next;
-		pnp_free_port(option->port);
-		pnp_free_irq(option->irq);
-		pnp_free_dma(option->dma);
-		pnp_free_mem(option->mem);
+	list_for_each_entry_safe(option, tmp, &dev->options, list) {
+		list_del(&option->list);
 		kfree(option);
-		option = next;
 	}
 }
 
@@ -668,66 +546,50 @@ struct pnp_resource *pnp_add_mem_resource(struct pnp_dev *dev,
 	return pnp_res;
 }
 
-static int pnp_possible_option(struct pnp_option *option, int type,
-			       resource_size_t start, resource_size_t size)
+/*
+ * Determine whether the specified resource is a possible configuration
+ * for this device.
+ */
+int pnp_possible_config(struct pnp_dev *dev, int type, resource_size_t start,
+			resource_size_t size)
 {
-	struct pnp_option *tmp;
+	struct pnp_option *option;
 	struct pnp_port *port;
 	struct pnp_mem *mem;
 	struct pnp_irq *irq;
 	struct pnp_dma *dma;
 
-	if (!option)
-		return 0;
+	list_for_each_entry(option, &dev->options, list) {
+		if (option->type != type)
+			continue;
 
-	for (tmp = option; tmp; tmp = tmp->next) {
-		switch (type) {
+		switch (option->type) {
 		case IORESOURCE_IO:
-			for (port = tmp->port; port; port = port->next) {
-				if (port->min == start && port->size == size)
-					return 1;
-			}
+			port = &option->u.port;
+			if (port->min == start && port->size == size)
+				return 1;
 			break;
 		case IORESOURCE_MEM:
-			for (mem = tmp->mem; mem; mem = mem->next) {
-				if (mem->min == start && mem->size == size)
-					return 1;
-			}
+			mem = &option->u.mem;
+			if (mem->min == start && mem->size == size)
+				return 1;
 			break;
 		case IORESOURCE_IRQ:
-			for (irq = tmp->irq; irq; irq = irq->next) {
-				if (start < PNP_IRQ_NR &&
-				    test_bit(start, irq->map.bits))
-					return 1;
-			}
+			irq = &option->u.irq;
+			if (start < PNP_IRQ_NR &&
+			    test_bit(start, irq->map.bits))
+				return 1;
 			break;
 		case IORESOURCE_DMA:
-			for (dma = tmp->dma; dma; dma = dma->next) {
-				if (dma->map & (1 << start))
-					return 1;
-			}
+			dma = &option->u.dma;
+			if (dma->map & (1 << start))
+				return 1;
 			break;
 		}
 	}
 
 	return 0;
 }
-
-/*
- * Determine whether the specified resource is a possible configuration
- * for this device.
- */
-int pnp_possible_config(struct pnp_dev *dev, int type, resource_size_t start,
-			resource_size_t size)
-{
-	if (pnp_possible_option(dev->independent, type, start, size))
-		return 1;
-
-	if (pnp_possible_option(dev->dependent, type, start, size))
-		return 1;
-
-	return 0;
-}
 EXPORT_SYMBOL(pnp_possible_config);
 
 /* format is: pnp_reserve_irq=irq1[,irq2] .... */
diff --git a/drivers/pnp/support.c b/drivers/pnp/support.c
index 0ad42db94884..bbf78ef4ba02 100644
--- a/drivers/pnp/support.c
+++ b/drivers/pnp/support.c
@@ -2,6 +2,8 @@
  * support.c - standard functions for the use of pnp protocol drivers
  *
  * Copyright 2003 Adam Belay <ambx1@neo.rr.com>
+ * Copyright (C) 2008 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
  */
 
 #include <linux/module.h>
@@ -117,3 +119,93 @@ void dbg_pnp_show_resources(struct pnp_dev *dev, char *desc)
 	}
 #endif
 }
+
+char *pnp_option_priority_name(struct pnp_option *option)
+{
+	switch (pnp_option_priority(option)) {
+	case PNP_RES_PRIORITY_PREFERRED:
+		return "preferred";
+	case PNP_RES_PRIORITY_ACCEPTABLE:
+		return "acceptable";
+	case PNP_RES_PRIORITY_FUNCTIONAL:
+		return "functional";
+	}
+	return "invalid";
+}
+
+void dbg_pnp_show_option(struct pnp_dev *dev, struct pnp_option *option)
+{
+#ifdef DEBUG
+	char buf[128];
+	int len = 0, i;
+	struct pnp_port *port;
+	struct pnp_mem *mem;
+	struct pnp_irq *irq;
+	struct pnp_dma *dma;
+
+	if (pnp_option_is_dependent(option))
+		len += snprintf(buf + len, sizeof(buf) - len,
+				"  dependent set %d (%s) ",
+				pnp_option_set(option),
+				pnp_option_priority_name(option));
+	else
+		len += snprintf(buf + len, sizeof(buf) - len, "  independent ");
+
+	switch (option->type) {
+	case IORESOURCE_IO:
+		port = &option->u.port;
+		len += snprintf(buf + len, sizeof(buf) - len, "io  min %#llx "
+				"max %#llx align %lld size %lld flags %#x",
+				(unsigned long long) port->min,
+				(unsigned long long) port->max,
+				(unsigned long long) port->align,
+				(unsigned long long) port->size, port->flags);
+		break;
+	case IORESOURCE_MEM:
+		mem = &option->u.mem;
+		len += snprintf(buf + len, sizeof(buf) - len, "mem min %#llx "
+				"max %#llx align %lld size %lld flags %#x",
+				(unsigned long long) mem->min,
+				(unsigned long long) mem->max,
+				(unsigned long long) mem->align,
+				(unsigned long long) mem->size, mem->flags);
+		break;
+	case IORESOURCE_IRQ:
+		irq = &option->u.irq;
+		len += snprintf(buf + len, sizeof(buf) - len, "irq");
+		if (bitmap_empty(irq->map.bits, PNP_IRQ_NR))
+			len += snprintf(buf + len, sizeof(buf) - len,
+					" <none>");
+		else {
+			for (i = 0; i < PNP_IRQ_NR; i++)
+				if (test_bit(i, irq->map.bits))
+					len += snprintf(buf + len,
+							sizeof(buf) - len,
+							" %d", i);
+		}
+		len += snprintf(buf + len, sizeof(buf) - len, " flags %#x",
+				irq->flags);
+		if (irq->flags & IORESOURCE_IRQ_OPTIONAL)
+			len += snprintf(buf + len, sizeof(buf) - len,
+					" (optional)");
+		break;
+	case IORESOURCE_DMA:
+		dma = &option->u.dma;
+		len += snprintf(buf + len, sizeof(buf) - len, "dma");
+		if (!dma->map)
+			len += snprintf(buf + len, sizeof(buf) - len,
+					" <none>");
+		else {
+			for (i = 0; i < 8; i++)
+				if (dma->map & (1 << i))
+					len += snprintf(buf + len,
+							sizeof(buf) - len,
+							" %d", i);
+		}
+		len += snprintf(buf + len, sizeof(buf) - len, " (bitmask %#x) "
+				"flags %#x", dma->map, dma->flags);
+		break;
+	}
+	dev_dbg(&dev->dev, "%s\n", buf);
+#endif
+}
diff --git a/include/linux/pnp.h b/include/linux/pnp.h
index 785126ffcc11..1ce54b63085d 100644
--- a/include/linux/pnp.h
+++ b/include/linux/pnp.h
@@ -1,6 +1,8 @@
 /*
  * Linux Plug and Play Support
  * Copyright by Adam Belay <ambx1@neo.rr.com>
+ * Copyright (C) 2008 Hewlett-Packard Development Company, L.P.
+ *	Bjorn Helgaas <bjorn.helgaas@hp.com>
  */
 
 #ifndef _LINUX_PNP_H
@@ -249,9 +251,9 @@ struct pnp_dev {
 
 	int active;
 	int capabilities;
-	struct pnp_option *independent;
-	struct pnp_option *dependent;
+	unsigned int num_dependent_sets;
 	struct list_head resources;
+	struct list_head options;
 
 	char name[PNP_NAME_LEN];	/* contains a human-readable name */
 	int flags;			/* used by protocols */
-- 
cgit v1.2.3


From d442cc44c0db56e84ef6aa244a88427d2efe06cd Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Wed, 16 Jul 2008 16:09:06 -0400
Subject: block: Trivial fix for blk_integrity_rq()

Fail integrity check gracefully when request does not have a bio
attached (BLOCK_PC).

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/blkdev.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 32a441b05fd5..88d68081a0f1 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -985,6 +985,9 @@ static inline int bdev_integrity_enabled(struct block_device *bdev, int rw)
 
 static inline int blk_integrity_rq(struct request *rq)
 {
+	if (rq->bio == NULL)
+		return 0;
+
 	return bio_integrity(rq->bio);
 }
 
-- 
cgit v1.2.3


From d49747bdfb2ddebea24d1580da55b79d093d48a9 Mon Sep 17 00:00:00 2001
From: Scott Wood <scottwood@freescale.com>
Date: Tue, 9 Oct 2007 12:37:13 -0500
Subject: powerpc/mpc83xx: Power Management support

Basic PM support for 83xx.  Standby is implemented as sleep.
Suspend-to-RAM is implemented as "deep sleep" (with the processor
turned off) on 831x.

Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/Kconfig                      |   2 +-
 arch/powerpc/platforms/83xx/Makefile      |   1 +
 arch/powerpc/platforms/83xx/suspend-asm.S | 533 ++++++++++++++++++++++++++++++
 arch/powerpc/platforms/83xx/suspend.c     | 388 ++++++++++++++++++++++
 arch/powerpc/sysdev/fsl_soc.h             |   1 +
 arch/powerpc/sysdev/ipic.c                |  71 ++++
 include/asm-powerpc/reg.h                 |   4 +
 include/linux/fsl_devices.h               |   6 +
 8 files changed, 1005 insertions(+), 1 deletion(-)
 create mode 100644 arch/powerpc/platforms/83xx/suspend-asm.S
 create mode 100644 arch/powerpc/platforms/83xx/suspend.c

(limited to 'include/linux')

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index de88972c5896..50be303e525d 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -199,7 +199,7 @@ config ARCH_HIBERNATION_POSSIBLE
 
 config ARCH_SUSPEND_POSSIBLE
 	def_bool y
-	depends on ADB_PMU || PPC_EFIKA || PPC_LITE5200
+	depends on ADB_PMU || PPC_EFIKA || PPC_LITE5200 || PPC_83xx
 
 config PPC_DCR_NATIVE
 	bool
diff --git a/arch/powerpc/platforms/83xx/Makefile b/arch/powerpc/platforms/83xx/Makefile
index f331fd7dd836..32c7ad13911c 100644
--- a/arch/powerpc/platforms/83xx/Makefile
+++ b/arch/powerpc/platforms/83xx/Makefile
@@ -3,6 +3,7 @@
 #
 obj-y				:= misc.o usb.o
 obj-$(CONFIG_PCI)		+= pci.o
+obj-$(CONFIG_SUSPEND)		+= suspend.o suspend-asm.o
 obj-$(CONFIG_MPC831x_RDB)	+= mpc831x_rdb.o
 obj-$(CONFIG_MPC832x_RDB)	+= mpc832x_rdb.o
 obj-$(CONFIG_MPC834x_MDS)	+= mpc834x_mds.o
diff --git a/arch/powerpc/platforms/83xx/suspend-asm.S b/arch/powerpc/platforms/83xx/suspend-asm.S
new file mode 100644
index 000000000000..1930543c98d3
--- /dev/null
+++ b/arch/powerpc/platforms/83xx/suspend-asm.S
@@ -0,0 +1,533 @@
+/*
+ * Enter and leave deep sleep state on MPC83xx
+ *
+ * Copyright (c) 2006-2008 Freescale Semiconductor, Inc.
+ * Author: Scott Wood <scottwood@freescale.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ */
+
+#include <asm/page.h>
+#include <asm/ppc_asm.h>
+#include <asm/reg.h>
+#include <asm/asm-offsets.h>
+
+#define SS_MEMSAVE	0x00 /* First 8 bytes of RAM */
+#define SS_HID		0x08 /* 3 HIDs */
+#define SS_IABR		0x14 /* 2 IABRs */
+#define SS_IBCR		0x1c
+#define SS_DABR		0x20 /* 2 DABRs */
+#define SS_DBCR		0x28
+#define SS_SP		0x2c
+#define SS_SR		0x30 /* 16 segment registers */
+#define SS_R2		0x70
+#define SS_MSR		0x74
+#define SS_SDR1		0x78
+#define SS_LR		0x7c
+#define SS_SPRG		0x80 /* 4 SPRGs */
+#define SS_DBAT		0x90 /* 8 DBATs */
+#define SS_IBAT		0xd0 /* 8 IBATs */
+#define SS_TB		0x110
+#define SS_CR		0x118
+#define SS_GPREG	0x11c /* r12-r31 */
+#define STATE_SAVE_SIZE 0x16c
+
+	.section .data
+	.align	5
+
+mpc83xx_sleep_save_area:
+	.space	STATE_SAVE_SIZE
+immrbase:
+	.long	0
+
+	.section .text
+	.align	5
+
+	/* r3 = physical address of IMMR */
+_GLOBAL(mpc83xx_enter_deep_sleep)
+	lis	r4, immrbase@ha
+	stw	r3, immrbase@l(r4)
+
+	/* The first 2 words of memory are used to communicate with the
+	 * bootloader, to tell it how to resume.
+	 *
+	 * The first word is the magic number 0xf5153ae5, and the second
+	 * is the pointer to mpc83xx_deep_resume.
+	 *
+	 * The original content of these two words is saved in SS_MEMSAVE.
+	 */
+
+	lis	r3, mpc83xx_sleep_save_area@h
+	ori	r3, r3, mpc83xx_sleep_save_area@l
+
+	lis	r4, KERNELBASE@h
+	lwz	r5, 0(r4)
+	lwz	r6, 4(r4)
+
+	stw	r5, SS_MEMSAVE+0(r3)
+	stw	r6, SS_MEMSAVE+4(r3)
+
+	mfspr	r5, SPRN_HID0
+	mfspr	r6, SPRN_HID1
+	mfspr	r7, SPRN_HID2
+
+	stw	r5, SS_HID+0(r3)
+	stw	r6, SS_HID+4(r3)
+	stw	r7, SS_HID+8(r3)
+
+	mfspr	r4, SPRN_IABR
+	mfspr	r5, SPRN_IABR2
+	mfspr	r6, SPRN_IBCR
+	mfspr	r7, SPRN_DABR
+	mfspr	r8, SPRN_DABR2
+	mfspr	r9, SPRN_DBCR
+
+	stw	r4, SS_IABR+0(r3)
+	stw	r5, SS_IABR+4(r3)
+	stw	r6, SS_IBCR(r3)
+	stw	r7, SS_DABR+0(r3)
+	stw	r8, SS_DABR+4(r3)
+	stw	r9, SS_DBCR(r3)
+
+	mfspr	r4, SPRN_SPRG0
+	mfspr	r5, SPRN_SPRG1
+	mfspr	r6, SPRN_SPRG2
+	mfspr	r7, SPRN_SPRG3
+	mfsdr1	r8
+
+	stw	r4, SS_SPRG+0(r3)
+	stw	r5, SS_SPRG+4(r3)
+	stw	r6, SS_SPRG+8(r3)
+	stw	r7, SS_SPRG+12(r3)
+	stw	r8, SS_SDR1(r3)
+
+	mfspr	r4, SPRN_DBAT0U
+	mfspr	r5, SPRN_DBAT0L
+	mfspr	r6, SPRN_DBAT1U
+	mfspr	r7, SPRN_DBAT1L
+
+	stw	r4, SS_DBAT+0x00(r3)
+	stw	r5, SS_DBAT+0x04(r3)
+	stw	r6, SS_DBAT+0x08(r3)
+	stw	r7, SS_DBAT+0x0c(r3)
+
+	mfspr	r4, SPRN_DBAT2U
+	mfspr	r5, SPRN_DBAT2L
+	mfspr	r6, SPRN_DBAT3U
+	mfspr	r7, SPRN_DBAT3L
+
+	stw	r4, SS_DBAT+0x10(r3)
+	stw	r5, SS_DBAT+0x14(r3)
+	stw	r6, SS_DBAT+0x18(r3)
+	stw	r7, SS_DBAT+0x1c(r3)
+
+	mfspr	r4, SPRN_DBAT4U
+	mfspr	r5, SPRN_DBAT4L
+	mfspr	r6, SPRN_DBAT5U
+	mfspr	r7, SPRN_DBAT5L
+
+	stw	r4, SS_DBAT+0x20(r3)
+	stw	r5, SS_DBAT+0x24(r3)
+	stw	r6, SS_DBAT+0x28(r3)
+	stw	r7, SS_DBAT+0x2c(r3)
+
+	mfspr	r4, SPRN_DBAT6U
+	mfspr	r5, SPRN_DBAT6L
+	mfspr	r6, SPRN_DBAT7U
+	mfspr	r7, SPRN_DBAT7L
+
+	stw	r4, SS_DBAT+0x30(r3)
+	stw	r5, SS_DBAT+0x34(r3)
+	stw	r6, SS_DBAT+0x38(r3)
+	stw	r7, SS_DBAT+0x3c(r3)
+
+	mfspr	r4, SPRN_IBAT0U
+	mfspr	r5, SPRN_IBAT0L
+	mfspr	r6, SPRN_IBAT1U
+	mfspr	r7, SPRN_IBAT1L
+
+	stw	r4, SS_IBAT+0x00(r3)
+	stw	r5, SS_IBAT+0x04(r3)
+	stw	r6, SS_IBAT+0x08(r3)
+	stw	r7, SS_IBAT+0x0c(r3)
+
+	mfspr	r4, SPRN_IBAT2U
+	mfspr	r5, SPRN_IBAT2L
+	mfspr	r6, SPRN_IBAT3U
+	mfspr	r7, SPRN_IBAT3L
+
+	stw	r4, SS_IBAT+0x10(r3)
+	stw	r5, SS_IBAT+0x14(r3)
+	stw	r6, SS_IBAT+0x18(r3)
+	stw	r7, SS_IBAT+0x1c(r3)
+
+	mfspr	r4, SPRN_IBAT4U
+	mfspr	r5, SPRN_IBAT4L
+	mfspr	r6, SPRN_IBAT5U
+	mfspr	r7, SPRN_IBAT5L
+
+	stw	r4, SS_IBAT+0x20(r3)
+	stw	r5, SS_IBAT+0x24(r3)
+	stw	r6, SS_IBAT+0x28(r3)
+	stw	r7, SS_IBAT+0x2c(r3)
+
+	mfspr	r4, SPRN_IBAT6U
+	mfspr	r5, SPRN_IBAT6L
+	mfspr	r6, SPRN_IBAT7U
+	mfspr	r7, SPRN_IBAT7L
+
+	stw	r4, SS_IBAT+0x30(r3)
+	stw	r5, SS_IBAT+0x34(r3)
+	stw	r6, SS_IBAT+0x38(r3)
+	stw	r7, SS_IBAT+0x3c(r3)
+
+	mfmsr	r4
+	mflr	r5
+	mfcr	r6
+
+	stw	r4, SS_MSR(r3)
+	stw	r5, SS_LR(r3)
+	stw	r6, SS_CR(r3)
+	stw	r1, SS_SP(r3)
+	stw	r2, SS_R2(r3)
+
+1:	mftbu	r4
+	mftb	r5
+	mftbu	r6
+	cmpw	r4, r6
+	bne	1b
+
+	stw	r4, SS_TB+0(r3)
+	stw	r5, SS_TB+4(r3)
+
+	stmw	r12, SS_GPREG(r3)
+
+	li	r4, 0
+	addi	r6, r3, SS_SR-4
+1:	mfsrin	r5, r4
+	stwu	r5, 4(r6)
+	addis	r4, r4, 0x1000
+	cmpwi	r4, 0
+	bne	1b
+
+	/* Disable machine checks and critical exceptions */
+	mfmsr	r4
+	rlwinm	r4, r4, 0, ~MSR_CE
+	rlwinm	r4, r4, 0, ~MSR_ME
+	mtmsr	r4
+	isync
+
+#define TMP_VIRT_IMMR		0xf0000000
+#define DEFAULT_IMMR_VALUE	0xff400000
+#define IMMRBAR_BASE		0x0000
+
+	lis	r4, immrbase@ha
+	lwz	r4, immrbase@l(r4)
+
+	/* Use DBAT0 to address the current IMMR space */
+
+	ori	r4, r4, 0x002a
+	mtspr	SPRN_DBAT0L, r4
+	lis	r8, TMP_VIRT_IMMR@h
+	ori	r4, r8, 0x001e	/* 1 MByte accessable from Kernel Space only */
+	mtspr	SPRN_DBAT0U, r4
+	isync
+
+	/* Use DBAT1 to address the original IMMR space */
+
+	lis	r4, DEFAULT_IMMR_VALUE@h
+	ori	r4, r4, 0x002a
+	mtspr	SPRN_DBAT1L, r4
+	lis	r9, (TMP_VIRT_IMMR + 0x01000000)@h
+	ori	r4, r9, 0x001e	/* 1 MByte accessable from Kernel Space only */
+	mtspr	SPRN_DBAT1U, r4
+	isync
+
+	/* Use DBAT2 to address the beginning of RAM.  This isn't done
+	 * using the normal virtual mapping, because with page debugging
+	 * enabled it will be read-only.
+	 */
+
+	li	r4, 0x0002
+	mtspr	SPRN_DBAT2L, r4
+	lis	r4, KERNELBASE@h
+	ori	r4, r4, 0x001e	/* 1 MByte accessable from Kernel Space only */
+	mtspr	SPRN_DBAT2U, r4
+	isync
+
+	/* Flush the cache with our BAT, as there will be TLB misses
+	 * otherwise if page debugging is enabled, and these misses
+	 * will disturb the PLRU algorithm.
+	 */
+
+	bl	__flush_disable_L1
+
+	/* Keep the i-cache enabled, so the hack below for low-boot
+	 * flash will work.
+	 */
+	mfspr	r3, SPRN_HID0
+	ori	r3, r3, HID0_ICE
+	mtspr	SPRN_HID0, r3
+	isync
+
+	lis	r6, 0xf515
+	ori	r6, r6, 0x3ae5
+
+	lis	r7, mpc83xx_deep_resume@h
+	ori	r7, r7, mpc83xx_deep_resume@l
+	tophys(r7, r7)
+
+	lis	r5, KERNELBASE@h
+	stw	r6, 0(r5)
+	stw	r7, 4(r5)
+
+	/* Reset BARs */
+
+	li	r4, 0
+	stw	r4, 0x0024(r8)
+	stw	r4, 0x002c(r8)
+	stw	r4, 0x0034(r8)
+	stw	r4, 0x003c(r8)
+	stw	r4, 0x0064(r8)
+	stw	r4, 0x006c(r8)
+
+	/* Rev 1 of the 8313 has problems with wakeup events that are
+	 * pending during the transition to deep sleep state (such as if
+	 * the PCI host sets the state to D3 and then D0 in rapid
+	 * succession).  This check shrinks the race window somewhat.
+	 *
+	 * See erratum PCI23, though the problem is not limited
+	 * to PCI.
+	 */
+
+	lwz	r3, 0x0b04(r8)
+	andi.	r3, r3, 1
+	bne-	mpc83xx_deep_resume
+
+	/* Move IMMR back to the default location, following the
+	 * procedure specified in the MPC8313 manual.
+	 */
+	lwz	r4, IMMRBAR_BASE(r8)
+	isync
+	lis	r4, DEFAULT_IMMR_VALUE@h
+	stw	r4, IMMRBAR_BASE(r8)
+	lis	r4, KERNELBASE@h
+	lwz	r4, 0(r4)
+	isync
+	lwz	r4, IMMRBAR_BASE(r9)
+	mr	r8, r9
+	isync
+
+	/* Check the Reset Configuration Word to see whether flash needs
+	 * to be mapped at a low address or a high address.
+	 */
+
+	lwz	r4, 0x0904(r8)
+	andis.	r4, r4, 0x0400
+	li	r4, 0
+	beq	boot_low
+	lis	r4, 0xff80
+boot_low:
+	stw	r4, 0x0020(r8)
+	lis	r7, 0x8000
+	ori	r7, r7, 0x0016
+
+	mfspr	r5, SPRN_HID0
+	rlwinm	r5, r5, 0, ~(HID0_DOZE | HID0_NAP)
+	oris	r5, r5, HID0_SLEEP@h
+	mtspr	SPRN_HID0, r5
+	isync
+
+	mfmsr	r5
+	oris	r5, r5, MSR_POW@h
+
+	/* Enable the flash mapping at the appropriate address.  This
+	 * mapping will override the RAM mapping if booting low, so there's
+	 * no need to disable the latter.  This must be done inside the same
+	 * cache line as setting MSR_POW, so that no instruction fetches
+	 * from RAM happen after the flash mapping is turned on.
+	 */
+
+	.align	5
+	stw	r7, 0x0024(r8)
+	sync
+	isync
+	mtmsr	r5
+	isync
+1:	b	1b
+
+mpc83xx_deep_resume:
+	lis	r4, 1f@h
+	ori	r4, r4, 1f@l
+	tophys(r4, r4)
+	mtsrr0	r4
+
+	mfmsr	r4
+	rlwinm	r4, r4, 0, ~(MSR_IR | MSR_DR)
+	mtsrr1	r4
+
+	rfi
+
+1:	tlbia
+	bl	__inval_enable_L1
+
+	lis	r3, mpc83xx_sleep_save_area@h
+	ori	r3, r3, mpc83xx_sleep_save_area@l
+	tophys(r3, r3)
+
+	lwz	r5, SS_MEMSAVE+0(r3)
+	lwz	r6, SS_MEMSAVE+4(r3)
+
+	stw	r5, 0(0)
+	stw	r6, 4(0)
+
+	lwz	r5, SS_HID+0(r3)
+	lwz	r6, SS_HID+4(r3)
+	lwz	r7, SS_HID+8(r3)
+
+	mtspr	SPRN_HID0, r5
+	mtspr	SPRN_HID1, r6
+	mtspr	SPRN_HID2, r7
+
+	lwz	r4, SS_IABR+0(r3)
+	lwz	r5, SS_IABR+4(r3)
+	lwz	r6, SS_IBCR(r3)
+	lwz	r7, SS_DABR+0(r3)
+	lwz	r8, SS_DABR+4(r3)
+	lwz	r9, SS_DBCR(r3)
+
+	mtspr	SPRN_IABR, r4
+	mtspr	SPRN_IABR2, r5
+	mtspr	SPRN_IBCR, r6
+	mtspr	SPRN_DABR, r7
+	mtspr	SPRN_DABR2, r8
+	mtspr	SPRN_DBCR, r9
+
+	li	r4, 0
+	addi	r6, r3, SS_SR-4
+1:	lwzu	r5, 4(r6)
+	mtsrin	r5, r4
+	addis	r4, r4, 0x1000
+	cmpwi	r4, 0
+	bne	1b
+
+	lwz	r4, SS_DBAT+0x00(r3)
+	lwz	r5, SS_DBAT+0x04(r3)
+	lwz	r6, SS_DBAT+0x08(r3)
+	lwz	r7, SS_DBAT+0x0c(r3)
+
+	mtspr	SPRN_DBAT0U, r4
+	mtspr	SPRN_DBAT0L, r5
+	mtspr	SPRN_DBAT1U, r6
+	mtspr	SPRN_DBAT1L, r7
+
+	lwz	r4, SS_DBAT+0x10(r3)
+	lwz	r5, SS_DBAT+0x14(r3)
+	lwz	r6, SS_DBAT+0x18(r3)
+	lwz	r7, SS_DBAT+0x1c(r3)
+
+	mtspr	SPRN_DBAT2U, r4
+	mtspr	SPRN_DBAT2L, r5
+	mtspr	SPRN_DBAT3U, r6
+	mtspr	SPRN_DBAT3L, r7
+
+	lwz	r4, SS_DBAT+0x20(r3)
+	lwz	r5, SS_DBAT+0x24(r3)
+	lwz	r6, SS_DBAT+0x28(r3)
+	lwz	r7, SS_DBAT+0x2c(r3)
+
+	mtspr	SPRN_DBAT4U, r4
+	mtspr	SPRN_DBAT4L, r5
+	mtspr	SPRN_DBAT5U, r6
+	mtspr	SPRN_DBAT5L, r7
+
+	lwz	r4, SS_DBAT+0x30(r3)
+	lwz	r5, SS_DBAT+0x34(r3)
+	lwz	r6, SS_DBAT+0x38(r3)
+	lwz	r7, SS_DBAT+0x3c(r3)
+
+	mtspr	SPRN_DBAT6U, r4
+	mtspr	SPRN_DBAT6L, r5
+	mtspr	SPRN_DBAT7U, r6
+	mtspr	SPRN_DBAT7L, r7
+
+	lwz	r4, SS_IBAT+0x00(r3)
+	lwz	r5, SS_IBAT+0x04(r3)
+	lwz	r6, SS_IBAT+0x08(r3)
+	lwz	r7, SS_IBAT+0x0c(r3)
+
+	mtspr	SPRN_IBAT0U, r4
+	mtspr	SPRN_IBAT0L, r5
+	mtspr	SPRN_IBAT1U, r6
+	mtspr	SPRN_IBAT1L, r7
+
+	lwz	r4, SS_IBAT+0x10(r3)
+	lwz	r5, SS_IBAT+0x14(r3)
+	lwz	r6, SS_IBAT+0x18(r3)
+	lwz	r7, SS_IBAT+0x1c(r3)
+
+	mtspr	SPRN_IBAT2U, r4
+	mtspr	SPRN_IBAT2L, r5
+	mtspr	SPRN_IBAT3U, r6
+	mtspr	SPRN_IBAT3L, r7
+
+	lwz	r4, SS_IBAT+0x20(r3)
+	lwz	r5, SS_IBAT+0x24(r3)
+	lwz	r6, SS_IBAT+0x28(r3)
+	lwz	r7, SS_IBAT+0x2c(r3)
+
+	mtspr	SPRN_IBAT4U, r4
+	mtspr	SPRN_IBAT4L, r5
+	mtspr	SPRN_IBAT5U, r6
+	mtspr	SPRN_IBAT5L, r7
+
+	lwz	r4, SS_IBAT+0x30(r3)
+	lwz	r5, SS_IBAT+0x34(r3)
+	lwz	r6, SS_IBAT+0x38(r3)
+	lwz	r7, SS_IBAT+0x3c(r3)
+
+	mtspr	SPRN_IBAT6U, r4
+	mtspr	SPRN_IBAT6L, r5
+	mtspr	SPRN_IBAT7U, r6
+	mtspr	SPRN_IBAT7L, r7
+
+	lwz	r4, SS_SPRG+0(r3)
+	lwz	r5, SS_SPRG+4(r3)
+	lwz	r6, SS_SPRG+8(r3)
+	lwz	r7, SS_SPRG+12(r3)
+	lwz	r8, SS_SDR1(r3)
+
+	mtspr	SPRN_SPRG0, r4
+	mtspr	SPRN_SPRG1, r5
+	mtspr	SPRN_SPRG2, r6
+	mtspr	SPRN_SPRG3, r7
+	mtsdr1	r8
+
+	lwz	r4, SS_MSR(r3)
+	lwz	r5, SS_LR(r3)
+	lwz	r6, SS_CR(r3)
+	lwz	r1, SS_SP(r3)
+	lwz	r2, SS_R2(r3)
+
+	mtsrr1	r4
+	mtsrr0	r5
+	mtcr	r6
+
+	li	r4, 0
+	mtspr	SPRN_TBWL, r4
+
+	lwz	r4, SS_TB+0(r3)
+	lwz	r5, SS_TB+4(r3)
+
+	mtspr	SPRN_TBWU, r4
+	mtspr	SPRN_TBWL, r5
+
+	lmw	r12, SS_GPREG(r3)
+
+	/* Kick decrementer */
+	li	r0, 1
+	mtdec	r0
+
+	rfi
diff --git a/arch/powerpc/platforms/83xx/suspend.c b/arch/powerpc/platforms/83xx/suspend.c
new file mode 100644
index 000000000000..08e65fc8b98c
--- /dev/null
+++ b/arch/powerpc/platforms/83xx/suspend.c
@@ -0,0 +1,388 @@
+/*
+ * MPC83xx suspend support
+ *
+ * Author: Scott Wood <scottwood@freescale.com>
+ *
+ * Copyright (c) 2006-2007 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/pm.h>
+#include <linux/types.h>
+#include <linux/ioport.h>
+#include <linux/interrupt.h>
+#include <linux/wait.h>
+#include <linux/kthread.h>
+#include <linux/freezer.h>
+#include <linux/suspend.h>
+#include <linux/fsl_devices.h>
+#include <linux/of_platform.h>
+
+#include <asm/reg.h>
+#include <asm/io.h>
+#include <asm/time.h>
+#include <asm/mpc6xx.h>
+
+#include <sysdev/fsl_soc.h>
+
+#define PMCCR1_NEXT_STATE       0x0C /* Next state for power management */
+#define PMCCR1_NEXT_STATE_SHIFT 2
+#define PMCCR1_CURR_STATE       0x03 /* Current state for power management*/
+#define IMMR_RCW_OFFSET         0x900
+#define RCW_PCI_HOST            0x80000000
+
+void mpc83xx_enter_deep_sleep(phys_addr_t immrbase);
+
+struct mpc83xx_pmc {
+	u32 config;
+#define PMCCR_DLPEN 2 /* DDR SDRAM low power enable */
+#define PMCCR_SLPEN 1 /* System low power enable */
+
+	u32 event;
+	u32 mask;
+/* All but PMCI are deep-sleep only */
+#define PMCER_GPIO   0x100
+#define PMCER_PCI    0x080
+#define PMCER_USB    0x040
+#define PMCER_ETSEC1 0x020
+#define PMCER_ETSEC2 0x010
+#define PMCER_TIMER  0x008
+#define PMCER_INT1   0x004
+#define PMCER_INT2   0x002
+#define PMCER_PMCI   0x001
+#define PMCER_ALL    0x1FF
+
+	/* deep-sleep only */
+	u32 config1;
+#define PMCCR1_USE_STATE  0x80000000
+#define PMCCR1_PME_EN     0x00000080
+#define PMCCR1_ASSERT_PME 0x00000040
+#define PMCCR1_POWER_OFF  0x00000020
+
+	/* deep-sleep only */
+	u32 config2;
+};
+
+struct mpc83xx_rcw {
+	u32 rcwlr;
+	u32 rcwhr;
+};
+
+struct mpc83xx_clock {
+	u32 spmr;
+	u32 occr;
+	u32 sccr;
+};
+
+struct pmc_type {
+	int has_deep_sleep;
+};
+
+static struct of_device *pmc_dev;
+static int has_deep_sleep, deep_sleeping;
+static int pmc_irq;
+static struct mpc83xx_pmc __iomem *pmc_regs;
+static struct mpc83xx_clock __iomem *clock_regs;
+static int is_pci_agent, wake_from_pci;
+static phys_addr_t immrbase;
+static int pci_pm_state;
+static DECLARE_WAIT_QUEUE_HEAD(agent_wq);
+
+int fsl_deep_sleep(void)
+{
+	return deep_sleeping;
+}
+
+static int mpc83xx_change_state(void)
+{
+	u32 curr_state;
+	u32 reg_cfg1 = in_be32(&pmc_regs->config1);
+
+	if (is_pci_agent) {
+		pci_pm_state = (reg_cfg1 & PMCCR1_NEXT_STATE) >>
+		               PMCCR1_NEXT_STATE_SHIFT;
+		curr_state = reg_cfg1 & PMCCR1_CURR_STATE;
+
+		if (curr_state != pci_pm_state) {
+			reg_cfg1 &= ~PMCCR1_CURR_STATE;
+			reg_cfg1 |= pci_pm_state;
+			out_be32(&pmc_regs->config1, reg_cfg1);
+
+			wake_up(&agent_wq);
+			return 1;
+		}
+	}
+
+	return 0;
+}
+
+static irqreturn_t pmc_irq_handler(int irq, void *dev_id)
+{
+	u32 event = in_be32(&pmc_regs->event);
+	int ret = IRQ_NONE;
+
+	if (mpc83xx_change_state())
+		ret = IRQ_HANDLED;
+
+	if (event) {
+		out_be32(&pmc_regs->event, event);
+		ret = IRQ_HANDLED;
+	}
+
+	return ret;
+}
+
+static int mpc83xx_suspend_enter(suspend_state_t state)
+{
+	int ret = -EAGAIN;
+
+	/* Don't go to sleep if there's a race where pci_pm_state changes
+	 * between the agent thread checking it and the PM code disabling
+	 * interrupts.
+	 */
+	if (wake_from_pci) {
+		if (pci_pm_state != (deep_sleeping ? 3 : 2))
+			goto out;
+
+		out_be32(&pmc_regs->config1,
+		         in_be32(&pmc_regs->config1) | PMCCR1_PME_EN);
+	}
+
+	/* Put the system into low-power mode and the RAM
+	 * into self-refresh mode once the core goes to
+	 * sleep.
+	 */
+
+	out_be32(&pmc_regs->config, PMCCR_SLPEN | PMCCR_DLPEN);
+
+	/* If it has deep sleep (i.e. it's an 831x or compatible),
+	 * disable power to the core upon entering sleep mode.  This will
+	 * require going through the boot firmware upon a wakeup event.
+	 */
+
+	if (deep_sleeping) {
+		out_be32(&pmc_regs->mask, PMCER_ALL);
+
+		out_be32(&pmc_regs->config1,
+		         in_be32(&pmc_regs->config1) | PMCCR1_POWER_OFF);
+
+		enable_kernel_fp();
+
+		mpc83xx_enter_deep_sleep(immrbase);
+
+		out_be32(&pmc_regs->config1,
+		         in_be32(&pmc_regs->config1) & ~PMCCR1_POWER_OFF);
+
+		out_be32(&pmc_regs->mask, PMCER_PMCI);
+	} else {
+		out_be32(&pmc_regs->mask, PMCER_PMCI);
+
+		mpc6xx_enter_standby();
+	}
+
+	ret = 0;
+
+out:
+	out_be32(&pmc_regs->config1,
+	         in_be32(&pmc_regs->config1) & ~PMCCR1_PME_EN);
+
+	return ret;
+}
+
+static void mpc83xx_suspend_finish(void)
+{
+	deep_sleeping = 0;
+}
+
+static int mpc83xx_suspend_valid(suspend_state_t state)
+{
+	return state == PM_SUSPEND_STANDBY || state == PM_SUSPEND_MEM;
+}
+
+static int mpc83xx_suspend_begin(suspend_state_t state)
+{
+	switch (state) {
+		case PM_SUSPEND_STANDBY:
+			deep_sleeping = 0;
+			return 0;
+
+		case PM_SUSPEND_MEM:
+			if (has_deep_sleep)
+				deep_sleeping = 1;
+
+			return 0;
+
+		default:
+			return -EINVAL;
+	}
+}
+
+static int agent_thread_fn(void *data)
+{
+	while (1) {
+		wait_event_interruptible(agent_wq, pci_pm_state >= 2);
+		try_to_freeze();
+
+		if (signal_pending(current) || pci_pm_state < 2)
+			continue;
+
+		/* With a preemptible kernel (or SMP), this could race with
+		 * a userspace-driven suspend request.  It's probably best
+		 * to avoid mixing the two with such a configuration (or
+		 * else fix it by adding a mutex to state_store that we can
+		 * synchronize with).
+		 */
+
+		wake_from_pci = 1;
+
+		pm_suspend(pci_pm_state == 3 ? PM_SUSPEND_MEM :
+		                               PM_SUSPEND_STANDBY);
+
+		wake_from_pci = 0;
+	}
+
+	return 0;
+}
+
+static void mpc83xx_set_agent(void)
+{
+	out_be32(&pmc_regs->config1, PMCCR1_USE_STATE);
+	out_be32(&pmc_regs->mask, PMCER_PMCI);
+
+	kthread_run(agent_thread_fn, NULL, "PCI power mgt");
+}
+
+static int mpc83xx_is_pci_agent(void)
+{
+	struct mpc83xx_rcw __iomem *rcw_regs;
+	int ret;
+
+	rcw_regs = ioremap(get_immrbase() + IMMR_RCW_OFFSET,
+	                   sizeof(struct mpc83xx_rcw));
+
+	if (!rcw_regs)
+		return -ENOMEM;
+
+	ret = !(in_be32(&rcw_regs->rcwhr) & RCW_PCI_HOST);
+
+	iounmap(rcw_regs);
+	return ret;
+}
+
+static struct platform_suspend_ops mpc83xx_suspend_ops = {
+	.valid = mpc83xx_suspend_valid,
+	.begin = mpc83xx_suspend_begin,
+	.enter = mpc83xx_suspend_enter,
+	.finish = mpc83xx_suspend_finish,
+};
+
+static int pmc_probe(struct of_device *ofdev,
+                     const struct of_device_id *match)
+{
+	struct device_node *np = ofdev->node;
+	struct resource res;
+	struct pmc_type *type = match->data;
+	int ret = 0;
+
+	if (!of_device_is_available(np))
+		return -ENODEV;
+
+	has_deep_sleep = type->has_deep_sleep;
+	immrbase = get_immrbase();
+	pmc_dev = ofdev;
+
+	is_pci_agent = mpc83xx_is_pci_agent();
+	if (is_pci_agent < 0)
+		return is_pci_agent;
+
+	ret = of_address_to_resource(np, 0, &res);
+	if (ret)
+		return -ENODEV;
+
+	pmc_irq = irq_of_parse_and_map(np, 0);
+	if (pmc_irq != NO_IRQ) {
+		ret = request_irq(pmc_irq, pmc_irq_handler, IRQF_SHARED,
+		                  "pmc", ofdev);
+
+		if (ret)
+			return -EBUSY;
+	}
+
+	pmc_regs = ioremap(res.start, sizeof(struct mpc83xx_pmc));
+
+	if (!pmc_regs) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = of_address_to_resource(np, 1, &res);
+	if (ret) {
+		ret = -ENODEV;
+		goto out_pmc;
+	}
+
+	clock_regs = ioremap(res.start, sizeof(struct mpc83xx_pmc));
+
+	if (!clock_regs) {
+		ret = -ENOMEM;
+		goto out_pmc;
+	}
+
+	if (is_pci_agent)
+		mpc83xx_set_agent();
+
+	suspend_set_ops(&mpc83xx_suspend_ops);
+	return 0;
+
+out_pmc:
+	iounmap(pmc_regs);
+out:
+	if (pmc_irq != NO_IRQ)
+		free_irq(pmc_irq, ofdev);
+
+	return ret;
+}
+
+static int pmc_remove(struct of_device *ofdev)
+{
+	return -EPERM;
+};
+
+static struct pmc_type pmc_types[] = {
+	{
+		.has_deep_sleep = 1,
+	},
+	{
+		.has_deep_sleep = 0,
+	}
+};
+
+static struct of_device_id pmc_match[] = {
+	{
+		.compatible = "fsl,mpc8313-pmc",
+		.data = &pmc_types[0],
+	},
+	{
+		.compatible = "fsl,mpc8349-pmc",
+		.data = &pmc_types[1],
+	},
+	{}
+};
+
+static struct of_platform_driver pmc_driver = {
+	.name = "mpc83xx-pmc",
+	.match_table = pmc_match,
+	.probe = pmc_probe,
+	.remove = pmc_remove
+};
+
+static int pmc_init(void)
+{
+	return of_register_platform_driver(&pmc_driver);
+}
+
+module_init(pmc_init);
diff --git a/arch/powerpc/sysdev/fsl_soc.h b/arch/powerpc/sysdev/fsl_soc.h
index 52c831fa1886..024299887352 100644
--- a/arch/powerpc/sysdev/fsl_soc.h
+++ b/arch/powerpc/sysdev/fsl_soc.h
@@ -10,6 +10,7 @@ extern u32 get_baudrate(void);
 extern u32 fsl_get_sys_freq(void);
 
 struct spi_board_info;
+struct device_node;
 
 extern int fsl_spi_init(struct spi_board_info *board_infos,
 			unsigned int num_board_infos,
diff --git a/arch/powerpc/sysdev/ipic.c b/arch/powerpc/sysdev/ipic.c
index caba1c0be5a7..88a983ece5c9 100644
--- a/arch/powerpc/sysdev/ipic.c
+++ b/arch/powerpc/sysdev/ipic.c
@@ -22,6 +22,7 @@
 #include <linux/device.h>
 #include <linux/bootmem.h>
 #include <linux/spinlock.h>
+#include <linux/fsl_devices.h>
 #include <asm/irq.h>
 #include <asm/io.h>
 #include <asm/prom.h>
@@ -889,8 +890,78 @@ unsigned int ipic_get_irq(void)
 	return irq_linear_revmap(primary_ipic->irqhost, irq);
 }
 
+#ifdef CONFIG_PM
+static struct {
+	u32 sicfr;
+	u32 siprr[2];
+	u32 simsr[2];
+	u32 sicnr;
+	u32 smprr[2];
+	u32 semsr;
+	u32 secnr;
+	u32 sermr;
+	u32 sercr;
+} ipic_saved_state;
+
+static int ipic_suspend(struct sys_device *sdev, pm_message_t state)
+{
+	struct ipic *ipic = primary_ipic;
+
+	ipic_saved_state.sicfr = ipic_read(ipic->regs, IPIC_SICFR);
+	ipic_saved_state.siprr[0] = ipic_read(ipic->regs, IPIC_SIPRR_A);
+	ipic_saved_state.siprr[1] = ipic_read(ipic->regs, IPIC_SIPRR_D);
+	ipic_saved_state.simsr[0] = ipic_read(ipic->regs, IPIC_SIMSR_H);
+	ipic_saved_state.simsr[1] = ipic_read(ipic->regs, IPIC_SIMSR_L);
+	ipic_saved_state.sicnr = ipic_read(ipic->regs, IPIC_SICNR);
+	ipic_saved_state.smprr[0] = ipic_read(ipic->regs, IPIC_SMPRR_A);
+	ipic_saved_state.smprr[1] = ipic_read(ipic->regs, IPIC_SMPRR_B);
+	ipic_saved_state.semsr = ipic_read(ipic->regs, IPIC_SEMSR);
+	ipic_saved_state.secnr = ipic_read(ipic->regs, IPIC_SECNR);
+	ipic_saved_state.sermr = ipic_read(ipic->regs, IPIC_SERMR);
+	ipic_saved_state.sercr = ipic_read(ipic->regs, IPIC_SERCR);
+
+	if (fsl_deep_sleep()) {
+		/* In deep sleep, make sure there can be no
+		 * pending interrupts, as this can cause
+		 * problems on 831x.
+		 */
+		ipic_write(ipic->regs, IPIC_SIMSR_H, 0);
+		ipic_write(ipic->regs, IPIC_SIMSR_L, 0);
+		ipic_write(ipic->regs, IPIC_SEMSR, 0);
+		ipic_write(ipic->regs, IPIC_SERMR, 0);
+	}
+
+	return 0;
+}
+
+static int ipic_resume(struct sys_device *sdev)
+{
+	struct ipic *ipic = primary_ipic;
+
+	ipic_write(ipic->regs, IPIC_SICFR, ipic_saved_state.sicfr);
+	ipic_write(ipic->regs, IPIC_SIPRR_A, ipic_saved_state.siprr[0]);
+	ipic_write(ipic->regs, IPIC_SIPRR_D, ipic_saved_state.siprr[1]);
+	ipic_write(ipic->regs, IPIC_SIMSR_H, ipic_saved_state.simsr[0]);
+	ipic_write(ipic->regs, IPIC_SIMSR_L, ipic_saved_state.simsr[1]);
+	ipic_write(ipic->regs, IPIC_SICNR, ipic_saved_state.sicnr);
+	ipic_write(ipic->regs, IPIC_SMPRR_A, ipic_saved_state.smprr[0]);
+	ipic_write(ipic->regs, IPIC_SMPRR_B, ipic_saved_state.smprr[1]);
+	ipic_write(ipic->regs, IPIC_SEMSR, ipic_saved_state.semsr);
+	ipic_write(ipic->regs, IPIC_SECNR, ipic_saved_state.secnr);
+	ipic_write(ipic->regs, IPIC_SERMR, ipic_saved_state.sermr);
+	ipic_write(ipic->regs, IPIC_SERCR, ipic_saved_state.sercr);
+
+	return 0;
+}
+#else
+#define ipic_suspend NULL
+#define ipic_resume NULL
+#endif
+
 static struct sysdev_class ipic_sysclass = {
 	.name = "ipic",
+	.suspend = ipic_suspend,
+	.resume = ipic_resume,
 };
 
 static struct sys_device device_ipic = {
diff --git a/include/asm-powerpc/reg.h b/include/asm-powerpc/reg.h
index bbccadfee0d6..c6d1ab650778 100644
--- a/include/asm-powerpc/reg.h
+++ b/include/asm-powerpc/reg.h
@@ -155,10 +155,12 @@
 #define   CTRL_RUNLATCH	0x1
 #define SPRN_DABR	0x3F5	/* Data Address Breakpoint Register */
 #define   DABR_TRANSLATION	(1UL << 2)
+#define SPRN_DABR2	0x13D	/* e300 */
 #define SPRN_DABRX	0x3F7	/* Data Address Breakpoint Register Extension */
 #define   DABRX_USER	(1UL << 0)
 #define   DABRX_KERNEL	(1UL << 1)
 #define SPRN_DAR	0x013	/* Data Address Register */
+#define SPRN_DBCR	0x136	/* e300 Data Breakpoint Control Reg */
 #define SPRN_DSISR	0x012	/* Data Storage Interrupt Status Register */
 #define   DSISR_NOHPTE		0x40000000	/* no translation found */
 #define   DSISR_PROTFAULT	0x08000000	/* protection fault */
@@ -264,6 +266,8 @@
 #define HID1_PS		(1<<16)		/* 750FX PLL selection */
 #define SPRN_HID2	0x3F8		/* Hardware Implementation Register 2 */
 #define SPRN_IABR	0x3F2	/* Instruction Address Breakpoint Register */
+#define SPRN_IABR2	0x3FA		/* 83xx */
+#define SPRN_IBCR	0x135		/* 83xx Insn Breakpoint Control Reg */
 #define SPRN_HID4	0x3F4		/* 970 HID4 */
 #define SPRN_HID5	0x3F6		/* 970 HID5 */
 #define SPRN_HID6	0x3F9	/* BE HID 6 */
diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h
index c415a496de3a..0472877d7ea8 100644
--- a/include/linux/fsl_devices.h
+++ b/include/linux/fsl_devices.h
@@ -125,4 +125,10 @@ struct mpc8xx_pcmcia_ops {
 	int(*voltage_set)(int slot, int vcc, int vpp);
 };
 
+/* Returns non-zero if the current suspend operation would
+ * lead to a deep sleep (i.e. power removed from the core,
+ * instead of just the clock).
+ */
+int fsl_deep_sleep(void);
+
 #endif /* _FSL_DEVICE_H_ */
-- 
cgit v1.2.3


From d87eb12785c14de1586e3bad86ca2c0991300339 Mon Sep 17 00:00:00 2001
From: Scott Wood <scottwood@freescale.com>
Date: Fri, 11 Jul 2008 18:04:45 -0500
Subject: gianfar: Add magic packet and suspend/resume support.

Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/sysdev/fsl_soc.c |   3 ++
 drivers/net/gianfar.c         | 122 +++++++++++++++++++++++++++++++++++++++++-
 drivers/net/gianfar.h         |  12 ++++-
 drivers/net/gianfar_ethtool.c |  41 +++++++++++++-
 include/linux/fsl_devices.h   |   1 +
 5 files changed, 173 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/sysdev/fsl_soc.c b/arch/powerpc/sysdev/fsl_soc.c
index ef4cb0d67a72..214388e11807 100644
--- a/arch/powerpc/sysdev/fsl_soc.c
+++ b/arch/powerpc/sysdev/fsl_soc.c
@@ -352,6 +352,9 @@ static int __init gfar_of_init(void)
 		else
 			gfar_data.interface = PHY_INTERFACE_MODE_MII;
 
+		if (of_get_property(np, "fsl,magic-packet", NULL))
+			gfar_data.device_flags |= FSL_GIANFAR_DEV_HAS_MAGIC_PACKET;
+
 		ph = of_get_property(np, "phy-handle", NULL);
 		if (ph == NULL) {
 			u32 *fixed_link;
diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index 25bdd0832df5..36f229ff52f2 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -143,6 +143,9 @@ static int gfar_process_frame(struct net_device *dev, struct sk_buff *skb, int l
 static void gfar_vlan_rx_register(struct net_device *netdev,
 		                struct vlan_group *grp);
 void gfar_halt(struct net_device *dev);
+#ifdef CONFIG_PM
+static void gfar_halt_nodisable(struct net_device *dev);
+#endif
 void gfar_start(struct net_device *dev);
 static void gfar_clear_exact_match(struct net_device *dev);
 static void gfar_set_mac_for_addr(struct net_device *dev, int num, u8 *addr);
@@ -216,6 +219,7 @@ static int gfar_probe(struct platform_device *pdev)
 
 	spin_lock_init(&priv->txlock);
 	spin_lock_init(&priv->rxlock);
+	spin_lock_init(&priv->bflock);
 
 	platform_set_drvdata(pdev, dev);
 
@@ -393,6 +397,103 @@ static int gfar_remove(struct platform_device *pdev)
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int gfar_suspend(struct platform_device *pdev, pm_message_t state)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	struct gfar_private *priv = netdev_priv(dev);
+	unsigned long flags;
+	u32 tempval;
+
+	int magic_packet = priv->wol_en &&
+		(priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET);
+
+	netif_device_detach(dev);
+
+	if (netif_running(dev)) {
+		spin_lock_irqsave(&priv->txlock, flags);
+		spin_lock(&priv->rxlock);
+
+		gfar_halt_nodisable(dev);
+
+		/* Disable Tx, and Rx if wake-on-LAN is disabled. */
+		tempval = gfar_read(&priv->regs->maccfg1);
+
+		tempval &= ~MACCFG1_TX_EN;
+
+		if (!magic_packet)
+			tempval &= ~MACCFG1_RX_EN;
+
+		gfar_write(&priv->regs->maccfg1, tempval);
+
+		spin_unlock(&priv->rxlock);
+		spin_unlock_irqrestore(&priv->txlock, flags);
+
+#ifdef CONFIG_GFAR_NAPI
+		napi_disable(&priv->napi);
+#endif
+
+		if (magic_packet) {
+			/* Enable interrupt on Magic Packet */
+			gfar_write(&priv->regs->imask, IMASK_MAG);
+
+			/* Enable Magic Packet mode */
+			tempval = gfar_read(&priv->regs->maccfg2);
+			tempval |= MACCFG2_MPEN;
+			gfar_write(&priv->regs->maccfg2, tempval);
+		} else {
+			phy_stop(priv->phydev);
+		}
+	}
+
+	return 0;
+}
+
+static int gfar_resume(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	struct gfar_private *priv = netdev_priv(dev);
+	unsigned long flags;
+	u32 tempval;
+	int magic_packet = priv->wol_en &&
+		(priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET);
+
+	if (!netif_running(dev)) {
+		netif_device_attach(dev);
+		return 0;
+	}
+
+	if (!magic_packet && priv->phydev)
+		phy_start(priv->phydev);
+
+	/* Disable Magic Packet mode, in case something
+	 * else woke us up.
+	 */
+
+	spin_lock_irqsave(&priv->txlock, flags);
+	spin_lock(&priv->rxlock);
+
+	tempval = gfar_read(&priv->regs->maccfg2);
+	tempval &= ~MACCFG2_MPEN;
+	gfar_write(&priv->regs->maccfg2, tempval);
+
+	gfar_start(dev);
+
+	spin_unlock(&priv->rxlock);
+	spin_unlock_irqrestore(&priv->txlock, flags);
+
+	netif_device_attach(dev);
+
+#ifdef CONFIG_GFAR_NAPI
+	napi_enable(&priv->napi);
+#endif
+
+	return 0;
+}
+#else
+#define gfar_suspend NULL
+#define gfar_resume NULL
+#endif
 
 /* Reads the controller's registers to determine what interface
  * connects it to the PHY.
@@ -549,8 +650,9 @@ static void init_registers(struct net_device *dev)
 }
 
 
+#ifdef CONFIG_PM
 /* Halt the receive and transmit queues */
-void gfar_halt(struct net_device *dev)
+static void gfar_halt_nodisable(struct net_device *dev)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	struct gfar __iomem *regs = priv->regs;
@@ -573,6 +675,15 @@ void gfar_halt(struct net_device *dev)
 			 (IEVENT_GRSC | IEVENT_GTSC)))
 			cpu_relax();
 	}
+}
+#endif
+
+/* Halt the receive and transmit queues */
+void gfar_halt(struct net_device *dev)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+	struct gfar __iomem *regs = priv->regs;
+	u32 tempval;
 
 	/* Disable Rx and Tx */
 	tempval = gfar_read(&regs->maccfg1);
@@ -1969,7 +2080,12 @@ static irqreturn_t gfar_error(int irq, void *dev_id)
 	u32 events = gfar_read(&priv->regs->ievent);
 
 	/* Clear IEVENT */
-	gfar_write(&priv->regs->ievent, IEVENT_ERR_MASK);
+	gfar_write(&priv->regs->ievent, events & IEVENT_ERR_MASK);
+
+	/* Magic Packet is not an error. */
+	if ((priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET) &&
+	    (events & IEVENT_MAG))
+		events &= ~IEVENT_MAG;
 
 	/* Hmm... */
 	if (netif_msg_rx_err(priv) || netif_msg_tx_err(priv))
@@ -2042,6 +2158,8 @@ MODULE_ALIAS("platform:fsl-gianfar");
 static struct platform_driver gfar_driver = {
 	.probe = gfar_probe,
 	.remove = gfar_remove,
+	.suspend = gfar_suspend,
+	.resume = gfar_resume,
 	.driver	= {
 		.name = "fsl-gianfar",
 		.owner = THIS_MODULE,
diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h
index 27f37c81e52c..5ee518a8782b 100644
--- a/drivers/net/gianfar.h
+++ b/drivers/net/gianfar.h
@@ -168,6 +168,7 @@ extern const char gfar_driver_version[];
 #define MACCFG2_GMII            0x00000200
 #define MACCFG2_HUGEFRAME	0x00000020
 #define MACCFG2_LENGTHCHECK	0x00000010
+#define MACCFG2_MPEN		0x00000008
 
 #define ECNTRL_INIT_SETTINGS	0x00001000
 #define ECNTRL_TBI_MODE         0x00000020
@@ -240,6 +241,7 @@ extern const char gfar_driver_version[];
 #define IEVENT_CRL		0x00020000
 #define IEVENT_XFUN		0x00010000
 #define IEVENT_RXB0		0x00008000
+#define IEVENT_MAG		0x00000800
 #define IEVENT_GRSC		0x00000100
 #define IEVENT_RXF0		0x00000080
 #define IEVENT_FIR		0x00000008
@@ -252,7 +254,8 @@ extern const char gfar_driver_version[];
 #define IEVENT_ERR_MASK         \
 (IEVENT_RXC | IEVENT_BSY | IEVENT_EBERR | IEVENT_MSRO | \
  IEVENT_BABT | IEVENT_TXC | IEVENT_TXE | IEVENT_LC \
- | IEVENT_CRL | IEVENT_XFUN | IEVENT_DPE | IEVENT_PERR)
+ | IEVENT_CRL | IEVENT_XFUN | IEVENT_DPE | IEVENT_PERR \
+ | IEVENT_MAG)
 
 #define IMASK_INIT_CLEAR	0x00000000
 #define IMASK_BABR              0x80000000
@@ -270,6 +273,7 @@ extern const char gfar_driver_version[];
 #define IMASK_CRL		0x00020000
 #define IMASK_XFUN		0x00010000
 #define IMASK_RXB0              0x00008000
+#define IMASK_MAG		0x00000800
 #define IMASK_GTSC              0x00000100
 #define IMASK_RXFEN0		0x00000080
 #define IMASK_FIR		0x00000008
@@ -737,10 +741,14 @@ struct gfar_private {
 	unsigned int fifo_starve;
 	unsigned int fifo_starve_off;
 
+	/* Bitfield update lock */
+	spinlock_t bflock;
+
 	unsigned char vlan_enable:1,
 		rx_csum_enable:1,
 		extended_hash:1,
-		bd_stash_en:1;
+		bd_stash_en:1,
+		wol_en:1; /* Wake-on-LAN enabled */
 	unsigned short padding;
 
 	unsigned int interruptTransmit;
diff --git a/drivers/net/gianfar_ethtool.c b/drivers/net/gianfar_ethtool.c
index 6007147cc1e9..fb7d3ccc0fdc 100644
--- a/drivers/net/gianfar_ethtool.c
+++ b/drivers/net/gianfar_ethtool.c
@@ -479,14 +479,13 @@ static int gfar_sringparam(struct net_device *dev, struct ethtool_ringparam *rva
 static int gfar_set_rx_csum(struct net_device *dev, uint32_t data)
 {
 	struct gfar_private *priv = netdev_priv(dev);
+	unsigned long flags;
 	int err = 0;
 
 	if (!(priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_CSUM))
 		return -EOPNOTSUPP;
 
 	if (dev->flags & IFF_UP) {
-		unsigned long flags;
-
 		/* Halt TX and RX, and process the frames which
 		 * have already been received */
 		spin_lock_irqsave(&priv->txlock, flags);
@@ -502,7 +501,9 @@ static int gfar_set_rx_csum(struct net_device *dev, uint32_t data)
 		stop_gfar(dev);
 	}
 
+	spin_lock_irqsave(&priv->bflock, flags);
 	priv->rx_csum_enable = data;
+	spin_unlock_irqrestore(&priv->bflock, flags);
 
 	if (dev->flags & IFF_UP)
 		err = startup_gfar(dev);
@@ -564,6 +565,38 @@ static void gfar_set_msglevel(struct net_device *dev, uint32_t data)
 	priv->msg_enable = data;
 }
 
+#ifdef CONFIG_PM
+static void gfar_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+
+	if (priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET) {
+		wol->supported = WAKE_MAGIC;
+		wol->wolopts = priv->wol_en ? WAKE_MAGIC : 0;
+	} else {
+		wol->supported = wol->wolopts = 0;
+	}
+}
+
+static int gfar_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct gfar_private *priv = netdev_priv(dev);
+	unsigned long flags;
+
+	if (!(priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_MAGIC_PACKET) &&
+	    wol->wolopts != 0)
+		return -EINVAL;
+
+	if (wol->wolopts & ~WAKE_MAGIC)
+		return -EINVAL;
+
+	spin_lock_irqsave(&priv->bflock, flags);
+	priv->wol_en = wol->wolopts & WAKE_MAGIC ? 1 : 0;
+	spin_unlock_irqrestore(&priv->bflock, flags);
+
+	return 0;
+}
+#endif
 
 const struct ethtool_ops gfar_ethtool_ops = {
 	.get_settings = gfar_gsettings,
@@ -585,4 +618,8 @@ const struct ethtool_ops gfar_ethtool_ops = {
 	.set_tx_csum = gfar_set_tx_csum,
 	.get_msglevel = gfar_get_msglevel,
 	.set_msglevel = gfar_set_msglevel,
+#ifdef CONFIG_PM
+	.get_wol = gfar_get_wol,
+	.set_wol = gfar_set_wol,
+#endif
 };
diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h
index 0472877d7ea8..4e625e0094c8 100644
--- a/include/linux/fsl_devices.h
+++ b/include/linux/fsl_devices.h
@@ -69,6 +69,7 @@ struct gianfar_mdio_data {
 #define FSL_GIANFAR_DEV_HAS_VLAN		0x00000020
 #define FSL_GIANFAR_DEV_HAS_EXTENDED_HASH	0x00000040
 #define FSL_GIANFAR_DEV_HAS_PADDING		0x00000080
+#define FSL_GIANFAR_DEV_HAS_MAGIC_PACKET	0x00000100
 
 /* Flags in gianfar_platform_data */
 #define FSL_GIANFAR_BRD_HAS_PHY_INTR	0x00000001 /* set or use a timer */
-- 
cgit v1.2.3


From b219108cbacee5f2eaeca63cba013688eeba3bd4 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Thu, 12 Jun 2008 08:32:13 -0500
Subject: fs_enet: Remove !CONFIG_PPC_CPM_NEW_BINDING code

Now that arch/ppc is gone we always define CONFIG_PPC_CPM_NEW_BINDING so
we can remove all the code associated with !CONFIG_PPC_CPM_NEW_BINDING.

Also fixed some asm/of_platform.h to linux/of_platform.h (and of_device.h)

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 drivers/net/fs_enet/Makefile       |   5 -
 drivers/net/fs_enet/fs_enet-main.c | 310 +------------------------------------
 drivers/net/fs_enet/fs_enet.h      |   4 -
 drivers/net/fs_enet/mac-fcc.c      |  67 +-------
 drivers/net/fs_enet/mac-fec.c      |  23 +--
 drivers/net/fs_enet/mac-scc.c      |  37 +----
 drivers/net/fs_enet/mii-bitbang.c  | 107 -------------
 drivers/net/fs_enet/mii-fec.c      | 144 +----------------
 include/linux/fs_enet_pd.h         |   4 -
 9 files changed, 5 insertions(+), 696 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/fs_enet/Makefile b/drivers/net/fs_enet/Makefile
index 1ffbe0756a0c..d4a305ee3455 100644
--- a/drivers/net/fs_enet/Makefile
+++ b/drivers/net/fs_enet/Makefile
@@ -8,12 +8,7 @@ fs_enet-$(CONFIG_FS_ENET_HAS_SCC) += mac-scc.o
 fs_enet-$(CONFIG_FS_ENET_HAS_FEC) += mac-fec.o
 fs_enet-$(CONFIG_FS_ENET_HAS_FCC) += mac-fcc.o
 
-ifeq ($(CONFIG_PPC_CPM_NEW_BINDING),y)
 obj-$(CONFIG_FS_ENET_MDIO_FEC) += mii-fec.o
 obj-$(CONFIG_FS_ENET_MDIO_FCC) += mii-bitbang.o
-else
-fs_enet-$(CONFIG_FS_ENET_MDIO_FEC) += mii-fec.o
-fs_enet-$(CONFIG_FS_ENET_MDIO_FCC) += mii-bitbang.o
-endif
 
 fs_enet-objs := fs_enet-main.o $(fs_enet-m)
diff --git a/drivers/net/fs_enet/fs_enet-main.c b/drivers/net/fs_enet/fs_enet-main.c
index 352574a3f056..ac2c48741d04 100644
--- a/drivers/net/fs_enet/fs_enet-main.c
+++ b/drivers/net/fs_enet/fs_enet-main.c
@@ -36,25 +36,17 @@
 #include <linux/fs.h>
 #include <linux/platform_device.h>
 #include <linux/phy.h>
+#include <linux/of_platform.h>
 
 #include <linux/vmalloc.h>
 #include <asm/pgtable.h>
 #include <asm/irq.h>
 #include <asm/uaccess.h>
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
-#include <linux/of_platform.h>
-#endif
-
 #include "fs_enet.h"
 
 /*************************************************/
 
-#ifndef CONFIG_PPC_CPM_NEW_BINDING
-static char version[] __devinitdata =
-    DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")" "\n";
-#endif
-
 MODULE_AUTHOR("Pantelis Antoniou <panto@intracom.gr>");
 MODULE_DESCRIPTION("Freescale Ethernet Driver");
 MODULE_LICENSE("GPL");
@@ -957,190 +949,6 @@ static int fs_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 extern int fs_mii_connect(struct net_device *dev);
 extern void fs_mii_disconnect(struct net_device *dev);
 
-#ifndef CONFIG_PPC_CPM_NEW_BINDING
-static struct net_device *fs_init_instance(struct device *dev,
-		struct fs_platform_info *fpi)
-{
-	struct net_device *ndev = NULL;
-	struct fs_enet_private *fep = NULL;
-	int privsize, i, r, err = 0, registered = 0;
-
-	fpi->fs_no = fs_get_id(fpi);
-	/* guard */
-	if ((unsigned int)fpi->fs_no >= FS_MAX_INDEX)
-		return ERR_PTR(-EINVAL);
-
-	privsize = sizeof(*fep) + (sizeof(struct sk_buff **) *
-			    (fpi->rx_ring + fpi->tx_ring));
-
-	ndev = alloc_etherdev(privsize);
-	if (!ndev) {
-		err = -ENOMEM;
-		goto err;
-	}
-
-	fep = netdev_priv(ndev);
-
-	fep->dev = dev;
-	dev_set_drvdata(dev, ndev);
-	fep->fpi = fpi;
-	if (fpi->init_ioports)
-		fpi->init_ioports((struct fs_platform_info *)fpi);
-
-#ifdef CONFIG_FS_ENET_HAS_FEC
-	if (fs_get_fec_index(fpi->fs_no) >= 0)
-		fep->ops = &fs_fec_ops;
-#endif
-
-#ifdef CONFIG_FS_ENET_HAS_SCC
-	if (fs_get_scc_index(fpi->fs_no) >=0)
-		fep->ops = &fs_scc_ops;
-#endif
-
-#ifdef CONFIG_FS_ENET_HAS_FCC
-	if (fs_get_fcc_index(fpi->fs_no) >= 0)
-		fep->ops = &fs_fcc_ops;
-#endif
-
-	if (fep->ops == NULL) {
-		printk(KERN_ERR DRV_MODULE_NAME
-		       ": %s No matching ops found (%d).\n",
-		       ndev->name, fpi->fs_no);
-		err = -EINVAL;
-		goto err;
-	}
-
-	r = (*fep->ops->setup_data)(ndev);
-	if (r != 0) {
-		printk(KERN_ERR DRV_MODULE_NAME
-		       ": %s setup_data failed\n",
-			ndev->name);
-		err = r;
-		goto err;
-	}
-
-	/* point rx_skbuff, tx_skbuff */
-	fep->rx_skbuff = (struct sk_buff **)&fep[1];
-	fep->tx_skbuff = fep->rx_skbuff + fpi->rx_ring;
-
-	/* init locks */
-	spin_lock_init(&fep->lock);
-	spin_lock_init(&fep->tx_lock);
-
-	/*
-	 * Set the Ethernet address.
-	 */
-	for (i = 0; i < 6; i++)
-		ndev->dev_addr[i] = fpi->macaddr[i];
-
-	r = (*fep->ops->allocate_bd)(ndev);
-
-	if (fep->ring_base == NULL) {
-		printk(KERN_ERR DRV_MODULE_NAME
-		       ": %s buffer descriptor alloc failed (%d).\n", ndev->name, r);
-		err = r;
-		goto err;
-	}
-
-	/*
-	 * Set receive and transmit descriptor base.
-	 */
-	fep->rx_bd_base = fep->ring_base;
-	fep->tx_bd_base = fep->rx_bd_base + fpi->rx_ring;
-
-	/* initialize ring size variables */
-	fep->tx_ring = fpi->tx_ring;
-	fep->rx_ring = fpi->rx_ring;
-
-	/*
-	 * The FEC Ethernet specific entries in the device structure.
-	 */
-	ndev->open = fs_enet_open;
-	ndev->hard_start_xmit = fs_enet_start_xmit;
-	ndev->tx_timeout = fs_timeout;
-	ndev->watchdog_timeo = 2 * HZ;
-	ndev->stop = fs_enet_close;
-	ndev->get_stats = fs_enet_get_stats;
-	ndev->set_multicast_list = fs_set_multicast_list;
-
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	ndev->poll_controller = fs_enet_netpoll;
-#endif
-
-	netif_napi_add(ndev, &fep->napi,
-		       fs_enet_rx_napi, fpi->napi_weight);
-
-	ndev->ethtool_ops = &fs_ethtool_ops;
-	ndev->do_ioctl = fs_ioctl;
-
-	init_timer(&fep->phy_timer_list);
-
-	netif_carrier_off(ndev);
-
-	err = register_netdev(ndev);
-	if (err != 0) {
-		printk(KERN_ERR DRV_MODULE_NAME
-		       ": %s register_netdev failed.\n", ndev->name);
-		goto err;
-	}
-	registered = 1;
-
-
-	return ndev;
-
-err:
-	if (ndev != NULL) {
-		if (registered)
-			unregister_netdev(ndev);
-
-		if (fep && fep->ops) {
-			(*fep->ops->free_bd)(ndev);
-			(*fep->ops->cleanup_data)(ndev);
-		}
-
-		free_netdev(ndev);
-	}
-
-	dev_set_drvdata(dev, NULL);
-
-	return ERR_PTR(err);
-}
-
-static int fs_cleanup_instance(struct net_device *ndev)
-{
-	struct fs_enet_private *fep;
-	const struct fs_platform_info *fpi;
-	struct device *dev;
-
-	if (ndev == NULL)
-		return -EINVAL;
-
-	fep = netdev_priv(ndev);
-	if (fep == NULL)
-		return -EINVAL;
-
-	fpi = fep->fpi;
-
-	unregister_netdev(ndev);
-
-	dma_free_coherent(fep->dev, (fpi->tx_ring + fpi->rx_ring) * sizeof(cbd_t),
-			  (void __force *)fep->ring_base, fep->ring_mem_addr);
-
-	/* reset it */
-	(*fep->ops->cleanup_data)(ndev);
-
-	dev = fep->dev;
-	if (dev != NULL) {
-		dev_set_drvdata(dev, NULL);
-		fep->dev = NULL;
-	}
-
-	free_netdev(ndev);
-
-	return 0;
-}
-#endif
-
 /**************************************************************************************/
 
 /* handy pointer to the immap */
@@ -1167,7 +975,6 @@ static void cleanup_immap(void)
 
 /**************************************************************************************/
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 static int __devinit find_phy(struct device_node *np,
                               struct fs_platform_info *fpi)
 {
@@ -1399,121 +1206,6 @@ static void __exit fs_cleanup(void)
 	of_unregister_platform_driver(&fs_enet_driver);
 	cleanup_immap();
 }
-#else
-static int __devinit fs_enet_probe(struct device *dev)
-{
-	struct net_device *ndev;
-
-	/* no fixup - no device */
-	if (dev->platform_data == NULL) {
-		printk(KERN_INFO "fs_enet: "
-				"probe called with no platform data; "
-				"remove unused devices\n");
-		return -ENODEV;
-	}
-
-	ndev = fs_init_instance(dev, dev->platform_data);
-	if (IS_ERR(ndev))
-		return PTR_ERR(ndev);
-	return 0;
-}
-
-static int fs_enet_remove(struct device *dev)
-{
-	return fs_cleanup_instance(dev_get_drvdata(dev));
-}
-
-static struct device_driver fs_enet_fec_driver = {
-	.name	  	= "fsl-cpm-fec",
-	.bus		= &platform_bus_type,
-	.probe		= fs_enet_probe,
-	.remove		= fs_enet_remove,
-#ifdef CONFIG_PM
-/*	.suspend	= fs_enet_suspend,	TODO */
-/*	.resume		= fs_enet_resume,	TODO */
-#endif
-};
-
-static struct device_driver fs_enet_scc_driver = {
-	.name	  	= "fsl-cpm-scc",
-	.bus		= &platform_bus_type,
-	.probe		= fs_enet_probe,
-	.remove		= fs_enet_remove,
-#ifdef CONFIG_PM
-/*	.suspend	= fs_enet_suspend,	TODO */
-/*	.resume		= fs_enet_resume,	TODO */
-#endif
-};
-
-static struct device_driver fs_enet_fcc_driver = {
-	.name	  	= "fsl-cpm-fcc",
-	.bus		= &platform_bus_type,
-	.probe		= fs_enet_probe,
-	.remove		= fs_enet_remove,
-#ifdef CONFIG_PM
-/*	.suspend	= fs_enet_suspend,	TODO */
-/*	.resume		= fs_enet_resume,	TODO */
-#endif
-};
-
-static int __init fs_init(void)
-{
-	int r;
-
-	printk(KERN_INFO
-			"%s", version);
-
-	r = setup_immap();
-	if (r != 0)
-		return r;
-
-#ifdef CONFIG_FS_ENET_HAS_FCC
-	/* let's insert mii stuff */
-	r = fs_enet_mdio_bb_init();
-
-	if (r != 0) {
-		printk(KERN_ERR DRV_MODULE_NAME
-			"BB PHY init failed.\n");
-		return r;
-	}
-	r = driver_register(&fs_enet_fcc_driver);
-	if (r != 0)
-		goto err;
-#endif
-
-#ifdef CONFIG_FS_ENET_HAS_FEC
-	r =  fs_enet_mdio_fec_init();
-	if (r != 0) {
-		printk(KERN_ERR DRV_MODULE_NAME
-			"FEC PHY init failed.\n");
-		return r;
-	}
-
-	r = driver_register(&fs_enet_fec_driver);
-	if (r != 0)
-		goto err;
-#endif
-
-#ifdef CONFIG_FS_ENET_HAS_SCC
-	r = driver_register(&fs_enet_scc_driver);
-	if (r != 0)
-		goto err;
-#endif
-
-	return 0;
-err:
-	cleanup_immap();
-	return r;
-}
-
-static void __exit fs_cleanup(void)
-{
-	driver_unregister(&fs_enet_fec_driver);
-	driver_unregister(&fs_enet_fcc_driver);
-	driver_unregister(&fs_enet_scc_driver);
-	cleanup_immap();
-}
-#endif
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
 static void fs_enet_netpoll(struct net_device *dev)
diff --git a/drivers/net/fs_enet/fs_enet.h b/drivers/net/fs_enet/fs_enet.h
index e05389c49bbb..db46d2e72329 100644
--- a/drivers/net/fs_enet/fs_enet.h
+++ b/drivers/net/fs_enet/fs_enet.h
@@ -138,10 +138,6 @@ struct fs_enet_private {
 };
 
 /***************************************************************************/
-#ifndef CONFIG_PPC_CPM_NEW_BINDING
-int fs_enet_mdio_bb_init(void);
-int fs_enet_mdio_fec_init(void);
-#endif
 
 void fs_init_bds(struct net_device *dev);
 void fs_cleanup_bds(struct net_device *dev);
diff --git a/drivers/net/fs_enet/mac-fcc.c b/drivers/net/fs_enet/mac-fcc.c
index 8268b3535b30..0a97fc2d97ec 100644
--- a/drivers/net/fs_enet/mac-fcc.c
+++ b/drivers/net/fs_enet/mac-fcc.c
@@ -33,6 +33,7 @@
 #include <linux/fs.h>
 #include <linux/platform_device.h>
 #include <linux/phy.h>
+#include <linux/of_device.h>
 
 #include <asm/immap_cpm2.h>
 #include <asm/mpc8260.h>
@@ -42,10 +43,6 @@
 #include <asm/irq.h>
 #include <asm/uaccess.h>
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
-#include <asm/of_device.h>
-#endif
-
 #include "fs_enet.h"
 
 /*************************************************/
@@ -87,7 +84,6 @@ static inline int fcc_cr_cmd(struct fs_enet_private *fep, u32 op)
 
 static int do_pd_setup(struct fs_enet_private *fep)
 {
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 	struct of_device *ofdev = to_of_device(fep->dev);
 	struct fs_platform_info *fpi = fep->fpi;
 	int ret = -EINVAL;
@@ -125,44 +121,6 @@ out_fccp:
 	iounmap(fep->fcc.fccp);
 out:
 	return ret;
-#else
-	struct platform_device *pdev = to_platform_device(fep->dev);
-	struct resource *r;
-
-	/* Fill out IRQ field */
-	fep->interrupt = platform_get_irq(pdev, 0);
-	if (fep->interrupt < 0)
-		return -EINVAL;
-
-	/* Attach the memory for the FCC Parameter RAM */
-	r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "fcc_pram");
-	fep->fcc.ep = ioremap(r->start, r->end - r->start + 1);
-	if (fep->fcc.ep == NULL)
-		return -EINVAL;
-
-	r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "fcc_regs");
-	fep->fcc.fccp = ioremap(r->start, r->end - r->start + 1);
-	if (fep->fcc.fccp == NULL)
-		return -EINVAL;
-
-	if (fep->fpi->fcc_regs_c) {
-		fep->fcc.fcccp = (void __iomem *)fep->fpi->fcc_regs_c;
-	} else {
-		r = platform_get_resource_byname(pdev, IORESOURCE_MEM,
-				"fcc_regs_c");
-		fep->fcc.fcccp = ioremap(r->start,
-				r->end - r->start + 1);
-	}
-
-	if (fep->fcc.fcccp == NULL)
-		return -EINVAL;
-
-	fep->fcc.mem = (void __iomem *)fep->fpi->mem_offset;
-	if (fep->fcc.mem == NULL)
-		return -EINVAL;
-
-	return 0;
-#endif
 }
 
 #define FCC_NAPI_RX_EVENT_MSK	(FCC_ENET_RXF | FCC_ENET_RXB)
@@ -173,17 +131,6 @@ out:
 static int setup_data(struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
-#ifndef CONFIG_PPC_CPM_NEW_BINDING
-	struct fs_platform_info *fpi = fep->fpi;
-
-	fpi->cp_command = (fpi->cp_page << 26) |
-	                  (fpi->cp_block << 21) |
-	                  (12 << 6);
-
-	fep->fcc.idx = fs_get_fcc_index(fpi->fs_no);
-	if ((unsigned int)fep->fcc.idx >= 3)	/* max 3 FCCs */
-		return -EINVAL;
-#endif
 
 	if (do_pd_setup(fep) != 0)
 		return -EINVAL;
@@ -304,9 +251,6 @@ static void restart(struct net_device *dev)
 	fcc_enet_t __iomem *ep = fep->fcc.ep;
 	dma_addr_t rx_bd_base_phys, tx_bd_base_phys;
 	u16 paddrh, paddrm, paddrl;
-#ifndef CONFIG_PPC_CPM_NEW_BINDING
-	u16 mem_addr;
-#endif
 	const unsigned char *mac;
 	int i;
 
@@ -338,19 +282,10 @@ static void restart(struct net_device *dev)
 	 * this area.
 	 */
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 	W16(ep, fen_genfcc.fcc_riptr, fpi->dpram_offset);
 	W16(ep, fen_genfcc.fcc_tiptr, fpi->dpram_offset + 32);
 
 	W16(ep, fen_padptr, fpi->dpram_offset + 64);
-#else
-	mem_addr = (u32) fep->fcc.mem;	/* de-fixup dpram offset */
-
-	W16(ep, fen_genfcc.fcc_riptr, (mem_addr & 0xffff));
-	W16(ep, fen_genfcc.fcc_tiptr, ((mem_addr + 32) & 0xffff));
-
-	W16(ep, fen_padptr, mem_addr + 64);
-#endif
 
 	/* fill with special symbol...  */
 	memset_io(fep->fcc.mem + fpi->dpram_offset + 64, 0x88, 32);
diff --git a/drivers/net/fs_enet/mac-fec.c b/drivers/net/fs_enet/mac-fec.c
index 8a311d1e435b..0a7d1c5c6524 100644
--- a/drivers/net/fs_enet/mac-fec.c
+++ b/drivers/net/fs_enet/mac-fec.c
@@ -32,6 +32,7 @@
 #include <linux/bitops.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
+#include <linux/of_device.h>
 
 #include <asm/irq.h>
 #include <asm/uaccess.h>
@@ -43,10 +44,6 @@
 #include <asm/cpm1.h>
 #endif
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
-#include <asm/of_device.h>
-#endif
-
 #include "fs_enet.h"
 #include "fec.h"
 
@@ -99,7 +96,6 @@ static int whack_reset(fec_t __iomem *fecp)
 
 static int do_pd_setup(struct fs_enet_private *fep)
 {
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 	struct of_device *ofdev = to_of_device(fep->dev);
 
 	fep->interrupt = of_irq_to_resource(ofdev->node, 0, NULL);
@@ -111,23 +107,6 @@ static int do_pd_setup(struct fs_enet_private *fep)
 		return -EINVAL;
 
 	return 0;
-#else
-	struct platform_device *pdev = to_platform_device(fep->dev);
-	struct resource	*r;
-
-	/* Fill out IRQ field */
-	fep->interrupt = platform_get_irq_byname(pdev,"interrupt");
-	if (fep->interrupt < 0)
-		return -EINVAL;
-
-	r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "regs");
-	fep->fec.fecp = ioremap(r->start, r->end - r->start + 1);
-
-	if(fep->fec.fecp == NULL)
-		return -EINVAL;
-
-	return 0;
-#endif
 }
 
 #define FEC_NAPI_RX_EVENT_MSK	(FEC_ENET_RXF | FEC_ENET_RXB)
diff --git a/drivers/net/fs_enet/mac-scc.c b/drivers/net/fs_enet/mac-scc.c
index e3557eca7b6d..029b3c7ef29c 100644
--- a/drivers/net/fs_enet/mac-scc.c
+++ b/drivers/net/fs_enet/mac-scc.c
@@ -32,6 +32,7 @@
 #include <linux/bitops.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
+#include <linux/of_platform.h>
 
 #include <asm/irq.h>
 #include <asm/uaccess.h>
@@ -43,10 +44,6 @@
 #include <asm/cpm1.h>
 #endif
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
-#include <linux/of_platform.h>
-#endif
-
 #include "fs_enet.h"
 
 /*************************************************/
@@ -99,7 +96,6 @@ static inline int scc_cr_cmd(struct fs_enet_private *fep, u32 op)
 
 static int do_pd_setup(struct fs_enet_private *fep)
 {
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 	struct of_device *ofdev = to_of_device(fep->dev);
 
 	fep->interrupt = of_irq_to_resource(ofdev->node, 0, NULL);
@@ -115,27 +111,6 @@ static int do_pd_setup(struct fs_enet_private *fep)
 		iounmap(fep->scc.sccp);
 		return -EINVAL;
 	}
-#else
-	struct platform_device *pdev = to_platform_device(fep->dev);
-	struct resource *r;
-
-	/* Fill out IRQ field */
-	fep->interrupt = platform_get_irq_byname(pdev, "interrupt");
-	if (fep->interrupt < 0)
-		return -EINVAL;
-
-	r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "regs");
-	fep->scc.sccp = ioremap(r->start, r->end - r->start + 1);
-
-	if (fep->scc.sccp == NULL)
-		return -EINVAL;
-
-	r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pram");
-	fep->scc.ep = ioremap(r->start, r->end - r->start + 1);
-
-	if (fep->scc.ep == NULL)
-		return -EINVAL;
-#endif
 
 	return 0;
 }
@@ -149,16 +124,6 @@ static int setup_data(struct net_device *dev)
 {
 	struct fs_enet_private *fep = netdev_priv(dev);
 
-#ifndef CONFIG_PPC_CPM_NEW_BINDING
-	struct fs_platform_info *fpi = fep->fpi;
-
-	fep->scc.idx = fs_get_scc_index(fpi->fs_no);
-	if ((unsigned int)fep->fcc.idx >= 4) /* max 4 SCCs */
-		return -EINVAL;
-
-	fpi->cp_command = fep->fcc.idx << 6;
-#endif
-
 	do_pd_setup(fep);
 
 	fep->scc.hthi = 0;
diff --git a/drivers/net/fs_enet/mii-bitbang.c b/drivers/net/fs_enet/mii-bitbang.c
index 1620030cd33c..be4b72f4f49a 100644
--- a/drivers/net/fs_enet/mii-bitbang.c
+++ b/drivers/net/fs_enet/mii-bitbang.c
@@ -22,10 +22,7 @@
 #include <linux/mii.h>
 #include <linux/platform_device.h>
 #include <linux/mdio-bitbang.h>
-
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 #include <linux/of_platform.h>
-#endif
 
 #include "fs_enet.h"
 
@@ -110,7 +107,6 @@ static struct mdiobb_ops bb_ops = {
 	.get_mdio_data = mdio_read,
 };
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 static int __devinit fs_mii_bitbang_init(struct mii_bus *bus,
                                          struct device_node *np)
 {
@@ -271,106 +267,3 @@ static void fs_enet_mdio_bb_exit(void)
 
 module_init(fs_enet_mdio_bb_init);
 module_exit(fs_enet_mdio_bb_exit);
-#else
-static int __devinit fs_mii_bitbang_init(struct bb_info *bitbang,
-                                         struct fs_mii_bb_platform_info *fmpi)
-{
-	bitbang->dir = (u32 __iomem *)fmpi->mdio_dir.offset;
-	bitbang->dat = (u32 __iomem *)fmpi->mdio_dat.offset;
-	bitbang->mdio_msk = 1U << (31 - fmpi->mdio_dat.bit);
-	bitbang->mdc_msk = 1U << (31 - fmpi->mdc_dat.bit);
-
-	return 0;
-}
-
-static int __devinit fs_enet_mdio_probe(struct device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct fs_mii_bb_platform_info *pdata;
-	struct mii_bus *new_bus;
-	struct bb_info *bitbang;
-	int err = 0;
-
-	if (NULL == dev)
-		return -EINVAL;
-
-	bitbang = kzalloc(sizeof(struct bb_info), GFP_KERNEL);
-
-	if (NULL == bitbang)
-		return -ENOMEM;
-
-	bitbang->ctrl.ops = &bb_ops;
-
-	new_bus = alloc_mdio_bitbang(&bitbang->ctrl);
-
-	if (NULL == new_bus)
-		return -ENOMEM;
-
-	new_bus->name = "BB MII Bus",
-	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id);
-
-	new_bus->phy_mask = ~0x9;
-	pdata = (struct fs_mii_bb_platform_info *)pdev->dev.platform_data;
-
-	if (NULL == pdata) {
-		printk(KERN_ERR "gfar mdio %d: Missing platform data!\n", pdev->id);
-		return -ENODEV;
-	}
-
-	/*set up workspace*/
-	fs_mii_bitbang_init(bitbang, pdata);
-
-	new_bus->priv = bitbang;
-
-	new_bus->irq = pdata->irq;
-
-	new_bus->dev = dev;
-	dev_set_drvdata(dev, new_bus);
-
-	err = mdiobus_register(new_bus);
-
-	if (0 != err) {
-		printk (KERN_ERR "%s: Cannot register as MDIO bus\n",
-				new_bus->name);
-		goto bus_register_fail;
-	}
-
-	return 0;
-
-bus_register_fail:
-	free_mdio_bitbang(new_bus);
-	kfree(bitbang);
-
-	return err;
-}
-
-static int fs_enet_mdio_remove(struct device *dev)
-{
-	struct mii_bus *bus = dev_get_drvdata(dev);
-
-	mdiobus_unregister(bus);
-
-	dev_set_drvdata(dev, NULL);
-
-	free_mdio_bitbang(bus);
-
-	return 0;
-}
-
-static struct device_driver fs_enet_bb_mdio_driver = {
-	.name = "fsl-bb-mdio",
-	.bus = &platform_bus_type,
-	.probe = fs_enet_mdio_probe,
-	.remove = fs_enet_mdio_remove,
-};
-
-int fs_enet_mdio_bb_init(void)
-{
-	return driver_register(&fs_enet_bb_mdio_driver);
-}
-
-void fs_enet_mdio_bb_exit(void)
-{
-	driver_unregister(&fs_enet_bb_mdio_driver);
-}
-#endif
diff --git a/drivers/net/fs_enet/mii-fec.c b/drivers/net/fs_enet/mii-fec.c
index 8f6a43b0e0ff..695f74cc4398 100644
--- a/drivers/net/fs_enet/mii-fec.c
+++ b/drivers/net/fs_enet/mii-fec.c
@@ -31,15 +31,12 @@
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/platform_device.h>
+#include <linux/of_platform.h>
 
 #include <asm/pgtable.h>
 #include <asm/irq.h>
 #include <asm/uaccess.h>
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
-#include <linux/of_platform.h>
-#endif
-
 #include "fs_enet.h"
 #include "fec.h"
 
@@ -51,52 +48,6 @@
 
 #define FEC_MII_LOOPS	10000
 
-#ifndef CONFIG_PPC_CPM_NEW_BINDING
-static int match_has_phy (struct device *dev, void* data)
-{
-	struct platform_device* pdev = container_of(dev, struct platform_device, dev);
-	struct fs_platform_info* fpi;
-	if(strcmp(pdev->name, (char*)data))
-	{
-	    return 0;
-	}
-
-	fpi = pdev->dev.platform_data;
-	if((fpi)&&(fpi->has_phy))
-		return 1;
-	return 0;
-}
-
-static int fs_mii_fec_init(struct fec_info* fec, struct fs_mii_fec_platform_info *fmpi)
-{
-	struct resource *r;
-	fec_t __iomem *fecp;
-	char* name = "fsl-cpm-fec";
-
-	/* we need fec in order to be useful */
-	struct platform_device *fec_pdev =
-		container_of(bus_find_device(&platform_bus_type, NULL, name, match_has_phy),
-				struct platform_device, dev);
-
-	if(fec_pdev == NULL) {
-		printk(KERN_ERR"Unable to find PHY for %s", name);
-		return -ENODEV;
-	}
-
-	r = platform_get_resource_byname(fec_pdev, IORESOURCE_MEM, "regs");
-
-	fec->fecp = fecp = ioremap(r->start,sizeof(fec_t));
-	fec->mii_speed = fmpi->mii_speed;
-
-	setbits32(&fecp->fec_r_cntrl, FEC_RCNTRL_MII_MODE);	/* MII enable */
-	setbits32(&fecp->fec_ecntrl, FEC_ECNTRL_PINMUX | FEC_ECNTRL_ETHER_EN);
-	out_be32(&fecp->fec_ievent, FEC_ENET_MII);
-	out_be32(&fecp->fec_mii_speed, fec->mii_speed);
-
-	return 0;
-}
-#endif
-
 static int fs_enet_fec_mii_read(struct mii_bus *bus , int phy_id, int location)
 {
 	struct fec_info* fec = bus->priv;
@@ -151,7 +102,6 @@ static int fs_enet_fec_mii_reset(struct mii_bus *bus)
 	return 0;
 }
 
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 static void __devinit add_phy(struct mii_bus *bus, struct device_node *np)
 {
 	const u32 *data;
@@ -286,95 +236,3 @@ static void fs_enet_mdio_fec_exit(void)
 
 module_init(fs_enet_mdio_fec_init);
 module_exit(fs_enet_mdio_fec_exit);
-#else
-static int __devinit fs_enet_fec_mdio_probe(struct device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct fs_mii_fec_platform_info *pdata;
-	struct mii_bus *new_bus;
-	struct fec_info *fec;
-	int err = 0;
-	if (NULL == dev)
-		return -EINVAL;
-	new_bus = kzalloc(sizeof(struct mii_bus), GFP_KERNEL);
-
-	if (NULL == new_bus)
-		return -ENOMEM;
-
-	fec = kzalloc(sizeof(struct fec_info), GFP_KERNEL);
-
-	if (NULL == fec)
-		return -ENOMEM;
-
-	new_bus->name = "FEC MII Bus",
-	new_bus->read = &fs_enet_fec_mii_read,
-	new_bus->write = &fs_enet_fec_mii_write,
-	new_bus->reset = &fs_enet_fec_mii_reset,
-	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id);
-
-	pdata = (struct fs_mii_fec_platform_info *)pdev->dev.platform_data;
-
-	if (NULL == pdata) {
-		printk(KERN_ERR "fs_enet FEC mdio %d: Missing platform data!\n", pdev->id);
-		return -ENODEV;
-	}
-
-	/*set up workspace*/
-
-	fs_mii_fec_init(fec, pdata);
-	new_bus->priv = fec;
-
-	new_bus->irq = pdata->irq;
-
-	new_bus->dev = dev;
-	dev_set_drvdata(dev, new_bus);
-
-	err = mdiobus_register(new_bus);
-
-	if (0 != err) {
-		printk (KERN_ERR "%s: Cannot register as MDIO bus\n",
-				new_bus->name);
-		goto bus_register_fail;
-	}
-
-	return 0;
-
-bus_register_fail:
-	kfree(new_bus);
-
-	return err;
-}
-
-
-static int fs_enet_fec_mdio_remove(struct device *dev)
-{
-	struct mii_bus *bus = dev_get_drvdata(dev);
-
-	mdiobus_unregister(bus);
-
-	dev_set_drvdata(dev, NULL);
-	kfree(bus->priv);
-
-	bus->priv = NULL;
-	kfree(bus);
-
-	return 0;
-}
-
-static struct device_driver fs_enet_fec_mdio_driver = {
-	.name = "fsl-cpm-fec-mdio",
-	.bus = &platform_bus_type,
-	.probe = fs_enet_fec_mdio_probe,
-	.remove = fs_enet_fec_mdio_remove,
-};
-
-int fs_enet_mdio_fec_init(void)
-{
-	return driver_register(&fs_enet_fec_mdio_driver);
-}
-
-void fs_enet_mdio_fec_exit(void)
-{
-	driver_unregister(&fs_enet_fec_mdio_driver);
-}
-#endif
diff --git a/include/linux/fs_enet_pd.h b/include/linux/fs_enet_pd.h
index 9bc045b8c478..0ba21ee0f58c 100644
--- a/include/linux/fs_enet_pd.h
+++ b/include/linux/fs_enet_pd.h
@@ -135,11 +135,7 @@ struct fs_platform_info {
 	u32 device_flags;
 
 	int phy_addr;		/* the phy address (-1 no phy) */
-#ifdef CONFIG_PPC_CPM_NEW_BINDING
 	char bus_id[16];
-#else
-	const char*	bus_id;
-#endif
 	int phy_irq;		/* the phy irq (if it exists)  */
 
 	const struct fs_mii_bus_info *bus_info;
-- 
cgit v1.2.3


From f470021adb9190819c03d6d8c5c860a17480aa6d Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Mon, 24 Mar 2008 18:36:23 -0700
Subject: ptrace children revamp

ptrace no longer fiddles with the children/sibling links, and the
old ptrace_children list is gone.  Now ptrace, whether of one's own
children or another's via PTRACE_ATTACH, just uses the new ptraced
list instead.

There should be no user-visible difference that matters.  The only
change is the order in which do_wait() sees multiple stopped
children and stopped ptrace attachees.  Since wait_task_stopped()
was changed earlier so it no longer reorders the children list, we
already know this won't cause any new problems.

Signed-off-by: Roland McGrath <roland@redhat.com>
---
 include/linux/init_task.h |   4 +-
 include/linux/sched.h     |  26 +++---
 kernel/exit.c             | 226 ++++++++++++++++++++++++----------------------
 kernel/fork.c             |   6 +-
 kernel/ptrace.c           |  37 +++++---
 5 files changed, 160 insertions(+), 139 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 9927a88674a3..93c45acf249a 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -140,8 +140,8 @@ extern struct group_info init_groups;
 		.nr_cpus_allowed = NR_CPUS,				\
 	},								\
 	.tasks		= LIST_HEAD_INIT(tsk.tasks),			\
-	.ptrace_children= LIST_HEAD_INIT(tsk.ptrace_children),		\
-	.ptrace_list	= LIST_HEAD_INIT(tsk.ptrace_list),		\
+	.ptraced	= LIST_HEAD_INIT(tsk.ptraced),			\
+	.ptrace_entry	= LIST_HEAD_INIT(tsk.ptrace_entry),		\
 	.real_parent	= &tsk,						\
 	.parent		= &tsk,						\
 	.children	= LIST_HEAD_INIT(tsk.children),			\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index ba2f859c6e4f..1941d8b5cf11 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1062,12 +1062,6 @@ struct task_struct {
 #endif
 
 	struct list_head tasks;
-	/*
-	 * ptrace_list/ptrace_children forms the list of my children
-	 * that were stolen by a ptracer.
-	 */
-	struct list_head ptrace_children;
-	struct list_head ptrace_list;
 
 	struct mm_struct *mm, *active_mm;
 
@@ -1089,18 +1083,25 @@ struct task_struct {
 	/* 
 	 * pointers to (original) parent process, youngest child, younger sibling,
 	 * older sibling, respectively.  (p->father can be replaced with 
-	 * p->parent->pid)
+	 * p->real_parent->pid)
 	 */
-	struct task_struct *real_parent; /* real parent process (when being debugged) */
-	struct task_struct *parent;	/* parent process */
+	struct task_struct *real_parent; /* real parent process */
+	struct task_struct *parent; /* recipient of SIGCHLD, wait4() reports */
 	/*
-	 * children/sibling forms the list of my children plus the
-	 * tasks I'm ptracing.
+	 * children/sibling forms the list of my natural children
 	 */
 	struct list_head children;	/* list of my children */
 	struct list_head sibling;	/* linkage in my parent's children list */
 	struct task_struct *group_leader;	/* threadgroup leader */
 
+	/*
+	 * ptraced is the list of tasks this task is using ptrace on.
+	 * This includes both natural children and PTRACE_ATTACH targets.
+	 * p->ptrace_entry is p's link on the p->parent->ptraced list.
+	 */
+	struct list_head ptraced;
+	struct list_head ptrace_entry;
+
 	/* PID/PID hash table linkage. */
 	struct pid_link pids[PIDTYPE_MAX];
 	struct list_head thread_group;
@@ -1876,9 +1877,6 @@ extern void wait_task_inactive(struct task_struct * p);
 #define wait_task_inactive(p)	do { } while (0)
 #endif
 
-#define remove_parent(p)	list_del_init(&(p)->sibling)
-#define add_parent(p)		list_add_tail(&(p)->sibling,&(p)->parent->children)
-
 #define next_task(p)	list_entry(rcu_dereference((p)->tasks.next), struct task_struct, tasks)
 
 #define for_each_process(p) \
diff --git a/kernel/exit.c b/kernel/exit.c
index 7453356a961f..1e909826a804 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -71,7 +71,7 @@ static void __unhash_process(struct task_struct *p)
 		__get_cpu_var(process_counts)--;
 	}
 	list_del_rcu(&p->thread_group);
-	remove_parent(p);
+	list_del_init(&p->sibling);
 }
 
 /*
@@ -152,6 +152,18 @@ static void delayed_put_task_struct(struct rcu_head *rhp)
 	put_task_struct(container_of(rhp, struct task_struct, rcu));
 }
 
+/*
+ * Do final ptrace-related cleanup of a zombie being reaped.
+ *
+ * Called with write_lock(&tasklist_lock) held.
+ */
+static void ptrace_release_task(struct task_struct *p)
+{
+	BUG_ON(!list_empty(&p->ptraced));
+	ptrace_unlink(p);
+	BUG_ON(!list_empty(&p->ptrace_entry));
+}
+
 void release_task(struct task_struct * p)
 {
 	struct task_struct *leader;
@@ -160,8 +172,7 @@ repeat:
 	atomic_dec(&p->user->processes);
 	proc_flush_task(p);
 	write_lock_irq(&tasklist_lock);
-	ptrace_unlink(p);
-	BUG_ON(!list_empty(&p->ptrace_list) || !list_empty(&p->ptrace_children));
+	ptrace_release_task(p);
 	__exit_signal(p);
 
 	/*
@@ -315,9 +326,8 @@ static void reparent_to_kthreadd(void)
 
 	ptrace_unlink(current);
 	/* Reparent to init */
-	remove_parent(current);
 	current->real_parent = current->parent = kthreadd_task;
-	add_parent(current);
+	list_move_tail(&current->sibling, &current->real_parent->children);
 
 	/* Set the exit signal to SIGCHLD so we signal init on exit */
 	current->exit_signal = SIGCHLD;
@@ -692,37 +702,71 @@ static void exit_mm(struct task_struct * tsk)
 	mmput(mm);
 }
 
-static void
-reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
+/*
+ * Detach all tasks we were using ptrace on.
+ * Any that need to be release_task'd are put on the @dead list.
+ *
+ * Called with write_lock(&tasklist_lock) held.
+ */
+static void ptrace_exit(struct task_struct *parent, struct list_head *dead)
 {
-	if (p->pdeath_signal)
-		/* We already hold the tasklist_lock here.  */
-		group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);
+	struct task_struct *p, *n;
 
-	/* Move the child from its dying parent to the new one.  */
-	if (unlikely(traced)) {
-		/* Preserve ptrace links if someone else is tracing this child.  */
-		list_del_init(&p->ptrace_list);
-		if (ptrace_reparented(p))
-			list_add(&p->ptrace_list, &p->real_parent->ptrace_children);
-	} else {
-		/* If this child is being traced, then we're the one tracing it
-		 * anyway, so let go of it.
+	list_for_each_entry_safe(p, n, &parent->ptraced, ptrace_entry) {
+		__ptrace_unlink(p);
+
+		if (p->exit_state != EXIT_ZOMBIE)
+			continue;
+
+		/*
+		 * If it's a zombie, our attachedness prevented normal
+		 * parent notification or self-reaping.  Do notification
+		 * now if it would have happened earlier.  If it should
+		 * reap itself, add it to the @dead list.  We can't call
+		 * release_task() here because we already hold tasklist_lock.
+		 *
+		 * If it's our own child, there is no notification to do.
 		 */
-		p->ptrace = 0;
-		remove_parent(p);
-		p->parent = p->real_parent;
-		add_parent(p);
+		if (!task_detached(p) && thread_group_empty(p)) {
+			if (!same_thread_group(p->real_parent, parent))
+				do_notify_parent(p, p->exit_signal);
+		}
 
-		if (task_is_traced(p)) {
+		if (task_detached(p)) {
 			/*
-			 * If it was at a trace stop, turn it into
-			 * a normal stop since it's no longer being
-			 * traced.
+			 * Mark it as in the process of being reaped.
 			 */
-			ptrace_untrace(p);
+			p->exit_state = EXIT_DEAD;
+			list_add(&p->ptrace_entry, dead);
 		}
 	}
+}
+
+/*
+ * Finish up exit-time ptrace cleanup.
+ *
+ * Called without locks.
+ */
+static void ptrace_exit_finish(struct task_struct *parent,
+			       struct list_head *dead)
+{
+	struct task_struct *p, *n;
+
+	BUG_ON(!list_empty(&parent->ptraced));
+
+	list_for_each_entry_safe(p, n, dead, ptrace_entry) {
+		list_del_init(&p->ptrace_entry);
+		release_task(p);
+	}
+}
+
+static void reparent_thread(struct task_struct *p, struct task_struct *father)
+{
+	if (p->pdeath_signal)
+		/* We already hold the tasklist_lock here.  */
+		group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);
+
+	list_move_tail(&p->sibling, &p->real_parent->children);
 
 	/* If this is a threaded reparent there is no need to
 	 * notify anyone anything has happened.
@@ -737,7 +781,8 @@ reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
 	/* If we'd notified the old parent about this child's death,
 	 * also notify the new parent.
 	 */
-	if (!traced && p->exit_state == EXIT_ZOMBIE &&
+	if (!ptrace_reparented(p) &&
+	    p->exit_state == EXIT_ZOMBIE &&
 	    !task_detached(p) && thread_group_empty(p))
 		do_notify_parent(p, p->exit_signal);
 
@@ -754,12 +799,15 @@ reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
 static void forget_original_parent(struct task_struct *father)
 {
 	struct task_struct *p, *n, *reaper = father;
-	struct list_head ptrace_dead;
-
-	INIT_LIST_HEAD(&ptrace_dead);
+	LIST_HEAD(ptrace_dead);
 
 	write_lock_irq(&tasklist_lock);
 
+	/*
+	 * First clean up ptrace if we were using it.
+	 */
+	ptrace_exit(father, &ptrace_dead);
+
 	do {
 		reaper = next_thread(reaper);
 		if (reaper == father) {
@@ -768,58 +816,19 @@ static void forget_original_parent(struct task_struct *father)
 		}
 	} while (reaper->flags & PF_EXITING);
 
-	/*
-	 * There are only two places where our children can be:
-	 *
-	 * - in our child list
-	 * - in our ptraced child list
-	 *
-	 * Search them and reparent children.
-	 */
 	list_for_each_entry_safe(p, n, &father->children, sibling) {
-		int ptrace;
-
-		ptrace = p->ptrace;
-
-		/* if father isn't the real parent, then ptrace must be enabled */
-		BUG_ON(father != p->real_parent && !ptrace);
-
-		if (father == p->real_parent) {
-			/* reparent with a reaper, real father it's us */
-			p->real_parent = reaper;
-			reparent_thread(p, father, 0);
-		} else {
-			/* reparent ptraced task to its real parent */
-			__ptrace_unlink (p);
-			if (p->exit_state == EXIT_ZOMBIE && !task_detached(p) &&
-			    thread_group_empty(p))
-				do_notify_parent(p, p->exit_signal);
-		}
-
-		/*
-		 * if the ptraced child is a detached zombie we must collect
-		 * it before we exit, or it will remain zombie forever since
-		 * we prevented it from self-reap itself while it was being
-		 * traced by us, to be able to see it in wait4.
-		 */
-		if (unlikely(ptrace && p->exit_state == EXIT_ZOMBIE && task_detached(p)))
-			list_add(&p->ptrace_list, &ptrace_dead);
-	}
-
-	list_for_each_entry_safe(p, n, &father->ptrace_children, ptrace_list) {
 		p->real_parent = reaper;
-		reparent_thread(p, father, 1);
+		if (p->parent == father) {
+			BUG_ON(p->ptrace);
+			p->parent = p->real_parent;
+		}
+		reparent_thread(p, father);
 	}
 
 	write_unlock_irq(&tasklist_lock);
 	BUG_ON(!list_empty(&father->children));
-	BUG_ON(!list_empty(&father->ptrace_children));
-
-	list_for_each_entry_safe(p, n, &ptrace_dead, ptrace_list) {
-		list_del_init(&p->ptrace_list);
-		release_task(p);
-	}
 
+	ptrace_exit_finish(father, &ptrace_dead);
 }
 
 /*
@@ -1180,13 +1189,6 @@ static int eligible_child(enum pid_type type, struct pid *pid, int options,
 			return 0;
 	}
 
-	/*
-	 * Do not consider detached threads that are
-	 * not ptraced:
-	 */
-	if (task_detached(p) && !p->ptrace)
-		return 0;
-
 	/* Wait for all children (clone and not) if __WALL is set;
 	 * otherwise, wait for clone children *only* if __WCLONE is
 	 * set; otherwise, wait for non-clone children *only*.  (Note:
@@ -1399,7 +1401,7 @@ static int wait_task_zombie(struct task_struct *p, int options,
  * the lock and this task is uninteresting.  If we return nonzero, we have
  * released the lock and the system call should return.
  */
-static int wait_task_stopped(struct task_struct *p,
+static int wait_task_stopped(int ptrace, struct task_struct *p,
 			     int options, struct siginfo __user *infop,
 			     int __user *stat_addr, struct rusage __user *ru)
 {
@@ -1407,7 +1409,7 @@ static int wait_task_stopped(struct task_struct *p,
 	uid_t uid = 0; /* unneeded, required by compiler */
 	pid_t pid;
 
-	if (!(p->ptrace & PT_PTRACED) && !(options & WUNTRACED))
+	if (!(options & WUNTRACED))
 		return 0;
 
 	exit_code = 0;
@@ -1416,7 +1418,7 @@ static int wait_task_stopped(struct task_struct *p,
 	if (unlikely(!task_is_stopped_or_traced(p)))
 		goto unlock_sig;
 
-	if (!(p->ptrace & PT_PTRACED) && p->signal->group_stop_count > 0)
+	if (!ptrace && p->signal->group_stop_count > 0)
 		/*
 		 * A group stop is in progress and this is the group leader.
 		 * We won't report until all threads have stopped.
@@ -1445,7 +1447,7 @@ unlock_sig:
 	 */
 	get_task_struct(p);
 	pid = task_pid_vnr(p);
-	why = (p->ptrace & PT_PTRACED) ? CLD_TRAPPED : CLD_STOPPED;
+	why = ptrace ? CLD_TRAPPED : CLD_STOPPED;
 	read_unlock(&tasklist_lock);
 
 	if (unlikely(options & WNOWAIT))
@@ -1536,7 +1538,7 @@ static int wait_task_continued(struct task_struct *p, int options,
  * Returns zero if the search for a child should continue;
  * then *@notask_error is 0 if @p is an eligible child, or still -ECHILD.
  */
-static int wait_consider_task(struct task_struct *parent,
+static int wait_consider_task(struct task_struct *parent, int ptrace,
 			      struct task_struct *p, int *notask_error,
 			      enum pid_type type, struct pid *pid, int options,
 			      struct siginfo __user *infop,
@@ -1546,6 +1548,15 @@ static int wait_consider_task(struct task_struct *parent,
 	if (ret <= 0)
 		return ret;
 
+	if (likely(!ptrace) && unlikely(p->ptrace)) {
+		/*
+		 * This child is hidden by ptrace.
+		 * We aren't allowed to see it now, but eventually we will.
+		 */
+		*notask_error = 0;
+		return 0;
+	}
+
 	if (p->exit_state == EXIT_DEAD)
 		return 0;
 
@@ -1562,7 +1573,8 @@ static int wait_consider_task(struct task_struct *parent,
 	*notask_error = 0;
 
 	if (task_is_stopped_or_traced(p))
-		return wait_task_stopped(p, options, infop, stat_addr, ru);
+		return wait_task_stopped(ptrace, p, options,
+					 infop, stat_addr, ru);
 
 	return wait_task_continued(p, options, infop, stat_addr, ru);
 }
@@ -1583,11 +1595,16 @@ static int do_wait_thread(struct task_struct *tsk, int *notask_error,
 	struct task_struct *p;
 
 	list_for_each_entry(p, &tsk->children, sibling) {
-		int ret = wait_consider_task(tsk, p, notask_error,
-					     type, pid, options,
-					     infop, stat_addr, ru);
-		if (ret)
-			return ret;
+		/*
+		 * Do not consider detached threads.
+		 */
+		if (!task_detached(p)) {
+			int ret = wait_consider_task(tsk, 0, p, notask_error,
+						     type, pid, options,
+						     infop, stat_addr, ru);
+			if (ret)
+				return ret;
+		}
 	}
 
 	return 0;
@@ -1601,21 +1618,16 @@ static int ptrace_do_wait(struct task_struct *tsk, int *notask_error,
 	struct task_struct *p;
 
 	/*
-	 * If we never saw an eligile child, check for children stolen by
-	 * ptrace.  We don't leave -ECHILD in *@notask_error if there are any,
-	 * because we will eventually be allowed to wait for them again.
+	 * Traditionally we see ptrace'd stopped tasks regardless of options.
 	 */
-	if (!*notask_error)
-		return 0;
+	options |= WUNTRACED;
 
-	list_for_each_entry(p, &tsk->ptrace_children, ptrace_list) {
-		int ret = eligible_child(type, pid, options, p);
-		if (unlikely(ret < 0))
+	list_for_each_entry(p, &tsk->ptraced, ptrace_entry) {
+		int ret = wait_consider_task(tsk, 1, p, notask_error,
+					     type, pid, options,
+					     infop, stat_addr, ru);
+		if (ret)
 			return ret;
-		if (ret) {
-			*notask_error = 0;
-			return 0;
-		}
 	}
 
 	return 0;
diff --git a/kernel/fork.c b/kernel/fork.c
index 4bd2f516401f..adefc1131f27 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1125,8 +1125,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	 */
 	p->group_leader = p;
 	INIT_LIST_HEAD(&p->thread_group);
-	INIT_LIST_HEAD(&p->ptrace_children);
-	INIT_LIST_HEAD(&p->ptrace_list);
+	INIT_LIST_HEAD(&p->ptrace_entry);
+	INIT_LIST_HEAD(&p->ptraced);
 
 	/* Now that the task is set up, run cgroup callbacks if
 	 * necessary. We need to run them before the task is visible
@@ -1198,7 +1198,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	}
 
 	if (likely(p->pid)) {
-		add_parent(p);
+		list_add_tail(&p->sibling, &p->real_parent->children);
 		if (unlikely(p->ptrace & PT_PTRACED))
 			__ptrace_link(p, current->parent);
 
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index e337390fce01..8392a9da6450 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -33,13 +33,9 @@
  */
 void __ptrace_link(struct task_struct *child, struct task_struct *new_parent)
 {
-	BUG_ON(!list_empty(&child->ptrace_list));
-	if (child->parent == new_parent)
-		return;
-	list_add(&child->ptrace_list, &child->parent->ptrace_children);
-	remove_parent(child);
+	BUG_ON(!list_empty(&child->ptrace_entry));
+	list_add(&child->ptrace_entry, &new_parent->ptraced);
 	child->parent = new_parent;
-	add_parent(child);
 }
  
 /*
@@ -73,12 +69,8 @@ void __ptrace_unlink(struct task_struct *child)
 	BUG_ON(!child->ptrace);
 
 	child->ptrace = 0;
-	if (ptrace_reparented(child)) {
-		list_del_init(&child->ptrace_list);
-		remove_parent(child);
-		child->parent = child->real_parent;
-		add_parent(child);
-	}
+	child->parent = child->real_parent;
+	list_del_init(&child->ptrace_entry);
 
 	if (task_is_traced(child))
 		ptrace_untrace(child);
@@ -492,15 +484,34 @@ int ptrace_traceme(void)
 	/*
 	 * Are we already being traced?
 	 */
+repeat:
 	task_lock(current);
 	if (!(current->ptrace & PT_PTRACED)) {
+		/*
+		 * See ptrace_attach() comments about the locking here.
+		 */
+		unsigned long flags;
+		if (!write_trylock_irqsave(&tasklist_lock, flags)) {
+			task_unlock(current);
+			do {
+				cpu_relax();
+			} while (!write_can_lock(&tasklist_lock));
+			goto repeat;
+		}
+
 		ret = security_ptrace(current->parent, current,
 				      PTRACE_MODE_ATTACH);
+
 		/*
 		 * Set the ptrace bit in the process ptrace flags.
+		 * Then link us on our parent's ptraced list.
 		 */
-		if (!ret)
+		if (!ret) {
 			current->ptrace |= PT_PTRACED;
+			__ptrace_link(current, current->real_parent);
+		}
+
+		write_unlock_irqrestore(&tasklist_lock, flags);
 	}
 	task_unlock(current);
 	return ret;
-- 
cgit v1.2.3


From f89ab8619e5320cc9c2576f5f8dcbaf6c0ba3950 Mon Sep 17 00:00:00 2001
From: Joel Becker <joel.becker@oracle.com>
Date: Thu, 17 Jul 2008 14:53:48 -0700
Subject: Revert "configfs: Allow ->make_item() and ->make_group() to return
 detailed errors."

This reverts commit 11c3b79218390a139f2d474ee1e983a672d5839a.  The code
will move to PTR_ERR().

Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 Documentation/filesystems/configfs/configfs.txt    | 10 ++---
 .../filesystems/configfs/configfs_example.c        | 14 +++----
 drivers/net/netconsole.c                           | 10 ++---
 fs/configfs/dir.c                                  | 13 ++++---
 fs/dlm/config.c                                    | 45 ++++++++--------------
 fs/ocfs2/cluster/heartbeat.c                       | 17 ++++----
 fs/ocfs2/cluster/nodemanager.c                     | 45 ++++++++--------------
 include/linux/configfs.h                           |  4 +-
 8 files changed, 64 insertions(+), 94 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/configfs/configfs.txt b/Documentation/filesystems/configfs/configfs.txt
index 15838d706ea2..44c97e6accb2 100644
--- a/Documentation/filesystems/configfs/configfs.txt
+++ b/Documentation/filesystems/configfs/configfs.txt
@@ -233,12 +233,10 @@ accomplished via the group operations specified on the group's
 config_item_type.
 
 	struct configfs_group_operations {
-		int (*make_item)(struct config_group *group,
-				 const char *name,
-				 struct config_item **new_item);
-		int (*make_group)(struct config_group *group,
-				  const char *name,
-				  struct config_group **new_group);
+		struct config_item *(*make_item)(struct config_group *group,
+						 const char *name);
+		struct config_group *(*make_group)(struct config_group *group,
+						   const char *name);
 		int (*commit_item)(struct config_item *item);
 		void (*disconnect_notify)(struct config_group *group,
 					  struct config_item *item);
diff --git a/Documentation/filesystems/configfs/configfs_example.c b/Documentation/filesystems/configfs/configfs_example.c
index 0b422acd470c..25151fd5c2c6 100644
--- a/Documentation/filesystems/configfs/configfs_example.c
+++ b/Documentation/filesystems/configfs/configfs_example.c
@@ -273,13 +273,13 @@ static inline struct simple_children *to_simple_children(struct config_item *ite
 	return item ? container_of(to_config_group(item), struct simple_children, group) : NULL;
 }
 
-static int simple_children_make_item(struct config_group *group, const char *name, struct config_item **new_item)
+static struct config_item *simple_children_make_item(struct config_group *group, const char *name)
 {
 	struct simple_child *simple_child;
 
 	simple_child = kzalloc(sizeof(struct simple_child), GFP_KERNEL);
 	if (!simple_child)
-		return -ENOMEM;
+		return NULL;
 
 
 	config_item_init_type_name(&simple_child->item, name,
@@ -287,8 +287,7 @@ static int simple_children_make_item(struct config_group *group, const char *nam
 
 	simple_child->storeme = 0;
 
-	*new_item = &simple_child->item;
-	return 0;
+	return &simple_child->item;
 }
 
 static struct configfs_attribute simple_children_attr_description = {
@@ -360,21 +359,20 @@ static struct configfs_subsystem simple_children_subsys = {
  * children of its own.
  */
 
-static int group_children_make_group(struct config_group *group, const char *name, struct config_group **new_group)
+static struct config_group *group_children_make_group(struct config_group *group, const char *name)
 {
 	struct simple_children *simple_children;
 
 	simple_children = kzalloc(sizeof(struct simple_children),
 				  GFP_KERNEL);
 	if (!simple_children)
-		return -ENOMEM;
+		return NULL;
 
 
 	config_group_init_type_name(&simple_children->group, name,
 				    &simple_children_type);
 
-	*new_group = &simple_children->group;
-	return 0;
+	return &simple_children->group;
 }
 
 static struct configfs_attribute group_children_attr_description = {
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 387a13395015..665341e43055 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -585,9 +585,8 @@ static struct config_item_type netconsole_target_type = {
  * Group operations and type for netconsole_subsys.
  */
 
-static int make_netconsole_target(struct config_group *group,
-				  const char *name,
-				  struct config_item **new_item)
+static struct config_item *make_netconsole_target(struct config_group *group,
+						  const char *name)
 {
 	unsigned long flags;
 	struct netconsole_target *nt;
@@ -599,7 +598,7 @@ static int make_netconsole_target(struct config_group *group,
 	nt = kzalloc(sizeof(*nt), GFP_KERNEL);
 	if (!nt) {
 		printk(KERN_ERR "netconsole: failed to allocate memory\n");
-		return -ENOMEM;
+		return NULL;
 	}
 
 	nt->np.name = "netconsole";
@@ -616,8 +615,7 @@ static int make_netconsole_target(struct config_group *group,
 	list_add(&nt->list, &target_list);
 	spin_unlock_irqrestore(&target_list_lock, flags);
 
-	*new_item = &nt->item;
-	return 0;
+	return &nt->item;
 }
 
 static void drop_netconsole_target(struct config_group *group,
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 0e64312a084c..614e382a6049 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -1073,24 +1073,25 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	group = NULL;
 	item = NULL;
 	if (type->ct_group_ops->make_group) {
-		ret = type->ct_group_ops->make_group(to_config_group(parent_item), name, &group);
-		if (!ret) {
+		group = type->ct_group_ops->make_group(to_config_group(parent_item), name);
+		if (group) {
 			link_group(to_config_group(parent_item), group);
 			item = &group->cg_item;
 		}
 	} else {
-		ret = type->ct_group_ops->make_item(to_config_group(parent_item), name, &item);
-		if (!ret)
+		item = type->ct_group_ops->make_item(to_config_group(parent_item), name);
+		if (item)
 			link_obj(parent_item, item);
 	}
 	mutex_unlock(&subsys->su_mutex);
 
 	kfree(name);
-	if (ret) {
+	if (!item) {
 		/*
-		 * If ret != 0, then link_obj() was never called.
+		 * If item == NULL, then link_obj() was never called.
 		 * There are no extra references to clean up.
 		 */
+		ret = -ENOMEM;
 		goto out_put;
 	}
 
diff --git a/fs/dlm/config.c b/fs/dlm/config.c
index 492d8caaaf25..eac23bd288b2 100644
--- a/fs/dlm/config.c
+++ b/fs/dlm/config.c
@@ -41,20 +41,16 @@ struct comm;
 struct nodes;
 struct node;
 
-static int make_cluster(struct config_group *, const char *,
-			struct config_group **);
+static struct config_group *make_cluster(struct config_group *, const char *);
 static void drop_cluster(struct config_group *, struct config_item *);
 static void release_cluster(struct config_item *);
-static int make_space(struct config_group *, const char *,
-		      struct config_group **);
+static struct config_group *make_space(struct config_group *, const char *);
 static void drop_space(struct config_group *, struct config_item *);
 static void release_space(struct config_item *);
-static int make_comm(struct config_group *, const char *,
-		     struct config_item **);
+static struct config_item *make_comm(struct config_group *, const char *);
 static void drop_comm(struct config_group *, struct config_item *);
 static void release_comm(struct config_item *);
-static int make_node(struct config_group *, const char *,
-		     struct config_item **);
+static struct config_item *make_node(struct config_group *, const char *);
 static void drop_node(struct config_group *, struct config_item *);
 static void release_node(struct config_item *);
 
@@ -396,8 +392,8 @@ static struct node *to_node(struct config_item *i)
 	return i ? container_of(i, struct node, item) : NULL;
 }
 
-static int make_cluster(struct config_group *g, const char *name,
-			struct config_group **new_g)
+static struct config_group *make_cluster(struct config_group *g,
+					 const char *name)
 {
 	struct cluster *cl = NULL;
 	struct spaces *sps = NULL;
@@ -435,15 +431,14 @@ static int make_cluster(struct config_group *g, const char *name,
 
 	space_list = &sps->ss_group;
 	comm_list = &cms->cs_group;
-	*new_g = &cl->group;
-	return 0;
+	return &cl->group;
 
  fail:
 	kfree(cl);
 	kfree(gps);
 	kfree(sps);
 	kfree(cms);
-	return -ENOMEM;
+	return NULL;
 }
 
 static void drop_cluster(struct config_group *g, struct config_item *i)
@@ -471,8 +466,7 @@ static void release_cluster(struct config_item *i)
 	kfree(cl);
 }
 
-static int make_space(struct config_group *g, const char *name,
-		      struct config_group **new_g)
+static struct config_group *make_space(struct config_group *g, const char *name)
 {
 	struct space *sp = NULL;
 	struct nodes *nds = NULL;
@@ -495,14 +489,13 @@ static int make_space(struct config_group *g, const char *name,
 	INIT_LIST_HEAD(&sp->members);
 	mutex_init(&sp->members_lock);
 	sp->members_count = 0;
-	*new_g = &sp->group;
-	return 0;
+	return &sp->group;
 
  fail:
 	kfree(sp);
 	kfree(gps);
 	kfree(nds);
-	return -ENOMEM;
+	return NULL;
 }
 
 static void drop_space(struct config_group *g, struct config_item *i)
@@ -529,21 +522,19 @@ static void release_space(struct config_item *i)
 	kfree(sp);
 }
 
-static int make_comm(struct config_group *g, const char *name,
-		     struct config_item **new_i)
+static struct config_item *make_comm(struct config_group *g, const char *name)
 {
 	struct comm *cm;
 
 	cm = kzalloc(sizeof(struct comm), GFP_KERNEL);
 	if (!cm)
-		return -ENOMEM;
+		return NULL;
 
 	config_item_init_type_name(&cm->item, name, &comm_type);
 	cm->nodeid = -1;
 	cm->local = 0;
 	cm->addr_count = 0;
-	*new_i = &cm->item;
-	return 0;
+	return &cm->item;
 }
 
 static void drop_comm(struct config_group *g, struct config_item *i)
@@ -563,15 +554,14 @@ static void release_comm(struct config_item *i)
 	kfree(cm);
 }
 
-static int make_node(struct config_group *g, const char *name,
-		     struct config_item **new_i)
+static struct config_item *make_node(struct config_group *g, const char *name)
 {
 	struct space *sp = to_space(g->cg_item.ci_parent);
 	struct node *nd;
 
 	nd = kzalloc(sizeof(struct node), GFP_KERNEL);
 	if (!nd)
-		return -ENOMEM;
+		return NULL;
 
 	config_item_init_type_name(&nd->item, name, &node_type);
 	nd->nodeid = -1;
@@ -583,8 +573,7 @@ static int make_node(struct config_group *g, const char *name,
 	sp->members_count++;
 	mutex_unlock(&sp->members_lock);
 
-	*new_i = &nd->item;
-	return 0;
+	return &nd->item;
 }
 
 static void drop_node(struct config_group *g, struct config_item *i)
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index 443d108211ab..f02ccb34604d 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -1489,28 +1489,25 @@ static struct o2hb_heartbeat_group *to_o2hb_heartbeat_group(struct config_group
 		: NULL;
 }
 
-static int o2hb_heartbeat_group_make_item(struct config_group *group,
-					  const char *name,
-					  struct config_item **new_item)
+static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *group,
+							  const char *name)
 {
 	struct o2hb_region *reg = NULL;
-	int ret = 0;
+	struct config_item *ret = NULL;
 
 	reg = kzalloc(sizeof(struct o2hb_region), GFP_KERNEL);
-	if (reg == NULL) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (reg == NULL)
+		goto out; /* ENOMEM */
 
 	config_item_init_type_name(&reg->hr_item, name, &o2hb_region_type);
 
-	*new_item = &reg->hr_item;
+	ret = &reg->hr_item;
 
 	spin_lock(&o2hb_live_lock);
 	list_add_tail(&reg->hr_all_item, &o2hb_all_regions);
 	spin_unlock(&o2hb_live_lock);
 out:
-	if (ret)
+	if (ret == NULL)
 		kfree(reg);
 
 	return ret;
diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c
index b364b7052e46..cfdb08b484ed 100644
--- a/fs/ocfs2/cluster/nodemanager.c
+++ b/fs/ocfs2/cluster/nodemanager.c
@@ -644,32 +644,27 @@ out:
 	return ret;
 }
 
-static int o2nm_node_group_make_item(struct config_group *group,
-				     const char *name,
-				     struct config_item **new_item)
+static struct config_item *o2nm_node_group_make_item(struct config_group *group,
+						     const char *name)
 {
 	struct o2nm_node *node = NULL;
-	int ret = 0;
+	struct config_item *ret = NULL;
 
-	if (strlen(name) > O2NM_MAX_NAME_LEN) {
-		ret = -ENAMETOOLONG;
-		goto out;
-	}
+	if (strlen(name) > O2NM_MAX_NAME_LEN)
+		goto out; /* ENAMETOOLONG */
 
 	node = kzalloc(sizeof(struct o2nm_node), GFP_KERNEL);
-	if (node == NULL) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (node == NULL)
+		goto out; /* ENOMEM */
 
 	strcpy(node->nd_name, name); /* use item.ci_namebuf instead? */
 	config_item_init_type_name(&node->nd_item, name, &o2nm_node_type);
 	spin_lock_init(&node->nd_lock);
 
-	*new_item = &node->nd_item;
+	ret = &node->nd_item;
 
 out:
-	if (ret)
+	if (ret == NULL)
 		kfree(node);
 
 	return ret;
@@ -756,31 +751,25 @@ static struct o2nm_cluster_group *to_o2nm_cluster_group(struct config_group *gro
 }
 #endif
 
-static int o2nm_cluster_group_make_group(struct config_group *group,
-					 const char *name,
-					 struct config_group **new_group)
+static struct config_group *o2nm_cluster_group_make_group(struct config_group *group,
+							  const char *name)
 {
 	struct o2nm_cluster *cluster = NULL;
 	struct o2nm_node_group *ns = NULL;
-	struct config_group *o2hb_group = NULL;
+	struct config_group *o2hb_group = NULL, *ret = NULL;
 	void *defs = NULL;
-	int ret = 0;
 
 	/* this runs under the parent dir's i_mutex; there can be only
 	 * one caller in here at a time */
-	if (o2nm_single_cluster) {
-		ret = -ENOSPC;
-		goto out;
-	}
+	if (o2nm_single_cluster)
+		goto out; /* ENOSPC */
 
 	cluster = kzalloc(sizeof(struct o2nm_cluster), GFP_KERNEL);
 	ns = kzalloc(sizeof(struct o2nm_node_group), GFP_KERNEL);
 	defs = kcalloc(3, sizeof(struct config_group *), GFP_KERNEL);
 	o2hb_group = o2hb_alloc_hb_set();
-	if (cluster == NULL || ns == NULL || o2hb_group == NULL || defs == NULL) {
-		ret = -ENOMEM;
+	if (cluster == NULL || ns == NULL || o2hb_group == NULL || defs == NULL)
 		goto out;
-	}
 
 	config_group_init_type_name(&cluster->cl_group, name,
 				    &o2nm_cluster_type);
@@ -797,11 +786,11 @@ static int o2nm_cluster_group_make_group(struct config_group *group,
 	cluster->cl_idle_timeout_ms    = O2NET_IDLE_TIMEOUT_MS_DEFAULT;
 	cluster->cl_keepalive_delay_ms = O2NET_KEEPALIVE_DELAY_MS_DEFAULT;
 
-	*new_group = &cluster->cl_group;
+	ret = &cluster->cl_group;
 	o2nm_single_cluster = cluster;
 
 out:
-	if (ret) {
+	if (ret == NULL) {
 		kfree(cluster);
 		kfree(ns);
 		o2hb_free_hb_set(o2hb_group);
diff --git a/include/linux/configfs.h b/include/linux/configfs.h
index 0488f937634a..3ae65b1bf90f 100644
--- a/include/linux/configfs.h
+++ b/include/linux/configfs.h
@@ -165,8 +165,8 @@ struct configfs_item_operations {
 };
 
 struct configfs_group_operations {
-	int (*make_item)(struct config_group *group, const char *name, struct config_item **new_item);
-	int (*make_group)(struct config_group *group, const char *name, struct config_group **new_group);
+	struct config_item *(*make_item)(struct config_group *group, const char *name);
+	struct config_group *(*make_group)(struct config_group *group, const char *name);
 	int (*commit_item)(struct config_item *item);
 	void (*disconnect_notify)(struct config_group *group, struct config_item *item);
 	void (*drop_item)(struct config_group *group, struct config_item *item);
-- 
cgit v1.2.3


From a6795e9ebb420d87af43789174689af0d66d1d35 Mon Sep 17 00:00:00 2001
From: Joel Becker <joel.becker@oracle.com>
Date: Thu, 17 Jul 2008 15:21:29 -0700
Subject: configfs: Allow ->make_item() and ->make_group() to return detailed
 errors.

The configfs operations ->make_item() and ->make_group() currently
return a new item/group.  A return of NULL signifies an error.  Because
of this, -ENOMEM is the only return code bubbled up the stack.

Multiple folks have requested the ability to return specific error codes
when these operations fail.  This patch adds that ability by changing the
->make_item/group() ops to return ERR_PTR() values.  These errors are
bubbled up appropriately.  NULL returns are changed to -ENOMEM for
compatibility.

Also updated are the in-kernel users of configfs.

This is a rework of reverted commit 11c3b79218390a139f2d474ee1e983a672d5839a.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 .../filesystems/configfs/configfs_example.c        |  4 ++--
 drivers/net/netconsole.c                           |  2 +-
 fs/configfs/dir.c                                  | 25 +++++++++++++---------
 fs/dlm/config.c                                    |  8 +++----
 fs/ocfs2/cluster/heartbeat.c                       | 10 ++-------
 fs/ocfs2/cluster/nodemanager.c                     | 16 +++++---------
 include/linux/configfs.h                           |  3 ++-
 7 files changed, 31 insertions(+), 37 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/configfs/configfs_example.c b/Documentation/filesystems/configfs/configfs_example.c
index 25151fd5c2c6..039648791701 100644
--- a/Documentation/filesystems/configfs/configfs_example.c
+++ b/Documentation/filesystems/configfs/configfs_example.c
@@ -279,7 +279,7 @@ static struct config_item *simple_children_make_item(struct config_group *group,
 
 	simple_child = kzalloc(sizeof(struct simple_child), GFP_KERNEL);
 	if (!simple_child)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 
 	config_item_init_type_name(&simple_child->item, name,
@@ -366,7 +366,7 @@ static struct config_group *group_children_make_group(struct config_group *group
 	simple_children = kzalloc(sizeof(struct simple_children),
 				  GFP_KERNEL);
 	if (!simple_children)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 
 	config_group_init_type_name(&simple_children->group, name,
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 665341e43055..e13966bb5f77 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -598,7 +598,7 @@ static struct config_item *make_netconsole_target(struct config_group *group,
 	nt = kzalloc(sizeof(*nt), GFP_KERNEL);
 	if (!nt) {
 		printk(KERN_ERR "netconsole: failed to allocate memory\n");
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 	}
 
 	nt->np.name = "netconsole";
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 614e382a6049..179589be063a 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -1027,9 +1027,10 @@ EXPORT_SYMBOL(configfs_undepend_item);
 
 static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
-	int ret, module_got = 0;
-	struct config_group *group;
-	struct config_item *item;
+	int ret = 0;
+	int module_got = 0;
+	struct config_group *group = NULL;
+	struct config_item *item = NULL;
 	struct config_item *parent_item;
 	struct configfs_subsystem *subsys;
 	struct configfs_dirent *sd;
@@ -1070,28 +1071,32 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	snprintf(name, dentry->d_name.len + 1, "%s", dentry->d_name.name);
 
 	mutex_lock(&subsys->su_mutex);
-	group = NULL;
-	item = NULL;
 	if (type->ct_group_ops->make_group) {
 		group = type->ct_group_ops->make_group(to_config_group(parent_item), name);
-		if (group) {
+		if (!group)
+			group = ERR_PTR(-ENOMEM);
+		if (!IS_ERR(group)) {
 			link_group(to_config_group(parent_item), group);
 			item = &group->cg_item;
-		}
+		} else
+			ret = PTR_ERR(group);
 	} else {
 		item = type->ct_group_ops->make_item(to_config_group(parent_item), name);
-		if (item)
+		if (!item)
+			item = ERR_PTR(-ENOMEM);
+		if (!IS_ERR(item))
 			link_obj(parent_item, item);
+		else
+			ret = PTR_ERR(item);
 	}
 	mutex_unlock(&subsys->su_mutex);
 
 	kfree(name);
-	if (!item) {
+	if (ret) {
 		/*
 		 * If item == NULL, then link_obj() was never called.
 		 * There are no extra references to clean up.
 		 */
-		ret = -ENOMEM;
 		goto out_put;
 	}
 
diff --git a/fs/dlm/config.c b/fs/dlm/config.c
index eac23bd288b2..c4e7d721bd8d 100644
--- a/fs/dlm/config.c
+++ b/fs/dlm/config.c
@@ -438,7 +438,7 @@ static struct config_group *make_cluster(struct config_group *g,
 	kfree(gps);
 	kfree(sps);
 	kfree(cms);
-	return NULL;
+	return ERR_PTR(-ENOMEM);
 }
 
 static void drop_cluster(struct config_group *g, struct config_item *i)
@@ -495,7 +495,7 @@ static struct config_group *make_space(struct config_group *g, const char *name)
 	kfree(sp);
 	kfree(gps);
 	kfree(nds);
-	return NULL;
+	return ERR_PTR(-ENOMEM);
 }
 
 static void drop_space(struct config_group *g, struct config_item *i)
@@ -528,7 +528,7 @@ static struct config_item *make_comm(struct config_group *g, const char *name)
 
 	cm = kzalloc(sizeof(struct comm), GFP_KERNEL);
 	if (!cm)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	config_item_init_type_name(&cm->item, name, &comm_type);
 	cm->nodeid = -1;
@@ -561,7 +561,7 @@ static struct config_item *make_node(struct config_group *g, const char *name)
 
 	nd = kzalloc(sizeof(struct node), GFP_KERNEL);
 	if (!nd)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	config_item_init_type_name(&nd->item, name, &node_type);
 	nd->nodeid = -1;
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index f02ccb34604d..7dce1612553e 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -1493,24 +1493,18 @@ static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *g
 							  const char *name)
 {
 	struct o2hb_region *reg = NULL;
-	struct config_item *ret = NULL;
 
 	reg = kzalloc(sizeof(struct o2hb_region), GFP_KERNEL);
 	if (reg == NULL)
-		goto out; /* ENOMEM */
+		return ERR_PTR(-ENOMEM);
 
 	config_item_init_type_name(&reg->hr_item, name, &o2hb_region_type);
 
-	ret = &reg->hr_item;
-
 	spin_lock(&o2hb_live_lock);
 	list_add_tail(&reg->hr_all_item, &o2hb_all_regions);
 	spin_unlock(&o2hb_live_lock);
-out:
-	if (ret == NULL)
-		kfree(reg);
 
-	return ret;
+	return &reg->hr_item;
 }
 
 static void o2hb_heartbeat_group_drop_item(struct config_group *group,
diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c
index cfdb08b484ed..816a3f61330c 100644
--- a/fs/ocfs2/cluster/nodemanager.c
+++ b/fs/ocfs2/cluster/nodemanager.c
@@ -648,26 +648,19 @@ static struct config_item *o2nm_node_group_make_item(struct config_group *group,
 						     const char *name)
 {
 	struct o2nm_node *node = NULL;
-	struct config_item *ret = NULL;
 
 	if (strlen(name) > O2NM_MAX_NAME_LEN)
-		goto out; /* ENAMETOOLONG */
+		return ERR_PTR(-ENAMETOOLONG);
 
 	node = kzalloc(sizeof(struct o2nm_node), GFP_KERNEL);
 	if (node == NULL)
-		goto out; /* ENOMEM */
+		return ERR_PTR(-ENOMEM);
 
 	strcpy(node->nd_name, name); /* use item.ci_namebuf instead? */
 	config_item_init_type_name(&node->nd_item, name, &o2nm_node_type);
 	spin_lock_init(&node->nd_lock);
 
-	ret = &node->nd_item;
-
-out:
-	if (ret == NULL)
-		kfree(node);
-
-	return ret;
+	return &node->nd_item;
 }
 
 static void o2nm_node_group_drop_item(struct config_group *group,
@@ -762,7 +755,7 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g
 	/* this runs under the parent dir's i_mutex; there can be only
 	 * one caller in here at a time */
 	if (o2nm_single_cluster)
-		goto out; /* ENOSPC */
+		return ERR_PTR(-ENOSPC);
 
 	cluster = kzalloc(sizeof(struct o2nm_cluster), GFP_KERNEL);
 	ns = kzalloc(sizeof(struct o2nm_node_group), GFP_KERNEL);
@@ -795,6 +788,7 @@ out:
 		kfree(ns);
 		o2hb_free_hb_set(o2hb_group);
 		kfree(defs);
+		ret = ERR_PTR(-ENOMEM);
 	}
 
 	return ret;
diff --git a/include/linux/configfs.h b/include/linux/configfs.h
index 3ae65b1bf90f..d62c19ff041c 100644
--- a/include/linux/configfs.h
+++ b/include/linux/configfs.h
@@ -148,7 +148,8 @@ struct configfs_attribute {
  * items.  If the item is a group, it may support mkdir(2).
  * Groups supply one of make_group() and make_item().  If the
  * group supports make_group(), one can create group children.  If it
- * supports make_item(), one can create config_item children.  If it has
+ * supports make_item(), one can create config_item children.  make_group()
+ * and make_item() return ERR_PTR() on errors.  If it has
  * default_groups on group->default_groups, it has automatically created
  * group children.  default_groups may coexist alongsize make_group() or
  * make_item(), but if the group wishes to have only default_groups
-- 
cgit v1.2.3


From d2c52b7983b95bb3fc2a784e479f832f142d4523 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Thu, 17 Jul 2008 17:59:55 -0700
Subject: async_tx: export async_tx_quiesce

Replace open coded "wait and acknowledge" instances with async_tx_quiesce.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_memcpy.c | 10 +---------
 crypto/async_tx/async_memset.c | 10 +---------
 crypto/async_tx/async_tx.c     | 29 ++++++++++++++++++++---------
 crypto/async_tx/async_xor.c    | 37 ++-----------------------------------
 include/linux/async_tx.h       |  2 ++
 5 files changed, 26 insertions(+), 62 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/async_tx/async_memcpy.c b/crypto/async_tx/async_memcpy.c
index a5eda80e8427..06a7f4be9736 100644
--- a/crypto/async_tx/async_memcpy.c
+++ b/crypto/async_tx/async_memcpy.c
@@ -73,15 +73,7 @@ async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
 		pr_debug("%s: (sync) len: %zu\n", __func__, len);
 
 		/* wait for any prerequisite operations */
-		if (depend_tx) {
-			/* if ack is already set then we cannot be sure
-			 * we are referring to the correct operation
-			 */
-			BUG_ON(async_tx_test_ack(depend_tx));
-			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for depend_tx\n",
-					__func__);
-		}
+		async_tx_quiesce(&depend_tx);
 
 		dest_buf = kmap_atomic(dest, KM_USER0) + dest_offset;
 		src_buf = kmap_atomic(src, KM_USER1) + src_offset;
diff --git a/crypto/async_tx/async_memset.c b/crypto/async_tx/async_memset.c
index 27a97dc90a7e..d48ed22ed1c3 100644
--- a/crypto/async_tx/async_memset.c
+++ b/crypto/async_tx/async_memset.c
@@ -72,15 +72,7 @@ async_memset(struct page *dest, int val, unsigned int offset,
 		dest_buf = (void *) (((char *) page_address(dest)) + offset);
 
 		/* wait for any prerequisite operations */
-		if (depend_tx) {
-			/* if ack is already set then we cannot be sure
-			 * we are referring to the correct operation
-			 */
-			BUG_ON(async_tx_test_ack(depend_tx));
-			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for depend_tx\n",
-					__func__);
-		}
+		async_tx_quiesce(&depend_tx);
 
 		memset(dest_buf, val, len);
 
diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 9325c61208a0..78a61e7f631a 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -607,15 +607,7 @@ async_trigger_callback(enum async_tx_flags flags,
 		pr_debug("%s: (sync)\n", __func__);
 
 		/* wait for any prerequisite operations */
-		if (depend_tx) {
-			/* if ack is already set then we cannot be sure
-			 * we are referring to the correct operation
-			 */
-			BUG_ON(async_tx_test_ack(depend_tx));
-			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for depend_tx\n",
-					__func__);
-		}
+		async_tx_quiesce(&depend_tx);
 
 		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
 	}
@@ -624,6 +616,25 @@ async_trigger_callback(enum async_tx_flags flags,
 }
 EXPORT_SYMBOL_GPL(async_trigger_callback);
 
+/**
+ * async_tx_quiesce - ensure tx is complete and freeable upon return
+ * @tx - transaction to quiesce
+ */
+void async_tx_quiesce(struct dma_async_tx_descriptor **tx)
+{
+	if (*tx) {
+		/* if ack is already set then we cannot be sure
+		 * we are referring to the correct operation
+		 */
+		BUG_ON(async_tx_test_ack(*tx));
+		if (dma_wait_for_async_tx(*tx) == DMA_ERROR)
+			panic("DMA_ERROR waiting for transaction\n");
+		async_tx_ack(*tx);
+		*tx = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(async_tx_quiesce);
+
 module_init(async_tx_init);
 module_exit(async_tx_exit);
 
diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index 19d16e452bcc..689ecce73ee1 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -30,24 +30,6 @@
 #include <linux/raid/xor.h>
 #include <linux/async_tx.h>
 
-/**
- * async_tx_quiesce - ensure tx is complete and freeable upon return
- * @tx - transaction to quiesce
- */
-static void async_tx_quiesce(struct dma_async_tx_descriptor **tx)
-{
-	if (*tx) {
-		/* if ack is already set then we cannot be sure
-		 * we are referring to the correct operation
-		 */
-		BUG_ON(async_tx_test_ack(*tx));
-		if (dma_wait_for_async_tx(*tx) == DMA_ERROR)
-			panic("DMA_ERROR waiting for transaction\n");
-		async_tx_ack(*tx);
-		*tx = NULL;
-       }
-}
-
 /* do_async_xor - dma map the pages and perform the xor with an engine.
  * 	This routine is marked __always_inline so it can be compiled away
  * 	when CONFIG_DMA_ENGINE=n
@@ -219,15 +201,7 @@ async_xor(struct page *dest, struct page **src_list, unsigned int offset,
 		}
 
 		/* wait for any prerequisite operations */
-		if (depend_tx) {
-			/* if ack is already set then we cannot be sure
-			 * we are referring to the correct operation
-			 */
-			BUG_ON(async_tx_test_ack(depend_tx));
-			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for depend_tx\n",
-					__func__);
-		}
+		async_tx_quiesce(&depend_tx);
 
 		do_sync_xor(dest, src_list, offset, src_cnt, len,
 			    flags, depend_tx, cb_fn, cb_param);
@@ -309,17 +283,10 @@ async_xor_zero_sum(struct page *dest, struct page **src_list,
 		tx = async_xor(dest, src_list, offset, src_cnt, len, xor_flags,
 			depend_tx, NULL, NULL);
 
-		if (tx) {
-			if (dma_wait_for_async_tx(tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for tx\n",
-					__func__);
-			async_tx_ack(tx);
-		}
+		async_tx_quiesce(&tx);
 
 		*result = page_is_zero(dest, offset, len) ? 0 : 1;
 
-		tx = NULL;
-
 		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
 	}
 
diff --git a/include/linux/async_tx.h b/include/linux/async_tx.h
index eb640f0acfac..9f0e7bd5bdc9 100644
--- a/include/linux/async_tx.h
+++ b/include/linux/async_tx.h
@@ -152,4 +152,6 @@ struct dma_async_tx_descriptor *
 async_trigger_callback(enum async_tx_flags flags,
 	struct dma_async_tx_descriptor *depend_tx,
 	dma_async_tx_callback cb_fn, void *cb_fn_param);
+
+void async_tx_quiesce(struct dma_async_tx_descriptor **tx);
 #endif /* _ASYNC_TX_H_ */
-- 
cgit v1.2.3


From 3dce01713723bbcc92562bd4488e8b840a4f786c Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Thu, 17 Jul 2008 17:59:55 -0700
Subject: async_tx: remove depend_tx from async_tx_sync_epilog

All callers of async_tx_sync_epilog have called async_tx_quiesce on the
depend_tx, so async_tx_sync_epilog need only call the callback to
complete the operation.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_memcpy.c | 2 +-
 crypto/async_tx/async_memset.c | 2 +-
 crypto/async_tx/async_tx.c     | 2 +-
 crypto/async_tx/async_xor.c    | 7 +++----
 include/linux/async_tx.h       | 9 +--------
 5 files changed, 7 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/async_tx/async_memcpy.c b/crypto/async_tx/async_memcpy.c
index 06a7f4be9736..ddccfb01c416 100644
--- a/crypto/async_tx/async_memcpy.c
+++ b/crypto/async_tx/async_memcpy.c
@@ -83,7 +83,7 @@ async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
 		kunmap_atomic(dest_buf, KM_USER0);
 		kunmap_atomic(src_buf, KM_USER1);
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
diff --git a/crypto/async_tx/async_memset.c b/crypto/async_tx/async_memset.c
index d48ed22ed1c3..5b5eb99bb244 100644
--- a/crypto/async_tx/async_memset.c
+++ b/crypto/async_tx/async_memset.c
@@ -76,7 +76,7 @@ async_memset(struct page *dest, int val, unsigned int offset,
 
 		memset(dest_buf, val, len);
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 78a61e7f631a..35869a37a6f2 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -609,7 +609,7 @@ async_trigger_callback(enum async_tx_flags flags,
 		/* wait for any prerequisite operations */
 		async_tx_quiesce(&depend_tx);
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index 689ecce73ee1..65974c6d3d7a 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -121,7 +121,6 @@ do_async_xor(struct dma_chan *chan, struct page *dest, struct page **src_list,
 static void
 do_sync_xor(struct page *dest, struct page **src_list, unsigned int offset,
 	    int src_cnt, size_t len, enum async_tx_flags flags,
-	    struct dma_async_tx_descriptor *depend_tx,
 	    dma_async_tx_callback cb_fn, void *cb_param)
 {
 	int i;
@@ -150,7 +149,7 @@ do_sync_xor(struct page *dest, struct page **src_list, unsigned int offset,
 		src_off += xor_src_cnt;
 	}
 
-	async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+	async_tx_sync_epilog(cb_fn, cb_param);
 }
 
 /**
@@ -204,7 +203,7 @@ async_xor(struct page *dest, struct page **src_list, unsigned int offset,
 		async_tx_quiesce(&depend_tx);
 
 		do_sync_xor(dest, src_list, offset, src_cnt, len,
-			    flags, depend_tx, cb_fn, cb_param);
+			    flags, cb_fn, cb_param);
 
 		return NULL;
 	}
@@ -287,7 +286,7 @@ async_xor_zero_sum(struct page *dest, struct page **src_list,
 
 		*result = page_is_zero(dest, offset, len) ? 0 : 1;
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
diff --git a/include/linux/async_tx.h b/include/linux/async_tx.h
index 9f0e7bd5bdc9..0f50d4cc4360 100644
--- a/include/linux/async_tx.h
+++ b/include/linux/async_tx.h
@@ -101,21 +101,14 @@ async_tx_find_channel(struct dma_async_tx_descriptor *depend_tx,
 
 /**
  * async_tx_sync_epilog - actions to take if an operation is run synchronously
- * @flags: async_tx flags
- * @depend_tx: transaction depends on depend_tx
  * @cb_fn: function to call when the transaction completes
  * @cb_fn_param: parameter to pass to the callback routine
  */
 static inline void
-async_tx_sync_epilog(unsigned long flags,
-	struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_fn_param)
+async_tx_sync_epilog(dma_async_tx_callback cb_fn, void *cb_fn_param)
 {
 	if (cb_fn)
 		cb_fn(cb_fn_param);
-
-	if (depend_tx && (flags & ASYNC_TX_DEP_ACK))
-		async_tx_ack(depend_tx);
 }
 
 void
-- 
cgit v1.2.3


From 0839875e0c197ded56bbae820e699f26d6fa2697 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Thu, 17 Jul 2008 17:59:56 -0700
Subject: async_tx: make async_tx_test_ack a boolean routine

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 include/linux/dmaengine.h | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 9b91d341e1fa..adb0b084eb5a 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -375,16 +375,14 @@ dma_cookie_t dma_async_memcpy_pg_to_pg(struct dma_chan *chan,
 void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx,
 	struct dma_chan *chan);
 
-static inline void
-async_tx_ack(struct dma_async_tx_descriptor *tx)
+static inline void async_tx_ack(struct dma_async_tx_descriptor *tx)
 {
 	tx->flags |= DMA_CTRL_ACK;
 }
 
-static inline int
-async_tx_test_ack(struct dma_async_tx_descriptor *tx)
+static inline bool async_tx_test_ack(struct dma_async_tx_descriptor *tx)
 {
-	return tx->flags & DMA_CTRL_ACK;
+	return (tx->flags & DMA_CTRL_ACK) == DMA_CTRL_ACK;
 }
 
 #define first_dma_cap(mask) __first_dma_cap(&(mask))
-- 
cgit v1.2.3


From e8a0464cc950972824e2e128028ae3db666ec1ed Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 00:34:19 -0700
Subject: netdev: Allocate multiple queues for TX.

alloc_netdev_mq() now allocates an array of netdev_queue
structures for TX, based upon the queue_count argument.

Furthermore, all accesses to the TX queues are now vectored
through the netdev_get_tx_queue() and netdev_for_each_tx_queue()
interfaces.  This makes it easy to grep the tree for all
things that want to get to a TX queue of a net device.

Problem spots which are not really multiqueue aware yet, and
only work with one queue, can easily be spotted by grepping
for all netdev_get_tx_queue() calls that pass in a zero index.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bonding/bond_main.c         |   6 +-
 drivers/net/hamradio/bpqether.c         |   6 +-
 drivers/net/ifb.c                       |  12 ++-
 drivers/net/macvlan.c                   |   6 +-
 drivers/net/wireless/hostap/hostap_hw.c |   6 +-
 include/linux/netdevice.h               |  69 ++++++++-----
 include/net/sch_generic.h               |  37 +++++--
 net/8021q/vlan_dev.c                    |  10 +-
 net/core/dev.c                          |  40 +++++--
 net/core/rtnetlink.c                    |   2 +-
 net/mac80211/main.c                     |   4 +-
 net/mac80211/wme.c                      |  12 +--
 net/netrom/af_netrom.c                  |   6 +-
 net/rose/af_rose.c                      |   6 +-
 net/sched/cls_api.c                     |   4 +-
 net/sched/sch_api.c                     |  32 ++++--
 net/sched/sch_generic.c                 | 178 +++++++++++++++++++++++---------
 net/sched/sch_teql.c                    |  21 ++--
 18 files changed, 320 insertions(+), 137 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index fd87dbe7999a..9737c06045d6 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5042,7 +5042,9 @@ static int bond_check_params(struct bond_params *params)
 
 static struct lock_class_key bonding_netdev_xmit_lock_key;
 
-static void bond_set_lockdep_class_one(struct netdev_queue *txq)
+static void bond_set_lockdep_class_one(struct net_device *dev,
+				       struct netdev_queue *txq,
+				       void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock,
 			  &bonding_netdev_xmit_lock_key);
@@ -5050,7 +5052,7 @@ static void bond_set_lockdep_class_one(struct netdev_queue *txq)
 
 static void bond_set_lockdep_class(struct net_device *dev)
 {
-	bond_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, bond_set_lockdep_class_one, NULL);
 }
 
 /* Create a new bond based on the specified name and bonding parameters.
diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c
index fb186b8c3d4d..b6500b2aacf2 100644
--- a/drivers/net/hamradio/bpqether.c
+++ b/drivers/net/hamradio/bpqether.c
@@ -124,14 +124,16 @@ static LIST_HEAD(bpq_devices);
  */
 static struct lock_class_key bpq_netdev_xmit_lock_key;
 
-static void bpq_set_lockdep_class_one(struct netdev_queue *txq)
+static void bpq_set_lockdep_class_one(struct net_device *dev,
+				      struct netdev_queue *txq,
+				      void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock, &bpq_netdev_xmit_lock_key);
 }
 
 static void bpq_set_lockdep_class(struct net_device *dev)
 {
-	bpq_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, bpq_set_lockdep_class_one, NULL);
 }
 
 /* ------------------------------------------------------------------------ */
diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index ccbd6554f6eb..897b05e79ed0 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -229,14 +229,20 @@ module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
 /*
- * dev_ifb->tx_queue.lock is usually taken after dev->rx_queue.lock,
+ * dev_ifb's TX queue lock is usually taken after dev->rx_queue.lock,
  * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev->tx_queue.lock with dev_ifb->rx_queue.lock.
+ * ifb doesn't take dev's TX queue lock with dev_ifb->rx_queue.lock.
  * But lockdep should know that ifb has different locks from dev.
  */
 static struct lock_class_key ifb_tx_queue_lock_key;
 static struct lock_class_key ifb_rx_queue_lock_key;
 
+static void set_tx_lockdep_key(struct net_device *dev,
+			       struct netdev_queue *txq,
+			       void *_unused)
+{
+	lockdep_set_class(&txq->lock, &ifb_tx_queue_lock_key);
+}
 
 static int __init ifb_init_one(int index)
 {
@@ -258,7 +264,7 @@ static int __init ifb_init_one(int index)
 	if (err < 0)
 		goto err;
 
-	lockdep_set_class(&dev_ifb->tx_queue.lock, &ifb_tx_queue_lock_key);
+	netdev_for_each_tx_queue(dev_ifb, set_tx_lockdep_key, NULL);
 	lockdep_set_class(&dev_ifb->rx_queue.lock, &ifb_rx_queue_lock_key);
 
 	return 0;
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 980001c2cf96..72745ce588c6 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -285,7 +285,9 @@ static struct lock_class_key macvlan_netdev_xmit_lock_key;
 #define MACVLAN_STATE_MASK \
 	((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT))
 
-static void macvlan_set_lockdep_class_one(struct netdev_queue *txq)
+static void macvlan_set_lockdep_class_one(struct net_device *dev,
+					  struct netdev_queue *txq,
+					  void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock,
 			  &macvlan_netdev_xmit_lock_key);
@@ -293,7 +295,7 @@ static void macvlan_set_lockdep_class_one(struct netdev_queue *txq)
 
 static void macvlan_set_lockdep_class(struct net_device *dev)
 {
-	macvlan_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, macvlan_set_lockdep_class_one, NULL);
 }
 
 static int macvlan_init(struct net_device *dev)
diff --git a/drivers/net/wireless/hostap/hostap_hw.c b/drivers/net/wireless/hostap/hostap_hw.c
index c1f4bb005d92..13d5882f1f21 100644
--- a/drivers/net/wireless/hostap/hostap_hw.c
+++ b/drivers/net/wireless/hostap/hostap_hw.c
@@ -3102,7 +3102,9 @@ static void prism2_clear_set_tim_queue(local_info_t *local)
  */
 static struct lock_class_key hostap_netdev_xmit_lock_key;
 
-static void prism2_set_lockdep_class_one(struct netdev_queue *txq)
+static void prism2_set_lockdep_class_one(struct net_device *dev,
+					 struct netdev_queue *txq,
+					 void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock,
 			  &hostap_netdev_xmit_lock_key);
@@ -3110,7 +3112,7 @@ static void prism2_set_lockdep_class_one(struct netdev_queue *txq)
 
 static void prism2_set_lockdep_class(struct net_device *dev)
 {
-	prism2_set_lockdep_class_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, prism2_set_lockdep_class_one, NULL);
 }
 
 static struct net_device *
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 570cf7affa72..f25d4f5a31b0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -463,7 +463,7 @@ struct netdev_queue {
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
 	struct netdev_queue	*next_sched;
-};
+} ____cacheline_aligned_in_smp;
 
 /*
  *	The DEVICE structure.
@@ -641,7 +641,9 @@ struct net_device
 	unsigned char		broadcast[MAX_ADDR_LEN];	/* hw bcast add	*/
 
 	struct netdev_queue	rx_queue;
-	struct netdev_queue	tx_queue ____cacheline_aligned_in_smp;
+
+	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
+	unsigned int		num_tx_queues;
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 /*
@@ -764,6 +766,25 @@ struct net_device
 #define	NETDEV_ALIGN		32
 #define	NETDEV_ALIGN_CONST	(NETDEV_ALIGN - 1)
 
+static inline
+struct netdev_queue *netdev_get_tx_queue(const struct net_device *dev,
+					 unsigned int index)
+{
+	return &dev->_tx[index];
+}
+
+static inline void netdev_for_each_tx_queue(struct net_device *dev,
+					    void (*f)(struct net_device *,
+						      struct netdev_queue *,
+						      void *),
+					    void *arg)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++)
+		f(dev, &dev->_tx[i], arg);
+}
+
 /*
  * Net namespace inlines
  */
@@ -977,7 +998,7 @@ static inline void netif_schedule_queue(struct netdev_queue *txq)
 
 static inline void netif_schedule(struct net_device *dev)
 {
-	netif_schedule_queue(&dev->tx_queue);
+	netif_schedule_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -993,7 +1014,7 @@ static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
 
 static inline void netif_start_queue(struct net_device *dev)
 {
-	netif_tx_start_queue(&dev->tx_queue);
+	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1017,7 +1038,7 @@ static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 
 static inline void netif_wake_queue(struct net_device *dev)
 {
-	netif_tx_wake_queue(&dev->tx_queue);
+	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1034,7 +1055,7 @@ static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
 
 static inline void netif_stop_queue(struct net_device *dev)
 {
-	netif_tx_stop_queue(&dev->tx_queue);
+	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1050,7 +1071,7 @@ static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
 
 static inline int netif_queue_stopped(const struct net_device *dev)
 {
-	return netif_tx_queue_stopped(&dev->tx_queue);
+	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1134,7 +1155,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 #endif
 	if (test_and_clear_bit(__QUEUE_STATE_XOFF,
 			       &dev->egress_subqueue[queue_index].state))
-		__netif_schedule(&dev->tx_queue);
+		__netif_schedule(netdev_get_tx_queue(dev, 0));
 }
 
 /**
@@ -1430,18 +1451,19 @@ static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 
 static inline void netif_tx_lock(struct net_device *dev)
 {
-	__netif_tx_lock(&dev->tx_queue, smp_processor_id());
-}
+	int cpu = smp_processor_id();
+	unsigned int i;
 
-static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
-{
-	spin_lock_bh(&txq->_xmit_lock);
-	txq->xmit_lock_owner = smp_processor_id();
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		__netif_tx_lock(txq, cpu);
+	}
 }
 
 static inline void netif_tx_lock_bh(struct net_device *dev)
 {
-	__netif_tx_lock_bh(&dev->tx_queue);
+	local_bh_disable();
+	netif_tx_lock(dev);
 }
 
 static inline int __netif_tx_trylock(struct netdev_queue *txq)
@@ -1454,7 +1476,7 @@ static inline int __netif_tx_trylock(struct netdev_queue *txq)
 
 static inline int netif_tx_trylock(struct net_device *dev)
 {
-	return __netif_tx_trylock(&dev->tx_queue);
+	return __netif_tx_trylock(netdev_get_tx_queue(dev, 0));
 }
 
 static inline void __netif_tx_unlock(struct netdev_queue *txq)
@@ -1465,18 +1487,19 @@ static inline void __netif_tx_unlock(struct netdev_queue *txq)
 
 static inline void netif_tx_unlock(struct net_device *dev)
 {
-	__netif_tx_unlock(&dev->tx_queue);
-}
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		__netif_tx_unlock(txq);
+	}
 
-static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
-{
-	txq->xmit_lock_owner = -1;
-	spin_unlock_bh(&txq->_xmit_lock);
 }
 
 static inline void netif_tx_unlock_bh(struct net_device *dev)
 {
-	__netif_tx_unlock_bh(&dev->tx_queue);
+	netif_tx_unlock(dev);
+	local_bh_enable();
 }
 
 #define HARD_TX_LOCK(dev, txq, cpu) {			\
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 5ba66b555578..b47f556c66f8 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -230,32 +230,47 @@ extern void tcf_destroy_chain(struct tcf_proto **fl);
 /* Reset all TX qdiscs of a device.  */
 static inline void qdisc_reset_all_tx(struct net_device *dev)
 {
-	qdisc_reset(dev->tx_queue.qdisc);
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++)
+		qdisc_reset(netdev_get_tx_queue(dev, i)->qdisc);
 }
 
 /* Are all TX queues of the device empty?  */
 static inline bool qdisc_all_tx_empty(const struct net_device *dev)
 {
-	const struct netdev_queue *txq = &dev->tx_queue;
-	const struct Qdisc *q = txq->qdisc;
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		const struct Qdisc *q = txq->qdisc;
 
-	return (q->q.qlen == 0);
+		if (q->q.qlen)
+			return false;
+	}
+	return true;
 }
 
 /* Are any of the TX qdiscs changing?  */
 static inline bool qdisc_tx_changing(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
-
-	return (txq->qdisc != txq->qdisc_sleeping);
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		if (txq->qdisc != txq->qdisc_sleeping)
+			return true;
+	}
+	return false;
 }
 
-/* Is the device using the noop qdisc?  */
+/* Is the device using the noop qdisc on all queues?  */
 static inline bool qdisc_tx_is_noop(const struct net_device *dev)
 {
-	const struct netdev_queue *txq = &dev->tx_queue;
-
-	return (txq->qdisc == &noop_qdisc);
+	unsigned int i;
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		if (txq->qdisc != &noop_qdisc)
+			return false;
+	}
+	return true;
 }
 
 static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 6b985f23fd9f..f42bc2b26b85 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -570,16 +570,18 @@ static void vlan_dev_set_rx_mode(struct net_device *vlan_dev)
  */
 static struct lock_class_key vlan_netdev_xmit_lock_key;
 
-static void vlan_dev_set_lockdep_one(struct netdev_queue *txq,
-				     int subclass)
+static void vlan_dev_set_lockdep_one(struct net_device *dev,
+				     struct netdev_queue *txq,
+				     void *_subclass)
 {
 	lockdep_set_class_and_subclass(&txq->_xmit_lock,
-				       &vlan_netdev_xmit_lock_key, subclass);
+				       &vlan_netdev_xmit_lock_key,
+				       *(int *)_subclass);
 }
 
 static void vlan_dev_set_lockdep_class(struct net_device *dev, int subclass)
 {
-	vlan_dev_set_lockdep_one(&dev->tx_queue, subclass);
+	netdev_for_each_tx_queue(dev, vlan_dev_set_lockdep_one, &subclass);
 }
 
 static const struct header_ops vlan_header_ops = {
diff --git a/net/core/dev.c b/net/core/dev.c
index 9b49f74a9820..69378f250695 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1666,6 +1666,12 @@ out_kfree_skb:
  *          --BLG
  */
 
+static struct netdev_queue *dev_pick_tx(struct net_device *dev,
+					struct sk_buff *skb)
+{
+	return netdev_get_tx_queue(dev, 0);
+}
+
 int dev_queue_xmit(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
@@ -1702,7 +1708,7 @@ int dev_queue_xmit(struct sk_buff *skb)
 	}
 
 gso:
-	txq = &dev->tx_queue;
+	txq = dev_pick_tx(dev, skb);
 	spin_lock_prefetch(&txq->lock);
 
 	/* Disable soft irqs for various locks below. Also
@@ -3788,8 +3794,9 @@ static void rollback_registered(struct net_device *dev)
 	dev_put(dev);
 }
 
-static void __netdev_init_queue_locks_one(struct netdev_queue *dev_queue,
-					  struct net_device *dev)
+static void __netdev_init_queue_locks_one(struct net_device *dev,
+					  struct netdev_queue *dev_queue,
+					  void *_unused)
 {
 	spin_lock_init(&dev_queue->_xmit_lock);
 	netdev_set_lockdep_class(&dev_queue->_xmit_lock, dev->type);
@@ -3798,8 +3805,8 @@ static void __netdev_init_queue_locks_one(struct netdev_queue *dev_queue,
 
 static void netdev_init_queue_locks(struct net_device *dev)
 {
-	__netdev_init_queue_locks_one(&dev->tx_queue, dev);
-	__netdev_init_queue_locks_one(&dev->rx_queue, dev);
+	netdev_for_each_tx_queue(dev, __netdev_init_queue_locks_one, NULL);
+	__netdev_init_queue_locks_one(dev, &dev->rx_queue, NULL);
 }
 
 /**
@@ -4119,7 +4126,8 @@ static struct net_device_stats *internal_stats(struct net_device *dev)
 }
 
 static void netdev_init_one_queue(struct net_device *dev,
-				  struct netdev_queue *queue)
+				  struct netdev_queue *queue,
+				  void *_unused)
 {
 	spin_lock_init(&queue->lock);
 	queue->dev = dev;
@@ -4127,8 +4135,8 @@ static void netdev_init_one_queue(struct net_device *dev,
 
 static void netdev_init_queues(struct net_device *dev)
 {
-	netdev_init_one_queue(dev, &dev->rx_queue);
-	netdev_init_one_queue(dev, &dev->tx_queue);
+	netdev_init_one_queue(dev, &dev->rx_queue, NULL);
+	netdev_for_each_tx_queue(dev, netdev_init_one_queue, NULL);
 }
 
 /**
@@ -4145,9 +4153,10 @@ static void netdev_init_queues(struct net_device *dev)
 struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 		void (*setup)(struct net_device *), unsigned int queue_count)
 {
-	void *p;
+	struct netdev_queue *tx;
 	struct net_device *dev;
 	int alloc_size;
+	void *p;
 
 	BUG_ON(strlen(name) >= sizeof(dev->name));
 
@@ -4167,11 +4176,22 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 		return NULL;
 	}
 
+	tx = kzalloc(sizeof(struct netdev_queue) * queue_count, GFP_KERNEL);
+	if (!tx) {
+		printk(KERN_ERR "alloc_netdev: Unable to allocate "
+		       "tx qdiscs.\n");
+		kfree(p);
+		return NULL;
+	}
+
 	dev = (struct net_device *)
 		(((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST);
 	dev->padded = (char *)dev - (char *)p;
 	dev_net_set(dev, &init_net);
 
+	dev->_tx = tx;
+	dev->num_tx_queues = queue_count;
+
 	if (sizeof_priv) {
 		dev->priv = ((char *)dev +
 			     ((sizeof(struct net_device) +
@@ -4205,6 +4225,8 @@ void free_netdev(struct net_device *dev)
 {
 	release_net(dev_net(dev));
 
+	kfree(dev->_tx);
+
 	/*  Compatibility with error handling in drivers */
 	if (dev->reg_state == NETREG_UNINITIALIZED) {
 		kfree((char *)dev - dev->padded);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 8ef9f1db610e..71edb8b36341 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -636,7 +636,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
 	if (dev->master)
 		NLA_PUT_U32(skb, IFLA_MASTER, dev->master->ifindex);
 
-	txq = &dev->tx_queue;
+	txq = netdev_get_tx_queue(dev, 0);
 	if (txq->qdisc_sleeping)
 		NLA_PUT_STRING(skb, IFLA_QDISC, txq->qdisc_sleeping->ops->id);
 
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index af0056e7e5b3..b486e634f4fe 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -621,7 +621,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid)
 
 	/* ensure that TX flow won't interrupt us
 	 * until the end of the call to requeue function */
-	txq = &local->mdev->tx_queue;
+	txq = netdev_get_tx_queue(local->mdev, 0);
 	spin_lock_bh(&txq->lock);
 
 	/* create a new queue for this aggregation */
@@ -862,7 +862,7 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid)
 
 	/* avoid ordering issues: we are the only one that can modify
 	 * the content of the qdiscs */
-	txq = &local->mdev->tx_queue;
+	txq = netdev_get_tx_queue(local->mdev, 0);
 	spin_lock_bh(&txq->lock);
 	/* remove the queue for this aggregation */
 	ieee80211_ht_agg_queue_remove(local, sta, tid, 1);
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index 6ae43a3c7726..f014cd38c2d0 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -574,7 +574,7 @@ static struct Qdisc_ops wme_qdisc_ops __read_mostly =
 
 void ieee80211_install_qdisc(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
 	struct Qdisc *qdisc;
 
 	qdisc = qdisc_create_dflt(dev, txq,
@@ -596,7 +596,7 @@ void ieee80211_install_qdisc(struct net_device *dev)
 
 int ieee80211_qdisc_installed(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
 
 	return txq->qdisc_sleeping->ops == &wme_qdisc_ops;
 }
@@ -617,7 +617,7 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 			struct sta_info *sta, u16 tid)
 {
 	int i;
-	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
 	struct ieee80211_sched_data *q =
 			qdisc_priv(txq->qdisc_sleeping);
 	DECLARE_MAC_BUF(mac);
@@ -652,14 +652,14 @@ int ieee80211_ht_agg_queue_add(struct ieee80211_local *local,
 }
 
 /**
- * the caller needs to hold local->mdev->tx_queue.lock
+ * the caller needs to hold netdev_get_tx_queue(local->mdev, X)->lock
  */
 void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 				   struct sta_info *sta, u16 tid,
 				   u8 requeue)
 {
 	struct ieee80211_hw *hw = &local->hw;
-	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
 	struct ieee80211_sched_data *q =
 		qdisc_priv(txq->qdisc_sleeping);
 	int agg_queue = sta->tid_to_tx_q[tid];
@@ -676,7 +676,7 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 
 void ieee80211_requeue(struct ieee80211_local *local, int queue)
 {
-	struct netdev_queue *txq = &local->mdev->tx_queue;
+	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, 0);
 	struct Qdisc *root_qd = txq->qdisc_sleeping;
 	struct ieee80211_sched_data *q = qdisc_priv(root_qd);
 	struct Qdisc *qdisc = q->queues[queue];
diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 819afc449e1e..d41be0d66eb0 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -74,14 +74,16 @@ static const struct proto_ops nr_proto_ops;
  */
 static struct lock_class_key nr_netdev_xmit_lock_key;
 
-static void nr_set_lockdep_one(struct netdev_queue *txq)
+static void nr_set_lockdep_one(struct net_device *dev,
+			       struct netdev_queue *txq,
+			       void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock, &nr_netdev_xmit_lock_key);
 }
 
 static void nr_set_lockdep_key(struct net_device *dev)
 {
-	nr_set_lockdep_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, nr_set_lockdep_one, NULL);
 }
 
 /*
diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index 7dbbc0891623..f3a691f34909 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c
@@ -75,14 +75,16 @@ ax25_address rose_callsign;
  */
 static struct lock_class_key rose_netdev_xmit_lock_key;
 
-static void rose_set_lockdep_one(struct netdev_queue *txq)
+static void rose_set_lockdep_one(struct net_device *dev,
+				 struct netdev_queue *txq,
+				 void *_unused)
 {
 	lockdep_set_class(&txq->_xmit_lock, &rose_netdev_xmit_lock_key);
 }
 
 static void rose_set_lockdep_key(struct net_device *dev)
 {
-	rose_set_lockdep_one(&dev->tx_queue);
+	netdev_for_each_tx_queue(dev, rose_set_lockdep_one, NULL);
 }
 
 /*
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index b483bbea6118..d0b0a9b14394 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -166,7 +166,7 @@ replay:
 
 	/* Find qdisc */
 	if (!parent) {
-		struct netdev_queue *dev_queue = &dev->tx_queue;
+		struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, 0);
 		q = dev_queue->qdisc_sleeping;
 		parent = q->handle;
 	} else {
@@ -410,7 +410,7 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 	if ((dev = dev_get_by_index(&init_net, tcm->tcm_ifindex)) == NULL)
 		return skb->len;
 
-	dev_queue = &dev->tx_queue;
+	dev_queue = netdev_get_tx_queue(dev, 0);
 	if (!tcm->tcm_parent)
 		q = dev_queue->qdisc_sleeping;
 	else
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 95873f8dd37c..830ccc544a15 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -183,9 +183,8 @@ EXPORT_SYMBOL(unregister_qdisc);
    (root qdisc, all its children, children of children etc.)
  */
 
-struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+static struct Qdisc *__qdisc_lookup(struct netdev_queue *dev_queue, u32 handle)
 {
-	struct netdev_queue *dev_queue = &dev->tx_queue;
 	struct Qdisc *q;
 
 	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
@@ -195,6 +194,19 @@ struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 	return NULL;
 }
 
+struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		struct Qdisc *q = __qdisc_lookup(txq, handle);
+		if (q)
+			return q;
+	}
+	return NULL;
+}
+
 static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
 {
 	unsigned long cl;
@@ -462,7 +474,7 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 		}
 
 	} else {
-		dev_queue = &dev->tx_queue;
+		dev_queue = netdev_get_tx_queue(dev, 0);
 		oqdisc = dev_queue->qdisc_sleeping;
 
 		/* Prune old scheduler */
@@ -742,7 +754,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 				q = dev->rx_queue.qdisc;
 			}
 		} else {
-			struct netdev_queue *dev_queue = &dev->tx_queue;
+			struct netdev_queue *dev_queue;
+			dev_queue = netdev_get_tx_queue(dev, 0);
 			q = dev_queue->qdisc_sleeping;
 		}
 		if (!q)
@@ -817,7 +830,8 @@ replay:
 				q = dev->rx_queue.qdisc;
 			}
 		} else {
-			struct netdev_queue *dev_queue = &dev->tx_queue;
+			struct netdev_queue *dev_queue;
+			dev_queue = netdev_get_tx_queue(dev, 0);
 			q = dev_queue->qdisc_sleeping;
 		}
 
@@ -899,7 +913,7 @@ create_n_graft:
 				 tcm->tcm_parent, tcm->tcm_parent,
 				 tca, &err);
 	else
-		q = qdisc_create(dev, &dev->tx_queue,
+		q = qdisc_create(dev, netdev_get_tx_queue(dev, 0),
 				 tcm->tcm_parent, tcm->tcm_handle,
 				 tca, &err);
 	if (q == NULL) {
@@ -1025,7 +1039,7 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		dev_queue = &dev->tx_queue;
+		dev_queue = netdev_get_tx_queue(dev, 0);
 		list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 			if (q_idx < s_q_idx) {
 				q_idx++;
@@ -1098,7 +1112,7 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg)
 
 	/* Step 1. Determine qdisc handle X:0 */
 
-	dev_queue = &dev->tx_queue;
+	dev_queue = netdev_get_tx_queue(dev, 0);
 	if (pid != TC_H_ROOT) {
 		u32 qid1 = TC_H_MAJ(pid);
 
@@ -1275,7 +1289,7 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	dev_queue = &dev->tx_queue;
+	dev_queue = netdev_get_tx_queue(dev, 0);
 	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
 		if (t < s_t || !q->ops->cl_ops ||
 		    (tcm->tcm_parent &&
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 243de935b182..4e2b865cbba0 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -40,20 +40,30 @@
  */
 
 void qdisc_lock_tree(struct net_device *dev)
-	__acquires(dev->tx_queue.lock)
 	__acquires(dev->rx_queue.lock)
 {
-	spin_lock_bh(&dev->tx_queue.lock);
+	unsigned int i;
+
+	local_bh_disable();
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		spin_lock(&txq->lock);
+	}
 	spin_lock(&dev->rx_queue.lock);
 }
 EXPORT_SYMBOL(qdisc_lock_tree);
 
 void qdisc_unlock_tree(struct net_device *dev)
 	__releases(dev->rx_queue.lock)
-	__releases(dev->tx_queue.lock)
 {
+	unsigned int i;
+
 	spin_unlock(&dev->rx_queue.lock);
-	spin_unlock_bh(&dev->tx_queue.lock);
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		spin_unlock(&txq->lock);
+	}
+	local_bh_enable();
 }
 EXPORT_SYMBOL(qdisc_unlock_tree);
 
@@ -212,22 +222,37 @@ void __qdisc_run(struct netdev_queue *txq)
 static void dev_watchdog(unsigned long arg)
 {
 	struct net_device *dev = (struct net_device *)arg;
-	struct netdev_queue *txq = &dev->tx_queue;
 
 	netif_tx_lock(dev);
-	if (txq->qdisc != &noop_qdisc) {
+	if (!qdisc_tx_is_noop(dev)) {
 		if (netif_device_present(dev) &&
 		    netif_running(dev) &&
 		    netif_carrier_ok(dev)) {
-			if (netif_queue_stopped(dev) &&
-			    time_after(jiffies, dev->trans_start + dev->watchdog_timeo)) {
+			int some_queue_stopped = 0;
+			unsigned int i;
+
+			for (i = 0; i < dev->num_tx_queues; i++) {
+				struct netdev_queue *txq;
+
+				txq = netdev_get_tx_queue(dev, i);
+				if (netif_tx_queue_stopped(txq)) {
+					some_queue_stopped = 1;
+					break;
+				}
+			}
 
-				printk(KERN_INFO "NETDEV WATCHDOG: %s: transmit timed out\n",
+			if (some_queue_stopped &&
+			    time_after(jiffies, (dev->trans_start +
+						 dev->watchdog_timeo))) {
+				printk(KERN_INFO "NETDEV WATCHDOG: %s: "
+				       "transmit timed out\n",
 				       dev->name);
 				dev->tx_timeout(dev);
 				WARN_ON_ONCE(1);
 			}
-			if (!mod_timer(&dev->watchdog_timer, round_jiffies(jiffies + dev->watchdog_timeo)))
+			if (!mod_timer(&dev->watchdog_timer,
+				       round_jiffies(jiffies +
+						     dev->watchdog_timeo)))
 				dev_hold(dev);
 		}
 	}
@@ -542,9 +567,55 @@ void qdisc_destroy(struct Qdisc *qdisc)
 }
 EXPORT_SYMBOL(qdisc_destroy);
 
+static bool dev_all_qdisc_sleeping_noop(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+
+		if (txq->qdisc_sleeping != &noop_qdisc)
+			return false;
+	}
+	return true;
+}
+
+static void attach_one_default_qdisc(struct net_device *dev,
+				     struct netdev_queue *dev_queue,
+				     void *_unused)
+{
+	struct Qdisc *qdisc;
+
+	if (dev->tx_queue_len) {
+		qdisc = qdisc_create_dflt(dev, dev_queue,
+					  &pfifo_fast_ops, TC_H_ROOT);
+		if (!qdisc) {
+			printk(KERN_INFO "%s: activation failed\n", dev->name);
+			return;
+		}
+		list_add_tail(&qdisc->list, &dev_queue->qdisc_list);
+	} else {
+		qdisc =  &noqueue_qdisc;
+	}
+	dev_queue->qdisc_sleeping = qdisc;
+}
+
+static void transition_one_qdisc(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 void *_need_watchdog)
+{
+	int *need_watchdog_p = _need_watchdog;
+
+	spin_lock_bh(&dev_queue->lock);
+	rcu_assign_pointer(dev_queue->qdisc, dev_queue->qdisc_sleeping);
+	if (dev_queue->qdisc != &noqueue_qdisc)
+		*need_watchdog_p = 1;
+	spin_unlock_bh(&dev_queue->lock);
+}
+
 void dev_activate(struct net_device *dev)
 {
-	struct netdev_queue *txq = &dev->tx_queue;
+	int need_watchdog;
 
 	/* No queueing discipline is attached to device;
 	   create default one i.e. pfifo_fast for devices,
@@ -552,39 +623,27 @@ void dev_activate(struct net_device *dev)
 	   virtual interfaces
 	 */
 
-	if (txq->qdisc_sleeping == &noop_qdisc) {
-		struct Qdisc *qdisc;
-		if (dev->tx_queue_len) {
-			qdisc = qdisc_create_dflt(dev, txq,
-						  &pfifo_fast_ops,
-						  TC_H_ROOT);
-			if (qdisc == NULL) {
-				printk(KERN_INFO "%s: activation failed\n", dev->name);
-				return;
-			}
-			list_add_tail(&qdisc->list, &txq->qdisc_list);
-		} else {
-			qdisc =  &noqueue_qdisc;
-		}
-		txq->qdisc_sleeping = qdisc;
-	}
+	if (dev_all_qdisc_sleeping_noop(dev))
+		netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL);
 
 	if (!netif_carrier_ok(dev))
 		/* Delay activation until next carrier-on event */
 		return;
 
-	spin_lock_bh(&txq->lock);
-	rcu_assign_pointer(txq->qdisc, txq->qdisc_sleeping);
-	if (txq->qdisc != &noqueue_qdisc) {
+	need_watchdog = 0;
+	netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog);
+
+	if (need_watchdog) {
 		dev->trans_start = jiffies;
 		dev_watchdog_up(dev);
 	}
-	spin_unlock_bh(&txq->lock);
 }
 
-static void dev_deactivate_queue(struct netdev_queue *dev_queue,
-				 struct Qdisc *qdisc_default)
+static void dev_deactivate_queue(struct net_device *dev,
+				 struct netdev_queue *dev_queue,
+				 void *_qdisc_default)
 {
+	struct Qdisc *qdisc_default = _qdisc_default;
 	struct Qdisc *qdisc;
 	struct sk_buff *skb;
 
@@ -603,12 +662,35 @@ static void dev_deactivate_queue(struct netdev_queue *dev_queue,
 	kfree_skb(skb);
 }
 
+static bool some_qdisc_is_running(struct net_device *dev, int lock)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *dev_queue;
+		int val;
+
+		dev_queue = netdev_get_tx_queue(dev, i);
+
+		if (lock)
+			spin_lock_bh(&dev_queue->lock);
+
+		val = test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state);
+
+		if (lock)
+			spin_unlock_bh(&dev_queue->lock);
+
+		if (val)
+			return true;
+	}
+	return false;
+}
+
 void dev_deactivate(struct net_device *dev)
 {
-	struct netdev_queue *dev_queue = &dev->tx_queue;
-	int running;
+	bool running;
 
-	dev_deactivate_queue(dev_queue, &noop_qdisc);
+	netdev_for_each_tx_queue(dev, dev_deactivate_queue, &noop_qdisc);
 
 	dev_watchdog_down(dev);
 
@@ -617,17 +699,14 @@ void dev_deactivate(struct net_device *dev)
 
 	/* Wait for outstanding qdisc_run calls. */
 	do {
-		while (test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state))
+		while (some_qdisc_is_running(dev, 0))
 			yield();
 
 		/*
 		 * Double-check inside queue lock to ensure that all effects
 		 * of the queue run are visible when we return.
 		 */
-		spin_lock_bh(&dev_queue->lock);
-		running = test_bit(__QUEUE_STATE_QDISC_RUNNING,
-				   &dev_queue->state);
-		spin_unlock_bh(&dev_queue->lock);
+		running = some_qdisc_is_running(dev, 1);
 
 		/*
 		 * The running flag should never be set at this point because
@@ -642,8 +721,10 @@ void dev_deactivate(struct net_device *dev)
 
 static void dev_init_scheduler_queue(struct net_device *dev,
 				     struct netdev_queue *dev_queue,
-				     struct Qdisc *qdisc)
+				     void *_qdisc)
 {
+	struct Qdisc *qdisc = _qdisc;
+
 	dev_queue->qdisc = qdisc;
 	dev_queue->qdisc_sleeping = qdisc;
 	INIT_LIST_HEAD(&dev_queue->qdisc_list);
@@ -652,18 +733,19 @@ static void dev_init_scheduler_queue(struct net_device *dev,
 void dev_init_scheduler(struct net_device *dev)
 {
 	qdisc_lock_tree(dev);
-	dev_init_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
+	netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc);
 	dev_init_scheduler_queue(dev, &dev->rx_queue, NULL);
 	qdisc_unlock_tree(dev);
 
 	setup_timer(&dev->watchdog_timer, dev_watchdog, (unsigned long)dev);
 }
 
-static void dev_shutdown_scheduler_queue(struct net_device *dev,
-					 struct netdev_queue *dev_queue,
-					 struct Qdisc *qdisc_default)
+static void shutdown_scheduler_queue(struct net_device *dev,
+				     struct netdev_queue *dev_queue,
+				     void *_qdisc_default)
 {
 	struct Qdisc *qdisc = dev_queue->qdisc_sleeping;
+	struct Qdisc *qdisc_default = _qdisc_default;
 
 	if (qdisc) {
 		dev_queue->qdisc = qdisc_default;
@@ -676,8 +758,8 @@ static void dev_shutdown_scheduler_queue(struct net_device *dev,
 void dev_shutdown(struct net_device *dev)
 {
 	qdisc_lock_tree(dev);
-	dev_shutdown_scheduler_queue(dev, &dev->tx_queue, &noop_qdisc);
-	dev_shutdown_scheduler_queue(dev, &dev->rx_queue, NULL);
+	netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
+	shutdown_scheduler_queue(dev, &dev->rx_queue, NULL);
 	BUG_TRAP(!timer_pending(&dev->watchdog_timer));
 	qdisc_unlock_tree(dev);
 }
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 8ac05981be20..44a2c3451f4d 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -111,7 +111,7 @@ teql_dequeue(struct Qdisc* sch)
 	struct sk_buff *skb;
 
 	skb = __skb_dequeue(&dat->q);
-	dat_queue = &dat->m->dev->tx_queue;
+	dat_queue = netdev_get_tx_queue(dat->m->dev, 0);
 	if (skb == NULL) {
 		struct net_device *m = qdisc_dev(dat_queue->qdisc);
 		if (m) {
@@ -155,10 +155,13 @@ teql_destroy(struct Qdisc* sch)
 				if (q == master->slaves) {
 					master->slaves = NEXT_SLAVE(q);
 					if (q == master->slaves) {
+						struct netdev_queue *txq;
+
+						txq = netdev_get_tx_queue(master->dev, 0);
 						master->slaves = NULL;
-						spin_lock_bh(&master->dev->tx_queue.lock);
-						qdisc_reset(master->dev->tx_queue.qdisc);
-						spin_unlock_bh(&master->dev->tx_queue.lock);
+						spin_lock_bh(&txq->lock);
+						qdisc_reset(txq->qdisc);
+						spin_unlock_bh(&txq->lock);
 					}
 				}
 				skb_queue_purge(&dat->q);
@@ -218,7 +221,8 @@ static int teql_qdisc_init(struct Qdisc *sch, struct nlattr *opt)
 static int
 __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *dev)
 {
-	struct teql_sched_data *q = qdisc_priv(dev->tx_queue.qdisc);
+	struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, 0);
+	struct teql_sched_data *q = qdisc_priv(dev_queue->qdisc);
 	struct neighbour *mn = skb->dst->neighbour;
 	struct neighbour *n = q->ncache;
 
@@ -254,7 +258,8 @@ __teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *
 static inline int teql_resolve(struct sk_buff *skb,
 			       struct sk_buff *skb_res, struct net_device *dev)
 {
-	if (dev->tx_queue.qdisc == &noop_qdisc)
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
+	if (txq->qdisc == &noop_qdisc)
 		return -ENODEV;
 
 	if (dev->header_ops == NULL ||
@@ -285,8 +290,10 @@ restart:
 
 	do {
 		struct net_device *slave = qdisc_dev(q);
+		struct netdev_queue *slave_txq;
 
-		if (slave->tx_queue.qdisc_sleeping != q)
+		slave_txq = netdev_get_tx_queue(slave, 0);
+		if (slave_txq->qdisc_sleeping != q)
 			continue;
 		if (netif_queue_stopped(slave) ||
 		    __netif_subqueue_stopped(slave, subq) ||
-- 
cgit v1.2.3


From 09e83b5d7d1878065e2453239b49b684cd0fe4e5 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 01:52:12 -0700
Subject: netdev: Kill NETIF_F_MULTI_QUEUE.

There is no need for a feature bit for something that
can be tested by simply checking the TX queue count.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/multiqueue.txt | 11 -----------
 drivers/net/cpmac.c                     |  1 -
 drivers/net/igb/igb_main.c              |  2 --
 drivers/net/ixgbe/ixgbe_main.c          |  2 --
 drivers/net/s2io.c                      |  2 --
 include/linux/netdevice.h               |  4 +---
 net/mac80211/main.c                     |  3 ---
 7 files changed, 1 insertion(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/networking/multiqueue.txt b/Documentation/networking/multiqueue.txt
index e6dc1ee9e8f1..d391ea631141 100644
--- a/Documentation/networking/multiqueue.txt
+++ b/Documentation/networking/multiqueue.txt
@@ -24,15 +24,4 @@ netif_{start|stop|wake}_subqueue() functions to manage each queue while the
 device is still operational.  netdev->queue_lock is still used when the device
 comes online or when it's completely shut down (unregister_netdev(), etc.).
 
-Finally, the base driver should indicate that it is a multiqueue device.  The
-feature flag NETIF_F_MULTI_QUEUE should be added to the netdev->features
-bitmap on device initialization.  Below is an example from e1000:
-
-#ifdef CONFIG_E1000_MQ
-	if ( (adapter->hw.mac.type == e1000_82571) ||
-	     (adapter->hw.mac.type == e1000_82572) ||
-	     (adapter->hw.mac.type == e1000_80003es2lan))
-		netdev->features |= NETIF_F_MULTI_QUEUE;
-#endif
-
 Author: Peter P. Waskiewicz Jr. <peter.p.waskiewicz.jr@intel.com>
diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index d630e2a72f42..7c7b54e4828e 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -1165,7 +1165,6 @@ static int __devinit cpmac_probe(struct platform_device *pdev)
 	dev->set_multicast_list = cpmac_set_multicast_list;
 	dev->tx_timeout         = cpmac_tx_timeout;
 	dev->ethtool_ops        = &cpmac_ethtool_ops;
-	dev->features |= NETIF_F_MULTI_QUEUE;
 
 	netif_napi_add(dev, &priv->napi, cpmac_poll, 64);
 
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 64a150a16f39..471c194cd54e 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -1155,8 +1155,6 @@ static int __devinit igb_probe(struct pci_dev *pdev,
 	if (pci_using_dac)
 		netdev->features |= NETIF_F_HIGHDMA;
 
-	netdev->features |= NETIF_F_MULTI_QUEUE;
-
 	netdev->features |= NETIF_F_LLTX;
 	adapter->en_mng_pt = igb_enable_mng_pass_thru(&adapter->hw);
 
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 7d8bf94d3783..e6df9233f5ef 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -3566,8 +3566,6 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	if (pci_using_dac)
 		netdev->features |= NETIF_F_HIGHDMA;
 
-	netdev->features |= NETIF_F_MULTI_QUEUE;
-
 	/* make sure the EEPROM is good */
 	if (ixgbe_validate_eeprom_checksum(hw, NULL) < 0) {
 		dev_err(&pdev->dev, "The EEPROM Checksum Is Not Valid\n");
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 517425dcb77c..5f0fcb04afff 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -7966,8 +7966,6 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre)
 		dev->features |= NETIF_F_UFO;
 		dev->features |= NETIF_F_HW_CSUM;
 	}
-	if (config->multiq)
-		dev->features |= NETIF_F_MULTI_QUEUE;
 	dev->tx_timeout = &s2io_tx_watchdog;
 	dev->watchdog_timeo = WATCH_DOG_TIMEOUT;
 	INIT_WORK(&sp->rst_timer_task, s2io_restart_nic);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f25d4f5a31b0..c02227b9dd7b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -533,7 +533,6 @@ struct net_device
 #define NETIF_F_LLTX		4096	/* LockLess TX - deprecated. Please */
 					/* do not use LLTX in new drivers */
 #define NETIF_F_NETNS_LOCAL	8192	/* Does not change network namespaces */
-#define NETIF_F_MULTI_QUEUE	16384	/* Has multiple TX/RX queues */
 #define NETIF_F_LRO		32768	/* large receive offload */
 
 	/* Segmentation offload features */
@@ -1163,11 +1162,10 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
  *	@dev: network device
  *
  * Check if device has multiple transmit queues
- * Always falls if NETDEVICE_MULTIQUEUE is not configured
  */
 static inline int netif_is_multiqueue(const struct net_device *dev)
 {
-	return (!!(NETIF_F_MULTI_QUEUE & dev->features));
+	return (dev->num_tx_queues > 1);
 }
 
 /* Use this variant when it is known for sure that it
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index b486e634f4fe..c74607eda1ee 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1678,9 +1678,6 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	if (!mdev)
 		goto fail_mdev_alloc;
 
-	if (ieee80211_num_queues(hw) > 1)
-		mdev->features |= NETIF_F_MULTI_QUEUE;
-
 	mwdev = netdev_priv(mdev);
 	mdev->ieee80211_ptr = mwdev;
 	mwdev->wiphy = local->hw.wiphy;
-- 
cgit v1.2.3


From 1d8ae3fdeb001b8f534a6782c261aba6ec1779f5 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 02:52:19 -0700
Subject: pkt_sched: Remove RR scheduler.

This actually fixes a bug added by the RR scheduler changes.  The
->bands and ->prio2band parameters were being set outside of the
sch_tree_lock() and thus could result in strange behavior and
inconsistencies.

It might be possible, in the new design (where there will be one qdisc
per device TX queue) to allow similar functionality via a TX hash
algorithm for RR but I really see no reason to export this aspect of
how these multiqueue cards actually implement the scheduling of the
the individual DMA TX rings and the single physical MAC/PHY port.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_sched.h |   9 ---
 net/sched/sch_prio.c      | 136 ++++++----------------------------------------
 2 files changed, 16 insertions(+), 129 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index dbb7ac37960d..87f4e0fa8f27 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -103,15 +103,6 @@ struct tc_prio_qopt
 	__u8	priomap[TC_PRIO_MAX+1];	/* Map: logical priority -> PRIO band */
 };
 
-enum
-{
-	TCA_PRIO_UNSPEC,
-	TCA_PRIO_MQ,
-	__TCA_PRIO_MAX
-};
-
-#define TCA_PRIO_MAX    (__TCA_PRIO_MAX - 1)
-
 /* TBF section */
 
 struct tc_tbf_qopt
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index 39157f7bc046..536ca474dc69 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -24,11 +24,9 @@
 struct prio_sched_data
 {
 	int bands;
-	int curband; /* for round-robin */
 	struct tcf_proto *filter_list;
 	u8  prio2band[TC_PRIO_MAX+1];
 	struct Qdisc *queues[TCQ_PRIO_BANDS];
-	int mq;
 };
 
 
@@ -55,17 +53,14 @@ prio_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr)
 		if (!q->filter_list || err < 0) {
 			if (TC_H_MAJ(band))
 				band = 0;
-			band = q->prio2band[band&TC_PRIO_MAX];
-			goto out;
+			return q->queues[q->prio2band[band&TC_PRIO_MAX]];
 		}
 		band = res.classid;
 	}
 	band = TC_H_MIN(band) - 1;
 	if (band >= q->bands)
-		band = q->prio2band[0];
-out:
-	if (q->mq)
-		skb_set_queue_mapping(skb, band);
+		return q->queues[q->prio2band[0]];
+
 	return q->queues[band];
 }
 
@@ -123,68 +118,23 @@ prio_requeue(struct sk_buff *skb, struct Qdisc* sch)
 }
 
 
-static struct sk_buff *
-prio_dequeue(struct Qdisc* sch)
+static struct sk_buff *prio_dequeue(struct Qdisc* sch)
 {
-	struct sk_buff *skb;
 	struct prio_sched_data *q = qdisc_priv(sch);
 	int prio;
-	struct Qdisc *qdisc;
 
 	for (prio = 0; prio < q->bands; prio++) {
-		/* Check if the target subqueue is available before
-		 * pulling an skb.  This way we avoid excessive requeues
-		 * for slower queues.
-		 */
-		if (!__netif_subqueue_stopped(qdisc_dev(sch),
-					      (q->mq ? prio : 0))) {
-			qdisc = q->queues[prio];
-			skb = qdisc->dequeue(qdisc);
-			if (skb) {
-				sch->q.qlen--;
-				return skb;
-			}
+		struct Qdisc *qdisc = q->queues[prio];
+		struct sk_buff *skb = qdisc->dequeue(qdisc);
+		if (skb) {
+			sch->q.qlen--;
+			return skb;
 		}
 	}
 	return NULL;
 
 }
 
-static struct sk_buff *rr_dequeue(struct Qdisc* sch)
-{
-	struct sk_buff *skb;
-	struct prio_sched_data *q = qdisc_priv(sch);
-	struct Qdisc *qdisc;
-	int bandcount;
-
-	/* Only take one pass through the queues.  If nothing is available,
-	 * return nothing.
-	 */
-	for (bandcount = 0; bandcount < q->bands; bandcount++) {
-		/* Check if the target subqueue is available before
-		 * pulling an skb.  This way we avoid excessive requeues
-		 * for slower queues.  If the queue is stopped, try the
-		 * next queue.
-		 */
-		if (!__netif_subqueue_stopped(qdisc_dev(sch),
-					      (q->mq ? q->curband : 0))) {
-			qdisc = q->queues[q->curband];
-			skb = qdisc->dequeue(qdisc);
-			if (skb) {
-				sch->q.qlen--;
-				q->curband++;
-				if (q->curband >= q->bands)
-					q->curband = 0;
-				return skb;
-			}
-		}
-		q->curband++;
-		if (q->curband >= q->bands)
-			q->curband = 0;
-	}
-	return NULL;
-}
-
 static unsigned int prio_drop(struct Qdisc* sch)
 {
 	struct prio_sched_data *q = qdisc_priv(sch);
@@ -229,45 +179,22 @@ static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
 {
 	struct prio_sched_data *q = qdisc_priv(sch);
 	struct tc_prio_qopt *qopt;
-	struct nlattr *tb[TCA_PRIO_MAX + 1];
-	int err;
 	int i;
 
-	err = nla_parse_nested_compat(tb, TCA_PRIO_MAX, opt, NULL, qopt,
-				      sizeof(*qopt));
-	if (err < 0)
-		return err;
-
-	q->bands = qopt->bands;
-	/* If we're multiqueue, make sure the number of incoming bands
-	 * matches the number of queues on the device we're associating with.
-	 * If the number of bands requested is zero, then set q->bands to
-	 * dev->egress_subqueue_count.  Also, the root qdisc must be the
-	 * only one that is enabled for multiqueue, since it's the only one
-	 * that interacts with the underlying device.
-	 */
-	q->mq = nla_get_flag(tb[TCA_PRIO_MQ]);
-	if (q->mq) {
-		if (sch->parent != TC_H_ROOT)
-			return -EINVAL;
-		if (netif_is_multiqueue(qdisc_dev(sch))) {
-			if (q->bands == 0)
-				q->bands = qdisc_dev(sch)->egress_subqueue_count;
-			else if (q->bands != qdisc_dev(sch)->egress_subqueue_count)
-				return -EINVAL;
-		} else
-			return -EOPNOTSUPP;
-	}
+	if (nla_len(opt) < sizeof(*qopt))
+		return -EINVAL;
+	qopt = nla_data(opt);
 
-	if (q->bands > TCQ_PRIO_BANDS || q->bands < 2)
+	if (qopt->bands > TCQ_PRIO_BANDS || qopt->bands < 2)
 		return -EINVAL;
 
 	for (i=0; i<=TC_PRIO_MAX; i++) {
-		if (qopt->priomap[i] >= q->bands)
+		if (qopt->priomap[i] >= qopt->bands)
 			return -EINVAL;
 	}
 
 	sch_tree_lock(sch);
+	q->bands = qopt->bands;
 	memcpy(q->prio2band, qopt->priomap, TC_PRIO_MAX+1);
 
 	for (i=q->bands; i<TCQ_PRIO_BANDS; i++) {
@@ -333,10 +260,6 @@ static int prio_dump(struct Qdisc *sch, struct sk_buff *skb)
 	nest = nla_nest_compat_start(skb, TCA_OPTIONS, sizeof(opt), &opt);
 	if (nest == NULL)
 		goto nla_put_failure;
-	if (q->mq) {
-		if (nla_put_flag(skb, TCA_PRIO_MQ) < 0)
-			goto nla_put_failure;
-	}
 	nla_nest_compat_end(skb, nest);
 
 	return skb->len;
@@ -509,44 +432,17 @@ static struct Qdisc_ops prio_qdisc_ops __read_mostly = {
 	.owner		=	THIS_MODULE,
 };
 
-static struct Qdisc_ops rr_qdisc_ops __read_mostly = {
-	.next		=	NULL,
-	.cl_ops		=	&prio_class_ops,
-	.id		=	"rr",
-	.priv_size	=	sizeof(struct prio_sched_data),
-	.enqueue	=	prio_enqueue,
-	.dequeue	=	rr_dequeue,
-	.requeue	=	prio_requeue,
-	.drop		=	prio_drop,
-	.init		=	prio_init,
-	.reset		=	prio_reset,
-	.destroy	=	prio_destroy,
-	.change		=	prio_tune,
-	.dump		=	prio_dump,
-	.owner		=	THIS_MODULE,
-};
-
 static int __init prio_module_init(void)
 {
-	int err;
-
-	err = register_qdisc(&prio_qdisc_ops);
-	if (err < 0)
-		return err;
-	err = register_qdisc(&rr_qdisc_ops);
-	if (err < 0)
-		unregister_qdisc(&prio_qdisc_ops);
-	return err;
+	return register_qdisc(&prio_qdisc_ops);
 }
 
 static void __exit prio_module_exit(void)
 {
 	unregister_qdisc(&prio_qdisc_ops);
-	unregister_qdisc(&rr_qdisc_ops);
 }
 
 module_init(prio_module_init)
 module_exit(prio_module_exit)
 
 MODULE_LICENSE("GPL");
-MODULE_ALIAS("sch_rr");
-- 
cgit v1.2.3


From fd2ea0a79faad824258af5dcec1927aa24d81c16 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 01:56:23 -0700
Subject: net: Use queue aware tests throughout.

This effectively "flips the switch" by making the core networking
and multiqueue-aware drivers use the new TX multiqueue structures.

Non-multiqueue drivers need no changes.  The interfaces they use such
as netif_stop_queue() degenerate into an operation on TX queue zero.
So everything "just works" for them.

Code that really wants to do "X" to all TX queues now invokes a
routine that does so, such as netif_tx_wake_all_queues(),
netif_tx_stop_all_queues(), etc.

pktgen and netpoll required a little bit more surgery than the others.

In particular the pktgen changes, whilst functional, could be largely
improved.  The initial check in pktgen_xmit() will sometimes check the
wrong queue, which is mostly harmless.  The thing to do is probably to
invoke fill_packet() earlier.

The bulk of the netpoll changes is to make the code operate solely on
the TX queue indicated by by the SKB queue mapping.

Setting of the SKB queue mapping is entirely confined inside of
net/core/dev.c:dev_pick_tx().  If we end up needing any kind of
special semantics (drops, for example) it will be implemented here.

Finally, we now have a "real_num_tx_queues" which is where the driver
indicates how many TX queues are actually active.

With IGB changes from Jeff Kirsher.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/cpmac.c               | 20 ++++------
 drivers/net/igb/igb_main.c        | 19 +++------
 drivers/net/ixgbe/ixgbe_ethtool.c | 10 +----
 drivers/net/ixgbe/ixgbe_main.c    | 15 +++----
 drivers/net/s2io.c                | 48 +++++++++--------------
 include/linux/netdevice.h         | 82 ++++++++++++++++++++++++++++++++++-----
 include/net/pkt_sched.h           |  4 +-
 net/core/dev.c                    | 28 ++++++-------
 net/core/netpoll.c                | 24 +++++++-----
 net/core/pktgen.c                 | 69 ++++++++++++++++++++------------
 net/sched/sch_generic.c           |  5 +--
 net/sched/sch_teql.c              |  6 +--
 12 files changed, 187 insertions(+), 143 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index 7c7b54e4828e..fbd4280c102c 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -544,7 +544,7 @@ fatal_error:
 
 	spin_unlock(&priv->rx_lock);
 	netif_rx_complete(priv->dev, napi);
-	netif_stop_queue(priv->dev);
+	netif_tx_stop_all_queues(priv->dev);
 	napi_disable(&priv->napi);
 
 	atomic_inc(&priv->reset_pending);
@@ -750,9 +750,7 @@ static void cpmac_hw_error(struct work_struct *work)
 	barrier();
 	atomic_dec(&priv->reset_pending);
 
-	for (i = 0; i < CPMAC_QUEUES; i++)
-		netif_wake_subqueue(priv->dev, i);
-	netif_wake_queue(priv->dev);
+	netif_tx_wake_all_queues(priv->dev);
 	cpmac_write(priv->regs, CPMAC_MAC_INT_ENABLE, 3);
 }
 
@@ -781,7 +779,7 @@ static void cpmac_check_status(struct net_device *dev)
 				     dev->name, tx_code, tx_channel, macstatus);
 		}
 
-		netif_stop_queue(dev);
+		netif_tx_stop_all_queues(dev);
 		cpmac_hw_stop(dev);
 		if (schedule_work(&priv->reset_work))
 			atomic_inc(&priv->reset_pending);
@@ -842,9 +840,7 @@ static void cpmac_tx_timeout(struct net_device *dev)
 	barrier();
 	atomic_dec(&priv->reset_pending);
 
-	netif_wake_queue(priv->dev);
-	for (i = 0; i < CPMAC_QUEUES; i++)
-		netif_wake_subqueue(dev, i);
+	netif_tx_wake_all_queues(priv->dev);
 }
 
 static int cpmac_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
@@ -935,7 +931,7 @@ static void cpmac_adjust_link(struct net_device *dev)
 
 	spin_lock(&priv->lock);
 	if (priv->phy->link) {
-		netif_start_queue(dev);
+		netif_tx_start_all_queues(dev);
 		if (priv->phy->duplex != priv->oldduplex) {
 			new_state = 1;
 			priv->oldduplex = priv->phy->duplex;
@@ -949,10 +945,10 @@ static void cpmac_adjust_link(struct net_device *dev)
 		if (!priv->oldlink) {
 			new_state = 1;
 			priv->oldlink = 1;
-			netif_schedule(dev);
+			netif_tx_schedule_all(dev);
 		}
 	} else if (priv->oldlink) {
-		netif_stop_queue(dev);
+		netif_tx_stop_all_queues(dev);
 		new_state = 1;
 		priv->oldlink = 0;
 		priv->oldspeed = 0;
@@ -1072,7 +1068,7 @@ static int cpmac_stop(struct net_device *dev)
 	struct cpmac_priv *priv = netdev_priv(dev);
 	struct resource *mem;
 
-	netif_stop_queue(dev);
+	netif_tx_stop_all_queues(dev);
 
 	cancel_work_sync(&priv->reset_work);
 	napi_disable(&priv->napi);
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 471c194cd54e..81bba6983dde 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -533,7 +533,7 @@ msi_only:
 		adapter->flags |= IGB_FLAG_HAS_MSI;
 
 	/* Notify the stack of the (possibly) reduced Tx Queue count. */
-	adapter->netdev->egress_subqueue_count = adapter->num_tx_queues;
+	adapter->netdev->real_num_tx_queues = adapter->num_tx_queues;
 	return;
 }
 
@@ -821,9 +821,7 @@ void igb_down(struct igb_adapter *adapter)
 	wr32(E1000_RCTL, rctl & ~E1000_RCTL_EN);
 	/* flush and sleep below */
 
-	netif_stop_queue(netdev);
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		netif_stop_subqueue(netdev, i);
+	netif_tx_stop_all_queues(netdev);
 
 	/* disable transmits in the hardware */
 	tctl = rd32(E1000_TCTL);
@@ -1266,9 +1264,7 @@ static int __devinit igb_probe(struct pci_dev *pdev,
 
 	/* tell the stack to leave us alone until igb_open() is called */
 	netif_carrier_off(netdev);
-	netif_stop_queue(netdev);
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		netif_stop_subqueue(netdev, i);
+	netif_tx_stop_all_queues(netdev);
 
 	strcpy(netdev->name, "eth%d");
 	err = register_netdev(netdev);
@@ -2315,7 +2311,6 @@ static void igb_watchdog_task(struct work_struct *work)
 	struct e1000_mac_info *mac = &adapter->hw.mac;
 	u32 link;
 	s32 ret_val;
-	int i;
 
 	if ((netif_carrier_ok(netdev)) &&
 	    (rd32(E1000_STATUS) & E1000_STATUS_LU))
@@ -2371,9 +2366,7 @@ static void igb_watchdog_task(struct work_struct *work)
 			}
 
 			netif_carrier_on(netdev);
-			netif_wake_queue(netdev);
-			for (i = 0; i < adapter->num_tx_queues; i++)
-				netif_wake_subqueue(netdev, i);
+			netif_tx_wake_all_queues(netdev);
 
 			if (!test_bit(__IGB_DOWN, &adapter->state))
 				mod_timer(&adapter->phy_info_timer,
@@ -2385,9 +2378,7 @@ static void igb_watchdog_task(struct work_struct *work)
 			adapter->link_duplex = 0;
 			dev_info(&adapter->pdev->dev, "NIC Link is Down\n");
 			netif_carrier_off(netdev);
-			netif_stop_queue(netdev);
-			for (i = 0; i < adapter->num_tx_queues; i++)
-				netif_stop_subqueue(netdev, i);
+			netif_tx_stop_all_queues(netdev);
 			if (!test_bit(__IGB_DOWN, &adapter->state))
 				mod_timer(&adapter->phy_info_timer,
 					  round_jiffies(jiffies + 2 * HZ));
diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index 81b769093d22..3efe5dda10af 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -252,16 +252,10 @@ static int ixgbe_set_tso(struct net_device *netdev, u32 data)
 		netdev->features |= NETIF_F_TSO;
 		netdev->features |= NETIF_F_TSO6;
 	} else {
-		struct ixgbe_adapter *adapter = netdev_priv(netdev);
-		int i;
-		netif_stop_queue(netdev);
-		for (i = 0; i < adapter->num_tx_queues; i++)
-			netif_stop_subqueue(netdev, i);
+		netif_tx_stop_all_queues(netdev);
 		netdev->features &= ~NETIF_F_TSO;
 		netdev->features &= ~NETIF_F_TSO6;
-		for (i = 0; i < adapter->num_tx_queues; i++)
-			netif_start_subqueue(netdev, i);
-		netif_start_queue(netdev);
+		netif_tx_start_all_queues(netdev);
 	}
 	return 0;
 }
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index e6df9233f5ef..6af8fb5c4b5f 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -2013,7 +2013,7 @@ void ixgbe_down(struct ixgbe_adapter *adapter)
 	del_timer_sync(&adapter->watchdog_timer);
 
 	netif_carrier_off(netdev);
-	netif_stop_queue(netdev);
+	netif_tx_stop_all_queues(netdev);
 
 	if (!pci_channel_offline(adapter->pdev))
 		ixgbe_reset(adapter);
@@ -2359,7 +2359,7 @@ try_msi:
 
 out:
 	/* Notify the stack of the (possibly) reduced Tx Queue count. */
-	adapter->netdev->egress_subqueue_count = adapter->num_tx_queues;
+	adapter->netdev->real_num_tx_queues = adapter->num_tx_queues;
 
 	return err;
 }
@@ -2896,7 +2896,6 @@ static void ixgbe_watchdog(unsigned long data)
 	struct net_device *netdev = adapter->netdev;
 	bool link_up;
 	u32 link_speed = 0;
-	int i;
 
 	adapter->hw.mac.ops.check_link(&adapter->hw, &(link_speed), &link_up);
 
@@ -2917,9 +2916,7 @@ static void ixgbe_watchdog(unsigned long data)
 				 (FLOW_TX ? "TX" : "None"))));
 
 			netif_carrier_on(netdev);
-			netif_wake_queue(netdev);
-			for (i = 0; i < adapter->num_tx_queues; i++)
-				netif_wake_subqueue(netdev, i);
+			netif_tx_wake_all_queues(netdev);
 		} else {
 			/* Force detection of hung controller */
 			adapter->detect_tx_hung = true;
@@ -2928,7 +2925,7 @@ static void ixgbe_watchdog(unsigned long data)
 		if (netif_carrier_ok(netdev)) {
 			DPRINTK(LINK, INFO, "NIC Link is Down\n");
 			netif_carrier_off(netdev);
-			netif_stop_queue(netdev);
+			netif_tx_stop_all_queues(netdev);
 		}
 	}
 
@@ -3631,9 +3628,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	ixgbe_start_hw(hw);
 
 	netif_carrier_off(netdev);
-	netif_stop_queue(netdev);
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		netif_stop_subqueue(netdev, i);
+	netif_tx_stop_all_queues(netdev);
 
 	ixgbe_napi_add_all(adapter);
 
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 5f0fcb04afff..9dae40ccf048 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -545,63 +545,53 @@ static struct pci_driver s2io_driver = {
 /* netqueue manipulation helper functions */
 static inline void s2io_stop_all_tx_queue(struct s2io_nic *sp)
 {
-	int i;
-	if (sp->config.multiq) {
-		for (i = 0; i < sp->config.tx_fifo_num; i++)
-			netif_stop_subqueue(sp->dev, i);
-	} else {
+	if (!sp->config.multiq) {
+		int i;
+
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_STOP;
-		netif_stop_queue(sp->dev);
 	}
+	netif_tx_stop_all_queues(sp->dev);
 }
 
 static inline void s2io_stop_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-	if (sp->config.multiq)
-		netif_stop_subqueue(sp->dev, fifo_no);
-	else {
+	if (!sp->config.multiq)
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_STOP;
-		netif_stop_queue(sp->dev);
-	}
+
+	netif_tx_stop_all_queues(sp->dev);
 }
 
 static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
 {
-	int i;
-	if (sp->config.multiq) {
-		for (i = 0; i < sp->config.tx_fifo_num; i++)
-			netif_start_subqueue(sp->dev, i);
-	} else {
+	if (!sp->config.multiq) {
+		int i;
+
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
-		netif_start_queue(sp->dev);
 	}
+	netif_tx_start_all_queues(sp->dev);
 }
 
 static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
 {
-	if (sp->config.multiq)
-		netif_start_subqueue(sp->dev, fifo_no);
-	else {
+	if (!sp->config.multiq)
 		sp->mac_control.fifos[fifo_no].queue_state =
 			FIFO_QUEUE_START;
-		netif_start_queue(sp->dev);
-	}
+
+	netif_tx_start_all_queues(sp->dev);
 }
 
 static inline void s2io_wake_all_tx_queue(struct s2io_nic *sp)
 {
-	int i;
-	if (sp->config.multiq) {
-		for (i = 0; i < sp->config.tx_fifo_num; i++)
-			netif_wake_subqueue(sp->dev, i);
-	} else {
+	if (!sp->config.multiq) {
+		int i;
+
 		for (i = 0; i < sp->config.tx_fifo_num; i++)
 			sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
-		netif_wake_queue(sp->dev);
 	}
+	netif_tx_wake_all_queues(sp->dev);
 }
 
 static inline void s2io_wake_tx_queue(
@@ -8691,5 +8681,5 @@ static void s2io_io_resume(struct pci_dev *pdev)
 	}
 
 	netif_device_attach(netdev);
-	netif_wake_queue(netdev);
+	netif_tx_wake_all_queues(netdev);
 }
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c02227b9dd7b..b5c1e7df64fc 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -642,7 +642,13 @@ struct net_device
 	struct netdev_queue	rx_queue;
 
 	struct netdev_queue	*_tx ____cacheline_aligned_in_smp;
+
+	/* Number of TX queues allocated at alloc_netdev_mq() time  */
 	unsigned int		num_tx_queues;
+
+	/* Number of TX queues currently active in device  */
+	unsigned int		real_num_tx_queues;
+
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
 
 /*
@@ -1000,6 +1006,14 @@ static inline void netif_schedule(struct net_device *dev)
 	netif_schedule_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_schedule_all(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++)
+		netif_schedule_queue(netdev_get_tx_queue(dev, i));
+}
+
 /**
  *	netif_start_queue - allow transmit
  *	@dev: network device
@@ -1016,6 +1030,16 @@ static inline void netif_start_queue(struct net_device *dev)
 	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_start_all_queues(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_start_queue(txq);
+	}
+}
+
 /**
  *	netif_wake_queue - restart transmit
  *	@dev: network device
@@ -1040,6 +1064,16 @@ static inline void netif_wake_queue(struct net_device *dev)
 	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_wake_all_queues(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_wake_queue(txq);
+	}
+}
+
 /**
  *	netif_stop_queue - stop transmitted packets
  *	@dev: network device
@@ -1057,6 +1091,16 @@ static inline void netif_stop_queue(struct net_device *dev)
 	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
 }
 
+static inline void netif_tx_stop_all_queues(struct net_device *dev)
+{
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_stop_queue(txq);
+	}
+}
+
 /**
  *	netif_queue_stopped - test if transmit queue is flowblocked
  *	@dev: network device
@@ -1100,7 +1144,8 @@ static inline int netif_running(const struct net_device *dev)
  */
 static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
 {
-	clear_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
+	clear_bit(__QUEUE_STATE_XOFF, &txq->state);
 }
 
 /**
@@ -1112,11 +1157,12 @@ static inline void netif_start_subqueue(struct net_device *dev, u16 queue_index)
  */
 static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 {
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
 #endif
-	set_bit(__QUEUE_STATE_XOFF, &dev->egress_subqueue[queue_index].state);
+	set_bit(__QUEUE_STATE_XOFF, &txq->state);
 }
 
 /**
@@ -1129,8 +1175,8 @@ static inline void netif_stop_subqueue(struct net_device *dev, u16 queue_index)
 static inline int __netif_subqueue_stopped(const struct net_device *dev,
 					 u16 queue_index)
 {
-	return test_bit(__QUEUE_STATE_XOFF,
-			&dev->egress_subqueue[queue_index].state);
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
+	return test_bit(__QUEUE_STATE_XOFF, &txq->state);
 }
 
 static inline int netif_subqueue_stopped(const struct net_device *dev,
@@ -1148,13 +1194,13 @@ static inline int netif_subqueue_stopped(const struct net_device *dev,
  */
 static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 {
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
 #ifdef CONFIG_NETPOLL_TRAP
 	if (netpoll_trap())
 		return;
 #endif
-	if (test_and_clear_bit(__QUEUE_STATE_XOFF,
-			       &dev->egress_subqueue[queue_index].state))
-		__netif_schedule(netdev_get_tx_queue(dev, 0));
+	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &txq->state))
+		__netif_schedule(txq);
 }
 
 /**
@@ -1198,7 +1244,8 @@ extern int		dev_set_mtu(struct net_device *, int);
 extern int		dev_set_mac_address(struct net_device *,
 					    struct sockaddr *);
 extern int		dev_hard_start_xmit(struct sk_buff *skb,
-					    struct net_device *dev);
+					    struct net_device *dev,
+					    struct netdev_queue *txq);
 
 extern int		netdev_budget;
 
@@ -1447,6 +1494,12 @@ static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 	txq->xmit_lock_owner = cpu;
 }
 
+static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
+{
+	spin_lock_bh(&txq->_xmit_lock);
+	txq->xmit_lock_owner = smp_processor_id();
+}
+
 static inline void netif_tx_lock(struct net_device *dev)
 {
 	int cpu = smp_processor_id();
@@ -1483,6 +1536,12 @@ static inline void __netif_tx_unlock(struct netdev_queue *txq)
 	spin_unlock(&txq->_xmit_lock);
 }
 
+static inline void __netif_tx_unlock_bh(struct netdev_queue *txq)
+{
+	txq->xmit_lock_owner = -1;
+	spin_unlock_bh(&txq->_xmit_lock);
+}
+
 static inline void netif_tx_unlock(struct net_device *dev)
 {
 	unsigned int i;
@@ -1514,8 +1573,13 @@ static inline void netif_tx_unlock_bh(struct net_device *dev)
 
 static inline void netif_tx_disable(struct net_device *dev)
 {
+	unsigned int i;
+
 	netif_tx_lock_bh(dev);
-	netif_stop_queue(dev);
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		netif_tx_stop_queue(txq);
+	}
 	netif_tx_unlock_bh(dev);
 }
 
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index d58c1a5eb845..cb9527815606 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -88,9 +88,7 @@ extern void __qdisc_run(struct netdev_queue *txq);
 
 static inline void qdisc_run(struct netdev_queue *txq)
 {
-	struct net_device *dev = txq->dev;
-
-	if (!netif_queue_stopped(dev) &&
+	if (!netif_tx_queue_stopped(txq) &&
 	    !test_and_set_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state))
 		__qdisc_run(txq);
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index 69378f250695..f027a1ac4fbb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1598,7 +1598,8 @@ static int dev_gso_segment(struct sk_buff *skb)
 	return 0;
 }
 
-int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
+int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
+			struct netdev_queue *txq)
 {
 	if (likely(!skb->next)) {
 		if (!list_empty(&ptype_all))
@@ -1627,9 +1628,7 @@ gso:
 			skb->next = nskb;
 			return rc;
 		}
-		if (unlikely((netif_queue_stopped(dev) ||
-			     netif_subqueue_stopped(dev, skb)) &&
-			     skb->next))
+		if (unlikely(netif_tx_queue_stopped(txq) && skb->next))
 			return NETDEV_TX_BUSY;
 	} while (skb->next);
 
@@ -1669,7 +1668,10 @@ out_kfree_skb:
 static struct netdev_queue *dev_pick_tx(struct net_device *dev,
 					struct sk_buff *skb)
 {
-	return netdev_get_tx_queue(dev, 0);
+	u16 queue_index = 0;
+
+	skb_set_queue_mapping(skb, queue_index);
+	return netdev_get_tx_queue(dev, queue_index);
 }
 
 int dev_queue_xmit(struct sk_buff *skb)
@@ -1737,8 +1739,6 @@ gso:
 		spin_lock(&txq->lock);
 		q = txq->qdisc;
 		if (q->enqueue) {
-			/* reset queue_mapping to zero */
-			skb_set_queue_mapping(skb, 0);
 			rc = q->enqueue(skb, q);
 			qdisc_run(txq);
 			spin_unlock(&txq->lock);
@@ -1768,10 +1768,9 @@ gso:
 
 			HARD_TX_LOCK(dev, txq, cpu);
 
-			if (!netif_queue_stopped(dev) &&
-			    !netif_subqueue_stopped(dev, skb)) {
+			if (!netif_tx_queue_stopped(txq)) {
 				rc = 0;
-				if (!dev_hard_start_xmit(skb, dev)) {
+				if (!dev_hard_start_xmit(skb, dev, txq)) {
 					HARD_TX_UNLOCK(dev, txq);
 					goto out;
 				}
@@ -4160,8 +4159,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 	BUG_ON(strlen(name) >= sizeof(dev->name));
 
-	alloc_size = sizeof(struct net_device) +
-		     sizeof(struct net_device_subqueue) * (queue_count - 1);
+	alloc_size = sizeof(struct net_device);
 	if (sizeof_priv) {
 		/* ensure 32-byte alignment of private area */
 		alloc_size = (alloc_size + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST;
@@ -4191,16 +4189,14 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name,
 
 	dev->_tx = tx;
 	dev->num_tx_queues = queue_count;
+	dev->real_num_tx_queues = queue_count;
 
 	if (sizeof_priv) {
 		dev->priv = ((char *)dev +
-			     ((sizeof(struct net_device) +
-			       (sizeof(struct net_device_subqueue) *
-				(queue_count - 1)) + NETDEV_ALIGN_CONST)
+			     ((sizeof(struct net_device) + NETDEV_ALIGN_CONST)
 			      & ~NETDEV_ALIGN_CONST));
 	}
 
-	dev->egress_subqueue_count = queue_count;
 	dev->gso_max_size = GSO_MAX_SIZE;
 
 	netdev_init_queues(dev);
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 8fb134da0346..c12720895ecf 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -58,25 +58,27 @@ static void queue_process(struct work_struct *work)
 
 	while ((skb = skb_dequeue(&npinfo->txq))) {
 		struct net_device *dev = skb->dev;
+		struct netdev_queue *txq;
 
 		if (!netif_device_present(dev) || !netif_running(dev)) {
 			__kfree_skb(skb);
 			continue;
 		}
 
+		txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
+
 		local_irq_save(flags);
-		netif_tx_lock(dev);
-		if ((netif_queue_stopped(dev) ||
-		     netif_subqueue_stopped(dev, skb)) ||
-		     dev->hard_start_xmit(skb, dev) != NETDEV_TX_OK) {
+		__netif_tx_lock(txq, smp_processor_id());
+		if (netif_tx_queue_stopped(txq) ||
+		    dev->hard_start_xmit(skb, dev) != NETDEV_TX_OK) {
 			skb_queue_head(&npinfo->txq, skb);
-			netif_tx_unlock(dev);
+			__netif_tx_unlock(txq);
 			local_irq_restore(flags);
 
 			schedule_delayed_work(&npinfo->tx_work, HZ/10);
 			return;
 		}
-		netif_tx_unlock(dev);
+		__netif_tx_unlock(txq);
 		local_irq_restore(flags);
 	}
 }
@@ -278,17 +280,19 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
 
 	/* don't get messages out of order, and no recursion */
 	if (skb_queue_len(&npinfo->txq) == 0 && !netpoll_owner_active(dev)) {
+		struct netdev_queue *txq;
 		unsigned long flags;
 
+		txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
+
 		local_irq_save(flags);
 		/* try until next clock tick */
 		for (tries = jiffies_to_usecs(1)/USEC_PER_POLL;
 		     tries > 0; --tries) {
-			if (netif_tx_trylock(dev)) {
-				if (!netif_queue_stopped(dev) &&
-				    !netif_subqueue_stopped(dev, skb))
+			if (__netif_tx_trylock(txq)) {
+				if (!netif_tx_queue_stopped(txq))
 					status = dev->hard_start_xmit(skb, dev);
-				netif_tx_unlock(dev);
+				__netif_tx_unlock(txq);
 
 				if (status == NETDEV_TX_OK)
 					break;
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index fdf537707e51..906802db4ed4 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2123,6 +2123,24 @@ static void get_ipsec_sa(struct pktgen_dev *pkt_dev, int flow)
 	}
 }
 #endif
+static void set_cur_queue_map(struct pktgen_dev *pkt_dev)
+{
+	if (pkt_dev->queue_map_min < pkt_dev->queue_map_max) {
+		__u16 t;
+		if (pkt_dev->flags & F_QUEUE_MAP_RND) {
+			t = random32() %
+				(pkt_dev->queue_map_max -
+				 pkt_dev->queue_map_min + 1)
+				+ pkt_dev->queue_map_min;
+		} else {
+			t = pkt_dev->cur_queue_map + 1;
+			if (t > pkt_dev->queue_map_max)
+				t = pkt_dev->queue_map_min;
+		}
+		pkt_dev->cur_queue_map = t;
+	}
+}
+
 /* Increment/randomize headers according to flags and current values
  * for IP src/dest, UDP src/dst port, MAC-Addr src/dst
  */
@@ -2325,19 +2343,7 @@ static void mod_cur_headers(struct pktgen_dev *pkt_dev)
 		pkt_dev->cur_pkt_size = t;
 	}
 
-	if (pkt_dev->queue_map_min < pkt_dev->queue_map_max) {
-		__u16 t;
-		if (pkt_dev->flags & F_QUEUE_MAP_RND) {
-			t = random32() %
-				(pkt_dev->queue_map_max - pkt_dev->queue_map_min + 1)
-				+ pkt_dev->queue_map_min;
-		} else {
-			t = pkt_dev->cur_queue_map + 1;
-			if (t > pkt_dev->queue_map_max)
-				t = pkt_dev->queue_map_min;
-		}
-		pkt_dev->cur_queue_map = t;
-	}
+	set_cur_queue_map(pkt_dev);
 
 	pkt_dev->flows[flow].count++;
 }
@@ -2458,7 +2464,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 	__be16 *vlan_encapsulated_proto = NULL;  /* packet type ID field (or len) for VLAN tag */
 	__be16 *svlan_tci = NULL;                /* Encapsulates priority and SVLAN ID */
 	__be16 *svlan_encapsulated_proto = NULL; /* packet type ID field (or len) for SVLAN tag */
-
+	u16 queue_map;
 
 	if (pkt_dev->nr_labels)
 		protocol = htons(ETH_P_MPLS_UC);
@@ -2469,6 +2475,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 	/* Update any of the values, used when we're incrementing various
 	 * fields.
 	 */
+	queue_map = pkt_dev->cur_queue_map;
 	mod_cur_headers(pkt_dev);
 
 	datalen = (odev->hard_header_len + 16) & ~0xf;
@@ -2507,7 +2514,7 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev,
 	skb->network_header = skb->tail;
 	skb->transport_header = skb->network_header + sizeof(struct iphdr);
 	skb_put(skb, sizeof(struct iphdr) + sizeof(struct udphdr));
-	skb_set_queue_mapping(skb, pkt_dev->cur_queue_map);
+	skb_set_queue_mapping(skb, queue_map);
 	iph = ip_hdr(skb);
 	udph = udp_hdr(skb);
 
@@ -2797,6 +2804,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 	__be16 *vlan_encapsulated_proto = NULL;  /* packet type ID field (or len) for VLAN tag */
 	__be16 *svlan_tci = NULL;                /* Encapsulates priority and SVLAN ID */
 	__be16 *svlan_encapsulated_proto = NULL; /* packet type ID field (or len) for SVLAN tag */
+	u16 queue_map;
 
 	if (pkt_dev->nr_labels)
 		protocol = htons(ETH_P_MPLS_UC);
@@ -2807,6 +2815,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 	/* Update any of the values, used when we're incrementing various
 	 * fields.
 	 */
+	queue_map = pkt_dev->cur_queue_map;
 	mod_cur_headers(pkt_dev);
 
 	skb = alloc_skb(pkt_dev->cur_pkt_size + 64 + 16 +
@@ -2844,7 +2853,7 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev,
 	skb->network_header = skb->tail;
 	skb->transport_header = skb->network_header + sizeof(struct ipv6hdr);
 	skb_put(skb, sizeof(struct ipv6hdr) + sizeof(struct udphdr));
-	skb_set_queue_mapping(skb, pkt_dev->cur_queue_map);
+	skb_set_queue_mapping(skb, queue_map);
 	iph = ipv6_hdr(skb);
 	udph = udp_hdr(skb);
 
@@ -3263,7 +3272,9 @@ static void pktgen_rem_thread(struct pktgen_thread *t)
 static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 {
 	struct net_device *odev = NULL;
+	struct netdev_queue *txq;
 	__u64 idle_start = 0;
+	u16 queue_map;
 	int ret;
 
 	odev = pkt_dev->odev;
@@ -3285,9 +3296,15 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		}
 	}
 
-	if ((netif_queue_stopped(odev) ||
-	     (pkt_dev->skb &&
-	      netif_subqueue_stopped(odev, pkt_dev->skb))) ||
+	if (!pkt_dev->skb) {
+		set_cur_queue_map(pkt_dev);
+		queue_map = pkt_dev->cur_queue_map;
+	} else {
+		queue_map = skb_get_queue_mapping(pkt_dev->skb);
+	}
+
+	txq = netdev_get_tx_queue(odev, queue_map);
+	if (netif_tx_queue_stopped(txq) ||
 	    need_resched()) {
 		idle_start = getCurUs();
 
@@ -3303,8 +3320,7 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 
 		pkt_dev->idle_acc += getCurUs() - idle_start;
 
-		if (netif_queue_stopped(odev) ||
-		    netif_subqueue_stopped(odev, pkt_dev->skb)) {
+		if (netif_tx_queue_stopped(txq)) {
 			pkt_dev->next_tx_us = getCurUs();	/* TODO */
 			pkt_dev->next_tx_ns = 0;
 			goto out;	/* Try the next interface */
@@ -3331,9 +3347,12 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		}
 	}
 
-	netif_tx_lock_bh(odev);
-	if (!netif_queue_stopped(odev) &&
-	    !netif_subqueue_stopped(odev, pkt_dev->skb)) {
+	/* fill_packet() might have changed the queue */
+	queue_map = skb_get_queue_mapping(pkt_dev->skb);
+	txq = netdev_get_tx_queue(odev, queue_map);
+
+	__netif_tx_lock_bh(txq);
+	if (!netif_tx_queue_stopped(txq)) {
 
 		atomic_inc(&(pkt_dev->skb->users));
 	      retry_now:
@@ -3377,7 +3396,7 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		pkt_dev->next_tx_ns = 0;
 	}
 
-	netif_tx_unlock_bh(odev);
+	__netif_tx_unlock_bh(txq);
 
 	/* If pkt_dev->count is zero, then run forever */
 	if ((pkt_dev->count != 0) && (pkt_dev->sofar >= pkt_dev->count)) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 4e2b865cbba0..2f575b9017d1 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -166,7 +166,7 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 
 	HARD_TX_LOCK(dev, txq, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
-		ret = dev_hard_start_xmit(skb, dev);
+		ret = dev_hard_start_xmit(skb, dev, txq);
 	HARD_TX_UNLOCK(dev, txq);
 
 	spin_lock(&txq->lock);
@@ -198,11 +198,10 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 
 void __qdisc_run(struct netdev_queue *txq)
 {
-	struct net_device *dev = txq->dev;
 	unsigned long start_time = jiffies;
 
 	while (qdisc_restart(txq)) {
-		if (netif_queue_stopped(dev))
+		if (netif_tx_queue_stopped(txq))
 			break;
 
 		/*
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index 44a2c3451f4d..ade3372221c7 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -295,8 +295,7 @@ restart:
 		slave_txq = netdev_get_tx_queue(slave, 0);
 		if (slave_txq->qdisc_sleeping != q)
 			continue;
-		if (netif_queue_stopped(slave) ||
-		    __netif_subqueue_stopped(slave, subq) ||
+		if (__netif_subqueue_stopped(slave, subq) ||
 		    !netif_running(slave)) {
 			busy = 1;
 			continue;
@@ -305,8 +304,7 @@ restart:
 		switch (teql_resolve(skb, skb_res, slave)) {
 		case 0:
 			if (netif_tx_trylock(slave)) {
-				if (!netif_queue_stopped(slave) &&
-				    !__netif_subqueue_stopped(slave, subq) &&
+				if (!__netif_subqueue_stopped(slave, subq) &&
 				    slave->hard_start_xmit(skb, slave) == 0) {
 					netif_tx_unlock(slave);
 					master->slaves = NEXT_SLAVE(q);
-- 
cgit v1.2.3


From 6b0fb1261a4655613bed5dac0e935e733969e999 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 02:58:10 -0700
Subject: netdev: Kill struct net_device_subqueue and netdev->egress_subqueue*

No longer used.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 12 ------------
 1 file changed, 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b5c1e7df64fc..a1c2c2204498 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -118,14 +118,6 @@ struct wireless_dev;
 
 #endif  /*  __KERNEL__  */
 
-struct net_device_subqueue
-{
-	/* Give a control state for each queue.  This struct may contain
-	 * per-queue locks in the future.
-	 */
-	unsigned long   state;
-};
-
 /*
  *	Network device statistics. Akin to the 2.0 ether stats but
  *	with byte counters.
@@ -761,10 +753,6 @@ struct net_device
 	/* for setting kernel sock attribute on TCP connection setup */
 #define GSO_MAX_SIZE		65536
 	unsigned int		gso_max_size;
-
-	/* The TX queue control structures */
-	unsigned int			egress_subqueue_count;
-	struct net_device_subqueue	egress_subqueue[1];
 };
 #define to_net_dev(d) container_of(d, struct net_device, dev)
 
-- 
cgit v1.2.3


From e3c50d5d25ac09efd9acbe2b2a3e365466de84ed Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 02:58:39 -0700
Subject: netdev: netdev_priv() can now be sane again.

The private area of a netdev is now at a fixed offset once more.

Unfortunately, some assumptions that netdev_priv() == netdev->priv
crept back into the tree.  In particular this happened in the
loopback driver.  Make it use netdev->ml_priv.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/loopback.c    | 8 ++++----
 include/linux/netdevice.h | 4 +++-
 2 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index 41b774baac4d..49f6bc036a92 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -153,7 +153,7 @@ static int loopback_xmit(struct sk_buff *skb, struct net_device *dev)
 	dev->last_rx = jiffies;
 
 	/* it's OK to use per_cpu_ptr() because BHs are off */
-	pcpu_lstats = netdev_priv(dev);
+	pcpu_lstats = dev->ml_priv;
 	lb_stats = per_cpu_ptr(pcpu_lstats, smp_processor_id());
 	lb_stats->bytes += skb->len;
 	lb_stats->packets++;
@@ -171,7 +171,7 @@ static struct net_device_stats *get_stats(struct net_device *dev)
 	unsigned long packets = 0;
 	int i;
 
-	pcpu_lstats = netdev_priv(dev);
+	pcpu_lstats = dev->ml_priv;
 	for_each_possible_cpu(i) {
 		const struct pcpu_lstats *lb_stats;
 
@@ -207,13 +207,13 @@ static int loopback_dev_init(struct net_device *dev)
 	if (!lstats)
 		return -ENOMEM;
 
-	dev->priv = lstats;
+	dev->ml_priv = lstats;
 	return 0;
 }
 
 static void loopback_dev_free(struct net_device *dev)
 {
-	struct pcpu_lstats *lstats = netdev_priv(dev);
+	struct pcpu_lstats *lstats = dev->ml_priv;
 
 	free_percpu(lstats);
 	free_netdev(dev);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a1c2c2204498..fdac1159253e 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -808,7 +808,9 @@ void dev_net_set(struct net_device *dev, struct net *net)
  */
 static inline void *netdev_priv(const struct net_device *dev)
 {
-	return dev->priv;
+	return (char *)dev + ((sizeof(struct net_device)
+			       + NETDEV_ALIGN_CONST)
+			      & ~NETDEV_ALIGN_CONST);
 }
 
 /* Set the sysfs physical device reference for the network logical device
-- 
cgit v1.2.3


From eae792b722fef08dcf3aee88266ee7def9710757 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 03:03:33 -0700
Subject: netdev: Add netdev->select_queue() method.

Devices or device layers can set this to control the queue selection
performed by dev_pick_tx().

This function runs under RCU protection, which allows overriding
functions to have some way of synchronizing with things like dynamic
->real_num_tx_queues adjustments.

This makes the spinlock prefetch in dev_queue_xmit() a little bit
less effective, but that's the price right now for correctness.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 3 +++
 net/core/dev.c            | 9 ++++++---
 2 files changed, 9 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fdac1159253e..9464e6452967 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -724,6 +724,9 @@ struct net_device
 	void                    (*poll_controller)(struct net_device *dev);
 #endif
 
+	u16			(*select_queue)(struct net_device *dev,
+						struct sk_buff *skb);
+
 #ifdef CONFIG_NET_NS
 	/* Network namespace this network device is inside */
 	struct net		*nd_net;
diff --git a/net/core/dev.c b/net/core/dev.c
index f027a1ac4fbb..7ca9564d2f44 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1670,6 +1670,9 @@ static struct netdev_queue *dev_pick_tx(struct net_device *dev,
 {
 	u16 queue_index = 0;
 
+	if (dev->select_queue)
+		queue_index = dev->select_queue(dev, skb);
+
 	skb_set_queue_mapping(skb, queue_index);
 	return netdev_get_tx_queue(dev, queue_index);
 }
@@ -1710,14 +1713,14 @@ int dev_queue_xmit(struct sk_buff *skb)
 	}
 
 gso:
-	txq = dev_pick_tx(dev, skb);
-	spin_lock_prefetch(&txq->lock);
-
 	/* Disable soft irqs for various locks below. Also
 	 * stops preemption for RCU.
 	 */
 	rcu_read_lock_bh();
 
+	txq = dev_pick_tx(dev, skb);
+	spin_lock_prefetch(&txq->lock);
+
 	/* Updates of qdisc are serialized by queue->lock.
 	 * The struct Qdisc which is pointed to by qdisc is now a
 	 * rcu structure - it may be accessed without acquiring
-- 
cgit v1.2.3


From 92831bc395ac8390bf759775c50cb6f90c6eb03d Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 03:48:01 -0700
Subject: netdev: Kill plain netif_schedule()

No more users.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9464e6452967..787fbfc5aebb 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -994,11 +994,6 @@ static inline void netif_schedule_queue(struct netdev_queue *txq)
 		__netif_schedule(txq);
 }
 
-static inline void netif_schedule(struct net_device *dev)
-{
-	netif_schedule_queue(netdev_get_tx_queue(dev, 0));
-}
-
 static inline void netif_tx_schedule_all(struct net_device *dev)
 {
 	unsigned int i;
-- 
cgit v1.2.3


From d3b753db7c4f1f37a98b51974d484fda5d86dab5 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 15 Jul 2008 20:14:35 -0700
Subject: pkt_sched: Move gso_skb into Qdisc.

We liberate any dangling gso_skb during qdisc destruction.

It really only matters for the root qdisc.  But when qdiscs
can be shared by multiple netdev_queue objects, we can't
have the gso_skb in the netdev_queue any more.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 -
 include/net/sch_generic.h |  1 +
 net/sched/sch_generic.c   | 19 +++++++++----------
 3 files changed, 10 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 787fbfc5aebb..0883fcf2d16a 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -449,7 +449,6 @@ struct netdev_queue {
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
 	unsigned long		state;
-	struct sk_buff		*gso_skb;
 	spinlock_t		_xmit_lock;
 	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index b47f556c66f8..b96c3d9e10a8 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -36,6 +36,7 @@ struct Qdisc
 	u32			handle;
 	u32			parent;
 	atomic_t		refcnt;
+	struct sk_buff		*gso_skb;
 	struct sk_buff_head	q;
 	struct netdev_queue	*dev_queue;
 	struct list_head	list;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 2f575b9017d1..2bd75befa066 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -77,7 +77,7 @@ static inline int dev_requeue_skb(struct sk_buff *skb,
 				  struct Qdisc *q)
 {
 	if (unlikely(skb->next))
-		dev_queue->gso_skb = skb;
+		q->gso_skb = skb;
 	else
 		q->ops->requeue(skb, q);
 
@@ -85,13 +85,12 @@ static inline int dev_requeue_skb(struct sk_buff *skb,
 	return 0;
 }
 
-static inline struct sk_buff *dequeue_skb(struct netdev_queue *dev_queue,
-					  struct Qdisc *q)
+static inline struct sk_buff *dequeue_skb(struct Qdisc *q)
 {
 	struct sk_buff *skb;
 
-	if ((skb = dev_queue->gso_skb))
-		dev_queue->gso_skb = NULL;
+	if ((skb = q->gso_skb))
+		q->gso_skb = NULL;
 	else
 		skb = q->dequeue(q);
 
@@ -155,10 +154,9 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 	struct sk_buff *skb;
 
 	/* Dequeue packet */
-	if (unlikely((skb = dequeue_skb(txq, q)) == NULL))
+	if (unlikely((skb = dequeue_skb(q)) == NULL))
 		return 0;
 
-
 	/* And release queue */
 	spin_unlock(&txq->lock);
 
@@ -643,8 +641,8 @@ static void dev_deactivate_queue(struct net_device *dev,
 				 void *_qdisc_default)
 {
 	struct Qdisc *qdisc_default = _qdisc_default;
+	struct sk_buff *skb = NULL;
 	struct Qdisc *qdisc;
-	struct sk_buff *skb;
 
 	spin_lock_bh(&dev_queue->lock);
 
@@ -652,9 +650,10 @@ static void dev_deactivate_queue(struct net_device *dev,
 	if (qdisc) {
 		dev_queue->qdisc = qdisc_default;
 		qdisc_reset(qdisc);
+
+		skb = qdisc->gso_skb;
+		qdisc->gso_skb = NULL;
 	}
-	skb = dev_queue->gso_skb;
-	dev_queue->gso_skb = NULL;
 
 	spin_unlock_bh(&dev_queue->lock);
 
-- 
cgit v1.2.3


From e2627c8c2241bce45e368e150654d076b58a4595 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 16 Jul 2008 00:56:32 -0700
Subject: pkt_sched: Make QDISC_RUNNING a qdisc state.

Currently it is associated with a netdev_queue, but when we have
qdisc sharing that no longer makes any sense.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  1 -
 include/net/pkt_sched.h   |  4 +++-
 include/net/sch_generic.h |  6 ++++++
 net/sched/sch_generic.c   | 18 ++++++++++--------
 4 files changed, 19 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 0883fcf2d16a..9240a95793be 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -441,7 +441,6 @@ static inline void napi_synchronize(const struct napi_struct *n)
 enum netdev_queue_state_t
 {
 	__QUEUE_STATE_XOFF,
-	__QUEUE_STATE_QDISC_RUNNING,
 };
 
 struct netdev_queue {
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index cb9527815606..06a442d85186 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -88,8 +88,10 @@ extern void __qdisc_run(struct netdev_queue *txq);
 
 static inline void qdisc_run(struct netdev_queue *txq)
 {
+	struct Qdisc *q = txq->qdisc;
+
 	if (!netif_tx_queue_stopped(txq) &&
-	    !test_and_set_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state))
+	    !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state))
 		__qdisc_run(txq);
 }
 
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index b96c3d9e10a8..bc2a09da21b1 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -23,6 +23,11 @@ struct qdisc_rate_table
 	int		refcnt;
 };
 
+enum qdisc_state_t
+{
+	__QDISC_STATE_RUNNING,
+};
+
 struct Qdisc
 {
 	int 			(*enqueue)(struct sk_buff *skb, struct Qdisc *dev);
@@ -36,6 +41,7 @@ struct Qdisc
 	u32			handle;
 	u32			parent;
 	atomic_t		refcnt;
+	unsigned long		state;
 	struct sk_buff		*gso_skb;
 	struct sk_buff_head	q;
 	struct netdev_queue	*dev_queue;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 2bd75befa066..ac208c2b2d10 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -130,8 +130,8 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 /*
  * NOTE: Called under queue->lock with locally disabled BH.
  *
- * __QUEUE_STATE_QDISC_RUNNING guarantees only one CPU can process
- * this queue at a time. queue->lock serializes queue accesses for
+ * __QDISC_STATE_RUNNING guarantees only one CPU can process
+ * this qdisc at a time. queue->lock serializes queue accesses for
  * this queue AND txq->qdisc pointer itself.
  *
  *  netif_tx_lock serializes accesses to device driver.
@@ -146,9 +146,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  *				>0 - queue is not empty.
  *
  */
-static inline int qdisc_restart(struct netdev_queue *txq)
+static inline int qdisc_restart(struct netdev_queue *txq,
+				struct Qdisc *q)
 {
-	struct Qdisc *q = txq->qdisc;
 	int ret = NETDEV_TX_BUSY;
 	struct net_device *dev;
 	struct sk_buff *skb;
@@ -168,7 +168,6 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 	HARD_TX_UNLOCK(dev, txq);
 
 	spin_lock(&txq->lock);
-	q = txq->qdisc;
 
 	switch (ret) {
 	case NETDEV_TX_OK:
@@ -197,8 +196,9 @@ static inline int qdisc_restart(struct netdev_queue *txq)
 void __qdisc_run(struct netdev_queue *txq)
 {
 	unsigned long start_time = jiffies;
+	struct Qdisc *q = txq->qdisc;
 
-	while (qdisc_restart(txq)) {
+	while (qdisc_restart(txq, q)) {
 		if (netif_tx_queue_stopped(txq))
 			break;
 
@@ -213,7 +213,7 @@ void __qdisc_run(struct netdev_queue *txq)
 		}
 	}
 
-	clear_bit(__QUEUE_STATE_QDISC_RUNNING, &txq->state);
+	clear_bit(__QDISC_STATE_RUNNING, &q->state);
 }
 
 static void dev_watchdog(unsigned long arg)
@@ -666,14 +666,16 @@ static bool some_qdisc_is_running(struct net_device *dev, int lock)
 
 	for (i = 0; i < dev->num_tx_queues; i++) {
 		struct netdev_queue *dev_queue;
+		struct Qdisc *q;
 		int val;
 
 		dev_queue = netdev_get_tx_queue(dev, i);
+		q = dev_queue->qdisc;
 
 		if (lock)
 			spin_lock_bh(&dev_queue->lock);
 
-		val = test_bit(__QUEUE_STATE_QDISC_RUNNING, &dev_queue->state);
+		val = test_bit(__QDISC_STATE_RUNNING, &q->state);
 
 		if (lock)
 			spin_unlock_bh(&dev_queue->lock);
-- 
cgit v1.2.3


From 37437bb2e1ae8af470dfcd5b4ff454110894ccaf Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 16 Jul 2008 02:15:04 -0700
Subject: pkt_sched: Schedule qdiscs instead of netdev_queue.

When we have shared qdiscs, packets come out of the qdiscs
for multiple transmit queues.

Therefore it doesn't make any sense to schedule the transmit
queue when logically we cannot know ahead of time the TX
queue of the SKB that the qdisc->dequeue() will give us.

Just for sanity I added a BUG check to make sure we never
get into a state where the noop_qdisc is scheduled.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 12 ++++-----
 include/net/pkt_sched.h   | 11 +++-----
 include/net/sch_generic.h |  2 ++
 net/core/dev.c            | 68 +++++++++++++++++++----------------------------
 net/sched/sch_api.c       |  3 +--
 net/sched/sch_cbq.c       |  2 +-
 net/sched/sch_generic.c   | 30 ++++++++++-----------
 7 files changed, 55 insertions(+), 73 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9240a95793be..1e839fa01434 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -275,7 +275,6 @@ enum netdev_state_t
 {
 	__LINK_STATE_START,
 	__LINK_STATE_PRESENT,
-	__LINK_STATE_SCHED,
 	__LINK_STATE_NOCARRIER,
 	__LINK_STATE_LINKWATCH_PENDING,
 	__LINK_STATE_DORMANT,
@@ -452,7 +451,6 @@ struct netdev_queue {
 	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
 	struct list_head	qdisc_list;
-	struct netdev_queue	*next_sched;
 } ____cacheline_aligned_in_smp;
 
 /*
@@ -969,7 +967,7 @@ static inline int unregister_gifconf(unsigned int family)
  */
 struct softnet_data
 {
-	struct netdev_queue	*output_queue;
+	struct Qdisc		*output_queue;
 	struct sk_buff_head	input_pkt_queue;
 	struct list_head	poll_list;
 	struct sk_buff		*completion_queue;
@@ -984,12 +982,12 @@ DECLARE_PER_CPU(struct softnet_data,softnet_data);
 
 #define HAVE_NETIF_QUEUE
 
-extern void __netif_schedule(struct netdev_queue *txq);
+extern void __netif_schedule(struct Qdisc *q);
 
 static inline void netif_schedule_queue(struct netdev_queue *txq)
 {
 	if (!test_bit(__QUEUE_STATE_XOFF, &txq->state))
-		__netif_schedule(txq);
+		__netif_schedule(txq->qdisc);
 }
 
 static inline void netif_tx_schedule_all(struct net_device *dev)
@@ -1042,7 +1040,7 @@ static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 	}
 #endif
 	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state))
-		__netif_schedule(dev_queue);
+		__netif_schedule(dev_queue->qdisc);
 }
 
 static inline void netif_wake_queue(struct net_device *dev)
@@ -1186,7 +1184,7 @@ static inline void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
 		return;
 #endif
 	if (test_and_clear_bit(__QUEUE_STATE_XOFF, &txq->state))
-		__netif_schedule(txq);
+		__netif_schedule(txq->qdisc);
 }
 
 /**
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 06a442d85186..e4e30052e4e2 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -84,15 +84,12 @@ extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r,
 		struct nlattr *tab);
 extern void qdisc_put_rtab(struct qdisc_rate_table *tab);
 
-extern void __qdisc_run(struct netdev_queue *txq);
+extern void __qdisc_run(struct Qdisc *q);
 
-static inline void qdisc_run(struct netdev_queue *txq)
+static inline void qdisc_run(struct Qdisc *q)
 {
-	struct Qdisc *q = txq->qdisc;
-
-	if (!netif_tx_queue_stopped(txq) &&
-	    !test_and_set_bit(__QDISC_STATE_RUNNING, &q->state))
-		__qdisc_run(txq);
+	if (!test_and_set_bit(__QDISC_STATE_RUNNING, &q->state))
+		__qdisc_run(q);
 }
 
 extern int tc_classify_compat(struct sk_buff *skb, struct tcf_proto *tp,
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 92417825d387..3cc4b5cd8c6a 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -26,6 +26,7 @@ struct qdisc_rate_table
 enum qdisc_state_t
 {
 	__QDISC_STATE_RUNNING,
+	__QDISC_STATE_SCHED,
 };
 
 struct Qdisc
@@ -45,6 +46,7 @@ struct Qdisc
 	struct sk_buff		*gso_skb;
 	struct sk_buff_head	q;
 	struct netdev_queue	*dev_queue;
+	struct Qdisc		*next_sched;
 	struct list_head	list;
 
 	struct gnet_stats_basic	bstats;
diff --git a/net/core/dev.c b/net/core/dev.c
index 467bfb325123..0b909b74f698 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1323,18 +1323,18 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
 }
 
 
-void __netif_schedule(struct netdev_queue *txq)
+void __netif_schedule(struct Qdisc *q)
 {
-	struct net_device *dev = txq->dev;
+	BUG_ON(q == &noop_qdisc);
 
-	if (!test_and_set_bit(__LINK_STATE_SCHED, &dev->state)) {
+	if (!test_and_set_bit(__QDISC_STATE_SCHED, &q->state)) {
 		struct softnet_data *sd;
 		unsigned long flags;
 
 		local_irq_save(flags);
 		sd = &__get_cpu_var(softnet_data);
-		txq->next_sched = sd->output_queue;
-		sd->output_queue = txq;
+		q->next_sched = sd->output_queue;
+		sd->output_queue = q;
 		raise_softirq_irqoff(NET_TX_SOFTIRQ);
 		local_irq_restore(flags);
 	}
@@ -1771,37 +1771,23 @@ gso:
 	rcu_read_lock_bh();
 
 	txq = dev_pick_tx(dev, skb);
-	spin_lock_prefetch(&txq->lock);
-
-	/* Updates of qdisc are serialized by queue->lock.
-	 * The struct Qdisc which is pointed to by qdisc is now a
-	 * rcu structure - it may be accessed without acquiring
-	 * a lock (but the structure may be stale.) The freeing of the
-	 * qdisc will be deferred until it's known that there are no
-	 * more references to it.
-	 *
-	 * If the qdisc has an enqueue function, we still need to
-	 * hold the queue->lock before calling it, since queue->lock
-	 * also serializes access to the device queue.
-	 */
-
 	q = rcu_dereference(txq->qdisc);
+
 #ifdef CONFIG_NET_CLS_ACT
 	skb->tc_verd = SET_TC_AT(skb->tc_verd,AT_EGRESS);
 #endif
 	if (q->enqueue) {
-		/* Grab device queue */
-		spin_lock(&txq->lock);
-		q = txq->qdisc;
-		if (q->enqueue) {
-			rc = q->enqueue(skb, q);
-			qdisc_run(txq);
-			spin_unlock(&txq->lock);
-
-			rc = rc == NET_XMIT_BYPASS ? NET_XMIT_SUCCESS : rc;
-			goto out;
-		}
-		spin_unlock(&txq->lock);
+		spinlock_t *root_lock = qdisc_root_lock(q);
+
+		spin_lock(root_lock);
+
+		rc = q->enqueue(skb, q);
+		qdisc_run(q);
+
+		spin_unlock(root_lock);
+
+		rc = rc == NET_XMIT_BYPASS ? NET_XMIT_SUCCESS : rc;
+		goto out;
 	}
 
 	/* The device has no queue. Common case for software devices:
@@ -1974,7 +1960,7 @@ static void net_tx_action(struct softirq_action *h)
 	}
 
 	if (sd->output_queue) {
-		struct netdev_queue *head;
+		struct Qdisc *head;
 
 		local_irq_disable();
 		head = sd->output_queue;
@@ -1982,18 +1968,20 @@ static void net_tx_action(struct softirq_action *h)
 		local_irq_enable();
 
 		while (head) {
-			struct netdev_queue *txq = head;
-			struct net_device *dev = txq->dev;
+			struct Qdisc *q = head;
+			spinlock_t *root_lock;
+
 			head = head->next_sched;
 
 			smp_mb__before_clear_bit();
-			clear_bit(__LINK_STATE_SCHED, &dev->state);
+			clear_bit(__QDISC_STATE_SCHED, &q->state);
 
-			if (spin_trylock(&txq->lock)) {
-				qdisc_run(txq);
-				spin_unlock(&txq->lock);
+			root_lock = qdisc_root_lock(q);
+			if (spin_trylock(root_lock)) {
+				qdisc_run(q);
+				spin_unlock(root_lock);
 			} else {
-				netif_schedule_queue(txq);
+				__netif_schedule(q);
 			}
 		}
 	}
@@ -4459,7 +4447,7 @@ static int dev_cpu_callback(struct notifier_block *nfb,
 			    void *ocpu)
 {
 	struct sk_buff **list_skb;
-	struct netdev_queue **list_net;
+	struct Qdisc **list_net;
 	struct sk_buff *skb;
 	unsigned int cpu, oldcpu = (unsigned long)ocpu;
 	struct softnet_data *sd, *oldsd;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 19c244a00839..8e8c5becc348 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -294,11 +294,10 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 {
 	struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
 						 timer);
-	struct netdev_queue *txq = wd->qdisc->dev_queue;
 
 	wd->qdisc->flags &= ~TCQ_F_THROTTLED;
 	smp_wmb();
-	netif_schedule_queue(txq);
+	__netif_schedule(wd->qdisc);
 
 	return HRTIMER_NORESTART;
 }
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index 37ae653db683..a3953bbe2d79 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -650,7 +650,7 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer)
 	}
 
 	sch->flags &= ~TCQ_F_THROTTLED;
-	netif_schedule_queue(sch->dev_queue);
+	__netif_schedule(sch);
 	return HRTIMER_NORESTART;
 }
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 739a8711ab30..dd5c4e70abe4 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -72,16 +72,14 @@ static inline int qdisc_qlen(struct Qdisc *q)
 	return q->q.qlen;
 }
 
-static inline int dev_requeue_skb(struct sk_buff *skb,
-				  struct netdev_queue *dev_queue,
-				  struct Qdisc *q)
+static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
 {
 	if (unlikely(skb->next))
 		q->gso_skb = skb;
 	else
 		q->ops->requeue(skb, q);
 
-	netif_schedule_queue(dev_queue);
+	__netif_schedule(q);
 	return 0;
 }
 
@@ -121,7 +119,7 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 		 * some time.
 		 */
 		__get_cpu_var(netdev_rx_stat).cpu_collision++;
-		ret = dev_requeue_skb(skb, dev_queue, q);
+		ret = dev_requeue_skb(skb, q);
 	}
 
 	return ret;
@@ -146,9 +144,9 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
  *				>0 - queue is not empty.
  *
  */
-static inline int qdisc_restart(struct netdev_queue *txq,
-				struct Qdisc *q)
+static inline int qdisc_restart(struct Qdisc *q)
 {
+	struct netdev_queue *txq;
 	int ret = NETDEV_TX_BUSY;
 	struct net_device *dev;
 	spinlock_t *root_lock;
@@ -163,7 +161,8 @@ static inline int qdisc_restart(struct netdev_queue *txq,
 	/* And release qdisc */
 	spin_unlock(root_lock);
 
-	dev = txq->dev;
+	dev = qdisc_dev(q);
+	txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
 
 	HARD_TX_LOCK(dev, txq, smp_processor_id());
 	if (!netif_subqueue_stopped(dev, skb))
@@ -189,29 +188,28 @@ static inline int qdisc_restart(struct netdev_queue *txq,
 			printk(KERN_WARNING "BUG %s code %d qlen %d\n",
 			       dev->name, ret, q->q.qlen);
 
-		ret = dev_requeue_skb(skb, txq, q);
+		ret = dev_requeue_skb(skb, q);
 		break;
 	}
 
+	if (ret && netif_tx_queue_stopped(txq))
+		ret = 0;
+
 	return ret;
 }
 
-void __qdisc_run(struct netdev_queue *txq)
+void __qdisc_run(struct Qdisc *q)
 {
 	unsigned long start_time = jiffies;
-	struct Qdisc *q = txq->qdisc;
-
-	while (qdisc_restart(txq, q)) {
-		if (netif_tx_queue_stopped(txq))
-			break;
 
+	while (qdisc_restart(q)) {
 		/*
 		 * Postpone processing if
 		 * 1. another process needs the CPU;
 		 * 2. we've been doing it for too long.
 		 */
 		if (need_resched() || jiffies != start_time) {
-			netif_schedule_queue(txq);
+			__netif_schedule(q);
 			break;
 		}
 	}
-- 
cgit v1.2.3


From ead81cc5fc6d996db6afb20f211241612610a07a Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 00:50:32 -0700
Subject: netdevice: Move qdisc_list back into net_device proper.

And give it it's own lock.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  3 ++-
 net/core/dev.c            |  2 ++
 net/sched/sch_api.c       | 31 +++++++------------------------
 net/sched/sch_generic.c   |  9 +++++++--
 4 files changed, 18 insertions(+), 27 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 1e839fa01434..3170bcef734b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -450,7 +450,6 @@ struct netdev_queue {
 	spinlock_t		_xmit_lock;
 	int			xmit_lock_owner;
 	struct Qdisc		*qdisc_sleeping;
-	struct list_head	qdisc_list;
 } ____cacheline_aligned_in_smp;
 
 /*
@@ -638,6 +637,8 @@ struct net_device
 	unsigned int		real_num_tx_queues;
 
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
+	spinlock_t		qdisc_list_lock;
+	struct list_head	qdisc_list;
 
 /*
  * One part is mostly used on xmit path (device)
diff --git a/net/core/dev.c b/net/core/dev.c
index 0b909b74f698..6741e344ac59 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3886,6 +3886,8 @@ int register_netdevice(struct net_device *dev)
 	net = dev_net(dev);
 
 	spin_lock_init(&dev->addr_list_lock);
+	spin_lock_init(&dev->qdisc_list_lock);
+	INIT_LIST_HEAD(&dev->qdisc_list);
 	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 8e8c5becc348..6958fe7c9a77 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -183,30 +183,17 @@ EXPORT_SYMBOL(unregister_qdisc);
    (root qdisc, all its children, children of children etc.)
  */
 
-static struct Qdisc *__qdisc_lookup(struct netdev_queue *dev_queue, u32 handle)
+struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 {
 	struct Qdisc *q;
 
-	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
+	list_for_each_entry(q, &dev->qdisc_list, list) {
 		if (q->handle == handle)
 			return q;
 	}
 	return NULL;
 }
 
-struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
-{
-	unsigned int i;
-
-	for (i = 0; i < dev->num_tx_queues; i++) {
-		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
-		struct Qdisc *q = __qdisc_lookup(txq, handle);
-		if (q)
-			return q;
-	}
-	return NULL;
-}
-
 static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
 {
 	unsigned long cl;
@@ -645,9 +632,9 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 				goto err_out3;
 			}
 		}
-		qdisc_lock_tree(dev);
-		list_add_tail(&sch->list, &dev_queue->qdisc_list);
-		qdisc_unlock_tree(dev);
+		spin_lock_bh(&dev->qdisc_list_lock);
+		list_add_tail(&sch->list, &dev->qdisc_list);
+		spin_unlock_bh(&dev->qdisc_list_lock);
 
 		return sch;
 	}
@@ -1032,14 +1019,12 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 	read_lock(&dev_base_lock);
 	idx = 0;
 	for_each_netdev(&init_net, dev) {
-		struct netdev_queue *dev_queue;
 		if (idx < s_idx)
 			goto cont;
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		dev_queue = netdev_get_tx_queue(dev, 0);
-		list_for_each_entry(q, &dev_queue->qdisc_list, list) {
+		list_for_each_entry(q, &dev->qdisc_list, list) {
 			if (q_idx < s_q_idx) {
 				q_idx++;
 				continue;
@@ -1269,7 +1254,6 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk
 static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
-	struct netdev_queue *dev_queue;
 	int t;
 	int s_t;
 	struct net_device *dev;
@@ -1288,8 +1272,7 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	dev_queue = netdev_get_tx_queue(dev, 0);
-	list_for_each_entry(q, &dev_queue->qdisc_list, list) {
+	list_for_each_entry(q, &dev->qdisc_list, list) {
 		if (t < s_t || !q->ops->cl_ops ||
 		    (tcm->tcm_parent &&
 		     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index efa418a1b34e..8cdf0b4a6a5a 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -563,11 +563,15 @@ static void __qdisc_destroy(struct rcu_head *head)
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
+	struct net_device *dev = qdisc_dev(qdisc);
+
 	if (qdisc->flags & TCQ_F_BUILTIN ||
 	    !atomic_dec_and_test(&qdisc->refcnt))
 		return;
 
+	spin_lock_bh(&dev->qdisc_list_lock);
 	list_del(&qdisc->list);
+	spin_unlock_bh(&dev->qdisc_list_lock);
 
 	call_rcu(&qdisc->q_rcu, __qdisc_destroy);
 }
@@ -599,7 +603,9 @@ static void attach_one_default_qdisc(struct net_device *dev,
 			printk(KERN_INFO "%s: activation failed\n", dev->name);
 			return;
 		}
-		list_add_tail(&qdisc->list, &dev_queue->qdisc_list);
+		spin_lock_bh(&dev->qdisc_list_lock);
+		list_add_tail(&qdisc->list, &dev->qdisc_list);
+		spin_unlock_bh(&dev->qdisc_list_lock);
 	} else {
 		qdisc =  &noqueue_qdisc;
 	}
@@ -738,7 +744,6 @@ static void dev_init_scheduler_queue(struct net_device *dev,
 
 	dev_queue->qdisc = qdisc;
 	dev_queue->qdisc_sleeping = qdisc;
-	INIT_LIST_HEAD(&dev_queue->qdisc_list);
 }
 
 void dev_init_scheduler(struct net_device *dev)
-- 
cgit v1.2.3


From 83874000929ed63aef30b44083a9f713135ff040 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 17 Jul 2008 00:53:03 -0700
Subject: pkt_sched: Kill netdev_queue lock.

We can simply use the qdisc->q.lock for all of the
qdisc tree synchronization.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c         | 20 --------------------
 include/linux/netdevice.h |  1 -
 include/net/sch_generic.h |  7 ++++++-
 net/core/dev.c            |  9 +++++----
 net/mac80211/wme.c        | 19 ++++++++++++-------
 net/sched/sch_generic.c   | 32 +++++++++++++++-----------------
 net/sched/sch_teql.c      |  7 +++++--
 7 files changed, 43 insertions(+), 52 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index 897b05e79ed0..0960e69b2da4 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -35,7 +35,6 @@
 #include <linux/moduleparam.h>
 #include <net/pkt_sched.h>
 #include <net/net_namespace.h>
-#include <linux/lockdep.h>
 
 #define TX_TIMEOUT  (2*HZ)
 
@@ -228,22 +227,6 @@ static struct rtnl_link_ops ifb_link_ops __read_mostly = {
 module_param(numifbs, int, 0);
 MODULE_PARM_DESC(numifbs, "Number of ifb devices");
 
-/*
- * dev_ifb's TX queue lock is usually taken after dev->rx_queue.lock,
- * reversely to e.g. qdisc_lock_tree(). It should be safe until
- * ifb doesn't take dev's TX queue lock with dev_ifb->rx_queue.lock.
- * But lockdep should know that ifb has different locks from dev.
- */
-static struct lock_class_key ifb_tx_queue_lock_key;
-static struct lock_class_key ifb_rx_queue_lock_key;
-
-static void set_tx_lockdep_key(struct net_device *dev,
-			       struct netdev_queue *txq,
-			       void *_unused)
-{
-	lockdep_set_class(&txq->lock, &ifb_tx_queue_lock_key);
-}
-
 static int __init ifb_init_one(int index)
 {
 	struct net_device *dev_ifb;
@@ -264,9 +247,6 @@ static int __init ifb_init_one(int index)
 	if (err < 0)
 		goto err;
 
-	netdev_for_each_tx_queue(dev_ifb, set_tx_lockdep_key, NULL);
-	lockdep_set_class(&dev_ifb->rx_queue.lock, &ifb_rx_queue_lock_key);
-
 	return 0;
 
 err:
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 3170bcef734b..9c5a68850114 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -443,7 +443,6 @@ enum netdev_queue_state_t
 };
 
 struct netdev_queue {
-	spinlock_t		lock;
 	struct net_device	*dev;
 	struct Qdisc		*qdisc;
 	unsigned long		state;
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 1eef8d0c9990..2902a42564f0 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -163,6 +163,11 @@ struct tcf_proto
 	struct tcf_proto_ops	*ops;
 };
 
+static inline spinlock_t *qdisc_lock(struct Qdisc *qdisc)
+{
+	return &qdisc->q.lock;
+}
+
 static inline struct Qdisc *qdisc_root(struct Qdisc *qdisc)
 {
 	return qdisc->dev_queue->qdisc;
@@ -172,7 +177,7 @@ static inline spinlock_t *qdisc_root_lock(struct Qdisc *qdisc)
 {
 	struct Qdisc *root = qdisc_root(qdisc);
 
-	return &root->dev_queue->lock;
+	return qdisc_lock(root);
 }
 
 static inline struct net_device *qdisc_dev(struct Qdisc *qdisc)
diff --git a/net/core/dev.c b/net/core/dev.c
index 6741e344ac59..32a13772c1cb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2080,10 +2080,12 @@ static int ing_filter(struct sk_buff *skb)
 
 	rxq = &dev->rx_queue;
 
-	spin_lock(&rxq->lock);
-	if ((q = rxq->qdisc) != NULL)
+	q = rxq->qdisc;
+	if (q) {
+		spin_lock(qdisc_lock(q));
 		result = q->enqueue(skb, q);
-	spin_unlock(&rxq->lock);
+		spin_unlock(qdisc_lock(q));
+	}
 
 	return result;
 }
@@ -4173,7 +4175,6 @@ static void netdev_init_one_queue(struct net_device *dev,
 				  struct netdev_queue *queue,
 				  void *_unused)
 {
-	spin_lock_init(&queue->lock);
 	queue->dev = dev;
 }
 
diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c
index b21cfec4b6ce..6e8099e77043 100644
--- a/net/mac80211/wme.c
+++ b/net/mac80211/wme.c
@@ -237,12 +237,14 @@ void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local,
 		ieee80211_requeue(local, agg_queue);
 	} else {
 		struct netdev_queue *txq;
+		spinlock_t *root_lock;
 
 		txq = netdev_get_tx_queue(local->mdev, agg_queue);
+		root_lock = qdisc_root_lock(txq->qdisc);
 
-		spin_lock_bh(&txq->lock);
+		spin_lock_bh(root_lock);
 		qdisc_reset(txq->qdisc);
-		spin_unlock_bh(&txq->lock);
+		spin_unlock_bh(root_lock);
 	}
 }
 
@@ -250,6 +252,7 @@ void ieee80211_requeue(struct ieee80211_local *local, int queue)
 {
 	struct netdev_queue *txq = netdev_get_tx_queue(local->mdev, queue);
 	struct sk_buff_head list;
+	spinlock_t *root_lock;
 	struct Qdisc *qdisc;
 	u32 len;
 
@@ -261,14 +264,15 @@ void ieee80211_requeue(struct ieee80211_local *local, int queue)
 
 	skb_queue_head_init(&list);
 
-	spin_lock(&txq->lock);
+	root_lock = qdisc_root_lock(qdisc);
+	spin_lock(root_lock);
 	for (len = qdisc->q.qlen; len > 0; len--) {
 		struct sk_buff *skb = qdisc->dequeue(qdisc);
 
 		if (skb)
 			__skb_queue_tail(&list, skb);
 	}
-	spin_unlock(&txq->lock);
+	spin_unlock(root_lock);
 
 	for (len = list.qlen; len > 0; len--) {
 		struct sk_buff *skb = __skb_dequeue(&list);
@@ -280,12 +284,13 @@ void ieee80211_requeue(struct ieee80211_local *local, int queue)
 
 		txq = netdev_get_tx_queue(local->mdev, new_queue);
 
-		spin_lock(&txq->lock);
 
 		qdisc = rcu_dereference(txq->qdisc);
-		qdisc->enqueue(skb, qdisc);
+		root_lock = qdisc_root_lock(qdisc);
 
-		spin_unlock(&txq->lock);
+		spin_lock(root_lock);
+		qdisc->enqueue(skb, qdisc);
+		spin_unlock(root_lock);
 	}
 
 out_unlock:
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 3d53e92ad9c8..8fc580b3e173 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -96,15 +96,15 @@ static inline int handle_dev_cpu_collision(struct sk_buff *skb,
 }
 
 /*
- * NOTE: Called under queue->lock with locally disabled BH.
+ * NOTE: Called under qdisc_lock(q) with locally disabled BH.
  *
  * __QDISC_STATE_RUNNING guarantees only one CPU can process
- * this qdisc at a time. queue->lock serializes queue accesses for
- * this queue AND txq->qdisc pointer itself.
+ * this qdisc at a time. qdisc_lock(q) serializes queue accesses for
+ * this queue.
  *
  *  netif_tx_lock serializes accesses to device driver.
  *
- *  queue->lock and netif_tx_lock are mutually exclusive,
+ *  qdisc_lock(q) and netif_tx_lock are mutually exclusive,
  *  if one is grabbed, another must be free.
  *
  * Note, that this procedure can be called by a watchdog timer
@@ -317,7 +317,6 @@ struct Qdisc_ops noop_qdisc_ops __read_mostly = {
 };
 
 static struct netdev_queue noop_netdev_queue = {
-	.lock		=	__SPIN_LOCK_UNLOCKED(noop_netdev_queue.lock),
 	.qdisc		=	&noop_qdisc,
 };
 
@@ -327,6 +326,7 @@ struct Qdisc noop_qdisc = {
 	.flags		=	TCQ_F_BUILTIN,
 	.ops		=	&noop_qdisc_ops,
 	.list		=	LIST_HEAD_INIT(noop_qdisc.list),
+	.q.lock		=	__SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
 	.dev_queue	=	&noop_netdev_queue,
 };
 EXPORT_SYMBOL(noop_qdisc);
@@ -498,7 +498,7 @@ errout:
 }
 EXPORT_SYMBOL(qdisc_create_dflt);
 
-/* Under queue->lock and BH! */
+/* Under qdisc_root_lock(qdisc) and BH! */
 
 void qdisc_reset(struct Qdisc *qdisc)
 {
@@ -526,10 +526,12 @@ static void __qdisc_destroy(struct rcu_head *head)
 	module_put(ops->owner);
 	dev_put(qdisc_dev(qdisc));
 
+	kfree_skb(qdisc->gso_skb);
+
 	kfree((char *) qdisc - qdisc->padded);
 }
 
-/* Under queue->lock and BH! */
+/* Under qdisc_root_lock(qdisc) and BH! */
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
@@ -586,13 +588,12 @@ static void transition_one_qdisc(struct net_device *dev,
 				 struct netdev_queue *dev_queue,
 				 void *_need_watchdog)
 {
+	struct Qdisc *new_qdisc = dev_queue->qdisc_sleeping;
 	int *need_watchdog_p = _need_watchdog;
 
-	spin_lock_bh(&dev_queue->lock);
-	rcu_assign_pointer(dev_queue->qdisc, dev_queue->qdisc_sleeping);
-	if (dev_queue->qdisc != &noqueue_qdisc)
+	rcu_assign_pointer(dev_queue->qdisc, new_qdisc);
+	if (new_qdisc != &noqueue_qdisc)
 		*need_watchdog_p = 1;
-	spin_unlock_bh(&dev_queue->lock);
 }
 
 void dev_activate(struct net_device *dev)
@@ -629,19 +630,16 @@ static void dev_deactivate_queue(struct net_device *dev,
 	struct sk_buff *skb = NULL;
 	struct Qdisc *qdisc;
 
-	spin_lock_bh(&dev_queue->lock);
-
 	qdisc = dev_queue->qdisc;
 	if (qdisc) {
+		spin_lock_bh(qdisc_lock(qdisc));
+
 		dev_queue->qdisc = qdisc_default;
 		qdisc_reset(qdisc);
 
-		skb = qdisc->gso_skb;
-		qdisc->gso_skb = NULL;
+		spin_unlock_bh(qdisc_lock(qdisc));
 	}
 
-	spin_unlock_bh(&dev_queue->lock);
-
 	kfree_skb(skb);
 }
 
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index ade3372221c7..8b0ff345f9da 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -156,12 +156,15 @@ teql_destroy(struct Qdisc* sch)
 					master->slaves = NEXT_SLAVE(q);
 					if (q == master->slaves) {
 						struct netdev_queue *txq;
+						spinlock_t *root_lock;
 
 						txq = netdev_get_tx_queue(master->dev, 0);
 						master->slaves = NULL;
-						spin_lock_bh(&txq->lock);
+
+						root_lock = qdisc_root_lock(txq->qdisc);
+						spin_lock_bh(root_lock);
 						qdisc_reset(txq->qdisc);
-						spin_unlock_bh(&txq->lock);
+						spin_unlock_bh(root_lock);
 					}
 				}
 				skb_queue_purge(&dat->q);
-- 
cgit v1.2.3


From de05c557b24c7dffc6d392e3db120cf11c9f6ae7 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:07:21 -0700
Subject: proc: consolidate per-net single_open callers

There are already 7 of them - time to kill some duplicate code.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c           | 24 +++++++++++++++++++
 include/linux/seq_file_net.h |  2 ++
 net/ipv4/fib_trie.c          | 13 +---------
 net/ipv4/proc.c              | 57 +++-----------------------------------------
 net/ipv6/proc.c              | 19 +--------------
 net/ipv6/route.c             | 26 ++------------------
 6 files changed, 33 insertions(+), 108 deletions(-)

(limited to 'include/linux')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index 83f357b30d71..ab232ad2e6f9 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -51,6 +51,30 @@ int seq_open_net(struct inode *ino, struct file *f,
 }
 EXPORT_SYMBOL_GPL(seq_open_net);
 
+int single_open_net(struct inode *inode, struct file *file,
+		int (*show)(struct seq_file *, void *))
+{
+	int err;
+	struct net *net;
+
+	err = -ENXIO;
+	net = get_proc_net(inode);
+	if (net == NULL)
+		goto err_net;
+
+	err = single_open(file, show, net);
+	if (err < 0)
+		goto err_open;
+
+	return 0;
+
+err_open:
+	put_net(net);
+err_net:
+	return err;
+}
+EXPORT_SYMBOL_GPL(single_open_net);
+
 int seq_release_net(struct inode *ino, struct file *f)
 {
 	struct seq_file *seq;
diff --git a/include/linux/seq_file_net.h b/include/linux/seq_file_net.h
index 4ac52542a563..87dcc0ecf6eb 100644
--- a/include/linux/seq_file_net.h
+++ b/include/linux/seq_file_net.h
@@ -14,6 +14,8 @@ struct seq_net_private {
 
 int seq_open_net(struct inode *, struct file *,
 		 const struct seq_operations *, int);
+int single_open_net(struct inode *, struct file *file,
+		int (*show)(struct seq_file *, void *));
 int seq_release_net(struct inode *, struct file *);
 static inline struct net *seq_file_net(struct seq_file *seq)
 {
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index f155a66d6ebf..6009df238ed9 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2251,18 +2251,7 @@ static int fib_triestat_seq_show(struct seq_file *seq, void *v)
 
 static int fib_triestat_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	net = get_proc_net(inode);
-	if (net == NULL)
-		return -ENXIO;
-	err = single_open(file, fib_triestat_seq_show, net);
-	if (err < 0) {
-		put_net(net);
-		return err;
-	}
-	return 0;
+	return single_open_net(inode, file, fib_triestat_seq_show);
 }
 
 static int fib_triestat_seq_release(struct inode *ino, struct file *f)
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 554390431a4e..daf5d3c80cef 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -71,24 +71,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 
 static int sockstat_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, sockstat_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, sockstat_seq_show);
 }
 
 static int sockstat_seq_release(struct inode *inode, struct file *file)
@@ -397,24 +380,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v)
 
 static int snmp_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, snmp_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, snmp_seq_show);
 }
 
 static int snmp_seq_release(struct inode *inode, struct file *file)
@@ -469,24 +435,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v)
 
 static int netstat_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, netstat_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, netstat_seq_show);
 }
 
 static int netstat_seq_release(struct inode *inode, struct file *file)
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index cbc7e514d3ec..29c5a79444c2 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -183,24 +183,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v)
 
 static int sockstat6_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net;
-
-	err = -ENXIO;
-	net = get_proc_net(inode);
-	if (net == NULL)
-		goto err_net;
-
-	err = single_open(file, sockstat6_seq_show, net);
-	if (err < 0)
-		goto err_open;
-
-	return 0;
-
-err_open:
-	put_net(net);
-err_net:
-	return err;
+	return single_open_net(inode, file, sockstat6_seq_show);
 }
 
 static int sockstat6_seq_release(struct inode *inode, struct file *file)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 5d6c166dfbb6..fb7ff8f0c6db 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2416,18 +2416,7 @@ static int ipv6_route_show(struct seq_file *m, void *v)
 
 static int ipv6_route_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net = get_proc_net(inode);
-	if (!net)
-		return -ENXIO;
-
-	err = single_open(file, ipv6_route_show, net);
-	if (err < 0) {
-		put_net(net);
-		return err;
-	}
-
-	return 0;
+	return single_open_net(inode, file, ipv6_route_show);
 }
 
 static int ipv6_route_release(struct inode *inode, struct file *file)
@@ -2463,18 +2452,7 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v)
 
 static int rt6_stats_seq_open(struct inode *inode, struct file *file)
 {
-	int err;
-	struct net *net = get_proc_net(inode);
-	if (!net)
-		return -ENXIO;
-
-	err = single_open(file, rt6_stats_seq_show, net);
-	if (err < 0) {
-		put_net(net);
-		return err;
-	}
-
-	return 0;
+	return single_open_net(inode, file, rt6_stats_seq_show);
 }
 
 static int rt6_stats_seq_release(struct inode *inode, struct file *file)
-- 
cgit v1.2.3


From b6fcbdb4f283f7ba67cec3cda6be23da8e959031 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 18 Jul 2008 04:07:44 -0700
Subject: proc: consolidate per-net single-release callers

They are symmetrical to single_open ones :)

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c           |  8 ++++++++
 include/linux/seq_file_net.h |  1 +
 net/ipv4/fib_trie.c          |  9 +--------
 net/ipv4/proc.c              | 30 +++---------------------------
 net/ipv6/proc.c              | 10 +---------
 net/ipv6/route.c             | 20 ++------------------
 6 files changed, 16 insertions(+), 62 deletions(-)

(limited to 'include/linux')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index ab232ad2e6f9..b224a28e0c15 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -87,6 +87,14 @@ int seq_release_net(struct inode *ino, struct file *f)
 }
 EXPORT_SYMBOL_GPL(seq_release_net);
 
+int single_release_net(struct inode *ino, struct file *f)
+{
+	struct seq_file *seq = f->private_data;
+	put_net(seq->private);
+	return single_release(ino, f);
+}
+EXPORT_SYMBOL_GPL(single_release_net);
+
 static struct net *get_proc_task_net(struct inode *dir)
 {
 	struct task_struct *task;
diff --git a/include/linux/seq_file_net.h b/include/linux/seq_file_net.h
index 87dcc0ecf6eb..32c89bbe24a2 100644
--- a/include/linux/seq_file_net.h
+++ b/include/linux/seq_file_net.h
@@ -17,6 +17,7 @@ int seq_open_net(struct inode *, struct file *,
 int single_open_net(struct inode *, struct file *file,
 		int (*show)(struct seq_file *, void *));
 int seq_release_net(struct inode *, struct file *);
+int single_release_net(struct inode *, struct file *);
 static inline struct net *seq_file_net(struct seq_file *seq)
 {
 #ifdef CONFIG_NET_NS
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 6009df238ed9..5cb72786a8af 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2254,19 +2254,12 @@ static int fib_triestat_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, fib_triestat_seq_show);
 }
 
-static int fib_triestat_seq_release(struct inode *ino, struct file *f)
-{
-	struct seq_file *seq = f->private_data;
-	put_net(seq->private);
-	return single_release(ino, f);
-}
-
 static const struct file_operations fib_triestat_fops = {
 	.owner	= THIS_MODULE,
 	.open	= fib_triestat_seq_open,
 	.read	= seq_read,
 	.llseek	= seq_lseek,
-	.release = fib_triestat_seq_release,
+	.release = single_release_net,
 };
 
 static struct node *fib_trie_get_idx(struct seq_file *seq, loff_t pos)
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index daf5d3c80cef..834356ea99df 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -74,20 +74,12 @@ static int sockstat_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, sockstat_seq_show);
 }
 
-static int sockstat_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations sockstat_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = sockstat_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = sockstat_seq_release,
+	.release = single_release_net,
 };
 
 /* snmp items */
@@ -383,20 +375,12 @@ static int snmp_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, snmp_seq_show);
 }
 
-static int snmp_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations snmp_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = snmp_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = snmp_seq_release,
+	.release = single_release_net,
 };
 
 
@@ -438,20 +422,12 @@ static int netstat_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, netstat_seq_show);
 }
 
-static int netstat_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations netstat_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = netstat_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = netstat_seq_release,
+	.release = single_release_net,
 };
 
 static __net_init int ip_proc_init_net(struct net *net)
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 29c5a79444c2..70940b3654a1 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -186,20 +186,12 @@ static int sockstat6_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, sockstat6_seq_show);
 }
 
-static int sockstat6_seq_release(struct inode *inode, struct file *file)
-{
-	struct net *net = ((struct seq_file *)file->private_data)->private;
-
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations sockstat6_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = sockstat6_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = sockstat6_seq_release,
+	.release = single_release_net,
 };
 
 static int snmp6_seq_open(struct inode *inode, struct file *file)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index fb7ff8f0c6db..cb8a51271b67 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2419,20 +2419,12 @@ static int ipv6_route_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, ipv6_route_show);
 }
 
-static int ipv6_route_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct net *net = seq->private;
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations ipv6_route_proc_fops = {
 	.owner		= THIS_MODULE,
 	.open		= ipv6_route_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= ipv6_route_release,
+	.release	= single_release_net,
 };
 
 static int rt6_stats_seq_show(struct seq_file *seq, void *v)
@@ -2455,20 +2447,12 @@ static int rt6_stats_seq_open(struct inode *inode, struct file *file)
 	return single_open_net(inode, file, rt6_stats_seq_show);
 }
 
-static int rt6_stats_seq_release(struct inode *inode, struct file *file)
-{
-	struct seq_file *seq = file->private_data;
-	struct net *net = (struct net *)seq->private;
-	put_net(net);
-	return single_release(inode, file);
-}
-
 static const struct file_operations rt6_stats_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open	 = rt6_stats_seq_open,
 	.read	 = seq_read,
 	.llseek	 = seq_lseek,
-	.release = rt6_stats_seq_release,
+	.release = single_release_net,
 };
 #endif	/* CONFIG_PROC_FS */
 
-- 
cgit v1.2.3


From e761b7725234276a802322549cee5255305a0930 Mon Sep 17 00:00:00 2001
From: Max Krasnyansky <maxk@qualcomm.com>
Date: Tue, 15 Jul 2008 04:43:49 -0700
Subject: cpu hotplug, sched: Introduce cpu_active_map and redo sched domain
 managment (take 2)

This is based on Linus' idea of creating cpu_active_map that prevents
scheduler load balancer from migrating tasks to the cpu that is going
down.

It allows us to simplify domain management code and avoid unecessary
domain rebuilds during cpu hotplug event handling.

Please ignore the cpusets part for now. It needs some more work in order
to avoid crazy lock nesting. Although I did simplfy and unify domain
reinitialization logic. We now simply call partition_sched_domains() in
all the cases. This means that we're using exact same code paths as in
cpusets case and hence the test below cover cpusets too.
Cpuset changes to make rebuild_sched_domains() callable from various
contexts are in the separate patch (right next after this one).

This not only boots but also easily handles
	while true; do make clean; make -j 8; done
and
	while true; do on-off-cpu 1; done
at the same time.
(on-off-cpu 1 simple does echo 0/1 > /sys/.../cpu1/online thing).

Suprisingly the box (dual-core Core2) is quite usable. In fact I'm typing
this on right now in gnome-terminal and things are moving just fine.

Also this is running with most of the debug features enabled (lockdep,
mutex, etc) no BUG_ONs or lockdep complaints so far.

I believe I addressed all of the Dmitry's comments for original Linus'
version. I changed both fair and rt balancer to mask out non-active cpus.
And replaced cpu_is_offline() with !cpu_active() in the main scheduler
code where it made sense (to me).

Signed-off-by: Max Krasnyanskiy <maxk@qualcomm.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Gregory Haskins <ghaskins@novell.com>
Cc: dmitry.adamushko@gmail.com
Cc: pj@sgi.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/cpumask.h |   6 ++-
 include/linux/cpuset.h  |   7 ++++
 init/main.c             |   7 ++++
 kernel/cpu.c            |  30 +++++++++++---
 kernel/cpuset.c         |   2 +-
 kernel/sched.c          | 108 ++++++++++++++++++++----------------------------
 kernel/sched_fair.c     |   3 ++
 kernel/sched_rt.c       |   7 ++++
 8 files changed, 99 insertions(+), 71 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index c24875bd9c5b..d614d2472798 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -359,13 +359,14 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
 
 /*
  * The following particular system cpumasks and operations manage
- * possible, present and online cpus.  Each of them is a fixed size
+ * possible, present, active and online cpus.  Each of them is a fixed size
  * bitmap of size NR_CPUS.
  *
  *  #ifdef CONFIG_HOTPLUG_CPU
  *     cpu_possible_map - has bit 'cpu' set iff cpu is populatable
  *     cpu_present_map  - has bit 'cpu' set iff cpu is populated
  *     cpu_online_map   - has bit 'cpu' set iff cpu available to scheduler
+ *     cpu_active_map   - has bit 'cpu' set iff cpu available to migration
  *  #else
  *     cpu_possible_map - has bit 'cpu' set iff cpu is populated
  *     cpu_present_map  - copy of cpu_possible_map
@@ -416,6 +417,7 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
 extern cpumask_t cpu_possible_map;
 extern cpumask_t cpu_online_map;
 extern cpumask_t cpu_present_map;
+extern cpumask_t cpu_active_map;
 
 #if NR_CPUS > 1
 #define num_online_cpus()	cpus_weight(cpu_online_map)
@@ -424,6 +426,7 @@ extern cpumask_t cpu_present_map;
 #define cpu_online(cpu)		cpu_isset((cpu), cpu_online_map)
 #define cpu_possible(cpu)	cpu_isset((cpu), cpu_possible_map)
 #define cpu_present(cpu)	cpu_isset((cpu), cpu_present_map)
+#define cpu_active(cpu)		cpu_isset((cpu), cpu_active_map)
 #else
 #define num_online_cpus()	1
 #define num_possible_cpus()	1
@@ -431,6 +434,7 @@ extern cpumask_t cpu_present_map;
 #define cpu_online(cpu)		((cpu) == 0)
 #define cpu_possible(cpu)	((cpu) == 0)
 #define cpu_present(cpu)	((cpu) == 0)
+#define cpu_active(cpu)		((cpu) == 0)
 #endif
 
 #define cpu_is_offline(cpu)	unlikely(!cpu_online(cpu))
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 038578362b47..e8f450c499b0 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -78,6 +78,8 @@ extern void cpuset_track_online_nodes(void);
 
 extern int current_cpuset_is_being_rebound(void);
 
+extern void rebuild_sched_domains(void);
+
 #else /* !CONFIG_CPUSETS */
 
 static inline int cpuset_init_early(void) { return 0; }
@@ -156,6 +158,11 @@ static inline int current_cpuset_is_being_rebound(void)
 	return 0;
 }
 
+static inline void rebuild_sched_domains(void)
+{
+	partition_sched_domains(0, NULL, NULL);
+}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */
diff --git a/init/main.c b/init/main.c
index edeace036fd9..dd25259530ea 100644
--- a/init/main.c
+++ b/init/main.c
@@ -415,6 +415,13 @@ static void __init smp_init(void)
 {
 	unsigned int cpu;
 
+	/*
+	 * Set up the current CPU as possible to migrate to.
+	 * The other ones will be done by cpu_up/cpu_down()
+	 */
+	cpu = smp_processor_id();
+	cpu_set(cpu, cpu_active_map);
+
 	/* FIXME: This should be done in userspace --RR */
 	for_each_present_cpu(cpu) {
 		if (num_online_cpus() >= setup_max_cpus)
diff --git a/kernel/cpu.c b/kernel/cpu.c
index cfb1d43ab801..a1ac7ea245d7 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -64,6 +64,8 @@ void __init cpu_hotplug_init(void)
 	cpu_hotplug.refcount = 0;
 }
 
+cpumask_t cpu_active_map;
+
 #ifdef CONFIG_HOTPLUG_CPU
 
 void get_online_cpus(void)
@@ -291,11 +293,20 @@ int __ref cpu_down(unsigned int cpu)
 	int err = 0;
 
 	cpu_maps_update_begin();
-	if (cpu_hotplug_disabled)
+
+	if (cpu_hotplug_disabled) {
 		err = -EBUSY;
-	else
-		err = _cpu_down(cpu, 0);
+		goto out;
+	}
+
+	cpu_clear(cpu, cpu_active_map);
+
+	err = _cpu_down(cpu, 0);
+
+	if (cpu_online(cpu))
+		cpu_set(cpu, cpu_active_map);
 
+out:
 	cpu_maps_update_done();
 	return err;
 }
@@ -355,11 +366,18 @@ int __cpuinit cpu_up(unsigned int cpu)
 	}
 
 	cpu_maps_update_begin();
-	if (cpu_hotplug_disabled)
+
+	if (cpu_hotplug_disabled) {
 		err = -EBUSY;
-	else
-		err = _cpu_up(cpu, 0);
+		goto out;
+	}
 
+	err = _cpu_up(cpu, 0);
+
+	if (cpu_online(cpu))
+		cpu_set(cpu, cpu_active_map);
+
+out:
 	cpu_maps_update_done();
 	return err;
 }
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 459d601947a8..3c3ef02f65f1 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -564,7 +564,7 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c)
  *	partition_sched_domains().
  */
 
-static void rebuild_sched_domains(void)
+void rebuild_sched_domains(void)
 {
 	struct kfifo *q;	/* queue of cpusets to be scanned */
 	struct cpuset *cp;	/* scans q */
diff --git a/kernel/sched.c b/kernel/sched.c
index 1ee18dbb4516..c237624a8a04 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2881,7 +2881,7 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu)
 
 	rq = task_rq_lock(p, &flags);
 	if (!cpu_isset(dest_cpu, p->cpus_allowed)
-	    || unlikely(cpu_is_offline(dest_cpu)))
+	    || unlikely(!cpu_active(dest_cpu)))
 		goto out;
 
 	/* force the process onto the specified CPU */
@@ -3849,7 +3849,7 @@ int select_nohz_load_balancer(int stop_tick)
 		/*
 		 * If we are going offline and still the leader, give up!
 		 */
-		if (cpu_is_offline(cpu) &&
+		if (!cpu_active(cpu) &&
 		    atomic_read(&nohz.load_balancer) == cpu) {
 			if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
 				BUG();
@@ -5876,7 +5876,7 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
 	struct rq *rq_dest, *rq_src;
 	int ret = 0, on_rq;
 
-	if (unlikely(cpu_is_offline(dest_cpu)))
+	if (unlikely(!cpu_active(dest_cpu)))
 		return ret;
 
 	rq_src = cpu_rq(src_cpu);
@@ -7553,18 +7553,6 @@ void __attribute__((weak)) arch_update_cpu_topology(void)
 {
 }
 
-/*
- * Free current domain masks.
- * Called after all cpus are attached to NULL domain.
- */
-static void free_sched_domains(void)
-{
-	ndoms_cur = 0;
-	if (doms_cur != &fallback_doms)
-		kfree(doms_cur);
-	doms_cur = &fallback_doms;
-}
-
 /*
  * Set up scheduler domains and groups. Callers must hold the hotplug lock.
  * For now this just excludes isolated cpus, but could be used to
@@ -7643,7 +7631,7 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur,
  * ownership of it and will kfree it when done with it. If the caller
  * failed the kmalloc call, then it can pass in doms_new == NULL,
  * and partition_sched_domains() will fallback to the single partition
- * 'fallback_doms'.
+ * 'fallback_doms', it also forces the domains to be rebuilt.
  *
  * Call with hotplug lock held
  */
@@ -7657,12 +7645,8 @@ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
 	/* always unregister in case we don't destroy any domains */
 	unregister_sched_domain_sysctl();
 
-	if (doms_new == NULL) {
-		ndoms_new = 1;
-		doms_new = &fallback_doms;
-		cpus_andnot(doms_new[0], cpu_online_map, cpu_isolated_map);
-		dattr_new = NULL;
-	}
+	if (doms_new == NULL)
+		ndoms_new = 0;
 
 	/* Destroy deleted domains */
 	for (i = 0; i < ndoms_cur; i++) {
@@ -7677,6 +7661,14 @@ match1:
 		;
 	}
 
+	if (doms_new == NULL) {
+		ndoms_cur = 0;
+		ndoms_new = 1;
+		doms_new = &fallback_doms;
+		cpus_andnot(doms_new[0], cpu_online_map, cpu_isolated_map);
+		dattr_new = NULL;
+	}
+
 	/* Build new domains */
 	for (i = 0; i < ndoms_new; i++) {
 		for (j = 0; j < ndoms_cur; j++) {
@@ -7707,17 +7699,10 @@ match2:
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
 int arch_reinit_sched_domains(void)
 {
-	int err;
-
 	get_online_cpus();
-	mutex_lock(&sched_domains_mutex);
-	detach_destroy_domains(&cpu_online_map);
-	free_sched_domains();
-	err = arch_init_sched_domains(&cpu_online_map);
-	mutex_unlock(&sched_domains_mutex);
+	rebuild_sched_domains();
 	put_online_cpus();
-
-	return err;
+	return 0;
 }
 
 static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
@@ -7783,14 +7768,30 @@ int sched_create_sysfs_power_savings_entries(struct sysdev_class *cls)
 }
 #endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */
 
+#ifndef CONFIG_CPUSETS
 /*
- * Force a reinitialization of the sched domains hierarchy. The domains
- * and groups cannot be updated in place without racing with the balancing
- * code, so we temporarily attach all running cpus to the NULL domain
- * which will prevent rebalancing while the sched domains are recalculated.
+ * Add online and remove offline CPUs from the scheduler domains.
+ * When cpusets are enabled they take over this function.
  */
 static int update_sched_domains(struct notifier_block *nfb,
 				unsigned long action, void *hcpu)
+{
+	switch (action) {
+	case CPU_ONLINE:
+	case CPU_ONLINE_FROZEN:
+	case CPU_DEAD:
+	case CPU_DEAD_FROZEN:
+		partition_sched_domains(0, NULL, NULL);
+		return NOTIFY_OK;
+
+	default:
+		return NOTIFY_DONE;
+	}
+}
+#endif
+
+static int update_runtime(struct notifier_block *nfb,
+				unsigned long action, void *hcpu)
 {
 	int cpu = (int)(long)hcpu;
 
@@ -7798,44 +7799,18 @@ static int update_sched_domains(struct notifier_block *nfb,
 	case CPU_DOWN_PREPARE:
 	case CPU_DOWN_PREPARE_FROZEN:
 		disable_runtime(cpu_rq(cpu));
-		/* fall-through */
-	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
-		detach_destroy_domains(&cpu_online_map);
-		free_sched_domains();
 		return NOTIFY_OK;
 
-
 	case CPU_DOWN_FAILED:
 	case CPU_DOWN_FAILED_FROZEN:
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
 		enable_runtime(cpu_rq(cpu));
-		/* fall-through */
-	case CPU_UP_CANCELED:
-	case CPU_UP_CANCELED_FROZEN:
-	case CPU_DEAD:
-	case CPU_DEAD_FROZEN:
-		/*
-		 * Fall through and re-initialise the domains.
-		 */
-		break;
+		return NOTIFY_OK;
+
 	default:
 		return NOTIFY_DONE;
 	}
-
-#ifndef CONFIG_CPUSETS
-	/*
-	 * Create default domain partitioning if cpusets are disabled.
-	 * Otherwise we let cpusets rebuild the domains based on the
-	 * current setup.
-	 */
-
-	/* The hotplug lock is already held by cpu_up/cpu_down */
-	arch_init_sched_domains(&cpu_online_map);
-#endif
-
-	return NOTIFY_OK;
 }
 
 void __init sched_init_smp(void)
@@ -7855,8 +7830,15 @@ void __init sched_init_smp(void)
 		cpu_set(smp_processor_id(), non_isolated_cpus);
 	mutex_unlock(&sched_domains_mutex);
 	put_online_cpus();
+
+#ifndef CONFIG_CPUSETS
 	/* XXX: Theoretical race here - CPU may be hotplugged now */
 	hotcpu_notifier(update_sched_domains, 0);
+#endif
+
+	/* RT runtime code needs to handle some hotplug events */
+	hotcpu_notifier(update_runtime, 0);
+
 	init_hrtick();
 
 	/* Move init over to a non-isolated CPU */
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index f2aa987027d6..d924c679dfac 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1004,6 +1004,8 @@ static void yield_task_fair(struct rq *rq)
  * not idle and an idle cpu is available.  The span of cpus to
  * search starts with cpus closest then further out as needed,
  * so we always favor a closer, idle cpu.
+ * Domains may include CPUs that are not usable for migration,
+ * hence we need to mask them out (cpu_active_map)
  *
  * Returns the CPU we should wake onto.
  */
@@ -1031,6 +1033,7 @@ static int wake_idle(int cpu, struct task_struct *p)
 		    || ((sd->flags & SD_WAKE_IDLE_FAR)
 			&& !task_hot(p, task_rq(p)->clock, sd))) {
 			cpus_and(tmp, sd->span, p->cpus_allowed);
+			cpus_and(tmp, tmp, cpu_active_map);
 			for_each_cpu_mask(i, tmp) {
 				if (idle_cpu(i)) {
 					if (i != task_cpu(p)) {
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index d3d1cccb3d7b..50735bb96149 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -933,6 +933,13 @@ static int find_lowest_rq(struct task_struct *task)
 	if (!cpupri_find(&task_rq(task)->rd->cpupri, task, lowest_mask))
 		return -1; /* No targets found */
 
+	/*
+	 * Only consider CPUs that are usable for migration.
+	 * I guess we might want to change cpupri_find() to ignore those
+	 * in the first place.
+	 */
+	cpus_and(*lowest_mask, *lowest_mask, cpu_active_map);
+
 	/*
 	 * At this point we have built a mask of cpus representing the
 	 * lowest priority tasks in the system.  Now we want to elect
-- 
cgit v1.2.3


From 1b427c153a08fdbc092c2bdbf845b92fda58d857 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 18 Jul 2008 14:01:39 +0200
Subject: sched: fix build error, provide partition_sched_domains()
 unconditionally
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

provide an empty partition_sched_domains() definition for the UP case:

 include/linux/cpuset.h: In function ‘rebuild_sched_domains':
 include/linux/cpuset.h:163: error: implicit declaration of function ‘partition_sched_domains'

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1941d8b5cf11..26da921530fe 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -824,7 +824,16 @@ extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
 				    struct sched_domain_attr *dattr_new);
 extern int arch_reinit_sched_domains(void);
 
-#endif	/* CONFIG_SMP */
+#else /* CONFIG_SMP */
+
+struct sched_domain_attr;
+
+static inline void
+partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+			struct sched_domain_attr *dattr_new)
+{
+}
+#endif	/* !CONFIG_SMP */
 
 struct io_context;			/* See blkdev.h */
 #define NGROUPS_SMALL		32
-- 
cgit v1.2.3


From 65c011845316d3c1381f478ca0d8265c43b3b039 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Tue, 15 Jul 2008 14:14:30 -0700
Subject: cpumask: Replace cpumask_of_cpu with cpumask_of_cpu_ptr

  * This patch replaces the dangerous lvalue version of cpumask_of_cpu
    with new cpumask_of_cpu_ptr macros.  These are patterned after the
    node_to_cpumask_ptr macros.

    In general terms, if there is a cpumask_of_cpu_map[] then a pointer to
    the cpumask_of_cpu_map[cpu] entry is used.  The cpumask_of_cpu_map
    is provided when there is a large NR_CPUS count, reducing
    greatly the amount of code generated and stack space used for
    cpumask_of_cpu().  The pointer to the cpumask_t value is needed for
    calling set_cpus_allowed_ptr() to reduce the amount of stack space
    needed to pass the cpumask_t value.

    If there isn't a cpumask_of_cpu_map[], then a temporary variable is
    declared and filled in with value from cpumask_of_cpu(cpu) as well as
    a pointer variable pointing to this temporary variable.  Afterwards,
    the pointer is used to reference the cpumask value.  The compiler
    will optimize out the extra dereference through the pointer as well
    as the stack space used for the pointer, resulting in identical code.

    A good example of the orthogonal usages is in net/sunrpc/svc.c:

	case SVC_POOL_PERCPU:
	{
		unsigned int cpu = m->pool_to[pidx];
		cpumask_of_cpu_ptr(cpumask, cpu);

		*oldmask = current->cpus_allowed;
		set_cpus_allowed_ptr(current, cpumask);
		return 1;
	}
	case SVC_POOL_PERNODE:
	{
		unsigned int node = m->pool_to[pidx];
		node_to_cpumask_ptr(nodecpumask, node);

		*oldmask = current->cpus_allowed;
		set_cpus_allowed_ptr(current, nodecpumask);
		return 1;
	}

Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/cstate.c                    |  3 ++-
 arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c       | 10 +++++---
 arch/x86/kernel/cpu/cpufreq/powernow-k8.c        | 15 +++++++----
 arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c |  9 ++++---
 arch/x86/kernel/cpu/cpufreq/speedstep-ich.c      |  3 ++-
 arch/x86/kernel/cpu/intel_cacheinfo.c            |  3 ++-
 arch/x86/kernel/microcode.c                      | 13 +++++++---
 arch/x86/kernel/reboot.c                         | 14 +++++++----
 drivers/acpi/processor_throttling.c              | 11 +++++---
 drivers/firmware/dcdbas.c                        |  3 ++-
 include/linux/cpumask.h                          | 32 ++++++++++++++++++++----
 kernel/stop_machine.c                            |  3 ++-
 kernel/trace/trace_sysprof.c                     |  4 ++-
 net/sunrpc/svc.c                                 |  3 ++-
 14 files changed, 91 insertions(+), 35 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c
index c2502eb9aa83..9220cf46aa10 100644
--- a/arch/x86/kernel/acpi/cstate.c
+++ b/arch/x86/kernel/acpi/cstate.c
@@ -73,6 +73,7 @@ int acpi_processor_ffh_cstate_probe(unsigned int cpu,
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr(new_mask, cpu);
 	int retval;
 	unsigned int eax, ebx, ecx, edx;
 	unsigned int edx_part;
@@ -91,7 +92,7 @@ int acpi_processor_ffh_cstate_probe(unsigned int cpu,
 
 	/* Make sure we are running on right CPU */
 	saved_mask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	retval = set_cpus_allowed_ptr(current, new_mask);
 	if (retval)
 		return -1;
 
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index dd097b835839..ff2fff56f0a8 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -200,10 +200,12 @@ static void drv_read(struct drv_cmd *cmd)
 static void drv_write(struct drv_cmd *cmd)
 {
 	cpumask_t saved_mask = current->cpus_allowed;
+	cpumask_of_cpu_ptr_declare(cpu_mask);
 	unsigned int i;
 
 	for_each_cpu_mask_nr(i, cmd->mask) {
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(i));
+		cpumask_of_cpu_ptr_next(cpu_mask, i);
+		set_cpus_allowed_ptr(current, cpu_mask);
 		do_drv_write(cmd);
 	}
 
@@ -267,11 +269,12 @@ static unsigned int get_measured_perf(unsigned int cpu)
 	} aperf_cur, mperf_cur;
 
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	unsigned int perf_percent;
 	unsigned int retval;
 
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, cpu_mask);
 	if (get_cpu() != cpu) {
 		/* We were not able to run on requested processor */
 		put_cpu();
@@ -337,6 +340,7 @@ static unsigned int get_measured_perf(unsigned int cpu)
 
 static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 {
+	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	struct acpi_cpufreq_data *data = per_cpu(drv_data, cpu);
 	unsigned int freq;
 	unsigned int cached_freq;
@@ -349,7 +353,7 @@ static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 	}
 
 	cached_freq = data->freq_table[data->acpi_data->state].frequency;
-	freq = extract_freq(get_cur_val(&cpumask_of_cpu(cpu)), data);
+	freq = extract_freq(get_cur_val(cpu_mask), data);
 	if (freq != cached_freq) {
 		/*
 		 * The dreaded BIOS frequency change behind our back.
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index c45ca6d4dce1..53c7b6936973 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -479,11 +479,12 @@ static int core_voltage_post_transition(struct powernow_k8_data *data, u32 reqvi
 static int check_supported_cpu(unsigned int cpu)
 {
 	cpumask_t oldmask;
+	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	u32 eax, ebx, ecx, edx;
 	unsigned int rc = 0;
 
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, cpu_mask);
 
 	if (smp_processor_id() != cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", cpu);
@@ -1016,6 +1017,7 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned i
 static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsigned relation)
 {
 	cpumask_t oldmask;
+	cpumask_of_cpu_ptr(cpu_mask, pol->cpu);
 	struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu);
 	u32 checkfid;
 	u32 checkvid;
@@ -1030,7 +1032,7 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi
 
 	/* only run on specific CPU from here on */
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu));
+	set_cpus_allowed_ptr(current, cpu_mask);
 
 	if (smp_processor_id() != pol->cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu);
@@ -1105,6 +1107,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 {
 	struct powernow_k8_data *data;
 	cpumask_t oldmask;
+	cpumask_of_cpu_ptr_declare(newmask);
 	int rc;
 
 	if (!cpu_online(pol->cpu))
@@ -1156,7 +1159,8 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 
 	/* only run on specific CPU from here on */
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu));
+	cpumask_of_cpu_ptr_next(newmask, pol->cpu);
+	set_cpus_allowed_ptr(current, newmask);
 
 	if (smp_processor_id() != pol->cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu);
@@ -1178,7 +1182,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 	set_cpus_allowed_ptr(current, &oldmask);
 
 	if (cpu_family == CPU_HW_PSTATE)
-		pol->cpus = cpumask_of_cpu(pol->cpu);
+		pol->cpus = *newmask;
 	else
 		pol->cpus = per_cpu(cpu_core_map, pol->cpu);
 	data->available_cores = &(pol->cpus);
@@ -1244,6 +1248,7 @@ static unsigned int powernowk8_get (unsigned int cpu)
 {
 	struct powernow_k8_data *data;
 	cpumask_t oldmask = current->cpus_allowed;
+	cpumask_of_cpu_ptr(newmask, cpu);
 	unsigned int khz = 0;
 	unsigned int first;
 
@@ -1253,7 +1258,7 @@ static unsigned int powernowk8_get (unsigned int cpu)
 	if (!data)
 		return -EINVAL;
 
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, newmask);
 	if (smp_processor_id() != cpu) {
 		printk(KERN_ERR PFX
 			"limiting to CPU %d failed in powernowk8_get\n", cpu);
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index 8b0dd6f2a1ac..fd561bb26c60 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -313,9 +313,10 @@ static unsigned int get_cur_freq(unsigned int cpu)
 	unsigned l, h;
 	unsigned clock_freq;
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr(new_mask, cpu);
 
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, new_mask);
 	if (smp_processor_id() != cpu)
 		return 0;
 
@@ -554,9 +555,11 @@ static int centrino_target (struct cpufreq_policy *policy,
 		 */
 
 		if (!cpus_empty(covered_cpus)) {
+			cpumask_of_cpu_ptr_declare(new_mask);
+
 			for_each_cpu_mask_nr(j, covered_cpus) {
-				set_cpus_allowed_ptr(current,
-						     &cpumask_of_cpu(j));
+				cpumask_of_cpu_ptr_next(new_mask, j);
+				set_cpus_allowed_ptr(current, new_mask);
 				wrmsr(MSR_IA32_PERF_CTL, oldmsr, h);
 			}
 		}
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
index 191f7263c61d..2f3728dc24f6 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
@@ -244,7 +244,8 @@ static unsigned int _speedstep_get(const cpumask_t *cpus)
 
 static unsigned int speedstep_get(unsigned int cpu)
 {
-	return _speedstep_get(&cpumask_of_cpu(cpu));
+	cpumask_of_cpu_ptr(newmask, cpu);
+	return _speedstep_get(newmask);
 }
 
 /**
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index a7b0f8f1736b..e4b8d189d7ed 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -516,6 +516,7 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 	unsigned long		j;
 	int			retval;
 	cpumask_t		oldmask;
+	cpumask_of_cpu_ptr(newmask, cpu);
 
 	if (num_cache_leaves == 0)
 		return -ENOENT;
@@ -526,7 +527,7 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 		return -ENOMEM;
 
 	oldmask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	retval = set_cpus_allowed_ptr(current, newmask);
 	if (retval)
 		goto out;
 
diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c
index 56b933119a04..58520169e35d 100644
--- a/arch/x86/kernel/microcode.c
+++ b/arch/x86/kernel/microcode.c
@@ -388,6 +388,7 @@ static int do_microcode_update (void)
 	void *new_mc = NULL;
 	int cpu;
 	cpumask_t old;
+	cpumask_of_cpu_ptr_declare(newmask);
 
 	old = current->cpus_allowed;
 
@@ -404,7 +405,8 @@ static int do_microcode_update (void)
 
 			if (!uci->valid)
 				continue;
-			set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+			cpumask_of_cpu_ptr_next(newmask, cpu);
+			set_cpus_allowed_ptr(current, newmask);
 			error = get_maching_microcode(new_mc, cpu);
 			if (error < 0)
 				goto out;
@@ -574,6 +576,7 @@ static int apply_microcode_check_cpu(int cpu)
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 	cpumask_t old;
+	cpumask_of_cpu_ptr(newmask, cpu);
 	unsigned int val[2];
 	int err = 0;
 
@@ -582,7 +585,7 @@ static int apply_microcode_check_cpu(int cpu)
 		return 0;
 
 	old = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, newmask);
 
 	/* Check if the microcode we have in memory matches the CPU */
 	if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
@@ -620,11 +623,12 @@ static int apply_microcode_check_cpu(int cpu)
 static void microcode_init_cpu(int cpu, int resume)
 {
 	cpumask_t old;
+	cpumask_of_cpu_ptr(newmask, cpu);
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 
 	old = current->cpus_allowed;
 
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	set_cpus_allowed_ptr(current, newmask);
 	mutex_lock(&microcode_mutex);
 	collect_cpu_info(cpu);
 	if (uci->valid && system_state == SYSTEM_RUNNING && !resume)
@@ -656,11 +660,12 @@ static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz)
 		return -EINVAL;
 	if (val == 1) {
 		cpumask_t old;
+		cpumask_of_cpu_ptr(newmask, cpu);
 
 		old = current->cpus_allowed;
 
 		get_online_cpus();
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+		set_cpus_allowed_ptr(current, newmask);
 
 		mutex_lock(&microcode_mutex);
 		if (uci->valid)
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index f8a62160e151..214bbdfc851e 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -403,24 +403,28 @@ void native_machine_shutdown(void)
 {
 	/* Stop the cpus and apics */
 #ifdef CONFIG_SMP
-	int reboot_cpu_id;
 
 	/* The boot cpu is always logical cpu 0 */
-	reboot_cpu_id = 0;
+	int reboot_cpu_id = 0;
+	cpumask_of_cpu_ptr(newmask, reboot_cpu_id);
 
 #ifdef CONFIG_X86_32
 	/* See if there has been given a command line override */
 	if ((reboot_cpu != -1) && (reboot_cpu < NR_CPUS) &&
-		cpu_online(reboot_cpu))
+		cpu_online(reboot_cpu)) {
 		reboot_cpu_id = reboot_cpu;
+		cpumask_of_cpu_ptr_next(newmask, reboot_cpu_id);
+	}
 #endif
 
 	/* Make certain the cpu I'm about to reboot on is online */
-	if (!cpu_online(reboot_cpu_id))
+	if (!cpu_online(reboot_cpu_id)) {
 		reboot_cpu_id = smp_processor_id();
+		cpumask_of_cpu_ptr_next(newmask, reboot_cpu_id);
+	}
 
 	/* Make certain I only run on the appropriate processor */
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(reboot_cpu_id));
+	set_cpus_allowed_ptr(current, newmask);
 
 	/* O.K Now that I'm on the appropriate processor,
 	 * stop all of the others.
diff --git a/drivers/acpi/processor_throttling.c b/drivers/acpi/processor_throttling.c
index a56fc6c4394b..a2c3f9cfa549 100644
--- a/drivers/acpi/processor_throttling.c
+++ b/drivers/acpi/processor_throttling.c
@@ -827,6 +827,7 @@ static int acpi_processor_get_throttling_ptc(struct acpi_processor *pr)
 static int acpi_processor_get_throttling(struct acpi_processor *pr)
 {
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr_declare(new_mask);
 	int ret;
 
 	if (!pr)
@@ -838,7 +839,8 @@ static int acpi_processor_get_throttling(struct acpi_processor *pr)
 	 * Migrate task to the cpu pointed by pr.
 	 */
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pr->id));
+	cpumask_of_cpu_ptr_next(new_mask, pr->id);
+	set_cpus_allowed_ptr(current, new_mask);
 	ret = pr->throttling.acpi_processor_get_throttling(pr);
 	/* restore the previous state */
 	set_cpus_allowed_ptr(current, &saved_mask);
@@ -987,6 +989,7 @@ static int acpi_processor_set_throttling_ptc(struct acpi_processor *pr,
 int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 {
 	cpumask_t saved_mask;
+	cpumask_of_cpu_ptr_declare(new_mask);
 	int ret = 0;
 	unsigned int i;
 	struct acpi_processor *match_pr;
@@ -1025,7 +1028,8 @@ int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 	 * it can be called only for the cpu pointed by pr.
 	 */
 	if (p_throttling->shared_type == DOMAIN_COORD_TYPE_SW_ANY) {
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(pr->id));
+		cpumask_of_cpu_ptr_next(new_mask, pr->id);
+		set_cpus_allowed_ptr(current, new_mask);
 		ret = p_throttling->acpi_processor_set_throttling(pr,
 						t_state.target_state);
 	} else {
@@ -1056,7 +1060,8 @@ int acpi_processor_set_throttling(struct acpi_processor *pr, int state)
 				continue;
 			}
 			t_state.cpu = i;
-			set_cpus_allowed_ptr(current, &cpumask_of_cpu(i));
+			cpumask_of_cpu_ptr_next(new_mask, i);
+			set_cpus_allowed_ptr(current, new_mask);
 			ret = match_pr->throttling.
 				acpi_processor_set_throttling(
 				match_pr, t_state.target_state);
diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c
index 25918f7dfd0f..0b624e927a6f 100644
--- a/drivers/firmware/dcdbas.c
+++ b/drivers/firmware/dcdbas.c
@@ -254,6 +254,7 @@ static ssize_t host_control_on_shutdown_store(struct device *dev,
 static int smi_request(struct smi_cmd *smi_cmd)
 {
 	cpumask_t old_mask;
+	cpumask_of_cpu_ptr(new_mask, 0);
 	int ret = 0;
 
 	if (smi_cmd->magic != SMI_CMD_MAGIC) {
@@ -264,7 +265,7 @@ static int smi_request(struct smi_cmd *smi_cmd)
 
 	/* SMI requires CPU 0 */
 	old_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(0));
+	set_cpus_allowed_ptr(current, new_mask);
 	if (smp_processor_id() != 0) {
 		dev_dbg(&dcdbas_pdev->dev, "%s: failed to get CPU 0\n",
 			__func__);
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 80226e776143..2dbd9a287e77 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -62,6 +62,15 @@
  * int next_cpu_nr(cpu, mask)		Next cpu past 'cpu', or nr_cpu_ids
  *
  * cpumask_t cpumask_of_cpu(cpu)	Return cpumask with bit 'cpu' set
+ *ifdef CONFIG_HAS_CPUMASK_OF_CPU
+ * cpumask_of_cpu_ptr_declare(v)	Declares cpumask_t *v
+ * cpumask_of_cpu_ptr_next(v, cpu)	Sets v = &cpumask_of_cpu_map[cpu]
+ * cpumask_of_cpu_ptr(v, cpu)		Combines above two operations
+ *else
+ * cpumask_of_cpu_ptr_declare(v)	Declares cpumask_t _v and *v = &_v
+ * cpumask_of_cpu_ptr_next(v, cpu)	Sets _v = cpumask_of_cpu(cpu)
+ * cpumask_of_cpu_ptr(v, cpu)		Combines above two operations
+ *endif
  * CPU_MASK_ALL				Initializer - all bits set
  * CPU_MASK_NONE			Initializer - no bits set
  * unsigned long *cpus_addr(mask)	Array of unsigned long's in mask
@@ -236,11 +245,16 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 
 #ifdef CONFIG_HAVE_CPUMASK_OF_CPU_MAP
 extern cpumask_t *cpumask_of_cpu_map;
-#define cpumask_of_cpu(cpu)    (cpumask_of_cpu_map[cpu])
-
+#define cpumask_of_cpu(cpu)	(cpumask_of_cpu_map[cpu])
+#define	cpumask_of_cpu_ptr(v, cpu)					\
+		const cpumask_t *v = &cpumask_of_cpu(cpu)
+#define	cpumask_of_cpu_ptr_declare(v)					\
+		const cpumask_t *v
+#define cpumask_of_cpu_ptr_next(v, cpu)					\
+					v = &cpumask_of_cpu(cpu)
 #else
 #define cpumask_of_cpu(cpu)						\
-(*({									\
+({									\
 	typeof(_unused_cpumask_arg_) m;					\
 	if (sizeof(m) == sizeof(unsigned long)) {			\
 		m.bits[0] = 1UL<<(cpu);					\
@@ -248,8 +262,16 @@ extern cpumask_t *cpumask_of_cpu_map;
 		cpus_clear(m);						\
 		cpu_set((cpu), m);					\
 	}								\
-	&m;								\
-}))
+	m;								\
+})
+#define	cpumask_of_cpu_ptr(v, cpu) 					\
+		cpumask_t _##v = cpumask_of_cpu(cpu);			\
+		const cpumask_t *v = &_##v
+#define	cpumask_of_cpu_ptr_declare(v)					\
+		cpumask_t _##v;						\
+		const cpumask_t *v = &_##v
+#define cpumask_of_cpu_ptr_next(v, cpu)					\
+					_##v = cpumask_of_cpu(cpu)
 #endif
 
 #define CPU_MASK_LAST_WORD BITMAP_LAST_WORD_MASK(NR_CPUS)
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index ba9b2054ecbd..738b411ff2d3 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -33,8 +33,9 @@ static int stopmachine(void *cpu)
 {
 	int irqs_disabled = 0;
 	int prepared = 0;
+	cpumask_of_cpu_ptr(cpumask, (int)(long)cpu);
 
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu((int)(long)cpu));
+	set_cpus_allowed_ptr(current, cpumask);
 
 	/* Ack: we are alive */
 	smp_mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */
diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c
index 2301e1e7c606..63528086337c 100644
--- a/kernel/trace/trace_sysprof.c
+++ b/kernel/trace/trace_sysprof.c
@@ -213,7 +213,9 @@ static void start_stack_timers(void)
 	int cpu;
 
 	for_each_online_cpu(cpu) {
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+		cpumask_of_cpu_ptr(new_mask, cpu);
+
+		set_cpus_allowed_ptr(current, new_mask);
 		start_stack_timer(cpu);
 	}
 	set_cpus_allowed_ptr(current, &saved_mask);
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index d43cf8ddff67..083d12688134 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -314,9 +314,10 @@ svc_pool_map_set_cpumask(unsigned int pidx, cpumask_t *oldmask)
 	case SVC_POOL_PERCPU:
 	{
 		unsigned int cpu = m->pool_to[pidx];
+		cpumask_of_cpu_ptr(cpumask, cpu);
 
 		*oldmask = current->cpus_allowed;
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+		set_cpus_allowed_ptr(current, cpumask);
 		return 1;
 	}
 	case SVC_POOL_PERNODE:
-- 
cgit v1.2.3


From 77586c2bdad0798cb24e35de5a878e7c6b200574 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Tue, 15 Jul 2008 14:14:36 -0700
Subject: cpumask: Provide a generic set of CPUMASK_ALLOC macros

  * Provide a generic set of CPUMASK_ALLOC macros patterned after the
    SCHED_CPUMASK_ALLOC macros.  This is used where multiple cpumask_t
    variables are declared on the stack to reduce the amount of stack
    space required.

Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/cpumask.h | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 2dbd9a287e77..72f9c32c12b0 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -75,6 +75,17 @@
  * CPU_MASK_NONE			Initializer - no bits set
  * unsigned long *cpus_addr(mask)	Array of unsigned long's in mask
  *
+ *if NR_CPUS > BITS_PER_LONG
+ *   CPUMASK_ALLOC(m)			Declares and allocates struct m *m =
+ *					   (struct m *)kmalloc(sizeof(*m), ...)
+ *   CPUMASK_FREE(m)			Macro for kfree(v)
+ *else
+ *   CPUMASK_ALLOC(m)			Declares struct m _m, *m = &_m
+ *   CPUMASK_FREE(m)			Nop
+ *endif
+ *   CPUMASK_VAR(v, m)			Declares cpumask_t *v =
+ *						m + offset(struct m, v)
+ *
  * int cpumask_scnprintf(buf, len, mask) Format cpumask for printing
  * int cpumask_parse_user(ubuf, ulen, mask)	Parse ascii string as cpumask
  * int cpulist_scnprintf(buf, len, mask) Format cpumask as list for printing
@@ -311,6 +322,16 @@ extern cpumask_t cpu_mask_all;
 
 #define cpus_addr(src) ((src).bits)
 
+#if NR_CPUS > BITS_PER_LONG
+#define	CPUMASK_ALLOC(m)	struct m *m = kmalloc(sizeof(*m), GFP_KERNEL)
+#define	CPUMASK_FREE(m)		kfree(m)
+#else
+#define	CPUMASK_ALLOC(m)	struct allmasks _m, *m = &_m
+#define	CPUMASK_FREE(m)
+#endif
+#define	CPUMASK_VAR(v, m) 	cpumask_t *v = (cpumask_t *)		\
+				((unsigned long)(m) + offsetof(struct m, v))
+
 #define cpumask_scnprintf(buf, len, src) \
 			__cpumask_scnprintf((buf), (len), &(src), NR_CPUS)
 static inline int __cpumask_scnprintf(char *buf, int len,
-- 
cgit v1.2.3


From ae79cdaacb5599781f8bb49f4bdd5723029669cf Mon Sep 17 00:00:00 2001
From: "venkatesh.pallipadi@intel.com" <venkatesh.pallipadi@intel.com>
Date: Fri, 18 Jul 2008 16:08:13 -0700
Subject: x86: Add a arch directory for x86 under debugfs

Add a directory for x86 arch under debugfs. Can be used to accumulate all
x86 specific debugfs files.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/kdebugfs.c | 8 ++++++++
 include/linux/debugfs.h    | 2 ++
 2 files changed, 10 insertions(+)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/kdebugfs.c b/arch/x86/kernel/kdebugfs.c
index c03205991718..f2d43bc75514 100644
--- a/arch/x86/kernel/kdebugfs.c
+++ b/arch/x86/kernel/kdebugfs.c
@@ -12,9 +12,13 @@
 #include <linux/init.h>
 #include <linux/io.h>
 #include <linux/mm.h>
+#include <linux/module.h>
 
 #include <asm/setup.h>
 
+struct dentry *arch_debugfs_dir;
+EXPORT_SYMBOL(arch_debugfs_dir);
+
 #ifdef CONFIG_DEBUG_BOOT_PARAMS
 struct setup_data_node {
 	u64 paddr;
@@ -209,6 +213,10 @@ static int __init arch_kdebugfs_init(void)
 {
 	int error = 0;
 
+	arch_debugfs_dir = debugfs_create_dir("x86", NULL);
+	if (!arch_debugfs_dir)
+		return -ENOMEM;
+
 #ifdef CONFIG_DEBUG_BOOT_PARAMS
 	error = boot_params_kdebugfs_init();
 #endif
diff --git a/include/linux/debugfs.h b/include/linux/debugfs.h
index 7266124361b4..32755cdf68db 100644
--- a/include/linux/debugfs.h
+++ b/include/linux/debugfs.h
@@ -26,6 +26,8 @@ struct debugfs_blob_wrapper {
 	unsigned long size;
 };
 
+extern struct dentry *arch_debugfs_dir;
+
 #if defined(CONFIG_DEBUG_FS)
 
 /* declared over in file.c */
-- 
cgit v1.2.3


From 8913336a7e8d56e984109a3137d6c0e3362596a4 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Fri, 18 Jul 2008 18:05:19 -0700
Subject: packet: add PACKET_RESERVE sockopt

Add new sockopt to reserve some headroom in the mmaped ring frames in
front of the packet payload. This can be used f.i. when the VLAN header
needs to be (re)constructed to avoid moving the entire payload.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_packet.h |  1 +
 net/packet/af_packet.c    | 29 ++++++++++++++++++++++++++---
 2 files changed, 27 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h
index a630295b255f..18db0668065a 100644
--- a/include/linux/if_packet.h
+++ b/include/linux/if_packet.h
@@ -45,6 +45,7 @@ struct sockaddr_ll
 #define PACKET_ORIGDEV			9
 #define PACKET_VERSION			10
 #define PACKET_HDRLEN			11
+#define PACKET_RESERVE			12
 
 struct tpacket_stats
 {
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index db792e02a37f..de73bcb5235a 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -188,6 +188,7 @@ struct packet_sock {
 	unsigned int		pg_vec_len;
 	enum tpacket_versions	tp_version;
 	unsigned int		tp_hdrlen;
+	unsigned int		tp_reserve;
 #endif
 };
 
@@ -635,11 +636,13 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
 		snaplen = res;
 
 	if (sk->sk_type == SOCK_DGRAM) {
-		macoff = netoff = TPACKET_ALIGN(po->tp_hdrlen) + 16;
+		macoff = netoff = TPACKET_ALIGN(po->tp_hdrlen) + 16 +
+				  po->tp_reserve;
 	} else {
 		unsigned maclen = skb_network_offset(skb);
 		netoff = TPACKET_ALIGN(po->tp_hdrlen +
-				       (maclen < 16 ? 16 : maclen));
+				       (maclen < 16 ? 16 : maclen)) +
+			po->tp_reserve;
 		macoff = netoff - maclen;
 	}
 
@@ -1448,6 +1451,19 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv
 			return -EINVAL;
 		}
 	}
+	case PACKET_RESERVE:
+	{
+		unsigned int val;
+
+		if (optlen != sizeof(val))
+			return -EINVAL;
+		if (po->pg_vec)
+			return -EBUSY;
+		if (copy_from_user(&val, optval, sizeof(val)))
+			return -EFAULT;
+		po->tp_reserve = val;
+		return 0;
+	}
 #endif
 	case PACKET_AUXDATA:
 	{
@@ -1547,6 +1563,12 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 		}
 		data = &val;
 		break;
+	case PACKET_RESERVE:
+		if (len > sizeof(unsigned int))
+			len = sizeof(unsigned int);
+		val = po->tp_reserve;
+		data = &val;
+		break;
 #endif
 	default:
 		return -ENOPROTOOPT;
@@ -1790,7 +1812,8 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing
 			return -EINVAL;
 		if (unlikely(req->tp_block_size & (PAGE_SIZE - 1)))
 			return -EINVAL;
-		if (unlikely(req->tp_frame_size < po->tp_hdrlen))
+		if (unlikely(req->tp_frame_size < po->tp_hdrlen +
+						  po->tp_reserve))
 			return -EINVAL;
 		if (unlikely(req->tp_frame_size & (TPACKET_ALIGNMENT - 1)))
 			return -EINVAL;
-- 
cgit v1.2.3


From 5ec461d083066441c1a280aa9e13d3bb5c3e008a Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Sat, 19 Jul 2008 00:40:25 -0400
Subject: Input: add microphone insert switch definition

Add a new switch type to the input API for reporting microphone
insertion. This will be used by the ALSA jack reporting API.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 include/linux/input.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/input.h b/include/linux/input.h
index 3e86b61d2e4c..1be28cbbc57b 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -642,6 +642,7 @@ struct input_absinfo {
 #define SW_RFKILL_ALL		0x03  /* rfkill master switch, type "any"
 					 set = radio enabled */
 #define SW_RADIO		SW_RFKILL_ALL	/* deprecated */
+#define SW_MICROPHONE_INSERT	0x04  /* set = inserted */
 #define SW_MAX			0x0f
 #define SW_CNT			(SW_MAX+1)
 
-- 
cgit v1.2.3


From 92c49890922d54cba4b1eadeb0b185773c2c9570 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg59@srcf.ucam.org>
Date: Sat, 19 Jul 2008 00:43:54 -0400
Subject: Input: add switch for dock events

Add a SW_DOCK switch to input.h.  ACPI docks currently send their docking
status as a uevent, but not all docks are ACPI or correspond to a device.
In that case, it makes more sense to simply generate an input event on
docking or undocking.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 include/linux/input.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/input.h b/include/linux/input.h
index 1be28cbbc57b..feaea76f0ef3 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -643,6 +643,7 @@ struct input_absinfo {
 					 set = radio enabled */
 #define SW_RADIO		SW_RFKILL_ALL	/* deprecated */
 #define SW_MICROPHONE_INSERT	0x04  /* set = inserted */
+#define SW_DOCK			0x05  /* set = plugged into dock */
 #define SW_MAX			0x0f
 #define SW_CNT			(SW_MAX+1)
 
-- 
cgit v1.2.3


From 3072367300aa8c779e3a14ee8e89de079e90f3ad Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 18 Jul 2008 22:50:15 -0700
Subject: pkt_sched: Manage qdisc list inside of root qdisc.

Idea is from Patrick McHardy.

Instead of managing the list of qdiscs on the device level, manage it
in the root qdisc of a netdev_queue.  This solves all kinds of
visibility issues during qdisc destruction.

The way to iterate over all qdiscs of a netdev_queue is to visit
the netdev_queue->qdisc, and then traverse it's list.

The only special case is to ignore builting qdiscs at the root when
dumping or doing a qdisc_lookup().  That was not needed previously
because builtin qdiscs were not added to the device's qdisc_list.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |   2 -
 net/core/dev.c            |   2 -
 net/sched/sch_api.c       | 175 ++++++++++++++++++++++++++++++++++------------
 net/sched/sch_generic.c   |  10 +--
 4 files changed, 133 insertions(+), 56 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9c5a68850114..812bcd8b4363 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -636,8 +636,6 @@ struct net_device
 	unsigned int		real_num_tx_queues;
 
 	unsigned long		tx_queue_len;	/* Max frames per queue allowed */
-	spinlock_t		qdisc_list_lock;
-	struct list_head	qdisc_list;
 
 /*
  * One part is mostly used on xmit path (device)
diff --git a/net/core/dev.c b/net/core/dev.c
index e54acde839da..065b9817e209 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3888,8 +3888,6 @@ int register_netdevice(struct net_device *dev)
 	net = dev_net(dev);
 
 	spin_lock_init(&dev->addr_list_lock);
-	spin_lock_init(&dev->qdisc_list_lock);
-	INIT_LIST_HEAD(&dev->qdisc_list);
 	netdev_init_queue_locks(dev);
 
 	dev->iflink = -1;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index b3ef8307204e..fb43731c9860 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -185,11 +185,20 @@ EXPORT_SYMBOL(unregister_qdisc);
 
 struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
 {
-	struct Qdisc *q;
+	unsigned int i;
+
+	for (i = 0; i < dev->num_tx_queues; i++) {
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
+		struct Qdisc *q, *txq_root = txq->qdisc;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
-		if (q->handle == handle)
-			return q;
+		if (!(txq_root->flags & TCQ_F_BUILTIN) &&
+		    txq_root->handle == handle)
+			return txq_root;
+
+		list_for_each_entry(q, &txq_root->list, list) {
+			if (q->handle == handle)
+				return q;
+		}
 	}
 	return NULL;
 }
@@ -676,9 +685,8 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 				goto err_out3;
 			}
 		}
-		spin_lock_bh(&dev->qdisc_list_lock);
-		list_add_tail(&sch->list, &dev->qdisc_list);
-		spin_unlock_bh(&dev->qdisc_list_lock);
+		if (parent)
+			list_add_tail(&sch->list, &dev_queue->qdisc->list);
 
 		return sch;
 	}
@@ -1037,13 +1045,57 @@ err_out:
 	return -EINVAL;
 }
 
+static bool tc_qdisc_dump_ignore(struct Qdisc *q)
+{
+	return (q->flags & TCQ_F_BUILTIN) ? true : false;
+}
+
+static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb,
+			      struct netlink_callback *cb,
+			      int *q_idx_p, int s_q_idx)
+{
+	int ret = 0, q_idx = *q_idx_p;
+	struct Qdisc *q;
+
+	if (!root)
+		return 0;
+
+	q = root;
+	if (q_idx < s_q_idx) {
+		q_idx++;
+	} else {
+		if (!tc_qdisc_dump_ignore(q) &&
+		    tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
+				  cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
+			goto done;
+		q_idx++;
+	}
+	list_for_each_entry(q, &root->list, list) {
+		if (q_idx < s_q_idx) {
+			q_idx++;
+			continue;
+		}
+		if (!tc_qdisc_dump_ignore(q) && 
+		    tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
+				  cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
+			goto done;
+		q_idx++;
+	}
+
+out:
+	*q_idx_p = q_idx;
+	return ret;
+done:
+	ret = -1;
+	goto out;
+}
+
 static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
 	int idx, q_idx;
 	int s_idx, s_q_idx;
 	struct net_device *dev;
-	struct Qdisc *q;
 
 	if (net != &init_net)
 		return 0;
@@ -1053,21 +1105,22 @@ static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb)
 	read_lock(&dev_base_lock);
 	idx = 0;
 	for_each_netdev(&init_net, dev) {
+		struct netdev_queue *dev_queue;
+
 		if (idx < s_idx)
 			goto cont;
 		if (idx > s_idx)
 			s_q_idx = 0;
 		q_idx = 0;
-		list_for_each_entry(q, &dev->qdisc_list, list) {
-			if (q_idx < s_q_idx) {
-				q_idx++;
-				continue;
-			}
-			if (tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).pid,
-					  cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWQDISC) <= 0)
-				goto done;
-			q_idx++;
-		}
+
+		dev_queue = netdev_get_tx_queue(dev, 0);
+		if (tc_dump_qdisc_root(dev_queue->qdisc, skb, cb, &q_idx, s_q_idx) < 0)
+			goto done;
+
+		dev_queue = &dev->rx_queue;
+		if (tc_dump_qdisc_root(dev_queue->qdisc, skb, cb, &q_idx, s_q_idx) < 0)
+			goto done;
+
 cont:
 		idx++;
 	}
@@ -1285,15 +1338,62 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk
 			      a->cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWTCLASS);
 }
 
+static int tc_dump_tclass_qdisc(struct Qdisc *q, struct sk_buff *skb,
+				struct tcmsg *tcm, struct netlink_callback *cb,
+				int *t_p, int s_t)
+{
+	struct qdisc_dump_args arg;
+
+	if (tc_qdisc_dump_ignore(q) ||
+	    *t_p < s_t || !q->ops->cl_ops ||
+	    (tcm->tcm_parent &&
+	     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
+		(*t_p)++;
+		return 0;
+	}
+	if (*t_p > s_t)
+		memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
+	arg.w.fn = qdisc_class_dump;
+	arg.skb = skb;
+	arg.cb = cb;
+	arg.w.stop  = 0;
+	arg.w.skip = cb->args[1];
+	arg.w.count = 0;
+	q->ops->cl_ops->walk(q, &arg.w);
+	cb->args[1] = arg.w.count;
+	if (arg.w.stop)
+		return -1;
+	(*t_p)++;
+	return 0;
+}
+
+static int tc_dump_tclass_root(struct Qdisc *root, struct sk_buff *skb,
+			       struct tcmsg *tcm, struct netlink_callback *cb,
+			       int *t_p, int s_t)
+{
+	struct Qdisc *q;
+
+	if (!root)
+		return 0;
+
+	if (tc_dump_tclass_qdisc(root, skb, tcm, cb, t_p, s_t) < 0)
+		return -1;
+
+	list_for_each_entry(q, &root->list, list) {
+		if (tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0)
+			return -1;
+	}
+
+	return 0;
+}
+
 static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
 	struct net *net = sock_net(skb->sk);
-	int t;
-	int s_t;
+	struct netdev_queue *dev_queue;
 	struct net_device *dev;
-	struct Qdisc *q;
-	struct tcmsg *tcm = (struct tcmsg*)NLMSG_DATA(cb->nlh);
-	struct qdisc_dump_args arg;
+	int t, s_t;
 
 	if (net != &init_net)
 		return 0;
@@ -1306,28 +1406,15 @@ static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb)
 	s_t = cb->args[0];
 	t = 0;
 
-	list_for_each_entry(q, &dev->qdisc_list, list) {
-		if (t < s_t || !q->ops->cl_ops ||
-		    (tcm->tcm_parent &&
-		     TC_H_MAJ(tcm->tcm_parent) != q->handle)) {
-			t++;
-			continue;
-		}
-		if (t > s_t)
-			memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0]));
-		arg.w.fn = qdisc_class_dump;
-		arg.skb = skb;
-		arg.cb = cb;
-		arg.w.stop  = 0;
-		arg.w.skip = cb->args[1];
-		arg.w.count = 0;
-		q->ops->cl_ops->walk(q, &arg.w);
-		cb->args[1] = arg.w.count;
-		if (arg.w.stop)
-			break;
-		t++;
-	}
+	dev_queue = netdev_get_tx_queue(dev, 0);
+	if (tc_dump_tclass_root(dev_queue->qdisc, skb, tcm, cb, &t, s_t) < 0)
+		goto done;
+
+	dev_queue = &dev->rx_queue;
+	if (tc_dump_tclass_root(dev_queue->qdisc, skb, tcm, cb, &t, s_t) < 0)
+		goto done;
 
+done:
 	cb->args[0] = t;
 
 	dev_put(dev);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index e244c462e6bd..14cc443d0490 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -480,15 +480,12 @@ static void __qdisc_destroy(struct rcu_head *head)
 
 void qdisc_destroy(struct Qdisc *qdisc)
 {
-	struct net_device *dev = qdisc_dev(qdisc);
-
 	if (qdisc->flags & TCQ_F_BUILTIN ||
 	    !atomic_dec_and_test(&qdisc->refcnt))
 		return;
 
-	spin_lock_bh(&dev->qdisc_list_lock);
-	list_del(&qdisc->list);
-	spin_unlock_bh(&dev->qdisc_list_lock);
+	if (qdisc->parent)
+		list_del(&qdisc->list);
 
 	call_rcu(&qdisc->q_rcu, __qdisc_destroy);
 }
@@ -520,9 +517,6 @@ static void attach_one_default_qdisc(struct net_device *dev,
 			printk(KERN_INFO "%s: activation failed\n", dev->name);
 			return;
 		}
-		spin_lock_bh(&dev->qdisc_list_lock);
-		list_add_tail(&qdisc->list, &dev->qdisc_list);
-		spin_unlock_bh(&dev->qdisc_list_lock);
 	} else {
 		qdisc =  &noqueue_qdisc;
 	}
-- 
cgit v1.2.3


From 4389dded7767d24290463f2a8302ba3253ebdd56 Mon Sep 17 00:00:00 2001
From: Adam Langley <agl@imperialviolet.org>
Date: Sat, 19 Jul 2008 00:07:02 -0700
Subject: tcp: Remove redundant checks when setting eff_sacks

Remove redundant checks when setting eff_sacks and make the number of SACKs a
compile time constant. Now that the options code knows how many SACK blocks can
fit in the header, we don't need to have the SACK code guessing at it.

Signed-off-by: Adam Langley <agl@imperialviolet.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/tcp.h  |  6 ++++++
 net/ipv4/tcp_input.c | 25 ++++++++++---------------
 2 files changed, 16 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 07e79bdb9cdf..2e2557388e36 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -224,6 +224,12 @@ struct tcp_options_received {
 	u16	mss_clamp;	/* Maximal mss, negotiated at connection setup */
 };
 
+/* This is the max number of SACKS that we'll generate and process. It's safe
+ * to increse this, although since:
+ *   size = TCPOLEN_SACK_BASE_ALIGNED (4) + n * TCPOLEN_SACK_PERBLOCK (8)
+ * only four options will fit in a standard TCP header */
+#define TCP_NUM_SACKS 4
+
 struct tcp_request_sock {
 	struct inet_request_sock 	req;
 #ifdef CONFIG_TCP_MD5SIG
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 88810bc01370..1f5e6049883e 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1423,10 +1423,10 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb,
 	unsigned char *ptr = (skb_transport_header(ack_skb) +
 			      TCP_SKB_CB(ack_skb)->sacked);
 	struct tcp_sack_block_wire *sp_wire = (struct tcp_sack_block_wire *)(ptr+2);
-	struct tcp_sack_block sp[4];
+	struct tcp_sack_block sp[TCP_NUM_SACKS];
 	struct tcp_sack_block *cache;
 	struct sk_buff *skb;
-	int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE) >> 3;
+	int num_sacks = min(TCP_NUM_SACKS, (ptr[1] - TCPOLEN_SACK_BASE) >> 3);
 	int used_sacks;
 	int reord = tp->packets_out;
 	int flag = 0;
@@ -3735,8 +3735,7 @@ static void tcp_dsack_set(struct sock *sk, u32 seq, u32 end_seq)
 		tp->rx_opt.dsack = 1;
 		tp->duplicate_sack[0].start_seq = seq;
 		tp->duplicate_sack[0].end_seq = end_seq;
-		tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + 1,
-					   4 - tp->rx_opt.tstamp_ok);
+		tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks + 1;
 	}
 }
 
@@ -3791,9 +3790,8 @@ static void tcp_sack_maybe_coalesce(struct tcp_sock *tp)
 			 * Decrease num_sacks.
 			 */
 			tp->rx_opt.num_sacks--;
-			tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks +
-						   tp->rx_opt.dsack,
-						   4 - tp->rx_opt.tstamp_ok);
+			tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks +
+					       tp->rx_opt.dsack;
 			for (i = this_sack; i < tp->rx_opt.num_sacks; i++)
 				sp[i] = sp[i + 1];
 			continue;
@@ -3843,7 +3841,7 @@ static void tcp_sack_new_ofo_skb(struct sock *sk, u32 seq, u32 end_seq)
 	 *
 	 * If the sack array is full, forget about the last one.
 	 */
-	if (this_sack >= 4) {
+	if (this_sack >= TCP_NUM_SACKS) {
 		this_sack--;
 		tp->rx_opt.num_sacks--;
 		sp--;
@@ -3856,8 +3854,7 @@ new_sack:
 	sp->start_seq = seq;
 	sp->end_seq = end_seq;
 	tp->rx_opt.num_sacks++;
-	tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack,
-				   4 - tp->rx_opt.tstamp_ok);
+	tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks + tp->rx_opt.dsack;
 }
 
 /* RCV.NXT advances, some SACKs should be eaten. */
@@ -3894,9 +3891,8 @@ static void tcp_sack_remove(struct tcp_sock *tp)
 	}
 	if (num_sacks != tp->rx_opt.num_sacks) {
 		tp->rx_opt.num_sacks = num_sacks;
-		tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks +
-					   tp->rx_opt.dsack,
-					   4 - tp->rx_opt.tstamp_ok);
+		tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks +
+				       tp->rx_opt.dsack;
 	}
 }
 
@@ -3975,8 +3971,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
 
 	if (tp->rx_opt.dsack) {
 		tp->rx_opt.dsack = 0;
-		tp->rx_opt.eff_sacks = min_t(unsigned int, tp->rx_opt.num_sacks,
-					     4 - tp->rx_opt.tstamp_ok);
+		tp->rx_opt.eff_sacks = tp->rx_opt.num_sacks;
 	}
 
 	/*  Queue data for delivery to the user.
-- 
cgit v1.2.3


From 230b183921ecbaa5fedc0d35ad6ba7bb64b6e06a Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sat, 19 Jul 2008 22:35:47 -0700
Subject: net: Use standard structures for generic socket address structures.

Use sockaddr_storage{} for generic socket address storage
and ensures proper alignment.
Use sockaddr{} for pointers to omit several casts.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/socket.h |  6 ++--
 include/net/compat.h   |  2 +-
 net/compat.c           |  2 +-
 net/core/iovec.c       |  2 +-
 net/socket.c           | 82 ++++++++++++++++++++++++++++----------------------
 5 files changed, 52 insertions(+), 42 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/socket.h b/include/linux/socket.h
index bd2b30a74e76..950af631e7fb 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -306,10 +306,10 @@ extern int csum_partial_copy_fromiovecend(unsigned char *kdata,
 					  int offset, 
 					  unsigned int len, __wsum *csump);
 
-extern int verify_iovec(struct msghdr *m, struct iovec *iov, char *address, int mode);
+extern int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, int mode);
 extern int memcpy_toiovec(struct iovec *v, unsigned char *kdata, int len);
-extern int move_addr_to_user(void *kaddr, int klen, void __user *uaddr, int __user *ulen);
-extern int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr);
+extern int move_addr_to_user(struct sockaddr *kaddr, int klen, void __user *uaddr, int __user *ulen);
+extern int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr *kaddr);
 extern int put_cmsg(struct msghdr*, int level, int type, int len, void *data);
 
 #endif
diff --git a/include/net/compat.h b/include/net/compat.h
index 164cb682e220..5bbf8bf9efea 100644
--- a/include/net/compat.h
+++ b/include/net/compat.h
@@ -32,7 +32,7 @@ extern int compat_sock_get_timestampns(struct sock *, struct timespec __user *);
 #endif /* defined(CONFIG_COMPAT) */
 
 extern int get_compat_msghdr(struct msghdr *, struct compat_msghdr __user *);
-extern int verify_compat_iovec(struct msghdr *, struct iovec *, char *, int);
+extern int verify_compat_iovec(struct msghdr *, struct iovec *, struct sockaddr *, int);
 extern asmlinkage long compat_sys_sendmsg(int,struct compat_msghdr __user *,unsigned);
 extern asmlinkage long compat_sys_recvmsg(int,struct compat_msghdr __user *,unsigned);
 extern asmlinkage long compat_sys_getsockopt(int, int, int, char __user *, int __user *);
diff --git a/net/compat.c b/net/compat.c
index c823f6f290cb..6e1b03b51933 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -75,7 +75,7 @@ int get_compat_msghdr(struct msghdr *kmsg, struct compat_msghdr __user *umsg)
 
 /* I've named the args so it is easy to tell whose space the pointers are in. */
 int verify_compat_iovec(struct msghdr *kern_msg, struct iovec *kern_iov,
-		   char *kern_address, int mode)
+		   struct sockaddr *kern_address, int mode)
 {
 	int tot_len;
 
diff --git a/net/core/iovec.c b/net/core/iovec.c
index 755c37fdaee7..4c9c0121c9da 100644
--- a/net/core/iovec.c
+++ b/net/core/iovec.c
@@ -36,7 +36,7 @@
  *	in any case.
  */
 
-int verify_iovec(struct msghdr *m, struct iovec *iov, char *address, int mode)
+int verify_iovec(struct msghdr *m, struct iovec *iov, struct sockaddr *address, int mode)
 {
 	int size, err, ct;
 
diff --git a/net/socket.c b/net/socket.c
index 81fe82513046..1ba57d888981 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -180,9 +180,9 @@ static DEFINE_PER_CPU(int, sockets_in_use) = 0;
  *	invalid addresses -EFAULT is returned. On a success 0 is returned.
  */
 
-int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr)
+int move_addr_to_kernel(void __user *uaddr, int ulen, struct sockaddr *kaddr)
 {
-	if (ulen < 0 || ulen > MAX_SOCK_ADDR)
+	if (ulen < 0 || ulen > sizeof(struct sockaddr_storage))
 		return -EINVAL;
 	if (ulen == 0)
 		return 0;
@@ -208,7 +208,7 @@ int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr)
  *	specified. Zero is returned for a success.
  */
 
-int move_addr_to_user(void *kaddr, int klen, void __user *uaddr,
+int move_addr_to_user(struct sockaddr *kaddr, int klen, void __user *uaddr,
 		      int __user *ulen)
 {
 	int err;
@@ -219,7 +219,7 @@ int move_addr_to_user(void *kaddr, int klen, void __user *uaddr,
 		return err;
 	if (len > klen)
 		len = klen;
-	if (len < 0 || len > MAX_SOCK_ADDR)
+	if (len < 0 || len > sizeof(struct sockaddr_storage))
 		return -EINVAL;
 	if (len) {
 		if (audit_sockaddr(klen, kaddr))
@@ -1342,20 +1342,20 @@ out_fd:
 asmlinkage long sys_bind(int fd, struct sockaddr __user *umyaddr, int addrlen)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (sock) {
-		err = move_addr_to_kernel(umyaddr, addrlen, address);
+		err = move_addr_to_kernel(umyaddr, addrlen, (struct sockaddr *)&address);
 		if (err >= 0) {
 			err = security_socket_bind(sock,
-						   (struct sockaddr *)address,
+						   (struct sockaddr *)&address,
 						   addrlen);
 			if (!err)
 				err = sock->ops->bind(sock,
 						      (struct sockaddr *)
-						      address, addrlen);
+						      &address, addrlen);
 		}
 		fput_light(sock->file, fput_needed);
 	}
@@ -1407,7 +1407,7 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 	struct socket *sock, *newsock;
 	struct file *newfile;
 	int err, len, newfd, fput_needed;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (!sock)
@@ -1446,13 +1446,13 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 		goto out_fd;
 
 	if (upeer_sockaddr) {
-		if (newsock->ops->getname(newsock, (struct sockaddr *)address,
+		if (newsock->ops->getname(newsock, (struct sockaddr *)&address,
 					  &len, 2) < 0) {
 			err = -ECONNABORTED;
 			goto out_fd;
 		}
-		err = move_addr_to_user(address, len, upeer_sockaddr,
-					upeer_addrlen);
+		err = move_addr_to_user((struct sockaddr *)&address,
+					len, upeer_sockaddr, upeer_addrlen);
 		if (err < 0)
 			goto out_fd;
 	}
@@ -1495,22 +1495,22 @@ asmlinkage long sys_connect(int fd, struct sockaddr __user *uservaddr,
 			    int addrlen)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (!sock)
 		goto out;
-	err = move_addr_to_kernel(uservaddr, addrlen, address);
+	err = move_addr_to_kernel(uservaddr, addrlen, (struct sockaddr *)&address);
 	if (err < 0)
 		goto out_put;
 
 	err =
-	    security_socket_connect(sock, (struct sockaddr *)address, addrlen);
+	    security_socket_connect(sock, (struct sockaddr *)&address, addrlen);
 	if (err)
 		goto out_put;
 
-	err = sock->ops->connect(sock, (struct sockaddr *)address, addrlen,
+	err = sock->ops->connect(sock, (struct sockaddr *)&address, addrlen,
 				 sock->file->f_flags);
 out_put:
 	fput_light(sock->file, fput_needed);
@@ -1527,7 +1527,7 @@ asmlinkage long sys_getsockname(int fd, struct sockaddr __user *usockaddr,
 				int __user *usockaddr_len)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int len, err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
@@ -1538,10 +1538,10 @@ asmlinkage long sys_getsockname(int fd, struct sockaddr __user *usockaddr,
 	if (err)
 		goto out_put;
 
-	err = sock->ops->getname(sock, (struct sockaddr *)address, &len, 0);
+	err = sock->ops->getname(sock, (struct sockaddr *)&address, &len, 0);
 	if (err)
 		goto out_put;
-	err = move_addr_to_user(address, len, usockaddr, usockaddr_len);
+	err = move_addr_to_user((struct sockaddr *)&address, len, usockaddr, usockaddr_len);
 
 out_put:
 	fput_light(sock->file, fput_needed);
@@ -1558,7 +1558,7 @@ asmlinkage long sys_getpeername(int fd, struct sockaddr __user *usockaddr,
 				int __user *usockaddr_len)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int len, err, fput_needed;
 
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
@@ -1570,10 +1570,10 @@ asmlinkage long sys_getpeername(int fd, struct sockaddr __user *usockaddr,
 		}
 
 		err =
-		    sock->ops->getname(sock, (struct sockaddr *)address, &len,
+		    sock->ops->getname(sock, (struct sockaddr *)&address, &len,
 				       1);
 		if (!err)
-			err = move_addr_to_user(address, len, usockaddr,
+			err = move_addr_to_user((struct sockaddr *)&address, len, usockaddr,
 						usockaddr_len);
 		fput_light(sock->file, fput_needed);
 	}
@@ -1591,7 +1591,7 @@ asmlinkage long sys_sendto(int fd, void __user *buff, size_t len,
 			   int addr_len)
 {
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err;
 	struct msghdr msg;
 	struct iovec iov;
@@ -1610,10 +1610,10 @@ asmlinkage long sys_sendto(int fd, void __user *buff, size_t len,
 	msg.msg_controllen = 0;
 	msg.msg_namelen = 0;
 	if (addr) {
-		err = move_addr_to_kernel(addr, addr_len, address);
+		err = move_addr_to_kernel(addr, addr_len, (struct sockaddr *)&address);
 		if (err < 0)
 			goto out_put;
-		msg.msg_name = address;
+		msg.msg_name = (struct sockaddr *)&address;
 		msg.msg_namelen = addr_len;
 	}
 	if (sock->file->f_flags & O_NONBLOCK)
@@ -1649,7 +1649,7 @@ asmlinkage long sys_recvfrom(int fd, void __user *ubuf, size_t size,
 	struct socket *sock;
 	struct iovec iov;
 	struct msghdr msg;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	int err, err2;
 	int fput_needed;
 
@@ -1663,14 +1663,15 @@ asmlinkage long sys_recvfrom(int fd, void __user *ubuf, size_t size,
 	msg.msg_iov = &iov;
 	iov.iov_len = size;
 	iov.iov_base = ubuf;
-	msg.msg_name = address;
-	msg.msg_namelen = MAX_SOCK_ADDR;
+	msg.msg_name = (struct sockaddr *)&address;
+	msg.msg_namelen = sizeof(address);
 	if (sock->file->f_flags & O_NONBLOCK)
 		flags |= MSG_DONTWAIT;
 	err = sock_recvmsg(sock, &msg, size, flags);
 
 	if (err >= 0 && addr != NULL) {
-		err2 = move_addr_to_user(address, msg.msg_namelen, addr, addr_len);
+		err2 = move_addr_to_user((struct sockaddr *)&address,
+					 msg.msg_namelen, addr, addr_len);
 		if (err2 < 0)
 			err = err2;
 	}
@@ -1790,7 +1791,7 @@ asmlinkage long sys_sendmsg(int fd, struct msghdr __user *msg, unsigned flags)
 	struct compat_msghdr __user *msg_compat =
 	    (struct compat_msghdr __user *)msg;
 	struct socket *sock;
-	char address[MAX_SOCK_ADDR];
+	struct sockaddr_storage address;
 	struct iovec iovstack[UIO_FASTIOV], *iov = iovstack;
 	unsigned char ctl[sizeof(struct cmsghdr) + 20]
 	    __attribute__ ((aligned(sizeof(__kernel_size_t))));
@@ -1828,9 +1829,13 @@ asmlinkage long sys_sendmsg(int fd, struct msghdr __user *msg, unsigned flags)
 
 	/* This will also move the address data into kernel space */
 	if (MSG_CMSG_COMPAT & flags) {
-		err = verify_compat_iovec(&msg_sys, iov, address, VERIFY_READ);
+		err = verify_compat_iovec(&msg_sys, iov,
+					  (struct sockaddr *)&address,
+					  VERIFY_READ);
 	} else
-		err = verify_iovec(&msg_sys, iov, address, VERIFY_READ);
+		err = verify_iovec(&msg_sys, iov,
+				   (struct sockaddr *)&address,
+				   VERIFY_READ);
 	if (err < 0)
 		goto out_freeiov;
 	total_len = err;
@@ -1901,7 +1906,7 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg,
 	int fput_needed;
 
 	/* kernel mode address */
-	char addr[MAX_SOCK_ADDR];
+	struct sockaddr_storage addr;
 
 	/* user mode address pointers */
 	struct sockaddr __user *uaddr;
@@ -1939,9 +1944,13 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg,
 	uaddr = (__force void __user *)msg_sys.msg_name;
 	uaddr_len = COMPAT_NAMELEN(msg);
 	if (MSG_CMSG_COMPAT & flags) {
-		err = verify_compat_iovec(&msg_sys, iov, addr, VERIFY_WRITE);
+		err = verify_compat_iovec(&msg_sys, iov,
+					  (struct sockaddr *)&addr,
+					  VERIFY_WRITE);
 	} else
-		err = verify_iovec(&msg_sys, iov, addr, VERIFY_WRITE);
+		err = verify_iovec(&msg_sys, iov,
+				   (struct sockaddr *)&addr,
+				   VERIFY_WRITE);
 	if (err < 0)
 		goto out_freeiov;
 	total_len = err;
@@ -1957,7 +1966,8 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg,
 	len = err;
 
 	if (uaddr != NULL) {
-		err = move_addr_to_user(addr, msg_sys.msg_namelen, uaddr,
+		err = move_addr_to_user((struct sockaddr *)&addr,
+					msg_sys.msg_namelen, uaddr,
 					uaddr_len);
 		if (err < 0)
 			goto out_freeiov;
-- 
cgit v1.2.3


From 175f9c1bba9b825d22b142d183c9e175488b260c Mon Sep 17 00:00:00 2001
From: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Date: Sun, 20 Jul 2008 00:08:47 -0700
Subject: net_sched: Add size table for qdiscs

Add size table functions for qdiscs and calculate packet size in
qdisc_enqueue().

Based on patch by Patrick McHardy
 http://marc.info/?l=linux-netdev&m=115201979221729&w=2

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/pkt_sched.h |  20 ++++++
 include/linux/rtnetlink.h |   1 +
 include/net/pkt_sched.h   |   1 +
 include/net/sch_generic.h |  25 +++++++-
 net/sched/sch_api.c       | 151 +++++++++++++++++++++++++++++++++++++++++++++-
 net/sched/sch_generic.c   |   1 +
 net/sched/sch_netem.c     |   5 +-
 7 files changed, 199 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index 87f4e0fa8f27..e5de421ac7b4 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -85,6 +85,26 @@ struct tc_ratespec
 
 #define TC_RTAB_SIZE	1024
 
+struct tc_sizespec {
+	unsigned char	cell_log;
+	unsigned char	size_log;
+	short		cell_align;
+	int		overhead;
+	unsigned int	linklayer;
+	unsigned int	mpu;
+	unsigned int	mtu;
+	unsigned int	tsize;
+};
+
+enum {
+	TCA_STAB_UNSPEC,
+	TCA_STAB_BASE,
+	TCA_STAB_DATA,
+	__TCA_STAB_MAX
+};
+
+#define TCA_STAB_MAX (__TCA_STAB_MAX - 1)
+
 /* FIFO section */
 
 struct tc_fifo_qopt
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index b358c704d102..f4d386c191f5 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -482,6 +482,7 @@ enum
 	TCA_RATE,
 	TCA_FCNT,
 	TCA_STATS2,
+	TCA_STAB,
 	__TCA_MAX
 };
 
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index e4e30052e4e2..6affcfaa123e 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -83,6 +83,7 @@ extern struct Qdisc *qdisc_lookup_class(struct net_device *dev, u32 handle);
 extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r,
 		struct nlattr *tab);
 extern void qdisc_put_rtab(struct qdisc_rate_table *tab);
+extern void qdisc_put_stab(struct qdisc_size_table *tab);
 
 extern void __qdisc_run(struct Qdisc *q);
 
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 8229520e088a..db9ad655eb8a 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -29,6 +29,13 @@ enum qdisc_state_t
 	__QDISC_STATE_SCHED,
 };
 
+struct qdisc_size_table {
+	struct list_head	list;
+	struct tc_sizespec	szopts;
+	int			refcnt;
+	u16			data[];
+};
+
 struct Qdisc
 {
 	int 			(*enqueue)(struct sk_buff *skb, struct Qdisc *dev);
@@ -39,6 +46,7 @@ struct Qdisc
 #define TCQ_F_INGRESS	4
 	int			padded;
 	struct Qdisc_ops	*ops;
+	struct qdisc_size_table	*stab;
 	u32			handle;
 	u32			parent;
 	atomic_t		refcnt;
@@ -165,6 +173,16 @@ struct tcf_proto
 	struct tcf_proto_ops	*ops;
 };
 
+struct qdisc_skb_cb {
+	unsigned int		pkt_len;
+	char			data[];
+};
+
+static inline struct qdisc_skb_cb *qdisc_skb_cb(struct sk_buff *skb)
+{
+	return (struct qdisc_skb_cb *)skb->cb;
+}
+
 static inline spinlock_t *qdisc_lock(struct Qdisc *qdisc)
 {
 	return &qdisc->q.lock;
@@ -257,6 +275,8 @@ extern struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
 				       struct netdev_queue *dev_queue,
 				       struct Qdisc_ops *ops, u32 parentid);
+extern void qdisc_calculate_pkt_len(struct sk_buff *skb,
+				   struct qdisc_size_table *stab);
 extern void tcf_destroy(struct tcf_proto *tp);
 extern void tcf_destroy_chain(struct tcf_proto **fl);
 
@@ -308,16 +328,19 @@ static inline bool qdisc_tx_is_noop(const struct net_device *dev)
 
 static inline unsigned int qdisc_pkt_len(struct sk_buff *skb)
 {
-	return skb->len;
+	return qdisc_skb_cb(skb)->pkt_len;
 }
 
 static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
+	if (sch->stab)
+		qdisc_calculate_pkt_len(skb, sch->stab);
 	return sch->enqueue(skb, sch);
 }
 
 static inline int qdisc_enqueue_root(struct sk_buff *skb, struct Qdisc *sch)
 {
+	qdisc_skb_cb(skb)->pkt_len = skb->len;
 	return qdisc_enqueue(skb, sch);
 }
 
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index fb43731c9860..5219d5f9d754 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -286,6 +286,129 @@ void qdisc_put_rtab(struct qdisc_rate_table *tab)
 }
 EXPORT_SYMBOL(qdisc_put_rtab);
 
+static LIST_HEAD(qdisc_stab_list);
+static DEFINE_SPINLOCK(qdisc_stab_lock);
+
+static const struct nla_policy stab_policy[TCA_STAB_MAX + 1] = {
+	[TCA_STAB_BASE]	= { .len = sizeof(struct tc_sizespec) },
+	[TCA_STAB_DATA] = { .type = NLA_BINARY },
+};
+
+static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt)
+{
+	struct nlattr *tb[TCA_STAB_MAX + 1];
+	struct qdisc_size_table *stab;
+	struct tc_sizespec *s;
+	unsigned int tsize = 0;
+	u16 *tab = NULL;
+	int err;
+
+	err = nla_parse_nested(tb, TCA_STAB_MAX, opt, stab_policy);
+	if (err < 0)
+		return ERR_PTR(err);
+	if (!tb[TCA_STAB_BASE])
+		return ERR_PTR(-EINVAL);
+
+	s = nla_data(tb[TCA_STAB_BASE]);
+
+	if (s->tsize > 0) {
+		if (!tb[TCA_STAB_DATA])
+			return ERR_PTR(-EINVAL);
+		tab = nla_data(tb[TCA_STAB_DATA]);
+		tsize = nla_len(tb[TCA_STAB_DATA]) / sizeof(u16);
+	}
+
+	if (!s || tsize != s->tsize || (!tab && tsize > 0))
+		return ERR_PTR(-EINVAL);
+
+	spin_lock(&qdisc_stab_lock);
+
+	list_for_each_entry(stab, &qdisc_stab_list, list) {
+		if (memcmp(&stab->szopts, s, sizeof(*s)))
+			continue;
+		if (tsize > 0 && memcmp(stab->data, tab, tsize * sizeof(u16)))
+			continue;
+		stab->refcnt++;
+		spin_unlock(&qdisc_stab_lock);
+		return stab;
+	}
+
+	spin_unlock(&qdisc_stab_lock);
+
+	stab = kmalloc(sizeof(*stab) + tsize * sizeof(u16), GFP_KERNEL);
+	if (!stab)
+		return ERR_PTR(-ENOMEM);
+
+	stab->refcnt = 1;
+	stab->szopts = *s;
+	if (tsize > 0)
+		memcpy(stab->data, tab, tsize * sizeof(u16));
+
+	spin_lock(&qdisc_stab_lock);
+	list_add_tail(&stab->list, &qdisc_stab_list);
+	spin_unlock(&qdisc_stab_lock);
+
+	return stab;
+}
+
+void qdisc_put_stab(struct qdisc_size_table *tab)
+{
+	if (!tab)
+		return;
+
+	spin_lock(&qdisc_stab_lock);
+
+	if (--tab->refcnt == 0) {
+		list_del(&tab->list);
+		kfree(tab);
+	}
+
+	spin_unlock(&qdisc_stab_lock);
+}
+EXPORT_SYMBOL(qdisc_put_stab);
+
+static int qdisc_dump_stab(struct sk_buff *skb, struct qdisc_size_table *stab)
+{
+	struct nlattr *nest;
+
+	nest = nla_nest_start(skb, TCA_STAB);
+	NLA_PUT(skb, TCA_STAB_BASE, sizeof(stab->szopts), &stab->szopts);
+	nla_nest_end(skb, nest);
+
+	return skb->len;
+
+nla_put_failure:
+	return -1;
+}
+
+void qdisc_calculate_pkt_len(struct sk_buff *skb, struct qdisc_size_table *stab)
+{
+	int pkt_len, slot;
+
+	pkt_len = skb->len + stab->szopts.overhead;
+	if (unlikely(!stab->szopts.tsize))
+		goto out;
+
+	slot = pkt_len + stab->szopts.cell_align;
+	if (unlikely(slot < 0))
+		slot = 0;
+
+	slot >>= stab->szopts.cell_log;
+	if (likely(slot < stab->szopts.tsize))
+		pkt_len = stab->data[slot];
+	else
+		pkt_len = stab->data[stab->szopts.tsize - 1] *
+				(slot / stab->szopts.tsize) +
+				stab->data[slot % stab->szopts.tsize];
+
+	pkt_len <<= stab->szopts.size_log;
+out:
+	if (unlikely(pkt_len < 1))
+		pkt_len = 1;
+	qdisc_skb_cb(skb)->pkt_len = pkt_len;
+}
+EXPORT_SYMBOL(qdisc_calculate_pkt_len);
+
 static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
 {
 	struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog,
@@ -613,6 +736,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 	struct nlattr *kind = tca[TCA_KIND];
 	struct Qdisc *sch;
 	struct Qdisc_ops *ops;
+	struct qdisc_size_table *stab;
 
 	ops = qdisc_lookup_ops(kind);
 #ifdef CONFIG_KMOD
@@ -670,6 +794,14 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 	sch->handle = handle;
 
 	if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS])) == 0) {
+		if (tca[TCA_STAB]) {
+			stab = qdisc_get_stab(tca[TCA_STAB]);
+			if (IS_ERR(stab)) {
+				err = PTR_ERR(stab);
+				goto err_out3;
+			}
+			sch->stab = stab;
+		}
 		if (tca[TCA_RATE]) {
 			err = gen_new_estimator(&sch->bstats, &sch->rate_est,
 						qdisc_root_lock(sch),
@@ -691,6 +823,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
 		return sch;
 	}
 err_out3:
+	qdisc_put_stab(sch->stab);
 	dev_put(dev);
 	kfree((char *) sch - sch->padded);
 err_out2:
@@ -702,15 +835,26 @@ err_out:
 
 static int qdisc_change(struct Qdisc *sch, struct nlattr **tca)
 {
-	if (tca[TCA_OPTIONS]) {
-		int err;
+	struct qdisc_size_table *stab = NULL;
+	int err = 0;
 
+	if (tca[TCA_OPTIONS]) {
 		if (sch->ops->change == NULL)
 			return -EINVAL;
 		err = sch->ops->change(sch, tca[TCA_OPTIONS]);
 		if (err)
 			return err;
 	}
+
+	if (tca[TCA_STAB]) {
+		stab = qdisc_get_stab(tca[TCA_STAB]);
+		if (IS_ERR(stab))
+			return PTR_ERR(stab);
+	}
+
+	qdisc_put_stab(sch->stab);
+	sch->stab = stab;
+
 	if (tca[TCA_RATE])
 		gen_replace_estimator(&sch->bstats, &sch->rate_est,
 				      qdisc_root_lock(sch), tca[TCA_RATE]);
@@ -994,6 +1138,9 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid,
 		goto nla_put_failure;
 	q->qstats.qlen = q->q.qlen;
 
+	if (q->stab && qdisc_dump_stab(skb, q->stab) < 0)
+		goto nla_put_failure;
+
 	if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS,
 					 TCA_XSTATS, qdisc_root_lock(q), &d) < 0)
 		goto nla_put_failure;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 522a41a9f904..27a51f04db49 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -469,6 +469,7 @@ static void __qdisc_destroy(struct rcu_head *head)
 	struct Qdisc *qdisc = container_of(head, struct Qdisc, q_rcu);
 	const struct Qdisc_ops  *ops = qdisc->ops;
 
+	qdisc_put_stab(qdisc->stab);
 	gen_kill_estimator(&qdisc->bstats, &qdisc->rate_est);
 	if (ops->reset)
 		ops->reset(qdisc);
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index ae49be00022f..a59085700678 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -84,8 +84,9 @@ struct netem_skb_cb {
 
 static inline struct netem_skb_cb *netem_skb_cb(struct sk_buff *skb)
 {
-	BUILD_BUG_ON(sizeof(skb->cb) < sizeof(struct netem_skb_cb));
-	return (struct netem_skb_cb *)skb->cb;
+	BUILD_BUG_ON(sizeof(skb->cb) <
+		sizeof(struct qdisc_skb_cb) + sizeof(struct netem_skb_cb));
+	return (struct netem_skb_cb *)qdisc_skb_cb(skb)->data;
 }
 
 /* init_crandom - initialize correlated random number generator
-- 
cgit v1.2.3


From 80422d3431cc990b967da129f9eb8e3e9989f841 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Fri, 18 Jul 2008 18:11:33 -0700
Subject: cpumask: Provide a generic set of CPUMASK_ALLOC macros, FIXUP

  * Rename CPUMASK_VAR --> CPUMASK_PTR (and simplify)

  * Fix a semantic error in CPUMASK_ALLOC

  * Add a bit of commentry to cpumask.h

Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/cpumask.h | 33 ++++++++++++++++++++++++++-------
 1 file changed, 26 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 72f9c32c12b0..30d59d1d0626 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -75,16 +75,36 @@
  * CPU_MASK_NONE			Initializer - no bits set
  * unsigned long *cpus_addr(mask)	Array of unsigned long's in mask
  *
+ * CPUMASK_ALLOC kmalloc's a structure that is a composite of many cpumask_t
+ * variables, and CPUMASK_PTR provides pointers to each field.
+ *
+ * The structure should be defined something like this:
+ * struct my_cpumasks {
+ *	cpumask_t mask1;
+ *	cpumask_t mask2;
+ * };
+ *
+ * Usage is then:
+ *	CPUMASK_ALLOC(my_cpumasks);
+ *	CPUMASK_PTR(mask1, my_cpumasks);
+ *	CPUMASK_PTR(mask2, my_cpumasks);
+ *
+ *	--- DO NOT reference cpumask_t pointers until this check ---
+ *	if (my_cpumasks == NULL)
+ *		"kmalloc failed"...
+ *
+ * References are now pointers to the cpumask_t variables (*mask1, ...)
+ *
  *if NR_CPUS > BITS_PER_LONG
  *   CPUMASK_ALLOC(m)			Declares and allocates struct m *m =
- *					   (struct m *)kmalloc(sizeof(*m), ...)
- *   CPUMASK_FREE(m)			Macro for kfree(v)
+ *						kmalloc(sizeof(*m), GFP_KERNEL)
+ *   CPUMASK_FREE(m)			Macro for kfree(m)
  *else
  *   CPUMASK_ALLOC(m)			Declares struct m _m, *m = &_m
  *   CPUMASK_FREE(m)			Nop
  *endif
- *   CPUMASK_VAR(v, m)			Declares cpumask_t *v =
- *						m + offset(struct m, v)
+ *   CPUMASK_PTR(v, m)			Declares cpumask_t *v = &(m->v)
+ * ------------------------------------------------------------------------
  *
  * int cpumask_scnprintf(buf, len, mask) Format cpumask for printing
  * int cpumask_parse_user(ubuf, ulen, mask)	Parse ascii string as cpumask
@@ -326,11 +346,10 @@ extern cpumask_t cpu_mask_all;
 #define	CPUMASK_ALLOC(m)	struct m *m = kmalloc(sizeof(*m), GFP_KERNEL)
 #define	CPUMASK_FREE(m)		kfree(m)
 #else
-#define	CPUMASK_ALLOC(m)	struct allmasks _m, *m = &_m
+#define	CPUMASK_ALLOC(m)	struct m _m, *m = &_m
 #define	CPUMASK_FREE(m)
 #endif
-#define	CPUMASK_VAR(v, m) 	cpumask_t *v = (cpumask_t *)		\
-				((unsigned long)(m) + offsetof(struct m, v))
+#define	CPUMASK_PTR(v, m) 	cpumask_t *v = &(m->v)
 
 #define cpumask_scnprintf(buf, len, src) \
 			__cpumask_scnprintf((buf), (len), &(src), NR_CPUS)
-- 
cgit v1.2.3


From 7cc8883074b040aa8c1ebd3a17463b0ea3a9ef16 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@qumranet.com>
Date: Tue, 13 May 2008 16:29:20 +0300
Subject: KVM: Remove decache_vcpus_on_cpu() and related callbacks

Obsoleted by the vmx-specific per-cpu list.

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/ia64/kvm/kvm-ia64.c   | 8 --------
 arch/powerpc/kvm/powerpc.c | 4 ----
 arch/s390/kvm/kvm-s390.c   | 4 ----
 arch/x86/kvm/svm.c         | 5 -----
 arch/x86/kvm/vmx.c         | 6 ------
 arch/x86/kvm/x86.c         | 8 --------
 include/asm-x86/kvm_host.h | 1 -
 include/linux/kvm_host.h   | 3 ---
 virt/kvm/kvm_main.c        | 1 -
 9 files changed, 40 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 68c978be9a51..7c504be57972 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1035,14 +1035,6 @@ static void kvm_free_vmm_area(void)
 	}
 }
 
-/*
- * Make sure that a cpu that is being hot-unplugged does not have any vcpus
- * cached on it. Leave it as blank for IA64.
- */
-void decache_vcpus_on_cpu(int cpu)
-{
-}
-
 static void vti_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 }
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 777e0f34e0ea..0513b359851b 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -240,10 +240,6 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
 }
 
-void decache_vcpus_on_cpu(int cpu)
-{
-}
-
 int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
                                     struct kvm_debug_guest *dbg)
 {
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 6558b09ff579..4585c8ac2b0c 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -79,10 +79,6 @@ void kvm_arch_hardware_disable(void *garbage)
 {
 }
 
-void decache_vcpus_on_cpu(int cpu)
-{
-}
-
 int kvm_arch_hardware_setup(void)
 {
 	return 0;
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 9390a31c06f4..238e8f3afaf4 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -709,10 +709,6 @@ static void svm_vcpu_put(struct kvm_vcpu *vcpu)
 	rdtscll(vcpu->arch.host_tsc);
 }
 
-static void svm_vcpu_decache(struct kvm_vcpu *vcpu)
-{
-}
-
 static void svm_cache_regs(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
@@ -1933,7 +1929,6 @@ static struct kvm_x86_ops svm_x86_ops = {
 	.prepare_guest_switch = svm_prepare_guest_switch,
 	.vcpu_load = svm_vcpu_load,
 	.vcpu_put = svm_vcpu_put,
-	.vcpu_decache = svm_vcpu_decache,
 
 	.set_guest_debug = svm_guest_debug,
 	.get_msr = svm_get_msr,
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 4d179d106376..b99bb37e5dec 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -692,11 +692,6 @@ static void vmx_fpu_deactivate(struct kvm_vcpu *vcpu)
 	update_exception_bitmap(vcpu);
 }
 
-static void vmx_vcpu_decache(struct kvm_vcpu *vcpu)
-{
-	vcpu_clear(to_vmx(vcpu));
-}
-
 static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu)
 {
 	return vmcs_readl(GUEST_RFLAGS);
@@ -3114,7 +3109,6 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.prepare_guest_switch = vmx_save_host_state,
 	.vcpu_load = vmx_vcpu_load,
 	.vcpu_put = vmx_vcpu_put,
-	.vcpu_decache = vmx_vcpu_decache,
 
 	.set_guest_debug = set_guest_debug,
 	.guest_debug_pre = kvm_guest_debug_pre,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 8c14ddcaba70..fd03b4465bcc 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -817,14 +817,6 @@ out:
 	return r;
 }
 
-/*
- * Make sure that a cpu that is being hot-unplugged does not have any vcpus
- * cached on it.
- */
-void decache_vcpus_on_cpu(int cpu)
-{
-}
-
 int kvm_dev_ioctl_check_extension(long ext)
 {
 	int r;
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 0df9d5fa281a..4bcdc7de07b5 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -380,7 +380,6 @@ struct kvm_x86_ops {
 	void (*prepare_guest_switch)(struct kvm_vcpu *vcpu);
 	void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu);
 	void (*vcpu_put)(struct kvm_vcpu *vcpu);
-	void (*vcpu_decache)(struct kvm_vcpu *vcpu);
 
 	int (*set_guest_debug)(struct kvm_vcpu *vcpu,
 			       struct kvm_debug_guest *dbg);
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index de9d1df4bba2..865dcbcb891f 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -135,9 +135,6 @@ void kvm_vcpu_uninit(struct kvm_vcpu *vcpu);
 void vcpu_load(struct kvm_vcpu *vcpu);
 void vcpu_put(struct kvm_vcpu *vcpu);
 
-void decache_vcpus_on_cpu(int cpu);
-
-
 int kvm_init(void *opaque, unsigned int vcpu_size,
 		  struct module *module);
 void kvm_exit(void);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index e4bf88a9ee4e..83a0e5ce6037 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1273,7 +1273,6 @@ static void hardware_disable(void *junk)
 	if (!cpu_isset(cpu, cpus_hardware_enabled))
 		return;
 	cpu_clear(cpu, cpus_hardware_enabled);
-	decache_vcpus_on_cpu(cpu);
 	kvm_arch_hardware_disable(NULL);
 }
 
-- 
cgit v1.2.3


From 92760499d01ef91518119908eb9b8798b6c9bd3f Mon Sep 17 00:00:00 2001
From: Laurent Vivier <Laurent.Vivier@bull.net>
Date: Fri, 30 May 2008 16:05:53 +0200
Subject: KVM: kvm_io_device: extend in_range() to manage len and write
 attribute

Modify member in_range() of structure kvm_io_device to pass length and the type
of the I/O (write or read).

This modification allows to use kvm_io_device with coalesced MMIO.

Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/ia64/kvm/kvm-ia64.c |  6 +++---
 arch/x86/kvm/i8254.c     |  6 ++++--
 arch/x86/kvm/i8259.c     |  3 ++-
 arch/x86/kvm/lapic.c     |  3 ++-
 arch/x86/kvm/x86.c       | 28 +++++++++++++++++-----------
 include/linux/kvm_host.h |  3 ++-
 virt/kvm/ioapic.c        |  3 ++-
 virt/kvm/iodev.h         |  8 +++++---
 virt/kvm/kvm_main.c      |  5 +++--
 9 files changed, 40 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 7c504be57972..bb58df7cc418 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -195,11 +195,11 @@ int kvm_dev_ioctl_check_extension(long ext)
 }
 
 static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu,
-					gpa_t addr)
+					gpa_t addr, int len, int is_write)
 {
 	struct kvm_io_device *dev;
 
-	dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr);
+	dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr, len, is_write);
 
 	return dev;
 }
@@ -231,7 +231,7 @@ static int handle_mmio(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	kvm_run->exit_reason = KVM_EXIT_MMIO;
 	return 0;
 mmio:
-	mmio_dev = vcpu_find_mmio_dev(vcpu, p->addr);
+	mmio_dev = vcpu_find_mmio_dev(vcpu, p->addr, p->size, !p->dir);
 	if (mmio_dev) {
 		if (!p->dir)
 			kvm_iodevice_write(mmio_dev, p->addr, p->size,
diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c
index 60074dc66bd7..9e3391e9a1b7 100644
--- a/arch/x86/kvm/i8254.c
+++ b/arch/x86/kvm/i8254.c
@@ -460,7 +460,8 @@ static void pit_ioport_read(struct kvm_io_device *this,
 	mutex_unlock(&pit_state->lock);
 }
 
-static int pit_in_range(struct kvm_io_device *this, gpa_t addr)
+static int pit_in_range(struct kvm_io_device *this, gpa_t addr,
+			int len, int is_write)
 {
 	return ((addr >= KVM_PIT_BASE_ADDRESS) &&
 		(addr < KVM_PIT_BASE_ADDRESS + KVM_PIT_MEM_LENGTH));
@@ -501,7 +502,8 @@ static void speaker_ioport_read(struct kvm_io_device *this,
 	mutex_unlock(&pit_state->lock);
 }
 
-static int speaker_in_range(struct kvm_io_device *this, gpa_t addr)
+static int speaker_in_range(struct kvm_io_device *this, gpa_t addr,
+			    int len, int is_write)
 {
 	return (addr == KVM_SPEAKER_BASE_ADDRESS);
 }
diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c
index ab29cf2def47..5857f59ad4aa 100644
--- a/arch/x86/kvm/i8259.c
+++ b/arch/x86/kvm/i8259.c
@@ -346,7 +346,8 @@ static u32 elcr_ioport_read(void *opaque, u32 addr1)
 	return s->elcr;
 }
 
-static int picdev_in_range(struct kvm_io_device *this, gpa_t addr)
+static int picdev_in_range(struct kvm_io_device *this, gpa_t addr,
+			   int len, int is_write)
 {
 	switch (addr) {
 	case 0x20:
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index e48d19394031..180ba7316da5 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -785,7 +785,8 @@ static void apic_mmio_write(struct kvm_io_device *this,
 
 }
 
-static int apic_mmio_range(struct kvm_io_device *this, gpa_t addr)
+static int apic_mmio_range(struct kvm_io_device *this, gpa_t addr,
+			   int len, int size)
 {
 	struct kvm_lapic *apic = (struct kvm_lapic *)this->private;
 	int ret = 0;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4c94fad7f01e..ab3f5552d694 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1797,13 +1797,14 @@ static void kvm_init_msr_list(void)
  * Only apic need an MMIO device hook, so shortcut now..
  */
 static struct kvm_io_device *vcpu_find_pervcpu_dev(struct kvm_vcpu *vcpu,
-						gpa_t addr)
+						gpa_t addr, int len,
+						int is_write)
 {
 	struct kvm_io_device *dev;
 
 	if (vcpu->arch.apic) {
 		dev = &vcpu->arch.apic->dev;
-		if (dev->in_range(dev, addr))
+		if (dev->in_range(dev, addr, len, is_write))
 			return dev;
 	}
 	return NULL;
@@ -1811,13 +1812,15 @@ static struct kvm_io_device *vcpu_find_pervcpu_dev(struct kvm_vcpu *vcpu,
 
 
 static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu,
-						gpa_t addr)
+						gpa_t addr, int len,
+						int is_write)
 {
 	struct kvm_io_device *dev;
 
-	dev = vcpu_find_pervcpu_dev(vcpu, addr);
+	dev = vcpu_find_pervcpu_dev(vcpu, addr, len, is_write);
 	if (dev == NULL)
-		dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr);
+		dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr, len,
+					  is_write);
 	return dev;
 }
 
@@ -1885,7 +1888,7 @@ mmio:
 	 * Is this MMIO handled locally?
 	 */
 	mutex_lock(&vcpu->kvm->lock);
-	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
+	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa, bytes, 0);
 	if (mmio_dev) {
 		kvm_iodevice_read(mmio_dev, gpa, bytes, val);
 		mutex_unlock(&vcpu->kvm->lock);
@@ -1940,7 +1943,7 @@ mmio:
 	 * Is this MMIO handled locally?
 	 */
 	mutex_lock(&vcpu->kvm->lock);
-	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
+	mmio_dev = vcpu_find_mmio_dev(vcpu, gpa, bytes, 1);
 	if (mmio_dev) {
 		kvm_iodevice_write(mmio_dev, gpa, bytes, val);
 		mutex_unlock(&vcpu->kvm->lock);
@@ -2317,9 +2320,10 @@ static void pio_string_write(struct kvm_io_device *pio_dev,
 }
 
 static struct kvm_io_device *vcpu_find_pio_dev(struct kvm_vcpu *vcpu,
-					       gpa_t addr)
+					       gpa_t addr, int len,
+					       int is_write)
 {
-	return kvm_io_bus_find_dev(&vcpu->kvm->pio_bus, addr);
+	return kvm_io_bus_find_dev(&vcpu->kvm->pio_bus, addr, len, is_write);
 }
 
 int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
@@ -2351,7 +2355,7 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 
 	kvm_x86_ops->skip_emulated_instruction(vcpu);
 
-	pio_dev = vcpu_find_pio_dev(vcpu, port);
+	pio_dev = vcpu_find_pio_dev(vcpu, port, size, !in);
 	if (pio_dev) {
 		kernel_pio(pio_dev, vcpu, vcpu->arch.pio_data);
 		complete_pio(vcpu);
@@ -2433,7 +2437,9 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
 		}
 	}
 
-	pio_dev = vcpu_find_pio_dev(vcpu, port);
+	pio_dev = vcpu_find_pio_dev(vcpu, port,
+				    vcpu->arch.pio.cur_count,
+				    !vcpu->arch.pio.in);
 	if (!vcpu->arch.pio.in) {
 		/* string PIO write */
 		ret = pio_copy_data(vcpu);
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 865dcbcb891f..499ff0604234 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -52,7 +52,8 @@ struct kvm_io_bus {
 
 void kvm_io_bus_init(struct kvm_io_bus *bus);
 void kvm_io_bus_destroy(struct kvm_io_bus *bus);
-struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, gpa_t addr);
+struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus,
+					  gpa_t addr, int len, int is_write);
 void kvm_io_bus_register_dev(struct kvm_io_bus *bus,
 			     struct kvm_io_device *dev);
 
diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c
index d0c668c6959e..c0d22870ee9c 100644
--- a/virt/kvm/ioapic.c
+++ b/virt/kvm/ioapic.c
@@ -307,7 +307,8 @@ void kvm_ioapic_update_eoi(struct kvm *kvm, int vector)
 			__kvm_ioapic_update_eoi(ioapic, i);
 }
 
-static int ioapic_in_range(struct kvm_io_device *this, gpa_t addr)
+static int ioapic_in_range(struct kvm_io_device *this, gpa_t addr,
+			   int len, int is_write)
 {
 	struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private;
 
diff --git a/virt/kvm/iodev.h b/virt/kvm/iodev.h
index c14e642027b2..55e8846ac3a6 100644
--- a/virt/kvm/iodev.h
+++ b/virt/kvm/iodev.h
@@ -27,7 +27,8 @@ struct kvm_io_device {
 		      gpa_t addr,
 		      int len,
 		      const void *val);
-	int (*in_range)(struct kvm_io_device *this, gpa_t addr);
+	int (*in_range)(struct kvm_io_device *this, gpa_t addr, int len,
+			int is_write);
 	void (*destructor)(struct kvm_io_device *this);
 
 	void             *private;
@@ -49,9 +50,10 @@ static inline void kvm_iodevice_write(struct kvm_io_device *dev,
 	dev->write(dev, addr, len, val);
 }
 
-static inline int kvm_iodevice_inrange(struct kvm_io_device *dev, gpa_t addr)
+static inline int kvm_iodevice_inrange(struct kvm_io_device *dev,
+				       gpa_t addr, int len, int is_write)
 {
-	return dev->in_range(dev, addr);
+	return dev->in_range(dev, addr, len, is_write);
 }
 
 static inline void kvm_iodevice_destructor(struct kvm_io_device *dev)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 83a0e5ce6037..9330fad2b918 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1350,14 +1350,15 @@ void kvm_io_bus_destroy(struct kvm_io_bus *bus)
 	}
 }
 
-struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, gpa_t addr)
+struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus,
+					  gpa_t addr, int len, int is_write)
 {
 	int i;
 
 	for (i = 0; i < bus->dev_count; i++) {
 		struct kvm_io_device *pos = bus->devs[i];
 
-		if (pos->in_range(pos, addr))
+		if (pos->in_range(pos, addr, len, is_write))
 			return pos;
 	}
 
-- 
cgit v1.2.3


From 5f94c1741bdc7a336553122036e8a779e616ccbf Mon Sep 17 00:00:00 2001
From: Laurent Vivier <Laurent.Vivier@bull.net>
Date: Fri, 30 May 2008 16:05:54 +0200
Subject: KVM: Add coalesced MMIO support (common part)

This patch adds all needed structures to coalesce MMIOs.
Until an architecture uses it, it is not compiled.

Coalesced MMIO introduces two ioctl() to define where are the MMIO zones that
can be coalesced:

- KVM_REGISTER_COALESCED_MMIO registers a coalesced MMIO zone.
  It requests one parameter (struct kvm_coalesced_mmio_zone) which defines
  a memory area where MMIOs can be coalesced until the next switch to
  user space. The maximum number of MMIO zones is KVM_COALESCED_MMIO_ZONE_MAX.

- KVM_UNREGISTER_COALESCED_MMIO cancels all registered zones inside
  the given bounds (bounds are also given by struct kvm_coalesced_mmio_zone).

The userspace client can check kernel coalesced MMIO availability by asking
ioctl(KVM_CHECK_EXTENSION) for the KVM_CAP_COALESCED_MMIO capability.
The ioctl() call to KVM_CAP_COALESCED_MMIO will return 0 if not supported,
or the page offset where will be stored the ring buffer.
The page offset depends on the architecture.

After an ioctl(KVM_RUN), the first page of the KVM memory mapped points to
a kvm_run structure. The offset given by KVM_CAP_COALESCED_MMIO is
an offset to the coalesced MMIO ring expressed in PAGE_SIZE relatively
to the address of the start of th kvm_run structure. The MMIO ring buffer
is defined by the structure kvm_coalesced_mmio_ring.

[akio: fix oops during guest shutdown]

Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>
Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/linux/kvm.h       |  29 +++++++++
 include/linux/kvm_host.h  |   4 ++
 virt/kvm/coalesced_mmio.c | 156 ++++++++++++++++++++++++++++++++++++++++++++++
 virt/kvm/coalesced_mmio.h |  23 +++++++
 virt/kvm/kvm_main.c       |  57 +++++++++++++++++
 5 files changed, 269 insertions(+)
 create mode 100644 virt/kvm/coalesced_mmio.c
 create mode 100644 virt/kvm/coalesced_mmio.h

(limited to 'include/linux')

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index a281afeddfbb..1c908ac29c6c 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -173,6 +173,30 @@ struct kvm_run {
 	};
 };
 
+/* for KVM_REGISTER_COALESCED_MMIO / KVM_UNREGISTER_COALESCED_MMIO */
+
+struct kvm_coalesced_mmio_zone {
+	__u64 addr;
+	__u32 size;
+	__u32 pad;
+};
+
+struct kvm_coalesced_mmio {
+	__u64 phys_addr;
+	__u32 len;
+	__u32 pad;
+	__u8  data[8];
+};
+
+struct kvm_coalesced_mmio_ring {
+	__u32 first, last;
+	struct kvm_coalesced_mmio coalesced_mmio[0];
+};
+
+#define KVM_COALESCED_MMIO_MAX \
+	((PAGE_SIZE - sizeof(struct kvm_coalesced_mmio_ring)) / \
+	 sizeof(struct kvm_coalesced_mmio))
+
 /* for KVM_TRANSLATE */
 struct kvm_translation {
 	/* in */
@@ -346,6 +370,7 @@ struct kvm_trace_rec {
 #define KVM_CAP_NOP_IO_DELAY 12
 #define KVM_CAP_PV_MMU 13
 #define KVM_CAP_MP_STATE 14
+#define KVM_CAP_COALESCED_MMIO 15
 
 /*
  * ioctls for VM fds
@@ -371,6 +396,10 @@ struct kvm_trace_rec {
 #define KVM_CREATE_PIT		  _IO(KVMIO,  0x64)
 #define KVM_GET_PIT		  _IOWR(KVMIO, 0x65, struct kvm_pit_state)
 #define KVM_SET_PIT		  _IOR(KVMIO,  0x66, struct kvm_pit_state)
+#define KVM_REGISTER_COALESCED_MMIO \
+			_IOW(KVMIO,  0x67, struct kvm_coalesced_mmio_zone)
+#define KVM_UNREGISTER_COALESCED_MMIO \
+			_IOW(KVMIO,  0x68, struct kvm_coalesced_mmio_zone)
 
 /*
  * ioctls for vcpu fds
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 499ff0604234..d220b4926c4a 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -117,6 +117,10 @@ struct kvm {
 	struct kvm_vm_stat stat;
 	struct kvm_arch arch;
 	atomic_t users_count;
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+	struct kvm_coalesced_mmio_dev *coalesced_mmio_dev;
+	struct kvm_coalesced_mmio_ring *coalesced_mmio_ring;
+#endif
 };
 
 /* The guest did something we don't support. */
diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c
new file mode 100644
index 000000000000..5ae620d32fac
--- /dev/null
+++ b/virt/kvm/coalesced_mmio.c
@@ -0,0 +1,156 @@
+/*
+ * KVM coalesced MMIO
+ *
+ * Copyright (c) 2008 Bull S.A.S.
+ *
+ *  Author: Laurent Vivier <Laurent.Vivier@bull.net>
+ *
+ */
+
+#include "iodev.h"
+
+#include <linux/kvm_host.h>
+#include <linux/kvm.h>
+
+#include "coalesced_mmio.h"
+
+static int coalesced_mmio_in_range(struct kvm_io_device *this,
+				   gpa_t addr, int len, int is_write)
+{
+	struct kvm_coalesced_mmio_dev *dev =
+				(struct kvm_coalesced_mmio_dev*)this->private;
+	struct kvm_coalesced_mmio_zone *zone;
+	int next;
+	int i;
+
+	if (!is_write)
+		return 0;
+
+	/* kvm->lock is taken by the caller and must be not released before
+         * dev.read/write
+         */
+
+	/* Are we able to batch it ? */
+
+	/* last is the first free entry
+	 * check if we don't meet the first used entry
+	 * there is always one unused entry in the buffer
+	 */
+
+	next = (dev->kvm->coalesced_mmio_ring->last + 1) %
+							KVM_COALESCED_MMIO_MAX;
+	if (next == dev->kvm->coalesced_mmio_ring->first) {
+		/* full */
+		return 0;
+	}
+
+	/* is it in a batchable area ? */
+
+	for (i = 0; i < dev->nb_zones; i++) {
+		zone = &dev->zone[i];
+
+		/* (addr,len) is fully included in
+		 * (zone->addr, zone->size)
+		 */
+
+		if (zone->addr <= addr &&
+		    addr + len <= zone->addr + zone->size)
+			return 1;
+	}
+	return 0;
+}
+
+static void coalesced_mmio_write(struct kvm_io_device *this,
+				 gpa_t addr, int len, const void *val)
+{
+	struct kvm_coalesced_mmio_dev *dev =
+				(struct kvm_coalesced_mmio_dev*)this->private;
+	struct kvm_coalesced_mmio_ring *ring = dev->kvm->coalesced_mmio_ring;
+
+	/* kvm->lock must be taken by caller before call to in_range()*/
+
+	/* copy data in first free entry of the ring */
+
+	ring->coalesced_mmio[ring->last].phys_addr = addr;
+	ring->coalesced_mmio[ring->last].len = len;
+	memcpy(ring->coalesced_mmio[ring->last].data, val, len);
+	smp_wmb();
+	ring->last = (ring->last + 1) % KVM_COALESCED_MMIO_MAX;
+}
+
+static void coalesced_mmio_destructor(struct kvm_io_device *this)
+{
+	kfree(this);
+}
+
+int kvm_coalesced_mmio_init(struct kvm *kvm)
+{
+	struct kvm_coalesced_mmio_dev *dev;
+
+	dev = kzalloc(sizeof(struct kvm_coalesced_mmio_dev), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+	dev->dev.write  = coalesced_mmio_write;
+	dev->dev.in_range  = coalesced_mmio_in_range;
+	dev->dev.destructor  = coalesced_mmio_destructor;
+	dev->dev.private  = dev;
+	dev->kvm = kvm;
+	kvm->coalesced_mmio_dev = dev;
+	kvm_io_bus_register_dev(&kvm->mmio_bus, &dev->dev);
+
+	return 0;
+}
+
+int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm,
+				         struct kvm_coalesced_mmio_zone *zone)
+{
+	struct kvm_coalesced_mmio_dev *dev = kvm->coalesced_mmio_dev;
+
+	if (dev == NULL)
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+	if (dev->nb_zones >= KVM_COALESCED_MMIO_ZONE_MAX) {
+		mutex_unlock(&kvm->lock);
+		return -ENOBUFS;
+	}
+
+	dev->zone[dev->nb_zones] = *zone;
+	dev->nb_zones++;
+
+	mutex_unlock(&kvm->lock);
+	return 0;
+}
+
+int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm,
+					   struct kvm_coalesced_mmio_zone *zone)
+{
+	int i;
+	struct kvm_coalesced_mmio_dev *dev = kvm->coalesced_mmio_dev;
+	struct kvm_coalesced_mmio_zone *z;
+
+	if (dev == NULL)
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+
+	i = dev->nb_zones;
+	while(i) {
+		z = &dev->zone[i - 1];
+
+		/* unregister all zones
+		 * included in (zone->addr, zone->size)
+		 */
+
+		if (zone->addr <= z->addr &&
+		    z->addr + z->size <= zone->addr + zone->size) {
+			dev->nb_zones--;
+			*z = dev->zone[dev->nb_zones];
+		}
+		i--;
+	}
+
+	mutex_unlock(&kvm->lock);
+
+	return 0;
+}
diff --git a/virt/kvm/coalesced_mmio.h b/virt/kvm/coalesced_mmio.h
new file mode 100644
index 000000000000..5ac0ec628461
--- /dev/null
+++ b/virt/kvm/coalesced_mmio.h
@@ -0,0 +1,23 @@
+/*
+ * KVM coalesced MMIO
+ *
+ * Copyright (c) 2008 Bull S.A.S.
+ *
+ *  Author: Laurent Vivier <Laurent.Vivier@bull.net>
+ *
+ */
+
+#define KVM_COALESCED_MMIO_ZONE_MAX 100
+
+struct kvm_coalesced_mmio_dev {
+	struct kvm_io_device dev;
+	struct kvm *kvm;
+	int nb_zones;
+	struct kvm_coalesced_mmio_zone zone[KVM_COALESCED_MMIO_ZONE_MAX];
+};
+
+int kvm_coalesced_mmio_init(struct kvm *kvm);
+int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm,
+                                       struct kvm_coalesced_mmio_zone *zone);
+int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm,
+                                         struct kvm_coalesced_mmio_zone *zone);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 9330fad2b918..7d10dfa0d388 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -47,6 +47,10 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+#include "coalesced_mmio.h"
+#endif
+
 MODULE_AUTHOR("Qumranet");
 MODULE_LICENSE("GPL");
 
@@ -185,10 +189,23 @@ EXPORT_SYMBOL_GPL(kvm_vcpu_uninit);
 static struct kvm *kvm_create_vm(void)
 {
 	struct kvm *kvm = kvm_arch_create_vm();
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+	struct page *page;
+#endif
 
 	if (IS_ERR(kvm))
 		goto out;
 
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
+	if (!page) {
+		kfree(kvm);
+		return ERR_PTR(-ENOMEM);
+	}
+	kvm->coalesced_mmio_ring =
+			(struct kvm_coalesced_mmio_ring *)page_address(page);
+#endif
+
 	kvm->mm = current->mm;
 	atomic_inc(&kvm->mm->mm_count);
 	spin_lock_init(&kvm->mmu_lock);
@@ -200,6 +217,9 @@ static struct kvm *kvm_create_vm(void)
 	spin_lock(&kvm_lock);
 	list_add(&kvm->vm_list, &vm_list);
 	spin_unlock(&kvm_lock);
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+	kvm_coalesced_mmio_init(kvm);
+#endif
 out:
 	return kvm;
 }
@@ -242,6 +262,10 @@ static void kvm_destroy_vm(struct kvm *kvm)
 	spin_unlock(&kvm_lock);
 	kvm_io_bus_destroy(&kvm->pio_bus);
 	kvm_io_bus_destroy(&kvm->mmio_bus);
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+	if (kvm->coalesced_mmio_ring != NULL)
+		free_page((unsigned long)kvm->coalesced_mmio_ring);
+#endif
 	kvm_arch_destroy_vm(kvm);
 	mmdrop(mm);
 }
@@ -825,6 +849,10 @@ static int kvm_vcpu_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 #ifdef CONFIG_X86
 	else if (vmf->pgoff == KVM_PIO_PAGE_OFFSET)
 		page = virt_to_page(vcpu->arch.pio_data);
+#endif
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+	else if (vmf->pgoff == KVM_COALESCED_MMIO_PAGE_OFFSET)
+		page = virt_to_page(vcpu->kvm->coalesced_mmio_ring);
 #endif
 	else
 		return VM_FAULT_SIGBUS;
@@ -1148,6 +1176,32 @@ static long kvm_vm_ioctl(struct file *filp,
 			goto out;
 		break;
 	}
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+	case KVM_REGISTER_COALESCED_MMIO: {
+		struct kvm_coalesced_mmio_zone zone;
+		r = -EFAULT;
+		if (copy_from_user(&zone, argp, sizeof zone))
+			goto out;
+		r = -ENXIO;
+		r = kvm_vm_ioctl_register_coalesced_mmio(kvm, &zone);
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
+	case KVM_UNREGISTER_COALESCED_MMIO: {
+		struct kvm_coalesced_mmio_zone zone;
+		r = -EFAULT;
+		if (copy_from_user(&zone, argp, sizeof zone))
+			goto out;
+		r = -ENXIO;
+		r = kvm_vm_ioctl_unregister_coalesced_mmio(kvm, &zone);
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
+#endif
 	default:
 		r = kvm_arch_vm_ioctl(filp, ioctl, arg);
 	}
@@ -1231,6 +1285,9 @@ static long kvm_dev_ioctl(struct file *filp,
 		r = PAGE_SIZE;     /* struct kvm_run */
 #ifdef CONFIG_X86
 		r += PAGE_SIZE;    /* pio data page */
+#endif
+#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET
+		r += PAGE_SIZE;    /* coalesced mmio ring page */
 #endif
 		break;
 	case KVM_TRACE_ENABLE:
-- 
cgit v1.2.3


From 9ef621d3be56e1188300476a8102ff54f7b6793f Mon Sep 17 00:00:00 2001
From: "Tan, Li" <li.tan@intel.com>
Date: Fri, 23 May 2008 14:54:09 +0800
Subject: KVM: Support mixed endian machines

Currently kvmtrace is not portable. This will prevent from copying a
trace file from big-endian target to little-endian workstation for analysis.
In the patch, kernel outputs metadata containing a magic number to trace
log, and changes 64-bit words to be u64 instead of a pair of u32s.

Signed-off-by: Tan Li <li.tan@intel.com>
Acked-by: Jerone Young <jyoung5@us.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 include/linux/kvm.h  |  4 ++--
 virt/kvm/kvm_trace.c | 18 ++++++++++++------
 2 files changed, 14 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 1c908ac29c6c..0ea064cbfbc8 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -318,14 +318,14 @@ struct kvm_trace_rec {
 	__u32 vcpu_id;
 	union {
 		struct {
-			__u32 cycle_lo, cycle_hi;
+			__u64 cycle_u64;
 			__u32 extra_u32[KVM_TRC_EXTRA_MAX];
 		} cycle;
 		struct {
 			__u32 extra_u32[KVM_TRC_EXTRA_MAX];
 		} nocycle;
 	} u;
-};
+} __attribute__((packed));
 
 #define KVMIO 0xAE
 
diff --git a/virt/kvm/kvm_trace.c b/virt/kvm/kvm_trace.c
index 0e495470788d..58141f31ea8f 100644
--- a/virt/kvm/kvm_trace.c
+++ b/virt/kvm/kvm_trace.c
@@ -72,11 +72,7 @@ static void kvm_add_trace(void *probe_private, void *call_data,
 	rec.cycle_in 	= p->cycle_in;
 
 	if (rec.cycle_in) {
-		u64 cycle = 0;
-
-		cycle = get_cycles();
-		rec.u.cycle.cycle_lo = (u32)cycle;
-		rec.u.cycle.cycle_hi = (u32)(cycle >> 32);
+		rec.u.cycle.cycle_u64 = get_cycles();
 
 		for (i = 0; i < rec.extra_u32; i++)
 			rec.u.cycle.extra_u32[i] = va_arg(*args, u32);
@@ -114,8 +110,18 @@ static int kvm_subbuf_start_callback(struct rchan_buf *buf, void *subbuf,
 {
 	struct kvm_trace *kt;
 
-	if (!relay_buf_full(buf))
+	if (!relay_buf_full(buf)) {
+		if (!prev_subbuf) {
+			/*
+			 * executed only once when the channel is opened
+			 * save metadata as first record
+			 */
+			subbuf_start_reserve(buf, sizeof(u32));
+			*(u32 *)subbuf = 0x12345678;
+		}
+
 		return 1;
+	}
 
 	kt = buf->chan->private_data;
 	atomic_inc(&kt->lost_records);
-- 
cgit v1.2.3


From 34d4cb8fca1f2a31be152b74797e6cd160ec9de6 Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Thu, 10 Jul 2008 20:49:31 -0300
Subject: KVM: MMU: nuke shadowed pgtable pages and ptes on memslot destruction

Flush the shadow mmu before removing regions to avoid stale entries.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 arch/ia64/kvm/kvm-ia64.c   | 3 +++
 arch/powerpc/kvm/powerpc.c | 4 ++++
 arch/s390/kvm/kvm-s390.c   | 4 ++++
 arch/x86/kvm/x86.c         | 5 +++++
 include/linux/kvm_host.h   | 1 +
 virt/kvm/kvm_main.c        | 3 +++
 6 files changed, 20 insertions(+)

(limited to 'include/linux')

diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 9408b30576d6..2672f4d278ac 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1455,6 +1455,9 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
 	return 0;
 }
 
+void kvm_arch_flush_shadow(struct kvm *kvm)
+{
+}
 
 long kvm_arch_dev_ioctl(struct file *filp,
 		unsigned int ioctl, unsigned long arg)
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index b850d2497027..53826a5f6c06 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -170,6 +170,10 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
 	return 0;
 }
 
+void kvm_arch_flush_shadow(struct kvm *kvm)
+{
+}
+
 struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id)
 {
 	struct kvm_vcpu *vcpu;
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 399acf3f64dd..1782cbcd2829 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -675,6 +675,10 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
 	return 0;
 }
 
+void kvm_arch_flush_shadow(struct kvm *kvm)
+{
+}
+
 gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
 {
 	return gfn;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index b131f3c0cf64..9f1cdb011cff 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -4032,6 +4032,11 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
 	return 0;
 }
 
+void kvm_arch_flush_shadow(struct kvm *kvm)
+{
+	kvm_mmu_zap_all(kvm);
+}
+
 int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 {
 	return vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index d220b4926c4a..07d68a8ae8e9 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -168,6 +168,7 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
 				struct kvm_userspace_memory_region *mem,
 				struct kvm_memory_slot old,
 				int user_alloc);
+void kvm_arch_flush_shadow(struct kvm *kvm);
 gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn);
 struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn);
 unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 9ccaf8f5402e..30b36368fcdf 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -405,6 +405,9 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	if (mem->slot >= kvm->nmemslots)
 		kvm->nmemslots = mem->slot + 1;
 
+	if (!npages)
+		kvm_arch_flush_shadow(kvm);
+
 	*memslot = new;
 
 	r = kvm_arch_set_memory_region(kvm, mem, old, user_alloc);
-- 
cgit v1.2.3


From 1d0ba5f3784612fe6e91a12e0dec37c797d4f07c Mon Sep 17 00:00:00 2001
From: Tobias Lorenz <tobias.lorenz@gmx.net>
Date: Mon, 26 May 2008 18:40:46 -0300
Subject: V4L/DVB (7942): Hardware frequency seek ioctl interface

Signed-off-by: Tobias Lorenz <tobias.lorenz@gmx.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/compat_ioctl32.c |  1 +
 drivers/media/video/videodev.c       | 12 ++++++++++++
 include/linux/videodev2.h            | 10 ++++++++++
 include/media/v4l2-dev.h             |  2 ++
 4 files changed, 25 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/media/video/compat_ioctl32.c b/drivers/media/video/compat_ioctl32.c
index cefd1381e8de..54de0cd482e9 100644
--- a/drivers/media/video/compat_ioctl32.c
+++ b/drivers/media/video/compat_ioctl32.c
@@ -884,6 +884,7 @@ long v4l_compat_ioctl32(struct file *file, unsigned int cmd, unsigned long arg)
 	case VIDIOC_G_INPUT32:
 	case VIDIOC_S_INPUT32:
 	case VIDIOC_TRY_FMT32:
+	case VIDIOC_S_HW_FREQ_SEEK:
 		ret = do_video_ioctl(file, cmd, arg);
 		break;
 
diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c
index 7649860a388d..52c56678ee69 100644
--- a/drivers/media/video/videodev.c
+++ b/drivers/media/video/videodev.c
@@ -278,6 +278,7 @@ static const char *v4l2_ioctls[] = {
 	[_IOC_NR(VIDIOC_DBG_G_REGISTER)]   = "VIDIOC_DBG_G_REGISTER",
 
 	[_IOC_NR(VIDIOC_G_CHIP_IDENT)]     = "VIDIOC_G_CHIP_IDENT",
+	[_IOC_NR(VIDIOC_S_HW_FREQ_SEEK)]   = "VIDIOC_S_HW_FREQ_SEEK",
 #endif
 };
 #define V4L2_IOCTLS ARRAY_SIZE(v4l2_ioctls)
@@ -1763,6 +1764,17 @@ static int __video_do_ioctl(struct inode *inode, struct file *file,
 		ret = vfd->vidioc_default(file, fh, cmd, arg);
 		break;
 	}
+	case VIDIOC_S_HW_FREQ_SEEK:
+	{
+		struct v4l2_hw_freq_seek *p = arg;
+		if (!vfd->vidioc_s_hw_freq_seek)
+			break;
+		dbgarg(cmd,
+			"tuner=%d, type=%d, seek_upward=%d, wrap_around=%d\n",
+			p->tuner, p->type, p->seek_upward, p->wrap_around);
+		ret = vfd->vidioc_s_hw_freq_seek(file, fh, p);
+		break;
+	}
 	} /* switch */
 
 	if (vfd->debug & V4L2_DEBUG_IOCTL_ARG) {
diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 4a535ea1e123..9385c823a978 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -246,6 +246,7 @@ struct v4l2_capability
 #define V4L2_CAP_SLICED_VBI_OUTPUT	0x00000080  /* Is a sliced VBI output device */
 #define V4L2_CAP_RDS_CAPTURE		0x00000100  /* RDS data capture */
 #define V4L2_CAP_VIDEO_OUTPUT_OVERLAY	0x00000200  /* Can do video output overlay */
+#define V4L2_CAP_HW_FREQ_SEEK		0x00000400  /* Can do hardware frequency seek  */
 
 #define V4L2_CAP_TUNER			0x00010000  /* has a tuner */
 #define V4L2_CAP_AUDIO			0x00020000  /* has audio support */
@@ -1156,6 +1157,14 @@ struct v4l2_frequency
 	__u32		      reserved[8];
 };
 
+struct v4l2_hw_freq_seek {
+	__u32		      tuner;
+	enum v4l2_tuner_type  type;
+	__u32		      seek_upward;
+	__u32		      wrap_around;
+	__u32		      reserved[8];
+};
+
 /*
  *	A U D I O
  */
@@ -1441,6 +1450,7 @@ struct v4l2_chip_ident {
 
 #define VIDIOC_G_CHIP_IDENT     _IOWR ('V', 81, struct v4l2_chip_ident)
 #endif
+#define VIDIOC_S_HW_FREQ_SEEK	_IOW  ('V', 82, struct v4l2_hw_freq_seek)
 
 #ifdef __OLD_VIDIOC_
 /* for compatibility, will go away some day */
diff --git a/include/media/v4l2-dev.h b/include/media/v4l2-dev.h
index 859f7a6f6f67..53b651ac685a 100644
--- a/include/media/v4l2-dev.h
+++ b/include/media/v4l2-dev.h
@@ -306,6 +306,8 @@ struct video_device
 	/* Log status ioctl */
 	int (*vidioc_log_status)       (struct file *file, void *fh);
 
+	int (*vidioc_s_hw_freq_seek)   (struct file *file, void *fh,
+					struct v4l2_hw_freq_seek *a);
 
 	/* Debugging ioctls */
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-- 
cgit v1.2.3


From 6a7eba24e4f0ff725d33159f6265e3a79d53a833 Mon Sep 17 00:00:00 2001
From: Jean-Francois Moine <moinejf@free.fr>
Date: Mon, 30 Jun 2008 15:50:11 -0300
Subject: V4L/DVB (8157): gspca: all subdrivers

- remaning subdrivers added
- remove the decoding helper and some specific frame decodings

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 Documentation/video4linux/gspca.txt |  306 ++---
 drivers/media/video/Makefile        |    1 +
 drivers/media/video/gspca/Kconfig   |    2 +-
 drivers/media/video/gspca/Makefile  |   26 +-
 drivers/media/video/gspca/conex.c   | 1059 +++++++++++++++++
 drivers/media/video/gspca/etoms.c   | 1062 +++++++++++++++++
 drivers/media/video/gspca/gspca.c   |  579 ++-------
 drivers/media/video/gspca/gspca.h   |   39 +-
 drivers/media/video/gspca/mars.c    |  455 +++++++
 drivers/media/video/gspca/ov519.c   | 2174 ++++++++++++++++++++++++++++++++++
 drivers/media/video/gspca/pac207.c  |   10 +-
 drivers/media/video/gspca/pac7311.c |  754 ++++++++++++
 drivers/media/video/gspca/sonixb.c  |  879 ++++++++++++++
 drivers/media/video/gspca/sonixj.c  | 1629 +++++++++++++++++++++++++
 drivers/media/video/gspca/spca500.c | 1195 +++++++++++++++++++
 drivers/media/video/gspca/spca501.c | 2219 +++++++++++++++++++++++++++++++++++
 drivers/media/video/gspca/spca505.c |  933 +++++++++++++++
 drivers/media/video/gspca/spca506.c |  830 +++++++++++++
 drivers/media/video/gspca/spca508.c | 1774 ++++++++++++++++++++++++++++
 drivers/media/video/gspca/spca561.c | 1025 ++++++++++++++++
 drivers/media/video/gspca/stk014.c  |  170 ++-
 drivers/media/video/gspca/sunplus.c | 1638 ++++++++++++++++++++++++++
 drivers/media/video/gspca/t613.c    | 1013 ++++++++++++++++
 drivers/media/video/gspca/tv8532.c  |  709 +++++++++++
 drivers/media/video/gspca/vc032x.c  | 1816 ++++++++++++++++++++++++++++
 drivers/media/video/gspca/zc3xx.c   |  649 +++++-----
 include/linux/videodev2.h           |    2 +
 27 files changed, 21964 insertions(+), 984 deletions(-)
 create mode 100644 drivers/media/video/gspca/conex.c
 create mode 100644 drivers/media/video/gspca/etoms.c
 create mode 100644 drivers/media/video/gspca/mars.c
 create mode 100644 drivers/media/video/gspca/ov519.c
 create mode 100644 drivers/media/video/gspca/pac7311.c
 create mode 100644 drivers/media/video/gspca/sonixb.c
 create mode 100644 drivers/media/video/gspca/sonixj.c
 create mode 100644 drivers/media/video/gspca/spca500.c
 create mode 100644 drivers/media/video/gspca/spca501.c
 create mode 100644 drivers/media/video/gspca/spca505.c
 create mode 100644 drivers/media/video/gspca/spca506.c
 create mode 100644 drivers/media/video/gspca/spca508.c
 create mode 100644 drivers/media/video/gspca/spca561.c
 create mode 100644 drivers/media/video/gspca/sunplus.c
 create mode 100644 drivers/media/video/gspca/t613.c
 create mode 100644 drivers/media/video/gspca/tv8532.c
 create mode 100644 drivers/media/video/gspca/vc032x.c

(limited to 'include/linux')

diff --git a/Documentation/video4linux/gspca.txt b/Documentation/video4linux/gspca.txt
index 9c404b56dbb3..37996e59d516 100644
--- a/Documentation/video4linux/gspca.txt
+++ b/Documentation/video4linux/gspca.txt
@@ -1,4 +1,4 @@
-Here the list of the known working cameras with gspca.
+List of the webcams know by gspca.
 
 The modules are:
 	gspca_main	main driver
@@ -6,106 +6,73 @@ The modules are:
 
 xxxx		vend:prod
 ----
-conex		0572:0041	Creative Notebook cx11646
-etoms		102c:6151	Qcam Sangha CIF
-etoms		102c:6251	Qcam xxxxxx VGA
-mars		093a:050f	Mars-Semi Pc-Camera
+spca501		0000:0000	MystFromOri Unknow Camera
+spca501		040a:0002	Kodak DVC-325
+spca500		040a:0300	Kodak EZ200
+zc3xx		041e:041e	Creative WebCam Live!
+spca500		041e:400a	Creative PC-CAM 300
+sunplus		041e:400b	Creative PC-CAM 600
+sunplus		041e:4012	PC-Cam350
+sunplus		041e:4013	Creative Pccam750
+zc3xx		041e:4017	Creative Webcam Mobile PD1090
+spca508		041e:4018	Creative Webcam Vista (PD1100)
+spca561		041e:401a	Creative Webcam Vista (PD1100)
+zc3xx		041e:401c	Creative NX
+spca505		041e:401d	Creative Webcam NX ULTRA
+zc3xx		041e:401e	Creative Nx Pro
+zc3xx		041e:401f	Creative Webcam Notebook PD1171
+pac207		041e:4028	Creative Webcam Vista Plus
+zc3xx		041e:4029	Creative WebCam Vista Pro
+zc3xx		041e:4034	Creative Instant P0620
+zc3xx		041e:4035	Creative Instant P0620D
+zc3xx		041e:4036	Creative Live !
+zc3xx		041e:403a	Creative Nx Pro 2
+spca561		041e:403b	Creative Webcam Vista (VF0010)
+zc3xx		041e:4051	Creative Live!Cam Notebook Pro (VF0250)
 ov519		041e:4052	Creative Live! VISTA IM
+zc3xx		041e:4053	Creative Live!Cam Video IM
 ov519		041e:405f	Creative Live! VISTA VF0330
 ov519		041e:4060	Creative Live! VISTA VF0350
 ov519		041e:4061	Creative Live! VISTA VF0400
 ov519		041e:4064	Creative Live! VISTA VF0420
 ov519		041e:4068	Creative Live! VISTA VF0470
-ov519		045e:028c	Micro$oft xbox cam
-ov519		054c:0154	Sonny toy4
-ov519		054c:0155	Sonny toy5
-ov519		05a9:0519	OmniVision
-ov519		05a9:4519	OmniVision
-ov519		05a9:8519	OmniVision
-ov519		05a9:0530	OmniVision
-pac207		041e:4028	Creative Webcam Vista Plus
-pac207		093a:2460	PAC207 Qtec Webcam 100
-pac207		093a:2463	Philips spc200nc pac207
-pac207		093a:2464	Labtec Webcam 1200
-pac207		093a:2468	PAC207
-pac207		093a:2470	Genius GF112
-pac207		093a:2471	PAC207 Genius VideoCam ge111
-pac207		093a:2472	PAC207 Genius VideoCam ge110
-pac7311		093a:2600	PAC7311 Typhoon
-pac7311		093a:2601	PAC7311 Phillips SPC610NC
-pac7311		093a:2603	PAC7312
-pac7311		093a:2608	PAC7311 Trust WB-3300p
-pac7311		093a:260e	PAC7311 Gigaware VGA PC Camera, Trust WB-3350p, SIGMA cam 2350
-pac7311		093a:260f	PAC7311 SnakeCam
-sonixb		0c45:6001	Genius VideoCAM NB
-sonixb		0c45:6005	Sweex Tas5110
-sonixb		0c45:6007	Sonix sn9c101 + Tas5110D
-sonixb		0c45:6009	spcaCam@120
-sonixb		0c45:600d	spcaCam@120
-sonixb		0c45:6011	MAX Webcam (Microdia - OV6650 - SN9C101G)
-sonixb		0c45:6019	Generic Sonix OV7630
-sonixb		0c45:6024	Generic Sonix Tas5130c
-sonixb		0c45:6025	Xcam Shanga
-sonixb		0c45:6028	Sonix Btc Pc380
-sonixb		0c45:6029	spcaCam@150
-sonixb		0c45:602c	Generic Sonix OV7630
-sonixb		0c45:602d	LIC-200 LG
-sonixb		0c45:602e	Genius VideoCam Messenger
+spca561		0458:7004	Genius VideoCAM Express V2
+sunplus		0458:7006	Genius Dsc 1.3 Smart
+zc3xx		0458:7007	Genius VideoCam V2
+zc3xx		0458:700c	Genius VideoCam V3
+zc3xx		0458:700f	Genius VideoCam Web V2
 sonixj		0458:7025	Genius Eye 311Q
 sonixj		045e:00f5	MicroSoft VX3000
 sonixj		045e:00f7	MicroSoft VX1000
-sonixj		0471:0327	Philips SPC 600 NC
-sonixj		0471:0328	Philips SPC 700 NC
-sonixj		0471:0330	Philips SPC 710NC
-sonixj		0c45:6040	Speed NVC 350K
-sonixj		0c45:607c	Sonix sn9c102p Hv7131R
-sonixj		0c45:60c0	Sangha Sn535
-sonixj		0c45:60ec	SN9C105+MO4000
-sonixj		0c45:60fb	Surfer NoName
-sonixj		0c45:60fc	LG-LIC300
-sonixj		0c45:612a	Avant Camera
-sonixj		0c45:612c	Typhoon Rasy Cam 1.3MPix
-sonixj		0c45:6130	Sonix Pccam
-sonixj		0c45:6138	Sn9c120 Mo4000
-sonixj		0c45:613b	Surfer SN-206
-sonixj		0c45:613c	Sonix Pccam168
-spca500		040a:0300	Kodak EZ200
-spca500		041e:400a	Creative PC-CAM 300
+ov519		045e:028c	Micro$oft xbox cam
+spca508		0461:0815	Micro Innovation IC200
+zc3xx		0461:0a00	MicroInnovation WebCam320
 spca500		046d:0890	Logitech QuickCam traveler
+vc032x		046d:0892	Logitech Orbicam
+vc032x		046d:0896	Logitech Orbicam
+zc3xx		046d:08a0	Logitech QC IM
+zc3xx		046d:08a1	Logitech QC IM 0x08A1 +sound
+zc3xx		046d:08a2	Labtec Webcam Pro
+zc3xx		046d:08a3	Logitech QC Chat
+zc3xx		046d:08a6	Logitech QCim
+zc3xx		046d:08a7	Logitech QuickCam Image
+zc3xx		046d:08a9	Logitech Notebook Deluxe
+zc3xx		046d:08aa	Labtec Webcam  Notebook
+zc3xx		046d:08ac	Logitech QuickCam Cool
+zc3xx		046d:08ad	Logitech QCCommunicate STX
+zc3xx		046d:08ae	Logitech QuickCam for Notebooks
+zc3xx		046d:08af	Logitech QuickCam Cool
+zc3xx		046d:08b9	Logitech QC IM ???
+zc3xx		046d:08d7	Logitech QCam STX
+zc3xx		046d:08d9	Logitech QuickCam IM/Connect
+zc3xx		046d:08d8	Logitech Notebook Deluxe
+zc3xx		046d:08da	Logitech QuickCam Messenger
+zc3xx		046d:08dd	Logitech QuickCam for Notebooks
 spca500		046d:0900	Logitech Inc. ClickSmart 310
 spca500		046d:0901	Logitech Inc. ClickSmart 510
-spca500		04a5:300c	Benq DC1016
-spca500		04fc:7333	PalmPixDC85
-spca500		055f:c200	Mustek Gsmart 300
-spca500		055f:c220	Gsmart Mini
-spca500		06bd:0404	Agfa CL20
-spca500		06be:0800	Optimedia
-spca500		084d:0003	D-Link DSC-350
-spca500		08ca:0103	Aiptek PocketDV
-spca500		2899:012c	Toptro Industrial
-spca500		8086:0630	Intel Pocket PC Camera
-spca501		040a:0002	Kodak DVC-325
-spca501		0497:c001	Smile International
-spca501		0506:00df	3Com HomeConnect Lite
-spca501		0733:0401	Intel Create and Share
-spca501		0733:0402	ViewQuest M318B
-spca501		1776:501c	Arowana 300K CMOS Camera
-spca501		0000:0000	MystFromOri Unknow Camera
-spca505		041e:401d	Creative Webcam NX ULTRA
-spca505		0733:0430	Intel PC Camera Pro
-spca506		06e1:a190	ADS Instant VCD
-spca506		0734:043b	3DeMon USB Capture aka
-spca506		99fa:8988	Grandtec V.cap
-spca506		99fa:8988	Grandtec V.cap
-spca508		041e:4018	Creative Webcam Vista (PD1100)
-spca508		0461:0815	Micro Innovation IC200
-spca508		0733:0110	ViewQuest VQ110
-spca508		0af9:0010	Hama USB Sightcam 100
-spca508		0af9:0011	Hama USB Sightcam 100
-spca508		8086:0110	Intel Easy PC Camera
-spca561		041e:401a	Creative Webcam Vista (PD1100)
-spca561		041e:403b	Creative Webcam Vista (VF0010)
-spca561		0458:7004	Genius VideoCAM Express V2
+sunplus		046d:0905	Logitech ClickSmart 820
+tv8532		046d:0920	QC Express
+tv8532		046d:0921	Labtec Webcam
 spca561		046d:0928	Logitech QC Express Etch2
 spca561		046d:0929	Labtec Webcam Elch2
 spca561		046d:092a	Logitech QC for Notebook
@@ -114,33 +81,42 @@ spca561		046d:092c	Logitech QC chat Elch2
 spca561		046d:092d	Logitech QC Elch2
 spca561		046d:092e	Logitech QC Elch2
 spca561		046d:092f	Logitech QC Elch2
-spca561		04fc:0561	Flexcam 100
-spca561		060b:a001	Maxell Compact Pc PM3
-spca561		10fd:7e50	FlyCam Usb 100
-spca561		abcd:cdee	Petcam
-stk014		05e1:0893	Syntek DV4000
-sunplus		041e:400b	Creative PC-CAM 600
-sunplus		041e:4012	PC-Cam350
-sunplus		041e:4013	Creative Pccam750
-sunplus		0458:7006	Genius Dsc 1.3 Smart
-sunplus		046d:0905	Logitech ClickSmart 820
 sunplus		046d:0960	Logitech ClickSmart 420
 sunplus		0471:0322	Philips DMVC1300K
+zc3xx		0471:0325	Philips SPC 200 NC
+zc3xx		0471:0326	Philips SPC 300 NC
+sonixj		0471:0327	Philips SPC 600 NC
+sonixj		0471:0328	Philips SPC 700 NC
+zc3xx		0471:032d	Philips spc210nc
+zc3xx		0471:032e	Philips spc315nc
+sonixj		0471:0330	Philips SPC 710NC
+spca501		0497:c001	Smile International
 sunplus		04a5:3003	Benq DC 1300
 sunplus		04a5:3008	Benq DC 1500
 sunplus		04a5:300a	Benq DC3410
+spca500		04a5:300c	Benq DC1016
 sunplus		04f1:1001	JVC GC A50
+spca561		04fc:0561	Flexcam 100
 sunplus		04fc:500c	Sunplus CA500C
 sunplus		04fc:504a	Aiptek Mini PenCam 1.3
 sunplus		04fc:504b	Maxell MaxPocket LE 1.3
 sunplus		04fc:5330	Digitrex 2110
 sunplus		04fc:5360	Sunplus Generic
+spca500		04fc:7333	PalmPixDC85
 sunplus		04fc:ffff	Pure DigitalDakota
+spca501		0506:00df	3Com HomeConnect Lite
 sunplus		052b:1513	Megapix V4
+tv8532		0545:808b	Veo Stingray
+tv8532		0545:8333	Veo Stingray
 sunplus		0546:3155	Polaroid PDC3070
 sunplus		0546:3191	Polaroid Ion 80
 sunplus		0546:3273	Polaroid PDC2030
+ov519		054c:0154	Sonny toy4
+ov519		054c:0155	Sonny toy5
+zc3xx		055f:c005	Mustek Wcam300A
+spca500		055f:c200	Mustek Gsmart 300
 sunplus		055f:c211	Kowa Bs888e Microcamera
+spca500		055f:c220	Gsmart Mini
 sunplus		055f:c230	Mustek Digicam 330K
 sunplus		055f:c232	Mustek MDC3500
 sunplus		055f:c360	Mustek DV4000 Mpeg4
@@ -152,14 +128,34 @@ sunplus		055f:c530	Mustek Gsmart LCD 3
 sunplus		055f:c540	Gsmart D30
 sunplus		055f:c630	Mustek MDC4000
 sunplus		055f:c650	Mustek MDC5500Z
+zc3xx		055f:d003	Mustek WCam300A
+zc3xx		055f:d004	Mustek WCam300 AN
+conex		0572:0041	Creative Notebook cx11646
+ov519		05a9:0519	OmniVision
+ov519		05a9:0530	OmniVision
+ov519		05a9:4519	OmniVision
+ov519		05a9:8519	OmniVision
 sunplus		05da:1018	Digital Dream Enigma 1.3
+stk014		05e1:0893	Syntek DV4000
+spca561		060b:a001	Maxell Compact Pc PM3
+zc3xx		0698:2003	CTX M730V built in
+spca500		06bd:0404	Agfa CL20
+spca500		06be:0800	Optimedia
 sunplus		06d6:0031	Trust 610 LCD PowerC@m Zoom
+spca506		06e1:a190	ADS Instant VCD
+spca508		0733:0110	ViewQuest VQ110
+spca501		0733:0401	Intel Create and Share
+spca501		0733:0402	ViewQuest M318B
+spca505		0733:0430	Intel PC Camera Pro
 sunplus		0733:1311	Digital Dream Epsilon 1.3
 sunplus		0733:1314	Mercury 2.1MEG Deluxe Classic Cam
 sunplus		0733:2211	Jenoptik jdc 21 LCD
 sunplus		0733:2221	Mercury Digital Pro 3.1p
 sunplus		0733:3261	Concord 3045 spca536a
 sunplus		0733:3281	Cyberpix S550V
+spca506		0734:043b	3DeMon USB Capture aka
+spca500		084d:0003	D-Link DSC-350
+spca500		08ca:0103	Aiptek PocketDV
 sunplus		08ca:0104	Aiptek PocketDVII 1.3
 sunplus		08ca:0106	Aiptek Pocket DV3100+
 sunplus		08ca:2008	Aiptek Mini PenCam 2 M
@@ -173,66 +169,72 @@ sunplus		08ca:2028	Aiptek PocketCam4M
 sunplus		08ca:2040	Aiptek PocketDV4100M
 sunplus		08ca:2042	Aiptek PocketDV5100
 sunplus		08ca:2060	Aiptek PocketDV5300
-sunplus		0d64:0303	Sunplus FashionCam DXG
-tv8532		046d:0920	QC Express
-tv8532		046d:0921	Labtec Webcam
-tv8532		0545:808b	Veo Stingray
-tv8532		0545:8333	Veo Stingray
 tv8532		0923:010f	ICM532 cams
-vc032x		046d:0892	Logitech Orbicam
-vc032x		046d:0896	Logitech Orbicam
+mars		093a:050f	Mars-Semi Pc-Camera
+pac207		093a:2460	PAC207 Qtec Webcam 100
+pac207		093a:2463	Philips spc200nc pac207
+pac207		093a:2464	Labtec Webcam 1200
+pac207		093a:2468	PAC207
+pac207		093a:2470	Genius GF112
+pac207		093a:2471	PAC207 Genius VideoCam ge111
+pac207		093a:2472	PAC207 Genius VideoCam ge110
+pac7311		093a:2600	PAC7311 Typhoon
+pac7311		093a:2601	PAC7311 Phillips SPC610NC
+pac7311		093a:2603	PAC7312
+pac7311		093a:2608	PAC7311 Trust WB-3300p
+pac7311		093a:260e	PAC7311 Gigaware VGA PC Camera, Trust WB-3350p, SIGMA cam 2350
+pac7311		093a:260f	PAC7311 SnakeCam
+pac7311		093a:2621	PAC731x
+zc3xx		0ac8:0302	Z-star Vimicro zc0302
 vc032x		0ac8:0321	Vimicro generic vc0321
 vc032x		0ac8:0323	Vimicro Vc0323
 vc032x		0ac8:0328	A4Tech PK-130MG
-vc032x		0ac8:c001	Sony embedded vimicro
-vc032x		0ac8:c002	Sony embedded vimicro
-vc032x		17ef:4802	Lenovo Vc0323+MI1310_SOC
-zc3xx		041e:041e	Creative WebCam Live!
-zc3xx		041e:4017	Creative Webcam Mobile PD1090
-zc3xx		041e:401c	Creative NX
-zc3xx		041e:401e	Creative Nx Pro
-zc3xx		041e:401f	Creative Webcam Notebook PD1171
-zc3xx		041e:4029	Creative WebCam Vista Pro
-zc3xx		041e:4034	Creative Instant P0620
-zc3xx		041e:4035	Creative Instant P0620D
-zc3xx		041e:4036	Creative Live !
-zc3xx		041e:403a	Creative Nx Pro 2
-zc3xx		041e:4051	Creative Live!Cam Notebook Pro (VF0250)
-zc3xx		041e:4053	Creative Live!Cam Video IM
-zc3xx		0458:7007	Genius VideoCam V2
-zc3xx		0458:700c	Genius VideoCam V3
-zc3xx		0458:700f	Genius VideoCam Web V2
-zc3xx		0461:0a00	MicroInnovation WebCam320
-zc3xx		046d:08a0	Logitech QC IM
-zc3xx		046d:08a1	Logitech QC IM 0x08A1 +sound
-zc3xx		046d:08a2	Labtec Webcam Pro
-zc3xx		046d:08a3	Logitech QC Chat
-zc3xx		046d:08a6	Logitech QCim
-zc3xx		046d:08a7	Logitech QuickCam Image
-zc3xx		046d:08a9	Logitech Notebook Deluxe
-zc3xx		046d:08aa	Labtec Webcam  Notebook
-zc3xx		046d:08ac	Logitech QuickCam Cool
-zc3xx		046d:08ad	Logitech QCCommunicate STX
-zc3xx		046d:08ae	Logitech QuickCam for Notebooks
-zc3xx		046d:08af	Logitech QuickCam Cool
-zc3xx		046d:08b9	Logitech QC IM ???
-zc3xx		046d:08d7	Logitech QCam STX
-zc3xx		046d:08d9	Logitech QuickCam IM/Connect
-zc3xx		046d:08d8	Logitech Notebook Deluxe
-zc3xx		046d:08da	Logitech QuickCam Messenger
-zc3xx		046d:08dd	Logitech QuickCam for Notebooks
-zc3xx		0471:0325	Philips SPC 200 NC
-zc3xx		0471:0326	Philips SPC 300 NC
-zc3xx		0471:032d	Philips spc210nc
-zc3xx		0471:032e	Philips spc315nc
-zc3xx		055f:c005	Mustek Wcam300A
-zc3xx		055f:d003	Mustek WCam300A
-zc3xx		055f:d004	Mustek WCam300 AN
-zc3xx		0698:2003	CTX M730V built in
-zc3xx		0ac8:0302	Z-star Vimicro zc0302
 zc3xx		0ac8:301b	Z-Star zc301b
 zc3xx		0ac8:303b	Vimicro 0x303b
 zc3xx		0ac8:305b	Z-star Vimicro zc0305b
 zc3xx		0ac8:307b	Ldlc VC302+Ov7620
+vc032x		0ac8:c001	Sony embedded vimicro
+vc032x		0ac8:c002	Sony embedded vimicro
+spca508		0af9:0010	Hama USB Sightcam 100
+spca508		0af9:0011	Hama USB Sightcam 100
+sonixb		0c45:6001	Genius VideoCAM NB
+sonixb		0c45:6005	Sweex Tas5110
+sonixb		0c45:6007	Sonix sn9c101 + Tas5110D
+sonixb		0c45:6009	spcaCam@120
+sonixb		0c45:600d	spcaCam@120
+sonixb		0c45:6011	MAX Webcam (Microdia - OV6650 - SN9C101G)
+sonixb		0c45:6019	Generic Sonix OV7630
+sonixb		0c45:6024	Generic Sonix Tas5130c
+sonixb		0c45:6025	Xcam Shanga
+sonixb		0c45:6028	Sonix Btc Pc380
+sonixb		0c45:6029	spcaCam@150
+sonixb		0c45:602c	Generic Sonix OV7630
+sonixb		0c45:602d	LIC-200 LG
+sonixb		0c45:602e	Genius VideoCam Messenger
+sonixj		0c45:6040	Speed NVC 350K
+sonixj		0c45:607c	Sonix sn9c102p Hv7131R
+sonixj		0c45:60c0	Sangha Sn535
+sonixj		0c45:60ec	SN9C105+MO4000
+sonixj		0c45:60fb	Surfer NoName
+sonixj		0c45:60fc	LG-LIC300
+sonixj		0c45:612a	Avant Camera
+sonixj		0c45:612c	Typhoon Rasy Cam 1.3MPix
+sonixj		0c45:6130	Sonix Pccam
+sonixj		0c45:6138	Sn9c120 Mo4000
+sonixj		0c45:613b	Surfer SN-206
+sonixj		0c45:613c	Sonix Pccam168
+sunplus		0d64:0303	Sunplus FashionCam DXG
+etoms		102c:6151	Qcam Sangha CIF
+etoms		102c:6251	Qcam xxxxxx VGA
 zc3xx		10fd:0128	Typhoon Webshot II USB 300k 0x0128
+spca561		10fd:7e50	FlyCam Usb 100
 zc3xx		10fd:8050	Typhoon Webshot II USB 300k
+spca501		1776:501c	Arowana 300K CMOS Camera
+t613		17a1:0128	T613/TAS5130A
+vc032x		17ef:4802	Lenovo Vc0323+MI1310_SOC
+pac207		2001:f115	D-Link DSB-C120
+spca500		2899:012c	Toptro Industrial
+spca508		8086:0110	Intel Easy PC Camera
+spca500		8086:0630	Intel Pocket PC Camera
+spca506		99fa:8988	Grandtec V.cap
+spca561		abcd:cdee	Petcam
diff --git a/drivers/media/video/Makefile b/drivers/media/video/Makefile
index 2ec920dc32e0..d5f6eea7d2fd 100644
--- a/drivers/media/video/Makefile
+++ b/drivers/media/video/Makefile
@@ -117,6 +117,7 @@ obj-$(CONFIG_USB_SN9C102)       += sn9c102/
 obj-$(CONFIG_USB_ET61X251)      += et61x251/
 obj-$(CONFIG_USB_PWC)           += pwc/
 obj-$(CONFIG_USB_ZC0301)        += zc0301/
+obj-$(CONFIG_USB_GSPCA)         += gspca/
 
 obj-$(CONFIG_USB_IBMCAM)        += usbvideo/
 obj-$(CONFIG_USB_KONICAWC)      += usbvideo/
diff --git a/drivers/media/video/gspca/Kconfig b/drivers/media/video/gspca/Kconfig
index a04e413e1258..42b90742b40b 100644
--- a/drivers/media/video/gspca/Kconfig
+++ b/drivers/media/video/gspca/Kconfig
@@ -2,7 +2,7 @@ config USB_GSPCA
 	tristate "USB GSPCA driver"
 	depends on VIDEO_V4L2
 	---help---
-	  Say Y here if you want support for various USB cameras.
+	  Say Y here if you want support for various USB webcams.
 
 	  See <file:Documentation/video4linux/gspca.txt> for more info.
 
diff --git a/drivers/media/video/gspca/Makefile b/drivers/media/video/gspca/Makefile
index d959f7771526..e68a8965297a 100644
--- a/drivers/media/video/gspca/Makefile
+++ b/drivers/media/video/gspca/Makefile
@@ -1,7 +1,29 @@
-obj-$(CONFIG_GSPCA)	+= gspca_main.o \
-	gspca_pac207.o gspca_stk014.o gspca_zc3xx.o
+obj-$(CONFIG_USB_GSPCA)	+= gspca_main.o \
+	gspca_conex.o gspca_etoms.o gspca_mars.o \
+	gspca_ov519.o gspca_pac207.o gspca_pac7311.o \
+	gspca_sonixb.o gspca_sonixj.o gspca_spca500.o gspca_spca501.o \
+	gspca_spca505.o gspca_spca506.o gspca_spca508.o gspca_spca561.o \
+	gspca_sunplus.o gspca_stk014.o gspca_t613.o gspca_tv8532.o \
+	gspca_vc032x.o gspca_zc3xx.o
 
 gspca_main-objs := gspca.o
+gspca_conex-objs := conex.o
+gspca_etoms-objs := etoms.o
+gspca_mars-objs := mars.o
+gspca_ov519-objs := ov519.o
 gspca_pac207-objs := pac207.o
+gspca_pac7311-objs := pac7311.o
+gspca_sonixb-objs := sonixb.o
+gspca_sonixj-objs := sonixj.o
+gspca_spca500-objs := spca500.o
+gspca_spca501-objs := spca501.o
+gspca_spca505-objs := spca505.o
+gspca_spca506-objs := spca506.o
+gspca_spca508-objs := spca508.o
+gspca_spca561-objs := spca561.o
 gspca_stk014-objs := stk014.o
+gspca_sunplus-objs := sunplus.o
+gspca_t613-objs := t613.o
+gspca_tv8532-objs := tv8532.o
+gspca_vc032x-objs := vc032x.o
 gspca_zc3xx-objs := zc3xx.o
diff --git a/drivers/media/video/gspca/conex.c b/drivers/media/video/gspca/conex.c
new file mode 100644
index 000000000000..b0294c9274e3
--- /dev/null
+++ b/drivers/media/video/gspca/conex.c
@@ -0,0 +1,1059 @@
+/*
+ *		Connexant Cx11646 library
+ *		Copyright (C) 2004 Michel Xhaard mxhaard@magic.fr
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "conex"
+
+#include "gspca.h"
+#define CONEX_CAM 1		/* special JPEG header */
+#include "jpeg.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA USB Conexant Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned char brightness;
+	unsigned char contrast;
+	unsigned char colors;
+
+	unsigned char qindex;
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id	 = V4L2_CID_BRIGHTNESS,
+		.type	 = V4L2_CTRL_TYPE_INTEGER,
+		.name	 = "Brightness",
+		.minimum = 0,
+		.maximum = 255,
+		.step	 = 1,
+		.default_value = 0xd4,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0x0a,
+		.maximum = 0x1f,
+		.step    = 1,
+		.default_value = 0x0c,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define SD_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Color",
+		.minimum = 0,
+		.maximum = 7,
+		.step    = 1,
+		.default_value = 3,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 176, 144, 3},
+	{V4L2_PIX_FMT_JPEG, 320, 240, 2},
+	{V4L2_PIX_FMT_JPEG, 352, 288, 1},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 0},
+};
+
+static void reg_r(struct usb_device *dev,
+			   __u16 index,
+			   __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,
+			index, buffer, length,
+			500);
+	PDEBUG(D_USBI, "reg read i:%02x -> %02x", index, *buffer);
+}
+
+static void reg_w(struct usb_device *dev,
+		  __u16 index,
+		  const __u8 *buffer, __u16 length)
+{
+	PDEBUG(D_USBO, "reg write i:%02x = %02x", index, *buffer);
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			0,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,
+			index, (__u8 *) buffer, length,
+			500);
+}
+
+static const __u8 cx_sensor_init[][4] = {
+	{0x88, 0x11, 0x01, 0x01},
+	{0x88, 0x12, 0x70, 0x01},
+	{0x88, 0x0f, 0x00, 0x01},
+	{0x88, 0x05, 0x01, 0x01},
+	{}
+};
+
+static const __u8 cx11646_fw1[][3] = {
+	{0x00, 0x02, 0x00},
+	{0x01, 0x43, 0x00},
+	{0x02, 0xA7, 0x00},
+	{0x03, 0x8B, 0x01},
+	{0x04, 0xE9, 0x02},
+	{0x05, 0x08, 0x04},
+	{0x06, 0x08, 0x05},
+	{0x07, 0x07, 0x06},
+	{0x08, 0xE7, 0x06},
+	{0x09, 0xC6, 0x07},
+	{0x0A, 0x86, 0x08},
+	{0x0B, 0x46, 0x09},
+	{0x0C, 0x05, 0x0A},
+	{0x0D, 0xA5, 0x0A},
+	{0x0E, 0x45, 0x0B},
+	{0x0F, 0xE5, 0x0B},
+	{0x10, 0x85, 0x0C},
+	{0x11, 0x25, 0x0D},
+	{0x12, 0xC4, 0x0D},
+	{0x13, 0x45, 0x0E},
+	{0x14, 0xE4, 0x0E},
+	{0x15, 0x64, 0x0F},
+	{0x16, 0xE4, 0x0F},
+	{0x17, 0x64, 0x10},
+	{0x18, 0xE4, 0x10},
+	{0x19, 0x64, 0x11},
+	{0x1A, 0xE4, 0x11},
+	{0x1B, 0x64, 0x12},
+	{0x1C, 0xE3, 0x12},
+	{0x1D, 0x44, 0x13},
+	{0x1E, 0xC3, 0x13},
+	{0x1F, 0x24, 0x14},
+	{0x20, 0xA3, 0x14},
+	{0x21, 0x04, 0x15},
+	{0x22, 0x83, 0x15},
+	{0x23, 0xE3, 0x15},
+	{0x24, 0x43, 0x16},
+	{0x25, 0xA4, 0x16},
+	{0x26, 0x23, 0x17},
+	{0x27, 0x83, 0x17},
+	{0x28, 0xE3, 0x17},
+	{0x29, 0x43, 0x18},
+	{0x2A, 0xA3, 0x18},
+	{0x2B, 0x03, 0x19},
+	{0x2C, 0x63, 0x19},
+	{0x2D, 0xC3, 0x19},
+	{0x2E, 0x22, 0x1A},
+	{0x2F, 0x63, 0x1A},
+	{0x30, 0xC3, 0x1A},
+	{0x31, 0x23, 0x1B},
+	{0x32, 0x83, 0x1B},
+	{0x33, 0xE2, 0x1B},
+	{0x34, 0x23, 0x1C},
+	{0x35, 0x83, 0x1C},
+	{0x36, 0xE2, 0x1C},
+	{0x37, 0x23, 0x1D},
+	{0x38, 0x83, 0x1D},
+	{0x39, 0xE2, 0x1D},
+	{0x3A, 0x23, 0x1E},
+	{0x3B, 0x82, 0x1E},
+	{0x3C, 0xC3, 0x1E},
+	{0x3D, 0x22, 0x1F},
+	{0x3E, 0x63, 0x1F},
+	{0x3F, 0xC1, 0x1F},
+	{}
+};
+static void cx11646_fw(struct gspca_dev*gspca_dev)
+{
+	__u8 val;
+	int i = 0;
+
+	val = 0x02;
+	reg_w(gspca_dev->dev, 0x006a, &val, 1);
+	while (cx11646_fw1[i][1]) {
+		reg_w(gspca_dev->dev, 0x006b, cx11646_fw1[i], 3);
+		i++;
+	}
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x006a, &val, 1);
+}
+
+static __u8 cxsensor[] = {
+	0x88, 0x12, 0x70, 0x01,
+	0x88, 0x0d, 0x02, 0x01,
+	0x88, 0x0f, 0x00, 0x01,
+	0x88, 0x03, 0x71, 0x01, 0x88, 0x04, 0x00, 0x01,	/* 3 */
+	0x88, 0x02, 0x10, 0x01,
+	0x88, 0x00, 0xD4, 0x01, 0x88, 0x01, 0x01, 0x01,	/* 5 */
+	0x88, 0x0B, 0x00, 0x01,
+	0x88, 0x0A, 0x0A, 0x01,
+	0x88, 0x00, 0x08, 0x01, 0x88, 0x01, 0x00, 0x01,	/* 8 */
+	0x88, 0x05, 0x01, 0x01,
+	0xA1, 0x18, 0x00, 0x01,
+	0x00
+};
+
+static __u8 reg20[] = { 0x10, 0x42, 0x81, 0x19, 0xd3, 0xff, 0xa7, 0xff };
+static __u8 reg28[] = { 0x87, 0x00, 0x87, 0x00, 0x8f, 0xff, 0xea, 0xff };
+static __u8 reg10[] = { 0xb1, 0xb1 };
+static __u8 reg71a[] = { 0x08, 0x18, 0x0a, 0x1e };	/* 640 */
+static __u8 reg71b[] = { 0x04, 0x0c, 0x05, 0x0f };
+	/* 352{0x04,0x0a,0x06,0x12}; //352{0x05,0x0e,0x06,0x11}; //352 */
+static __u8 reg71c[] = { 0x02, 0x07, 0x03, 0x09 };
+					/* 320{0x04,0x0c,0x05,0x0f}; //320 */
+static __u8 reg71d[] = { 0x02, 0x07, 0x03, 0x09 };	/* 176 */
+static __u8 reg7b[] = { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff };
+
+static void cx_sensor(struct gspca_dev*gspca_dev)
+{
+	__u8 val = 0;
+	int i = 0;
+	__u8 bufread[] = { 0, 0, 0, 0, 0, 0, 0, 0 };
+	int length = 0;
+	__u8 *ptsensor = cxsensor;
+
+	reg_w(gspca_dev->dev, 0x0020, reg20, 8);
+	reg_w(gspca_dev->dev, 0x0028, reg28, 8);
+	reg_w(gspca_dev->dev, 0x0010, reg10, 8);
+	val = 0x03;
+	reg_w(gspca_dev->dev, 0x0092, &val, 1);
+
+	switch (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+	case 0:
+		reg_w(gspca_dev->dev, 0x0071, reg71a, 4);
+		break;
+	case 1:
+		reg_w(gspca_dev->dev, 0x0071, reg71b, 4);
+		break;
+	default:
+/* case 2: */
+		reg_w(gspca_dev->dev, 0x0071, reg71c, 4);
+		break;
+	case 3:
+		reg_w(gspca_dev->dev, 0x0071, reg71d, 4);
+		break;
+	}
+	reg_w(gspca_dev->dev, 0x007b, reg7b, 6);
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x00f8, &val, 1);
+	reg_w(gspca_dev->dev, 0x0010, reg10, 8);
+	val = 0x41;
+	reg_w(gspca_dev->dev, 0x0098, &val, 1);
+	for (i = 0; i < 11; i++) {
+		if (i == 3 || i == 5 || i == 8)
+			length = 8;
+		else
+			length = 4;
+		reg_w(gspca_dev->dev, 0x00e5, ptsensor, length);
+		if (length == 4)
+			reg_r(gspca_dev->dev, 0x00e8, &val, 1);
+		else
+			reg_r(gspca_dev->dev, 0x00e8, bufread, length);
+		ptsensor += length;
+	}
+	reg_r(gspca_dev->dev, 0x00e7, bufread, 8);
+}
+
+static __u8 cx_inits_176[] = {
+	0x33, 0x81, 0xB0, 0x00, 0x90, 0x00, 0x0A, 0x03,	/* 176x144 */
+	0x00, 0x03, 0x03, 0x03, 0x1B, 0x05, 0x30, 0x03,
+	0x65, 0x15, 0x18, 0x25, 0x03, 0x25, 0x08, 0x30,
+	0x3B, 0x25, 0x10, 0x00, 0x04, 0x00, 0x00, 0x00,
+	0xDC, 0xFF, 0xEE, 0xFF, 0xC5, 0xFF, 0xBF, 0xFF,
+	0xF7, 0xFF, 0x88, 0xFF, 0x66, 0x02, 0x28, 0x02,
+	0x1E, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+static __u8 cx_inits_320[] = {
+	0x7f, 0x7f, 0x40, 0x01, 0xf0, 0x00, 0x02, 0x01,
+	0x00, 0x01, 0x01, 0x01, 0x10, 0x00, 0x02, 0x01,
+	0x65, 0x45, 0xfa, 0x4c, 0x2c, 0xdf, 0xb9, 0x81,
+	0x30, 0x00, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00,
+	0xe2, 0xff, 0xf1, 0xff, 0xc2, 0xff, 0xbc, 0xff,
+	0xf5, 0xff, 0x6d, 0xff, 0xf6, 0x01, 0x43, 0x02,
+	0xd3, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+static __u8 cx_inits_352[] = {
+	0x2e, 0x7c, 0x60, 0x01, 0x20, 0x01, 0x05, 0x03,
+	0x00, 0x06, 0x03, 0x06, 0x1b, 0x10, 0x05, 0x3b,
+	0x30, 0x25, 0x18, 0x25, 0x08, 0x30, 0x03, 0x25,
+	0x3b, 0x30, 0x25, 0x1b, 0x10, 0x05, 0x00, 0x00,
+	0xe3, 0xff, 0xf1, 0xff, 0xc2, 0xff, 0xbc, 0xff,
+	0xf5, 0xff, 0x6b, 0xff, 0xee, 0x01, 0x43, 0x02,
+	0xe4, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+static __u8 cx_inits_640[] = {
+	0x7e, 0x7e, 0x80, 0x02, 0xe0, 0x01, 0x01, 0x01,
+	0x00, 0x02, 0x01, 0x02, 0x10, 0x30, 0x01, 0x01,
+	0x65, 0x45, 0xf7, 0x52, 0x2c, 0xdf, 0xb9, 0x81,
+	0x30, 0x00, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00,
+	0xe2, 0xff, 0xf1, 0xff, 0xc2, 0xff, 0xbc, 0xff,
+	0xf6, 0xff, 0x7b, 0xff, 0x01, 0x02, 0x43, 0x02,
+	0x77, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+
+static int cx11646_initsize(struct gspca_dev *gspca_dev)
+{
+	__u8 *cxinit;
+	__u8 val;
+	static const __u8 reg12[] = { 0x08, 0x05, 0x07, 0x04, 0x24 };
+	static const __u8 reg17[] =
+			{ 0x0a, 0x00, 0xf2, 0x01, 0x0f, 0x00, 0x97, 0x02 };
+
+	switch (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+	case 0:
+		cxinit = cx_inits_640;
+		break;
+	case 1:
+		cxinit = cx_inits_352;
+		break;
+	default:
+/*	case 2: */
+		cxinit = cx_inits_320;
+		break;
+	case 3:
+		cxinit = cx_inits_176;
+		break;
+	}
+	val = 0x01;
+	reg_w(gspca_dev->dev, 0x009a, &val, 1);
+	val = 0x10;
+	reg_w(gspca_dev->dev, 0x0010, &val, 1);
+	reg_w(gspca_dev->dev, 0x0012, reg12, 5);
+	reg_w(gspca_dev->dev, 0x0017, reg17, 8);
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x00c0, &val, 1);
+	val = 0x04;
+	reg_w(gspca_dev->dev, 0x00c1, &val, 1);
+	val = 0x04;
+	reg_w(gspca_dev->dev, 0x00c2, &val, 1);
+
+	reg_w(gspca_dev->dev, 0x0061, cxinit, 8);
+	cxinit += 8;
+	reg_w(gspca_dev->dev, 0x00ca, cxinit, 8);
+	cxinit += 8;
+	reg_w(gspca_dev->dev, 0x00d2, cxinit, 8);
+	cxinit += 8;
+	reg_w(gspca_dev->dev, 0x00da, cxinit, 6);
+	cxinit += 8;
+	reg_w(gspca_dev->dev, 0x0041, cxinit, 8);
+	cxinit += 8;
+	reg_w(gspca_dev->dev, 0x0049, cxinit, 8);
+	cxinit += 8;
+	reg_w(gspca_dev->dev, 0x0051, cxinit, 2);
+
+	reg_r(gspca_dev->dev, 0x0010, &val, 1);
+	return val;
+}
+
+static __u8 cx_jpeg_init[][8] = {
+	{0xff, 0xd8, 0xff, 0xdb, 0x00, 0x84, 0x00, 0x15},	/* 1 */
+	{0x0f, 0x10, 0x12, 0x10, 0x0d, 0x15, 0x12, 0x11},
+	{0x12, 0x18, 0x16, 0x15, 0x19, 0x20, 0x35, 0x22},
+	{0x20, 0x1d, 0x1d, 0x20, 0x41, 0x2e, 0x31, 0x26},
+	{0x35, 0x4d, 0x43, 0x51, 0x4f, 0x4b, 0x43, 0x4a},
+	{0x49, 0x55, 0x5F, 0x79, 0x67, 0x55, 0x5A, 0x73},
+	{0x5B, 0x49, 0x4A, 0x6A, 0x90, 0x6B, 0x73, 0x7D},
+	{0x81, 0x88, 0x89, 0x88, 0x52, 0x66, 0x95, 0xA0},
+	{0x94, 0x84, 0x9E, 0x79, 0x85, 0x88, 0x83, 0x01},
+	{0x15, 0x0F, 0x10, 0x12, 0x10, 0x0D, 0x15, 0x12},
+	{0x11, 0x12, 0x18, 0x16, 0x15, 0x19, 0x20, 0x35},
+	{0x22, 0x20, 0x1D, 0x1D, 0x20, 0x41, 0x2E, 0x31},
+	{0x26, 0x35, 0x4D, 0x43, 0x51, 0x4F, 0x4B, 0x43},
+	{0x4A, 0x49, 0x55, 0x5F, 0x79, 0x67, 0x55, 0x5A},
+	{0x73, 0x5B, 0x49, 0x4A, 0x6A, 0x90, 0x6B, 0x73},
+	{0x7D, 0x81, 0x88, 0x89, 0x88, 0x52, 0x66, 0x95},
+	{0xA0, 0x94, 0x84, 0x9E, 0x79, 0x85, 0x88, 0x83},
+	{0xFF, 0xC4, 0x01, 0xA2, 0x00, 0x00, 0x01, 0x05},
+	{0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x00, 0x00},
+	{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x02},
+	{0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0A},
+	{0x0B, 0x01, 0x00, 0x03, 0x01, 0x01, 0x01, 0x01},
+	{0x01, 0x01, 0x01, 0x01, 0x01, 0x00, 0x00, 0x00},
+	{0x00, 0x00, 0x00, 0x01, 0x02, 0x03, 0x04, 0x05},
+	{0x06, 0x07, 0x08, 0x09, 0x0A, 0x0B, 0x10, 0x00},
+	{0x02, 0x01, 0x03, 0x03, 0x02, 0x04, 0x03, 0x05},
+	{0x05, 0x04, 0x04, 0x00, 0x00, 0x01, 0x7D, 0x01},
+	{0x02, 0x03, 0x00, 0x04, 0x11, 0x05, 0x12, 0x21},
+	{0x31, 0x41, 0x06, 0x13, 0x51, 0x61, 0x07, 0x22},
+	{0x71, 0x14, 0x32, 0x81, 0x91, 0xA1, 0x08, 0x23},
+	{0x42, 0xB1, 0xC1, 0x15, 0x52, 0xD1, 0xF0, 0x24},
+	{0x33, 0x62, 0x72, 0x82, 0x09, 0x0A, 0x16, 0x17},
+	{0x18, 0x19, 0x1A, 0x25, 0x26, 0x27, 0x28, 0x29},
+	{0x2A, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3A},
+	{0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4A},
+	{0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A},
+	{0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A},
+	{0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A},
+	{0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8A},
+	{0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, 0x99},
+	{0x9A, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8},
+	{0xA9, 0xAA, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7},
+	{0xB8, 0xB9, 0xBA, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6},
+	{0xC7, 0xC8, 0xC9, 0xCA, 0xD2, 0xD3, 0xD4, 0xD5},
+	{0xD6, 0xD7, 0xD8, 0xD9, 0xDA, 0xE1, 0xE2, 0xE3},
+	{0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xEA, 0xF1},
+	{0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9},
+	{0xFA, 0x11, 0x00, 0x02, 0x01, 0x02, 0x04, 0x04},
+	{0x03, 0x04, 0x07, 0x05, 0x04, 0x04, 0x00, 0x01},
+	{0x02, 0x77, 0x00, 0x01, 0x02, 0x03, 0x11, 0x04},
+	{0x05, 0x21, 0x31, 0x06, 0x12, 0x41, 0x51, 0x07},
+	{0x61, 0x71, 0x13, 0x22, 0x32, 0x81, 0x08, 0x14},
+	{0x42, 0x91, 0xA1, 0xB1, 0xC1, 0x09, 0x23, 0x33},
+	{0x52, 0xF0, 0x15, 0x62, 0x72, 0xD1, 0x0A, 0x16},
+	{0x24, 0x34, 0xE1, 0x25, 0xF1, 0x17, 0x18, 0x19},
+	{0x1A, 0x26, 0x27, 0x28, 0x29, 0x2A, 0x35, 0x36},
+	{0x37, 0x38, 0x39, 0x3A, 0x43, 0x44, 0x45, 0x46},
+	{0x47, 0x48, 0x49, 0x4A, 0x53, 0x54, 0x55, 0x56},
+	{0x57, 0x58, 0x59, 0x5A, 0x63, 0x64, 0x65, 0x66},
+	{0x67, 0x68, 0x69, 0x6A, 0x73, 0x74, 0x75, 0x76},
+	{0x77, 0x78, 0x79, 0x7A, 0x82, 0x83, 0x84, 0x85},
+	{0x86, 0x87, 0x88, 0x89, 0x8A, 0x92, 0x93, 0x94},
+	{0x95, 0x96, 0x97, 0x98, 0x99, 0x9A, 0xA2, 0xA3},
+	{0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xAA, 0xB2},
+	{0xB3, 0xB4, 0xB5, 0xB6, 0xB7, 0xB8, 0xB9, 0xBA},
+	{0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9},
+	{0xCA, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7, 0xD8},
+	{0xD9, 0xDA, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7},
+	{0xE8, 0xE9, 0xEA, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6},
+	{0xF7, 0xF8, 0xF9, 0xFA, 0xFF, 0x20, 0x00, 0x1F},
+	{0x02, 0x0C, 0x00, 0x00, 0x0A, 0x00, 0x00, 0x00},
+	{0x00, 0x00, 0x11, 0x00, 0x11, 0x22, 0x00, 0x22},
+	{0x22, 0x11, 0x22, 0x22, 0x11, 0x33, 0x33, 0x11},
+	{0x44, 0x66, 0x22, 0x55, 0x66, 0xFF, 0xDD, 0x00},
+	{0x04, 0x00, 0x14, 0xFF, 0xC0, 0x00, 0x11, 0x08},
+	{0x00, 0xF0, 0x01, 0x40, 0x03, 0x00, 0x21, 0x00},
+	{0x01, 0x11, 0x01, 0x02, 0x11, 0x01, 0xFF, 0xDA},
+	{0x00, 0x0C, 0x03, 0x00, 0x00, 0x01, 0x11, 0x02},
+	{0x11, 0x00, 0x3F, 0x00, 0xFF, 0xD9, 0x00, 0x00}	/* 79 */
+};
+
+
+static __u8 cxjpeg_640[][8] = {
+	{0xff, 0xd8, 0xff, 0xdb, 0x00, 0x84, 0x00, 0x10},	/* 1 */
+	{0x0b, 0x0c, 0x0e, 0x0c, 0x0a, 0x10, 0x0e, 0x0d},
+	{0x0e, 0x12, 0x11, 0x10, 0x13, 0x18, 0x28, 0x1a},
+	{0x18, 0x16, 0x16, 0x18, 0x31, 0x23, 0x25, 0x1d},
+	{0x28, 0x3a, 0x33, 0x3D, 0x3C, 0x39, 0x33, 0x38},
+	{0x37, 0x40, 0x48, 0x5C, 0x4E, 0x40, 0x44, 0x57},
+	{0x45, 0x37, 0x38, 0x50, 0x6D, 0x51, 0x57, 0x5F},
+	{0x62, 0x67, 0x68, 0x67, 0x3E, 0x4D, 0x71, 0x79},
+	{0x70, 0x64, 0x78, 0x5C, 0x65, 0x67, 0x63, 0x01},
+	{0x10, 0x0B, 0x0C, 0x0E, 0x0C, 0x0A, 0x10, 0x0E},
+	{0x0D, 0x0E, 0x12, 0x11, 0x10, 0x13, 0x18, 0x28},
+	{0x1A, 0x18, 0x16, 0x16, 0x18, 0x31, 0x23, 0x25},
+	{0x1D, 0x28, 0x3A, 0x33, 0x3D, 0x3C, 0x39, 0x33},
+	{0x38, 0x37, 0x40, 0x48, 0x5C, 0x4E, 0x40, 0x44},
+	{0x57, 0x45, 0x37, 0x38, 0x50, 0x6D, 0x51, 0x57},
+	{0x5F, 0x62, 0x67, 0x68, 0x67, 0x3E, 0x4D, 0x71},
+	{0x79, 0x70, 0x64, 0x78, 0x5C, 0x65, 0x67, 0x63},
+	{0xFF, 0x20, 0x00, 0x1F, 0x00, 0x83, 0x00, 0x00},
+	{0x0A, 0x00, 0x00, 0x00, 0x00, 0x00, 0x11, 0x00},
+	{0x11, 0x22, 0x00, 0x22, 0x22, 0x11, 0x22, 0x22},
+	{0x11, 0x33, 0x33, 0x11, 0x44, 0x66, 0x22, 0x55},
+	{0x66, 0xFF, 0xDD, 0x00, 0x04, 0x00, 0x28, 0xFF},
+	{0xC0, 0x00, 0x11, 0x08, 0x01, 0xE0, 0x02, 0x80},
+	{0x03, 0x00, 0x21, 0x00, 0x01, 0x11, 0x01, 0x02},
+	{0x11, 0x01, 0xFF, 0xDA, 0x00, 0x0C, 0x03, 0x00},
+	{0x00, 0x01, 0x11, 0x02, 0x11, 0x00, 0x3F, 0x00},
+	{0xFF, 0xD9, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}	/* 27 */
+};
+static __u8 cxjpeg_352[][8] = {
+	{0xff, 0xd8, 0xff, 0xdb, 0x00, 0x84, 0x00, 0x0d},
+	{0x09, 0x09, 0x0b, 0x09, 0x08, 0x0D, 0x0b, 0x0a},
+	{0x0b, 0x0e, 0x0d, 0x0d, 0x0f, 0x13, 0x1f, 0x14},
+	{0x13, 0x11, 0x11, 0x13, 0x26, 0x1b, 0x1d, 0x17},
+	{0x1F, 0x2D, 0x28, 0x30, 0x2F, 0x2D, 0x28, 0x2C},
+	{0x2B, 0x32, 0x38, 0x48, 0x3D, 0x32, 0x35, 0x44},
+	{0x36, 0x2B, 0x2C, 0x3F, 0x55, 0x3F, 0x44, 0x4A},
+	{0x4D, 0x50, 0x51, 0x50, 0x30, 0x3C, 0x58, 0x5F},
+	{0x58, 0x4E, 0x5E, 0x48, 0x4F, 0x50, 0x4D, 0x01},
+	{0x0D, 0x09, 0x09, 0x0B, 0x09, 0x08, 0x0D, 0x0B},
+	{0x0A, 0x0B, 0x0E, 0x0D, 0x0D, 0x0F, 0x13, 0x1F},
+	{0x14, 0x13, 0x11, 0x11, 0x13, 0x26, 0x1B, 0x1D},
+	{0x17, 0x1F, 0x2D, 0x28, 0x30, 0x2F, 0x2D, 0x28},
+	{0x2C, 0x2B, 0x32, 0x38, 0x48, 0x3D, 0x32, 0x35},
+	{0x44, 0x36, 0x2B, 0x2C, 0x3F, 0x55, 0x3F, 0x44},
+	{0x4A, 0x4D, 0x50, 0x51, 0x50, 0x30, 0x3C, 0x58},
+	{0x5F, 0x58, 0x4E, 0x5E, 0x48, 0x4F, 0x50, 0x4D},
+	{0xFF, 0x20, 0x00, 0x1F, 0x01, 0x83, 0x00, 0x00},
+	{0x0A, 0x00, 0x00, 0x00, 0x00, 0x00, 0x11, 0x00},
+	{0x11, 0x22, 0x00, 0x22, 0x22, 0x11, 0x22, 0x22},
+	{0x11, 0x33, 0x33, 0x11, 0x44, 0x66, 0x22, 0x55},
+	{0x66, 0xFF, 0xDD, 0x00, 0x04, 0x00, 0x16, 0xFF},
+	{0xC0, 0x00, 0x11, 0x08, 0x01, 0x20, 0x01, 0x60},
+	{0x03, 0x00, 0x21, 0x00, 0x01, 0x11, 0x01, 0x02},
+	{0x11, 0x01, 0xFF, 0xDA, 0x00, 0x0C, 0x03, 0x00},
+	{0x00, 0x01, 0x11, 0x02, 0x11, 0x00, 0x3F, 0x00},
+	{0xFF, 0xD9, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}
+};
+static __u8 cxjpeg_320[][8] = {
+	{0xff, 0xd8, 0xff, 0xdb, 0x00, 0x84, 0x00, 0x05},
+	{0x03, 0x04, 0x04, 0x04, 0x03, 0x05, 0x04, 0x04},
+	{0x04, 0x05, 0x05, 0x05, 0x06, 0x07, 0x0c, 0x08},
+	{0x07, 0x07, 0x07, 0x07, 0x0f, 0x0b, 0x0b, 0x09},
+	{0x0C, 0x11, 0x0F, 0x12, 0x12, 0x11, 0x0f, 0x11},
+	{0x11, 0x13, 0x16, 0x1C, 0x17, 0x13, 0x14, 0x1A},
+	{0x15, 0x11, 0x11, 0x18, 0x21, 0x18, 0x1A, 0x1D},
+	{0x1D, 0x1F, 0x1F, 0x1F, 0x13, 0x17, 0x22, 0x24},
+	{0x22, 0x1E, 0x24, 0x1C, 0x1E, 0x1F, 0x1E, 0x01},
+	{0x05, 0x03, 0x04, 0x04, 0x04, 0x03, 0x05, 0x04},
+	{0x04, 0x04, 0x05, 0x05, 0x05, 0x06, 0x07, 0x0C},
+	{0x08, 0x07, 0x07, 0x07, 0x07, 0x0F, 0x0B, 0x0B},
+	{0x09, 0x0C, 0x11, 0x0F, 0x12, 0x12, 0x11, 0x0F},
+	{0x11, 0x11, 0x13, 0x16, 0x1C, 0x17, 0x13, 0x14},
+	{0x1A, 0x15, 0x11, 0x11, 0x18, 0x21, 0x18, 0x1A},
+	{0x1D, 0x1D, 0x1F, 0x1F, 0x1F, 0x13, 0x17, 0x22},
+	{0x24, 0x22, 0x1E, 0x24, 0x1C, 0x1E, 0x1F, 0x1E},
+	{0xFF, 0x20, 0x00, 0x1F, 0x02, 0x0C, 0x00, 0x00},
+	{0x0A, 0x00, 0x00, 0x00, 0x00, 0x00, 0x11, 0x00},
+	{0x11, 0x22, 0x00, 0x22, 0x22, 0x11, 0x22, 0x22},
+	{0x11, 0x33, 0x33, 0x11, 0x44, 0x66, 0x22, 0x55},
+	{0x66, 0xFF, 0xDD, 0x00, 0x04, 0x00, 0x14, 0xFF},
+	{0xC0, 0x00, 0x11, 0x08, 0x00, 0xF0, 0x01, 0x40},
+	{0x03, 0x00, 0x21, 0x00, 0x01, 0x11, 0x01, 0x02},
+	{0x11, 0x01, 0xFF, 0xDA, 0x00, 0x0C, 0x03, 0x00},
+	{0x00, 0x01, 0x11, 0x02, 0x11, 0x00, 0x3F, 0x00},
+	{0xFF, 0xD9, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}	/* 27 */
+};
+static __u8 cxjpeg_176[][8] = {
+	{0xff, 0xd8, 0xff, 0xdb, 0x00, 0x84, 0x00, 0x0d},
+	{0x09, 0x09, 0x0B, 0x09, 0x08, 0x0D, 0x0B, 0x0A},
+	{0x0B, 0x0E, 0x0D, 0x0D, 0x0F, 0x13, 0x1F, 0x14},
+	{0x13, 0x11, 0x11, 0x13, 0x26, 0x1B, 0x1D, 0x17},
+	{0x1F, 0x2D, 0x28, 0x30, 0x2F, 0x2D, 0x28, 0x2C},
+	{0x2B, 0x32, 0x38, 0x48, 0x3D, 0x32, 0x35, 0x44},
+	{0x36, 0x2B, 0x2C, 0x3F, 0x55, 0x3F, 0x44, 0x4A},
+	{0x4D, 0x50, 0x51, 0x50, 0x30, 0x3C, 0x58, 0x5F},
+	{0x58, 0x4E, 0x5E, 0x48, 0x4F, 0x50, 0x4D, 0x01},
+	{0x0D, 0x09, 0x09, 0x0B, 0x09, 0x08, 0x0D, 0x0B},
+	{0x0A, 0x0B, 0x0E, 0x0D, 0x0D, 0x0F, 0x13, 0x1F},
+	{0x14, 0x13, 0x11, 0x11, 0x13, 0x26, 0x1B, 0x1D},
+	{0x17, 0x1F, 0x2D, 0x28, 0x30, 0x2F, 0x2D, 0x28},
+	{0x2C, 0x2B, 0x32, 0x38, 0x48, 0x3D, 0x32, 0x35},
+	{0x44, 0x36, 0x2B, 0x2C, 0x3F, 0x55, 0x3F, 0x44},
+	{0x4A, 0x4D, 0x50, 0x51, 0x50, 0x30, 0x3C, 0x58},
+	{0x5F, 0x58, 0x4E, 0x5E, 0x48, 0x4F, 0x50, 0x4D},
+	{0xFF, 0x20, 0x00, 0x1F, 0x03, 0xA1, 0x00, 0x00},
+	{0x0A, 0x00, 0x00, 0x00, 0x00, 0x00, 0x11, 0x00},
+	{0x11, 0x22, 0x00, 0x22, 0x22, 0x11, 0x22, 0x22},
+	{0x11, 0x33, 0x33, 0x11, 0x44, 0x66, 0x22, 0x55},
+	{0x66, 0xFF, 0xDD, 0x00, 0x04, 0x00, 0x0B, 0xFF},
+	{0xC0, 0x00, 0x11, 0x08, 0x00, 0x90, 0x00, 0xB0},
+	{0x03, 0x00, 0x21, 0x00, 0x01, 0x11, 0x01, 0x02},
+	{0x11, 0x01, 0xFF, 0xDA, 0x00, 0x0C, 0x03, 0x00},
+	{0x00, 0x01, 0x11, 0x02, 0x11, 0x00, 0x3F, 0x00},
+	{0xFF, 0xD9, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}
+};
+static __u8 cxjpeg_qtable[][8] = {	/* 640 take with the zcx30x part */
+	{0xff, 0xd8, 0xff, 0xdb, 0x00, 0x84, 0x00, 0x08},
+	{0x06, 0x06, 0x07, 0x06, 0x05, 0x08, 0x07, 0x07},
+	{0x07, 0x09, 0x09, 0x08, 0x0a, 0x0c, 0x14, 0x0a},
+	{0x0c, 0x0b, 0x0b, 0x0c, 0x19, 0x12, 0x13, 0x0f},
+	{0x14, 0x1d, 0x1a, 0x1f, 0x1e, 0x1d, 0x1a, 0x1c},
+	{0x1c, 0x20, 0x24, 0x2e, 0x27, 0x20, 0x22, 0x2c},
+	{0x23, 0x1c, 0x1c, 0x28, 0x37, 0x29, 0x2c, 0x30},
+	{0x31, 0x34, 0x34, 0x34, 0x1f, 0x27, 0x39, 0x3d},
+	{0x38, 0x32, 0x3c, 0x2e, 0x33, 0x34, 0x32, 0x01},
+	{0x09, 0x09, 0x09, 0x0c, 0x0b, 0x0c, 0x18, 0x0a},
+	{0x0a, 0x18, 0x32, 0x21, 0x1c, 0x21, 0x32, 0x32},
+	{0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32},
+	{0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32},
+	{0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32},
+	{0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32},
+	{0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32},
+	{0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32, 0x32},
+	{0xFF, 0xD9, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}	/* 18 */
+};
+
+
+static void cx11646_jpegInit(struct gspca_dev*gspca_dev)
+{
+	__u8 val;
+	int i;
+	int length;
+
+	val = 0x01;
+	reg_w(gspca_dev->dev, 0x00c0, &val, 1);
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x00c3, &val, 1);
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x00c0, &val, 1);
+	reg_r(gspca_dev->dev, 0x0001, &val, 1);
+	length = 8;
+	for (i = 0; i < 79; i++) {
+		if (i == 78)
+			length = 6;
+		reg_w(gspca_dev->dev, 0x0008, cx_jpeg_init[i], length);
+	}
+	reg_r(gspca_dev->dev, 0x0002, &val, 1);
+	val = 0x14;
+	reg_w(gspca_dev->dev, 0x0055, &val, 1);
+}
+
+static __u8 reg12[] = { 0x0a, 0x05, 0x07, 0x04, 0x19 };
+static __u8 regE5_8[] = { 0x88, 0x00, 0xd4, 0x01, 0x88, 0x01, 0x01, 0x01 };
+static __u8 regE5a[] = { 0x88, 0x0a, 0x0c, 0x01 };
+static __u8 regE5b[] = { 0x88, 0x0b, 0x12, 0x01 };
+static __u8 regE5c[] = { 0x88, 0x05, 0x01, 0x01 };
+static __u8 reg51[] = { 0x77, 0x03 };
+static __u8 reg70 = 0x03;
+
+static void cx11646_jpeg(struct gspca_dev*gspca_dev)
+{
+	__u8 val;
+	int i;
+	int length = 8;
+	__u8 Reg55 = 0x14;
+	__u8 bufread[8] = { 0, 0, 0, 0, 0, 0, 0, 0 };
+	int retry = 50;
+
+	val = 0x01;
+	reg_w(gspca_dev->dev, 0x00c0, &val, 1);
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x00c3, &val, 1);
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x00c0, &val, 1);
+	reg_r(gspca_dev->dev, 0x0001, &val, 1);
+	switch (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+	case 0:
+		for (i = 0; i < 27; i++) {
+			if (i == 26)
+				length = 2;
+			reg_w(gspca_dev->dev, 0x0008,
+					cxjpeg_640[i], length);
+		}
+		Reg55 = 0x28;
+		break;
+	case 1:
+		for (i = 0; i < 27; i++) {
+			if (i == 26)
+				length = 2;
+			reg_w(gspca_dev->dev, 0x0008,
+					cxjpeg_352[i], length);
+		}
+		Reg55 = 0x16;
+		break;
+	default:
+/*	case 2: */
+		for (i = 0; i < 27; i++) {
+			if (i == 26)
+				length = 2;
+			reg_w(gspca_dev->dev, 0x0008,
+					cxjpeg_320[i], length);
+		}
+		Reg55 = 0x14;
+		break;
+	case 3:
+		for (i = 0; i < 27; i++) {
+			if (i == 26)
+				length = 2;
+			reg_w(gspca_dev->dev, 0x0008,
+					cxjpeg_176[i], length);
+		}
+		Reg55 = 0x0B;
+		break;
+	}
+
+	reg_r(gspca_dev->dev, 0x0002, &val, 1);
+	val = Reg55;
+	reg_w(gspca_dev->dev, 0x0055, &val, 1);
+	reg_r(gspca_dev->dev, 0x0002, &val, 1);
+	reg_w(gspca_dev->dev, 0x0010, reg10, 2);
+	val = 0x02;
+	reg_w(gspca_dev->dev, 0x0054, &val, 1);
+	val = 0x01;
+	reg_w(gspca_dev->dev, 0x0054, &val, 1);
+	val = 0x94;
+	reg_w(gspca_dev->dev, 0x0000, &val, 1);
+	val = 0xc0;
+	reg_w(gspca_dev->dev, 0x0053, &val, 1);
+	val = 0xe1;
+	reg_w(gspca_dev->dev, 0x00fc, &val, 1);
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x0000, &val, 1);
+	/* wait for completion */
+	while (retry--) {
+		reg_r(gspca_dev->dev, 0x0002, &val, 1);
+							/* 0x07 until 0x00 */
+		if (val == 0x00)
+			break;
+		val = 0x00;
+		reg_w(gspca_dev->dev, 0x0053, &val, 1);
+	}
+	if (retry == 0)
+		PDEBUG(D_ERR, "Damned Errors sending jpeg Table");
+	/* send the qtable now */
+	reg_r(gspca_dev->dev, 0x0001, &val, 1);		/* -> 0x18 */
+	length = 8;
+	for (i = 0; i < 18; i++) {
+		if (i == 17)
+			length = 2;
+		reg_w(gspca_dev->dev, 0x0008,
+				cxjpeg_qtable[i], length);
+
+	}
+	reg_r(gspca_dev->dev, 0x0002, &val, 1);	/* 0x00 */
+	reg_r(gspca_dev->dev, 0x0053, &val, 1);	/* 0x00 */
+	val = 0x02;
+	reg_w(gspca_dev->dev, 0x0054, &val, 1);
+	val = 0x01;
+	reg_w(gspca_dev->dev, 0x0054, &val, 1);
+	val = 0x94;
+	reg_w(gspca_dev->dev, 0x0000, &val, 1);
+	val = 0xc0;
+	reg_w(gspca_dev->dev, 0x0053, &val, 1);
+
+	reg_r(gspca_dev->dev, 0x0038, &val, 1);	/* 0x40 */
+	reg_r(gspca_dev->dev, 0x0038, &val, 1);	/* 0x40 */
+	reg_r(gspca_dev->dev, 0x001f, &val, 1);	/* 0x38 */
+	reg_w(gspca_dev->dev, 0x0012, reg12, 5);
+	reg_w(gspca_dev->dev, 0x00e5, regE5_8, 8);
+	reg_r(gspca_dev->dev, 0x00e8, bufread, 8);
+	reg_w(gspca_dev->dev, 0x00e5, regE5a, 4);
+	reg_r(gspca_dev->dev, 0x00e8, &val, 1);	/* 0x00 */
+	val = 0x01;
+	reg_w(gspca_dev->dev, 0x009a, &val, 1);
+	reg_w(gspca_dev->dev, 0x00e5, regE5b, 4);
+	reg_r(gspca_dev->dev, 0x00e8, &val, 1);	/* 0x00 */
+	reg_w(gspca_dev->dev, 0x00e5, regE5c, 4);
+	reg_r(gspca_dev->dev, 0x00e8, &val, 1);	/* 0x00 */
+
+	reg_w(gspca_dev->dev, 0x0051, reg51, 2);
+	reg_w(gspca_dev->dev, 0x0010, reg10, 2);
+	reg_w(gspca_dev->dev, 0x0070, &reg70, 1);
+}
+
+static void cx11646_init1(struct gspca_dev *gspca_dev)
+{
+	__u8 val;
+	int i = 0;
+
+	val = 0;
+	reg_w(gspca_dev->dev, 0x0010, &val, 1);
+	reg_w(gspca_dev->dev, 0x0053, &val, 1);
+	reg_w(gspca_dev->dev, 0x0052, &val, 1);
+	val = 0x2f;
+	reg_w(gspca_dev->dev, 0x009b, &val, 1);
+	val = 0x10;
+	reg_w(gspca_dev->dev, 0x009c, &val, 1);
+	reg_r(gspca_dev->dev, 0x0098, &val, 1);
+	val = 0x40;
+	reg_w(gspca_dev->dev, 0x0098, &val, 1);
+	reg_r(gspca_dev->dev, 0x0099, &val, 1);
+	val = 0x07;
+	reg_w(gspca_dev->dev, 0x0099, &val, 1);
+	val = 0x40;
+	reg_w(gspca_dev->dev, 0x0039, &val, 1);
+	val = 0xff;
+	reg_w(gspca_dev->dev, 0x003c, &val, 1);
+	val = 0x1f;
+	reg_w(gspca_dev->dev, 0x003f, &val, 1);
+	val = 0x40;
+	reg_w(gspca_dev->dev, 0x003d, &val, 1);
+/* val= 0x60; */
+/* reg_w(gspca_dev->dev,0x00,0x00,0x003d,&val,1); */
+	reg_r(gspca_dev->dev, 0x0099, &val, 1);			/* ->0x07 */
+
+	while (cx_sensor_init[i][0]) {
+		reg_w(gspca_dev->dev, 0x00e5, cx_sensor_init[i], 1);
+		reg_r(gspca_dev->dev, 0x00e8, &val, 1);		/* -> 0x00 */
+		if (i == 1) {
+			val = 1;
+			reg_w(gspca_dev->dev, 0x00ed, &val, 1);
+			reg_r(gspca_dev->dev, 0x00ed, &val, 1);	/* -> 0x01 */
+		}
+		i++;
+	}
+	val = 0x00;
+	reg_w(gspca_dev->dev, 0x00c3, &val, 1);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	cam->cam_mode = vga_mode;
+	cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+
+	sd->qindex = 0;			/* set the quantization table */
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	cx11646_init1(gspca_dev);
+	cx11646_initsize(gspca_dev);
+	cx11646_fw(gspca_dev);
+	cx_sensor(gspca_dev);
+	cx11646_jpegInit(gspca_dev);
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	cx11646_initsize(gspca_dev);
+	cx11646_fw(gspca_dev);
+	cx_sensor(gspca_dev);
+	cx11646_jpeg(gspca_dev);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+	int retry = 50;
+	__u8 val;
+
+	val = 0;
+	reg_w(gspca_dev->dev, 0x0000, &val, 1);
+	reg_r(gspca_dev->dev, 0x0002, &val, 1);
+	val = 0;
+	reg_w(gspca_dev->dev, 0x0053, &val, 1);
+
+	while (retry--) {
+/*		reg_r (gspca_dev->dev,0x00,0x00,0x0002,&val,1);*/
+		reg_r(gspca_dev->dev, 0x0053, &val, 1);
+		if (val == 0)
+			break;
+	}
+	val = 0;
+	reg_w(gspca_dev->dev, 0x0000, &val, 1);
+	reg_r(gspca_dev->dev, 0x0002, &val, 1);
+
+	val = 0;
+	reg_w(gspca_dev->dev, 0x0010, &val, 1);
+	reg_r(gspca_dev->dev, 0x0033, &val, 1);
+	val = 0xe0;
+	reg_w(gspca_dev->dev, 0x00fc, &val, 1);
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	if (data[0] == 0xff && data[1] == 0xd8) {
+
+		/* start of frame */
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					data, 0);
+
+		/* put the JPEG header in the new frame */
+		jpeg_put_header(gspca_dev, frame,
+				((struct sd *) gspca_dev)->qindex,
+				0x22);
+		data += 2;
+		len -= 2;
+	}
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+}
+
+static void setbrightness(struct gspca_dev*gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 regE5cbx[] = { 0x88, 0x00, 0xd4, 0x01, 0x88, 0x01, 0x01, 0x01 };
+	__u8 reg51c[] = { 0x77, 0x03 };
+	__u8 bright;
+	__u8 colors;
+	__u8 val;
+	__u8 bufread[8];
+
+	bright = sd->brightness;
+	colors = sd->colors;
+	regE5cbx[2] = bright;
+	reg51c[1] = colors;
+	reg_w(gspca_dev->dev, 0x00e5, regE5cbx, 8);
+	reg_r(gspca_dev->dev, 0x00e8, bufread, 8);
+	reg_w(gspca_dev->dev, 0x00e5, regE5c, 4);
+	reg_r(gspca_dev->dev, 0x00e8, &val, 1);	/* 0x00 */
+
+	reg_w(gspca_dev->dev, 0x0051, reg51c, 2);
+	reg_w(gspca_dev->dev, 0x0010, reg10, 2);
+	reg_w(gspca_dev->dev, 0x0070, &reg70, 1);
+}
+
+static void setcontrast(struct gspca_dev*gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 regE5acx[] = { 0x88, 0x0a, 0x0c, 0x01 };	/* seem MSB */
+	/* __u8 regE5bcx[]={0x88,0x0b,0x12,0x01}; // LSB */
+	__u8 reg51c[] = { 0x77, 0x03 };
+	__u8 val;
+
+	reg51c[1] = sd->colors;
+	regE5acx[2] = sd->contrast;
+	reg_w(gspca_dev->dev, 0x00e5, regE5acx, 4);
+	reg_r(gspca_dev->dev, 0x00e8, &val, 1);	/* 0x00 */
+	reg_w(gspca_dev->dev, 0x0051, reg51c, 2);
+	reg_w(gspca_dev->dev, 0x0010, reg10, 2);
+	reg_w(gspca_dev->dev, 0x0070, &reg70, 1);
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming) {
+		setbrightness(gspca_dev);
+		setcontrast(gspca_dev);
+	}
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->colors;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x0572, 0x0041), DVNM("Creative Notebook cx11646")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/etoms.c b/drivers/media/video/gspca/etoms.c
new file mode 100644
index 000000000000..c479f638413e
--- /dev/null
+++ b/drivers/media/video/gspca/etoms.c
@@ -0,0 +1,1062 @@
+/*
+ * Etoms Et61x151 GPL Linux driver by Michel Xhaard (09/09/2004)
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "etoms"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("Etoms USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned char brightness;
+	unsigned char contrast;
+	unsigned char colors;
+	unsigned char autogain;
+
+	char sensor;
+#define SENSOR_PAS106 0
+#define SENSOR_TAS5130CXX 1
+	signed char ag_cnt;
+#define AG_CNT_START 13
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	 {
+	  .id = V4L2_CID_BRIGHTNESS,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Brightness",
+	  .minimum = 1,
+	  .maximum = 127,
+	  .step = 1,
+	  .default_value = 63,
+	  },
+	 .set = sd_setbrightness,
+	 .get = sd_getbrightness,
+	 },
+#define SD_CONTRAST 1
+	{
+	 {
+	  .id = V4L2_CID_CONTRAST,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Contrast",
+	  .minimum = 0,
+	  .maximum = 255,
+	  .step = 1,
+	  .default_value = 127,
+	  },
+	 .set = sd_setcontrast,
+	 .get = sd_getcontrast,
+	 },
+#define SD_COLOR 2
+	{
+	 {
+	  .id = V4L2_CID_SATURATION,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Color",
+	  .minimum = 0,
+	  .maximum = 15,
+	  .step = 1,
+	  .default_value = 7,
+	  },
+	 .set = sd_setcolors,
+	 .get = sd_getcolors,
+	 },
+#define SD_AUTOGAIN 3
+	{
+	 {
+	  .id = V4L2_CID_AUTOGAIN,
+	  .type = V4L2_CTRL_TYPE_BOOLEAN,
+	  .name = "Auto Gain",
+	  .minimum = 0,
+	  .maximum = 1,
+	  .step = 1,
+	  .default_value = 1,
+	  },
+	 .set = sd_setautogain,
+	 .get = sd_getautogain,
+	 },
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_SBGGR8, 320, 240, 1},
+/*	{V4L2_PIX_FMT_SBGGR8, 640, 480, 0}, */
+};
+
+static struct cam_mode sif_mode[] = {
+	{V4L2_PIX_FMT_SBGGR8, 176, 144, 1},
+	{V4L2_PIX_FMT_SBGGR8, 352, 288, 0},
+};
+
+#define ETOMS_ALT_SIZE_1000   12
+
+#define ET_GPIO_DIR_CTRL 0x04	/* Control IO bit[0..5] (0 in  1 out) */
+#define ET_GPIO_OUT 0x05	/* Only IO data */
+#define ET_GPIO_IN 0x06		/* Read Only IO data */
+#define ET_RESET_ALL 0x03
+#define ET_ClCK 0x01
+#define ET_CTRL 0x02		/* enable i2c OutClck Powerdown mode */
+
+#define ET_COMP 0x12		/* Compression register */
+#define ET_MAXQt 0x13
+#define ET_MINQt 0x14
+#define ET_COMP_VAL0 0x02
+#define ET_COMP_VAL1 0x03
+
+#define ET_REG1d 0x1d
+#define ET_REG1e 0x1e
+#define ET_REG1f 0x1f
+#define ET_REG20 0x20
+#define ET_REG21 0x21
+#define ET_REG22 0x22
+#define ET_REG23 0x23
+#define ET_REG24 0x24
+#define ET_REG25 0x25
+/* base registers for luma calculation */
+#define ET_LUMA_CENTER 0x39
+
+#define ET_G_RED 0x4d
+#define ET_G_GREEN1 0x4e
+#define ET_G_BLUE 0x4f
+#define ET_G_GREEN2 0x50
+#define ET_G_GR_H 0x51
+#define ET_G_GB_H 0x52
+
+#define ET_O_RED 0x34
+#define ET_O_GREEN1 0x35
+#define ET_O_BLUE 0x36
+#define ET_O_GREEN2 0x37
+
+#define ET_SYNCHRO 0x68
+#define ET_STARTX 0x69
+#define ET_STARTY 0x6a
+#define ET_WIDTH_LOW 0x6b
+#define ET_HEIGTH_LOW 0x6c
+#define ET_W_H_HEIGTH 0x6d
+
+#define ET_REG6e 0x6e		/* OBW */
+#define ET_REG6f 0x6f		/* OBW */
+#define ET_REG70 0x70		/* OBW_AWB */
+#define ET_REG71 0x71		/* OBW_AWB */
+#define ET_REG72 0x72		/* OBW_AWB */
+#define ET_REG73 0x73		/* Clkdelay ns */
+#define ET_REG74 0x74		/* test pattern */
+#define ET_REG75 0x75		/* test pattern */
+
+#define ET_I2C_CLK 0x8c
+#define ET_PXL_CLK 0x60
+
+#define ET_I2C_BASE 0x89
+#define ET_I2C_COUNT 0x8a
+#define ET_I2C_PREFETCH 0x8b
+#define ET_I2C_REG 0x88
+#define ET_I2C_DATA7 0x87
+#define ET_I2C_DATA6 0x86
+#define ET_I2C_DATA5 0x85
+#define ET_I2C_DATA4 0x84
+#define ET_I2C_DATA3 0x83
+#define ET_I2C_DATA2 0x82
+#define ET_I2C_DATA1 0x81
+#define ET_I2C_DATA0 0x80
+
+#define PAS106_REG2 0x02	/* pxlClk = systemClk/(reg2) */
+#define PAS106_REG3 0x03	/* line/frame H [11..4] */
+#define PAS106_REG4 0x04	/* line/frame L [3..0] */
+#define PAS106_REG5 0x05	/* exposure time line offset(default 5) */
+#define PAS106_REG6 0x06	/* exposure time pixel offset(default 6) */
+#define PAS106_REG7 0x07	/* signbit Dac (default 0) */
+#define PAS106_REG9 0x09
+#define PAS106_REG0e 0x0e	/* global gain [4..0](default 0x0e) */
+#define PAS106_REG13 0x13	/* end i2c write */
+
+static __u8 GainRGBG[] = { 0x80, 0x80, 0x80, 0x80, 0x00, 0x00 };
+
+static __u8 I2c2[] = { 0x08, 0x08, 0x08, 0x08, 0x0d };
+
+static __u8 I2c3[] = { 0x12, 0x05 };
+
+static __u8 I2c4[] = { 0x41, 0x08 };
+
+static void Et_RegRead(struct usb_device *dev,
+		       __u16 index, __u8 *buffer, int len)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_INTERFACE,
+			0, index, buffer, len, 500);
+}
+
+static void Et_RegWrite(struct usb_device *dev,
+			__u16 index, __u8 *buffer, __u16 len)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			0,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_INTERFACE,
+			0, index, buffer, len, 500);
+}
+
+static int Et_i2cwrite(struct usb_device *dev, __u8 reg, __u8 * buffer,
+		       __u16 length, __u8 mode)
+{
+/* buffer should be [D0..D7] */
+	int i, j;
+	__u8 base = 0x40;	/* sensor base for the pas106 */
+	__u8 ptchcount = 0;
+
+	ptchcount = (((length & 0x07) << 4) | (mode & 0x03));
+/* set the base address */
+	Et_RegWrite(dev, ET_I2C_BASE, &base, 1);
+/* set count and prefetch */
+	Et_RegWrite(dev, ET_I2C_COUNT, &ptchcount, 1);
+/* set the register base */
+	Et_RegWrite(dev, ET_I2C_REG, &reg, 1);
+	j = length - 1;
+	for (i = 0; i < length; i++) {
+		Et_RegWrite(dev, (ET_I2C_DATA0 + j), &buffer[j], 1);
+		j--;
+	}
+	return 0;
+}
+
+static int Et_i2cread(struct usb_device *dev, __u8 reg, __u8 * buffer,
+		      __u16 length, __u8 mode)
+{
+/* buffer should be [D0..D7] */
+	int i, j;
+	__u8 base = 0x40;	/* sensor base for the pas106 */
+	__u8 ptchcount;
+	__u8 prefetch = 0x02;
+
+	ptchcount = (((length & 0x07) << 4) | (mode & 0x03));
+/* set the base address */
+	Et_RegWrite(dev, ET_I2C_BASE, &base, 1);
+/* set count and prefetch */
+	Et_RegWrite(dev, ET_I2C_COUNT, &ptchcount, 1);
+/* set the register base */
+	Et_RegWrite(dev, ET_I2C_REG, &reg, 1);
+	Et_RegWrite(dev, ET_I2C_PREFETCH, &prefetch, 1);
+	prefetch = 0x00;
+	Et_RegWrite(dev, ET_I2C_PREFETCH, &prefetch, 1);
+	j = length - 1;
+	for (i = 0; i < length; i++) {
+		Et_RegRead(dev, (ET_I2C_DATA0 + j), &buffer[j], 1);
+		j--;
+	}
+	return 0;
+}
+
+static int Et_WaitStatus(struct usb_device *dev)
+{
+	__u8 bytereceived;
+	int retry = 10;
+
+	while (retry--) {
+		Et_RegRead(dev, ET_ClCK, &bytereceived, 1);
+		if (bytereceived != 0)
+			return 1;
+	}
+	return 0;
+}
+
+static int Et_videoOff(struct usb_device *dev)
+{
+	int err;
+	__u8 stopvideo = 0;
+
+	Et_RegWrite(dev, ET_GPIO_OUT, &stopvideo, 1);
+	err = Et_WaitStatus(dev);
+	if (!err)
+		PDEBUG(D_ERR, "timeout Et_waitStatus VideoON");
+	return err;
+}
+
+static int Et_videoOn(struct usb_device *dev)
+{
+	int err;
+	__u8 startvideo = 0x10;	/* set Bit5 */
+
+	Et_RegWrite(dev, ET_GPIO_OUT, &startvideo, 1);
+	err = Et_WaitStatus(dev);
+	if (!err)
+		PDEBUG(D_ERR, "timeout Et_waitStatus VideoOFF");
+	return err;
+}
+
+static void Et_init2(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 value = 0x00;
+	__u8 received = 0x00;
+	__u8 FormLine[] = { 0x84, 0x03, 0x14, 0xf4, 0x01, 0x05 };
+
+	PDEBUG(D_STREAM, "Open Init2 ET");
+	value = 0x2f;
+	Et_RegWrite(dev, ET_GPIO_DIR_CTRL, &value, 1);
+	value = 0x10;
+	Et_RegWrite(dev, ET_GPIO_OUT, &value, 1);
+	Et_RegRead(dev, ET_GPIO_IN, &received, 1);
+	value = 0x14;		/* 0x14 // 0x16 enabled pattern */
+	Et_RegWrite(dev, ET_ClCK, &value, 1);
+	value = 0x1b;
+	Et_RegWrite(dev, ET_CTRL, &value, 1);
+
+	/*  compression et subsampling */
+	if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode)
+		value = ET_COMP_VAL1;	/* 320 */
+	else
+		value = ET_COMP_VAL0;	/* 640 */
+	Et_RegWrite(dev, ET_COMP, &value, 1);
+	value = 0x1f;
+	Et_RegWrite(dev, ET_MAXQt, &value, 1);
+	value = 0x04;
+	Et_RegWrite(dev, ET_MINQt, &value, 1);
+	/* undocumented registers */
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG1d, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG1e, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG1f, &value, 1);
+	value = 0x35;
+	Et_RegWrite(dev, ET_REG20, &value, 1);
+	value = 0x01;
+	Et_RegWrite(dev, ET_REG21, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_REG22, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG23, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG24, &value, 1);
+	value = 0x0f;
+	Et_RegWrite(dev, ET_REG25, &value, 1);
+	/* colors setting */
+	value = 0x11;
+	Et_RegWrite(dev, 0x30, &value, 1);	/* 0x30 */
+	value = 0x40;
+	Et_RegWrite(dev, 0x31, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, 0x32, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_O_RED, &value, 1);	/* 0x34 */
+	value = 0x00;
+	Et_RegWrite(dev, ET_O_GREEN1, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_O_BLUE, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_O_GREEN2, &value, 1);
+	/*************/
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_RED, &value, 1);	/* 0x4d */
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_GREEN1, &value, 1);
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_BLUE, &value, 1);
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_GREEN2, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_G_GR_H, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_G_GB_H, &value, 1);	/* 0x52 */
+	/* Window control registers */
+
+	value = 0x80;		/* use cmc_out */
+	Et_RegWrite(dev, 0x61, &value, 1);
+
+	value = 0x02;
+	Et_RegWrite(dev, 0x62, &value, 1);
+	value = 0x03;
+	Et_RegWrite(dev, 0x63, &value, 1);
+	value = 0x14;
+	Et_RegWrite(dev, 0x64, &value, 1);
+	value = 0x0e;
+	Et_RegWrite(dev, 0x65, &value, 1);
+	value = 0x02;
+	Et_RegWrite(dev, 0x66, &value, 1);
+	value = 0x02;
+	Et_RegWrite(dev, 0x67, &value, 1);
+
+	/**************************************/
+	value = 0x8f;
+	Et_RegWrite(dev, ET_SYNCHRO, &value, 1);	/* 0x68 */
+	value = 0x69;		/* 0x6a //0x69 */
+	Et_RegWrite(dev, ET_STARTX, &value, 1);
+	value = 0x0d;		/* 0x0d //0x0c */
+	Et_RegWrite(dev, ET_STARTY, &value, 1);
+	value = 0x80;
+	Et_RegWrite(dev, ET_WIDTH_LOW, &value, 1);
+	value = 0xe0;
+	Et_RegWrite(dev, ET_HEIGTH_LOW, &value, 1);
+	value = 0x60;
+	Et_RegWrite(dev, ET_W_H_HEIGTH, &value, 1);	/* 6d */
+	value = 0x86;
+	Et_RegWrite(dev, ET_REG6e, &value, 1);
+	value = 0x01;
+	Et_RegWrite(dev, ET_REG6f, &value, 1);
+	value = 0x26;
+	Et_RegWrite(dev, ET_REG70, &value, 1);
+	value = 0x7a;
+	Et_RegWrite(dev, ET_REG71, &value, 1);
+	value = 0x01;
+	Et_RegWrite(dev, ET_REG72, &value, 1);
+	/* Clock Pattern registers ***************** */
+	value = 0x00;
+	Et_RegWrite(dev, ET_REG73, &value, 1);
+	value = 0x18;		/* 0x28 */
+	Et_RegWrite(dev, ET_REG74, &value, 1);
+	value = 0x0f;		/* 0x01 */
+	Et_RegWrite(dev, ET_REG75, &value, 1);
+	/**********************************************/
+	value = 0x20;
+	Et_RegWrite(dev, 0x8a, &value, 1);
+	value = 0x0f;
+	Et_RegWrite(dev, 0x8d, &value, 1);
+	value = 0x08;
+	Et_RegWrite(dev, 0x8e, &value, 1);
+	/**************************************/
+	value = 0x08;
+	Et_RegWrite(dev, 0x03, &value, 1);
+	value = 0x03;
+	Et_RegWrite(dev, ET_PXL_CLK, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, 0x81, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, 0x80, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, 0x81, &value, 1);
+	value = 0x20;
+	Et_RegWrite(dev, 0x80, &value, 1);
+	value = 0x01;
+	Et_RegWrite(dev, 0x03, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, 0x03, &value, 1);
+	value = 0x08;
+	Et_RegWrite(dev, 0x03, &value, 1);
+	/********************************************/
+
+	/* Et_RegRead(dev,0x0,ET_I2C_BASE,&received,1);
+					 always 0x40 as the pas106 ??? */
+	/* set the sensor */
+	if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+		value = 0x04;	/* 320 */
+		Et_RegWrite(dev, ET_PXL_CLK, &value, 1);
+		/* now set by fifo the FormatLine setting */
+		Et_RegWrite(dev, 0x62, FormLine, 6);
+	} else {		/* 640 */
+		/* setting PixelClock
+		   0x03 mean 24/(3+1) = 6 Mhz
+		   0x05 -> 24/(5+1) = 4 Mhz
+		   0x0b -> 24/(11+1) = 2 Mhz
+		   0x17 -> 24/(23+1) = 1 Mhz
+		 */
+		value = 0x1e;	/* 0x17 */
+		Et_RegWrite(dev, ET_PXL_CLK, &value, 1);
+		/* now set by fifo the FormatLine setting */
+		Et_RegWrite(dev, 0x62, FormLine, 6);
+	}
+
+	/* set exposure times [ 0..0x78] 0->longvalue 0x78->shortvalue */
+	value = 0x47;		/* 0x47; */
+	Et_RegWrite(dev, 0x81, &value, 1);
+	value = 0x40;		/* 0x40; */
+	Et_RegWrite(dev, 0x80, &value, 1);
+	/* Pedro change */
+	/* Brightness change Brith+ decrease value */
+	/* Brigth- increase value */
+	/* original value = 0x70; */
+	value = 0x30;		/* 0x20; */
+	Et_RegWrite(dev, 0x81, &value, 1);	/* set brightness */
+	value = 0x20;		/* 0x20; */
+	Et_RegWrite(dev, 0x80, &value, 1);
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	static __u8 I2cc[] = { 0x05, 0x02, 0x02, 0x05, 0x0d };
+	__u8 i2cflags = 0x01;
+	/* __u8 green = 0; */
+	__u8 colors = sd->colors;
+
+	I2cc[3] = colors;	/* red */
+	I2cc[0] = 15 - colors;	/* blue */
+	/* green = 15 - ((((7*I2cc[0]) >> 2 ) + I2cc[3]) >> 1); */
+	/* I2cc[1] = I2cc[2] = green; */
+	if (sd->sensor == SENSOR_PAS106) {
+		Et_i2cwrite(dev, PAS106_REG13, &i2cflags, 1, 3);
+		Et_i2cwrite(dev, PAS106_REG9, I2cc, sizeof(I2cc), 1);
+	}
+/*	PDEBUG(D_CONF , "Etoms red %d blue %d green %d",
+		I2cc[3], I2cc[0], green); */
+}
+
+static void getcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	/* __u8 valblue = 0; */
+	__u8 valred;
+
+	if (sd->sensor == SENSOR_PAS106) {
+		/* Et_i2cread(gspca_dev->dev,PAS106_REG9,&valblue,1,1); */
+		Et_i2cread(gspca_dev->dev, PAS106_REG9 + 3, &valred, 1, 1);
+		sd->colors = valred & 0x0f;
+	}
+}
+
+static void Et_init1(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 value = 0x00;
+	__u8 received = 0x00;
+/*	__u8 I2c0 [] ={0x0a,0x12,0x05,0x22,0xac,0x00,0x01,0x00}; */
+	__u8 I2c0[] = { 0x0a, 0x12, 0x05, 0x6d, 0xcd, 0x00, 0x01, 0x00 };
+						/* try 1/120 0x6d 0xcd 0x40 */
+/*	__u8 I2c0 [] ={0x0a,0x12,0x05,0xfe,0xfe,0xc0,0x01,0x00};
+						 * 1/60000 hmm ?? */
+
+	PDEBUG(D_STREAM, "Open Init1 ET");
+	value = 7;
+	Et_RegWrite(dev, ET_GPIO_DIR_CTRL, &value, 1);
+	Et_RegRead(dev, ET_GPIO_IN, &received, 1);
+	value = 1;
+	Et_RegWrite(dev, ET_RESET_ALL, &value, 1);
+	value = 0;
+	Et_RegWrite(dev, ET_RESET_ALL, &value, 1);
+	value = 0x10;
+	Et_RegWrite(dev, ET_ClCK, &value, 1);
+	value = 0x19;
+	Et_RegWrite(dev, ET_CTRL, &value, 1);
+	/*   compression et subsampling */
+	if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode)
+		value = ET_COMP_VAL1;
+	else
+		value = ET_COMP_VAL0;
+
+	PDEBUG(D_STREAM, "Open mode %d Compression %d",
+	       gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode,
+	       value);
+	Et_RegWrite(dev, ET_COMP, &value, 1);
+	value = 0x1d;
+	Et_RegWrite(dev, ET_MAXQt, &value, 1);
+	value = 0x02;
+	Et_RegWrite(dev, ET_MINQt, &value, 1);
+	/* undocumented registers */
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG1d, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG1e, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG1f, &value, 1);
+	value = 0x35;
+	Et_RegWrite(dev, ET_REG20, &value, 1);
+	value = 0x01;
+	Et_RegWrite(dev, ET_REG21, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_REG22, &value, 1);
+	value = 0xf7;
+	Et_RegWrite(dev, ET_REG23, &value, 1);
+	value = 0xff;
+	Et_RegWrite(dev, ET_REG24, &value, 1);
+	value = 0x07;
+	Et_RegWrite(dev, ET_REG25, &value, 1);
+	/* colors setting */
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_RED, &value, 1);
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_GREEN1, &value, 1);
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_BLUE, &value, 1);
+	value = 0x80;
+	Et_RegWrite(dev, ET_G_GREEN2, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_G_GR_H, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_G_GB_H, &value, 1);
+	/* Window control registers */
+	value = 0xf0;
+	Et_RegWrite(dev, ET_SYNCHRO, &value, 1);
+	value = 0x56;		/* 0x56 */
+	Et_RegWrite(dev, ET_STARTX, &value, 1);
+	value = 0x05;		/* 0x04 */
+	Et_RegWrite(dev, ET_STARTY, &value, 1);
+	value = 0x60;
+	Et_RegWrite(dev, ET_WIDTH_LOW, &value, 1);
+	value = 0x20;
+	Et_RegWrite(dev, ET_HEIGTH_LOW, &value, 1);
+	value = 0x50;
+	Et_RegWrite(dev, ET_W_H_HEIGTH, &value, 1);
+	value = 0x86;
+	Et_RegWrite(dev, ET_REG6e, &value, 1);
+	value = 0x01;
+	Et_RegWrite(dev, ET_REG6f, &value, 1);
+	value = 0x86;
+	Et_RegWrite(dev, ET_REG70, &value, 1);
+	value = 0x14;
+	Et_RegWrite(dev, ET_REG71, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_REG72, &value, 1);
+	/* Clock Pattern registers */
+	value = 0x00;
+	Et_RegWrite(dev, ET_REG73, &value, 1);
+	value = 0x00;
+	Et_RegWrite(dev, ET_REG74, &value, 1);
+	value = 0x0a;
+	Et_RegWrite(dev, ET_REG75, &value, 1);
+	value = 0x04;
+	Et_RegWrite(dev, ET_I2C_CLK, &value, 1);
+	value = 0x01;
+	Et_RegWrite(dev, ET_PXL_CLK, &value, 1);
+	/* set the sensor */
+	if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+		I2c0[0] = 0x06;
+		Et_i2cwrite(dev, PAS106_REG2, I2c0, sizeof(I2c0), 1);
+		Et_i2cwrite(dev, PAS106_REG9, I2c2, sizeof(I2c2), 1);
+		value = 0x06;
+		Et_i2cwrite(dev, PAS106_REG2, &value, 1, 1);
+		Et_i2cwrite(dev, PAS106_REG3, I2c3, sizeof(I2c3), 1);
+		/* value = 0x1f; */
+		value = 0x04;
+		Et_i2cwrite(dev, PAS106_REG0e, &value, 1, 1);
+	} else {
+		I2c0[0] = 0x0a;
+
+		Et_i2cwrite(dev, PAS106_REG2, I2c0, sizeof(I2c0), 1);
+		Et_i2cwrite(dev, PAS106_REG9, I2c2, sizeof(I2c2), 1);
+		value = 0x0a;
+
+		Et_i2cwrite(dev, PAS106_REG2, &value, 1, 1);
+		Et_i2cwrite(dev, PAS106_REG3, I2c3, sizeof(I2c3), 1);
+		value = 0x04;
+		/* value = 0x10; */
+		Et_i2cwrite(dev, PAS106_REG0e, &value, 1, 1);
+		/* bit 2 enable bit 1:2 select 0 1 2 3
+		   value = 0x07;                                * curve 0 *
+		   Et_i2cwrite(dev,PAS106_REG0f,&value,1,1);
+		 */
+	}
+
+/*	value = 0x01; */
+/*	value = 0x22; */
+/*	Et_i2cwrite(dev, PAS106_REG5, &value, 1, 1); */
+	/* magnetude and sign bit for DAC */
+	Et_i2cwrite(dev, PAS106_REG7, I2c4, sizeof I2c4, 1);
+	/* now set by fifo the whole colors setting */
+	Et_RegWrite(dev, ET_G_RED, GainRGBG, 6);
+	getcolors(gspca_dev);
+	setcolors(gspca_dev);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+		     const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+	__u16 vendor;
+	__u16 product;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+/*	switch (vendor) { */
+/*	case 0x102c:		* Etoms */
+		switch (product) {
+		case 0x6151:
+			sd->sensor = SENSOR_PAS106;	/* Etoms61x151 */
+			break;
+		case 0x6251:
+			sd->sensor = SENSOR_TAS5130CXX;	/* Etoms61x251 */
+			break;
+/*		} */
+/*		break; */
+	}
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 1;
+	if (sd->sensor == SENSOR_PAS106) {
+		cam->cam_mode = sif_mode;
+		cam->nmodes = sizeof sif_mode / sizeof sif_mode[0];
+	} else {
+		cam->cam_mode = vga_mode;
+		cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	}
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	sd->autogain = sd_ctrls[SD_AUTOGAIN].qctrl.default_value;
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int err;
+	__u8 value;
+
+	PDEBUG(D_STREAM, "Initialize ET1");
+	if (sd->sensor == SENSOR_PAS106)
+		Et_init1(gspca_dev);
+	else
+		Et_init2(gspca_dev);
+	value = 0x08;
+	Et_RegWrite(dev, ET_RESET_ALL, &value, 1);
+	err = Et_videoOff(dev);
+	PDEBUG(D_STREAM, "Et_Init_VideoOff %d", err);
+	return 0;
+}
+
+/* -- start the camera -- */
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int err;
+	__u8 value;
+
+	if (sd->sensor == SENSOR_PAS106)
+		Et_init1(gspca_dev);
+	else
+		Et_init2(gspca_dev);
+
+	value = 0x08;
+	Et_RegWrite(dev, ET_RESET_ALL, &value, 1);
+	err = Et_videoOn(dev);
+	PDEBUG(D_STREAM, "Et_VideoOn %d", err);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	int err;
+
+	err = Et_videoOff(gspca_dev->dev);
+	PDEBUG(D_STREAM, "Et_VideoOff %d", err);
+
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int i;
+	__u8 brightness = sd->brightness;
+
+	for (i = 0; i < 4; i++)
+		Et_RegWrite(gspca_dev->dev, (ET_O_RED + i), &brightness, 1);
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int i;
+	int brightness = 0;
+	__u8 value = 0;
+
+	for (i = 0; i < 4; i++) {
+		Et_RegRead(gspca_dev->dev, (ET_O_RED + i), &value, 1);
+		brightness += value;
+	}
+	sd->brightness = brightness >> 3;
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 RGBG[] = { 0x80, 0x80, 0x80, 0x80, 0x00, 0x00 };
+	__u8 contrast = sd->contrast;
+
+	memset(RGBG, contrast, sizeof RGBG - 2);
+	Et_RegWrite(gspca_dev->dev, ET_G_RED, RGBG, 6);
+}
+
+static void getcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int i;
+	int contrast = 0;
+	__u8 value = 0;
+
+	for (i = 0; i < 4; i++) {
+		Et_RegRead(gspca_dev->dev, (ET_G_RED + i), &value, 1);
+		contrast += value;
+	}
+	sd->contrast = contrast >> 2;
+}
+
+static __u8 Et_getgainG(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 value = 0;
+
+	if (sd->sensor == SENSOR_PAS106) {
+		Et_i2cread(gspca_dev->dev, PAS106_REG0e, &value, 1, 1);
+		PDEBUG(D_CONF, "Etoms gain G %d", value);
+		return value;
+	}
+	return 0x1f;
+}
+
+static void Et_setgainG(struct gspca_dev *gspca_dev, __u8 gain)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 i2cflags = 0x01;
+
+	if (sd->sensor == SENSOR_PAS106) {
+		Et_i2cwrite(dev, PAS106_REG13, &i2cflags, 1, 3);
+		Et_i2cwrite(dev, PAS106_REG0e, &gain, 1, 1);
+	}
+}
+
+#define BLIMIT(bright) \
+	(__u8)((bright > 0x1f)?0x1f:((bright < 4)?3:bright))
+#define LIMIT(color) \
+	(unsigned char)((color > 0xff)?0xff:((color < 0)?0:color))
+
+static void setautogain(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 GRBG[] = { 0, 0, 0, 0 };
+	__u8 luma = 0;
+	__u8 luma_mean = 128;
+	__u8 luma_delta = 20;
+	__u8 spring = 4;
+	int Gbright = 0;
+	__u8 r, g, b;
+
+	Gbright = Et_getgainG(gspca_dev);
+	Et_RegRead(dev, ET_LUMA_CENTER, GRBG, 4);
+	g = (GRBG[0] + GRBG[3]) >> 1;
+	r = GRBG[1];
+	b = GRBG[2];
+	r = ((r << 8) - (r << 4) - (r << 3)) >> 10;
+	b = ((b << 7) >> 10);
+	g = ((g << 9) + (g << 7) + (g << 5)) >> 10;
+	luma = LIMIT(r + g + b);
+	PDEBUG(D_FRAM, "Etoms luma G %d", luma);
+	if (luma < luma_mean - luma_delta || luma > luma_mean + luma_delta) {
+		Gbright += (luma_mean - luma) >> spring;
+		Gbright = BLIMIT(Gbright);
+		PDEBUG(D_FRAM, "Etoms Gbright %d", Gbright);
+		Et_setgainG(gspca_dev, (__u8) Gbright);
+	}
+}
+
+#undef BLIMIT
+#undef LIMIT
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd;
+	int seqframe;
+
+	seqframe = data[0] & 0x3f;
+	len = (int) (((data[0] & 0xc0) << 2) | data[1]);
+	if (seqframe == 0x3f) {
+		PDEBUG(D_FRAM,
+		       "header packet found datalength %d !!", len);
+		PDEBUG(D_FRAM, "G %d R %d G %d B %d",
+		       data[2], data[3], data[4], data[5]);
+		data += 30;
+		/* don't change datalength as the chips provided it */
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					data, 0);
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame, data, len);
+		sd = (struct sd *) gspca_dev;
+		if (sd->ag_cnt >= 0) {
+			if (--sd->ag_cnt < 0) {
+				sd->ag_cnt = AG_CNT_START;
+				setautogain(gspca_dev);
+			}
+		}
+		return;
+	}
+	if (len) {
+		data += 8;
+		gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+	} else {			/* Drop Packet */
+		gspca_dev->last_packet_type = DISCARD_PACKET;
+	}
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcontrast(gspca_dev);
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcolors(gspca_dev);
+	*val = sd->colors;
+	return 0;
+}
+
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->autogain = val;
+	if (val)
+		sd->ag_cnt = AG_CNT_START;
+	else
+		sd->ag_cnt = -1;
+	return 0;
+}
+
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autogain;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x102c, 0x6151), DVNM("Qcam Sangha CIF")},
+	{USB_DEVICE(0x102c, 0x6251), DVNM("Qcam xxxxxx VGA")},
+	{}
+};
+
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+		    const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+			       THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/gspca.c b/drivers/media/video/gspca/gspca.c
index 423ebbdc4b4f..5583c53e4863 100644
--- a/drivers/media/video/gspca/gspca.c
+++ b/drivers/media/video/gspca/gspca.c
@@ -35,24 +35,24 @@
 
 #include "gspca.h"
 
-/* option */
-#define GSPCA_HLP 0
+#undef CONFIG_VIDEO_V4L1_COMPAT
 
 /* global values */
 #define DEF_NURBS 2		/* default number of URBs (mmap) */
+#define USR_NURBS 5		/* default number of URBs (userptr) */
 
 MODULE_AUTHOR("Jean-Francois Moine <http://moinejf.free.fr>");
 MODULE_DESCRIPTION("GSPCA USB Camera Driver");
 MODULE_LICENSE("GPL");
 
-#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(0, 2, 15)
-static const char version[] = "0.2.15";
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
 
 static int video_nr = -1;
 
 static int comp_fac = 30;	/* Buffer size ratio when compressed in % */
 
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 int gspca_debug = D_ERR | D_PROBE;
 EXPORT_SYMBOL(gspca_debug);
 
@@ -81,224 +81,7 @@ static void PDEBUG_MODE(char *txt, __u32 pixfmt, int w, int h)
 #define GSPCA_MEMORY_NO 0	/* V4L2_MEMORY_xxx starts from 1 */
 #define GSPCA_MEMORY_READ 7
 
-#ifndef GSPCA_HLP
 #define BUF_ALL_FLAGS (V4L2_BUF_FLAG_QUEUED | V4L2_BUF_FLAG_DONE)
-#else
-#define GSPCA_BUF_FLAG_DECODE	0x1000	/* internal buffer flag */
-#define BUF_ALL_FLAGS (V4L2_BUF_FLAG_QUEUED | V4L2_BUF_FLAG_DONE \
-			| GSPCA_BUF_FLAG_DECODE)
-
-static int autostart = 4;
-module_param(autostart, int, 0644);
-MODULE_PARM_DESC(autostart,
-		"Automatically start the helper process");
-
-/* try to start the helper process */
-static void start_hlp(void)
-{
-	int ret;
-	static char *argv[] = {"gspca_hlp", NULL};
-	static char *env[] = {NULL};
-
-	if (autostart <= 0) {
-		if (autostart < 0)
-			PDEBUG(D_ERR|D_PROBE, "Too many helper restart");
-		return;
-	}
-	autostart--;
-	if (autostart == 0)
-		autostart = -1;
-	ret = call_usermodehelper("/sbin/gspca_hlp", argv, env,
-				UMH_WAIT_EXEC);
-	if (ret != 0)
-		PDEBUG(D_ERR|D_PROBE,
-			"/sbin/gspca_hlp start failed %d", ret);
-}
-
-/* /dev/gspca_hlp stuff */
-#include <linux/miscdevice.h>
-#include "gspca_hlp.h"
-
-/* !! possible decodings defined in decoder.c */
-static __u32 bayer_to_tb[] = {
-	V4L2_PIX_FMT_SBGGR8,
-	V4L2_PIX_FMT_YUYV,
-	V4L2_PIX_FMT_YUV420,
-	V4L2_PIX_FMT_RGB24,
-	V4L2_PIX_FMT_BGR24,
-	V4L2_PIX_FMT_RGB565,
-};
-static __u32 jpeg_to_tb[] = {
-	V4L2_PIX_FMT_JPEG,
-	V4L2_PIX_FMT_YUYV,
-	V4L2_PIX_FMT_YUV420,
-	V4L2_PIX_FMT_RGB24,
-	V4L2_PIX_FMT_BGR24,
-	V4L2_PIX_FMT_RGB565,
-};
-
-/* /dev/gspca_hlp device */
-struct hlp_dev {
-	struct gspca_dev *gspca_dev;	/* associated device */
-	struct gspca_frame *frame;	/* frame being decoded */
-	__u32 pixfmt;			/* webcam pixel format */
-	atomic_t nevent;		/* nb of frames ready to decode */
-	wait_queue_head_t wq;		/* wait queue */
-	char fr_d;			/* next frame to decode */
-} *hlp;
-
-static int hlp_open(struct inode *inode, struct file *file)
-{
-	struct hlp_dev *hlp_dev;
-
-	PDEBUG(D_CONF, "hlp open");
-	if (hlp != 0)
-		return -EBUSY;
-	hlp_dev = kzalloc(sizeof *hlp_dev, GFP_KERNEL);
-	if (hlp_dev == NULL) {
-		err("couldn't kzalloc hlp struct");
-		return -EIO;
-	}
-	init_waitqueue_head(&hlp_dev->wq);
-	file->private_data = hlp_dev;
-	hlp = hlp_dev;
-	return 0;
-}
-
-static int hlp_close(struct inode *inode, struct file *file)
-{
-	struct gspca_dev *gspca_dev;
-	int mode;
-
-	PDEBUG(D_CONF, "hlp close");
-	file->private_data = NULL;
-
-	/* stop decoding */
-	gspca_dev = hlp->gspca_dev;
-	if (gspca_dev != 0) {
-		mode = gspca_dev->curr_mode;
-		gspca_dev->pixfmt = gspca_dev->cam.cam_mode[mode].pixfmt;
-	}
-
-	/* destroy the helper structure */
-	kfree(hlp);
-	hlp = 0;
-
-	/* try to restart the helper process */
-	start_hlp();
-	return 0;
-}
-
-static ssize_t hlp_read(struct file *file, char __user *buf,
-		    size_t cnt, loff_t *ppos)
-{
-	struct hlp_dev *hlp_dev = file->private_data;
-	struct gspca_dev *gspca_dev;
-	struct gspca_frame *frame;
-	struct gspca_hlp_read_hd head;
-	int i, j, len, ret;
-
-	PDEBUG(D_FRAM, "hlp read (%d)", cnt);
-
-	/* check / wait till a frame is ready */
-	for (;;) {
-		gspca_dev = hlp_dev->gspca_dev;
-		if (gspca_dev != 0 && gspca_dev->streaming) {
-			i = hlp_dev->fr_d;	/* frame to decode */
-			j = gspca_dev->fr_queue[i];
-			frame = &gspca_dev->frame[j];
-			if (frame->v4l2_buf.flags & GSPCA_BUF_FLAG_DECODE)
-				break;
-		}
-		ret = wait_event_interruptible(hlp_dev->wq,
-					atomic_read(&hlp_dev->nevent) > 0);
-		if (ret < 0) {			/* helper process is killed */
-			autostart = 0;		/* don't restart it */
-			return ret;
-		}
-	}
-	atomic_dec(&hlp_dev->nevent);
-	hlp_dev->fr_d = (i + 1) % gspca_dev->nframes;
-	PDEBUG(D_FRAM, "hlp read q:%d i:%d d:%d o:%d",
-		gspca_dev->fr_q,
-		gspca_dev->fr_i,
-		hlp_dev->fr_d,
-		gspca_dev->fr_o);
-
-	hlp_dev->frame = frame;		/* memorize the current frame */
-	len = frame->v4l2_buf.bytesused;
-	if (cnt < sizeof head - sizeof head.data + len)
-/*fixme: special errno?*/
-		return -EINVAL;
-	head.pixfmt_out = gspca_dev->pixfmt;
-	head.pixfmt_in = hlp_dev->pixfmt;
-	head.width = gspca_dev->width;
-	head.height = gspca_dev->height;
-	copy_to_user(buf, &head, sizeof head);
-	copy_to_user(buf + sizeof head - sizeof head.data,
-			frame->data, len);
-	return sizeof head - sizeof head.data + len;
-}
-
-static ssize_t hlp_write(struct file *file,
-			const char __user *buf,
-			size_t cnt, loff_t *ppos)
-{
-	struct hlp_dev *hlp_dev = file->private_data;
-	struct gspca_dev *gspca_dev;
-	struct gspca_frame *frame;
-
-	PDEBUG(D_FRAM, "hlp write (%d)", cnt);
-	gspca_dev = hlp_dev->gspca_dev;
-	if (gspca_dev == 0)
-		return cnt;
-	if (mutex_lock_interruptible(&gspca_dev->queue_lock))
-		return -ERESTARTSYS;
-	if (!gspca_dev->streaming)
-		goto out;
-	frame = hlp_dev->frame;
-	hlp_dev->frame = 0;
-	if (frame == 0)
-		goto out;
-	if (cnt > frame->v4l2_buf.length) {
-		PDEBUG(D_ERR|D_FRAM, "bad frame size %d - %d",
-			cnt, frame->v4l2_buf.length);
-		cnt = -EINVAL;
-		goto out;
-	}
-	copy_from_user(frame->data, buf, cnt);
-	frame->v4l2_buf.bytesused = cnt;
-	frame->v4l2_buf.flags &= ~(V4L2_BUF_FLAG_QUEUED
-				  | GSPCA_BUF_FLAG_DECODE);
-	frame->v4l2_buf.flags |= V4L2_BUF_FLAG_DONE;
-	mutex_unlock(&gspca_dev->queue_lock);
-	atomic_inc(&gspca_dev->nevent);
-	wake_up_interruptible(&gspca_dev->wq);	/* event = new frame */
-	PDEBUG(D_FRAM, "hlp write q:%d i:%d d:%d o:%d",
-		gspca_dev->fr_q,
-		gspca_dev->fr_i,
-		hlp_dev->fr_d,
-		gspca_dev->fr_o);
-	return cnt;
-out:
-	mutex_unlock(&gspca_dev->queue_lock);
-	return cnt;
-}
-
-static struct file_operations hlp_fops = {
-	.owner = THIS_MODULE,
-	.open = hlp_open,
-	.release = hlp_close,
-	.read = hlp_read,
-	.write = hlp_write,
-	.llseek = no_llseek
-};
-static struct miscdevice hlp_device = {
-	.minor = MISC_DYNAMIC_MINOR,
-	.name = "gspca_hlp",
-	.fops = &hlp_fops,
-};
-#endif
 
 /*
  * VMA operations.
@@ -331,10 +114,14 @@ static void fill_frame(struct gspca_dev *gspca_dev,
 			struct urb *urb)
 {
 	struct gspca_frame *frame;
-	unsigned char *data;	/* address of data in the iso message */
+	__u8 *data;		/* address of data in the iso message */
 	int i, j, len, st;
 	cam_pkt_op pkt_scan;
 
+	if (urb->status != 0) {
+		PDEBUG(D_ERR|D_PACK, "urb status: %d", urb->status);
+		return;		/* disconnection ? */
+	}
 	pkt_scan = gspca_dev->sd_desc->pkt_scan;
 	for (i = 0; i < urb->number_of_packets; i++) {
 
@@ -350,20 +137,21 @@ static void fill_frame(struct gspca_dev *gspca_dev,
 
 		/* check the packet status and length */
 		len = urb->iso_frame_desc[i].actual_length;
+		if (len == 0)
+			continue;
 		st = urb->iso_frame_desc[i].status;
 		if (st) {
-			PDEBUG(D_ERR, "ISOC data error: [%d] len=%d, status=%d",
+			PDEBUG(D_ERR,
+				"ISOC data error: [%d] len=%d, status=%d",
 				i, len, st);
 			gspca_dev->last_packet_type = DISCARD_PACKET;
 			continue;
 		}
-		if (len == 0)
-			continue;
 
 		/* let the packet be analyzed by the subdriver */
 		PDEBUG(D_PACK, "packet [%d] o:%d l:%d",
 			i, urb->iso_frame_desc[i].offset, len);
-		data = (unsigned char *) urb->transfer_buffer
+		data = (__u8 *) urb->transfer_buffer
 					+ urb->iso_frame_desc[i].offset;
 		pkt_scan(gspca_dev, frame, data, len);
 	}
@@ -390,7 +178,8 @@ static void fill_frame(struct gspca_dev *gspca_dev,
  *	  buffers are in user space (userptr). The frame detection
  *	  and copy is done by the application.
  */
-static void isoc_irq_mmap(struct urb *urb)
+static void isoc_irq_mmap(struct urb *urb
+)
 {
 	struct gspca_dev *gspca_dev = (struct gspca_dev *) urb->context;
 
@@ -400,7 +189,8 @@ static void isoc_irq_mmap(struct urb *urb)
 	fill_frame(gspca_dev, urb);
 }
 
-static void isoc_irq_user(struct urb *urb)
+static void isoc_irq_user(struct urb *urb
+)
 {
 	struct gspca_dev *gspca_dev = (struct gspca_dev *) urb->context;
 	int i;
@@ -459,7 +249,7 @@ static void isoc_transfer(struct gspca_dev *gspca_dev)
 struct gspca_frame *gspca_frame_add(struct gspca_dev *gspca_dev,
 				    int packet_type,
 				    struct gspca_frame *frame,
-				    unsigned char *data,
+				    __u8 *data,
 				    int len)
 {
 	int i, j;
@@ -503,23 +293,10 @@ struct gspca_frame *gspca_frame_add(struct gspca_dev *gspca_dev,
 	/* if last packet, wake the application and advance in the queue */
 	if (packet_type == LAST_PACKET) {
 		frame->v4l2_buf.bytesused = frame->data_end - frame->data;
-#ifndef GSPCA_HLP
 		frame->v4l2_buf.flags &= ~V4L2_BUF_FLAG_QUEUED;
 		frame->v4l2_buf.flags |= V4L2_BUF_FLAG_DONE;
 		atomic_inc(&gspca_dev->nevent);
 		wake_up_interruptible(&gspca_dev->wq);	/* event = new frame */
-#else /*GSPCA_HLP*/
-		if (hlp != 0 && hlp->gspca_dev == gspca_dev) {
-			frame->v4l2_buf.flags |= GSPCA_BUF_FLAG_DECODE;
-			atomic_inc(&hlp->nevent);
-			wake_up_interruptible(&hlp->wq);
-		} else {
-			frame->v4l2_buf.flags &= ~V4L2_BUF_FLAG_QUEUED;
-			frame->v4l2_buf.flags |= V4L2_BUF_FLAG_DONE;
-			atomic_inc(&gspca_dev->nevent);
-			wake_up_interruptible(&gspca_dev->wq);	/* new frame */
-		}
-#endif /*GSPCA_HLP*/
 		i = (gspca_dev->fr_i + 1) % gspca_dev->nframes;
 		gspca_dev->fr_i = i;
 		PDEBUG(D_FRAM, "frame complete len:%d q:%d i:%d o:%d",
@@ -581,13 +358,13 @@ static void rvfree(void *mem, unsigned long size)
 static __u32 get_v4l2_depth(__u32 pixfmt)
 {
 	switch (pixfmt) {
-	case V4L2_PIX_FMT_BGR32:
+/*	case V4L2_PIX_FMT_BGR32:
 	case V4L2_PIX_FMT_RGB32:
-		return 32;
+		return 32; */
 	case V4L2_PIX_FMT_RGB24:	/* 'RGB3' */
 	case V4L2_PIX_FMT_BGR24:
 		return 24;
-	case V4L2_PIX_FMT_RGB565:	/* 'RGBP' */
+/*	case V4L2_PIX_FMT_RGB565:	 * 'RGBP' */
 	case V4L2_PIX_FMT_YUYV:		/* 'YUYV' packed 4.2.2 */
 	case V4L2_PIX_FMT_YYUV:		/* 'YYUV' */
 		return 16;
@@ -596,6 +373,9 @@ static __u32 get_v4l2_depth(__u32 pixfmt)
 	case V4L2_PIX_FMT_MJPEG:
 	case V4L2_PIX_FMT_JPEG:
 	case V4L2_PIX_FMT_SBGGR8:	/* 'BA81' Bayer */
+	case V4L2_PIX_FMT_SN9C10X:	/* 'S910' SN9C10x compression */
+	case V4L2_PIX_FMT_SPCA501:	/* 'S501' YUYV per line */
+	case V4L2_PIX_FMT_SPCA561:	/* 'S561' compressed BGGR bayer */
 		return 8;
 	}
 	PDEBUG(D_ERR|D_CONF, "Unknown pixel format %c%c%c%c",
@@ -603,7 +383,7 @@ static __u32 get_v4l2_depth(__u32 pixfmt)
 		(pixfmt >> 8) & 0xff,
 		(pixfmt >> 16) & 0xff,
 		pixfmt >> 24);
-	return -EINVAL;
+	return 24;
 }
 
 static int gspca_get_buff_size(struct gspca_dev *gspca_dev)
@@ -632,7 +412,7 @@ static int frame_alloc(struct gspca_dev *gspca_dev,
 		count = GSPCA_MAX_FRAMES;
 	/* if compressed (JPEG), reduce the buffer size */
 	if (gspca_is_compressed(gspca_dev->pixfmt))
-		frsz = (frsz * comp_fac) / 100 + 600;	/* plus JPEG header */
+		frsz = (frsz * comp_fac) / 100 + 600; /* (+ JPEG header sz) */
 	frsz = PAGE_ALIGN(frsz);
 	PDEBUG(D_STREAM, "new fr_sz: %d", frsz);
 	gspca_dev->frsz = frsz;
@@ -660,17 +440,6 @@ static int frame_alloc(struct gspca_dev *gspca_dev,
 		}
 	}
 	gspca_dev->fr_i = gspca_dev->fr_o = gspca_dev->fr_q = 0;
-#ifdef GSPCA_HLP
-	{
-		struct hlp_dev *hlp_dev;
-
-		hlp_dev = hlp;
-		if (hlp != 0 && hlp_dev->gspca_dev == gspca_dev) {
-			hlp_dev->fr_d = 0;
-			atomic_set(&hlp_dev->nevent, 0);
-		}
-	}
-#endif /*GSPCA_HLP*/
 	gspca_dev->last_packet_type = DISCARD_PACKET;
 	gspca_dev->sequence = 0;
 	atomic_set(&gspca_dev->nevent, 0);
@@ -752,13 +521,14 @@ struct usb_host_endpoint *get_isoc_ep(struct gspca_dev *gspca_dev)
 	int i, ret;
 
 	intf = usb_ifnum_to_if(gspca_dev->dev, gspca_dev->iface);
+	ep = NULL;
 	i = gspca_dev->alt;			/* previous alt setting */
 	while (--i > 0) {			/* alt 0 is unusable */
 		ep = alt_isoc(&intf->altsetting[i], gspca_dev->cam.epaddr);
 		if (ep)
 			break;
 	}
-	if (i <= 0) {
+	if (ep == NULL) {
 		err("no ISOC endpoint found");
 		return NULL;
 	}
@@ -796,11 +566,14 @@ static int create_urbs(struct gspca_dev *gspca_dev,
 		"isoc %d pkts size %d (bsize:%d)", npkt, psize, bsize);
 /*fixme:change for userptr*/
 /*fixme:don't submit all URBs when userptr*/
-	gspca_dev->nurbs = nurbs = DEF_NURBS;
-	if (gspca_dev->memory == V4L2_MEMORY_MMAP)
+	if (gspca_dev->memory == V4L2_MEMORY_MMAP) {
 		usb_complete = isoc_irq_mmap;
-	else
+		nurbs = DEF_NURBS;
+	} else {
 		usb_complete = isoc_irq_user;
+		nurbs = USR_NURBS;
+	}
+	gspca_dev->nurbs = nurbs;
 	for (n = 0; n < nurbs; n++) {
 		urb = usb_alloc_urb(npkt, GFP_KERNEL);
 		if (!urb) {
@@ -904,16 +677,6 @@ static void gspca_stream_off(struct gspca_dev *gspca_dev)
 {
 	gspca_dev->streaming = 0;
 	atomic_set(&gspca_dev->nevent, 0);
-#ifdef GSPCA_HLP
-	{
-		struct hlp_dev *hlp_dev;
-
-		hlp_dev = hlp;
-		if (hlp_dev != 0
-		    && hlp_dev->gspca_dev == gspca_dev)
-			atomic_set(&hlp_dev->nevent, 0);
-	}
-#endif
 	if (gspca_dev->present) {
 		gspca_dev->sd_desc->stopN(gspca_dev);
 		destroy_urbs(gspca_dev);
@@ -979,15 +742,11 @@ static int vidioc_enum_fmt_cap(struct file *file, void  *priv,
 				struct v4l2_fmtdesc *fmtdesc)
 {
 	struct gspca_dev *gspca_dev = priv;
-	int i;
-#ifndef GSPCA_HLP
-	int j, index;
+	int i, j, index;
 	__u32 fmt_tb[8];
-#endif
 
 	PDEBUG(D_CONF, "enum fmt cap");
 
-#ifndef GSPCA_HLP
 	/* give an index to each format */
 	index = 0;
 	j = 0;
@@ -1013,36 +772,6 @@ static int vidioc_enum_fmt_cap(struct file *file, void  *priv,
 	fmtdesc->pixelformat = fmt_tb[index];
 	if (gspca_is_compressed(fmt_tb[index]))
 		fmtdesc->flags = V4L2_FMT_FLAG_COMPRESSED;
-#else /*GSPCA_HLP*/
-	/* !! code tied to the decoding functions in decoder.c */
-	i = gspca_dev->cam.nmodes - 1;
-	if (fmtdesc->index == 0) {	/* (assume one format per subdriver) */
-		fmtdesc->pixelformat = gspca_dev->cam.cam_mode[i].pixfmt;
-		if (gspca_is_compressed(fmtdesc->pixelformat))
-			fmtdesc->flags = V4L2_FMT_FLAG_COMPRESSED;
-	} else {
-		if (hlp == 0
-		    || (hlp->gspca_dev != 0
-			&& hlp->gspca_dev != gspca_dev))
-			return -EINVAL;
-		switch (gspca_dev->cam.cam_mode[i].pixfmt) {
-		case V4L2_PIX_FMT_JPEG:
-			if (fmtdesc->index >= sizeof jpeg_to_tb
-						 / sizeof jpeg_to_tb[0])
-				return -EINVAL;
-			fmtdesc->pixelformat = jpeg_to_tb[fmtdesc->index];
-			break;
-		case V4L2_PIX_FMT_SBGGR8:
-			if (fmtdesc->index >= sizeof bayer_to_tb
-						 / sizeof bayer_to_tb[0])
-				return -EINVAL;
-			fmtdesc->pixelformat = bayer_to_tb[fmtdesc->index];
-			break;
-		default:
-			return -EINVAL;
-		}
-	}
-#endif /*GSPCA_HLP*/
 	fmtdesc->type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
 	fmtdesc->description[0] = fmtdesc->pixelformat & 0xff;
 	fmtdesc->description[1] = (fmtdesc->pixelformat >> 8) & 0xff;
@@ -1057,25 +786,12 @@ static int vidioc_g_fmt_cap(struct file *file, void *priv,
 {
 	struct gspca_dev *gspca_dev = priv;
 
-#ifdef GSPCA_HLP
-	int i;
-
-	/* if the pixel format is not the one of the device and
-	 * if the helper is inactive or busy, restore */
-	i = gspca_dev->curr_mode;
-	if (gspca_dev->pixfmt != gspca_dev->cam.cam_mode[i].pixfmt) {
-		struct hlp_dev *hlp_dev;
-
-		hlp_dev = hlp;
-		if (hlp_dev == 0 || hlp_dev->gspca_dev != gspca_dev)
-			gspca_dev->pixfmt = gspca_dev->cam.cam_mode[i].pixfmt;
-	}
-#endif /*GSPCA_HLP*/
-
+	if (fmt->type != V4L2_BUF_TYPE_VIDEO_CAPTURE)
+		return -EINVAL;
 	fmt->fmt.pix.width = gspca_dev->width;
 	fmt->fmt.pix.height = gspca_dev->height;
 	fmt->fmt.pix.pixelformat = gspca_dev->pixfmt;
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 	if (gspca_debug & D_CONF) {
 		PDEBUG_MODE("get fmt cap",
 			fmt->fmt.pix.pixelformat,
@@ -1099,13 +815,15 @@ static int try_fmt_cap(struct gspca_dev *gspca_dev,
 {
 	int w, h, mode, mode2, frsz;
 
+	if (fmt->type != V4L2_BUF_TYPE_VIDEO_CAPTURE)
+		return -EINVAL;
 	w = fmt->fmt.pix.width;
 	h = fmt->fmt.pix.height;
 
 	/* (luvcview problem) */
 	if (fmt->fmt.pix.pixelformat == V4L2_PIX_FMT_MJPEG)
 		fmt->fmt.pix.pixelformat = V4L2_PIX_FMT_JPEG;
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 	if (gspca_debug & D_CONF)
 		PDEBUG_MODE("try fmt cap", fmt->fmt.pix.pixelformat, w, h);
 #endif
@@ -1121,44 +839,11 @@ static int try_fmt_cap(struct gspca_dev *gspca_dev,
 		if (mode2 >= 0) {
 			mode = mode2;
 		} else {
-			__u32 pixfmt;
-
-			pixfmt = gspca_dev->cam.cam_mode[mode].pixfmt;
-#ifndef GSPCA_HLP
 
 			/* no chance, return this mode */
-			fmt->fmt.pix.pixelformat = pixfmt;
-#else /*GSPCA_HLP*/
-			if (hlp != 0
-			    && (hlp->gspca_dev == 0
-				|| hlp->gspca_dev == gspca_dev)
-/* decoding works for JPEG and Bayer only */
-			    && (pixfmt == V4L2_PIX_FMT_JPEG
-				|| pixfmt == V4L2_PIX_FMT_SBGGR8)) {
-				switch (fmt->fmt.pix.pixelformat) {
-				case V4L2_PIX_FMT_YUYV:		/* 'YUYV' */
-				case V4L2_PIX_FMT_BGR24:	/* 'BGR3' */
-				case V4L2_PIX_FMT_RGB24:	/* 'RGB3' */
-				case V4L2_PIX_FMT_YUV420:	/* 'YU12' */
-				case V4L2_PIX_FMT_RGB565:	/* 'RGBP' */
-					break;
-				default: {
-					/* return any of the supported fmt's */
-					__u8 u;
-
-					u = get_jiffies_64();
-					u %= sizeof bayer_to_tb
-						/ sizeof bayer_to_tb[0] - 1;
-					fmt->fmt.pix.pixelformat =
-							bayer_to_tb[u + 1];
-					break;
-				    }
-				}
-			} else {
-				fmt->fmt.pix.pixelformat = pixfmt;
-			}
-#endif /*GSPCA_HLP*/
-#ifdef GSPCA_DEBUG
+			fmt->fmt.pix.pixelformat =
+					gspca_dev->cam.cam_mode[mode].pixfmt;
+#ifdef VIDEO_ADV_DEBUG
 			if (gspca_debug & D_CONF) {
 				PDEBUG_MODE("new format",
 					fmt->fmt.pix.pixelformat,
@@ -1198,7 +883,17 @@ static int vidioc_s_fmt_cap(struct file *file, void *priv,
 	struct gspca_dev *gspca_dev = priv;
 	int ret;
 
-#ifdef GSPCA_DEBUG
+#ifdef CONFIG_VIDEO_V4L1_COMPAT
+	/* if v4l1 got JPEG */
+	if (fmt->fmt.pix.pixelformat == 0
+	    && gspca_dev->streaming) {
+		fmt->fmt.pix.width = gspca_dev->width;
+		fmt->fmt.pix.height = gspca_dev->height;
+		fmt->fmt.pix.pixelformat = gspca_dev->pixfmt;
+		return 0;
+	}
+#endif
+#ifdef VIDEO_ADV_DEBUG
 	if (gspca_debug & D_CONF) {
 		PDEBUG_MODE("set fmt cap",
 			fmt->fmt.pix.pixelformat,
@@ -1218,14 +913,8 @@ static int vidioc_s_fmt_cap(struct file *file, void *priv,
 		goto out;
 	}
 
-#ifndef GSPCA_HLP
 	if (ret == gspca_dev->curr_mode)
 		goto out;			/* same mode */
-#else /*GSPCA_HLP*/
-	if (ret == gspca_dev->curr_mode
-	    && gspca_dev->pixfmt == fmt->fmt.pix.pixelformat)
-		goto out;			/* same mode */
-#endif /*GSPCA_HLP*/
 
 	if (gspca_dev->streaming) {
 		ret = -EBUSY;
@@ -1236,26 +925,6 @@ static int vidioc_s_fmt_cap(struct file *file, void *priv,
 	gspca_dev->pixfmt = fmt->fmt.pix.pixelformat;
 	gspca_dev->curr_mode = ret;
 
-#ifdef GSPCA_HLP
-	/* if frame decoding is required */
-	if (gspca_dev->pixfmt != gspca_dev->cam.cam_mode[ret].pixfmt) {
-		struct hlp_dev *hlp_dev;
-
-		hlp_dev = hlp;
-		if (hlp_dev == 0
-		    || (hlp_dev->gspca_dev != 0
-			&& hlp_dev->gspca_dev != gspca_dev)) { /* helper busy */
-			fmt->fmt.pix.pixelformat =
-				gspca_dev->pixfmt =
-					gspca_dev->cam.cam_mode[ret].pixfmt;
-		} else {				/* helper active */
-			hlp_dev->gspca_dev = gspca_dev;
-			hlp_dev->pixfmt = gspca_dev->cam.cam_mode[ret].pixfmt;
-			hlp_dev->fr_d = gspca_dev->fr_i;
-		}
-	} else if (hlp != 0 && hlp->gspca_dev == gspca_dev)
-		hlp->gspca_dev = 0;
-#endif /*GSPCA_HLP*/
 	ret = 0;
 out:
 	mutex_unlock(&gspca_dev->queue_lock);
@@ -1294,7 +963,7 @@ static int dev_open(struct inode *inode, struct file *file)
 	}
 	gspca_dev->users++;
 	file->private_data = gspca_dev;
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 	/* activate the v4l2 debug */
 	if (gspca_debug & D_V4L2)
 		gspca_dev->vdev.debug |= 3;
@@ -1329,22 +998,6 @@ static int dev_close(struct inode *inode, struct file *file)
 		frame_free(gspca_dev);
 		gspca_dev->capt_file = 0;
 		gspca_dev->memory = GSPCA_MEMORY_NO;
-#ifdef GSPCA_HLP
-		{
-			struct hlp_dev *hlp_dev;
-			int mode;
-
-			hlp_dev = hlp;
-			if (hlp_dev != 0
-			    && hlp_dev->gspca_dev == gspca_dev) {
-				hlp_dev->gspca_dev = 0;
-				hlp_dev->frame = 0;
-				mode = gspca_dev->curr_mode;
-				gspca_dev->pixfmt =
-					gspca_dev->cam.cam_mode[mode].pixfmt;
-			}
-		}
-#endif /*GSPCA_HLP*/
 	}
 	file->private_data = NULL;
 	mutex_unlock(&gspca_dev->queue_lock);
@@ -1370,23 +1023,38 @@ static int vidioc_querycap(struct file *file, void  *priv,
 	return 0;
 }
 
+/* the use of V4L2_CTRL_FLAG_NEXT_CTRL asks for the controls to be sorted */
 static int vidioc_queryctrl(struct file *file, void *priv,
 			   struct v4l2_queryctrl *q_ctrl)
 {
 	struct gspca_dev *gspca_dev = priv;
 	int i;
-
-	PDEBUG(D_CONF, "queryctrl");
+	u32 id;
+
+	id = q_ctrl->id;
+	if (id & V4L2_CTRL_FLAG_NEXT_CTRL) {
+		id &= V4L2_CTRL_ID_MASK;
+		id++;
+		for (i = 0; i < gspca_dev->sd_desc->nctrls; i++) {
+			if (id >= gspca_dev->sd_desc->ctrls[i].qctrl.id) {
+				memcpy(q_ctrl,
+					&gspca_dev->sd_desc->ctrls[i].qctrl,
+					sizeof *q_ctrl);
+				return 0;
+			}
+		}
+		return -EINVAL;
+	}
 	for (i = 0; i < gspca_dev->sd_desc->nctrls; i++) {
-		if (q_ctrl->id == gspca_dev->sd_desc->ctrls[i].qctrl.id) {
+		if (id == gspca_dev->sd_desc->ctrls[i].qctrl.id) {
 			memcpy(q_ctrl,
 				&gspca_dev->sd_desc->ctrls[i].qctrl,
 				sizeof *q_ctrl);
 			return 0;
 		}
 	}
-	if (q_ctrl->id >= V4L2_CID_BASE
-	    && q_ctrl->id <= V4L2_CID_LASTP1) {
+	if (id >= V4L2_CID_BASE
+	    && id <= V4L2_CID_LASTP1) {
 		q_ctrl->flags |= V4L2_CTRL_FLAG_DISABLED;
 		return 0;
 	}
@@ -1489,13 +1157,8 @@ static int vidioc_reqbufs(struct file *file, void *priv,
 		return -EINVAL;
 	switch (rb->memory) {
 	case V4L2_MEMORY_MMAP:
-		break;
 	case V4L2_MEMORY_USERPTR:
-#ifdef GSPCA_HLP
-		if (hlp == 0 || hlp->gspca_dev != gspca_dev)
-			break;
-#endif
-		return -EINVAL;
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -1578,7 +1241,7 @@ static int vidioc_streamon(struct file *file, void *priv,
 		if (ret < 0)
 			goto out;
 	}
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 	if (gspca_debug & D_STREAM) {
 		PDEBUG_MODE("stream on OK",
 			gspca_dev->pixfmt,
@@ -1657,7 +1320,7 @@ static int vidioc_g_parm(struct file *filp, void *priv,
 {
 	struct gspca_dev *gspca_dev = priv;
 
-	memset(parm, 0, sizeof parm);
+	memset(parm, 0, sizeof *parm);
 	parm->type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
 	parm->parm.capture.readbuffers = gspca_dev->nbufread;
 	return 0;
@@ -1677,6 +1340,12 @@ static int vidioc_s_parm(struct file *filp, void *priv,
 	return 0;
 }
 
+static int vidioc_s_std(struct file *filp, void *priv,
+			v4l2_std_id *parm)
+{
+	return 0;
+}
+
 #ifdef CONFIG_VIDEO_V4L1_COMPAT
 static int vidiocgmbuf(struct file *file, void *priv,
 			struct video_mbuf *mbuf)
@@ -1686,29 +1355,32 @@ static int vidiocgmbuf(struct file *file, void *priv,
 
 	PDEBUG(D_STREAM, "cgmbuf");
 	if (gspca_dev->nframes == 0) {
-		struct v4l2_requestbuffers rb;
 		int ret;
-		__u32 pixfmt;
-		short width, height;
-
-		/* as the final format is not yet defined, allocate
-		   buffers with the max size */
-		pixfmt = gspca_dev->pixfmt;
-		width = gspca_dev->width;
-		height = gspca_dev->height;
-		gspca_dev->pixfmt = V4L2_PIX_FMT_BGR32;
-		gspca_dev->width = 640;
-		gspca_dev->height = 480;
-		memset(&rb, 0, sizeof rb);
-		rb.count = 4;
-		rb.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
-		rb.memory = V4L2_MEMORY_MMAP;
-		ret = vidioc_reqbufs(file, priv, &rb);
-		gspca_dev->pixfmt = pixfmt;
-		gspca_dev->width = width;
-		gspca_dev->height = height;
-		if (ret != 0)
-			return ret;
+
+		{
+			struct v4l2_format fmt;
+
+			memset(&fmt, 0, sizeof fmt);
+			fmt.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+			i = gspca_dev->cam.nmodes - 1;	/* highest mode */
+			fmt.fmt.pix.width = gspca_dev->cam.cam_mode[i].width;
+			fmt.fmt.pix.height = gspca_dev->cam.cam_mode[i].height;
+			fmt.fmt.pix.pixelformat = V4L2_PIX_FMT_BGR24;
+			ret = vidioc_s_fmt_cap(file, priv, &fmt);
+			if (ret != 0)
+				return ret;
+		}
+		{
+			struct v4l2_requestbuffers rb;
+
+			memset(&rb, 0, sizeof rb);
+			rb.count = 4;
+			rb.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+			rb.memory = V4L2_MEMORY_MMAP;
+			ret = vidioc_reqbufs(file, priv, &rb);
+			if (ret != 0)
+				return ret;
+		}
 	}
 	mbuf->frames = gspca_dev->nframes;
 	mbuf->size = gspca_dev->frsz * gspca_dev->nframes;
@@ -1951,7 +1623,7 @@ static int vidioc_qbuf(struct file *file, void *priv,
 
 	if (frame->v4l2_buf.memory == V4L2_MEMORY_USERPTR) {
 		frame->data = frame->data_end =
-				(unsigned char *) v4l2_buf->m.userptr;
+				(__u8 *) v4l2_buf->m.userptr;
 		frame->v4l2_buf.m.userptr = v4l2_buf->m.userptr;
 		frame->v4l2_buf.length = v4l2_buf->length;
 	}
@@ -2154,6 +1826,9 @@ static struct file_operations dev_fops = {
 	.read = dev_read,
 	.mmap = dev_mmap,
 	.ioctl = video_ioctl2,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = v4l_compat_ioctl32,
+#endif
 	.llseek = no_llseek,
 	.poll	= dev_poll,
 };
@@ -2186,6 +1861,7 @@ static struct video_device gspca_template = {
 	.vidioc_s_jpegcomp	= vidioc_s_jpegcomp,
 	.vidioc_g_parm		= vidioc_g_parm,
 	.vidioc_s_parm		= vidioc_s_parm,
+	.vidioc_s_std		= vidioc_s_std,
 #ifdef CONFIG_VIDEO_V4L1_COMPAT
 	.vidiocgmbuf          = vidiocgmbuf,
 #endif
@@ -2207,12 +1883,8 @@ int gspca_dev_probe(struct usb_interface *intf,
 	struct gspca_dev *gspca_dev;
 	struct usb_device *dev = interface_to_usbdev(intf);
 	int ret;
-	__u16 vendor;
-	__u16 product;
 
-	vendor = id->idVendor;
-	product = id->idProduct;
-	PDEBUG(D_PROBE, "probing %04x:%04x", vendor, product);
+	PDEBUG(D_PROBE, "probing %04x:%04x", id->idVendor, id->idProduct);
 
 	/* we don't handle multi-config cameras */
 	if (dev->descriptor.bNumConfigurations != 1)
@@ -2309,35 +1981,24 @@ EXPORT_SYMBOL(gspca_disconnect);
 /* -- module insert / remove -- */
 static int __init gspca_init(void)
 {
-#ifdef GSPCA_HLP
-	int ret;
-
-	/* create /dev/gspca_hlp */
-	ret = misc_register(&hlp_device);
-	if (ret < 0)
-		err("misc_register err %d", ret);
-	start_hlp();		/* try to start the helper process */
-#endif
 	info("main v%s registered", version);
 	return 0;
 }
 static void __exit gspca_exit(void)
 {
-#ifdef GSPCA_HLP
-	misc_deregister(&hlp_device);
-#endif
 	info("main deregistered");
 }
 
 module_init(gspca_init);
 module_exit(gspca_exit);
 
+#ifdef VIDEO_ADV_DEBUG
 module_param_named(debug, gspca_debug, int, 0644);
 MODULE_PARM_DESC(debug,
 		"Debug (bit) 0x01:error 0x02:probe 0x04:config"
 		" 0x08:stream 0x10:frame 0x20:packet 0x40:USBin 0x80:USBout"
 		" 0x0100: v4l2");
-
+#endif
 module_param(comp_fac, int, 0644);
 MODULE_PARM_DESC(comp_fac,
 		"Buffer size ratio when compressed in percent");
diff --git a/drivers/media/video/gspca/gspca.h b/drivers/media/video/gspca/gspca.h
index e69d8472a284..1581fa808b6f 100644
--- a/drivers/media/video/gspca/gspca.h
+++ b/drivers/media/video/gspca/gspca.h
@@ -9,7 +9,26 @@
 #include <media/v4l2-common.h>
 #include <linux/mutex.h>
 
-#ifdef GSPCA_DEBUG
+/* values in 2.6.27 */
+#ifndef V4L2_PIX_FMT_SPCA501
+#define V4L2_PIX_FMT_SPCA501 v4l2_fourcc('S', '5', '0', '1')
+#endif
+#ifndef V4L2_PIX_FMT_SPCA561
+#define V4L2_PIX_FMT_SPCA561 v4l2_fourcc('S', '5', '6', '1')
+#endif
+
+/* values in 2.6.26 */
+#ifndef V4L2_CID_POWER_LINE_FREQUENCY
+#define V4L2_CID_POWER_LINE_FREQUENCY  (V4L2_CID_BASE+24)
+#endif
+#ifndef V4L2_CID_WHITE_BALANCE_TEMPERATURE
+#define V4L2_CID_WHITE_BALANCE_TEMPERATURE (V4L2_CID_BASE + 26)
+#endif
+#ifndef V4L2_CID_SHARPNESS
+#define V4L2_CID_SHARPNESS  (V4L2_CID_BASE+27)
+#endif
+
+#ifdef VIDEO_ADV_DEBUG
 /* GSPCA our debug messages */
 extern int gspca_debug;
 #define PDEBUG(level, fmt, args...) \
@@ -47,7 +66,7 @@ extern int gspca_debug;
 
 #define GSPCA_MAX_FRAMES 16	/* maximum number of video frame buffers */
 /* ISOC transfers */
-#define MAX_NURBS 32		/* max number of URBs (read & userptr) */
+#define MAX_NURBS 16		/* max number of URBs */
 #define ISO_MAX_PKT 32		/* max number of packets in an ISOC transfer */
 #define ISO_MAX_SIZE 0x8000	/* max size of one URB buffer (32 Kb) */
 
@@ -79,7 +98,7 @@ typedef int (*cam_qmnu_op) (struct gspca_dev *,
 			struct v4l2_querymenu *);
 typedef void (*cam_pkt_op) (struct gspca_dev *gspca_dev,
 				struct gspca_frame *frame,
-				unsigned char *data,
+				__u8 *data,
 				int len);
 
 struct ctrl {
@@ -116,8 +135,8 @@ struct sd_desc {
 #define LAST_PACKET	3
 
 struct gspca_frame {
-	unsigned char *data;		/* frame buffer */
-	unsigned char *data_end;	/* end of frame while filling */
+	__u8 *data;			/* frame buffer */
+	__u8 *data_end;			/* end of frame while filling */
 	int vma_use_count;
 	struct v4l2_buffer v4l2_buf;
 };
@@ -135,7 +154,7 @@ struct gspca_dev {
 
 	__u8 *frbuf;				/* buffer for nframes */
 	struct gspca_frame frame[GSPCA_MAX_FRAMES];
-	unsigned int frsz;			/* frame size */
+	__u32 frsz;				/* frame size */
 	char nframes;				/* number of frames */
 	char fr_i;				/* frame being filled */
 	char fr_q;				/* next frame to queue */
@@ -145,10 +164,10 @@ struct gspca_dev {
 
 	__u8 iface;			/* USB interface number */
 	__u8 alt;			/* USB alternate setting */
-	unsigned char curr_mode;	/* current camera mode */
+	__u8 curr_mode;			/* current camera mode */
 	__u32 pixfmt;			/* current mode parameters */
-	short width;
-	short height;
+	__u16 width;
+	__u16 height;
 
 	atomic_t nevent;		/* number of frames done */
 	wait_queue_head_t wq;		/* wait queue */
@@ -176,6 +195,6 @@ void gspca_disconnect(struct usb_interface *intf);
 struct gspca_frame *gspca_frame_add(struct gspca_dev *gspca_dev,
 				    int packet_type,
 				    struct gspca_frame *frame,
-				    unsigned char *data,
+				    __u8 *data,
 				    int len);
 #endif /* GSPCAV2_H */
diff --git a/drivers/media/video/gspca/mars.c b/drivers/media/video/gspca/mars.c
new file mode 100644
index 000000000000..48b861d68299
--- /dev/null
+++ b/drivers/media/video/gspca/mars.c
@@ -0,0 +1,455 @@
+/*
+ *		Mars-Semi MR97311A library
+ *		Copyright (C) 2005 <bradlch@hotmail.com>
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "mars"
+
+#include "gspca.h"
+#include "jpeg.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/Mars USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	char qindex;
+};
+
+/* V4L2 controls supported by the driver */
+static struct ctrl sd_ctrls[] = {
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 320, 240, 2},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 1},
+};
+
+/* MI Register table //elvis */
+enum {
+	REG_HW_MI_0,
+	REG_HW_MI_1,
+	REG_HW_MI_2,
+	REG_HW_MI_3,
+	REG_HW_MI_4,
+	REG_HW_MI_5,
+	REG_HW_MI_6,
+	REG_HW_MI_7,
+	REG_HW_MI_9 = 0x09,
+	REG_HW_MI_B = 0x0B,
+	REG_HW_MI_C,
+	REG_HW_MI_D,
+	REG_HW_MI_1E = 0x1E,
+	REG_HW_MI_20 = 0x20,
+	REG_HW_MI_2B = 0x2B,
+	REG_HW_MI_2C,
+	REG_HW_MI_2D,
+	REG_HW_MI_2E,
+	REG_HW_MI_35 = 0x35,
+	REG_HW_MI_5F = 0x5f,
+	REG_HW_MI_60,
+	REG_HW_MI_61,
+	REG_HW_MI_62,
+	REG_HW_MI_63,
+	REG_HW_MI_64,
+	REG_HW_MI_F1 = 0xf1,
+	ATTR_TOTAL_MI_REG = 242
+};
+
+static int pcam_reg_write(struct usb_device *dev,
+			  __u16 index, unsigned char *value, int length)
+{
+	int rc;
+
+	rc = usb_control_msg(dev,
+			 usb_sndbulkpipe(dev, 4),
+			 0x12,
+/* ?? 0xc8 = USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_??? !? */
+			 0xc8,
+			 0,		/* value */
+			 index, value, length, 500);
+	PDEBUG(D_USBO, "reg write: 0x%02X , result = 0x%x", index, rc);
+
+	if (rc < 0)
+		PDEBUG(D_ERR, "reg write: error %d", rc);
+	return rc;
+}
+
+static void MISensor_BulkWrite(struct usb_device *dev, unsigned short *pch,
+				   char Address)
+{
+	int result;
+	unsigned char data[6];
+
+	data[0] = 0x1f;
+	data[1] = 0;
+	data[2] = Address;
+	data[3] = *pch >> 8;		/* high byte */
+	data[4] = *pch;			/* low byte */
+	data[5] = 0;
+
+	result = usb_control_msg(dev,
+				 usb_sndbulkpipe(dev, 4),
+				 0x12,
+/* ?? 0xc8 = USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_??? !? */
+				 0xc8,
+				 0,		/* value */
+				 Address,	/* index */
+				 data, 5, 500);
+	PDEBUG(D_USBO, "bulk write 0x%02x = 0x%04x", Address, *pch);
+
+	if (result < 0)
+		PDEBUG(D_ERR, "reg write: error %d", result);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	cam->cam_mode = vga_mode;
+	cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	sd->qindex = 1;			/* set the quantization table */
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int err_code;
+	__u8 data[12];
+	__u16 MI_buf[242];
+	int h_size, v_size;
+	int intpipe;
+	/* struct usb_device *dev = pcam->dev; */
+	memset(data, 0, sizeof data);
+	memset(MI_buf, 0, sizeof MI_buf);
+
+	PDEBUG(D_STREAM, "camera start, iface %d, alt 8", gspca_dev->iface);
+	if (usb_set_interface(dev, gspca_dev->iface, 8) < 0) {
+		PDEBUG(D_ERR|D_STREAM, "Set packet size: set interface error");
+		return;
+	}
+
+	data[0] = 0x01;		/* address */
+	data[1] = 0x01;
+
+	err_code = pcam_reg_write(dev, data[0], data, 0x02);
+	if (err_code < 0)
+		return;
+
+	/*
+	   Initialize the MR97113 chip register
+	 */
+	data[0] = 0x00;		/* address */
+	data[1] = 0x0c | 0x01;	/* reg 0 */
+	data[2] = 0x01;		/* reg 1 */
+	h_size = gspca_dev->width;
+	v_size = gspca_dev->height;
+	data[3] = h_size / 8;	/* h_size , reg 2 */
+	data[4] = v_size / 8;	/* v_size , reg 3 */
+	data[5] = 0x30;		/* reg 4, MI, PAS5101 :
+				 *	0x30 for 24mhz , 0x28 for 12mhz */
+	data[6] = 4;		/* reg 5, H start */
+	data[7] = 0xc0;		/* reg 6, gamma 1.5 */
+	data[8] = 3;		/* reg 7, V start */
+/*	if(h_size == 320 ) */
+/*		data[9]= 0x56;	 * reg 8, 24MHz, 2:1 scale down */
+/*	else */
+	data[9] = 0x52;		/* reg 8, 24MHz, no scale down */
+	data[10] = 0x5d;	/* reg 9, I2C device address
+				 *	[for PAS5101 (0x40)] [for MI (0x5d)] */
+
+	err_code = pcam_reg_write(dev, data[0], data, 0x0b);
+	if (err_code < 0)
+		return;
+
+	data[0] = 0x23;		/* address */
+	data[1] = 0x09;		/* reg 35, append frame header */
+
+	err_code = pcam_reg_write(dev, data[0], data, 0x02);
+	if (err_code < 0) {
+		PDEBUG(D_ERR, "Register write failed");
+		return;
+	}
+
+	data[0] = 0x3C;		/* address */
+/*	if (pcam->width == 1280) */
+/*		data[1] = 200;	 * reg 60, pc-cam frame size
+				 *	(unit: 4KB) 800KB */
+/*	else */
+	data[1] = 50;		/* 50 reg 60, pc-cam frame size
+				 *	(unit: 4KB) 200KB */
+	err_code = pcam_reg_write(dev, data[0], data, 0x02);
+	if (err_code < 0)
+		return;
+
+	if (0) {			/* fixed dark-gain */
+		data[1] = 0;		/* reg 94, Y Gain (1.75) */
+		data[2] = 0;		/* reg 95, UV Gain (1.75) */
+		data[3] = 0x3f;		/* reg 96, Y Gain/UV Gain/disable auto dark-gain */
+		data[4] = 0;		/* reg 97, set fixed dark level */
+		data[5] = 0;		/* reg 98, don't care */
+	} else {			/* auto dark-gain */
+		data[1] = 0;		/* reg 94, Y Gain (auto) */
+		data[2] = 0;		/* reg 95, UV Gain (1.75) */
+		data[3] = 0x78;		/* reg 96, Y Gain/UV Gain/disable auto dark-gain */
+		switch (gspca_dev->width) {
+/*		case 1280: */
+/*			data[4] = 154;
+				 * reg 97, %3 shadow point (unit: 256 pixel) */
+/*			data[5] = 51;
+				 * reg 98, %1 highlight point
+				 *	(uint: 256 pixel) */
+/*			break; */
+		default:
+/*		case 640: */
+			data[4] = 36;	/* reg 97, %3 shadow point
+					 *	(unit: 256 pixel) */
+			data[5] = 12;	/* reg 98, %1 highlight point
+					 *	(uint: 256 pixel) */
+			break;
+		case 320:
+			data[4] = 9;	/* reg 97, %3 shadow point
+					 *	(unit: 256 pixel) */
+			data[5] = 3;	/* reg 98, %1 highlight point
+					 *	(uint: 256 pixel) */
+			break;
+		}
+	}
+	/* auto dark-gain */
+	data[0] = 0x5e;		/* address */
+
+	err_code = pcam_reg_write(dev, data[0], data, 0x06);
+	if (err_code < 0)
+		return;
+
+	data[0] = 0x67;
+	data[1] = 0x13;		/* reg 103, first pixel B, disable sharpness */
+	err_code = pcam_reg_write(dev, data[0], data, 0x02);
+	if (err_code < 0)
+		return;
+
+	/*
+	 * initialize the value of MI sensor...
+	 */
+	MI_buf[REG_HW_MI_1] = 0x000a;
+	MI_buf[REG_HW_MI_2] = 0x000c;
+	MI_buf[REG_HW_MI_3] = 0x0405;
+	MI_buf[REG_HW_MI_4] = 0x0507;
+	/* mi_Attr_Reg_[REG_HW_MI_5]	 = 0x01ff;//13 */
+	MI_buf[REG_HW_MI_5] = 0x0013;	/* 13 */
+	MI_buf[REG_HW_MI_6] = 0x001f;	/* vertical blanking */
+	/* mi_Attr_Reg_[REG_HW_MI_6]	 = 0x0400;  // vertical blanking */
+	MI_buf[REG_HW_MI_7] = 0x0002;
+	/* mi_Attr_Reg_[REG_HW_MI_9]	 = 0x015f; */
+	/* mi_Attr_Reg_[REG_HW_MI_9]	 = 0x030f; */
+	MI_buf[REG_HW_MI_9] = 0x0374;
+	MI_buf[REG_HW_MI_B] = 0x0000;
+	MI_buf[REG_HW_MI_C] = 0x0000;
+	MI_buf[REG_HW_MI_D] = 0x0000;
+	MI_buf[REG_HW_MI_1E] = 0x8000;
+/* mi_Attr_Reg_[REG_HW_MI_20]	  = 0x1104; */
+	MI_buf[REG_HW_MI_20] = 0x1104;	/* 0x111c; */
+	MI_buf[REG_HW_MI_2B] = 0x0008;
+/* mi_Attr_Reg_[REG_HW_MI_2C]	  = 0x000f; */
+	MI_buf[REG_HW_MI_2C] = 0x001f;	/* lita suggest */
+	MI_buf[REG_HW_MI_2D] = 0x0008;
+	MI_buf[REG_HW_MI_2E] = 0x0008;
+	MI_buf[REG_HW_MI_35] = 0x0051;
+	MI_buf[REG_HW_MI_5F] = 0x0904;	/* fail to write */
+	MI_buf[REG_HW_MI_60] = 0x0000;
+	MI_buf[REG_HW_MI_61] = 0x0000;
+	MI_buf[REG_HW_MI_62] = 0x0498;
+	MI_buf[REG_HW_MI_63] = 0x0000;
+	MI_buf[REG_HW_MI_64] = 0x0000;
+	MI_buf[REG_HW_MI_F1] = 0x0001;
+	/* changing while setting up the different value of dx/dy */
+
+	if (gspca_dev->width != 1280) {
+		MI_buf[0x01] = 0x010a;
+		MI_buf[0x02] = 0x014c;
+		MI_buf[0x03] = 0x01e5;
+		MI_buf[0x04] = 0x0287;
+	}
+	MI_buf[0x20] = 0x1104;
+
+	MISensor_BulkWrite(dev, MI_buf + 1, 1);
+	MISensor_BulkWrite(dev, MI_buf + 2, 2);
+	MISensor_BulkWrite(dev, MI_buf + 3, 3);
+	MISensor_BulkWrite(dev, MI_buf + 4, 4);
+	MISensor_BulkWrite(dev, MI_buf + 5, 5);
+	MISensor_BulkWrite(dev, MI_buf + 6, 6);
+	MISensor_BulkWrite(dev, MI_buf + 7, 7);
+	MISensor_BulkWrite(dev, MI_buf + 9, 9);
+	MISensor_BulkWrite(dev, MI_buf + 0x0b, 0x0b);
+	MISensor_BulkWrite(dev, MI_buf + 0x0c, 0x0c);
+	MISensor_BulkWrite(dev, MI_buf + 0x0d, 0x0d);
+	MISensor_BulkWrite(dev, MI_buf + 0x1e, 0x1e);
+	MISensor_BulkWrite(dev, MI_buf + 0x20, 0x20);
+	MISensor_BulkWrite(dev, MI_buf + 0x2b, 0x2b);
+	MISensor_BulkWrite(dev, MI_buf + 0x2c, 0x2c);
+	MISensor_BulkWrite(dev, MI_buf + 0x2d, 0x2d);
+	MISensor_BulkWrite(dev, MI_buf + 0x2e, 0x2e);
+	MISensor_BulkWrite(dev, MI_buf + 0x35, 0x35);
+	MISensor_BulkWrite(dev, MI_buf + 0x5f, 0x5f);
+	MISensor_BulkWrite(dev, MI_buf + 0x60, 0x60);
+	MISensor_BulkWrite(dev, MI_buf + 0x61, 0x61);
+	MISensor_BulkWrite(dev, MI_buf + 0x62, 0x62);
+	MISensor_BulkWrite(dev, MI_buf + 0x63, 0x63);
+	MISensor_BulkWrite(dev, MI_buf + 0x64, 0x64);
+	MISensor_BulkWrite(dev, MI_buf + 0xf1, 0xf1);
+
+	intpipe = usb_sndintpipe(dev, 0);
+	err_code = usb_clear_halt(dev, intpipe);
+
+	data[0] = 0x00;
+	data[1] = 0x4d;		/* ISOC transfering enable... */
+	pcam_reg_write(dev, data[0], data, 0x02);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	int result;
+	__u8 data[2];
+
+	data[0] = 1;
+	data[1] = 0;
+	result = pcam_reg_write(gspca_dev->dev, data[0], data, 2);
+	if (result < 0)
+		PDEBUG(D_ERR, "Camera Stop failed");
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int p;
+
+	if (len < 6) {
+/*		gspca_dev->last_packet_type = DISCARD_PACKET; */
+		return;
+	}
+	for (p = 0; p < len - 6; p++) {
+		if (data[0 + p] == 0xff
+		    && data[1 + p] == 0xff
+		    && data[2 + p] == 0x00
+		    && data[3 + p] == 0xff
+		    && data[4 + p] == 0x96) {
+			if (data[5 + p] == 0x64
+			    || data[5 + p] == 0x65
+			    || data[5 + p] == 0x66
+			    || data[5 + p] == 0x67) {
+				PDEBUG(D_PACK, "sof offset: %d leng: %d",
+					p, len);
+				frame = gspca_frame_add(gspca_dev, LAST_PACKET,
+							frame, data, 0);
+
+				/* put the JPEG header */
+				jpeg_put_header(gspca_dev, frame,
+						sd->qindex, 0x21);
+				data += 16;
+				len -= 16;
+				break;
+			}
+		}
+	}
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x093a, 0x050f), DVNM("Mars-Semi Pc-Camera")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
new file mode 100644
index 000000000000..7d6237f18ba0
--- /dev/null
+++ b/drivers/media/video/gspca/ov519.c
@@ -0,0 +1,2174 @@
+/**
+ * OV519 driver
+ *
+ * Copyright (C) 2008 Jean-Francois Moine (http://moinejf.free.fr)
+ *
+ * (This module is adapted from the ov51x-jpeg package)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+#define MODULE_NAME "ov519"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Jean-Francois Moine <http://moinejf.free.fr>");
+MODULE_DESCRIPTION("OV519 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* global parameters */
+static int frame_rate;
+
+/* Number of times to retry a failed I2C transaction. Increase this if you
+ * are getting "Failed to read sensor ID..." */
+static int i2c_detect_tries = 10;
+
+/* ov519 device descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;		/* !! must be the first item */
+
+	/* Determined by sensor type */
+	short maxwidth;
+	short maxheight;
+
+	unsigned char primary_i2c_slave;	/* I2C write id of sensor */
+
+	unsigned char brightness;
+	unsigned char contrast;
+	unsigned char colors;
+
+	char compress;		/* Should the next frame be compressed? */
+	char compress_inited;	/* Are compression params uploaded? */
+	char stopped;		/* Streaming is temporarily paused */
+
+	char frame_rate;	/* current Framerate (OV519 only) */
+	char clockdiv;		/* clockdiv override for OV519 only */
+
+	char sensor;		/* Type of image sensor chip (SEN_*) */
+#define SEN_UNKNOWN 0
+#define SEN_OV6620 1
+#define SEN_OV6630 2
+#define SEN_OV7610 3
+#define SEN_OV7620 4
+#define SEN_OV7630 5
+#define SEN_OV7640 6
+#define SEN_OV7670 7
+#define SEN_OV76BE 8
+#define SEN_OV8610 9
+
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define SD_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Saturation",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 320, 240},
+	{V4L2_PIX_FMT_JPEG, 640, 480},
+};
+static struct cam_mode sif_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 176, 144},
+	{V4L2_PIX_FMT_JPEG, 352, 288},
+};
+
+/* OV519 Camera interface register numbers */
+#define OV519_CAM_H_SIZE		0x10
+#define OV519_CAM_V_SIZE		0x11
+#define OV519_CAM_X_OFFSETL		0x12
+#define OV519_CAM_X_OFFSETH		0x13
+#define OV519_CAM_Y_OFFSETL		0x14
+#define OV519_CAM_Y_OFFSETH		0x15
+#define OV519_CAM_DIVIDER		0x16
+#define OV519_CAM_DFR			0x20
+#define OV519_CAM_FORMAT		0x25
+
+/* OV519 System Controller register numbers */
+#define OV519_SYS_RESET1 0x51
+#define OV519_SYS_EN_CLK1 0x54
+
+#define OV519_GPIO_DATA_OUT0		0x71
+#define OV519_GPIO_IO_CTRL0		0x72
+
+#define OV511_ENDPOINT_ADDRESS  1	/* Isoc endpoint number */
+
+/* I2C registers */
+#define R51x_I2C_W_SID		0x41
+#define R51x_I2C_SADDR_3	0x42
+#define R51x_I2C_SADDR_2	0x43
+#define R51x_I2C_R_SID		0x44
+#define R51x_I2C_DATA		0x45
+#define R518_I2C_CTL		0x47	/* OV518(+) only */
+
+/* I2C ADDRESSES */
+#define OV7xx0_SID   0x42
+#define OV8xx0_SID   0xa0
+#define OV6xx0_SID   0xc0
+
+/* OV7610 registers */
+#define OV7610_REG_GAIN		0x00	/* gain setting (5:0) */
+#define OV7610_REG_SAT		0x03	/* saturation */
+#define OV8610_REG_HUE		0x04	/* 04 reserved */
+#define OV7610_REG_CNT		0x05	/* Y contrast */
+#define OV7610_REG_BRT		0x06	/* Y brightness */
+#define OV7610_REG_COM_C	0x14	/* misc common regs */
+#define OV7610_REG_ID_HIGH	0x1c	/* manufacturer ID MSB */
+#define OV7610_REG_ID_LOW	0x1d	/* manufacturer ID LSB */
+#define OV7610_REG_COM_I	0x29	/* misc settings */
+
+/* OV7670 registers */
+#define OV7670_REG_GAIN        0x00    /* Gain lower 8 bits (rest in vref) */
+#define OV7670_REG_BLUE        0x01    /* blue gain */
+#define OV7670_REG_RED         0x02    /* red gain */
+#define OV7670_REG_VREF        0x03    /* Pieces of GAIN, VSTART, VSTOP */
+#define OV7670_REG_COM1        0x04    /* Control 1 */
+#define OV7670_REG_AECHH       0x07    /* AEC MS 5 bits */
+#define OV7670_REG_COM3        0x0c    /* Control 3 */
+#define OV7670_REG_COM4        0x0d    /* Control 4 */
+#define OV7670_REG_COM5        0x0e    /* All "reserved" */
+#define OV7670_REG_COM6        0x0f    /* Control 6 */
+#define OV7670_REG_AECH        0x10    /* More bits of AEC value */
+#define OV7670_REG_CLKRC       0x11    /* Clock control */
+#define OV7670_REG_COM7        0x12    /* Control 7 */
+#define   OV7670_COM7_FMT_VGA    0x00
+#define   OV7670_COM7_YUV        0x00    /* YUV */
+#define   OV7670_COM7_FMT_QVGA   0x10    /* QVGA format */
+#define   OV7670_COM7_FMT_MASK   0x38
+#define   OV7670_COM7_RESET      0x80    /* Register reset */
+#define OV7670_REG_COM8        0x13    /* Control 8 */
+#define   OV7670_COM8_AEC        0x01    /* Auto exposure enable */
+#define   OV7670_COM8_AWB        0x02    /* White balance enable */
+#define   OV7670_COM8_AGC        0x04    /* Auto gain enable */
+#define   OV7670_COM8_BFILT      0x20    /* Band filter enable */
+#define   OV7670_COM8_AECSTEP    0x40    /* Unlimited AEC step size */
+#define   OV7670_COM8_FASTAEC    0x80    /* Enable fast AGC/AEC */
+#define OV7670_REG_COM9        0x14    /* Control 9  - gain ceiling */
+#define OV7670_REG_COM10       0x15    /* Control 10 */
+#define OV7670_REG_HSTART      0x17    /* Horiz start high bits */
+#define OV7670_REG_HSTOP       0x18    /* Horiz stop high bits */
+#define OV7670_REG_VSTART      0x19    /* Vert start high bits */
+#define OV7670_REG_VSTOP       0x1a    /* Vert stop high bits */
+#define OV7670_REG_MVFP        0x1e    /* Mirror / vflip */
+#define   OV7670_MVFP_MIRROR     0x20    /* Mirror image */
+#define OV7670_REG_AEW         0x24    /* AGC upper limit */
+#define OV7670_REG_AEB         0x25    /* AGC lower limit */
+#define OV7670_REG_VPT         0x26    /* AGC/AEC fast mode op region */
+#define OV7670_REG_HREF        0x32    /* HREF pieces */
+#define OV7670_REG_TSLB        0x3a    /* lots of stuff */
+#define OV7670_REG_COM11       0x3b    /* Control 11 */
+#define   OV7670_COM11_EXP       0x02
+#define   OV7670_COM11_HZAUTO    0x10    /* Auto detect 50/60 Hz */
+#define OV7670_REG_COM12       0x3c    /* Control 12 */
+#define OV7670_REG_COM13       0x3d    /* Control 13 */
+#define   OV7670_COM13_GAMMA     0x80    /* Gamma enable */
+#define   OV7670_COM13_UVSAT     0x40    /* UV saturation auto adjustment */
+#define OV7670_REG_COM14       0x3e    /* Control 14 */
+#define OV7670_REG_EDGE        0x3f    /* Edge enhancement factor */
+#define OV7670_REG_COM15       0x40    /* Control 15 */
+#define   OV7670_COM15_R00FF     0xc0    /*            00 to FF */
+#define OV7670_REG_COM16       0x41    /* Control 16 */
+#define   OV7670_COM16_AWBGAIN   0x08    /* AWB gain enable */
+#define OV7670_REG_BRIGHT      0x55    /* Brightness */
+#define OV7670_REG_CONTRAS     0x56    /* Contrast control */
+#define OV7670_REG_GFIX        0x69    /* Fix gain control */
+#define OV7670_REG_RGB444      0x8c    /* RGB 444 control */
+#define OV7670_REG_HAECC1      0x9f    /* Hist AEC/AGC control 1 */
+#define OV7670_REG_HAECC2      0xa0    /* Hist AEC/AGC control 2 */
+#define OV7670_REG_BD50MAX     0xa5    /* 50hz banding step limit */
+#define OV7670_REG_HAECC3      0xa6    /* Hist AEC/AGC control 3 */
+#define OV7670_REG_HAECC4      0xa7    /* Hist AEC/AGC control 4 */
+#define OV7670_REG_HAECC5      0xa8    /* Hist AEC/AGC control 5 */
+#define OV7670_REG_HAECC6      0xa9    /* Hist AEC/AGC control 6 */
+#define OV7670_REG_HAECC7      0xaa    /* Hist AEC/AGC control 7 */
+#define OV7670_REG_BD60MAX     0xab    /* 60hz banding step limit */
+
+struct ovsensor_window {
+	short x;
+	short y;
+	short width;
+	short height;
+/*	int format; */
+	short quarter;		/* Scale width and height down 2x */
+	short clockdiv;		/* Clock divisor setting */
+};
+
+static unsigned char ov7670_abs_to_sm(unsigned char v)
+{
+	if (v > 127)
+		return v & 0x7f;
+	return (128 - v) | 0x80;
+}
+
+/* Write a OV519 register */
+static int reg_w(struct sd *sd, __u16 index, __u8 value)
+{
+	int ret;
+	__u8 buf[4];
+
+	buf[0] = value;
+	ret = usb_control_msg(sd->gspca_dev.dev,
+			usb_sndctrlpipe(sd->gspca_dev.dev, 0),
+			1,			/* REQ_IO (ov518/519) */
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0, index,
+			&buf[0], 1, 500);
+	if (ret < 0)
+		PDEBUG(D_ERR, "Write reg [%02x] %02x failed", index, value);
+	return ret;
+}
+
+/* Read from a OV519 register */
+/* returns: negative is error, pos or zero is data */
+static int reg_r(struct sd *sd, __u16 index)
+{
+	int ret;
+	__u8 buf[4];
+
+	ret = usb_control_msg(sd->gspca_dev.dev,
+			usb_rcvctrlpipe(sd->gspca_dev.dev, 0),
+			1,			/* REQ_IO */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0, index, &buf[0], 1, 500);
+
+	if (ret >= 0)
+		ret = buf[0];
+	else
+		PDEBUG(D_ERR, "Read reg [0x%02x] failed", index);
+	return ret;
+}
+
+/* Read 8 values from a OV519 register */
+static int reg_r8(struct sd *sd,
+		__u16 index)
+{
+	int ret;
+	__u8 buf[8];
+
+	ret = usb_control_msg(sd->gspca_dev.dev,
+			usb_rcvctrlpipe(sd->gspca_dev.dev, 0),
+			1,			/* REQ_IO */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0, index, &buf[0], 8, 500);
+
+	if (ret >= 0)
+		ret = buf[0];
+	else
+		PDEBUG(D_ERR, "Read reg 8 [0x%02x] failed", index);
+	return ret;
+}
+
+/*
+ * Writes bits at positions specified by mask to an OV51x reg. Bits that are in
+ * the same position as 1's in "mask" are cleared and set to "value". Bits
+ * that are in the same position as 0's in "mask" are preserved, regardless
+ * of their respective state in "value".
+ */
+static int reg_w_mask(struct sd *sd,
+			__u16 index,
+			__u8 value,
+			__u8 mask)
+{
+	int ret;
+	__u8 oldval;
+
+	if (mask != 0xff) {
+		value &= mask;			/* Enforce mask on value */
+		ret = reg_r(sd, index);
+		if (ret < 0)
+			return ret;
+
+		oldval = ret & ~mask;		/* Clear the masked bits */
+		value |= oldval;		/* Set the desired bits */
+	}
+	return reg_w(sd, index, value);
+}
+
+/*
+ * The OV518 I2C I/O procedure is different, hence, this function.
+ * This is normally only called from i2c_w(). Note that this function
+ * always succeeds regardless of whether the sensor is present and working.
+ */
+static int i2c_w(struct sd *sd,
+		__u8 reg,
+		__u8 value)
+{
+	int rc;
+
+	PDEBUG(D_USBO, "i2c 0x%02x -> [0x%02x]", value, reg);
+
+	/* Select camera register */
+	rc = reg_w(sd, R51x_I2C_SADDR_3, reg);
+	if (rc < 0)
+		return rc;
+
+	/* Write "value" to I2C data port of OV511 */
+	rc = reg_w(sd, R51x_I2C_DATA, value);
+	if (rc < 0)
+		return rc;
+
+	/* Initiate 3-byte write cycle */
+	rc = reg_w(sd, R518_I2C_CTL, 0x01);
+
+	/* wait for write complete */
+	msleep(4);
+	if (rc < 0)
+		return rc;
+	return reg_r8(sd, R518_I2C_CTL);
+}
+
+/*
+ * returns: negative is error, pos or zero is data
+ *
+ * The OV518 I2C I/O procedure is different, hence, this function.
+ * This is normally only called from i2c_r(). Note that this function
+ * always succeeds regardless of whether the sensor is present and working.
+ */
+static int i2c_r(struct sd *sd, __u8 reg)
+{
+	int rc, value;
+
+	/* Select camera register */
+	rc = reg_w(sd, R51x_I2C_SADDR_2, reg);
+	if (rc < 0)
+		return rc;
+
+	/* Initiate 2-byte write cycle */
+	rc = reg_w(sd, R518_I2C_CTL, 0x03);
+	if (rc < 0)
+		return rc;
+
+	/* Initiate 2-byte read cycle */
+	rc = reg_w(sd, R518_I2C_CTL, 0x05);
+	if (rc < 0)
+		return rc;
+	value = reg_r(sd, R51x_I2C_DATA);
+	PDEBUG(D_USBI, "i2c [0x%02X] -> 0x%02X", reg, value);
+	return value;
+}
+
+/* Writes bits at positions specified by mask to an I2C reg. Bits that are in
+ * the same position as 1's in "mask" are cleared and set to "value". Bits
+ * that are in the same position as 0's in "mask" are preserved, regardless
+ * of their respective state in "value".
+ */
+static int i2c_w_mask(struct sd *sd,
+		   __u8 reg,
+		   __u8 value,
+		   __u8 mask)
+{
+	int rc;
+	__u8 oldval;
+
+	value &= mask;			/* Enforce mask on value */
+	rc = i2c_r(sd, reg);
+	if (rc < 0)
+		return rc;
+	oldval = rc & ~mask;		/* Clear the masked bits */
+	value |= oldval;		/* Set the desired bits */
+	return i2c_w(sd, reg, value);
+}
+
+/* Temporarily stops OV511 from functioning. Must do this before changing
+ * registers while the camera is streaming */
+static inline int ov51x_stop(struct sd *sd)
+{
+	PDEBUG(D_STREAM, "stopping");
+	sd->stopped = 1;
+	return reg_w(sd, OV519_SYS_RESET1, 0x0f);
+}
+
+/* Restarts OV511 after ov511_stop() is called. Has no effect if it is not
+ * actually stopped (for performance). */
+static inline int ov51x_restart(struct sd *sd)
+{
+	PDEBUG(D_STREAM, "restarting");
+	if (!sd->stopped)
+		return 0;
+	sd->stopped = 0;
+
+	/* Reinitialize the stream */
+	return reg_w(sd, OV519_SYS_RESET1, 0x00);
+}
+
+/* This does an initial reset of an OmniVision sensor and ensures that I2C
+ * is synchronized. Returns <0 on failure.
+ */
+static int init_ov_sensor(struct sd *sd)
+{
+	int i, success;
+
+	/* Reset the sensor */
+	if (i2c_w(sd, 0x12, 0x80) < 0)
+		return -EIO;
+
+	/* Wait for it to initialize */
+	msleep(150);
+
+	for (i = 0, success = 0; i < i2c_detect_tries && !success; i++) {
+		if (i2c_r(sd, OV7610_REG_ID_HIGH) == 0x7f &&
+		    i2c_r(sd, OV7610_REG_ID_LOW) == 0xa2) {
+			success = 1;
+			continue;
+		}
+
+		/* Reset the sensor */
+		if (i2c_w(sd, 0x12, 0x80) < 0)
+			return -EIO;
+		/* Wait for it to initialize */
+		msleep(150);
+		/* Dummy read to sync I2C */
+		if (i2c_r(sd, 0x00) < 0)
+			return -EIO;
+	}
+	if (!success)
+		return -EIO;
+	PDEBUG(D_PROBE, "I2C synced in %d attempt(s)", i);
+	return 0;
+}
+
+/* Switch on standard JPEG compression. Returns 0 for success. */
+static int ov519_init_compression(struct sd *sd)
+{
+	if (!sd->compress_inited) {
+		if (reg_w_mask(sd, OV519_SYS_EN_CLK1, 1 << 2, 1 << 2) < 0) {
+			PDEBUG(D_ERR, "Error switching to compressed mode");
+			return -EIO;
+		}
+		sd->compress_inited = 1;
+	}
+	return 0;
+}
+
+/* Set the read and write slave IDs. The "slave" argument is the write slave,
+ * and the read slave will be set to (slave + 1).
+ * This should not be called from outside the i2c I/O functions.
+ * Sets I2C read and write slave IDs. Returns <0 for error
+ */
+static int ov51x_set_slave_ids(struct sd *sd,
+				__u8 slave)
+{
+	int rc;
+
+	rc = reg_w(sd, R51x_I2C_W_SID, slave);
+	if (rc < 0)
+		return rc;
+	return reg_w(sd, R51x_I2C_R_SID, slave + 1);
+}
+
+struct ov_regvals {
+	__u8 reg;
+	__u8 val;
+};
+struct ov_i2c_regvals {
+	__u8 reg;
+	__u8 val;
+};
+
+static int write_regvals(struct sd *sd,
+			 struct ov_regvals *regvals,
+			 int n)
+{
+	int rc;
+
+	while (--n >= 0) {
+		rc = reg_w(sd, regvals->reg, regvals->val);
+		if (rc < 0)
+			return rc;
+		regvals++;
+	}
+	return 0;
+}
+
+static int write_i2c_regvals(struct sd *sd,
+			     struct ov_i2c_regvals *regvals,
+			     int n)
+{
+	int rc;
+
+	while (--n >= 0) {
+		rc = i2c_w(sd, regvals->reg, regvals->val);
+		if (rc < 0)
+			return rc;
+		regvals++;
+	}
+	return 0;
+}
+
+/****************************************************************************
+ *
+ * OV511 and sensor configuration
+ *
+ ***************************************************************************/
+
+/* This initializes the OV8110, OV8610 sensor. The OV8110 uses
+ * the same register settings as the OV8610, since they are very similar.
+ */
+static int ov8xx0_configure(struct sd *sd)
+{
+	int rc;
+	static struct ov_i2c_regvals norm_8610[] = {
+		{ 0x12, 0x80 },
+		{ 0x00, 0x00 },
+		{ 0x01, 0x80 },
+		{ 0x02, 0x80 },
+		{ 0x03, 0xc0 },
+		{ 0x04, 0x30 },
+		{ 0x05, 0x30 }, /* was 0x10, new from windrv 090403 */
+		{ 0x06, 0x70 }, /* was 0x80, new from windrv 090403 */
+		{ 0x0a, 0x86 },
+		{ 0x0b, 0xb0 },
+		{ 0x0c, 0x20 },
+		{ 0x0d, 0x20 },
+		{ 0x11, 0x01 },
+		{ 0x12, 0x25 },
+		{ 0x13, 0x01 },
+		{ 0x14, 0x04 },
+		{ 0x15, 0x01 }, /* Lin and Win think different about UV order */
+		{ 0x16, 0x03 },
+		{ 0x17, 0x38 }, /* was 0x2f, new from windrv 090403 */
+		{ 0x18, 0xea }, /* was 0xcf, new from windrv 090403 */
+		{ 0x19, 0x02 }, /* was 0x06, new from windrv 090403 */
+		{ 0x1a, 0xf5 },
+		{ 0x1b, 0x00 },
+		{ 0x20, 0xd0 }, /* was 0x90, new from windrv 090403 */
+		{ 0x23, 0xc0 }, /* was 0x00, new from windrv 090403 */
+		{ 0x24, 0x30 }, /* was 0x1d, new from windrv 090403 */
+		{ 0x25, 0x50 }, /* was 0x57, new from windrv 090403 */
+		{ 0x26, 0xa2 },
+		{ 0x27, 0xea },
+		{ 0x28, 0x00 },
+		{ 0x29, 0x00 },
+		{ 0x2a, 0x80 },
+		{ 0x2b, 0xc8 }, /* was 0xcc, new from windrv 090403 */
+		{ 0x2c, 0xac },
+		{ 0x2d, 0x45 }, /* was 0xd5, new from windrv 090403 */
+		{ 0x2e, 0x80 },
+		{ 0x2f, 0x14 }, /* was 0x01, new from windrv 090403 */
+		{ 0x4c, 0x00 },
+		{ 0x4d, 0x30 }, /* was 0x10, new from windrv 090403 */
+		{ 0x60, 0x02 }, /* was 0x01, new from windrv 090403 */
+		{ 0x61, 0x00 }, /* was 0x09, new from windrv 090403 */
+		{ 0x62, 0x5f }, /* was 0xd7, new from windrv 090403 */
+		{ 0x63, 0xff },
+		{ 0x64, 0x53 }, /* new windrv 090403 says 0x57,
+				 * maybe thats wrong */
+		{ 0x65, 0x00 },
+		{ 0x66, 0x55 },
+		{ 0x67, 0xb0 },
+		{ 0x68, 0xc0 }, /* was 0xaf, new from windrv 090403 */
+		{ 0x69, 0x02 },
+		{ 0x6a, 0x22 },
+		{ 0x6b, 0x00 },
+		{ 0x6c, 0x99 }, /* was 0x80, old windrv says 0x00, but
+				   deleting bit7 colors the first images red */
+		{ 0x6d, 0x11 }, /* was 0x00, new from windrv 090403 */
+		{ 0x6e, 0x11 }, /* was 0x00, new from windrv 090403 */
+		{ 0x6f, 0x01 },
+		{ 0x70, 0x8b },
+		{ 0x71, 0x00 },
+		{ 0x72, 0x14 },
+		{ 0x73, 0x54 },
+		{ 0x74, 0x00 },/* 0x60? - was 0x00, new from windrv 090403 */
+		{ 0x75, 0x0e },
+		{ 0x76, 0x02 }, /* was 0x02, new from windrv 090403 */
+		{ 0x77, 0xff },
+		{ 0x78, 0x80 },
+		{ 0x79, 0x80 },
+		{ 0x7a, 0x80 },
+		{ 0x7b, 0x10 }, /* was 0x13, new from windrv 090403 */
+		{ 0x7c, 0x00 },
+		{ 0x7d, 0x08 }, /* was 0x09, new from windrv 090403 */
+		{ 0x7e, 0x08 }, /* was 0xc0, new from windrv 090403 */
+		{ 0x7f, 0xfb },
+		{ 0x80, 0x28 },
+		{ 0x81, 0x00 },
+		{ 0x82, 0x23 },
+		{ 0x83, 0x0b },
+		{ 0x84, 0x00 },
+		{ 0x85, 0x62 }, /* was 0x61, new from windrv 090403 */
+		{ 0x86, 0xc9 },
+		{ 0x87, 0x00 },
+		{ 0x88, 0x00 },
+		{ 0x89, 0x01 },
+		{ 0x12, 0x20 },
+		{ 0x12, 0x25 }, /* was 0x24, new from windrv 090403 */
+	};
+
+	PDEBUG(D_PROBE, "starting ov8xx0 configuration");
+
+	if (init_ov_sensor(sd) < 0)
+		PDEBUG(D_ERR|D_PROBE, "Failed to read sensor ID");
+	else
+		PDEBUG(D_PROBE, "OV86x0 initialized");
+
+	/* Detect sensor (sub)type */
+	rc = i2c_r(sd, OV7610_REG_COM_I);
+	if (rc < 0) {
+		PDEBUG(D_ERR, "Error detecting sensor type");
+		return -1;
+	}
+	if ((rc & 3) == 1) {
+		PDEBUG(D_PROBE, "Sensor is an OV8610");
+		sd->sensor = SEN_OV8610;
+	} else {
+		PDEBUG(D_ERR, "Unknown image sensor version: %d", rc & 3);
+		return -1;
+	}
+	PDEBUG(D_PROBE, "Writing 8610 registers");
+	if (write_i2c_regvals(sd,
+			norm_8610,
+			sizeof norm_8610 / sizeof norm_8610[0]))
+		return -1;
+
+	/* Set sensor-specific vars */
+	sd->maxwidth = 640;
+	sd->maxheight = 480;
+	return 0;
+}
+
+/* This initializes the OV7610, OV7620, or OV76BE sensor. The OV76BE uses
+ * the same register settings as the OV7610, since they are very similar.
+ */
+static int ov7xx0_configure(struct sd *sd)
+{
+	int rc, high, low;
+
+	/* Lawrence Glaister <lg@jfm.bc.ca> reports:
+	 *
+	 * Register 0x0f in the 7610 has the following effects:
+	 *
+	 * 0x85 (AEC method 1): Best overall, good contrast range
+	 * 0x45 (AEC method 2): Very overexposed
+	 * 0xa5 (spec sheet default): Ok, but the black level is
+	 *	shifted resulting in loss of contrast
+	 * 0x05 (old driver setting): very overexposed, too much
+	 *	contrast
+	 */
+	static struct ov_i2c_regvals norm_7610[] = {
+		{ 0x10, 0xff },
+		{ 0x16, 0x06 },
+		{ 0x28, 0x24 },
+		{ 0x2b, 0xac },
+		{ 0x12, 0x00 },
+		{ 0x38, 0x81 },
+		{ 0x28, 0x24 },	/* 0c */
+		{ 0x0f, 0x85 },	/* lg's setting */
+		{ 0x15, 0x01 },
+		{ 0x20, 0x1c },
+		{ 0x23, 0x2a },
+		{ 0x24, 0x10 },
+		{ 0x25, 0x8a },
+		{ 0x26, 0xa2 },
+		{ 0x27, 0xc2 },
+		{ 0x2a, 0x04 },
+		{ 0x2c, 0xfe },
+		{ 0x2d, 0x93 },
+		{ 0x30, 0x71 },
+		{ 0x31, 0x60 },
+		{ 0x32, 0x26 },
+		{ 0x33, 0x20 },
+		{ 0x34, 0x48 },
+		{ 0x12, 0x24 },
+		{ 0x11, 0x01 },
+		{ 0x0c, 0x24 },
+		{ 0x0d, 0x24 },
+	};
+
+	static struct ov_i2c_regvals norm_7620[] = {
+		{ 0x00, 0x00 },		/* gain */
+		{ 0x01, 0x80 },		/* blue gain */
+		{ 0x02, 0x80 },		/* red gain */
+		{ 0x03, 0xc0 },		/* OV7670_REG_VREF */
+		{ 0x06, 0x60 },
+		{ 0x07, 0x00 },
+		{ 0x0c, 0x24 },
+		{ 0x0c, 0x24 },
+		{ 0x0d, 0x24 },
+		{ 0x11, 0x01 },
+		{ 0x12, 0x24 },
+		{ 0x13, 0x01 },
+		{ 0x14, 0x84 },
+		{ 0x15, 0x01 },
+		{ 0x16, 0x03 },
+		{ 0x17, 0x2f },
+		{ 0x18, 0xcf },
+		{ 0x19, 0x06 },
+		{ 0x1a, 0xf5 },
+		{ 0x1b, 0x00 },
+		{ 0x20, 0x18 },
+		{ 0x21, 0x80 },
+		{ 0x22, 0x80 },
+		{ 0x23, 0x00 },
+		{ 0x26, 0xa2 },
+		{ 0x27, 0xea },
+		{ 0x28, 0x20 },
+		{ 0x29, 0x00 },
+		{ 0x2a, 0x10 },
+		{ 0x2b, 0x00 },
+		{ 0x2c, 0x88 },
+		{ 0x2d, 0x91 },
+		{ 0x2e, 0x80 },
+		{ 0x2f, 0x44 },
+		{ 0x60, 0x27 },
+		{ 0x61, 0x02 },
+		{ 0x62, 0x5f },
+		{ 0x63, 0xd5 },
+		{ 0x64, 0x57 },
+		{ 0x65, 0x83 },
+		{ 0x66, 0x55 },
+		{ 0x67, 0x92 },
+		{ 0x68, 0xcf },
+		{ 0x69, 0x76 },
+		{ 0x6a, 0x22 },
+		{ 0x6b, 0x00 },
+		{ 0x6c, 0x02 },
+		{ 0x6d, 0x44 },
+		{ 0x6e, 0x80 },
+		{ 0x6f, 0x1d },
+		{ 0x70, 0x8b },
+		{ 0x71, 0x00 },
+		{ 0x72, 0x14 },
+		{ 0x73, 0x54 },
+		{ 0x74, 0x00 },
+		{ 0x75, 0x8e },
+		{ 0x76, 0x00 },
+		{ 0x77, 0xff },
+		{ 0x78, 0x80 },
+		{ 0x79, 0x80 },
+		{ 0x7a, 0x80 },
+		{ 0x7b, 0xe2 },
+		{ 0x7c, 0x00 },
+	};
+
+	/* 7640 and 7648. The defaults should be OK for most registers. */
+	static struct ov_i2c_regvals norm_7640[] = {
+		{ 0x12, 0x80 },
+		{ 0x12, 0x14 },
+	};
+
+	/* 7670. Defaults taken from OmniVision provided data,
+	*  as provided by Jonathan Corbet of OLPC		*/
+	static struct ov_i2c_regvals norm_7670[] = {
+		{ OV7670_REG_COM7, OV7670_COM7_RESET },
+		{ OV7670_REG_TSLB, 0x04 },		/* OV */
+		{ OV7670_REG_COM7, OV7670_COM7_FMT_VGA }, /* VGA */
+		{ OV7670_REG_CLKRC, 0x1 },
+	/*
+	 * Set the hardware window.  These values from OV don't entirely
+	 * make sense - hstop is less than hstart.  But they work...
+	 */
+		{ OV7670_REG_HSTART, 0x13 },	{ OV7670_REG_HSTOP, 0x01 },
+		{ OV7670_REG_HREF, 0xb6 },	{ OV7670_REG_VSTART, 0x02 },
+		{ OV7670_REG_VSTOP, 0x7a },	{ OV7670_REG_VREF, 0x0a },
+
+		{ OV7670_REG_COM3, 0 },	{ OV7670_REG_COM14, 0 },
+	/* Mystery scaling numbers */
+		{ 0x70, 0x3a },		{ 0x71, 0x35 },
+		{ 0x72, 0x11 },		{ 0x73, 0xf0 },
+		{ 0xa2, 0x02 },
+/* jfm */
+/* { OV7670_REG_COM10, 0x0 }, */
+
+	/* Gamma curve values */
+		{ 0x7a, 0x20 },
+/* jfm:win 7b=1c */
+		{ 0x7b, 0x10 },
+/* jfm:win 7c=28 */
+		{ 0x7c, 0x1e },
+/* jfm:win 7d=3c */
+		{ 0x7d, 0x35 },
+		{ 0x7e, 0x5a },		{ 0x7f, 0x69 },
+		{ 0x80, 0x76 },		{ 0x81, 0x80 },
+		{ 0x82, 0x88 },		{ 0x83, 0x8f },
+		{ 0x84, 0x96 },		{ 0x85, 0xa3 },
+		{ 0x86, 0xaf },		{ 0x87, 0xc4 },
+		{ 0x88, 0xd7 },		{ 0x89, 0xe8 },
+
+	/* AGC and AEC parameters.  Note we start by disabling those features,
+	   then turn them only after tweaking the values. */
+		{ OV7670_REG_COM8, OV7670_COM8_FASTAEC
+				 | OV7670_COM8_AECSTEP
+				 | OV7670_COM8_BFILT },
+		{ OV7670_REG_GAIN, 0 },	{ OV7670_REG_AECH, 0 },
+		{ OV7670_REG_COM4, 0x40 }, /* magic reserved bit */
+/* jfm:win 14=38 */
+		{ OV7670_REG_COM9, 0x18 }, /* 4x gain + magic rsvd bit */
+		{ OV7670_REG_BD50MAX, 0x05 },	{ OV7670_REG_BD60MAX, 0x07 },
+		{ OV7670_REG_AEW, 0x95 },	{ OV7670_REG_AEB, 0x33 },
+		{ OV7670_REG_VPT, 0xe3 },	{ OV7670_REG_HAECC1, 0x78 },
+		{ OV7670_REG_HAECC2, 0x68 },
+/* jfm:win a1=0b */
+		{ 0xa1, 0x03 }, /* magic */
+		{ OV7670_REG_HAECC3, 0xd8 },	{ OV7670_REG_HAECC4, 0xd8 },
+		{ OV7670_REG_HAECC5, 0xf0 },	{ OV7670_REG_HAECC6, 0x90 },
+		{ OV7670_REG_HAECC7, 0x94 },
+		{ OV7670_REG_COM8, OV7670_COM8_FASTAEC
+				| OV7670_COM8_AECSTEP
+				| OV7670_COM8_BFILT
+				| OV7670_COM8_AGC
+				| OV7670_COM8_AEC },
+
+	/* Almost all of these are magic "reserved" values.  */
+		{ OV7670_REG_COM5, 0x61 },	{ OV7670_REG_COM6, 0x4b },
+		{ 0x16, 0x02 },
+/* jfm */
+/*		{ OV7670_REG_MVFP, 0x07|OV7670_MVFP_MIRROR }, */
+		{ OV7670_REG_MVFP, 0x07 },
+		{ 0x21, 0x02 },		{ 0x22, 0x91 },
+		{ 0x29, 0x07 },		{ 0x33, 0x0b },
+		{ 0x35, 0x0b },		{ 0x37, 0x1d },
+		{ 0x38, 0x71 },		{ 0x39, 0x2a },
+		{ OV7670_REG_COM12, 0x78 },	{ 0x4d, 0x40 },
+		{ 0x4e, 0x20 },		{ OV7670_REG_GFIX, 0 },
+		{ 0x6b, 0x4a },		{ 0x74, 0x10 },
+		{ 0x8d, 0x4f },		{ 0x8e, 0 },
+		{ 0x8f, 0 },		{ 0x90, 0 },
+		{ 0x91, 0 },		{ 0x96, 0 },
+		{ 0x9a, 0 },		{ 0xb0, 0x84 },
+		{ 0xb1, 0x0c },		{ 0xb2, 0x0e },
+		{ 0xb3, 0x82 },		{ 0xb8, 0x0a },
+
+	/* More reserved magic, some of which tweaks white balance */
+		{ 0x43, 0x0a },		{ 0x44, 0xf0 },
+		{ 0x45, 0x34 },		{ 0x46, 0x58 },
+		{ 0x47, 0x28 },		{ 0x48, 0x3a },
+		{ 0x59, 0x88 },		{ 0x5a, 0x88 },
+		{ 0x5b, 0x44 },		{ 0x5c, 0x67 },
+		{ 0x5d, 0x49 },		{ 0x5e, 0x0e },
+		{ 0x6c, 0x0a },		{ 0x6d, 0x55 },
+		{ 0x6e, 0x11 },		{ 0x6f, 0x9f },
+						/* "9e for advance AWB" */
+		{ 0x6a, 0x40 },		{ OV7670_REG_BLUE, 0x40 },
+		{ OV7670_REG_RED, 0x60 },
+		{ OV7670_REG_COM8, OV7670_COM8_FASTAEC
+				| OV7670_COM8_AECSTEP
+				| OV7670_COM8_BFILT
+				| OV7670_COM8_AGC
+				| OV7670_COM8_AEC
+				| OV7670_COM8_AWB },
+
+	/* Matrix coefficients */
+		{ 0x4f, 0x80 },		{ 0x50, 0x80 },
+		{ 0x51, 0 },		{ 0x52, 0x22 },
+		{ 0x53, 0x5e },		{ 0x54, 0x80 },
+		{ 0x58, 0x9e },
+
+		{ OV7670_REG_COM16, OV7670_COM16_AWBGAIN },
+		{ OV7670_REG_EDGE, 0 },
+		{ 0x75, 0x05 },		{ 0x76, 0xe1 },
+		{ 0x4c, 0 },		{ 0x77, 0x01 },
+		{ OV7670_REG_COM13, 0xc3 },	{ 0x4b, 0x09 },
+		{ 0xc9, 0x60 },		{ OV7670_REG_COM16, 0x38 },
+		{ 0x56, 0x40 },
+
+		{ 0x34, 0x11 },
+		{ OV7670_REG_COM11, OV7670_COM11_EXP|OV7670_COM11_HZAUTO },
+		{ 0xa4, 0x88 },		{ 0x96, 0 },
+		{ 0x97, 0x30 },		{ 0x98, 0x20 },
+		{ 0x99, 0x30 },		{ 0x9a, 0x84 },
+		{ 0x9b, 0x29 },		{ 0x9c, 0x03 },
+		{ 0x9d, 0x4c },		{ 0x9e, 0x3f },
+		{ 0x78, 0x04 },
+
+	/* Extra-weird stuff.  Some sort of multiplexor register */
+		{ 0x79, 0x01 },		{ 0xc8, 0xf0 },
+		{ 0x79, 0x0f },		{ 0xc8, 0x00 },
+		{ 0x79, 0x10 },		{ 0xc8, 0x7e },
+		{ 0x79, 0x0a },		{ 0xc8, 0x80 },
+		{ 0x79, 0x0b },		{ 0xc8, 0x01 },
+		{ 0x79, 0x0c },		{ 0xc8, 0x0f },
+		{ 0x79, 0x0d },		{ 0xc8, 0x20 },
+		{ 0x79, 0x09 },		{ 0xc8, 0x80 },
+		{ 0x79, 0x02 },		{ 0xc8, 0xc0 },
+		{ 0x79, 0x03 },		{ 0xc8, 0x40 },
+		{ 0x79, 0x05 },		{ 0xc8, 0x30 },
+		{ 0x79, 0x26 },
+
+	/* Format YUV422 */
+		{ OV7670_REG_COM7, OV7670_COM7_YUV },  /* Selects YUV mode */
+		{ OV7670_REG_RGB444, 0 },	/* No RGB444 please */
+		{ OV7670_REG_COM1, 0 },
+		{ OV7670_REG_COM15, OV7670_COM15_R00FF },
+		{ OV7670_REG_COM9, 0x18 },
+				/* 4x gain ceiling; 0x8 is reserved bit */
+		{ 0x4f, 0x80 }, 	/* "matrix coefficient 1" */
+		{ 0x50, 0x80 }, 	/* "matrix coefficient 2" */
+		{ 0x52, 0x22 }, 	/* "matrix coefficient 4" */
+		{ 0x53, 0x5e }, 	/* "matrix coefficient 5" */
+		{ 0x54, 0x80 }, 	/* "matrix coefficient 6" */
+		{ OV7670_REG_COM13, OV7670_COM13_GAMMA|OV7670_COM13_UVSAT },
+};
+
+	PDEBUG(D_PROBE, "starting OV7xx0 configuration");
+
+/* jfm:already done? */
+	if (init_ov_sensor(sd) < 0)
+		PDEBUG(D_ERR, "Failed to read sensor ID");
+	else
+		PDEBUG(D_PROBE, "OV7xx0 initialized");
+
+	/* Detect sensor (sub)type */
+	rc = i2c_r(sd, OV7610_REG_COM_I);
+
+	/* add OV7670 here
+	 * it appears to be wrongly detected as a 7610 by default */
+	if (rc < 0) {
+		PDEBUG(D_ERR, "Error detecting sensor type");
+		return -1;
+	}
+	if ((rc & 3) == 3) {
+		/* quick hack to make OV7670s work */
+		high = i2c_r(sd, 0x0a);
+		low = i2c_r(sd, 0x0b);
+		/* info("%x, %x", high, low); */
+		if (high == 0x76 && low == 0x73) {
+			PDEBUG(D_PROBE, "Sensor is an OV7670");
+			sd->sensor = SEN_OV7670;
+		} else {
+			PDEBUG(D_PROBE, "Sensor is an OV7610");
+			sd->sensor = SEN_OV7610;
+		}
+	} else if ((rc & 3) == 1) {
+		/* I don't know what's different about the 76BE yet. */
+		if (i2c_r(sd, 0x15) & 1)
+			PDEBUG(D_PROBE, "Sensor is an OV7620AE");
+		else
+			PDEBUG(D_PROBE, "Sensor is an OV76BE");
+
+		/* OV511+ will return all zero isoc data unless we
+		 * configure the sensor as a 7620. Someone needs to
+		 * find the exact reg. setting that causes this. */
+		sd->sensor = SEN_OV76BE;
+	} else if ((rc & 3) == 0) {
+		/* try to read product id registers */
+		high = i2c_r(sd, 0x0a);
+		if (high < 0) {
+			PDEBUG(D_ERR, "Error detecting camera chip PID");
+			return high;
+		}
+		low = i2c_r(sd, 0x0b);
+		if (low < 0) {
+			PDEBUG(D_ERR, "Error detecting camera chip VER");
+			return low;
+		}
+		if (high == 0x76) {
+			if (low == 0x30) {
+				PDEBUG(D_PROBE, "Sensor is an OV7630/OV7635");
+				sd->sensor = SEN_OV7630;
+			} else if (low == 0x40) {
+				PDEBUG(D_PROBE, "Sensor is an OV7645");
+				sd->sensor = SEN_OV7640; /* FIXME */
+			} else if (low == 0x45) {
+				PDEBUG(D_PROBE, "Sensor is an OV7645B");
+				sd->sensor = SEN_OV7640; /* FIXME */
+			} else if (low == 0x48) {
+				PDEBUG(D_PROBE, "Sensor is an OV7648");
+				sd->sensor = SEN_OV7640; /* FIXME */
+			} else {
+				PDEBUG(D_PROBE, "Unknown sensor: 0x76%X", low);
+				return -1;
+			}
+		} else {
+			PDEBUG(D_PROBE, "Sensor is an OV7620");
+			sd->sensor = SEN_OV7620;
+		}
+	} else {
+		PDEBUG(D_ERR, "Unknown image sensor version: %d", rc & 3);
+		return -1;
+	}
+
+	if (sd->sensor == SEN_OV7620) {
+		PDEBUG(D_PROBE, "Writing 7620 registers");
+		if (write_i2c_regvals(sd, norm_7620,
+				sizeof norm_7620 / sizeof norm_7620[0]))
+			return -1;
+	} else if (sd->sensor == SEN_OV7630) {
+		PDEBUG(D_ERR, "7630 is not supported by this driver version");
+		return -1;
+	} else if (sd->sensor == SEN_OV7640) {
+		PDEBUG(D_PROBE, "Writing 7640 registers");
+		if (write_i2c_regvals(sd, norm_7640,
+				sizeof norm_7640 / sizeof norm_7640[0]))
+			return -1;
+	} else if (sd->sensor == SEN_OV7670) {
+		PDEBUG(D_PROBE, "Writing 7670 registers");
+		if (write_i2c_regvals(sd, norm_7670,
+				sizeof norm_7670 / sizeof norm_7670[0]))
+			return -1;
+	} else {
+		PDEBUG(D_PROBE, "Writing 7610 registers");
+		if (write_i2c_regvals(sd, norm_7610,
+				sizeof norm_7610 / sizeof norm_7610[0]))
+			return -1;
+	}
+
+	/* Set sensor-specific vars */
+	sd->maxwidth = 640;
+	sd->maxheight = 480;
+	return 0;
+}
+
+/* This initializes the OV6620, OV6630, OV6630AE, or OV6630AF sensor. */
+static int ov6xx0_configure(struct sd *sd)
+{
+	int rc;
+	static struct ov_i2c_regvals norm_6x20[] = {
+		{ 0x12, 0x80 }, /* reset */
+		{ 0x11, 0x01 },
+		{ 0x03, 0x60 },
+		{ 0x05, 0x7f }, /* For when autoadjust is off */
+		{ 0x07, 0xa8 },
+		/* The ratio of 0x0c and 0x0d  controls the white point */
+		{ 0x0c, 0x24 },
+		{ 0x0d, 0x24 },
+		{ 0x0f, 0x15 }, /* COMS */
+		{ 0x10, 0x75 }, /* AEC Exposure time */
+		{ 0x12, 0x24 }, /* Enable AGC */
+		{ 0x14, 0x04 },
+		/* 0x16: 0x06 helps frame stability with moving objects */
+		{ 0x16, 0x06 },
+/*		{ 0x20, 0x30 },  * Aperture correction enable */
+		{ 0x26, 0xb2 }, /* BLC enable */
+		/* 0x28: 0x05 Selects RGB format if RGB on */
+		{ 0x28, 0x05 },
+		{ 0x2a, 0x04 }, /* Disable framerate adjust */
+/*		{ 0x2b, 0xac },  * Framerate; Set 2a[7] first */
+		{ 0x2d, 0x99 },
+		{ 0x33, 0xa0 }, /* Color Processing Parameter */
+		{ 0x34, 0xd2 }, /* Max A/D range */
+		{ 0x38, 0x8b },
+		{ 0x39, 0x40 },
+
+		{ 0x3c, 0x39 }, /* Enable AEC mode changing */
+		{ 0x3c, 0x3c }, /* Change AEC mode */
+		{ 0x3c, 0x24 }, /* Disable AEC mode changing */
+
+		{ 0x3d, 0x80 },
+		/* These next two registers (0x4a, 0x4b) are undocumented.
+		 * They control the color balance */
+		{ 0x4a, 0x80 },
+		{ 0x4b, 0x80 },
+		{ 0x4d, 0xd2 }, /* This reduces noise a bit */
+		{ 0x4e, 0xc1 },
+		{ 0x4f, 0x04 },
+/* Do 50-53 have any effect? */
+/* Toggle 0x12[2] off and on here? */
+	};
+
+	static struct ov_i2c_regvals norm_6x30[] = {
+		{ 0x12, 0x80 }, /* Reset */
+		{ 0x00, 0x1f }, /* Gain */
+		{ 0x01, 0x99 }, /* Blue gain */
+		{ 0x02, 0x7c }, /* Red gain */
+		{ 0x03, 0xc0 }, /* Saturation */
+		{ 0x05, 0x0a }, /* Contrast */
+		{ 0x06, 0x95 }, /* Brightness */
+		{ 0x07, 0x2d }, /* Sharpness */
+		{ 0x0c, 0x20 },
+		{ 0x0d, 0x20 },
+		{ 0x0e, 0x20 },
+		{ 0x0f, 0x05 },
+		{ 0x10, 0x9a },
+		{ 0x11, 0x00 }, /* Pixel clock = fastest */
+		{ 0x12, 0x24 }, /* Enable AGC and AWB */
+		{ 0x13, 0x21 },
+		{ 0x14, 0x80 },
+		{ 0x15, 0x01 },
+		{ 0x16, 0x03 },
+		{ 0x17, 0x38 },
+		{ 0x18, 0xea },
+		{ 0x19, 0x04 },
+		{ 0x1a, 0x93 },
+		{ 0x1b, 0x00 },
+		{ 0x1e, 0xc4 },
+		{ 0x1f, 0x04 },
+		{ 0x20, 0x20 },
+		{ 0x21, 0x10 },
+		{ 0x22, 0x88 },
+		{ 0x23, 0xc0 }, /* Crystal circuit power level */
+		{ 0x25, 0x9a }, /* Increase AEC black ratio */
+		{ 0x26, 0xb2 }, /* BLC enable */
+		{ 0x27, 0xa2 },
+		{ 0x28, 0x00 },
+		{ 0x29, 0x00 },
+		{ 0x2a, 0x84 }, /* 60 Hz power */
+		{ 0x2b, 0xa8 }, /* 60 Hz power */
+		{ 0x2c, 0xa0 },
+		{ 0x2d, 0x95 }, /* Enable auto-brightness */
+		{ 0x2e, 0x88 },
+		{ 0x33, 0x26 },
+		{ 0x34, 0x03 },
+		{ 0x36, 0x8f },
+		{ 0x37, 0x80 },
+		{ 0x38, 0x83 },
+		{ 0x39, 0x80 },
+		{ 0x3a, 0x0f },
+		{ 0x3b, 0x3c },
+		{ 0x3c, 0x1a },
+		{ 0x3d, 0x80 },
+		{ 0x3e, 0x80 },
+		{ 0x3f, 0x0e },
+		{ 0x40, 0x00 }, /* White bal */
+		{ 0x41, 0x00 }, /* White bal */
+		{ 0x42, 0x80 },
+		{ 0x43, 0x3f }, /* White bal */
+		{ 0x44, 0x80 },
+		{ 0x45, 0x20 },
+		{ 0x46, 0x20 },
+		{ 0x47, 0x80 },
+		{ 0x48, 0x7f },
+		{ 0x49, 0x00 },
+		{ 0x4a, 0x00 },
+		{ 0x4b, 0x80 },
+		{ 0x4c, 0xd0 },
+		{ 0x4d, 0x10 }, /* U = 0.563u, V = 0.714v */
+		{ 0x4e, 0x40 },
+		{ 0x4f, 0x07 }, /* UV avg., col. killer: max */
+		{ 0x50, 0xff },
+		{ 0x54, 0x23 }, /* Max AGC gain: 18dB */
+		{ 0x55, 0xff },
+		{ 0x56, 0x12 },
+		{ 0x57, 0x81 },
+		{ 0x58, 0x75 },
+		{ 0x59, 0x01 }, /* AGC dark current comp.: +1 */
+		{ 0x5a, 0x2c },
+		{ 0x5b, 0x0f }, /* AWB chrominance levels */
+		{ 0x5c, 0x10 },
+		{ 0x3d, 0x80 },
+		{ 0x27, 0xa6 },
+		{ 0x12, 0x20 }, /* Toggle AWB */
+		{ 0x12, 0x24 },
+	};
+
+	PDEBUG(D_PROBE, "starting sensor configuration");
+
+	if (init_ov_sensor(sd) < 0) {
+		PDEBUG(D_ERR, "Failed to read sensor ID.");
+		return -1;
+	}
+	PDEBUG(D_PROBE, "OV6xx0 sensor detected");
+
+	/* Detect sensor (sub)type */
+	rc = i2c_r(sd, OV7610_REG_COM_I);
+	if (rc < 0) {
+		PDEBUG(D_ERR, "Error detecting sensor type");
+		return -1;
+	}
+
+	/* Ugh. The first two bits are the version bits, but
+	 * the entire register value must be used. I guess OVT
+	 * underestimated how many variants they would make. */
+	if (rc == 0x00) {
+		sd->sensor = SEN_OV6630;
+		PDEBUG(D_ERR,
+			"WARNING: Sensor is an OV66308. Your camera may have");
+		PDEBUG(D_ERR, "been misdetected in previous driver versions.");
+	} else if (rc == 0x01) {
+		sd->sensor = SEN_OV6620;
+		PDEBUG(D_PROBE, "Sensor is an OV6620");
+	} else if (rc == 0x02) {
+		sd->sensor = SEN_OV6630;
+		PDEBUG(D_PROBE, "Sensor is an OV66308AE");
+	} else if (rc == 0x03) {
+		sd->sensor = SEN_OV6630;
+		PDEBUG(D_PROBE, "Sensor is an OV66308AF");
+	} else if (rc == 0x90) {
+		sd->sensor = SEN_OV6630;
+		PDEBUG(D_ERR,
+			"WARNING: Sensor is an OV66307. Your camera may have");
+		PDEBUG(D_ERR, "been misdetected in previous driver versions.");
+	} else {
+		PDEBUG(D_ERR, "FATAL: Unknown sensor version: 0x%02x", rc);
+		return -1;
+	}
+
+	/* Set sensor-specific vars */
+	sd->maxwidth = 352;
+	sd->maxheight = 288;
+
+	if (sd->sensor == SEN_OV6620) {
+		PDEBUG(D_PROBE, "Writing 6x20 registers");
+		if (write_i2c_regvals(sd, norm_6x20,
+				sizeof norm_6x20 / sizeof norm_6x20[0]))
+			return -1;
+	} else {
+		PDEBUG(D_PROBE, "Writing 6x30 registers");
+		if (write_i2c_regvals(sd, norm_6x30,
+				sizeof norm_6x30 / sizeof norm_6x30[0]))
+			return -1;
+	}
+	return 0;
+}
+
+/* Turns on or off the LED. Only has an effect with OV511+/OV518(+)/OV519 */
+static void ov51x_led_control(struct sd *sd, int on)
+{
+	PDEBUG(D_STREAM, "LED (%s)", on ? "on" : "off");
+
+/*	if (sd->bridge == BRG_OV511PLUS) */
+/*		reg_w(sd, R511_SYS_LED_CTL, on ? 1 : 0); */
+/*	else if (sd->bridge == BRG_OV519) */
+		reg_w_mask(sd, OV519_GPIO_DATA_OUT0, !on, 1);	/* 0 / 1 */
+/*	else if (sd->bclass == BCL_OV518) */
+/*		reg_w_mask(sd, R518_GPIO_OUT, on ? 0x02 : 0x00, 0x02); */
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+
+/* (from ov519_configure) */
+	static struct ov_regvals init_519[] = {
+		{ 0x5a,  0x6d }, /* EnableSystem */
+/* jfm trace usbsnoop3-1.txt */
+/* jfm 53 = fb */
+		{ 0x53,  0x9b },
+		{ 0x54,  0xff }, /* set bit2 to enable jpeg */
+		{ 0x5d,  0x03 },
+		{ 0x49,  0x01 },
+		{ 0x48,  0x00 },
+		/* Set LED pin to output mode. Bit 4 must be cleared or sensor
+		 * detection will fail. This deserves further investigation. */
+		{ OV519_GPIO_IO_CTRL0,   0xee },
+		{ 0x51,  0x0f }, /* SetUsbInit */
+		{ 0x51,  0x00 },
+		{ 0x22,  0x00 },
+		/* windows reads 0x55 at this point*/
+	};
+
+	if (write_regvals(sd, init_519,
+			sizeof init_519 / sizeof init_519[0]))
+		goto error;
+/* jfm: not seen in windows trace */
+	if (ov519_init_compression(sd))
+		goto error;
+	ov51x_led_control(sd, 0);	/* turn LED off */
+
+	/* Test for 76xx */
+	sd->primary_i2c_slave = OV7xx0_SID;
+	if (ov51x_set_slave_ids(sd, OV7xx0_SID) < 0)
+		goto error;
+
+	/* The OV519 must be more aggressive about sensor detection since
+	 * I2C write will never fail if the sensor is not present. We have
+	 * to try to initialize the sensor to detect its presence */
+	if (init_ov_sensor(sd) < 0) {
+		/* Test for 6xx0 */
+		sd->primary_i2c_slave = OV6xx0_SID;
+		if (ov51x_set_slave_ids(sd, OV6xx0_SID) < 0)
+			goto error;
+
+		if (init_ov_sensor(sd) < 0) {
+			/* Test for 8xx0 */
+			sd->primary_i2c_slave = OV8xx0_SID;
+			if (ov51x_set_slave_ids(sd, OV8xx0_SID) < 0)
+				goto error;
+
+			if (init_ov_sensor(sd) < 0) {
+				PDEBUG(D_ERR,
+					"Can't determine sensor slave IDs");
+				goto error;
+			} else {
+				if (ov8xx0_configure(sd) < 0) {
+					PDEBUG(D_ERR,
+					   "Failed to configure OV8xx0 sensor");
+					goto error;
+				}
+			}
+		} else {
+			if (ov6xx0_configure(sd) < 0) {
+				PDEBUG(D_ERR, "Failed to configure OV6xx0");
+				goto error;
+			}
+		}
+	} else {
+		if (ov7xx0_configure(sd) < 0) {
+			PDEBUG(D_ERR, "Failed to configure OV7xx0");
+			goto error;
+		}
+	}
+
+	cam = &gspca_dev->cam;
+	cam->epaddr = OV511_ENDPOINT_ADDRESS;
+	if (sd->maxwidth == 640) {
+		cam->cam_mode = vga_mode;
+		cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	} else {
+		cam->cam_mode = sif_mode;
+		cam->nmodes = sizeof sif_mode / sizeof sif_mode[0];
+	}
+	cam->dev_name = (char *) id->driver_info;
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	return 0;
+error:
+	PDEBUG(D_ERR, "OV519 Config failed");
+	return -EBUSY;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	return 0;
+}
+
+/* Sets up the OV519 with the given image parameters
+ *
+ * OV519 needs a completely different approach, until we can figure out what
+ * the individual registers do.
+ *
+ * Do not put any sensor-specific code in here (including I2C I/O functions)
+ */
+static int ov519_mode_init_regs(struct sd *sd,
+				int width, int height)
+{
+	static struct ov_regvals mode_init_519_ov7670[] = {
+		{ 0x5d,	0x03 }, /* Turn off suspend mode */
+		{ 0x53,	0x9f }, /* was 9b in 1.65-1.08 */
+		{ 0x54,	0x0f }, /* bit2 (jpeg enable) */
+		{ 0xa2,	0x20 }, /* a2-a5 are undocumented */
+		{ 0xa3,	0x18 },
+		{ 0xa4,	0x04 },
+		{ 0xa5,	0x28 },
+		{ 0x37,	0x00 },	/* SetUsbInit */
+		{ 0x55,	0x02 }, /* 4.096 Mhz audio clock */
+		/* Enable both fields, YUV Input, disable defect comp (why?) */
+		{ 0x20,	0x0c },
+		{ 0x21,	0x38 },
+		{ 0x22,	0x1d },
+		{ 0x17,	0x50 }, /* undocumented */
+		{ 0x37,	0x00 }, /* undocumented */
+		{ 0x40,	0xff }, /* I2C timeout counter */
+		{ 0x46,	0x00 }, /* I2C clock prescaler */
+		{ 0x59,	0x04 },	/* new from windrv 090403 */
+		{ 0xff,	0x00 }, /* undocumented */
+		/* windows reads 0x55 at this point, why? */
+	};
+
+	static struct ov_regvals mode_init_519[] = {
+		{ 0x5d,	0x03 }, /* Turn off suspend mode */
+		{ 0x53,	0x9f }, /* was 9b in 1.65-1.08 */
+		{ 0x54,	0x0f }, /* bit2 (jpeg enable) */
+		{ 0xa2,	0x20 }, /* a2-a5 are undocumented */
+		{ 0xa3,	0x18 },
+		{ 0xa4,	0x04 },
+		{ 0xa5,	0x28 },
+		{ 0x37,	0x00 },	/* SetUsbInit */
+		{ 0x55,	0x02 }, /* 4.096 Mhz audio clock */
+		/* Enable both fields, YUV Input, disable defect comp (why?) */
+		{ 0x22,	0x1d },
+		{ 0x17,	0x50 }, /* undocumented */
+		{ 0x37,	0x00 }, /* undocumented */
+		{ 0x40,	0xff }, /* I2C timeout counter */
+		{ 0x46,	0x00 }, /* I2C clock prescaler */
+		{ 0x59,	0x04 },	/* new from windrv 090403 */
+		{ 0xff,	0x00 }, /* undocumented */
+		/* windows reads 0x55 at this point, why? */
+	};
+
+/* int hi_res; */
+
+	PDEBUG(D_CONF, "mode init %dx%d", width, height);
+
+/*	if (width >= 800 && height >= 600)
+		hi_res = 1;
+	else
+		hi_res = 0; */
+
+/*	if (ov51x_stop(sd) < 0)
+		return -EIO; */
+
+	/******** Set the mode ********/
+	if (sd->sensor != SEN_OV7670) {
+		if (write_regvals(sd, mode_init_519,
+			    sizeof mode_init_519 / sizeof mode_init_519[0]))
+			return -EIO;
+	} else {
+		if (write_regvals(sd, mode_init_519_ov7670,
+				sizeof mode_init_519_ov7670
+					/ sizeof mode_init_519_ov7670[0]))
+			return -EIO;
+	}
+
+	if (sd->sensor == SEN_OV7640) {
+		/* Select 8-bit input mode */
+		reg_w_mask(sd, OV519_CAM_DFR, 0x10, 0x10);
+	}
+
+	reg_w(sd, OV519_CAM_H_SIZE,	width >> 4);
+	reg_w(sd, OV519_CAM_V_SIZE,	height >> 3);
+	reg_w(sd, OV519_CAM_X_OFFSETL,	0x00);
+	reg_w(sd, OV519_CAM_X_OFFSETH,	0x00);
+	reg_w(sd, OV519_CAM_Y_OFFSETL,	0x00);
+	reg_w(sd, OV519_CAM_Y_OFFSETH,	0x00);
+	reg_w(sd, OV519_CAM_DIVIDER,	0x00);
+	reg_w(sd, OV519_CAM_FORMAT,	0x03); /* YUV422 */
+	reg_w(sd, 0x26,			0x00); /* Undocumented */
+
+	/******** Set the framerate ********/
+	if (frame_rate > 0)
+		sd->frame_rate = frame_rate;
+
+/* FIXME: These are only valid at the max resolution. */
+	sd->clockdiv = 0;
+	if (sd->sensor == SEN_OV7640) {
+		switch (sd->frame_rate) {
+/*jfm: default was 30 fps */
+		case 30:
+			reg_w(sd, 0xa4, 0x0c);
+			reg_w(sd, 0x23, 0xff);
+			break;
+		case 25:
+			reg_w(sd, 0xa4, 0x0c);
+			reg_w(sd, 0x23, 0x1f);
+			break;
+		case 20:
+			reg_w(sd, 0xa4, 0x0c);
+			reg_w(sd, 0x23, 0x1b);
+			break;
+		default:
+/*		case 15: */
+			reg_w(sd, 0xa4, 0x04);
+			reg_w(sd, 0x23, 0xff);
+			sd->clockdiv = 1;
+			break;
+		case 10:
+			reg_w(sd, 0xa4, 0x04);
+			reg_w(sd, 0x23, 0x1f);
+			sd->clockdiv = 1;
+			break;
+		case 5:
+			reg_w(sd, 0xa4, 0x04);
+			reg_w(sd, 0x23, 0x1b);
+			sd->clockdiv = 1;
+			break;
+		}
+	} else if (sd->sensor == SEN_OV8610) {
+		switch (sd->frame_rate) {
+		default:	/* 15 fps */
+/*		case 15: */
+			reg_w(sd, 0xa4, 0x06);
+			reg_w(sd, 0x23, 0xff);
+			break;
+		case 10:
+			reg_w(sd, 0xa4, 0x06);
+			reg_w(sd, 0x23, 0x1f);
+			break;
+		case 5:
+			reg_w(sd, 0xa4, 0x06);
+			reg_w(sd, 0x23, 0x1b);
+			break;
+		}
+		sd->clockdiv = 0;
+	} else if (sd->sensor == SEN_OV7670) { /* guesses, based on 7640 */
+		PDEBUG(D_STREAM, "Setting framerate to %d fps",
+				 (sd->frame_rate == 0) ? 15 : sd->frame_rate);
+		switch (sd->frame_rate) {
+		case 30:
+			reg_w(sd, 0xa4, 0x10);
+			reg_w(sd, 0x23, 0xff);
+			break;
+		case 20:
+			reg_w(sd, 0xa4, 0x10);
+			reg_w(sd, 0x23, 0x1b);
+			break;
+		default: /* 15 fps */
+/*			case 15: */
+			reg_w(sd, 0xa4, 0x10);
+			reg_w(sd, 0x23, 0xff);
+			sd->clockdiv = 1;
+			break;
+		}
+	}
+
+/*	if (ov51x_restart(sd) < 0)
+		return -EIO; */
+
+	/* Reset it just for good measure */
+/*	if (ov51x_reset(sd, OV511_RESET_NOREGS) < 0)
+		return -EIO; */
+	return 0;
+}
+
+static int mode_init_ov_sensor_regs(struct sd *sd,
+				struct ovsensor_window *win)
+{
+	int qvga = win->quarter;
+
+	/******** Mode (VGA/QVGA) and sensor specific regs ********/
+	switch (sd->sensor) {
+	case SEN_OV8610:
+		/* For OV8610 qvga means qsvga */
+		i2c_w_mask(sd, OV7610_REG_COM_C, qvga ? (1 << 5) : 0, 1 << 5);
+		break;
+	case SEN_OV7610:
+		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
+		break;
+	case SEN_OV7620:
+/*		i2c_w(sd, 0x2b, 0x00); */
+		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
+		i2c_w_mask(sd, 0x28, qvga ? 0x00 : 0x20, 0x20);
+		i2c_w(sd, 0x24, qvga ? 0x20 : 0x3a);
+		i2c_w(sd, 0x25, qvga ? 0x30 : 0x60);
+		i2c_w_mask(sd, 0x2d, qvga ? 0x40 : 0x00, 0x40);
+		i2c_w_mask(sd, 0x67, qvga ? 0xf0 : 0x90, 0xf0);
+		i2c_w_mask(sd, 0x74, qvga ? 0x20 : 0x00, 0x20);
+		break;
+	case SEN_OV76BE:
+/*		i2c_w(sd, 0x2b, 0x00); */
+		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
+		break;
+	case SEN_OV7640:
+/*		i2c_w(sd, 0x2b, 0x00); */
+		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
+		i2c_w_mask(sd, 0x28, qvga ? 0x00 : 0x20, 0x20);
+/*		i2c_w(sd, 0x24, qvga ? 0x20 : 0x3a); */
+/*		i2c_w(sd, 0x25, qvga ? 0x30 : 0x60); */
+/*		i2c_w_mask(sd, 0x2d, qvga ? 0x40 : 0x00, 0x40); */
+/*		i2c_w_mask(sd, 0x67, qvga ? 0xf0 : 0x90, 0xf0); */
+/*		i2c_w_mask(sd, 0x74, qvga ? 0x20 : 0x00, 0x20); */
+		break;
+	case SEN_OV7670:
+		/* set COM7_FMT_VGA or COM7_FMT_QVGA
+		 * do we need to set anything else?
+		 *	HSTART etc are set in set_ov_sensor_window itself */
+		i2c_w_mask(sd, OV7670_REG_COM7,
+			 qvga ? OV7670_COM7_FMT_QVGA : OV7670_COM7_FMT_VGA,
+			 OV7670_COM7_FMT_MASK);
+		break;
+	case SEN_OV6620:
+		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
+		break;
+	case SEN_OV6630:
+		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/******** Palette-specific regs ********/
+/* Need to do work here for the OV7670 */
+
+		if (sd->sensor == SEN_OV7610 || sd->sensor == SEN_OV76BE) {
+			/* not valid on the OV6620/OV7620/6630? */
+			i2c_w_mask(sd, 0x0e, 0x00, 0x40);
+		}
+
+		/* The OV518 needs special treatment. Although both the OV518
+		 * and the OV6630 support a 16-bit video bus, only the 8 bit Y
+		 * bus is actually used. The UV bus is tied to ground.
+		 * Therefore, the OV6630 needs to be in 8-bit multiplexed
+		 * output mode */
+
+		/* OV7640 is 8-bit only */
+
+		if (sd->sensor != SEN_OV6630 && sd->sensor != SEN_OV7640)
+			i2c_w_mask(sd, 0x13, 0x00, 0x20);
+/*	} */
+
+	/******** Clock programming ********/
+	/* The OV6620 needs special handling. This prevents the
+	 * severe banding that normally occurs */
+	if (sd->sensor == SEN_OV6620) {
+
+		/* Clock down */
+		i2c_w(sd, 0x2a, 0x04);
+		i2c_w(sd, 0x11, win->clockdiv);
+		i2c_w(sd, 0x2a, 0x84);
+		/* This next setting is critical. It seems to improve
+		 * the gain or the contrast. The "reserved" bits seem
+		 * to have some effect in this case. */
+		i2c_w(sd, 0x2d, 0x85);
+	} else if (win->clockdiv >= 0) {
+		i2c_w(sd, 0x11, win->clockdiv);
+	}
+
+	/******** Special Features ********/
+/* no evidence this is possible with OV7670, either */
+	/* Test Pattern */
+	if (sd->sensor != SEN_OV7640 && sd->sensor != SEN_OV7670)
+		i2c_w_mask(sd, 0x12, 0x00, 0x02);
+
+	/* Enable auto white balance */
+	if (sd->sensor == SEN_OV7670)
+		i2c_w_mask(sd, OV7670_REG_COM8, OV7670_COM8_AWB,
+				OV7670_COM8_AWB);
+	else
+		i2c_w_mask(sd, 0x12, 0x04, 0x04);
+
+	/* This will go away as soon as ov51x_mode_init_sensor_regs() */
+	/* is fully tested. */
+	/* 7620/6620/6630? don't have register 0x35, so play it safe */
+	if (sd->sensor == SEN_OV7610 || sd->sensor == SEN_OV76BE) {
+		if (win->width == 640 /*&& win->height == 480*/)
+			i2c_w(sd, 0x35, 0x9e);
+		else
+			i2c_w(sd, 0x35, 0x1e);
+	}
+	return 0;
+}
+
+static int set_ov_sensor_window(struct sd *sd,
+				struct ovsensor_window *win)
+{
+	int hwsbase, hwebase, vwsbase, vwebase, hwscale, vwscale;
+	int ret, hstart, hstop, vstop, vstart;
+	__u8 v;
+
+	/* The different sensor ICs handle setting up of window differently.
+	 * IF YOU SET IT WRONG, YOU WILL GET ALL ZERO ISOC DATA FROM OV51x!! */
+	switch (sd->sensor) {
+	case SEN_OV8610:
+		hwsbase = 0x1e;
+		hwebase = 0x1e;
+		vwsbase = 0x02;
+		vwebase = 0x02;
+		break;
+	case SEN_OV7610:
+	case SEN_OV76BE:
+		hwsbase = 0x38;
+		hwebase = 0x3a;
+		vwsbase = vwebase = 0x05;
+		break;
+	case SEN_OV6620:
+	case SEN_OV6630:
+		hwsbase = 0x38;
+		hwebase = 0x3a;
+		vwsbase = 0x05;
+		vwebase = 0x06;
+		break;
+	case SEN_OV7620:
+		hwsbase = 0x2f;		/* From 7620.SET (spec is wrong) */
+		hwebase = 0x2f;
+		vwsbase = vwebase = 0x05;
+		break;
+	case SEN_OV7640:
+		hwsbase = 0x1a;
+		hwebase = 0x1a;
+		vwsbase = vwebase = 0x03;
+		break;
+	case SEN_OV7670:
+		/*handling of OV7670 hardware sensor start and stop values
+		 * is very odd, compared to the other OV sensors */
+		vwsbase = vwebase = hwebase = hwsbase = 0x00;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	switch (sd->sensor) {
+	case SEN_OV6620:
+	case SEN_OV6630:
+		if (win->quarter) {	/* QCIF */
+			hwscale = 0;
+			vwscale = 0;
+		} else {		/* CIF */
+			hwscale = 1;
+			vwscale = 1;	/* The datasheet says 0;
+					 * it's wrong */
+		}
+		break;
+	case SEN_OV8610:
+		if (win->quarter) {	/* QSVGA */
+			hwscale = 1;
+			vwscale = 1;
+		} else {		/* SVGA */
+			hwscale = 2;
+			vwscale = 2;
+		}
+		break;
+	default:			/* SEN_OV7xx0 */
+		if (win->quarter) {	/* QVGA */
+			hwscale = 1;
+			vwscale = 0;
+		} else {		/* VGA */
+			hwscale = 2;
+			vwscale = 1;
+		}
+	}
+
+	ret = mode_init_ov_sensor_regs(sd, win);
+	if (ret < 0)
+		return ret;
+
+	if (sd->sensor == SEN_OV8610) {
+		i2c_w_mask(sd, 0x2d, 0x05, 0x40);
+				/* old 0x95, new 0x05 from windrv 090403 */
+						/* bits 5-7: reserved */
+		i2c_w_mask(sd, 0x28, 0x20, 0x20);
+					/* bit 5: progressive mode on */
+	}
+
+	/* The below is wrong for OV7670s because their window registers
+	 * only store the high bits in 0x17 to 0x1a */
+
+	/* SRH Use sd->max values instead of requested win values */
+	/* SCS Since we're sticking with only the max hardware widths
+	 * for a given mode */
+	/* I can hard code this for OV7670s */
+	/* Yes, these numbers do look odd, but they're tested and work! */
+	if (sd->sensor == SEN_OV7670) {
+		if (win->quarter) {	/* QVGA from ov7670.c by
+					 * Jonathan Corbet */
+			hstart = 164;
+			hstop = 20;
+			vstart = 14;
+			vstop = 494;
+		} else {		/* VGA */
+			hstart = 158;
+			hstop = 14;
+			vstart = 10;
+			vstop = 490;
+		}
+		/* OV7670 hardware window registers are split across
+		 * multiple locations */
+		i2c_w(sd, OV7670_REG_HSTART, (hstart >> 3) & 0xff);
+		i2c_w(sd, OV7670_REG_HSTOP, (hstop >> 3) & 0xff);
+		v = i2c_r(sd, OV7670_REG_HREF);
+		v = (v & 0xc0) | ((hstop & 0x7) << 3) | (hstart & 0x07);
+		msleep(10);	/* need to sleep between read and write to
+				 * same reg! */
+		i2c_w(sd, OV7670_REG_HREF, v);
+
+		i2c_w(sd, OV7670_REG_VSTART, (vstart >> 2) & 0xff);
+		i2c_w(sd, OV7670_REG_VSTOP, (vstop >> 2) & 0xff);
+		v = i2c_r(sd, OV7670_REG_VREF);
+		v = (v & 0xc0) | ((vstop & 0x3) << 2) | (vstart & 0x03);
+		msleep(10);	/* need to sleep between read and write to
+				 * same reg! */
+		i2c_w(sd, OV7670_REG_VREF, v);
+
+	} else {
+		i2c_w(sd, 0x17, hwsbase + (win->x >> hwscale));
+		i2c_w(sd, 0x18, hwebase + ((win->x + win->width) >> hwscale));
+		i2c_w(sd, 0x19, vwsbase + (win->y >> vwscale));
+		i2c_w(sd, 0x1a, vwebase + ((win->y + win->height) >> vwscale));
+	}
+	return 0;
+}
+
+static int ov_sensor_mode_setup(struct sd *sd,
+				int width, int height)
+{
+	struct ovsensor_window win;
+
+/*	win.format = mode; */
+
+	/* Unless subcapture is enabled,
+	 * center the image window and downsample
+	 * if possible to increase the field of view */
+	/* NOTE: OV518(+) and OV519 does downsampling on its own */
+	win.width = width;
+	win.height = height;
+	if (width == sd->maxwidth)
+		win.quarter = 0;
+	else
+		win.quarter = 1;
+
+	/* Center it */
+	win.x = (win.width - width) / 2;
+	win.y = (win.height - height) / 2;
+
+	/* Clock is determined by OV519 frame rate code */
+	win.clockdiv = sd->clockdiv;
+
+	PDEBUG(D_CONF, "Setting clock divider to %d", win.clockdiv);
+	return set_ov_sensor_window(sd, &win);
+}
+
+/* -- start the camera -- */
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int ret;
+
+
+	ret = ov519_mode_init_regs(sd, gspca_dev->width, gspca_dev->height);
+	if (ret < 0)
+		goto out;
+	ret = ov_sensor_mode_setup(sd, gspca_dev->width, gspca_dev->height);
+	if (ret < 0)
+		goto out;
+
+	ret = ov51x_restart((struct sd *) gspca_dev);
+	if (ret < 0)
+		goto out;
+	PDEBUG(D_STREAM, "camera started alt: 0x%02x", gspca_dev->alt);
+	ov51x_led_control(sd, 1);
+	return;
+out:
+	PDEBUG(D_ERR, "camera start error:%d", ret);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	ov51x_stop((struct sd *) gspca_dev);
+	ov51x_led_control((struct sd *) gspca_dev, 0);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	/* Header of ov519 is 16 bytes:
+	 *     Byte     Value      Description
+	 *	0	0xff	magic
+	 *	1	0xff	magic
+	 *	2	0xff	magic
+	 *	3	0xXX	0x50 = SOF, 0x51 = EOF
+	 *	9	0xXX	0x01 initial frame without data,
+	 *			0x00 standard frame with image
+	 *	14	Lo	in EOF: length of image data / 8
+	 *	15	Hi
+	 */
+
+	if (data[0] == 0xff && data[1] == 0xff && data[2] == 0xff) {
+		switch (data[3]) {
+		case 0x50:		/* start of frame */
+#define HDRSZ 16
+			data += HDRSZ;
+			len -= HDRSZ;
+#undef HDRSZ
+			if (data[0] == 0xff || data[1] == 0xd8)
+				gspca_frame_add(gspca_dev, FIRST_PACKET, frame,
+						data, len);
+			else
+				gspca_dev->last_packet_type = DISCARD_PACKET;
+			return;
+		case 0x51:		/* end of frame */
+			if (data[9] != 0)
+				gspca_dev->last_packet_type = DISCARD_PACKET;
+			gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					data, 0);
+			return;
+		}
+	}
+
+	/* intermediate packet */
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame,
+			data, len);
+}
+
+/* -- management routines -- */
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int val;
+/*	int was_streaming; */
+
+	val = sd->brightness;
+	PDEBUG(D_CONF, "brightness:%d", val);
+/*	was_streaming = gspca_dev->streaming;
+ *	if (was_streaming)
+ *		ov51x_stop(sd); */
+	switch (sd->sensor) {
+	case SEN_OV8610:
+	case SEN_OV7610:
+	case SEN_OV76BE:
+	case SEN_OV6620:
+	case SEN_OV6630:
+	case SEN_OV7640:
+		i2c_w(sd, OV7610_REG_BRT, val);
+		break;
+	case SEN_OV7620:
+		/* 7620 doesn't like manual changes when in auto mode */
+/*fixme
+ *		if (!sd->auto_brt) */
+			i2c_w(sd, OV7610_REG_BRT, val);
+		break;
+	case SEN_OV7670:
+/*jfm - from windblows
+ *		i2c_w_mask(sd, OV7670_REG_COM8, 0, OV7670_COM8_AEC); */
+		i2c_w(sd, OV7670_REG_BRIGHT, ov7670_abs_to_sm(val));
+		break;
+	}
+/*	if (was_streaming)
+ *		ov51x_restart(sd); */
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int val;
+/*	int was_streaming; */
+
+	val = sd->contrast;
+	PDEBUG(D_CONF, "contrast:%d", val);
+/*	was_streaming = gspca_dev->streaming;
+	if (was_streaming)
+		ov51x_stop(sd); */
+	switch (sd->sensor) {
+	case SEN_OV7610:
+	case SEN_OV6620:
+		i2c_w(sd, OV7610_REG_CNT, val);
+		break;
+	case SEN_OV6630:
+		i2c_w_mask(sd, OV7610_REG_CNT, val >> 4, 0x0f);
+	case SEN_OV8610: {
+		static __u8 ctab[] = {
+			0x03, 0x09, 0x0b, 0x0f, 0x53, 0x6f, 0x35, 0x7f
+		};
+
+		/* Use Y gamma control instead. Bit 0 enables it. */
+		i2c_w(sd, 0x64, ctab[val >> 5]);
+		break;
+	    }
+	case SEN_OV7620: {
+		static __u8 ctab[] = {
+			0x01, 0x05, 0x09, 0x11, 0x15, 0x35, 0x37, 0x57,
+			0x5b, 0xa5, 0xa7, 0xc7, 0xc9, 0xcf, 0xef, 0xff
+		};
+
+		/* Use Y gamma control instead. Bit 0 enables it. */
+		i2c_w(sd, 0x64, ctab[val >> 4]);
+		break;
+	    }
+	case SEN_OV7640:
+		/* Use gain control instead. */
+		i2c_w(sd, OV7610_REG_GAIN, val >> 2);
+		break;
+	case SEN_OV7670:
+		/* check that this isn't just the same as ov7610 */
+		i2c_w(sd, OV7670_REG_CONTRAS, val >> 1);
+		break;
+	}
+/*	if (was_streaming)
+		ov51x_restart(sd); */
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int val;
+/*	int was_streaming; */
+
+	val = sd->colors;
+	PDEBUG(D_CONF, "saturation:%d", val);
+/*	was_streaming = gspca_dev->streaming;
+	if (was_streaming)
+		ov51x_stop(sd); */
+	switch (sd->sensor) {
+	case SEN_OV8610:
+	case SEN_OV7610:
+	case SEN_OV76BE:
+	case SEN_OV6620:
+	case SEN_OV6630:
+		i2c_w(sd, OV7610_REG_SAT, val);
+		break;
+	case SEN_OV7620:
+		/* Use UV gamma control instead. Bits 0 & 7 are reserved. */
+/*		rc = ov_i2c_write(sd->dev, 0x62, (val >> 9) & 0x7e);
+		if (rc < 0)
+			goto out; */
+		i2c_w(sd, OV7610_REG_SAT, val);
+		break;
+	case SEN_OV7640:
+		i2c_w(sd, OV7610_REG_SAT, val & 0xf0);
+		break;
+	case SEN_OV7670:
+		/* supported later once I work out how to do it
+		 * transparently fail now! */
+		/* set REG_COM13 values for UV sat auto mode */
+		break;
+	}
+/*	if (was_streaming)
+		ov51x_restart(sd); */
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->colors;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x041e, 0x4052), DVNM("Creative Live! VISTA IM")},
+	{USB_DEVICE(0x041e, 0x405f), DVNM("Creative Live! VISTA VF0330")},
+	{USB_DEVICE(0x041e, 0x4060), DVNM("Creative Live! VISTA VF0350")},
+	{USB_DEVICE(0x041e, 0x4061), DVNM("Creative Live! VISTA VF0400")},
+	{USB_DEVICE(0x041e, 0x4064), DVNM("Creative Live! VISTA VF0420")},
+	{USB_DEVICE(0x041e, 0x4068), DVNM("Creative Live! VISTA VF0470")},
+	{USB_DEVICE(0x045e, 0x028c), DVNM("Microsoft xbox cam")},
+	{USB_DEVICE(0x054c, 0x0154), DVNM("Sonny toy4")},
+	{USB_DEVICE(0x054c, 0x0155), DVNM("Sonny toy5")},
+	{USB_DEVICE(0x05a9, 0x0519), DVNM("OmniVision")},
+	{USB_DEVICE(0x05a9, 0x0530), DVNM("OmniVision")},
+	{USB_DEVICE(0x05a9, 0x4519), DVNM("OmniVision")},
+	{USB_DEVICE(0x05a9, 0x8519), DVNM("OmniVision")},
+	{}
+};
+#undef DVNAME
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
+
+module_param(frame_rate, int, 0644);
+MODULE_PARM_DESC(frame_rate, "Frame rate (5, 10, 15, 20 or 30 fps)");
diff --git a/drivers/media/video/gspca/pac207.c b/drivers/media/video/gspca/pac207.c
index 482ef4a6afc0..72a5b89cd59d 100644
--- a/drivers/media/video/gspca/pac207.c
+++ b/drivers/media/video/gspca/pac207.c
@@ -27,8 +27,8 @@
 
 #include "gspca.h"
 
-#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(0, 2, 15)
-static const char version[] = "0.2.15";
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
 
 MODULE_AUTHOR("Hans de Goede <j.w.r.degoede@hhs.nl>");
 MODULE_DESCRIPTION("Pixart PAC207");
@@ -297,7 +297,6 @@ static int sd_open(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 
 	sd->autogain = 1;
-
 	return 0;
 }
 
@@ -338,7 +337,7 @@ static void sd_start(struct gspca_dev *gspca_dev)
 
 	pac207_write_reg(gspca_dev, 0x13, 0x01); /* Bit 0, auto clear */
 	pac207_write_reg(gspca_dev, 0x1c, 0x01); /* not documented */
-	udelay(1000); /* taken from gspca */
+	msleep(10);
 	pac207_write_reg(gspca_dev, 0x40, 0x01); /* Start ISO pipe */
 
 	sd->sof_read = 0;
@@ -743,8 +742,7 @@ static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 				PDEBUG(D_STREAM, "Incomplete frame");
 		}
 		pac207_decode_frame_init(gspca_dev);
-		gspca_frame_add(gspca_dev, FIRST_PACKET, frame, NULL,
-				0);
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame, NULL, 0);
 		len -= sof - data;
 		data = sof;
 	}
diff --git a/drivers/media/video/gspca/pac7311.c b/drivers/media/video/gspca/pac7311.c
new file mode 100644
index 000000000000..8f51976db995
--- /dev/null
+++ b/drivers/media/video/gspca/pac7311.c
@@ -0,0 +1,754 @@
+/*
+ *		Pixart PAC7311 library
+ *		Copyright (C) 2005 Thomas Kaiser thomas@kaiser-linux.li
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "pac7311"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Thomas Kaiser thomas@kaiser-linux.li");
+MODULE_DESCRIPTION("Pixart PAC7311");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;		/* !! must be the first item */
+
+	int avg_lum;
+
+	unsigned char brightness;
+#define BRIGHTNESS_MAX 0x20
+	unsigned char contrast;
+	unsigned char colors;
+	unsigned char autogain;
+
+	char ffseq;
+	signed char ag_cnt;
+#define AG_CNT_START 13
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = BRIGHTNESS_MAX,
+		.step    = 1,
+		.default_value = 0x10,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define SD_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Color",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+#define SD_AUTOGAIN 3
+	{
+	    {
+		.id      = V4L2_CID_AUTOGAIN,
+		.type    = V4L2_CTRL_TYPE_BOOLEAN,
+		.name    = "Auto Gain",
+		.minimum = 0,
+		.maximum = 1,
+		.step    = 1,
+		.default_value = 1,
+	    },
+	    .set = sd_setautogain,
+	    .get = sd_getautogain,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 160, 120, 2},
+	{V4L2_PIX_FMT_JPEG, 320, 240, 1},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 0},
+};
+
+#define PAC7311_JPEG_HEADER_SIZE (sizeof pac7311_jpeg_header)	/* (594) */
+
+const unsigned char pac7311_jpeg_header[] = {
+	0xff, 0xd8,
+	0xff, 0xe0, 0x00, 0x03, 0x20,
+	0xff, 0xc0, 0x00, 0x11, 0x08,
+		0x01, 0xe0,			/* 12: height */
+		0x02, 0x80,			/* 14: width */
+		0x03,				/* 16 */
+			0x01, 0x21, 0x00,
+			0x02, 0x11, 0x01,
+			0x03, 0x11, 0x01,
+	0xff, 0xdb, 0x00, 0x84,
+	0x00, 0x10, 0x0b, 0x0c, 0x0e, 0x0c, 0x0a, 0x10, 0x0e, 0x0d,
+	0x0e, 0x12, 0x11, 0x10, 0x13, 0x18, 0x28, 0x1a, 0x18, 0x16,
+	0x16, 0x18, 0x31, 0x23, 0x25, 0x1d, 0x28, 0x3a, 0x33, 0x3d,
+	0x3c, 0x39, 0x33, 0x38, 0x37, 0x40, 0x48, 0x5c, 0x4e, 0x40,
+	0x44, 0x57, 0x45, 0x37, 0x38, 0x50, 0x6d, 0x51, 0x57, 0x5f,
+	0x62, 0x67, 0x68, 0x67, 0x3e, 0x4d, 0x71, 0x79, 0x70, 0x64,
+	0x78, 0x5c, 0x65, 0x67, 0x63, 0x01, 0x11, 0x12, 0x12, 0x18,
+	0x15, 0x18, 0x2f, 0x1a, 0x1a, 0x2f, 0x63, 0x42, 0x38, 0x42,
+	0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63,
+	0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63,
+	0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63,
+	0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63,
+	0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63, 0x63,
+	0xff, 0xc4, 0x01, 0xa2, 0x00, 0x00, 0x01, 0x05, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08,
+	0x09, 0x0a, 0x0b, 0x10, 0x00, 0x02, 0x01, 0x03, 0x03, 0x02,
+	0x04, 0x03, 0x05, 0x05, 0x04, 0x04, 0x00, 0x00, 0x01, 0x7d,
+	0x01, 0x02, 0x03, 0x00, 0x04, 0x11, 0x05, 0x12, 0x21, 0x31,
+	0x41, 0x06, 0x13, 0x51, 0x61, 0x07, 0x22, 0x71, 0x14, 0x32,
+	0x81, 0x91, 0xa1, 0x08, 0x23, 0x42, 0xb1, 0xc1, 0x15, 0x52,
+	0xd1, 0xf0, 0x24, 0x33, 0x62, 0x72, 0x82, 0x09, 0x0a, 0x16,
+	0x17, 0x18, 0x19, 0x1a, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a,
+	0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3a, 0x43, 0x44, 0x45,
+	0x46, 0x47, 0x48, 0x49, 0x4a, 0x53, 0x54, 0x55, 0x56, 0x57,
+	0x58, 0x59, 0x5a, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69,
+	0x6a, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x83,
+	0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x92, 0x93, 0x94,
+	0x95, 0x96, 0x97, 0x98, 0x99, 0x9a, 0xa2, 0xa3, 0xa4, 0xa5,
+	0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6,
+	0xb7, 0xb8, 0xb9, 0xba, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7,
+	0xc8, 0xc9, 0xca, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, 0xd8,
+	0xd9, 0xda, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8,
+	0xe9, 0xea, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8,
+	0xf9, 0xfa, 0x01, 0x00, 0x03, 0x01, 0x01, 0x01, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a,
+	0x0b, 0x11, 0x00, 0x02, 0x01, 0x02, 0x04, 0x04, 0x03, 0x04,
+	0x07, 0x05, 0x04, 0x04, 0x00, 0x01, 0x02, 0x77, 0x00, 0x01,
+	0x02, 0x03, 0x11, 0x04, 0x05, 0x21, 0x31, 0x06, 0x12, 0x41,
+	0x51, 0x07, 0x61, 0x71, 0x13, 0x22, 0x32, 0x81, 0x08, 0x14,
+	0x42, 0x91, 0xa1, 0xb1, 0xc1, 0x09, 0x23, 0x33, 0x52, 0xf0,
+	0x15, 0x62, 0x72, 0xd1, 0x0a, 0x16, 0x24, 0x34, 0xe1, 0x25,
+	0xf1, 0x17, 0x18, 0x19, 0x1a, 0x26, 0x27, 0x28, 0x29, 0x2a,
+	0x35, 0x36, 0x37, 0x38, 0x39, 0x3a, 0x43, 0x44, 0x45, 0x46,
+	0x47, 0x48, 0x49, 0x4a, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58,
+	0x59, 0x5a, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6a,
+	0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x82, 0x83,
+	0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8a, 0x92, 0x93, 0x94,
+	0x95, 0x96, 0x97, 0x98, 0x99, 0x9a, 0xa2, 0xa3, 0xa4, 0xa5,
+	0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6,
+	0xb7, 0xb8, 0xb9, 0xba, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7,
+	0xc8, 0xc9, 0xca, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, 0xd8,
+	0xd9, 0xda, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9,
+	0xea, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xf9, 0xfa,
+	0xff, 0xda, 0x00, 0x0c, 0x03, 0x01, 0x00, 0x02, 0x11, 0x03,
+	0x11, 0x00, 0x3f, 0x00
+};
+
+static void reg_w(struct usb_device *dev,
+			    __u16 req,
+			    __u16 value,
+			    __u16 index,
+			    __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			req,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, buffer, length,
+			500);
+}
+
+static void pac7311_reg_read(struct usb_device *dev, __u16 index,
+			    __u8 *buffer)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,			/* request */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,			/* value */
+			index, buffer, 1,
+			500);
+}
+
+static void pac7311_reg_write(struct usb_device *dev,
+			      __u16 index,
+			      __u8 value)
+{
+	__u8 buf;
+
+	buf = value;
+	reg_w(dev, 0x00, value, index, &buf, 1);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	struct cam *cam;
+
+	PDEBUG(D_CONF, "Find Sensor PAC7311");
+	pac7311_reg_write(dev, 0x78, 0x40); /* Bit_0=start stream, Bit_7=LED */
+	pac7311_reg_write(dev, 0x78, 0x40); /* Bit_0=start stream, Bit_7=LED */
+	pac7311_reg_write(dev, 0x78, 0x44); /* Bit_0=start stream, Bit_7=LED */
+	pac7311_reg_write(dev, 0xff, 0x04);
+	pac7311_reg_write(dev, 0x27, 0x80);
+	pac7311_reg_write(dev, 0x28, 0xca);
+	pac7311_reg_write(dev, 0x29, 0x53);
+	pac7311_reg_write(dev, 0x2a, 0x0e);
+	pac7311_reg_write(dev, 0xff, 0x01);
+	pac7311_reg_write(dev, 0x3e, 0x20);
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x05;
+	cam->cam_mode = vga_mode;
+	cam->nmodes = ARRAY_SIZE(vga_mode);
+
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	sd->autogain = sd_ctrls[SD_AUTOGAIN].qctrl.default_value;
+	return 0;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int brightness;
+
+/*jfm: inverted?*/
+	brightness = BRIGHTNESS_MAX - sd->brightness;
+	pac7311_reg_write(gspca_dev->dev, 0xff, 0x04);
+	/* pac7311_reg_write(gspca_dev->dev, 0x0e, 0x00); */
+	pac7311_reg_write(gspca_dev->dev, 0x0f, brightness);
+	/* load registers to sensor (Bit 0, auto clear) */
+	pac7311_reg_write(gspca_dev->dev, 0x11, 0x01);
+	PDEBUG(D_CONF|D_STREAM, "brightness: %i", brightness);
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	pac7311_reg_write(gspca_dev->dev, 0xff, 0x01);
+	pac7311_reg_write(gspca_dev->dev, 0x80, sd->contrast);
+	/* load registers to sensor (Bit 0, auto clear) */
+	pac7311_reg_write(gspca_dev->dev, 0x11, 0x01);
+	PDEBUG(D_CONF|D_STREAM, "contrast: %i", sd->contrast);
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	pac7311_reg_write(gspca_dev->dev, 0xff, 0x01);
+	pac7311_reg_write(gspca_dev->dev, 0x10, sd->colors);
+	/* load registers to sensor (Bit 0, auto clear) */
+	pac7311_reg_write(gspca_dev->dev, 0x11, 0x01);
+	PDEBUG(D_CONF|D_STREAM, "color: %i", sd->colors);
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	pac7311_reg_write(gspca_dev->dev, 0x78, 0x00);	/* Turn on LED */
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	pac7311_reg_write(dev, 0xff, 0x01);
+	reg_w(dev, 0x01, 0, 0x0002, "\x48\x0a\x40\x08\x00\x00\x08\x00", 8);
+	reg_w(dev, 0x01, 0, 0x000a, "\x06\xff\x11\xff\x5a\x30\x90\x4c", 8);
+	reg_w(dev, 0x01, 0, 0x0012, "\x00\x07\x00\x0a\x10\x00\xa0\x10", 8);
+	reg_w(dev, 0x01, 0, 0x001a, "\x02\x00\x00\x00\x00\x0b\x01\x00", 8);
+	reg_w(dev, 0x01, 0, 0x0022, "\x00\x00\x00\x00\x00\x00\x00\x00", 8);
+	reg_w(dev, 0x01, 0, 0x002a, "\x00\x00\x00", 3);
+	reg_w(dev, 0x01, 0, 0x003e, "\x00\x00\x78\x52\x4a\x52\x78\x6e", 8);
+	reg_w(dev, 0x01, 0, 0x0046, "\x48\x46\x48\x6e\x5f\x49\x42\x49", 8);
+	reg_w(dev, 0x01, 0, 0x004e, "\x5f\x5f\x49\x42\x49\x5f\x6e\x48", 8);
+	reg_w(dev, 0x01, 0, 0x0056, "\x46\x48\x6e\x78\x52\x4a\x52\x78", 8);
+	reg_w(dev, 0x01, 0, 0x005e, "\x00\x00\x09\x1b\x34\x49\x5c\x9b", 8);
+	reg_w(dev, 0x01, 0, 0x0066, "\xd0\xff", 2);
+	reg_w(dev, 0x01, 0, 0x0078, "\x44\x00\xf2\x01\x01\x80", 6);
+	reg_w(dev, 0x01, 0, 0x007f, "\x2a\x1c\x00\xc8\x02\x58\x03\x84", 8);
+	reg_w(dev, 0x01, 0, 0x0087, "\x12\x00\x1a\x04\x08\x0c\x10\x14", 8);
+	reg_w(dev, 0x01, 0, 0x008f, "\x18\x20", 2);
+	reg_w(dev, 0x01, 0, 0x0096, "\x01\x08\x04", 3);
+	reg_w(dev, 0x01, 0, 0x00a0, "\x44\x44\x44\x04", 4);
+	reg_w(dev, 0x01, 0, 0x00f0, "\x01\x00\x00\x00\x22\x00\x20\x00", 8);
+	reg_w(dev, 0x01, 0, 0x00f8, "\x3f\x00\x0a\x01\x00", 5);
+
+	pac7311_reg_write(dev, 0xff, 0x04);
+	pac7311_reg_write(dev, 0x02, 0x04);
+	pac7311_reg_write(dev, 0x03, 0x54);
+	pac7311_reg_write(dev, 0x04, 0x07);
+	pac7311_reg_write(dev, 0x05, 0x2b);
+	pac7311_reg_write(dev, 0x06, 0x09);
+	pac7311_reg_write(dev, 0x07, 0x0f);
+	pac7311_reg_write(dev, 0x08, 0x09);
+	pac7311_reg_write(dev, 0x09, 0x00);
+	pac7311_reg_write(dev, 0x0c, 0x07);
+	pac7311_reg_write(dev, 0x0d, 0x00);
+	pac7311_reg_write(dev, 0x0e, 0x00);
+	pac7311_reg_write(dev, 0x0f, 0x62);
+	pac7311_reg_write(dev, 0x10, 0x08);
+	pac7311_reg_write(dev, 0x12, 0x07);
+	pac7311_reg_write(dev, 0x13, 0x00);
+	pac7311_reg_write(dev, 0x14, 0x00);
+	pac7311_reg_write(dev, 0x15, 0x00);
+	pac7311_reg_write(dev, 0x16, 0x00);
+	pac7311_reg_write(dev, 0x17, 0x00);
+	pac7311_reg_write(dev, 0x18, 0x00);
+	pac7311_reg_write(dev, 0x19, 0x00);
+	pac7311_reg_write(dev, 0x1a, 0x00);
+	pac7311_reg_write(dev, 0x1b, 0x03);
+	pac7311_reg_write(dev, 0x1c, 0xa0);
+	pac7311_reg_write(dev, 0x1d, 0x01);
+	pac7311_reg_write(dev, 0x1e, 0xf4);
+	pac7311_reg_write(dev, 0x21, 0x00);
+	pac7311_reg_write(dev, 0x22, 0x08);
+	pac7311_reg_write(dev, 0x24, 0x03);
+	pac7311_reg_write(dev, 0x26, 0x00);
+	pac7311_reg_write(dev, 0x27, 0x01);
+	pac7311_reg_write(dev, 0x28, 0xca);
+	pac7311_reg_write(dev, 0x29, 0x10);
+	pac7311_reg_write(dev, 0x2a, 0x06);
+	pac7311_reg_write(dev, 0x2b, 0x78);
+	pac7311_reg_write(dev, 0x2c, 0x00);
+	pac7311_reg_write(dev, 0x2d, 0x00);
+	pac7311_reg_write(dev, 0x2e, 0x00);
+	pac7311_reg_write(dev, 0x2f, 0x00);
+	pac7311_reg_write(dev, 0x30, 0x23);
+	pac7311_reg_write(dev, 0x31, 0x28);
+	pac7311_reg_write(dev, 0x32, 0x04);
+	pac7311_reg_write(dev, 0x33, 0x11);
+	pac7311_reg_write(dev, 0x34, 0x00);
+	pac7311_reg_write(dev, 0x35, 0x00);
+	pac7311_reg_write(dev, 0x11, 0x01);
+	setcontrast(gspca_dev);
+	setbrightness(gspca_dev);
+	setcolors(gspca_dev);
+
+	/* set correct resolution */
+	switch (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+	case 2:					/* 160x120 */
+		pac7311_reg_write(dev, 0xff, 0x04);
+		pac7311_reg_write(dev, 0x02, 0x03);
+		pac7311_reg_write(dev, 0xff, 0x01);
+		pac7311_reg_write(dev, 0x08, 0x09);
+		pac7311_reg_write(dev, 0x17, 0x20);
+		pac7311_reg_write(dev, 0x1b, 0x00);
+/*		pac7311_reg_write(dev, 0x80, 0x69); */
+		pac7311_reg_write(dev, 0x87, 0x10);
+		break;
+	case 1:					/* 320x240 */
+		pac7311_reg_write(dev, 0xff, 0x04);
+		pac7311_reg_write(dev, 0x02, 0x03);
+		pac7311_reg_write(dev, 0xff, 0x01);
+		pac7311_reg_write(dev, 0x08, 0x09);
+		pac7311_reg_write(dev, 0x17, 0x30);
+/*		pac7311_reg_write(dev, 0x80, 0x3f); */
+		pac7311_reg_write(dev, 0x87, 0x11);
+		break;
+	case 0:					/* 640x480 */
+		pac7311_reg_write(dev, 0xff, 0x04);
+		pac7311_reg_write(dev, 0x02, 0x03);
+		pac7311_reg_write(dev, 0xff, 0x01);
+		pac7311_reg_write(dev, 0x08, 0x08);
+		pac7311_reg_write(dev, 0x17, 0x00);
+/*		pac7311_reg_write(dev, 0x80, 0x1c); */
+		pac7311_reg_write(dev, 0x87, 0x12);
+		break;
+	}
+
+	/* start stream */
+	pac7311_reg_write(dev, 0xff, 0x01);
+	pac7311_reg_write(dev, 0x78, 0x04);
+	pac7311_reg_write(dev, 0x78, 0x05);
+
+	if (sd->autogain) {
+		sd->ag_cnt = AG_CNT_START;
+		sd->avg_lum = 0;
+	} else {
+		sd->ag_cnt = -1;
+	}
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	pac7311_reg_write(dev, 0xff, 0x04);
+	pac7311_reg_write(dev, 0x27, 0x80);
+	pac7311_reg_write(dev, 0x28, 0xca);
+	pac7311_reg_write(dev, 0x29, 0x53);
+	pac7311_reg_write(dev, 0x2a, 0x0e);
+	pac7311_reg_write(dev, 0xff, 0x01);
+	pac7311_reg_write(dev, 0x3e, 0x20);
+	pac7311_reg_write(dev, 0x78, 0x04); /* Bit_0=start stream, Bit_7=LED */
+	pac7311_reg_write(dev, 0x78, 0x44); /* Bit_0=start stream, Bit_7=LED */
+	pac7311_reg_write(dev, 0x78, 0x44); /* Bit_0=start stream, Bit_7=LED */
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+/* this function is called at close time */
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	pac7311_reg_write(dev, 0xff, 0x04);
+	pac7311_reg_write(dev, 0x27, 0x80);
+	pac7311_reg_write(dev, 0x28, 0xca);
+	pac7311_reg_write(dev, 0x29, 0x53);
+	pac7311_reg_write(dev, 0x2a, 0x0e);
+	pac7311_reg_write(dev, 0xff, 0x01);
+	pac7311_reg_write(dev, 0x3e, 0x20);
+	pac7311_reg_write(dev, 0x78, 0x04); /* Bit_0=start stream, Bit_7=LED */
+	pac7311_reg_write(dev, 0x78, 0x44); /* Bit_0=start stream, Bit_7=LED */
+	pac7311_reg_write(dev, 0x78, 0x44); /* Bit_0=start stream, Bit_7=LED */
+}
+
+static void setautogain(struct gspca_dev *gspca_dev, int luma)
+{
+	int luma_mean = 128;
+	int luma_delta = 20;
+	__u8 spring = 5;
+	__u8 Pxclk;
+	int Gbright;
+
+	pac7311_reg_read(gspca_dev->dev, 0x02, &Pxclk);
+	Gbright = Pxclk;
+	PDEBUG(D_FRAM, "luma mean %d", luma);
+	if (luma < luma_mean - luma_delta ||
+	    luma > luma_mean + luma_delta) {
+		Gbright += (luma_mean - luma) >> spring;
+		if (Gbright > 0x1a)
+			Gbright = 0x1a;
+		else if (Gbright < 4)
+			Gbright = 4;
+		PDEBUG(D_FRAM, "gbright %d", Gbright);
+		pac7311_reg_write(gspca_dev->dev, 0xff, 0x04);
+		pac7311_reg_write(gspca_dev->dev, 0x0f, Gbright);
+		/* load registers to sensor (Bit 0, auto clear) */
+		pac7311_reg_write(gspca_dev->dev, 0x11, 0x01);
+	}
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	unsigned char tmpbuf[4];
+	int i, p, ffseq;
+
+/*	if (len < 5) { */
+	if (len < 6) {
+/*		gspca_dev->last_packet_type = DISCARD_PACKET; */
+		return;
+	}
+
+	ffseq = sd->ffseq;
+
+	for (p = 0; p < len - 6; p++) {
+		if ((data[0 + p] == 0xff)
+		    && (data[1 + p] == 0xff)
+		    && (data[2 + p] == 0x00)
+		    && (data[3 + p] == 0xff)
+		    && (data[4 + p] == 0x96)) {
+
+			/* start of frame */
+			if (sd->ag_cnt >= 0 && p > 28) {
+				sd->avg_lum += data[p - 23];
+				if (--sd->ag_cnt < 0) {
+					sd->ag_cnt = AG_CNT_START;
+					setautogain(gspca_dev,
+						sd->avg_lum / AG_CNT_START);
+					sd->avg_lum = 0;
+				}
+			}
+
+			/* copy the end of data to the current frame */
+			frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+						data, p);
+
+			/* put the JPEG header in the new frame */
+			gspca_frame_add(gspca_dev, FIRST_PACKET, frame,
+					(unsigned char *) pac7311_jpeg_header,
+					12);
+			tmpbuf[0] = gspca_dev->height >> 8;
+			tmpbuf[1] = gspca_dev->height & 0xff;
+			tmpbuf[2] = gspca_dev->width >> 8;
+			tmpbuf[3] = gspca_dev->width & 0xff;
+			gspca_frame_add(gspca_dev, INTER_PACKET, frame,
+					tmpbuf, 4);
+			gspca_frame_add(gspca_dev, INTER_PACKET, frame,
+				(unsigned char *) &pac7311_jpeg_header[16],
+				PAC7311_JPEG_HEADER_SIZE - 16);
+
+			data += p + 7;
+			len -= p + 7;
+			ffseq = 0;
+			break;
+		}
+	}
+
+	/* remove the 'ff ff ff xx' sequences */
+	switch (ffseq) {
+	case 3:
+		data += 1;
+		len -= 1;
+		break;
+	case 2:
+		if (data[0] == 0xff) {
+			data += 2;
+			len -= 2;
+			frame->data_end -= 2;
+		}
+		break;
+	case 1:
+		if (data[0] == 0xff
+		    && data[1] == 0xff) {
+			data += 3;
+			len -= 3;
+			frame->data_end -= 1;
+		}
+		break;
+	}
+	for (i = 0; i < len - 4; i++) {
+		if (data[i] == 0xff
+		    && data[i + 1] == 0xff
+		    && data[i + 2] == 0xff) {
+			memmove(&data[i], &data[i + 4], len - i - 4);
+			len -= 4;
+		}
+	}
+	ffseq = 0;
+	if (data[len - 4] == 0xff) {
+		if (data[len - 3] == 0xff
+		    && data[len - 2] == 0xff) {
+			len -= 4;
+		}
+	} else if (data[len - 3] == 0xff) {
+		if (data[len - 2] == 0xff
+		    && data[len - 1] == 0xff)
+			ffseq = 3;
+	} else if (data[len - 2] == 0xff) {
+		if (data[len - 1] == 0xff)
+			ffseq = 2;
+	} else if (data[len - 1] == 0xff)
+		ffseq = 1;
+	sd->ffseq = ffseq;
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+/*	__u8 brightness = 0;
+
+	pac7311_reg_read(gspca_dev->dev, 0x0008, &brightness);
+	spca50x->brightness = brightness;
+	return spca50x->brightness;	*/
+/*	PDEBUG(D_CONF, "Called pac7311_getbrightness: Not implemented yet"); */
+}
+
+
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+/*	getcontrast(gspca_dev); */
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+/*	getcolors(gspca_dev); */
+	*val = sd->colors;
+	return 0;
+}
+
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->autogain = val;
+	if (val) {
+		sd->ag_cnt = AG_CNT_START;
+		sd->avg_lum = 0;
+	} else {
+		sd->ag_cnt = -1;
+	}
+	return 0;
+}
+
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autogain;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x093a, 0x2600), DVNM("Typhoon")},
+	{USB_DEVICE(0x093a, 0x2601), DVNM("Philips SPC610NC")},
+	{USB_DEVICE(0x093a, 0x2603), DVNM("PAC7312")},
+	{USB_DEVICE(0x093a, 0x2608), DVNM("Trust WB-3300p")},
+	{USB_DEVICE(0x093a, 0x260e), DVNM("Gigaware VGA PC Camera, Trust WB-3350p, SIGMA cam 2350")},
+	{USB_DEVICE(0x093a, 0x260f), DVNM("SnakeCam")},
+	{USB_DEVICE(0x093a, 0x2621), DVNM("PAC731x")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/sonixb.c b/drivers/media/video/gspca/sonixb.c
new file mode 100644
index 000000000000..d26255ddfd5b
--- /dev/null
+++ b/drivers/media/video/gspca/sonixb.c
@@ -0,0 +1,879 @@
+/*
+ *		sonix sn9c102 (bayer) library
+ *		Copyright (C) 2003 2004 Michel Xhaard mxhaard@magic.fr
+ * Add Pas106 Stefano Mozzi (C) 2004
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "sonixb"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SN9C102 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned char brightness;
+	unsigned char contrast;
+
+	char sensor;			/* Type of image sensor chip */
+#define SENSOR_HV7131R 0
+#define SENSOR_OV6650 1
+#define SENSOR_OV7630 2
+#define SENSOR_OV7630_3 3
+#define SENSOR_PAS106 4
+#define SENSOR_PAS202 5
+#define SENSOR_TAS5110 6
+#define SENSOR_TAS5130CXX 7
+};
+
+#define COMP2 0x8f
+#define COMP 0xc7		/* 0x87 //0x07 */
+#define COMP1 0xc9		/* 0x89 //0x09 */
+
+#define MCK_INIT 0x63
+#define MCK_INIT1 0x20		/*fixme: Bayer - 0x50 for JPEG ??*/
+
+#define SYS_CLK 0x04
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+};
+
+/* fixme: should have V4L2_PIX_FMT_SN9C10X */
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_SN9C10X, 160, 120, 2},
+	{V4L2_PIX_FMT_SN9C10X, 320, 240, 1},
+	{V4L2_PIX_FMT_SN9C10X, 640, 480, 0},
+};
+static struct cam_mode sif_mode[] = {
+	{V4L2_PIX_FMT_SN9C10X, 176, 144, 1},
+	{V4L2_PIX_FMT_SN9C10X, 352, 288, 0},
+};
+
+static const __u8 probe_ov7630[] = {0x08, 0x44};
+
+static const __u8 initHv7131[] = {
+	0x46, 0x77, 0x00, 0x04, 0x00, 0x00, 0x00, 0x80, 0x11, 0x00, 0x00, 0x00,
+	0x00, 0x00,
+	0x00, 0x00, 0x00, 0x03, 0x01, 0x00,	/* shift from 0x02 0x01 0x00 */
+	0x28, 0x1e, 0x60, 0x8a, 0x20,
+	0x1d, 0x10, 0x02, 0x03, 0x0f, 0x0c
+};
+static const __u8 hv7131_sensor_init[][8] = {
+	{0xc0, 0x11, 0x31, 0x38, 0x2a, 0x2e, 0x00, 0x10},
+	{0xa0, 0x11, 0x01, 0x08, 0x2a, 0x2e, 0x00, 0x10},
+	{0xb0, 0x11, 0x20, 0x00, 0xd0, 0x2e, 0x00, 0x10},
+	{0xc0, 0x11, 0x25, 0x03, 0x0e, 0x28, 0x00, 0x16},
+	{0xa0, 0x11, 0x30, 0x10, 0x0e, 0x28, 0x00, 0x15},
+};
+static const __u8 initOv6650[] = {
+	0x44, 0x44, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80,
+	0x60, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x00, 0x02, 0x01, 0x0a, 0x16, 0x12, 0x68, 0x0b,
+	0x10, 0x1d, 0x10, 0x00, 0x06, 0x1f, 0x00
+};
+static const __u8 ov6650_sensor_init[][8] =
+{
+	/* Bright, contrast, etc are set througth SCBB interface.
+	 * AVCAP on win2 do not send any data on this 	controls. */
+	/* Anyway, some registers appears to alter bright and constrat */
+	{0xa0, 0x60, 0x12, 0x80, 0x00, 0x00, 0x00, 0x10},
+	{0xd0, 0x60, 0x11, 0xc0, 0x1b, 0x18, 0xc1, 0x10},
+	{0xb0, 0x60, 0x15, 0x00, 0x02, 0x18, 0xc1, 0x10},
+/*	{0xa0, 0x60, 0x1b, 0x01, 0x02, 0x18, 0xc1, 0x10},
+		 * THIS SET GREEN SCREEN
+		 * (pixels could be innverted in decode kind of "brg",
+		 * but blue wont be there. Avoid this data ... */
+	{0xd0, 0x60, 0x26, 0x01, 0x14, 0xd8, 0xa4, 0x10}, /* format out? */
+	{0xd0, 0x60, 0x26, 0x01, 0x14, 0xd8, 0xa4, 0x10},
+	{0xa0, 0x60, 0x30, 0x3d, 0x0A, 0xd8, 0xa4, 0x10},
+	{0xb0, 0x60, 0x60, 0x66, 0x68, 0xd8, 0xa4, 0x10},
+	{0xa0, 0x60, 0x68, 0x04, 0x68, 0xd8, 0xa4, 0x10},
+	{0xd0, 0x60, 0x17, 0x24, 0xd6, 0x04, 0x94, 0x10}, /* Clipreg */
+	{0xa0, 0x60, 0x10, 0x5d, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x2d, 0x0a, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x32, 0x00, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x33, 0x40, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x11, 0xc0, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x00, 0x16, 0x99, 0x04, 0x94, 0x15}, /* bright / Lumino */
+	{0xa0, 0x60, 0x2b, 0xab, 0x99, 0x04, 0x94, 0x15},
+							/* ?flicker o brillo */
+	{0xa0, 0x60, 0x2d, 0x2a, 0x99, 0x04, 0x94, 0x15},
+	{0xa0, 0x60, 0x2d, 0x2b, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x32, 0x00, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x33, 0x00, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x10, 0x57, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x2d, 0x2b, 0x99, 0x04, 0x94, 0x16},
+	{0xa0, 0x60, 0x32, 0x00, 0x99, 0x04, 0x94, 0x16},
+		/* Low Light (Enabled: 0x32 0x1 | Disabled: 0x32 0x00) */
+	{0xa0, 0x60, 0x33, 0x29, 0x99, 0x04, 0x94, 0x16},
+		/* Low Ligth (Enabled: 0x33 0x13 | Disabled: 0x33 0x29) */
+/*	{0xa0, 0x60, 0x11, 0xc1, 0x99, 0x04, 0x94, 0x16}, */
+	{0xa0, 0x60, 0x00, 0x17, 0x99, 0x04, 0x94, 0x15}, /* clip? r */
+	{0xa0, 0x60, 0x00, 0x18, 0x99, 0x04, 0x94, 0x15}, /* clip? r */
+};
+static const __u8 initOv7630[] = {
+	0x04, 0x44, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80,	/* r01 .. r08 */
+	0x21, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* r09 .. r10 */
+	0x00, 0x02, 0x01, 0x0a,				/* r11 .. r14 */
+	0x28, 0x1e,			/* H & V sizes     r15 .. r16 */
+	0x68, COMP1, MCK_INIT1,				/* r17 .. r19 */
+	0x1d, 0x10, 0x02, 0x03, 0x0f, 0x0c		/* r1a .. r1f */
+};
+static const __u8 initOv7630_3[] = {
+	0x44, 0x44, 0x00, 0x1a, 0x20, 0x20, 0x20, 0x80,	/* r01 .. r08 */
+	0x21, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x04,	/* r09 .. r10 */
+	0x00, 0x02, 0x01, 0x0a,				/* r11 .. r14 */
+	0x28, 0x1e,			/* H & V sizes     r15 .. r16 */
+	0x68, COMP1, MCK_INIT1,				/* r17 .. r19 */
+	0x1d, 0x10, 0x02, 0x03, 0x0f, 0x0c		/* r1a .. r1f */
+};
+static const __u8 ov7630_sensor_init_com[][8] = {
+	{0xa0, 0x21, 0x12, 0x80, 0x00, 0x00, 0x00, 0x10},
+	{0xb0, 0x21, 0x01, 0x77, 0x3a, 0x00, 0x00, 0x10},
+/*	{0xd0, 0x21, 0x12, 0x7c, 0x01, 0x80, 0x34, 0x10},	   jfm */
+	{0xd0, 0x21, 0x12, 0x78, 0x00, 0x80, 0x34, 0x10},	/* jfm */
+	{0xa0, 0x21, 0x1b, 0x04, 0x00, 0x80, 0x34, 0x10},
+	{0xa0, 0x21, 0x20, 0x44, 0x00, 0x80, 0x34, 0x10},
+	{0xa0, 0x21, 0x23, 0xee, 0x00, 0x80, 0x34, 0x10},
+	{0xd0, 0x21, 0x26, 0xa0, 0x9a, 0xa0, 0x30, 0x10},
+	{0xb0, 0x21, 0x2a, 0x80, 0x00, 0xa0, 0x30, 0x10},
+	{0xb0, 0x21, 0x2f, 0x3d, 0x24, 0xa0, 0x30, 0x10},
+	{0xa0, 0x21, 0x32, 0x86, 0x24, 0xa0, 0x30, 0x10},
+/*	{0xb0, 0x21, 0x60, 0xa9, 0x4a, 0xa0, 0x30, 0x10},	   jfm */
+	{0xb0, 0x21, 0x60, 0xa9, 0x42, 0xa0, 0x30, 0x10},	/* jfm */
+	{0xa0, 0x21, 0x65, 0x00, 0x42, 0xa0, 0x30, 0x10},
+	{0xa0, 0x21, 0x69, 0x38, 0x42, 0xa0, 0x30, 0x10},
+	{0xc0, 0x21, 0x6f, 0x88, 0x0b, 0x00, 0x30, 0x10},
+	{0xc0, 0x21, 0x74, 0x21, 0x8e, 0x00, 0x30, 0x10},
+	{0xa0, 0x21, 0x7d, 0xf7, 0x8e, 0x00, 0x30, 0x10},
+	{0xd0, 0x21, 0x17, 0x1c, 0xbd, 0x06, 0xf6, 0x10},
+};
+static const __u8 ov7630_sensor_init[][8] = {
+	{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00}, /* delay 200ms */
+	{0xa0, 0x21, 0x11, 0x01, 0xbd, 0x06, 0xf6, 0x10},	/* jfm */
+	{0xa0, 0x21, 0x10, 0x57, 0xbd, 0x06, 0xf6, 0x16},
+	{0xa0, 0x21, 0x76, 0x02, 0xbd, 0x06, 0xf6, 0x16},
+	{0xa0, 0x21, 0x00, 0x10, 0xbd, 0x06, 0xf6, 0x15},	/* gain */
+};
+static const __u8 ov7630_sensor_init_3[][8] = {
+	{0xa0, 0x21, 0x10, 0x36, 0xbd, 0x06, 0xf6, 0x16},	/* exposure */
+	{0xa0, 0x21, 0x76, 0x03, 0xbd, 0x06, 0xf6, 0x16},
+	{0xa0, 0x21, 0x11, 0x01, 0xbd, 0x06, 0xf6, 0x16},
+	{0xa0, 0x21, 0x00, 0x10, 0xbd, 0x06, 0xf6, 0x15},	/* gain */
+/*	{0xb0, 0x21, 0x2a, 0xc0, 0x3c, 0x06, 0xf6, 0x1d},
+		* a0 1c,a0 1f,c0 3c frame rate ?line interval from ov6630 */
+/*	{0xb0, 0x21, 0x2a, 0xa0, 0x1f, 0x06, 0xf6, 0x1d},	 * from win */
+	{0xb0, 0x21, 0x2a, 0xa0, 0x1c, 0x06, 0xf6, 0x1d},
+};
+
+static const __u8 initPas106[] = {
+	0x04, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x81, 0x40, 0x00, 0x00, 0x00,
+	0x00, 0x00,
+	0x00, 0x00, 0x00, 0x05, 0x01, 0x00,
+	0x16, 0x12, 0x28, COMP1, MCK_INIT1,
+	0x18, 0x10, 0x04, 0x03, 0x11, 0x0c
+};
+/* compression 0x86 mckinit1 0x2b */
+static const __u8 pas106_data[][2] = {
+	{0x02, 0x04},		/* Pixel Clock Divider 6 */
+	{0x03, 0x13},		/* Frame Time MSB */
+/*	{0x03, 0x12},		 * Frame Time MSB */
+	{0x04, 0x06},		/* Frame Time LSB */
+/*	{0x04, 0x05},		 * Frame Time LSB */
+	{0x05, 0x65},		/* Shutter Time Line Offset */
+/*	{0x05, 0x6d},		 * Shutter Time Line Offset */
+/*	{0x06, 0xb1},		 * Shutter Time Pixel Offset */
+	{0x06, 0xcd},		/* Shutter Time Pixel Offset */
+	{0x07, 0xc1},		/* Black Level Subtract Sign */
+/*	{0x07, 0x00},		 * Black Level Subtract Sign */
+	{0x08, 0x06},		/* Black Level Subtract Level */
+	{0x08, 0x06},		/* Black Level Subtract Level */
+/*	{0x08, 0x01},		 * Black Level Subtract Level */
+	{0x09, 0x05},		/* Color Gain B Pixel 5 a */
+	{0x0a, 0x04},		/* Color Gain G1 Pixel 1 5 */
+	{0x0b, 0x04},		/* Color Gain G2 Pixel 1 0 5 */
+	{0x0c, 0x05},		/* Color Gain R Pixel 3 1 */
+	{0x0d, 0x00},		/* Color GainH  Pixel */
+	{0x0e, 0x0e},		/* Global Gain */
+	{0x0f, 0x00},		/* Contrast */
+	{0x10, 0x06},		/* H&V synchro polarity */
+	{0x11, 0x06},		/* ?default */
+	{0x12, 0x06},		/* DAC scale */
+	{0x14, 0x02},		/* ?default */
+	{0x13, 0x01},		/* Validate Settings */
+};
+static const __u8 initPas202[] = {
+	0x44, 0x44, 0x21, 0x30, 0x00, 0x00, 0x00, 0x80, 0x40, 0x00, 0x00, 0x00,
+	0x00, 0x00,
+	0x00, 0x00, 0x00, 0x07, 0x03, 0x0a,	/* 6 */
+	0x28, 0x1e, 0x28, 0x89, 0x30,
+	0x00, 0x00, 0x02, 0x03, 0x0f, 0x0c
+};
+static const __u8 pas202_sensor_init[][8] = {
+	{0xa0, 0x40, 0x02, 0x03, 0x00, 0x00, 0x00, 0x10},
+	{0xd0, 0x40, 0x04, 0x07, 0x34, 0x00, 0x09, 0x10},
+	{0xd0, 0x40, 0x08, 0x01, 0x00, 0x00, 0x01, 0x10},
+	{0xd0, 0x40, 0x0C, 0x00, 0x0C, 0x00, 0x32, 0x10},
+	{0xd0, 0x40, 0x10, 0x00, 0x01, 0x00, 0x63, 0x10},
+	{0xa0, 0x40, 0x15, 0x70, 0x01, 0x00, 0x63, 0x10},
+	{0xa0, 0x40, 0x18, 0x00, 0x01, 0x00, 0x63, 0x10},
+	{0xa0, 0x40, 0x11, 0x01, 0x01, 0x00, 0x63, 0x10},
+	{0xa0, 0x40, 0x03, 0x56, 0x01, 0x00, 0x63, 0x10},
+	{0xa0, 0x40, 0x11, 0x01, 0x01, 0x00, 0x63, 0x10},
+	{0xb0, 0x40, 0x04, 0x07, 0x2a, 0x00, 0x63, 0x10},
+	{0xb0, 0x40, 0x0e, 0x00, 0x3d, 0x00, 0x63, 0x10},
+
+	{0xa0, 0x40, 0x11, 0x01, 0x3d, 0x00, 0x63, 0x16},
+	{0xa0, 0x40, 0x10, 0x08, 0x3d, 0x00, 0x63, 0x15},
+	{0xa0, 0x40, 0x02, 0x04, 0x3d, 0x00, 0x63, 0x16},
+	{0xa0, 0x40, 0x11, 0x01, 0x3d, 0x00, 0x63, 0x16},
+	{0xb0, 0x40, 0x0e, 0x00, 0x31, 0x00, 0x63, 0x16},
+	{0xa0, 0x40, 0x11, 0x01, 0x31, 0x00, 0x63, 0x16},
+	{0xa0, 0x40, 0x10, 0x0e, 0x31, 0x00, 0x63, 0x15},
+	{0xa0, 0x40, 0x11, 0x01, 0x31, 0x00, 0x63, 0x16},
+};
+
+static const __u8 initTas5110[] = {
+	0x44, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x20, 0x11, 0x00, 0x00, 0x00,
+	0x00, 0x00,
+	0x00, 0x01, 0x00, 0x46, 0x09, 0x0a,	/* shift from 0x45 0x09 0x0a */
+	0x16, 0x12, 0x60, 0x86, 0x2b,
+	0x14, 0x0a, 0x02, 0x02, 0x09, 0x07
+};
+static const __u8 tas5110_sensor_init[][8] = {
+	{0x30, 0x11, 0x00, 0x00, 0x0c, 0x00, 0x00, 0x10},
+	{0x30, 0x11, 0x02, 0x20, 0xa9, 0x00, 0x00, 0x10},
+	{0xa0, 0x61, 0x9a, 0xca, 0x00, 0x00, 0x00, 0x17},
+};
+
+static const __u8 initTas5130[] = {
+	0x04, 0x03, 0x00, 0x00, 0x00, 0x00, 0x00, 0x20, 0x11, 0x00, 0x00, 0x00,
+	0x00, 0x00,
+	0x00, 0x01, 0x00, 0x69, 0x0c, 0x0a,
+	0x28, 0x1e, 0x60, COMP, MCK_INIT,
+	0x18, 0x10, 0x04, 0x03, 0x11, 0x0c
+};
+static const __u8 tas5130_sensor_init[][8] = {
+/* 	{0x30, 0x11, 0x00, 0x40, 0x47, 0x00, 0x00, 0x10},
+					* shutter 0x47 short exposure? */
+	{0x30, 0x11, 0x00, 0x40, 0x01, 0x00, 0x00, 0x10},
+					/* shutter 0x01 long exposure */
+	{0x30, 0x11, 0x02, 0x20, 0x70, 0x00, 0x00, 0x10},
+};
+
+static void reg_r(struct usb_device *dev,
+			 __u16 value, __u8 *buffer)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,			/* request */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_INTERFACE,
+			value,
+			0,			/* index */
+			buffer, 1,
+			500);
+}
+
+static void reg_w(struct usb_device *dev,
+			  __u16 value,
+			  const __u8 *buffer,
+			  __u16 len)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			0x08,			/* request */
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_INTERFACE,
+			value,
+			0,			/* index */
+			(__u8 *) buffer, len,
+			500);
+}
+
+static int i2c_w(struct usb_device *dev, const __u8 *buffer)
+{
+	int retry = 60;
+	__u8 ByteReceive;
+
+	/* is i2c ready */
+	reg_w(dev, 0x08, buffer, 8);
+	while (retry--) {
+		msleep(10);
+		reg_r(dev, 0x08, &ByteReceive);
+		if (ByteReceive == 4)
+			return 0;
+	}
+	return -1;
+}
+
+static void i2c_w_vector(struct usb_device *dev,
+			const __u8 buffer[][8], int len)
+{
+	for (;;) {
+		reg_w(dev, 0x08, *buffer, 8);
+		len -= 8;
+		if (len <= 0)
+			break;
+		buffer++;
+	}
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 value;
+
+	switch (sd->sensor) {
+	case SENSOR_OV6650: {
+		__u8 i2cOV6650[] =
+			{0xa0, 0x60, 0x06, 0x11, 0x99, 0x04, 0x94, 0x15};
+
+		i2cOV6650[3] = sd->brightness;
+		if (i2c_w(gspca_dev->dev, i2cOV6650) < 0)
+			 goto err;
+		break;
+	    }
+	case  SENSOR_OV7630: {
+		__u8 i2cOV[] =
+			{0xa0, 0x21, 0x06, 0x36, 0xbd, 0x06, 0xf6, 0x16};
+
+		/* change reg 0x06 */
+		i2cOV[3] = sd->brightness;
+		if (i2c_w(gspca_dev->dev, i2cOV) < 0)
+			goto err;
+		break;
+	    }
+	case SENSOR_PAS106: {
+		__u8 i2c1[] =
+			{0xa1, 0x40, 0x00, 0x00, 0x00, 0x00, 0x00, 0x14};
+
+		i2c1[3] = sd->brightness >> 3;
+		i2c1[2] = 0x0e;
+		if (i2c_w(gspca_dev->dev, i2c1) < 0)
+			goto err;
+		i2c1[3] = 0x01;
+		i2c1[2] = 0x13;
+		if (i2c_w(gspca_dev->dev, i2c1) < 0)
+			goto err;
+		break;
+	    }
+	case SENSOR_PAS202: {
+		/* __u8 i2cpexpo1[] =
+			{0xb0, 0x40, 0x04, 0x07, 0x2a, 0x00, 0x63, 0x16}; */
+		__u8 i2cpexpo[] =
+			{0xb0, 0x40, 0x0e, 0x01, 0xab, 0x00, 0x63, 0x16};
+		__u8 i2cp202[] =
+			{0xa0, 0x40, 0x10, 0x0e, 0x31, 0x00, 0x63, 0x15};
+		static __u8 i2cpdoit[] =
+			{0xa0, 0x40, 0x11, 0x01, 0x31, 0x00, 0x63, 0x16};
+
+		/* change reg 0x10 */
+		i2cpexpo[4] = 0xff - sd->brightness;
+/*		if(i2c_w(gspca_dev->dev,i2cpexpo1) < 0)
+			goto err; */
+/*		if(i2c_w(gspca_dev->dev,i2cpdoit) < 0)
+			goto err; */
+		if (i2c_w(gspca_dev->dev, i2cpexpo) < 0)
+			goto err;
+		if (i2c_w(gspca_dev->dev, i2cpdoit) < 0)
+			goto err;
+		i2cp202[3] = sd->brightness >> 3;
+		if (i2c_w(gspca_dev->dev, i2cp202) < 0)
+			goto err;
+		if (i2c_w(gspca_dev->dev, i2cpdoit) < 0)
+			goto err;
+		break;
+	    }
+	case SENSOR_TAS5130CXX:
+	case SENSOR_TAS5110: {
+		__u8 i2c[] =
+			{0x30, 0x11, 0x02, 0x20, 0x70, 0x00, 0x00, 0x10};
+
+		value = 0xff - sd->brightness;
+		i2c[4] = value;
+		PDEBUG(D_CONF, "brightness %d : %d", value, i2c[4]);
+		if (i2c_w(gspca_dev->dev, i2c) < 0)
+			goto err;
+		break;
+	    }
+	}
+	return;
+err:
+	PDEBUG(D_ERR, "i2c error brightness");
+}
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 gain;
+	__u8 rgb_value;
+
+	gain = sd->contrast >> 4;
+	/* red and blue gain */
+	rgb_value = gain << 4 | gain;
+	reg_w(gspca_dev->dev, 0x10, &rgb_value, 1);
+	/* green gain */
+	rgb_value = gain;
+	reg_w(gspca_dev->dev, 0x11, &rgb_value, 1);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+/*	__u16 vendor; */
+	__u16 product;
+	int sif = 0;
+
+/*	vendor = id->idVendor; */
+	product = id->idProduct;
+/*	switch (vendor) { */
+/*	case 0x0c45:				 * Sonix */
+		switch (product) {
+		case 0x6001:			/* SN9C102 */
+		case 0x6005:			/* SN9C101 */
+		case 0x6007:			/* SN9C101 */
+			sd->sensor = SENSOR_TAS5110;
+			sif = 1;
+			break;
+		case 0x6009:			/* SN9C101 */
+		case 0x600d:			/* SN9C101 */
+		case 0x6029:			/* SN9C101 */
+			sd->sensor = SENSOR_PAS106;
+			sif = 1;
+			break;
+		case 0x6011:			/* SN9C101 - SN9C101G */
+			sd->sensor = SENSOR_OV6650;
+			sif = 1;
+			break;
+		case 0x6019:			/* SN9C101 */
+		case 0x602c:			/* SN9C102 */
+		case 0x602e:			/* SN9C102 */
+			sd->sensor = SENSOR_OV7630;
+			break;
+		case 0x60b0:			/* SN9C103 */
+			sd->sensor = SENSOR_OV7630_3;
+			break;
+		case 0x6024:			/* SN9C102 */
+		case 0x6025:			/* SN9C102 */
+			sd->sensor = SENSOR_TAS5130CXX;
+			break;
+		case 0x6028:			/* SN9C102 */
+			sd->sensor = SENSOR_PAS202;
+			break;
+		case 0x602d:			/* SN9C102 */
+			sd->sensor = SENSOR_HV7131R;
+			break;
+		case 0x60af:			/* SN9C103 */
+			sd->sensor = SENSOR_PAS202;
+			break;
+		}
+/*		break; */
+/*	} */
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	if (!sif) {
+		cam->cam_mode = vga_mode;
+		cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	} else {
+		cam->cam_mode = sif_mode;
+		cam->nmodes = sizeof sif_mode / sizeof sif_mode[0];
+	}
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	if (sd->sensor == SENSOR_OV7630_3)	/* jfm: from win trace */
+		reg_w(gspca_dev->dev, 0x01, probe_ov7630, sizeof probe_ov7630);
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	__u8 ByteReceive;
+
+	reg_r(gspca_dev->dev, 0x00, &ByteReceive);
+	if (ByteReceive != 0x10)
+		return -ENODEV;
+	return 0;
+}
+
+static void pas106_i2cinit(struct usb_device *dev)
+{
+	int i;
+	const __u8 *data;
+	__u8 i2c1[] = { 0xa1, 0x40, 0x00, 0x00, 0x00, 0x00, 0x00, 0x14 };
+
+	i = ARRAY_SIZE(pas106_data);
+	data = pas106_data[0];
+	while (--i >= 0) {
+		memcpy(&i2c1[2], data, 2);
+					/* copy 2 bytes from the template */
+		if (i2c_w(dev, i2c1) < 0)
+			PDEBUG(D_ERR, "i2c error pas106");
+		data += 2;
+	}
+}
+
+/* -- start the camera -- */
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int mode, l;
+	const __u8 *sn9c10x;
+	__u8 reg01, reg17;
+	__u8 reg17_19[3];
+
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+		sn9c10x = initHv7131;
+		reg17_19[0] = 0x60;
+		reg17_19[1] = (mode << 4) | 0x8a;
+		reg17_19[2] = 0x20;
+		break;
+	case SENSOR_OV6650:
+		sn9c10x = initOv6650;
+		reg17_19[0] = 0x68;
+		reg17_19[1] = (mode << 4) | 0x8b;
+		reg17_19[2] = 0x20;
+		break;
+	case SENSOR_OV7630:
+		sn9c10x = initOv7630;
+		reg17_19[0] = 0x68;
+		reg17_19[1] = (mode << 4) | COMP2;
+		reg17_19[2] = MCK_INIT1;
+		break;
+	case SENSOR_OV7630_3:
+		sn9c10x = initOv7630_3;
+		reg17_19[0] = 0x68;
+		reg17_19[1] = (mode << 4) | COMP2;
+		reg17_19[2] = MCK_INIT1;
+		break;
+	case SENSOR_PAS106:
+		sn9c10x = initPas106;
+		reg17_19[0] = 0x24;		/* 0x28 */
+		reg17_19[1] = (mode << 4) | COMP1;
+		reg17_19[2] = MCK_INIT1;
+		break;
+	case SENSOR_PAS202:
+		sn9c10x = initPas202;
+		reg17_19[0] = mode ? 0x24 : 0x20;
+		reg17_19[1] = (mode << 4) | 0x89;
+		reg17_19[2] = 0x20;
+		break;
+	case SENSOR_TAS5110:
+		sn9c10x = initTas5110;
+		reg17_19[0] = 0x60;
+		reg17_19[1] = (mode << 4) | 0x86;
+		reg17_19[2] = 0x2b;		/* 0xf3; */
+		break;
+	default:
+/*	case SENSOR_TAS5130CXX: */
+		sn9c10x = initTas5130;
+		reg17_19[0] = 0x60;
+		reg17_19[1] = (mode << 4) | COMP;
+		reg17_19[2] = mode ? 0x23 : 0x43;
+		break;
+	}
+	switch (sd->sensor) {
+	case SENSOR_OV7630:
+		reg01 = 0x06;
+		reg17 = 0x29;
+		l = 0x10;
+		break;
+	case SENSOR_OV7630_3:
+		reg01 = 0x44;
+		reg17 = 0x68;
+		l = 0x10;
+		break;
+	default:
+		reg01 = sn9c10x[0];
+		reg17 = sn9c10x[0x17 - 1];
+		l = 0x1f;
+		break;
+	}
+
+	/* reg 0x01 bit 2 video transfert on */
+	reg_w(dev, 0x01, &reg01, 1);
+	/* reg 0x17 SensorClk enable inv Clk 0x60 */
+	reg_w(dev, 0x17, &reg17, 1);
+/*fixme: for ov7630 102
+	reg_w(dev, 0x01, {0x06, sn9c10x[1]}, 2); */
+	/* Set the registers from the template */
+	reg_w(dev, 0x01, sn9c10x, l);
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+		i2c_w_vector(dev, hv7131_sensor_init,
+				sizeof hv7131_sensor_init);
+		break;
+	case SENSOR_OV6650:
+		i2c_w_vector(dev, ov6650_sensor_init,
+				sizeof ov6650_sensor_init);
+		break;
+	case SENSOR_OV7630:
+		i2c_w_vector(dev, ov7630_sensor_init_com,
+				sizeof ov7630_sensor_init_com);
+		msleep(200);
+		i2c_w_vector(dev, ov7630_sensor_init,
+				sizeof ov7630_sensor_init);
+		break;
+	case SENSOR_OV7630_3:
+		i2c_w_vector(dev, ov7630_sensor_init_com,
+				sizeof ov7630_sensor_init_com);
+		msleep(200);
+		i2c_w_vector(dev, ov7630_sensor_init_3,
+				sizeof ov7630_sensor_init_3);
+		break;
+	case SENSOR_PAS106:
+		pas106_i2cinit(dev);
+		break;
+	case SENSOR_PAS202:
+		i2c_w_vector(dev, pas202_sensor_init,
+				sizeof pas202_sensor_init);
+		break;
+	case SENSOR_TAS5110:
+		i2c_w_vector(dev, tas5110_sensor_init,
+				sizeof tas5110_sensor_init);
+		break;
+	default:
+/*	case SENSOR_TAS5130CXX: */
+		i2c_w_vector(dev, tas5130_sensor_init,
+				sizeof tas5130_sensor_init);
+		break;
+	}
+	/* H_size V_size  0x28, 0x1e maybe 640x480 */
+	reg_w(dev, 0x15, &sn9c10x[0x15 - 1], 2);
+	/* compression register */
+	reg_w(dev, 0x18, &reg17_19[1], 1);
+	/* H_start */		/*fixme: not ov7630*/
+	reg_w(dev, 0x12, &sn9c10x[0x12 - 1], 1);
+	/* V_START */		/*fixme: not ov7630*/
+	reg_w(dev, 0x13, &sn9c10x[0x13 - 1], 1);
+	/* reset 0x17 SensorClk enable inv Clk 0x60 */
+				/*fixme: ov7630 [17]=68 8f (+20 if 102)*/
+	reg_w(dev, 0x17, &reg17_19[0], 1);
+	/*MCKSIZE ->3 */	/*fixme: not ov7630*/
+	reg_w(dev, 0x19, &reg17_19[2], 1);
+	/* AE_STRX AE_STRY AE_ENDX AE_ENDY */
+	reg_w(dev, 0x1c, &sn9c10x[0x1c - 1], 4);
+	/* Enable video transfert */
+	reg_w(dev, 0x01, &sn9c10x[0], 1);
+	/* Compression */
+	reg_w(dev, 0x18, &reg17_19[1], 2);
+	msleep(20);
+
+	setcontrast(gspca_dev);
+	setbrightness(gspca_dev);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	__u8 ByteSend = 0;
+
+	ByteSend = 0x09;	/* 0X00 */
+	reg_w(gspca_dev->dev, 0x01, &ByteSend, 1);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	int p;
+
+	if (len > 6 && len < 24) {
+		for (p = 0; p < len - 6; p++) {
+			if (data[0 + p] == 0xff
+			    && data[1 + p] == 0xff
+			    && data[2 + p] == 0x00
+			    && data[3 + p] == 0xc4
+			    && data[4 + p] == 0xc4
+			    && data[5 + p] == 0x96) {	/* start of frame */
+				frame = gspca_frame_add(gspca_dev,
+							LAST_PACKET,
+							frame,
+							data, 0);
+				data += 12;
+				len -= 12;
+				gspca_frame_add(gspca_dev, FIRST_PACKET,
+						frame, data, len);
+				return;
+			}
+		}
+	}
+	gspca_frame_add(gspca_dev, INTER_PACKET,
+			frame, data, len);
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->contrast;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x0c45, 0x6001), DVNM("Genius VideoCAM NB")},
+	{USB_DEVICE(0x0c45, 0x6005), DVNM("Sweex Tas5110")},
+	{USB_DEVICE(0x0c45, 0x6007), DVNM("Sonix sn9c101 + Tas5110D")},
+	{USB_DEVICE(0x0c45, 0x6009), DVNM("spcaCam@120")},
+	{USB_DEVICE(0x0c45, 0x600d), DVNM("spcaCam@120")},
+	{USB_DEVICE(0x0c45, 0x6011), DVNM("MAX Webcam Microdia-OV6650-SN9C101G")},
+	{USB_DEVICE(0x0c45, 0x6019), DVNM("Generic Sonix OV7630")},
+	{USB_DEVICE(0x0c45, 0x6024), DVNM("Generic Sonix Tas5130c")},
+	{USB_DEVICE(0x0c45, 0x6025), DVNM("Xcam Shanga")},
+	{USB_DEVICE(0x0c45, 0x6028), DVNM("Sonix Btc Pc380")},
+	{USB_DEVICE(0x0c45, 0x6029), DVNM("spcaCam@150")},
+	{USB_DEVICE(0x0c45, 0x602c), DVNM("Generic Sonix OV7630")},
+	{USB_DEVICE(0x0c45, 0x602d), DVNM("LIC-200 LG")},
+	{USB_DEVICE(0x0c45, 0x602e), DVNM("Genius VideoCam Messenger")},
+	{USB_DEVICE(0x0c45, 0x60af), DVNM("Trust WB3100P")},
+	{USB_DEVICE(0x0c45, 0x60b0), DVNM("Genius VideoCam Look")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
new file mode 100644
index 000000000000..6180bc565ca1
--- /dev/null
+++ b/drivers/media/video/gspca/sonixj.c
@@ -0,0 +1,1629 @@
+/*
+ *		Sonix sn9c102p sn9c105 sn9c120 (jpeg) library
+ *		Copyright (C) 2005 Michel Xhaard mxhaard@magic.fr
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "sonixj"
+
+#include "gspca.h"
+#include "jpeg.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SONIX JPEG USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	int avg_lum;
+	unsigned int exposure;
+
+	unsigned short brightness;
+	unsigned char contrast;
+	unsigned char colors;
+	unsigned char autogain;
+
+	signed char ag_cnt;
+#define AG_CNT_START 13
+
+	char qindex;
+	char sensor;			/* Type of image sensor chip */
+#define SENSOR_HV7131R 0
+#define SENSOR_MI0360 1
+#define SENSOR_MO4000 2
+#define SENSOR_OV7648 3
+#define SENSOR_OV7660 4
+	unsigned char customid;
+#define SN9C102P 0
+#define SN9C105 1
+#define SN9C110 2
+#define SN9C120 3
+#define SN9C325 4
+	unsigned char i2c_base;
+	unsigned char i2c_ctrl_reg;
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 0xffff,
+		.step    = 1,
+		.default_value = 0x7fff,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 127,
+		.step    = 1,
+		.default_value = 63,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define SD_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Color",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+#define SD_AUTOGAIN 3
+	{
+	    {
+		.id      = V4L2_CID_AUTOGAIN,
+		.type    = V4L2_CTRL_TYPE_BOOLEAN,
+		.name    = "Auto Gain",
+		.minimum = 0,
+		.maximum = 1,
+		.step    = 1,
+		.default_value = 1,
+	    },
+	    .set = sd_setautogain,
+	    .get = sd_getautogain,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 160, 120, 2},
+	{V4L2_PIX_FMT_JPEG, 320, 240, 1},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 0},
+};
+
+/*Data from sn9c102p+hv71331r */
+static __u8 sn_hv7131[] = {
+	0x00, 0x03, 0x64, 0x00, 0x1A, 0x20, 0x20, 0x20, 0xA1, 0x11,
+/*	reg0  reg1  reg2  reg3  reg4  reg5  reg6  reg7  reg8  reg9 */
+	0x02, 0x09, 0x00, 0x00, 0x00, 0x10, 0x03, 0x00,		/* 00 */
+/*	rega  regb  regc  regd  rege  regf  reg10 reg11 */
+	0x00, 0x01, 0x03, 0x28, 0x1e, 0x41, 0x0a, 0x00, 0x00, 0x00,
+/*	reg12 reg13 reg14 reg15 reg16 reg17 reg18 reg19 reg1a reg1b */
+	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+/*	reg1c reg1d reg1e reg1f reg20 reg21 reg22 reg23 */
+};
+
+static __u8 sn_mi0360[] = {
+	0x00, 0x61, 0x44, 0x00, 0x1a, 0x20, 0x20, 0x20, 0xb1, 0x5d,
+/*	reg0  reg1  reg2  reg3  reg4  reg5  reg6  reg7  reg8  reg9 */
+	0x07, 0x00, 0x00, 0x00, 0x00, 0x10, 0x03, 0x00,
+/*	rega  regb  regc  regd  rege  regf  reg10 reg11 */
+	0x00, 0x02, 0x0a, 0x28, 0x1e, 0x61, 0x06, 0x00, 0x00, 0x00,
+/*	reg12 reg13 reg14 reg15 reg16 reg17 reg18 reg19 reg1a reg1b */
+	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+/*	reg1c reg1d reg1e reg1f reg20 reg21 reg22 reg23 */
+};
+
+static __u8 sn_mo4000[] = {
+	0x12,	0x23,	0x60,	0x00,	0x1A,	0x00,	0x20,	0x18,	0x81,
+/*	reg0	reg1	reg2	reg3	reg4	reg5	reg6	reg7	reg8 */
+	0x21,	0x00,	0x00,	0x00,	0x00,	0x00,	0x00,	0x03,	0x00,
+/*	reg9	rega	regb	regc	regd	rege	regf	reg10	reg11*/
+	0x0b,	0x0f,	0x14,	0x28,	0x1e,	0x40,	0x08,	0x00,	0x00,
+/*	reg12	reg13	reg14	reg15	reg16	reg17	reg18	reg19	reg1a*/
+	0x00,	0x00,	0x00,	0x00,	0x00,	0x08,	0x25,	0x39,	0x4b,
+/*	reg1b	reg1c	reg1d	reg1e	reg1f	reg20	reg21	reg22	reg23*/
+	0x5c,	0x6b,	0x79,	0x87,	0x95,	0xa2,	0xaf,	0xbb,	0xc7,
+	0xd3,	0xdf,	0xea,	0xf5
+};
+
+static __u8 sn_ov7648[] = {
+	0x00, 0x21, 0x62, 0x00, 0x1a, 0x20, 0x20, 0x20, 0xA1, 0x6E, 0x18, 0x65,
+	0x00, 0x00, 0x00, 0x10, 0x03, 0x00, 0x00, 0x06, 0x06, 0x28, 0x1E, 0x82,
+	0x07, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+
+static __u8 sn_ov7660[]	= {
+/*	reg0	reg1	reg2	reg3	reg4	reg5	reg6	reg7	reg8 */
+	0x00,	0x61,	0x40,	0x00,	0x1a,	0x00,	0x00,	0x00,	0x81,
+/* 	reg9	rega	regb	regc	regd	rege	regf	reg10	reg11*/
+	0x21,	0x00,	0x00,	0x00,	0x00,	0x00,	0x00,	0x03,	0x00,
+/*	reg12	reg13	reg14	reg15	reg16	reg17	reg18	reg19	reg1a*/
+	0x01,	0x01,	0x08,	0x28,	0x1e,	0x20,	0x07,	0x00,	0x00,
+/*	reg1b	reg1c	reg1d	reg1e	reg1f	reg20	reg21	reg22	reg23*/
+	0x00,	0x00,	0x00,	0x00,	0x00,	0x00,	0x00,	0x00,	0x00
+};
+
+/* sequence specific to the sensors - !! index = SENSOR_xxx */
+static __u8 *sn_tb[] = {
+	sn_hv7131,
+	sn_mi0360,
+	sn_mo4000,
+	sn_ov7648,
+	sn_ov7660
+};
+
+static __u8 regsn20[] = {
+	0x00, 0x2d, 0x46, 0x5a, 0x6c, 0x7c, 0x8b, 0x99,
+	0xa6, 0xb2, 0xbf, 0xca, 0xd5, 0xe0, 0xeb, 0xf5, 0xff
+};
+static __u8 regsn20_sn9c325[] = {
+	0x0a, 0x3a, 0x56, 0x6c, 0x7e, 0x8d, 0x9a, 0xa4,
+	0xaf, 0xbb, 0xc5, 0xcd, 0xd5, 0xde, 0xe8, 0xed, 0xf5
+};
+
+static __u8 reg84[] = {
+	0x14, 0x00, 0x27, 0x00, 0x07, 0x00, 0xe5, 0x0f,
+	0xe4, 0x0f, 0x38, 0x00, 0x3e, 0x00, 0xc3, 0x0f,
+/*	0x00, 0x00, 0x00, 0x00, 0x00 */
+	0xf7, 0x0f, 0x0a, 0x00, 0x00
+};
+static __u8 reg84_sn9c325[] = {
+	0x14, 0x00, 0x27, 0x00, 0x07, 0x00, 0xe4, 0x0f,
+	0xd3, 0x0f, 0x4b, 0x00, 0x48, 0x00, 0xc0, 0x0f,
+	0xf8, 0x0f, 0x00, 0x00, 0x00
+};
+
+static __u8 hv7131r_sensor_init[][8] = {
+	{0xC1, 0x11, 0x01, 0x08, 0x01, 0x00, 0x00, 0x10},
+	{0xB1, 0x11, 0x34, 0x17, 0x7F, 0x00, 0x00, 0x10},
+	{0xD1, 0x11, 0x40, 0xFF, 0x7F, 0x7F, 0x7F, 0x10},
+	{0x91, 0x11, 0x44, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x11, 0x10, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x11, 0x14, 0x01, 0xE2, 0x02, 0x82, 0x10},
+	{0x91, 0x11, 0x18, 0x00, 0x00, 0x00, 0x00, 0x10},
+
+	{0xA1, 0x11, 0x01, 0x08, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x01, 0x08, 0x00, 0x00, 0x00, 0x10},
+	{0xC1, 0x11, 0x25, 0x00, 0x61, 0xA8, 0x00, 0x10},
+	{0xA1, 0x11, 0x30, 0x22, 0x00, 0x00, 0x00, 0x10},
+	{0xC1, 0x11, 0x31, 0x20, 0x2E, 0x20, 0x00, 0x10},
+	{0xC1, 0x11, 0x25, 0x00, 0xC3, 0x50, 0x00, 0x10},
+	{0xA1, 0x11, 0x30, 0x07, 0x00, 0x00, 0x00, 0x10}, /* gain14 */
+	{0xC1, 0x11, 0x31, 0x10, 0x10, 0x10, 0x00, 0x10}, /* r g b 101a10 */
+
+	{0xA1, 0x11, 0x01, 0x08, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x20, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x21, 0xD0, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x22, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x23, 0x09, 0x00, 0x00, 0x00, 0x10},
+
+	{0xA1, 0x11, 0x01, 0x08, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x20, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x21, 0xD0, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x22, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x11, 0x23, 0x10, 0x00, 0x00, 0x00, 0x10},
+	{0, 0, 0, 0, 0, 0, 0, 0}
+};
+static __u8 mi0360_sensor_init[][8] = {
+	{0xB1, 0x5D, 0x07, 0x00, 0x02, 0x00, 0x00, 0x10},
+	{0xB1, 0x5D, 0x0D, 0x00, 0x01, 0x00, 0x00, 0x10},
+	{0xB1, 0x5D, 0x0D, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x01, 0x00, 0x08, 0x00, 0x16, 0x10},
+	{0xD1, 0x5D, 0x03, 0x01, 0xE2, 0x02, 0x82, 0x10},
+	{0xD1, 0x5D, 0x05, 0x00, 0x09, 0x00, 0x53, 0x10},
+	{0xB1, 0x5D, 0x0D, 0x00, 0x02, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x0A, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x0C, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x0E, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x10, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x12, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x14, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x16, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x18, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x1A, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x1C, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xB1, 0x5D, 0x32, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x20, 0x91, 0x01, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x22, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x24, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x26, 0x00, 0x00, 0x00, 0x24, 0x10},
+	{0xD1, 0x5D, 0x2F, 0xF7, 0xB0, 0x00, 0x04, 0x10},
+	{0xD1, 0x5D, 0x31, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x33, 0x00, 0x00, 0x01, 0x00, 0x10},
+	{0xB1, 0x5D, 0x3D, 0x06, 0x8F, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x40, 0x01, 0xE0, 0x00, 0xD1, 0x10},
+	{0xB1, 0x5D, 0x44, 0x00, 0x82, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x58, 0x00, 0x78, 0x00, 0x43, 0x10},
+	{0xD1, 0x5D, 0x5A, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x5C, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x5E, 0x00, 0x00, 0xA3, 0x1D, 0x10},
+	{0xB1, 0x5D, 0x62, 0x04, 0x11, 0x00, 0x00, 0x10},
+
+	{0xB1, 0x5D, 0x20, 0x91, 0x01, 0x00, 0x00, 0x10},
+	{0xB1, 0x5D, 0x20, 0x11, 0x01, 0x00, 0x00, 0x10},
+	{0xB1, 0x5D, 0x09, 0x00, 0x64, 0x00, 0x00, 0x10},
+	{0xD1, 0x5D, 0x2B, 0x00, 0xA0, 0x00, 0xB0, 0x10},
+	{0xD1, 0x5D, 0x2D, 0x00, 0xA0, 0x00, 0xA0, 0x10},
+
+	{0xB1, 0x5D, 0x0A, 0x00, 0x02, 0x00, 0x00, 0x10}, /* sensor clck ?2 */
+	{0xB1, 0x5D, 0x06, 0x00, 0x30, 0x00, 0x00, 0x10},
+	{0xB1, 0x5D, 0x05, 0x00, 0x0A, 0x00, 0x00, 0x10},
+	{0xB1, 0x5D, 0x09, 0x02, 0x35, 0x00, 0x00, 0x10}, /* exposure 2 */
+
+	{0xD1, 0x5D, 0x2B, 0x00, 0xB9, 0x00, 0xE3, 0x10},
+	{0xD1, 0x5D, 0x2D, 0x00, 0x5f, 0x00, 0xB9, 0x10}, /* 42 */
+/*	{0xB1, 0x5D, 0x35, 0x00, 0x67, 0x00, 0x00, 0x10}, * gain orig */
+/*	{0xB1, 0x5D, 0x35, 0x00, 0x20, 0x00, 0x00, 0x10}, * gain */
+	{0xB1, 0x5D, 0x07, 0x00, 0x03, 0x00, 0x00, 0x10}, /* update */
+	{0xB1, 0x5D, 0x07, 0x00, 0x02, 0x00, 0x00, 0x10}, /* sensor on */
+	{0, 0, 0, 0, 0, 0, 0, 0}
+};
+static __u8 mo4000_sensor_init[][8] = {
+	{0xa1, 0x21, 0x01, 0x02, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x02, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x03, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x04, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x05, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x05, 0x04, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x06, 0x80, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x06, 0x81, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x0e, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x11, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x11, 0x20, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x11, 0x30, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x11, 0x38, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x11, 0x38, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x12, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x10, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x0f, 0x20, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x10, 0x20, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x11, 0x38, 0x00, 0x00, 0x00, 0x10},
+	{0, 0, 0, 0, 0, 0, 0, 0}
+};
+static __u8 ov7660_sensor_init[][8] = {
+	{0xa1, 0x21, 0x12, 0x80, 0x00, 0x00, 0x00, 0x10}, /* reset SCCB */
+	{0xa1, 0x21, 0x12, 0x05, 0x00, 0x00, 0x00, 0x10},
+						/* Outformat ?? rawRGB */
+	{0xa1, 0x21, 0x13, 0xb8, 0x00, 0x00, 0x00, 0x10}, /* init COM8 */
+/*	{0xd1, 0x21, 0x00, 0x01, 0x74, 0x92, 0x00, 0x10},
+						 * GAIN BLUE RED VREF */
+	{0xd1, 0x21, 0x00, 0x01, 0x74, 0x74, 0x00, 0x10},
+						/* GAIN BLUE RED VREF */
+	{0xd1, 0x21, 0x04, 0x00, 0x7d, 0x62, 0x00, 0x10},
+						/* COM 1 BAVE GEAVE AECHH */
+	{0xb1, 0x21, 0x08, 0x83, 0x01, 0x00, 0x00, 0x10}, /* RAVE COM2 */
+	{0xd1, 0x21, 0x0c, 0x00, 0x08, 0x04, 0x4f, 0x10}, /* COM 3 4 5 6 */
+/*	{0xd1, 0x21, 0x10, 0x7f, 0x40, 0x05, 0xf8, 0x10},
+						 * AECH CLKRC COM7 COM8 */
+	{0xd1, 0x21, 0x10, 0x7f, 0x40, 0x05, 0xff, 0x10},
+						/* AECH CLKRC COM7 COM8 */
+	{0xc1, 0x21, 0x14, 0x2c, 0x00, 0x02, 0x00, 0x10}, /* COM9 COM10 */
+	{0xd1, 0x21, 0x17, 0x10, 0x60, 0x02, 0x7b, 0x10},
+						/* HSTART HSTOP VSTRT VSTOP */
+	{0xa1, 0x21, 0x1b, 0x02, 0x00, 0x00, 0x00, 0x10}, /* PSHFT */
+	{0xb1, 0x21, 0x1e, 0x01, 0x0e, 0x00, 0x00, 0x10}, /* MVFP LAEC */
+	{0xd1, 0x21, 0x20, 0x07, 0x07, 0x07, 0x07, 0x10},
+					/* BOS GBOS GROS ROS (BGGR offset) */
+/*	{0xd1, 0x21, 0x24, 0x68, 0x58, 0xd4, 0x80, 0x10},
+						 * AEW AEB VPT BBIAS */
+	{0xd1, 0x21, 0x24, 0x78, 0x68, 0xd4, 0x80, 0x10},
+						/* AEW AEB VPT BBIAS */
+	{0xd1, 0x21, 0x28, 0x80, 0x30, 0x00, 0x00, 0x10},
+						/* GbBIAS RSVD EXHCH EXHCL */
+	{0xd1, 0x21, 0x2c, 0x80, 0x00, 0x00, 0x62, 0x10},
+						/* RBIAS ADVFL ASDVFH YAVE */
+	{0xc1, 0x21, 0x30, 0x08, 0x30, 0xb4, 0x00, 0x10},
+						/* HSYST HSYEN HREF */
+	{0xd1, 0x21, 0x33, 0x00, 0x07, 0x84, 0x00, 0x10}, /* reserved */
+	{0xd1, 0x21, 0x37, 0x0c, 0x02, 0x43, 0x00, 0x10},
+						/* ADC ACOM OFON TSLB */
+	{0xd1, 0x21, 0x3b, 0x02, 0x6c, 0x19, 0x0e, 0x10},
+						/* COM11 COM12 COM13 COM14 */
+	{0xd1, 0x21, 0x3f, 0x41, 0xc1, 0x22, 0x08, 0x10},
+						/* EDGE COM15 COM16 COM17 */
+	{0xd1, 0x21, 0x43, 0xf0, 0x10, 0x78, 0xa8, 0x10}, /* reserved */
+	{0xd1, 0x21, 0x47, 0x60, 0x80, 0x00, 0x00, 0x10}, /* reserved */
+	{0xd1, 0x21, 0x4b, 0x00, 0x00, 0x00, 0x00, 0x10}, /* reserved */
+	{0xd1, 0x21, 0x4f, 0x46, 0x36, 0x0f, 0x17, 0x10}, /* MTX 1 2 3 4 */
+	{0xd1, 0x21, 0x53, 0x7f, 0x96, 0x40, 0x40, 0x10}, /* MTX 5 6 7 8 */
+	{0xb1, 0x21, 0x57, 0x40, 0x0f, 0x00, 0x00, 0x10}, /* MTX9 MTXS */
+	{0xd1, 0x21, 0x59, 0xba, 0x9a, 0x22, 0xb9, 0x10}, /* reserved */
+	{0xd1, 0x21, 0x5d, 0x9b, 0x10, 0xf0, 0x05, 0x10}, /* reserved */
+	{0xa1, 0x21, 0x61, 0x60, 0x00, 0x00, 0x00, 0x10}, /* reserved */
+	{0xd1, 0x21, 0x62, 0x00, 0x00, 0x50, 0x30, 0x10},
+						/* LCC1 LCC2 LCC3 LCC4 */
+	{0xa1, 0x21, 0x66, 0x00, 0x00, 0x00, 0x00, 0x10}, /* LCC5 */
+	{0xd1, 0x21, 0x67, 0x80, 0x7a, 0x90, 0x80, 0x10},
+	{0xa1, 0x21, 0x6b, 0x0a, 0x00, 0x00, 0x00, 0x10},
+					/* band gap reference [0..3] DBLV */
+	{0xd1, 0x21, 0x6c, 0x30, 0x48, 0x80, 0x74, 0x10}, /* gamma curve */
+	{0xd1, 0x21, 0x70, 0x64, 0x60, 0x5c, 0x58, 0x10}, /* gamma curve */
+	{0xd1, 0x21, 0x74, 0x54, 0x4c, 0x40, 0x38, 0x10}, /* gamma curve */
+	{0xd1, 0x21, 0x78, 0x34, 0x30, 0x2f, 0x2b, 0x10}, /* gamma curve */
+	{0xd1, 0x21, 0x7c, 0x03, 0x07, 0x17, 0x34, 0x10}, /* gamma curve */
+	{0xd1, 0x21, 0x80, 0x41, 0x4d, 0x58, 0x63, 0x10}, /* gamma curve */
+	{0xd1, 0x21, 0x84, 0x6e, 0x77, 0x87, 0x95, 0x10}, /* gamma curve */
+	{0xc1, 0x21, 0x88, 0xaf, 0xc7, 0xdf, 0x00, 0x10}, /* gamma curve */
+	{0xc1, 0x21, 0x8b, 0x99, 0x99, 0xcf, 0x00, 0x10}, /* reserved */
+	{0xb1, 0x21, 0x92, 0x00, 0x00, 0x00, 0x00, 0x10},
+/****** (some exchanges in the win trace) ******/
+	{0xa1, 0x21, 0x1e, 0x01, 0x00, 0x00, 0x00, 0x10},
+						/* bits[3..0]reserved */
+	{0xa1, 0x21, 0x1e, 0x01, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x03, 0x00, 0x00, 0x00, 0x00, 0x10},
+						/* VREF vertical frame ctrl */
+	{0xa1, 0x21, 0x03, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x10, 0x20, 0x00, 0x00, 0x00, 0x10}, /* 0x20 */
+	{0xa1, 0x21, 0x2d, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x2e, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x00, 0x1f, 0x00, 0x00, 0x00, 0x10},
+/*	{0xb1, 0x21, 0x01, 0x78, 0x78, 0x00, 0x00, 0x10}, */
+/****** (some exchanges in the win trace) ******/
+	{0xa1, 0x21, 0x93, 0x00, 0x00, 0x00, 0x00, 0x10},/* dummy line hight */
+	{0xa1, 0x21, 0x92, 0x25, 0x00, 0x00, 0x00, 0x10},/* dummy line low */
+	{0xa1, 0x21, 0x2a, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x2b, 0x00, 0x00, 0x00, 0x00, 0x10},
+/*	{0xa1, 0x21, 0x02, 0x90, 0x00, 0x00, 0x00, 0x10}, */
+/****** (some exchanges in the win trace) ******/
+/**********startsensor KO if changed !!****/
+	{0xa1, 0x21, 0x93, 0x01, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x92, 0xff, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x2a, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xa1, 0x21, 0x2b, 0xc3, 0x00, 0x00, 0x00, 0x10},
+/* here may start the isoc exchanges */
+	{0, 0, 0, 0, 0, 0, 0, 0}
+};
+/* reg0x04		reg0x07		reg 0x10 */
+/* expo  = (COM1 & 0x02) | (AECHH & 0x2f <<10) [ (AECh << 2) */
+
+static __u8 ov7648_sensor_init[][8] = {
+	{0xC1, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x00},
+	{0xC1, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x00},
+	{0xC1, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x00},
+	{0xA1, 0x6E, 0x3F, 0x20, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x3F, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x3E, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x04, 0x02, 0xB1, 0x02, 0x39, 0x10},
+	{0xD1, 0x6E, 0x08, 0x00, 0x01, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x0C, 0x02, 0x7F, 0x01, 0xE0, 0x10},
+	{0xD1, 0x6E, 0x12, 0x03, 0x02, 0x00, 0x03, 0x10},
+	{0xD1, 0x6E, 0x16, 0x85, 0x40, 0x4A, 0x40, 0x10},
+	{0xC1, 0x6E, 0x1A, 0x00, 0x80, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x1D, 0x08, 0x03, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x23, 0x00, 0xB0, 0x00, 0x94, 0x10},
+	{0xD1, 0x6E, 0x27, 0x58, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x2D, 0x14, 0x35, 0x61, 0x84, 0x10},
+	{0xD1, 0x6E, 0x31, 0xA2, 0xBD, 0xD8, 0xFF, 0x10},
+	{0xD1, 0x6E, 0x35, 0x06, 0x1E, 0x12, 0x02, 0x10},
+	{0xD1, 0x6E, 0x39, 0xAA, 0x53, 0x37, 0xD5, 0x10},
+	{0xA1, 0x6E, 0x3D, 0xF2, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x3E, 0x00, 0x00, 0x80, 0x03, 0x10},
+	{0xD1, 0x6E, 0x42, 0x03, 0x00, 0x00, 0x00, 0x10},
+	{0xC1, 0x6E, 0x46, 0x00, 0x80, 0x80, 0x00, 0x10},
+	{0xD1, 0x6E, 0x4B, 0x02, 0xEF, 0x08, 0xCD, 0x10},
+	{0xD1, 0x6E, 0x4F, 0x00, 0xD0, 0x00, 0xA0, 0x10},
+	{0xD1, 0x6E, 0x53, 0x01, 0xAA, 0x01, 0x40, 0x10},
+	{0xD1, 0x6E, 0x5A, 0x50, 0x04, 0x30, 0x03, 0x10},
+	{0xA1, 0x6E, 0x5E, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x5F, 0x10, 0x40, 0xFF, 0x00, 0x10},
+  /*	{0xD1, 0x6E, 0x63, 0x40, 0x40, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x67, 0x00, 0x00, 0x00, 0x00, 0x10},
+ * This is currently setting a
+ * blue tint, and some things more , i leave it here for future test if
+ * somene is having problems with color on this sensor
+	{0xD1, 0x6E, 0x6B, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xD1, 0x6E, 0x6F, 0x00, 0x00, 0x00, 0x00, 0x10},
+	{0xC1, 0x6E, 0x73, 0x10, 0x80, 0xEB, 0x00, 0x10},
+	{0xA1, 0x6E, 0x1E, 0x03, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x15, 0x01, 0x00, 0x00, 0x00, 0x10},
+	{0xC1, 0x6E, 0x16, 0x40, 0x40, 0x40, 0x00, 0x10},
+	{0xA1, 0x6E, 0x1D, 0x08, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x06, 0x02, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x07, 0xB5, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x18, 0x6B, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x1D, 0x08, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x06, 0x02, 0x00, 0x00, 0x00, 0x10},
+	{0xA1, 0x6E, 0x07, 0xB8, 0x00, 0x00, 0x00, 0x10},  */
+	{0xC1, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x00},
+	{0xA1, 0x6E, 0x06, 0x03, 0x00, 0x00, 0x00, 0x10}, /* Bright... */
+	{0xA1, 0x6E, 0x07, 0x66, 0x00, 0x00, 0x00, 0x10}, /* B.. */
+	{0xC1, 0x6E, 0x1A, 0x03, 0x65, 0x90, 0x00, 0x10}, /* Bright/Witen....*/
+/*	{0xC1, 0x6E, 0x16, 0x45, 0x40, 0x60, 0x00, 0x10},  * Bright/Witene */
+	{0, 0, 0, 0, 0, 0, 0, 0}
+};
+
+static __u8 qtable4[] = {
+	0x06, 0x04, 0x04, 0x06, 0x04, 0x04, 0x06, 0x06, 0x06, 0x06, 0x08, 0x06,
+	0x06, 0x08, 0x0A, 0x11,
+	0x0A, 0x0A, 0x08, 0x08, 0x0A, 0x15, 0x0F, 0x0F, 0x0C, 0x11, 0x19, 0x15,
+	0x19, 0x19, 0x17, 0x15,
+	0x17, 0x17, 0x1B, 0x1D, 0x25, 0x21, 0x1B, 0x1D, 0x23, 0x1D, 0x17, 0x17,
+	0x21, 0x2E, 0x21, 0x23,
+	0x27, 0x29, 0x2C, 0x2C, 0x2C, 0x19, 0x1F, 0x30, 0x32, 0x2E, 0x29, 0x32,
+	0x25, 0x29, 0x2C, 0x29,
+	0x06, 0x08, 0x08, 0x0A, 0x08, 0x0A, 0x13, 0x0A, 0x0A, 0x13, 0x29, 0x1B,
+	0x17, 0x1B, 0x29, 0x29,
+	0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29,
+	0x29, 0x29, 0x29, 0x29,
+	0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29,
+	0x29, 0x29, 0x29, 0x29,
+	0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29, 0x29,
+	0x29, 0x29, 0x29, 0x29
+};
+
+static void reg_r(struct usb_device *dev,
+		  __u16 value,
+		  __u8 *buffer, int len)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_INTERFACE,
+			value, 0,
+			buffer, len,
+			500);
+}
+
+static void reg_w(struct usb_device *dev,
+			  __u16 value,
+			  __u8 *buffer,
+			  int len)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			0x08,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_INTERFACE,
+			value, 0,
+			buffer, len,
+			500);
+}
+
+/* write 2 bytes */
+static void i2c_w2(struct gspca_dev *gspca_dev,
+		   __u8 *buffer)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 mode[8];
+
+	/* is i2c ready */
+	mode[0] = sd->i2c_ctrl_reg | (2 << 4);
+	mode[1] = sd->i2c_base;
+	mode[2] = buffer[0];
+	mode[3] = buffer[1];
+	mode[4] = 0;
+	mode[5] = 0;
+	mode[6] = 0;
+	mode[7] = 0x10;
+	reg_w(dev, 0x08, mode, 8);
+}
+
+/* write 8 bytes */
+static void i2c_w8(struct usb_device *dev, __u8 *buffer)
+{
+	reg_w(dev, 0x08, buffer, 8);
+	msleep(1);
+}
+
+/* read 5 bytes */
+static void i2c_r5(struct gspca_dev *gspca_dev, __u8 reg,
+			     __u8 *buffer)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 mode[8];
+
+	mode[0] = sd->i2c_ctrl_reg | 0x10;
+	mode[1] = sd->i2c_base;
+	mode[2] = reg;
+	mode[3] = 0;
+	mode[4] = 0;
+	mode[5] = 0;
+	mode[6] = 0;
+	mode[7] = 0x10;
+	i2c_w8(dev, mode);
+	mode[0] = sd->i2c_ctrl_reg | (5 << 4) | 0x02;
+	mode[2] = 0;
+	i2c_w8(dev, mode);
+	reg_r(dev, 0x0a, buffer, 5);
+}
+
+static int probesensor(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 reg02;
+	static __u8 datasend[] = { 2, 0 };
+	/* reg val1 val2 val3 val4 */
+	__u8 datarecd[6];
+
+	i2c_w2(gspca_dev, datasend);
+/* should write 0xa1 0x11 0x02 0x00 0x00 0x00 0x00 the 0x10 is add by i2cw */
+	msleep(10);
+	reg02 = 0x66;
+	reg_w(dev, 0x02, &reg02, 1);		/* Gpio on */
+	msleep(10);
+	i2c_r5(gspca_dev, 0, datarecd);		/* read sensor id */
+	if (datarecd[0] == 0x02
+	    && datarecd[1] == 0x09
+	    && datarecd[2] == 0x01
+	    && datarecd[3] == 0x00
+	    && datarecd[4] == 0x00) {
+		PDEBUG(D_PROBE, "Find Sensor sn9c102P HV7131R");
+		sd->sensor = SENSOR_HV7131R;
+		return SENSOR_HV7131R;
+	}
+	PDEBUG(D_PROBE, "Find Sensor %d %d %d",
+		datarecd[0], datarecd[1], datarecd[2]);
+	PDEBUG(D_PROBE, "Sensor sn9c102P Not found");
+	return -ENODEV;
+}
+
+static int configure_gpio(struct gspca_dev *gspca_dev,
+			  __u8 *sn9c1xx)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 data;
+	__u8 regF1;
+	__u8 *reg9a;
+	static __u8 reg9a_def[] =
+		{0x08, 0x40, 0x20, 0x10, 0x00, 0x04};
+	static __u8 reg9a_sn9c120[] =			/* from win trace */
+		{0x00, 0x40, 0x38, 0x30, 0x00, 0x20};
+	static __u8 reg9a_sn9c325[] =
+		{0x0a, 0x40, 0x38, 0x30, 0x00, 0x20};
+
+
+	regF1 = 0x00;
+	reg_w(dev, 0xf1, &regF1, 1);
+
+	reg_w(dev, 0x01, &sn9c1xx[0], 1);	/*fixme:jfm was [1] en v1*/
+
+	/* configure gpio */
+	reg_w(dev, 0x01, &sn9c1xx[1], 2);
+	reg_w(dev, 0x08, &sn9c1xx[8], 2);
+	reg_w(dev, 0x17, &sn9c1xx[0x17], 3);
+	switch (sd->customid) {
+	case SN9C325:
+		reg9a = reg9a_sn9c325;
+		break;
+	case SN9C120:
+		reg9a = reg9a_sn9c120;
+		break;
+	default:
+		reg9a = reg9a_def;
+		break;
+	}
+	reg_w(dev, 0x9a, reg9a, 6);
+
+	data = 0x60;				/*fixme:jfm 60 00 00 (3) */
+	reg_w(dev, 0xd4, &data, 1);
+
+	reg_w(dev, 0x03, &sn9c1xx[3], 0x0f);
+
+	switch (sd->customid) {
+	case SN9C120:				/* from win trace */
+		data = 0x61;
+		reg_w(dev, 0x01, &data, 1);
+		data = 0x20;
+		reg_w(dev, 0x17, &data, 1);
+		data = 0x60;
+		reg_w(dev, 0x01, &data, 1);
+		break;
+	case SN9C325:
+		data = 0x43;
+		reg_w(dev, 0x01, &data, 1);
+		data = 0xae;
+		reg_w(dev, 0x17, &data, 1);
+		data = 0x42;
+		reg_w(dev, 0x01, &data, 1);
+		break;
+	default:
+		data = 0x43;
+		reg_w(dev, 0x01, &data, 1);
+		data = 0x61;
+		reg_w(dev, 0x17, &data, 1);
+		data = 0x42;
+		reg_w(dev, 0x01, &data, 1);
+	}
+
+	if (sd->sensor == SENSOR_HV7131R) {
+		if (probesensor(gspca_dev) < 0)
+			return -ENODEV;
+	}
+	return 0;
+}
+
+static void hv7131R_InitSensor(struct gspca_dev *gspca_dev)
+{
+	int i = 0;
+	struct usb_device *dev = gspca_dev->dev;
+	static __u8 SetSensorClk[] =		/* 0x08 Mclk */
+		{ 0xa1, 0x11, 0x01, 0x18, 0x00, 0x00, 0x00, 0x10 };
+
+	while (hv7131r_sensor_init[i][0]) {
+		i2c_w8(dev, hv7131r_sensor_init[i]);
+		i++;
+	}
+	i2c_w8(dev, SetSensorClk);
+}
+
+static void mi0360_InitSensor(struct gspca_dev *gspca_dev)
+{
+	int i = 0;
+	struct usb_device *dev = gspca_dev->dev;
+
+	while (mi0360_sensor_init[i][0]) {
+		i2c_w8(dev, mi0360_sensor_init[i]);
+		i++;
+	}
+}
+
+static void mo4000_InitSensor(struct gspca_dev *gspca_dev)
+{
+	int i = 0;
+	struct usb_device *dev = gspca_dev->dev;
+
+	while (mo4000_sensor_init[i][0]) {
+		i2c_w8(dev, mo4000_sensor_init[i]);
+		i++;
+	}
+}
+
+static void ov7648_InitSensor(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int i = 0;
+
+	while (ov7648_sensor_init[i][0]) {
+		i2c_w8(dev, ov7648_sensor_init[i]);
+		i++;
+	}
+}
+
+static void ov7660_InitSensor(struct gspca_dev *gspca_dev)
+{
+	int i = 0;
+	struct usb_device *dev = gspca_dev->dev;
+
+	while (ov7660_sensor_init[i][0]) {
+		i2c_w8(dev, ov7660_sensor_init[i]);
+		i++;
+	}
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+	__u16 vendor;
+	__u16 product;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+	sd->sensor = -1;
+	switch (vendor) {
+	case 0x0458:				/* Genius */
+/*		switch (product) {
+		case 0x7025: */
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_MI0360;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x5d;
+/*			break;
+		} */
+		break;
+	case 0x045e:
+/*		switch (product) {
+		case 0x00f5:
+		case 0x00f7: */
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_OV7660;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x21;
+/*			break;
+		} */
+		break;
+	case 0x0471:				/* Philips */
+/*		switch (product) {
+		case 0x0327:
+		case 0x0328:
+		case 0x0330: */
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_MI0360;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x5d;
+/*			break;
+		} */
+		break;
+	case 0x0c45:				/* Sonix */
+		switch (product) {
+		case 0x6040:
+			sd->customid = SN9C102P;
+			sd->sensor = SENSOR_MI0360;	/* from BW600.inf */
+/*			sd->sensor = SENSOR_HV7131R;	 * gspcav1 value */
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x11;
+			break;
+/*		case 0x607a:				* from BW600.inf
+			sd->customid = SN9C102P;
+			sd->sensor = SENSOR_OV7648;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x607c:
+			sd->customid = SN9C102P;
+			sd->sensor = SENSOR_HV7131R;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x11;
+			break;
+/*		case 0x607e:				* from BW600.inf
+			sd->customid = SN9C102P;
+			sd->sensor = SENSOR_OV7630;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x60c0:
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_MI0360;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x5d;
+			break;
+/*		case 0x60c8:				* from BW600.inf
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_OM6801;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+/*		case 0x60cc:				* from BW600.inf
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_HV7131GP;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x60ec:
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_MO4000;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x21;
+			break;
+/*		case 0x60ef:				* from BW600.inf
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_ICM105C;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+/*		case 0x60fa:				* from BW600.inf
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_OV7648;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x60fb:
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_OV7660;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x21;
+			break;
+		case 0x60fc:
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_HV7131R;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x11;
+			break;
+/*		case 0x60fe:				* from BW600.inf
+			sd->customid = SN9C105;
+			sd->sensor = SENSOR_OV7630;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+/*		case 0x6108:				* from BW600.inf
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_OM6801;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+/*		case 0x6122:				* from BW600.inf
+			sd->customid = SN9C110;
+			sd->sensor = SENSOR_ICM105C;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x612a:
+/*			sd->customid = SN9C110;		 * in BW600.inf */
+			sd->customid = SN9C325;
+			sd->sensor = SENSOR_OV7648;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x21;
+			break;
+/*		case 0x6123:				* from BW600.inf
+			sd->customid = SN9C110;
+			sd->sensor = SENSOR_SanyoCCD;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x612c:
+			sd->customid = SN9C110;
+			sd->sensor = SENSOR_MO4000;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x21;
+			break;
+/*		case 0x612e:				* from BW600.inf
+			sd->customid = SN9C110;
+			sd->sensor = SENSOR_OV7630;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+/*		case 0x612f:				* from BW600.inf
+			sd->customid = SN9C110;
+			sd->sensor = SENSOR_ICM105C;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x6130:
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_MI0360;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x5d;
+			break;
+		case 0x6138:
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_MO4000;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x21;
+			break;
+/*		case 0x613a:				* from BW600.inf
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_OV7648;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		case 0x613b:
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_OV7660;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x21;
+			break;
+		case 0x613c:
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_HV7131R;
+			sd->i2c_ctrl_reg = 0x81;
+			sd->i2c_base = 0x11;
+			break;
+/*		case 0x613e:				* from BW600.inf
+			sd->customid = SN9C120;
+			sd->sensor = SENSOR_OV7630;
+			sd->i2c_ctrl_reg = 0x??;
+			sd->i2c_base = 0x??;
+			break; */
+		}
+		break;
+	}
+	if (sd->sensor < 0) {
+		PDEBUG(D_ERR, "Invalid vendor/product %04x:%04x",
+			vendor, product);
+		return -EINVAL;
+	}
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	cam->cam_mode = vga_mode;
+	cam->nmodes = ARRAY_SIZE(vga_mode);
+	sd->qindex = 4;			/* set the quantization table */
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	sd->autogain = sd_ctrls[SD_AUTOGAIN].qctrl.default_value;
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+/*	__u8 *sn9c1xx; */
+	__u8 regF1;
+	__u8 regGpio[] = { 0x29, 0x74 };
+
+	/* setup a selector by customid */
+	regF1 = 0x01;
+	reg_w(dev, 0xf1, &regF1, 1);
+	reg_r(dev, 0x00, &regF1, 1);	/* -> regF1 = 0x00 */
+	reg_w(dev, 0xf1, &regF1, 1);
+	reg_r(dev, 0x00, &regF1, 1);
+	switch (sd->customid) {
+	case SN9C102P:
+		if (regF1 != 0x11)
+			return -ENODEV;
+		reg_w(dev, 0x02, &regGpio[1], 1);
+		break;
+	case SN9C105:
+		if (regF1 != 0x11)
+			return -ENODEV;
+		reg_w(dev, 0x02, regGpio, 2);
+		break;
+	case SN9C110:
+		if (regF1 != 0x12)
+			return -ENODEV;
+		regGpio[1] = 0x62;
+		reg_w(dev, 0x02, &regGpio[1], 1);
+		break;
+	case SN9C120:
+		if (regF1 != 0x12)
+			return -ENODEV;
+		regGpio[1] = 0x70;
+		reg_w(dev, 0x02, regGpio, 2);
+		break;
+	default:
+/*	case SN9C325: */
+		if (regF1 != 0x12)
+			return -ENODEV;
+		regGpio[1] = 0x62;
+		reg_w(dev, 0x02, &regGpio[1], 1);
+		break;
+	}
+
+	regF1 = 0x01;
+	reg_w(dev, 0xf1, &regF1, 1);
+
+	return 0;
+}
+
+static unsigned int setexposure(struct gspca_dev *gspca_dev,
+				unsigned int expo)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	static __u8 doit[] =			/* update sensor */
+		{ 0xb1, 0x5d, 0x07, 0x00, 0x03, 0x00, 0x00, 0x10 };
+	static __u8 sensorgo[] =		/* sensor on */
+		{ 0xb1, 0x5d, 0x07, 0x00, 0x02, 0x00, 0x00, 0x10 };
+	static __u8 gainMo[] =
+		{ 0xa1, 0x21, 0x00, 0x10, 0x00, 0x00, 0x00, 0x1d };
+
+	switch (sd->sensor) {
+	case SENSOR_HV7131R: {
+		__u8 Expodoit[] =
+			{ 0xc1, 0x11, 0x25, 0x07, 0x27, 0xc0, 0x00, 0x16 };
+
+		Expodoit[3] = expo >> 16;
+		Expodoit[4] = expo >> 8;
+		Expodoit[5] = expo;
+		i2c_w8(gspca_dev->dev, Expodoit);
+		break;
+	    }
+	case SENSOR_MI0360: {
+		__u8 expoMi[] =	 /* exposure 0x0635 -> 4 fp/s 0x10 */
+			{ 0xb1, 0x5d, 0x09, 0x06, 0x35, 0x00, 0x00, 0x16 };
+
+		if (expo > 0x0635)
+			expo = 0x0635;
+		else if (expo < 0x0001)
+			expo = 0x0001;
+		expoMi[3] = expo >> 8;
+		expoMi[4] = expo;
+		i2c_w8(gspca_dev->dev, expoMi);
+		i2c_w8(gspca_dev->dev, doit);
+		i2c_w8(gspca_dev->dev, sensorgo);
+		break;
+	    }
+	case SENSOR_MO4000: {
+		__u8 expoMof[] =
+			{ 0xa1, 0x21, 0x0f, 0x20, 0x00, 0x00, 0x00, 0x10 };
+		__u8 expoMo10[] =
+			{ 0xa1, 0x21, 0x10, 0x20, 0x00, 0x00, 0x00, 0x10 };
+
+		if (expo > 0x1fff)
+			expo = 0x1fff;
+		else if (expo < 0x0001)
+			expo = 0x0001;
+		expoMof[3] = (expo & 0x03fc) >> 2;
+		i2c_w8(gspca_dev->dev, expoMof);
+		expoMo10[3] = ((expo & 0x1c00) >> 10)
+				| ((expo & 0x0003) << 4);
+		i2c_w8(gspca_dev->dev, expoMo10);
+		i2c_w8(gspca_dev->dev, gainMo);
+		PDEBUG(D_CONF," set exposure %d",
+			((expoMo10[3] & 0x07) << 10)
+			| (expoMof[3] << 2)
+			| ((expoMo10[3] & 0x30) >> 4));
+		break;
+	    }
+	}
+	return expo;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	unsigned int expo;
+	__u8 k2;
+
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+		expo = sd->brightness << 4;
+		if (expo > 0x002dc6c0)
+			expo = 0x002dc6c0;
+		else if (expo < 0x02a0)
+			expo = 0x02a0;
+		sd->exposure = setexposure(gspca_dev, expo);
+		break;
+	case SENSOR_MI0360:
+		expo = sd->brightness >> 4;
+		sd->exposure = setexposure(gspca_dev, expo);
+		break;
+	case SENSOR_MO4000:
+		expo = sd->brightness >> 4;
+		sd->exposure = setexposure(gspca_dev, expo);
+		break;
+	case SENSOR_OV7660:
+		return;				/*jfm??*/
+	}
+
+	k2 = sd->brightness >> 10;
+	reg_w(gspca_dev->dev, 0x96, &k2, 1);
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 k2;
+	__u8 contrast[] = { 0x00, 0x00, 0x28, 0x00, 0x07, 0x00 };
+
+	if (sd->sensor == SENSOR_OV7660)
+		return;				/*jfm??*/
+	k2 = sd->contrast;
+	contrast[2] = k2;
+	contrast[0] = (k2 + 1) >> 1;
+	contrast[4] = (k2 + 1) / 5;
+	reg_w(gspca_dev->dev, 0x84, contrast, 6);
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 data;
+	int colour;
+
+	colour = sd->colors - 128;
+	if (colour > 0)
+		data = (colour + 32) & 0x7f;	/* blue */
+	else
+		data = (-colour + 32) & 0x7f;	/* red */
+	reg_w(gspca_dev->dev, 0x05, &data, 1);
+}
+
+/* -- start the camera -- */
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int i;
+	__u8 data;
+	__u8 reg1;
+	__u8 reg17;
+	__u8 *sn9c1xx;
+	int mode;
+	static __u8 DC29[] = { 0x6a, 0x50, 0x00, 0x00, 0x50, 0x3c };
+	static __u8 C0[] = { 0x2d, 0x2d, 0x3a, 0x05, 0x04, 0x3f };
+	static __u8 CA[] = { 0x28, 0xd8, 0x14, 0xec };
+	static __u8 CA_sn9c120[] = { 0x14, 0xec, 0x0a, 0xf6 }; /* SN9C120 */
+	static __u8 CE[] = { 0x32, 0xdd, 0x2d, 0xdd };	/* MI0360 */
+	static __u8 CE_sn9c325[] =
+			{ 0x32, 0xdd, 0x32, 0xdd };	/* OV7648 - SN9C325 */
+
+	sn9c1xx = sn_tb[(int) sd->sensor];
+	configure_gpio(gspca_dev, sn9c1xx);
+
+/*fixme:jfm this sequence should appear at end of sd_start */
+/* with
+	data = 0x44;
+	reg_w(dev, 0x01, &data, 1); */
+	reg_w(dev, 0x15, &sn9c1xx[0x15], 1);
+	reg_w(dev, 0x16, &sn9c1xx[0x16], 1);
+	reg_w(dev, 0x12, &sn9c1xx[0x12], 1);
+	reg_w(dev, 0x13, &sn9c1xx[0x13], 1);
+	reg_w(dev, 0x18, &sn9c1xx[0x18], 1);
+	reg_w(dev, 0xd2, &DC29[0], 1);
+	reg_w(dev, 0xd3, &DC29[1], 1);
+	reg_w(dev, 0xc6, &DC29[2], 1);
+	reg_w(dev, 0xc7, &DC29[3], 1);
+	reg_w(dev, 0xc8, &DC29[4], 1);
+	reg_w(dev, 0xc9, &DC29[5], 1);
+/*fixme:jfm end of ending sequence */
+	reg_w(dev, 0x18, &sn9c1xx[0x18], 1);
+	if (sd->customid == SN9C325)
+		data = 0xae;
+	else
+		data = 0x60;
+	reg_w(dev, 0x17, &data, 1);
+	reg_w(dev, 0x05, &sn9c1xx[5], 1);
+	reg_w(dev, 0x07, &sn9c1xx[7], 1);
+	reg_w(dev, 0x06, &sn9c1xx[6], 1);
+	reg_w(dev, 0x14, &sn9c1xx[0x14], 1);
+	if (sd->customid == SN9C325) {
+		reg_w(dev, 0x20, regsn20_sn9c325, 0x11);
+		for (i = 0; i < 8; i++)
+			reg_w(dev, 0x84, reg84_sn9c325, 0x15);
+		data = 0x0a;
+		reg_w(dev, 0x9a, &data, 1);
+		data = 0x60;
+		reg_w(dev, 0x99, &data, 1);
+	} else {
+		reg_w(dev, 0x20, regsn20, 0x11);
+		for (i = 0; i < 8; i++)
+			reg_w(dev, 0x84, reg84, 0x15);
+		data = 0x08;
+		reg_w(dev, 0x9a, &data, 1);
+		data = 0x59;
+		reg_w(dev, 0x99, &data, 1);
+	}
+
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+	reg1 = 0x02;
+	reg17 = 0x61;
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+		hv7131R_InitSensor(gspca_dev);
+		if (mode)
+			reg1 = 0x46;	/* 320 clk 48Mhz */
+		else
+			reg1 = 0x06;	/* 640 clk 24Mz */
+		break;
+	case SENSOR_MI0360:
+		mi0360_InitSensor(gspca_dev);
+		if (mode)
+			reg1 = 0x46;	/* 320 clk 48Mhz */
+		else
+			reg1 = 0x06;	/* 640 clk 24Mz */
+		break;
+	case SENSOR_MO4000:
+		mo4000_InitSensor(gspca_dev);
+		if (mode) {
+/*			reg1 = 0x46;	 * 320 clk 48Mhz 60fp/s */
+			reg1 = 0x06;	/* clk 24Mz */
+		} else {
+			reg17 = 0x22;	/* 640 MCKSIZE */
+			reg1 = 0x06;	/* 640 clk 24Mz */
+		}
+		break;
+	case SENSOR_OV7648:
+		reg17 = 0xa2;
+		reg1 = 0x44;
+		ov7648_InitSensor(gspca_dev);
+/*		if (mode)
+			;		 * 320x2...
+		else
+			;		 * 640x... */
+		break;
+	default:
+/*	case SENSOR_OV7660: */
+		ov7660_InitSensor(gspca_dev);
+		if (mode) {
+/*			reg17 = 0x21;	 * 320 */
+/*			reg1 = 0x44; */
+			reg1 = 0x46;
+		} else {
+			reg17 = 0xa2;	/* 640 */
+			reg1 = 0x40;
+		}
+		break;
+	}
+	reg_w(dev, 0xc0, C0, 6);
+	switch (sd->customid) {
+	case SN9C120:			/*jfm ?? */
+		reg_w(dev, 0xca, CA_sn9c120, 4);
+		break;
+	default:
+		reg_w(dev, 0xca, CA, 4);
+		break;
+	}
+	switch (sd->customid) {
+	case SN9C120:			/*jfm ?? */
+	case SN9C325:
+		reg_w(dev, 0xce, CE_sn9c325, 4);
+		break;
+	default:
+		reg_w(dev, 0xce, CE, 4);
+					/* ?? {0x1e, 0xdd, 0x2d, 0xe7} */
+		break;
+	}
+
+	/* here change size mode 0 -> VGA; 1 -> CIF */
+	data = 0x40 | sn9c1xx[0x18] | (mode << 4);
+	reg_w(dev, 0x18, &data, 1);
+
+	reg_w(dev, 0x100, qtable4, 0x40);
+	reg_w(dev, 0x140, qtable4 + 0x40, 0x40);
+
+	data = sn9c1xx[0x18] | (mode << 4);
+	reg_w(dev, 0x18, &data, 1);
+
+	reg_w(dev, 0x17, &reg17, 1);
+	reg_w(dev, 0x01, &reg1, 1);
+	setbrightness(gspca_dev);
+	setcontrast(gspca_dev);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	static __u8 stophv7131[] =
+		{ 0xa1, 0x11, 0x02, 0x09, 0x00, 0x00, 0x00, 0x10 };
+	static __u8 stopmi0360[] =
+		{ 0xb1, 0x5d, 0x07, 0x00, 0x00, 0x00, 0x00, 0x10 };
+	__u8 regF1;
+	__u8 data;
+	__u8 *sn9c1xx;
+
+	data = 0x0b;
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+		i2c_w8(dev, stophv7131);
+		data = 0x2b;
+		break;
+	case SENSOR_MI0360:
+		i2c_w8(dev, stopmi0360);
+		data = 0x29;
+		break;
+	case SENSOR_MO4000:
+		break;
+	case SENSOR_OV7648:
+		data = 0x29;
+		break;
+	default:
+/*	case SENSOR_OV7660: */
+		break;
+	}
+	sn9c1xx = sn_tb[(int) sd->sensor];
+	reg_w(dev, 0x01, &sn9c1xx[1], 1);
+	reg_w(dev, 0x17, &sn9c1xx[0x17], 1);
+	reg_w(dev, 0x01, &sn9c1xx[1], 1);
+	reg_w(dev, 0x01, &data, 1);
+	regF1 = 0x01;
+	reg_w(dev, 0xf1, &regF1, 1);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void setautogain(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	/* Thanks S., without your advice, autobright should not work :) */
+	int delta;
+	int expotimes = 0;
+	__u8 luma_mean = 130;
+	__u8 luma_delta = 20;
+
+	delta = sd->avg_lum;
+	if (delta < luma_mean - luma_delta ||
+	    delta > luma_mean + luma_delta) {
+		switch (sd->sensor) {
+		case SENSOR_HV7131R:
+			expotimes = sd->exposure >> 8;
+			expotimes += (luma_mean - delta) >> 4;
+			if (expotimes < 0)
+				expotimes = 0;
+			sd->exposure = setexposure(gspca_dev,
+					(unsigned int) (expotimes << 8));
+			break;
+		case SENSOR_MO4000:
+		case SENSOR_MI0360:
+			expotimes = sd->exposure;
+			expotimes += (luma_mean - delta) >> 6;
+			if (expotimes < 0)
+				expotimes = 0;
+			sd->exposure = setexposure(gspca_dev,
+						   (unsigned int) expotimes);
+			setcolors(gspca_dev);
+			break;
+		}
+	}
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int sof, avg_lum;
+
+	sof = len - 64;
+	if (sof >= 0 && data[sof] == 0xff && data[sof + 1] == 0xd9) {
+
+		/* end of frame */
+		gspca_frame_add(gspca_dev, LAST_PACKET,
+				frame, data, sof + 2);
+		if (sd->ag_cnt < 0)
+			return;
+		if (--sd->ag_cnt >= 0)
+			return;
+		sd->ag_cnt = AG_CNT_START;
+/* w1 w2 w3 */
+/* w4 w5 w6 */
+/* w7 w8 */
+/* w4 */
+		avg_lum = ((data[sof + 29] << 8) | data[sof + 30]) >> 6;
+/* w6 */
+		avg_lum += ((data[sof + 33] << 8) | data[sof + 34]) >> 6;
+/* w2 */
+		avg_lum += ((data[sof + 25] << 8) | data[sof + 26]) >> 6;
+/* w8 */
+		avg_lum += ((data[sof + 37] << 8) | data[sof + 38]) >> 6;
+/* w5 */
+		avg_lum += ((data[sof + 31] << 8) | data[sof + 32]) >> 4;
+		avg_lum >>= 4;
+		sd->avg_lum = avg_lum;
+		PDEBUG(D_PACK, "mean lum %d", avg_lum);
+		setautogain(gspca_dev);
+		return;
+	}
+	if (gspca_dev->last_packet_type == LAST_PACKET) {
+
+		/* put the JPEG 422 header */
+		jpeg_put_header(gspca_dev, frame, sd->qindex, 0x21);
+	}
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+}
+
+static unsigned int getexposure(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 hexpo, mexpo, lexpo;
+	__u8 expo[6];
+
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+		/* read sensor exposure */
+		i2c_r5(gspca_dev, 0x25, expo);
+		return (expo[0] << 16) | (expo[1] << 8) | expo[2];
+	case SENSOR_MI0360:
+		/* read sensor exposure */
+		i2c_r5(gspca_dev, 0x09, expo);
+		return (expo[0] << 8) | expo[1];
+	case SENSOR_MO4000:
+		i2c_r5(gspca_dev, 0x0e, expo);
+		hexpo = 0;		/* expo[1] & 0x07; */
+		mexpo = 0x40;		/* expo[2] &0xff; */
+		lexpo = (expo[1] & 0x30) >> 4;
+		PDEBUG(D_CONF, "exposure %d",
+			(hexpo << 10) | (mexpo << 2) | lexpo);
+		return (hexpo << 10) | (mexpo << 2) | lexpo;
+	default:
+/*	case SENSOR_OV7660: */
+		/* read sensor exposure */
+		i2c_r5(gspca_dev, 0x04, expo);
+		hexpo = expo[3] & 0x2f;
+		lexpo = expo[0] & 0x02;
+		i2c_r5(gspca_dev, 0x08, expo);
+		mexpo = expo[2];
+		return (hexpo << 10) | (mexpo << 2) | lexpo;
+	}
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	/* hardcoded registers seem not readable */
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+/*		sd->brightness = 0x7fff; */
+		sd->brightness = getexposure(gspca_dev) >> 4;
+		break;
+	case SENSOR_MI0360:
+		sd->brightness = getexposure(gspca_dev) << 4;
+		break;
+	case SENSOR_MO4000:
+/*		sd->brightness = 0x1fff; */
+		sd->brightness = getexposure(gspca_dev) << 4;
+		break;
+	}
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->colors;
+	return 0;
+}
+
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->autogain = val;
+	if (val)
+		sd->ag_cnt = AG_CNT_START;
+	else
+		sd->ag_cnt = -1;
+	return 0;
+}
+
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autogain;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x0458, 0x7025), DVNM("Genius Eye 311Q")},
+	{USB_DEVICE(0x045e, 0x00f5), DVNM("MicroSoft VX3000")},
+	{USB_DEVICE(0x045e, 0x00f7), DVNM("MicroSoft VX1000")},
+	{USB_DEVICE(0x0471, 0x0327), DVNM("Philips SPC 600 NC")},
+	{USB_DEVICE(0x0471, 0x0328), DVNM("Philips SPC 700 NC")},
+	{USB_DEVICE(0x0471, 0x0330), DVNM("Philips SPC 710NC")},
+	{USB_DEVICE(0x0c45, 0x6040), DVNM("Speed NVC 350K")},
+	{USB_DEVICE(0x0c45, 0x607c), DVNM("Sonix sn9c102p Hv7131R")},
+	{USB_DEVICE(0x0c45, 0x60c0), DVNM("Sangha Sn535")},
+	{USB_DEVICE(0x0c45, 0x60ec), DVNM("SN9C105+MO4000")},
+	{USB_DEVICE(0x0c45, 0x60fb), DVNM("Surfer NoName")},
+	{USB_DEVICE(0x0c45, 0x60fc), DVNM("LG-LIC300")},
+	{USB_DEVICE(0x0c45, 0x612a), DVNM("Avant Camera")},
+	{USB_DEVICE(0x0c45, 0x612c), DVNM("Typhoon Rasy Cam 1.3MPix")},
+	{USB_DEVICE(0x0c45, 0x6130), DVNM("Sonix Pccam")},
+	{USB_DEVICE(0x0c45, 0x6138), DVNM("Sn9c120 Mo4000")},
+	{USB_DEVICE(0x0c45, 0x613b), DVNM("Surfer SN-206")},
+	{USB_DEVICE(0x0c45, 0x613c), DVNM("Sonix Pccam168")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+		    const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	info("v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	info("deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/spca500.c b/drivers/media/video/gspca/spca500.c
new file mode 100644
index 000000000000..c0dd969a3106
--- /dev/null
+++ b/drivers/media/video/gspca/spca500.c
@@ -0,0 +1,1195 @@
+/*
+ * SPCA500 chip based cameras initialization data
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#define MODULE_NAME "spca500"
+
+#include "gspca.h"
+#include "jpeg.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SPCA500 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;		/* !! must be the first item */
+
+	unsigned char packet[ISO_MAX_SIZE + 128];
+				 /* !! no more than 128 ff in an ISO packet */
+
+	unsigned char brightness;
+	unsigned char contrast;
+	unsigned char colors;
+
+	char qindex;
+	char subtype;
+#define AgfaCl20 0
+#define AiptekPocketDV 1
+#define BenqDC1016 2
+#define CreativePCCam300 3
+#define DLinkDSC350 4
+#define Gsmartmini 5
+#define IntelPocketPCCamera 6
+#define KodakEZ200 7
+#define LogitechClickSmart310 8
+#define LogitechClickSmart510 9
+#define LogitechTraveler 10
+#define MustekGsmart300 11
+#define Optimedia 12
+#define PalmPixDC85 13
+#define ToptroIndus 14
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0x7f,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define SD_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Color",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 320, 240, 1},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 0},
+};
+
+static struct cam_mode sif_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 176, 144, 1},
+	{V4L2_PIX_FMT_JPEG, 352, 288, 0},
+};
+
+/* Frame packet header offsets for the spca500 */
+#define SPCA500_OFFSET_PADDINGLB 2
+#define SPCA500_OFFSET_PADDINGHB 3
+#define SPCA500_OFFSET_MODE      4
+#define SPCA500_OFFSET_IMGWIDTH  5
+#define SPCA500_OFFSET_IMGHEIGHT 6
+#define SPCA500_OFFSET_IMGMODE   7
+#define SPCA500_OFFSET_QTBLINDEX 8
+#define SPCA500_OFFSET_FRAMSEQ   9
+#define SPCA500_OFFSET_CDSPINFO  10
+#define SPCA500_OFFSET_GPIO      11
+#define SPCA500_OFFSET_AUGPIO    12
+#define SPCA500_OFFSET_DATA      16
+
+
+static __u16 spca500_visual_defaults[][3] = {
+	{0x00, 0x0003, 0x816b},	/* SSI not active sync with vsync,
+				 * hue (H byte) = 0,
+				 * saturation/hue enable,
+				 * brightness/contrast enable.
+				 */
+	{0x00, 0x0000, 0x8167},	/* brightness = 0 */
+	{0x00, 0x0020, 0x8168},	/* contrast = 0 */
+	{0x00, 0x0003, 0x816b},	/* SSI not active sync with vsync,
+				 * hue (H byte) = 0, saturation/hue enable,
+				 * brightness/contrast enable.
+				 * was 0x0003, now 0x0000.
+				 */
+	{0x00, 0x0000, 0x816a},	/* hue (L byte) = 0 */
+	{0x00, 0x0020, 0x8169},	/* saturation = 0x20 */
+	{0x00, 0x0050, 0x8157},	/* edge gain high threshold */
+	{0x00, 0x0030, 0x8158},	/* edge gain low threshold */
+	{0x00, 0x0028, 0x8159},	/* edge bandwidth high threshold */
+	{0x00, 0x000a, 0x815a},	/* edge bandwidth low threshold */
+	{0x00, 0x0001, 0x8202},	/* clock rate compensation = 1/25 sec/frame */
+	{0x0c, 0x0004, 0x0000},
+	/* set interface */
+
+	{0, 0, 0}
+};
+static __u16 Clicksmart510_defaults[][3] = {
+	{0x00, 0x00, 0x8211},
+	{0x00, 0x01, 0x82c0},
+	{0x00, 0x10, 0x82cb},
+	{0x00, 0x0f, 0x800d},
+	{0x00, 0x82, 0x8225},
+	{0x00, 0x21, 0x8228},
+	{0x00, 0x00, 0x8203},
+	{0x00, 0x00, 0x8204},
+	{0x00, 0x08, 0x8205},
+	{0x00, 0xf8, 0x8206},
+	{0x00, 0x28, 0x8207},
+	{0x00, 0xa0, 0x8208},
+	{0x00, 0x08, 0x824a},
+	{0x00, 0x08, 0x8214},
+	{0x00, 0x80, 0x82c1},
+	{0x00, 0x00, 0x82c2},
+	{0x00, 0x00, 0x82ca},
+	{0x00, 0x80, 0x82c1},
+	{0x00, 0x04, 0x82c2},
+	{0x00, 0x00, 0x82ca},
+	{0x00, 0xfc, 0x8100},
+	{0x00, 0xfc, 0x8105},
+	{0x00, 0x30, 0x8101},
+	{0x00, 0x00, 0x8102},
+	{0x00, 0x00, 0x8103},
+	{0x00, 0x66, 0x8107},
+	{0x00, 0x00, 0x816b},
+	{0x00, 0x00, 0x8155},
+	{0x00, 0x01, 0x8156},
+	{0x00, 0x60, 0x8157},
+	{0x00, 0x40, 0x8158},
+	{0x00, 0x0a, 0x8159},
+	{0x00, 0x06, 0x815a},
+	{0x00, 0x00, 0x813f},
+	{0x00, 0x00, 0x8200},
+	{0x00, 0x19, 0x8201},
+	{0x00, 0x00, 0x82c1},
+	{0x00, 0xa0, 0x82c2},
+	{0x00, 0x00, 0x82ca},
+	{0x00, 0x00, 0x8117},
+	{0x00, 0x00, 0x8118},
+	{0x00, 0x65, 0x8119},
+	{0x00, 0x00, 0x811a},
+	{0x00, 0x00, 0x811b},
+	{0x00, 0x55, 0x811c},
+	{0x00, 0x65, 0x811d},
+	{0x00, 0x55, 0x811e},
+	{0x00, 0x16, 0x811f},
+	{0x00, 0x19, 0x8120},
+	{0x00, 0x80, 0x8103},
+	{0x00, 0x83, 0x816b},
+	{0x00, 0x25, 0x8168},
+	{0x00, 0x01, 0x820f},
+	{0x00, 0xff, 0x8115},
+	{0x00, 0x48, 0x8116},
+	{0x00, 0x50, 0x8151},
+	{0x00, 0x40, 0x8152},
+	{0x00, 0x78, 0x8153},
+	{0x00, 0x40, 0x8154},
+	{0x00, 0x00, 0x8167},
+	{0x00, 0x20, 0x8168},
+	{0x00, 0x00, 0x816a},
+	{0x00, 0x03, 0x816b},
+	{0x00, 0x20, 0x8169},
+	{0x00, 0x60, 0x8157},
+	{0x00, 0x00, 0x8190},
+	{0x00, 0x00, 0x81a1},
+	{0x00, 0x00, 0x81b2},
+	{0x00, 0x27, 0x8191},
+	{0x00, 0x27, 0x81a2},
+	{0x00, 0x27, 0x81b3},
+	{0x00, 0x4b, 0x8192},
+	{0x00, 0x4b, 0x81a3},
+	{0x00, 0x4b, 0x81b4},
+	{0x00, 0x66, 0x8193},
+	{0x00, 0x66, 0x81a4},
+	{0x00, 0x66, 0x81b5},
+	{0x00, 0x79, 0x8194},
+	{0x00, 0x79, 0x81a5},
+	{0x00, 0x79, 0x81b6},
+	{0x00, 0x8a, 0x8195},
+	{0x00, 0x8a, 0x81a6},
+	{0x00, 0x8a, 0x81b7},
+	{0x00, 0x9b, 0x8196},
+	{0x00, 0x9b, 0x81a7},
+	{0x00, 0x9b, 0x81b8},
+	{0x00, 0xa6, 0x8197},
+	{0x00, 0xa6, 0x81a8},
+	{0x00, 0xa6, 0x81b9},
+	{0x00, 0xb2, 0x8198},
+	{0x00, 0xb2, 0x81a9},
+	{0x00, 0xb2, 0x81ba},
+	{0x00, 0xbe, 0x8199},
+	{0x00, 0xbe, 0x81aa},
+	{0x00, 0xbe, 0x81bb},
+	{0x00, 0xc8, 0x819a},
+	{0x00, 0xc8, 0x81ab},
+	{0x00, 0xc8, 0x81bc},
+	{0x00, 0xd2, 0x819b},
+	{0x00, 0xd2, 0x81ac},
+	{0x00, 0xd2, 0x81bd},
+	{0x00, 0xdb, 0x819c},
+	{0x00, 0xdb, 0x81ad},
+	{0x00, 0xdb, 0x81be},
+	{0x00, 0xe4, 0x819d},
+	{0x00, 0xe4, 0x81ae},
+	{0x00, 0xe4, 0x81bf},
+	{0x00, 0xed, 0x819e},
+	{0x00, 0xed, 0x81af},
+	{0x00, 0xed, 0x81c0},
+	{0x00, 0xf7, 0x819f},
+	{0x00, 0xf7, 0x81b0},
+	{0x00, 0xf7, 0x81c1},
+	{0x00, 0xff, 0x81a0},
+	{0x00, 0xff, 0x81b1},
+	{0x00, 0xff, 0x81c2},
+	{0x00, 0x03, 0x8156},
+	{0x00, 0x00, 0x8211},
+	{0x00, 0x20, 0x8168},
+	{0x00, 0x01, 0x8202},
+	{0x00, 0x30, 0x8101},
+	{0x00, 0x00, 0x8111},
+	{0x00, 0x00, 0x8112},
+	{0x00, 0x00, 0x8113},
+	{0x00, 0x00, 0x8114},
+	{}
+};
+
+static unsigned char qtable_creative_pccam[2][64] = {
+	{				/* Q-table Y-components */
+	 0x05, 0x03, 0x03, 0x05, 0x07, 0x0c, 0x0f, 0x12,
+	 0x04, 0x04, 0x04, 0x06, 0x08, 0x11, 0x12, 0x11,
+	 0x04, 0x04, 0x05, 0x07, 0x0c, 0x11, 0x15, 0x11,
+	 0x04, 0x05, 0x07, 0x09, 0x0f, 0x1a, 0x18, 0x13,
+	 0x05, 0x07, 0x0b, 0x11, 0x14, 0x21, 0x1f, 0x17,
+	 0x07, 0x0b, 0x11, 0x13, 0x18, 0x1f, 0x22, 0x1c,
+	 0x0f, 0x13, 0x17, 0x1a, 0x1f, 0x24, 0x24, 0x1e,
+	 0x16, 0x1c, 0x1d, 0x1d, 0x22, 0x1e, 0x1f, 0x1e},
+	{				/* Q-table C-components */
+	 0x05, 0x05, 0x07, 0x0e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x05, 0x06, 0x08, 0x14, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x07, 0x08, 0x11, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x0e, 0x14, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e}
+};
+
+static unsigned char qtable_kodak_ez200[2][64] = {
+	{				/* Q-table Y-components */
+	 0x02, 0x01, 0x01, 0x02, 0x02, 0x04, 0x05, 0x06,
+	 0x01, 0x01, 0x01, 0x02, 0x03, 0x06, 0x06, 0x06,
+	 0x01, 0x01, 0x02, 0x02, 0x04, 0x06, 0x07, 0x06,
+	 0x01, 0x02, 0x02, 0x03, 0x05, 0x09, 0x08, 0x06,
+	 0x02, 0x02, 0x04, 0x06, 0x07, 0x0b, 0x0a, 0x08,
+	 0x02, 0x04, 0x06, 0x06, 0x08, 0x0a, 0x0b, 0x09,
+	 0x05, 0x06, 0x08, 0x09, 0x0a, 0x0c, 0x0c, 0x0a,
+	 0x07, 0x09, 0x0a, 0x0a, 0x0b, 0x0a, 0x0a, 0x0a},
+	{				/* Q-table C-components */
+	 0x02, 0x02, 0x02, 0x05, 0x0a, 0x0a, 0x0a, 0x0a,
+	 0x02, 0x02, 0x03, 0x07, 0x0a, 0x0a, 0x0a, 0x0a,
+	 0x02, 0x03, 0x06, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a,
+	 0x05, 0x07, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a,
+	 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a,
+	 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a,
+	 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a,
+	 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a}
+};
+
+static unsigned char qtable_pocketdv[2][64] = {
+	{			/* Q-table Y-components start registers 0x8800 */
+	 0x06, 0x04, 0x04, 0x06, 0x0a, 0x10, 0x14, 0x18,
+	 0x05, 0x05, 0x06, 0x08, 0x0a, 0x17, 0x18, 0x16,
+	 0x06, 0x05, 0x06, 0x0a, 0x10, 0x17, 0x1c, 0x16,
+	 0x06, 0x07, 0x09, 0x0c, 0x14, 0x23, 0x20, 0x19,
+	 0x07, 0x09, 0x0f, 0x16, 0x1b, 0x2c, 0x29, 0x1f,
+	 0x0a, 0x0e, 0x16, 0x1a, 0x20, 0x2a, 0x2d, 0x25,
+	 0x14, 0x1a, 0x1f, 0x23, 0x29, 0x30, 0x30, 0x28,
+	 0x1d, 0x25, 0x26, 0x27, 0x2d, 0x28, 0x29, 0x28,
+	 },
+	{			/* Q-table C-components start registers 0x8840 */
+	 0x07, 0x07, 0x0a, 0x13, 0x28, 0x28, 0x28, 0x28,
+	 0x07, 0x08, 0x0a, 0x1a, 0x28, 0x28, 0x28, 0x28,
+	 0x0a, 0x0a, 0x16, 0x28, 0x28, 0x28, 0x28, 0x28,
+	 0x13, 0x1a, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28,
+	 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28,
+	 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28,
+	 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28,
+	 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28, 0x28}
+};
+
+static void spca5xxRegRead(struct usb_device *dev,
+			   __u16 index,
+			   __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,		/* value */
+			index, buffer, length, 500);
+}
+
+static int reg_write(struct usb_device *dev,
+		     __u16 req, __u16 index, __u16 value)
+{
+	int ret;
+
+	ret = usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			req,
+			USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, NULL, 0, 500);
+	PDEBUG(D_USBO, "reg write: [0x%02x] = 0x%02x, 0x%x",
+		index, value, ret);
+	if (ret < 0)
+		PDEBUG(D_ERR, "reg write: error %d", ret);
+	return ret;
+}
+
+/* returns: negative is error, pos or zero is data */
+static int reg_read(struct usb_device *dev,
+			__u16 req,	/* bRequest */
+			__u16 index,	/* wIndex */
+			__u16 length)	/* wLength (1 or 2 only) */
+{
+	int ret;
+	__u8 buf[2];
+
+	buf[1] = 0;
+	ret = usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			req,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,		/* value */
+			index,
+			buf, length,
+			500);		/* timeout */
+	if (ret < 0) {
+		PDEBUG(D_ERR, "reg_read err %d", ret);
+		return -1;
+	}
+	return (buf[1] << 8) + buf[0];
+}
+
+/*
+ * Simple function to wait for a given 8-bit value to be returned from
+ * a reg_read call.
+ * Returns: negative is error or timeout, zero is success.
+ */
+static int reg_readwait(struct usb_device *dev,
+			__u16 reg, __u16 index, __u16 value)
+{
+	int ret, cnt = 20;
+
+	while (--cnt > 0) {
+		ret = reg_read(dev, reg, index, 1);
+		if (ret == value)
+			return 0;
+		msleep(50);
+	}
+	return -EIO;
+}
+
+static int write_vector(struct gspca_dev *gspca_dev,
+				__u16 data[][3])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int ret, i = 0;
+
+	while (data[i][0] != 0 || data[i][1] != 0 || data[i][2] != 0) {
+		ret = reg_write(dev, data[i][0], data[i][2], data[i][1]);
+		if (ret < 0)
+			return ret;
+		i++;
+	}
+	return 0;
+}
+
+static int spca50x_setup_qtable(struct gspca_dev *gspca_dev,
+				unsigned int request,
+				unsigned int ybase,
+				unsigned int cbase,
+				unsigned char qtable[2][64])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int i, err;
+
+	/* loop over y components */
+	for (i = 0; i < 64; i++) {
+		err = reg_write(dev, request, ybase + i, qtable[0][i]);
+		if (err < 0)
+			return err;
+	}
+
+	/* loop over c components */
+	for (i = 0; i < 64; i++) {
+		err = reg_write(dev, request, cbase + i, qtable[1][i]);
+		if (err < 0)
+			return err;
+	}
+	return 0;
+}
+
+static void spca500_ping310(struct gspca_dev *gspca_dev)
+{
+	__u8 Data[2];
+
+	spca5xxRegRead(gspca_dev->dev, 0x0d04, Data, 2);
+	PDEBUG(D_PACK, "ClickSmart310 ping 0x0d04 0x%02x 0x%02x",
+		Data[0], Data[1]);
+}
+
+static void spca500_clksmart310_init(struct gspca_dev *gspca_dev)
+{
+	__u8 Data[2];
+
+	spca5xxRegRead(gspca_dev->dev, 0x0d05, Data, 2);
+	PDEBUG(D_PACK, "ClickSmart310 init 0x0d05 0x%02x 0x%02x", Data[0],
+		Data[1]);
+	reg_write(gspca_dev->dev, 0x00, 0x8167, 0x5a);
+	spca500_ping310(gspca_dev);
+
+	reg_write(gspca_dev->dev, 0x00, 0x8168, 0x22);
+	reg_write(gspca_dev->dev, 0x00, 0x816a, 0xc0);
+	reg_write(gspca_dev->dev, 0x00, 0x816b, 0x0b);
+	reg_write(gspca_dev->dev, 0x00, 0x8169, 0x25);
+	reg_write(gspca_dev->dev, 0x00, 0x8157, 0x5b);
+	reg_write(gspca_dev->dev, 0x00, 0x8158, 0x5b);
+	reg_write(gspca_dev->dev, 0x00, 0x813f, 0x03);
+	reg_write(gspca_dev->dev, 0x00, 0x8151, 0x4a);
+	reg_write(gspca_dev->dev, 0x00, 0x8153, 0x78);
+	reg_write(gspca_dev->dev, 0x00, 0x0d01, 0x04);
+						/* 00 for adjust shutter */
+	reg_write(gspca_dev->dev, 0x00, 0x0d02, 0x01);
+	reg_write(gspca_dev->dev, 0x00, 0x8169, 0x25);
+	reg_write(gspca_dev->dev, 0x00, 0x0d01, 0x02);
+}
+
+static void spca500_setmode(struct gspca_dev *gspca_dev,
+			__u8 xmult, __u8 ymult)
+{
+	int mode;
+
+	/* set x multiplier */
+	reg_write(gspca_dev->dev, 0, 0x8001, xmult);
+
+	/* set y multiplier */
+	reg_write(gspca_dev->dev, 0, 0x8002, ymult);
+
+	/* use compressed mode, VGA, with mode specific subsample */
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+	reg_write(gspca_dev->dev, 0, 0x8003, mode << 4);
+}
+
+static int spca500_full_reset(struct gspca_dev *gspca_dev)
+{
+	int err;
+
+	/* send the reset command */
+	err = reg_write(gspca_dev->dev, 0xe0, 0x0001, 0x0000);
+	if (err < 0)
+		return err;
+
+	/* wait for the reset to complete */
+	err = reg_readwait(gspca_dev->dev, 0x06, 0x0000, 0x0000);
+	if (err < 0)
+		return err;
+	err = reg_write(gspca_dev->dev, 0xe0, 0x0000, 0x0000);
+	if (err < 0)
+		return err;
+	err = reg_readwait(gspca_dev->dev, 0x06, 0, 0);
+	if (err < 0) {
+		PDEBUG(D_ERR, "reg_readwait() failed");
+		return err;
+	}
+	/* all ok */
+	return 0;
+}
+
+/* Synchro the Bridge with sensor */
+/* Maybe that will work on all spca500 chip */
+/* because i only own a clicksmart310 try for that chip */
+/* using spca50x_set_packet_size() cause an Ooops here */
+/* usb_set_interface from kernel 2.6.x clear all the urb stuff */
+/* up-port the same feature as in 2.4.x kernel */
+static int spca500_synch310(struct gspca_dev *gspca_dev)
+{
+	__u8 Data;
+
+	if (usb_set_interface(gspca_dev->dev, gspca_dev->iface, 0) < 0) {
+		PDEBUG(D_ERR, "Set packet size: set interface error");
+		goto error;
+	}
+	spca500_ping310(gspca_dev);
+
+	spca5xxRegRead(gspca_dev->dev, 0x0d00, &Data, 1);
+
+	/* need alt setting here */
+	PDEBUG(D_PACK, "ClickSmart310 sync alt: %d", gspca_dev->alt);
+
+	/* Windoze use pipe with altsetting 6 why 7 here */
+	if (usb_set_interface(gspca_dev->dev,
+				gspca_dev->iface,
+				gspca_dev->alt) < 0) {
+		PDEBUG(D_ERR, "Set packet size: set interface error");
+		goto error;
+	}
+	return 0;
+error:
+	return -EBUSY;
+}
+
+static void spca500_reinit(struct gspca_dev *gspca_dev)
+{
+	int err;
+	__u8 Data;
+
+	/* some unknow command from Aiptek pocket dv and family300 */
+
+	reg_write(gspca_dev->dev, 0x00, 0x0d01, 0x01);
+	reg_write(gspca_dev->dev, 0x00, 0x0d03, 0x00);
+	reg_write(gspca_dev->dev, 0x00, 0x0d02, 0x01);
+
+	/* enable drop packet */
+	reg_write(gspca_dev->dev, 0x00, 0x850a, 0x0001);
+
+	err = spca50x_setup_qtable(gspca_dev, 0x00, 0x8800, 0x8840,
+				 qtable_pocketdv);
+	if (err < 0)
+		PDEBUG(D_ERR|D_STREAM, "spca50x_setup_qtable failed on init");
+
+	/* set qtable index */
+	reg_write(gspca_dev->dev, 0x00, 0x8880, 2);
+	/* family cam Quicksmart stuff */
+	reg_write(gspca_dev->dev, 0x00, 0x800a, 0x00);
+	/* Set agc transfer: synced inbetween frames */
+	reg_write(gspca_dev->dev, 0x00, 0x820f, 0x01);
+	/* Init SDRAM - needed for SDRAM access */
+	reg_write(gspca_dev->dev, 0x00, 0x870a, 0x04);
+	/*Start init sequence or stream */
+
+	reg_write(gspca_dev->dev, 0, 0x8003, 0x00);
+	/* switch to video camera mode */
+	reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+	msleep(2000);
+	if (reg_readwait(gspca_dev->dev, 0, 0x8000, 0x44) != 0)
+		spca5xxRegRead(gspca_dev->dev, 0x816b, &Data, 1);
+	reg_write(gspca_dev->dev, 0x00, 0x816b, Data);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+	__u16 vendor;
+	__u16 product;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+	switch (vendor) {
+	case 0x040a:		/* Kodak cameras */
+/*		switch (product) { */
+/*		case 0x0300: */
+			sd->subtype = KodakEZ200;
+/*			break; */
+/*		} */
+		break;
+	case 0x041e:		/* Creative cameras */
+/*		switch (product) { */
+/*		case 0x400a: */
+			sd->subtype = CreativePCCam300;
+/*			break; */
+/*		} */
+		break;
+	case 0x046d:		/* Logitech Labtec */
+		switch (product) {
+		case 0x0890:
+			sd->subtype = LogitechTraveler;
+			break;
+		case 0x0900:
+			sd->subtype = LogitechClickSmart310;
+			break;
+		case 0x0901:
+			sd->subtype = LogitechClickSmart510;
+			break;
+		}
+		break;
+	case 0x04a5:		/* Benq */
+/*		switch (product) { */
+/*		case 0x300c: */
+			sd->subtype = BenqDC1016;
+/*			break; */
+/*		} */
+		break;
+	case 0x04fc:		/* SunPlus */
+/*		switch (product) { */
+/*		case 0x7333: */
+			sd->subtype = PalmPixDC85;
+/*			break; */
+/*		} */
+		break;
+	case 0x055f:		/* Mustek cameras */
+		switch (product) {
+		case 0xc200:
+			sd->subtype = MustekGsmart300;
+			break;
+		case 0xc220:
+			sd->subtype = Gsmartmini;
+			break;
+		}
+		break;
+	case 0x06bd:		/* Agfa Cl20 */
+/*		switch (product) { */
+/*		case 0x0404: */
+			sd->subtype = AgfaCl20;
+/*			break; */
+/*		} */
+		break;
+	case 0x06be:		/* Optimedia */
+/*		switch (product) { */
+/*		case 0x0800: */
+			sd->subtype = Optimedia;
+/*			break; */
+/*		} */
+		break;
+	case 0x084d:		/* D-Link / Minton */
+/*		switch (product) { */
+/*		case 0x0003:	 * DSC-350 / S-Cam F5 */
+			sd->subtype = DLinkDSC350;
+/*			break; */
+/*		} */
+		break;
+	case 0x08ca:		/* Aiptek */
+/*		switch (product) { */
+/*		case 0x0103: */
+			sd->subtype = AiptekPocketDV;
+/*			break; */
+/*		} */
+		break;
+	case 0x2899:		/* ToptroIndustrial */
+/*		switch (product) { */
+/*		case 0x012c: */
+			sd->subtype = ToptroIndus;
+/*			break; */
+/*		} */
+		break;
+	case 0x8086:		/* Intel */
+/*		switch (product) { */
+/*		case 0x0630:	 * Pocket PC Camera */
+			sd->subtype = IntelPocketPCCamera;
+/*			break; */
+/*		} */
+		break;
+	}
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	if (sd->subtype != LogitechClickSmart310) {
+		cam->cam_mode = vga_mode;
+		cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	} else {
+		cam->cam_mode = sif_mode;
+		cam->nmodes = sizeof sif_mode / sizeof sif_mode[0];
+	}
+	sd->qindex = 5;
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	/* initialisation of spca500 based cameras is deferred */
+	PDEBUG(D_STREAM, "SPCA500 init");
+	if (sd->subtype == LogitechClickSmart310)
+		spca500_clksmart310_init(gspca_dev);
+/*	else
+		spca500_initialise(gspca_dev); */
+	PDEBUG(D_STREAM, "SPCA500 init done");
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int err;
+	__u8 Data;
+	__u8 xmult, ymult;
+
+	if (sd->subtype == LogitechClickSmart310) {
+		xmult = 0x16;
+		ymult = 0x12;
+	} else {
+		xmult = 0x28;
+		ymult = 0x1e;
+	}
+
+	/* is there a sensor here ? */
+	spca5xxRegRead(gspca_dev->dev, 0x8a04, &Data, 1);
+	PDEBUG(D_STREAM, "Spca500 Sensor Address 0x%02X", Data);
+	PDEBUG(D_STREAM, "Spca500 curr_mode: %d Xmult: 0x%02X, Ymult: 0x%02X",
+		gspca_dev->curr_mode, xmult, ymult);
+
+	/* setup qtable */
+	switch (sd->subtype) {
+	case LogitechClickSmart310:
+		 spca500_setmode(gspca_dev, xmult, ymult);
+
+		/* enable drop packet */
+		reg_write(gspca_dev->dev, 0x00, 0x850a, 0x0001);
+		reg_write(gspca_dev->dev, 0x00, 0x8880, 3);
+		err = spca50x_setup_qtable(gspca_dev,
+					   0x00, 0x8800, 0x8840,
+					   qtable_creative_pccam);
+		if (err < 0)
+			PDEBUG(D_ERR, "spca50x_setup_qtable failed");
+		/* Init SDRAM - needed for SDRAM access */
+		reg_write(gspca_dev->dev, 0x00, 0x870a, 0x04);
+
+		/* switch to video camera mode */
+		reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+		msleep(500);
+		if (reg_readwait(gspca_dev->dev, 0, 0x8000, 0x44) != 0)
+			PDEBUG(D_ERR, "reg_readwait() failed");
+
+		spca5xxRegRead(gspca_dev->dev, 0x816b, &Data, 1);
+		reg_write(gspca_dev->dev, 0x00, 0x816b, Data);
+
+		spca500_synch310(gspca_dev);
+
+		write_vector(gspca_dev, spca500_visual_defaults);
+		spca500_setmode(gspca_dev, xmult, ymult);
+		/* enable drop packet */
+		reg_write(gspca_dev->dev, 0x00, 0x850a, 0x0001);
+			PDEBUG(D_ERR, "failed to enable drop packet");
+		reg_write(gspca_dev->dev, 0x00, 0x8880, 3);
+		err = spca50x_setup_qtable(gspca_dev,
+					   0x00, 0x8800, 0x8840,
+					   qtable_creative_pccam);
+		if (err < 0)
+			PDEBUG(D_ERR, "spca50x_setup_qtable failed");
+
+		/* Init SDRAM - needed for SDRAM access */
+		reg_write(gspca_dev->dev, 0x00, 0x870a, 0x04);
+
+		/* switch to video camera mode */
+		reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+
+		if (reg_readwait(gspca_dev->dev, 0, 0x8000, 0x44) != 0)
+			PDEBUG(D_ERR, "reg_readwait() failed");
+
+		spca5xxRegRead(gspca_dev->dev, 0x816b, &Data, 1);
+		reg_write(gspca_dev->dev, 0x00, 0x816b, Data);
+		break;
+	case CreativePCCam300:		/* Creative PC-CAM 300 640x480 CCD */
+	case IntelPocketPCCamera:	/* FIXME: Temporary fix for
+					 *	Intel Pocket PC Camera
+					 *	- NWG (Sat 29th March 2003) */
+
+		/* do a full reset */
+		if ((err = spca500_full_reset(gspca_dev)) < 0)
+			PDEBUG(D_ERR, "spca500_full_reset failed");
+
+		/* enable drop packet */
+		err = reg_write(gspca_dev->dev, 0x00, 0x850a, 0x0001);
+		if (err < 0)
+			PDEBUG(D_ERR, "failed to enable drop packet");
+		reg_write(gspca_dev->dev, 0x00, 0x8880, 3);
+		err = spca50x_setup_qtable(gspca_dev,
+					   0x00, 0x8800, 0x8840,
+					   qtable_creative_pccam);
+		if (err < 0)
+			PDEBUG(D_ERR, "spca50x_setup_qtable failed");
+
+		spca500_setmode(gspca_dev, xmult, ymult);
+		reg_write(gspca_dev->dev, 0x20, 0x0001, 0x0004);
+
+		/* switch to video camera mode */
+		reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+
+		if (reg_readwait(gspca_dev->dev, 0, 0x8000, 0x44) != 0)
+			PDEBUG(D_ERR, "reg_readwait() failed");
+
+		spca5xxRegRead(gspca_dev->dev, 0x816b, &Data, 1);
+		reg_write(gspca_dev->dev, 0x00, 0x816b, Data);
+
+		/* write_vector(gspca_dev, spca500_visual_defaults); */
+		break;
+	case KodakEZ200:		/* Kodak EZ200 */
+
+		/* do a full reset */
+		err = spca500_full_reset(gspca_dev);
+		if (err < 0)
+			PDEBUG(D_ERR, "spca500_full_reset failed");
+		/* enable drop packet */
+		reg_write(gspca_dev->dev, 0x00, 0x850a, 0x0001);
+		reg_write(gspca_dev->dev, 0x00, 0x8880, 0);
+		err = spca50x_setup_qtable(gspca_dev,
+					   0x00, 0x8800, 0x8840,
+					   qtable_kodak_ez200);
+		if (err < 0)
+			PDEBUG(D_ERR, "spca50x_setup_qtable failed");
+		spca500_setmode(gspca_dev, xmult, ymult);
+
+		reg_write(gspca_dev->dev, 0x20, 0x0001, 0x0004);
+
+		/* switch to video camera mode */
+		reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+
+		if (reg_readwait(gspca_dev->dev, 0, 0x8000, 0x44) != 0)
+			PDEBUG(D_ERR, "reg_readwait() failed");
+
+		spca5xxRegRead(gspca_dev->dev, 0x816b, &Data, 1);
+		reg_write(gspca_dev->dev, 0x00, 0x816b, Data);
+
+		/* write_vector(gspca_dev, spca500_visual_defaults); */
+		break;
+
+	case BenqDC1016:
+	case DLinkDSC350:		/* FamilyCam 300 */
+	case AiptekPocketDV:		/* Aiptek PocketDV */
+	case Gsmartmini:		/*Mustek Gsmart Mini */
+	case MustekGsmart300:		/* Mustek Gsmart 300 */
+	case PalmPixDC85:
+	case Optimedia:
+	case ToptroIndus:
+	case AgfaCl20:
+		spca500_reinit(gspca_dev);
+		reg_write(gspca_dev->dev, 0x00, 0x0d01, 0x01);
+		/* enable drop packet */
+		reg_write(gspca_dev->dev, 0x00, 0x850a, 0x0001);
+
+		err = spca50x_setup_qtable(gspca_dev,
+				   0x00, 0x8800, 0x8840, qtable_pocketdv);
+		if (err < 0)
+			PDEBUG(D_ERR, "spca50x_setup_qtable failed");
+		reg_write(gspca_dev->dev, 0x00, 0x8880, 2);
+
+		/* familycam Quicksmart pocketDV stuff */
+		reg_write(gspca_dev->dev, 0x00, 0x800a, 0x00);
+		/* Set agc transfer: synced inbetween frames */
+		reg_write(gspca_dev->dev, 0x00, 0x820f, 0x01);
+		/* Init SDRAM - needed for SDRAM access */
+		reg_write(gspca_dev->dev, 0x00, 0x870a, 0x04);
+
+		spca500_setmode(gspca_dev,xmult,ymult);
+		/* switch to video camera mode */
+		reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+
+		reg_readwait(gspca_dev->dev, 0, 0x8000, 0x44);
+
+		spca5xxRegRead(gspca_dev->dev, 0x816b, &Data, 1);
+		reg_write(gspca_dev->dev, 0x00, 0x816b, Data);
+		break;
+	case LogitechTraveler:
+	case LogitechClickSmart510:
+		reg_write(gspca_dev->dev, 0x02, 0x00, 0x00);
+		/* enable drop packet */
+		reg_write(gspca_dev->dev, 0x00, 0x850a, 0x0001);
+
+		err = spca50x_setup_qtable(gspca_dev,
+					0x00, 0x8800,
+					0x8840, qtable_creative_pccam);
+		if (err < 0)
+			PDEBUG(D_ERR, "spca50x_setup_qtable failed");
+		reg_write(gspca_dev->dev, 0x00, 0x8880, 3);
+		reg_write(gspca_dev->dev, 0x00, 0x800a, 0x00);
+		/* Init SDRAM - needed for SDRAM access */
+		reg_write(gspca_dev->dev, 0x00, 0x870a, 0x04);
+
+		spca500_setmode(gspca_dev, xmult, ymult);
+
+		/* switch to video camera mode */
+		reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+		reg_readwait(gspca_dev->dev, 0, 0x8000, 0x44);
+
+		spca5xxRegRead(gspca_dev->dev, 0x816b, &Data, 1);
+		reg_write(gspca_dev->dev, 0x00, 0x816b, Data);
+		write_vector(gspca_dev, Clicksmart510_defaults);
+		break;
+	}
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	__u8 data = 0;
+
+	reg_write(gspca_dev->dev, 0, 0x8003, 0x00);
+
+	/* switch to video camera mode */
+	reg_write(gspca_dev->dev, 0x00, 0x8000, 0x0004);
+	spca5xxRegRead(gspca_dev->dev, 0x8000, &data, 1);
+	PDEBUG(D_STREAM, "stop SPCA500 done reg8000: 0x%2x", data);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int i;
+	unsigned char *s, *d;
+	static unsigned char ffd9[] = {0xff, 0xd9};
+
+/* frames are jpeg 4.1.1 without 0xff escape */
+	if (data[0] == 0xff) {
+		if (data[1] != 0x01) {	/* drop packet */
+/*			gspca_dev->last_packet_type = DISCARD_PACKET; */
+			return;
+		}
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					ffd9, 2);
+
+		/* put the JPEG header in the new frame */
+		jpeg_put_header(gspca_dev, frame,
+				((struct sd *) gspca_dev)->qindex,
+				0x22);
+
+		data += SPCA500_OFFSET_DATA;
+		len -= SPCA500_OFFSET_DATA;
+	} else {
+		data += 1;
+		len -= 1;
+	}
+
+	/* add 0x00 after 0xff */
+	for (i = len; --i >= 0; )
+		if (data[i] == 0xff)
+			break;
+	if (i < 0) {			/* no 0xff */
+		gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+		return;
+	}
+	s = data;
+	d = sd->packet;
+	for (i = 0; i < len; i++) {
+		*d++ = *s++;
+		if (s[-1] == 0xff)
+			*d++ = 0x00;
+	}
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame,
+			sd->packet, d - sd->packet);
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	reg_write(gspca_dev->dev, 0x00, 0x8167,
+			(__u8) (sd->brightness - 128));
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = reg_read(gspca_dev->dev, 0x00, 0x8167, 1) + 128;
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	reg_write(gspca_dev->dev, 0x00, 0x8168, sd->contrast >> 2);
+}
+
+static void getcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = reg_read(gspca_dev->dev, 0x0, 0x8168, 1) << 2;
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	reg_write(gspca_dev->dev, 0x00, 0x8169, sd->colors >> 2);
+}
+
+static void getcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = reg_read(gspca_dev->dev, 0x0, 0x8169, 1) << 2;
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcontrast(gspca_dev);
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcolors(gspca_dev);
+	*val = sd->colors;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = sizeof sd_ctrls / sizeof sd_ctrls[0],
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x040a, 0x0300), DVNM("Kodak EZ200")},
+	{USB_DEVICE(0x041e, 0x400a), DVNM("Creative PC-CAM 300")},
+	{USB_DEVICE(0x046d, 0x0890), DVNM("Logitech QuickCam traveler")},
+	{USB_DEVICE(0x046d, 0x0900), DVNM("Logitech Inc. ClickSmart 310")},
+	{USB_DEVICE(0x046d, 0x0901), DVNM("Logitech Inc. ClickSmart 510")},
+	{USB_DEVICE(0x04a5, 0x300c), DVNM("Benq DC1016")},
+	{USB_DEVICE(0x04fc, 0x7333), DVNM("PalmPixDC85")},
+	{USB_DEVICE(0x055f, 0xc200), DVNM("Mustek Gsmart 300")},
+	{USB_DEVICE(0x055f, 0xc220), DVNM("Gsmart Mini")},
+	{USB_DEVICE(0x06bd, 0x0404), DVNM("Agfa CL20")},
+	{USB_DEVICE(0x06be, 0x0800), DVNM("Optimedia")},
+	{USB_DEVICE(0x084d, 0x0003), DVNM("D-Link DSC-350")},
+	{USB_DEVICE(0x08ca, 0x0103), DVNM("Aiptek PocketDV")},
+	{USB_DEVICE(0x2899, 0x012c), DVNM("Toptro Industrial")},
+	{USB_DEVICE(0x8086, 0x0630), DVNM("Intel Pocket PC Camera")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/spca501.c b/drivers/media/video/gspca/spca501.c
new file mode 100644
index 000000000000..c6468cf3506a
--- /dev/null
+++ b/drivers/media/video/gspca/spca501.c
@@ -0,0 +1,2219 @@
+/*
+ * SPCA501 chip based cameras initialization data
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#define MODULE_NAME "spca501"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SPCA501 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned short contrast;
+	__u8 brightness;
+	__u8 colors;
+
+	char subtype;
+#define Arowana300KCMOSCamera 0
+#define IntelCreateAndShare 1
+#define KodakDVC325 2
+#define MystFromOriUnknownCamera 3
+#define SmileIntlCamera 4
+#define ThreeComHomeConnectLite 5
+#define ViewQuestM318B 6
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define MY_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 127,
+		.step    = 1,
+		.default_value = 63,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define MY_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 0xffff,
+		.step    = 1,
+		.default_value = 0xaa00,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define MY_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Color",
+		.minimum = 0,
+		.maximum = 63,
+		.step    = 1,
+		.default_value = 31,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_SPCA501, 160, 120, 2},
+	{V4L2_PIX_FMT_SPCA501, 320, 240, 1},
+	{V4L2_PIX_FMT_SPCA501, 640, 480, 0},
+};
+
+#define SPCA50X_REG_USB 0x2	/* spca505 501 */
+/*
+ * Data to initialize a SPCA501. From a capture file provided by Bill Roehl
+ * With SPCA501 chip description
+ */
+#define CCDSP_SET		/* set CCDSP parameters */
+#define TG_SET			/* set time generator set */
+#undef DSPWIN_SET		/* set DSP windows parameters */
+#undef ALTER_GAMA	/* Set alternate set to YUV transform coeffs. */
+#define SPCA501_SNAPBIT 0x80
+#define SPCA501_SNAPCTRL 0x10
+/* Frame packet header offsets for the spca501 */
+#define SPCA501_OFFSET_GPIO   1
+#define SPCA501_OFFSET_TYPE   2
+#define SPCA501_OFFSET_TURN3A 3
+#define SPCA501_OFFSET_FRAMSEQ 4
+#define SPCA501_OFFSET_COMPRESS 5
+#define SPCA501_OFFSET_QUANT 6
+#define SPCA501_OFFSET_QUANT2 7
+#define SPCA501_OFFSET_DATA 8
+
+#define SPCA501_PROP_COMP_ENABLE(d) ((d) & 1)
+#define SPCA501_PROP_SNAP(d) ((d) & 0x40)
+#define SPCA501_PROP_SNAP_CTRL(d) ((d) & 0x10)
+#define SPCA501_PROP_COMP_THRESH(d) (((d) & 0x0e) >> 1)
+#define SPCA501_PROP_COMP_QUANT(d) (((d) & 0x70) >> 4)
+
+/* SPCA501 CCDSP control */
+#define SPCA501_REG_CCDSP 0x01
+/* SPCA501 control/status registers */
+#define SPCA501_REG_CTLRL 0x02
+
+/* registers for color correction and YUV transformation */
+#define SPCA501_A11 0x08
+#define SPCA501_A12 0x09
+#define SPCA501_A13 0x0A
+#define SPCA501_A21 0x0B
+#define SPCA501_A22 0x0C
+#define SPCA501_A23 0x0D
+#define SPCA501_A31 0x0E
+#define SPCA501_A32 0x0F
+#define SPCA501_A33 0x10
+
+/* Data for video camera initialization before capturing */
+static __u16 spca501_open_data[][3] = {
+	/* bmRequest,value,index */
+
+	{0x2, 0x50, 0x00},	/* C/S enable soft reset */
+	{0x2, 0x40, 0x00},	/* C/S disable soft reset */
+	{0x2, 0x02, 0x05},	/* C/S general purpose I/O data */
+	{0x2, 0x03, 0x05},	/* C/S general purpose I/O data */
+
+#ifdef CCDSP_SET
+	{0x1, 0x38, 0x01},	/* CCDSP options */
+	{0x1, 0x05, 0x02}, /* CCDSP Optical black level for user settings */
+	{0x1, 0xC0, 0x03},	/* CCDSP Optical black settings */
+
+	{0x1, 0x67, 0x07},
+	{0x1, 0x63, 0x3f},	/* CCDSP CCD gamma enable */
+	{0x1, 0x03, 0x56},	/* Add gamma correction */
+
+	{0x1, 0xFF, 0x15},	/* CCDSP High luminance for white balance */
+	{0x1, 0x01, 0x16},	/* CCDSP Low luminance for white balance */
+
+/* Color correction and RGB-to-YUV transformation coefficients changing */
+#ifdef ALTER_GAMA
+	{0x0, 0x00, 0x08},	/* A11 */
+	{0x0, 0x00, 0x09},	/* A12 */
+	{0x0, 0x90, 0x0A},	/* A13 */
+	{0x0, 0x12, 0x0B},	/* A21 */
+	{0x0, 0x00, 0x0C},	/* A22 */
+	{0x0, 0x00, 0x0D},	/* A23 */
+	{0x0, 0x00, 0x0E},	/* A31 */
+	{0x0, 0x02, 0x0F},	/* A32 */
+	{0x0, 0x00, 0x10},	/* A33 */
+#else
+	{0x1, 0x2a, 0x08},	/* A11 0x31 */
+	{0x1, 0xf8, 0x09},	/* A12 f8 */
+	{0x1, 0xf8, 0x0A},	/* A13 f8 */
+	{0x1, 0xf8, 0x0B},	/* A21 f8 */
+	{0x1, 0x14, 0x0C},	/* A22 0x14 */
+	{0x1, 0xf8, 0x0D},	/* A23 f8 */
+	{0x1, 0xf8, 0x0E},	/* A31 f8 */
+	{0x1, 0xf8, 0x0F},	/* A32 f8 */
+	{0x1, 0x20, 0x10},	/* A33 0x20 */
+#endif
+	{0x1, 0x00, 0x11},	/* R offset */
+	{0x1, 0x00, 0x12},	/* G offset */
+	{0x1, 0x00, 0x13},	/* B offset */
+	{0x1, 0x00, 0x14},	/* GB offset */
+
+#endif
+
+#ifdef TG_SET
+	/* Time generator manipulations */
+	{0x0, 0xfc, 0x0},	/* Set up high bits of shutter speed */
+	{0x0, 0x01, 0x1},	/* Set up low bits of shutter speed */
+
+	{0x0, 0xe4, 0x04},	/* DCLK*2 clock phase adjustment */
+	{0x0, 0x08, 0x05},	/* ADCK phase adjustment, inv. ext. VB */
+	{0x0, 0x03, 0x06},	/* FR phase adjustment */
+	{0x0, 0x01, 0x07},	/* FCDS phase adjustment */
+	{0x0, 0x39, 0x08},	/* FS phase adjustment */
+	{0x0, 0x88, 0x0a},	/* FH1 phase and delay adjustment */
+	{0x0, 0x03, 0x0f},	/* pixel identification */
+	{0x0, 0x00, 0x11},	/* clock source selection (default) */
+
+	/*VERY strange manipulations with
+	 * select DMCLP or OBPX to be ADCLP output (0x0C)
+	 * OPB always toggle or not (0x0D) but they allow
+	 * us to set up brightness
+	 */
+	{0x0, 0x01, 0x0c},
+	{0x0, 0xe0, 0x0d},
+	/* Done */
+#endif
+
+#ifdef DSPWIN_SET
+	{0x1, 0xa0, 0x01},	/* Setting image processing parameters */
+	{0x1, 0x1c, 0x17},	/* Changing Windows positions X1 */
+	{0x1, 0xe2, 0x19},	/* X2 */
+	{0x1, 0x1c, 0x1b},	/* X3 */
+	{0x1, 0xe2, 0x1d},	/* X4 */
+	{0x1, 0x5f, 0x1f},	/* X5 */
+	{0x1, 0x32, 0x20},	/* Y5 */
+	{0x1, 0x01, 0x10},	/* Changing A33 */
+#endif
+
+	{0x2, 0x204a, 0x07},/* Setting video compression & resolution 160x120 */
+	{0x2, 0x94, 0x06},	/* Setting video no compression */
+	{}
+};
+
+/*
+   The SPCAxxx docs from Sunplus document these values
+   in tables, one table per register number.  In the data
+   below, dmRequest is the register number, index is the Addr,
+   and value is a combination of Bit values.
+   Bit  Value (hex)
+   0    01
+   1    02
+   2    04
+   3    08
+   4    10
+   5    20
+   6    40
+   7    80
+ */
+
+/* Data for chip initialization (set default values) */
+static __u16 spca501_init_data[][3] = {
+	/* Set all the values to powerup defaults */
+	/* bmRequest,value,index */
+	{0x0, 0xAA, 0x00},
+	{0x0, 0x02, 0x01},
+	{0x0, 0x01, 0x02},
+	{0x0, 0x02, 0x03},
+	{0x0, 0xCE, 0x04},
+	{0x0, 0x00, 0x05},
+	{0x0, 0x00, 0x06},
+	{0x0, 0x00, 0x07},
+	{0x0, 0x00, 0x08},
+	{0x0, 0x00, 0x09},
+	{0x0, 0x90, 0x0A},
+	{0x0, 0x12, 0x0B},
+	{0x0, 0x00, 0x0C},
+	{0x0, 0x00, 0x0D},
+	{0x0, 0x00, 0x0E},
+	{0x0, 0x02, 0x0F},
+	{0x0, 0x00, 0x10},
+	{0x0, 0x00, 0x11},
+	{0x0, 0x00, 0x12},
+	{0x0, 0x00, 0x13},
+	{0x0, 0x00, 0x14},
+	{0x0, 0x00, 0x15},
+	{0x0, 0x00, 0x16},
+	{0x0, 0x00, 0x17},
+	{0x0, 0x00, 0x18},
+	{0x0, 0x00, 0x19},
+	{0x0, 0x00, 0x1A},
+	{0x0, 0x00, 0x1B},
+	{0x0, 0x00, 0x1C},
+	{0x0, 0x00, 0x1D},
+	{0x0, 0x00, 0x1E},
+	{0x0, 0x00, 0x1F},
+	{0x0, 0x00, 0x20},
+	{0x0, 0x00, 0x21},
+	{0x0, 0x00, 0x22},
+	{0x0, 0x00, 0x23},
+	{0x0, 0x00, 0x24},
+	{0x0, 0x00, 0x25},
+	{0x0, 0x00, 0x26},
+	{0x0, 0x00, 0x27},
+	{0x0, 0x00, 0x28},
+	{0x0, 0x00, 0x29},
+	{0x0, 0x00, 0x2A},
+	{0x0, 0x00, 0x2B},
+	{0x0, 0x00, 0x2C},
+	{0x0, 0x00, 0x2D},
+	{0x0, 0x00, 0x2E},
+	{0x0, 0x00, 0x2F},
+	{0x0, 0x00, 0x30},
+	{0x0, 0x00, 0x31},
+	{0x0, 0x00, 0x32},
+	{0x0, 0x00, 0x33},
+	{0x0, 0x00, 0x34},
+	{0x0, 0x00, 0x35},
+	{0x0, 0x00, 0x36},
+	{0x0, 0x00, 0x37},
+	{0x0, 0x00, 0x38},
+	{0x0, 0x00, 0x39},
+	{0x0, 0x00, 0x3A},
+	{0x0, 0x00, 0x3B},
+	{0x0, 0x00, 0x3C},
+	{0x0, 0x00, 0x3D},
+	{0x0, 0x00, 0x3E},
+	{0x0, 0x00, 0x3F},
+	{0x0, 0x00, 0x40},
+	{0x0, 0x00, 0x41},
+	{0x0, 0x00, 0x42},
+	{0x0, 0x00, 0x43},
+	{0x0, 0x00, 0x44},
+	{0x0, 0x00, 0x45},
+	{0x0, 0x00, 0x46},
+	{0x0, 0x00, 0x47},
+	{0x0, 0x00, 0x48},
+	{0x0, 0x00, 0x49},
+	{0x0, 0x00, 0x4A},
+	{0x0, 0x00, 0x4B},
+	{0x0, 0x00, 0x4C},
+	{0x0, 0x00, 0x4D},
+	{0x0, 0x00, 0x4E},
+	{0x0, 0x00, 0x4F},
+	{0x0, 0x00, 0x50},
+	{0x0, 0x00, 0x51},
+	{0x0, 0x00, 0x52},
+	{0x0, 0x00, 0x53},
+	{0x0, 0x00, 0x54},
+	{0x0, 0x00, 0x55},
+	{0x0, 0x00, 0x56},
+	{0x0, 0x00, 0x57},
+	{0x0, 0x00, 0x58},
+	{0x0, 0x00, 0x59},
+	{0x0, 0x00, 0x5A},
+	{0x0, 0x00, 0x5B},
+	{0x0, 0x00, 0x5C},
+	{0x0, 0x00, 0x5D},
+	{0x0, 0x00, 0x5E},
+	{0x0, 0x00, 0x5F},
+	{0x0, 0x00, 0x60},
+	{0x0, 0x00, 0x61},
+	{0x0, 0x00, 0x62},
+	{0x0, 0x00, 0x63},
+	{0x0, 0x00, 0x64},
+	{0x0, 0x00, 0x65},
+	{0x0, 0x00, 0x66},
+	{0x0, 0x00, 0x67},
+	{0x0, 0x00, 0x68},
+	{0x0, 0x00, 0x69},
+	{0x0, 0x00, 0x6A},
+	{0x0, 0x00, 0x6B},
+	{0x0, 0x00, 0x6C},
+	{0x0, 0x00, 0x6D},
+	{0x0, 0x00, 0x6E},
+	{0x0, 0x00, 0x6F},
+	{0x0, 0x00, 0x70},
+	{0x0, 0x00, 0x71},
+	{0x0, 0x00, 0x72},
+	{0x0, 0x00, 0x73},
+	{0x0, 0x00, 0x74},
+	{0x0, 0x00, 0x75},
+	{0x0, 0x00, 0x76},
+	{0x0, 0x00, 0x77},
+	{0x0, 0x00, 0x78},
+	{0x0, 0x00, 0x79},
+	{0x0, 0x00, 0x7A},
+	{0x0, 0x00, 0x7B},
+	{0x0, 0x00, 0x7C},
+	{0x0, 0x00, 0x7D},
+	{0x0, 0x00, 0x7E},
+	{0x0, 0x00, 0x7F},
+	{0x0, 0x00, 0x80},
+	{0x0, 0x00, 0x81},
+	{0x0, 0x00, 0x82},
+	{0x0, 0x00, 0x83},
+	{0x0, 0x00, 0x84},
+	{0x0, 0x00, 0x85},
+	{0x0, 0x00, 0x86},
+	{0x0, 0x00, 0x87},
+	{0x0, 0x00, 0x88},
+	{0x0, 0x00, 0x89},
+	{0x0, 0x00, 0x8A},
+	{0x0, 0x00, 0x8B},
+	{0x0, 0x00, 0x8C},
+	{0x0, 0x00, 0x8D},
+	{0x0, 0x00, 0x8E},
+	{0x0, 0x00, 0x8F},
+	{0x0, 0x00, 0x90},
+	{0x0, 0x00, 0x91},
+	{0x0, 0x00, 0x92},
+	{0x0, 0x00, 0x93},
+	{0x0, 0x00, 0x94},
+	{0x0, 0x00, 0x95},
+	{0x0, 0x00, 0x96},
+	{0x0, 0x00, 0x97},
+	{0x0, 0x00, 0x98},
+	{0x0, 0x00, 0x99},
+	{0x0, 0x00, 0x9A},
+	{0x0, 0x00, 0x9B},
+	{0x0, 0x00, 0x9C},
+	{0x0, 0x00, 0x9D},
+	{0x0, 0x00, 0x9E},
+	{0x0, 0x00, 0x9F},
+	{0x0, 0x00, 0xA0},
+	{0x0, 0x00, 0xA1},
+	{0x0, 0x00, 0xA2},
+	{0x0, 0x00, 0xA3},
+	{0x0, 0x00, 0xA4},
+	{0x0, 0x00, 0xA5},
+	{0x0, 0x00, 0xA6},
+	{0x0, 0x00, 0xA7},
+	{0x0, 0x00, 0xA8},
+	{0x0, 0x00, 0xA9},
+	{0x0, 0x00, 0xAA},
+	{0x0, 0x00, 0xAB},
+	{0x0, 0x00, 0xAC},
+	{0x0, 0x00, 0xAD},
+	{0x0, 0x00, 0xAE},
+	{0x0, 0x00, 0xAF},
+	{0x0, 0x00, 0xB0},
+	{0x0, 0x00, 0xB1},
+	{0x0, 0x00, 0xB2},
+	{0x0, 0x00, 0xB3},
+	{0x0, 0x00, 0xB4},
+	{0x0, 0x00, 0xB5},
+	{0x0, 0x00, 0xB6},
+	{0x0, 0x00, 0xB7},
+	{0x0, 0x00, 0xB8},
+	{0x0, 0x00, 0xB9},
+	{0x0, 0x00, 0xBA},
+	{0x0, 0x00, 0xBB},
+	{0x0, 0x00, 0xBC},
+	{0x0, 0x00, 0xBD},
+	{0x0, 0x00, 0xBE},
+	{0x0, 0x00, 0xBF},
+	{0x0, 0x00, 0xC0},
+	{0x0, 0x00, 0xC1},
+	{0x0, 0x00, 0xC2},
+	{0x0, 0x00, 0xC3},
+	{0x0, 0x00, 0xC4},
+	{0x0, 0x00, 0xC5},
+	{0x0, 0x00, 0xC6},
+	{0x0, 0x00, 0xC7},
+	{0x0, 0x00, 0xC8},
+	{0x0, 0x00, 0xC9},
+	{0x0, 0x00, 0xCA},
+	{0x0, 0x00, 0xCB},
+	{0x0, 0x00, 0xCC},
+	{0x1, 0xF4, 0x00},
+	{0x1, 0x38, 0x01},
+	{0x1, 0x40, 0x02},
+	{0x1, 0x0A, 0x03},
+	{0x1, 0x40, 0x04},
+	{0x1, 0x40, 0x05},
+	{0x1, 0x40, 0x06},
+	{0x1, 0x67, 0x07},
+	{0x1, 0x31, 0x08},
+	{0x1, 0x00, 0x09},
+	{0x1, 0x00, 0x0A},
+	{0x1, 0x00, 0x0B},
+	{0x1, 0x14, 0x0C},
+	{0x1, 0x00, 0x0D},
+	{0x1, 0x00, 0x0E},
+	{0x1, 0x00, 0x0F},
+	{0x1, 0x1E, 0x10},
+	{0x1, 0x00, 0x11},
+	{0x1, 0x00, 0x12},
+	{0x1, 0x00, 0x13},
+	{0x1, 0x00, 0x14},
+	{0x1, 0xFF, 0x15},
+	{0x1, 0x01, 0x16},
+	{0x1, 0x32, 0x17},
+	{0x1, 0x23, 0x18},
+	{0x1, 0xCE, 0x19},
+	{0x1, 0x23, 0x1A},
+	{0x1, 0x32, 0x1B},
+	{0x1, 0x8D, 0x1C},
+	{0x1, 0xCE, 0x1D},
+	{0x1, 0x8D, 0x1E},
+	{0x1, 0x00, 0x1F},
+	{0x1, 0x00, 0x20},
+	{0x1, 0xFF, 0x3E},
+	{0x1, 0x02, 0x3F},
+	{0x1, 0x00, 0x40},
+	{0x1, 0x00, 0x41},
+	{0x1, 0x00, 0x42},
+	{0x1, 0x00, 0x43},
+	{0x1, 0x00, 0x44},
+	{0x1, 0x00, 0x45},
+	{0x1, 0x00, 0x46},
+	{0x1, 0x00, 0x47},
+	{0x1, 0x00, 0x48},
+	{0x1, 0x00, 0x49},
+	{0x1, 0x00, 0x4A},
+	{0x1, 0x00, 0x4B},
+	{0x1, 0x00, 0x4C},
+	{0x1, 0x00, 0x4D},
+	{0x1, 0x00, 0x4E},
+	{0x1, 0x00, 0x4F},
+	{0x1, 0x00, 0x50},
+	{0x1, 0x00, 0x51},
+	{0x1, 0x00, 0x52},
+	{0x1, 0x00, 0x53},
+	{0x1, 0x00, 0x54},
+	{0x1, 0x00, 0x55},
+	{0x1, 0x00, 0x56},
+	{0x1, 0x00, 0x57},
+	{0x1, 0x00, 0x58},
+	{0x1, 0x00, 0x59},
+	{0x1, 0x00, 0x5A},
+	{0x2, 0x03, 0x00},
+	{0x2, 0x00, 0x01},
+	{0x2, 0x00, 0x05},
+	{0x2, 0x00, 0x06},
+	{0x2, 0x00, 0x07},
+	{0x2, 0x00, 0x10},
+	{0x2, 0x00, 0x11},
+	/* Strange - looks like the 501 driver doesn't do anything
+	 * at insert time except read the EEPROM
+	 */
+	{}
+};
+
+/* Data for video camera init before capture.
+ * Capture and decoding by Colin Peart.
+ * This is is for the 3com HomeConnect Lite which is spca501a based.
+ */
+static __u16 spca501_3com_open_data[][3] = {
+	/* bmRequest,value,index */
+	{0x2, 0x0050, 0x0000},	/* C/S Enable TG soft reset, timing mode=010 */
+	{0x2, 0x0043, 0x0000},	/* C/S Disable TG soft reset, timing mode=010 */
+	{0x2, 0x0002, 0x0005},	/* C/S GPIO */
+	{0x2, 0x0003, 0x0005},	/* C/S GPIO */
+
+#ifdef CCDSP_SET
+	{0x1, 0x0020, 0x0001},	/* CCDSP Options */
+
+	{0x1, 0x0020, 0x0002},	/* CCDSP Black Level */
+	{0x1, 0x006e, 0x0007},	/* CCDSP Gamma options */
+	{0x1, 0x0090, 0x0015},	/* CCDSP Luminance Low */
+	{0x1, 0x00ff, 0x0016},	/* CCDSP Luminance High */
+	{0x1, 0x0003, 0x003F},	/* CCDSP Gamma correction toggle */
+
+#ifdef ALTER_GAMMA
+	{0x1, 0x0010, 0x0008},	/* CCDSP YUV A11 */
+	{0x1, 0x0000, 0x0009},	/* CCDSP YUV A12 */
+	{0x1, 0x0000, 0x000a},	/* CCDSP YUV A13 */
+	{0x1, 0x0000, 0x000b},	/* CCDSP YUV A21 */
+	{0x1, 0x0010, 0x000c},	/* CCDSP YUV A22 */
+	{0x1, 0x0000, 0x000d},	/* CCDSP YUV A23 */
+	{0x1, 0x0000, 0x000e},	/* CCDSP YUV A31 */
+	{0x1, 0x0000, 0x000f},	/* CCDSP YUV A32 */
+	{0x1, 0x0010, 0x0010},	/* CCDSP YUV A33 */
+	{0x1, 0x0000, 0x0011},	/* CCDSP R Offset */
+	{0x1, 0x0000, 0x0012},	/* CCDSP G Offset */
+	{0x1, 0x0001, 0x0013},	/* CCDSP B Offset */
+	{0x1, 0x0001, 0x0014},	/* CCDSP BG Offset */
+	{0x1, 0x003f, 0x00C1},	/* CCDSP Gamma Correction Enable */
+#endif
+#endif
+
+#ifdef TG_SET
+	{0x0, 0x00fc, 0x0000},	/* TG Shutter Speed High Bits */
+	{0x0, 0x0000, 0x0001},	/* TG Shutter Speed Low Bits */
+	{0x0, 0x00e4, 0x0004},	/* TG DCLK*2 Adjust */
+	{0x0, 0x0008, 0x0005},	/* TG ADCK Adjust */
+	{0x0, 0x0003, 0x0006},	/* TG FR Phase Adjust */
+	{0x0, 0x0001, 0x0007},	/* TG FCDS Phase Adjust */
+	{0x0, 0x0039, 0x0008},	/* TG FS Phase Adjust */
+	{0x0, 0x0088, 0x000a},	/* TG MH1 */
+	{0x0, 0x0003, 0x000f},	/* TG Pixel ID */
+
+	/* Like below, unexplained toglleing */
+	{0x0, 0x0080, 0x000c},
+	{0x0, 0x0000, 0x000d},
+	{0x0, 0x0080, 0x000c},
+	{0x0, 0x0004, 0x000d},
+	{0x0, 0x0000, 0x000c},
+	{0x0, 0x0000, 0x000d},
+	{0x0, 0x0040, 0x000c},
+	{0x0, 0x0017, 0x000d},
+	{0x0, 0x00c0, 0x000c},
+	{0x0, 0x0000, 0x000d},
+	{0x0, 0x0080, 0x000c},
+	{0x0, 0x0006, 0x000d},
+	{0x0, 0x0080, 0x000c},
+	{0x0, 0x0004, 0x000d},
+	{0x0, 0x0002, 0x0003},
+#endif
+
+#ifdef DSPWIN_SET
+	{0x1, 0x001c, 0x0017},	/* CCDSP W1 Start X */
+	{0x1, 0x00e2, 0x0019},	/* CCDSP W2 Start X */
+	{0x1, 0x001c, 0x001b},	/* CCDSP W3 Start X */
+	{0x1, 0x00e2, 0x001d},	/* CCDSP W4 Start X */
+	{0x1, 0x00aa, 0x001f},	/* CCDSP W5 Start X */
+	{0x1, 0x0070, 0x0020},	/* CCDSP W5 Start Y */
+#endif
+	{0x0, 0x0001, 0x0010},	/* TG Start Clock */
+
+/*	{0x2, 0x006a, 0x0001},	 * C/S Enable ISOSYNCH Packet Engine */
+	{0x2, 0x0068, 0x0001},	/* C/S Diable ISOSYNCH Packet Engine */
+	{0x2, 0x0000, 0x0005},
+	{0x2, 0x0043, 0x0000},	/* C/S Set Timing Mode, Disable TG soft reset */
+	{0x2, 0x0043, 0x0000},	/* C/S Set Timing Mode, Disable TG soft reset */
+	{0x2, 0x0002, 0x0005},	/* C/S GPIO */
+	{0x2, 0x0003, 0x0005},	/* C/S GPIO */
+
+	{0x2, 0x006a, 0x0001},	/* C/S Enable ISOSYNCH Packet Engine */
+	{}
+};
+
+/*
+ * Data used to initialize a SPCA501C with HV7131B sensor.
+ * From a capture file taken with USBSnoop v 1.5
+ * I have a "SPCA501C pc camera chipset" manual by sunplus, but some
+ * of the value meanings are obscure or simply "reserved".
+ * to do list:
+ * 1) Understand what every value means
+ * 2) Understand why some values seem to appear more than once
+ * 3) Write a small comment for each line of the following arrays.
+ */
+static __u16 spca501c_arowana_open_data[][3] = {
+	/* bmRequest,value,index */
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x01, 0x0006, 0x0011},
+	{0x01, 0x00ff, 0x0012},
+	{0x01, 0x0014, 0x0013},
+	{0x01, 0x0000, 0x0014},
+	{0x01, 0x0042, 0x0051},
+	{0x01, 0x0040, 0x0052},
+	{0x01, 0x0051, 0x0053},
+	{0x01, 0x0040, 0x0054},
+	{0x01, 0x0000, 0x0055},
+	{0x00, 0x0025, 0x0000},
+	{0x00, 0x0026, 0x0000},
+	{0x00, 0x0001, 0x0000},
+	{0x00, 0x0027, 0x0000},
+	{0x00, 0x008a, 0x0000},
+	{}
+};
+
+static __u16 spca501c_arowana_init_data[][3] = {
+	/* bmRequest,value,index */
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x01, 0x0006, 0x0011},
+	{0x01, 0x00ff, 0x0012},
+	{0x01, 0x0014, 0x0013},
+	{0x01, 0x0000, 0x0014},
+	{0x01, 0x0042, 0x0051},
+	{0x01, 0x0040, 0x0052},
+	{0x01, 0x0051, 0x0053},
+	{0x01, 0x0040, 0x0054},
+	{0x01, 0x0000, 0x0055},
+	{0x00, 0x0025, 0x0000},
+	{0x00, 0x0026, 0x0000},
+	{0x00, 0x0001, 0x0000},
+	{0x00, 0x0027, 0x0000},
+	{0x00, 0x008a, 0x0000},
+	{0x02, 0x0000, 0x0005},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x2000, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0015, 0x0001},
+	{0x05, 0x00ea, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0023, 0x0001},
+	{0x05, 0x0003, 0x0000},
+	{0x05, 0x0030, 0x0001},
+	{0x05, 0x002b, 0x0000},
+	{0x05, 0x0031, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0032, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0033, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0034, 0x0001},
+	{0x05, 0x0002, 0x0000},
+	{0x05, 0x0050, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0051, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0052, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0054, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x00, 0x0000, 0x0001},
+	{0x00, 0x0000, 0x0002},
+	{0x00, 0x000c, 0x0003},
+	{0x00, 0x0000, 0x0004},
+	{0x00, 0x0090, 0x0005},
+	{0x00, 0x0000, 0x0006},
+	{0x00, 0x0040, 0x0007},
+	{0x00, 0x00c0, 0x0008},
+	{0x00, 0x004a, 0x0009},
+	{0x00, 0x0000, 0x000a},
+	{0x00, 0x0000, 0x000b},
+	{0x00, 0x0001, 0x000c},
+	{0x00, 0x0001, 0x000d},
+	{0x00, 0x0000, 0x000e},
+	{0x00, 0x0002, 0x000f},
+	{0x00, 0x0001, 0x0010},
+	{0x00, 0x0000, 0x0011},
+	{0x00, 0x0000, 0x0012},
+	{0x00, 0x0002, 0x0020},
+	{0x00, 0x0080, 0x0021},
+	{0x00, 0x0001, 0x0022},
+	{0x00, 0x00e0, 0x0023},
+	{0x00, 0x0000, 0x0024},
+	{0x00, 0x00d5, 0x0025},
+	{0x00, 0x0000, 0x0026},
+	{0x00, 0x000b, 0x0027},
+	{0x00, 0x0000, 0x0046},
+	{0x00, 0x0000, 0x0047},
+	{0x00, 0x0000, 0x0048},
+	{0x00, 0x0000, 0x0049},
+	{0x00, 0x0008, 0x004a},
+	{0xff, 0x0000, 0x00d0},
+	{0xff, 0x00d8, 0x00d1},
+	{0xff, 0x0000, 0x00d4},
+	{0xff, 0x0000, 0x00d5},
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0028, 0x0001},
+	{0x01, 0x0000, 0x0002},
+	{0x01, 0x000a, 0x0003},
+	{0x01, 0x0040, 0x0004},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0x00fd, 0x000a},
+	{0x01, 0x0038, 0x000b},
+	{0x01, 0x00d1, 0x000c},
+	{0x01, 0x00f7, 0x000d},
+	{0x01, 0x00ed, 0x000e},
+	{0x01, 0x00d8, 0x000f},
+	{0x01, 0x0038, 0x0010},
+	{0x01, 0x00ff, 0x0015},
+	{0x01, 0x0001, 0x0016},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x0000, 0x001f},
+	{0x01, 0x0000, 0x0020},
+	{0x01, 0x00ff, 0x003e},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0000, 0x0040},
+	{0x01, 0x0035, 0x0041},
+	{0x01, 0x0053, 0x0042},
+	{0x01, 0x0069, 0x0043},
+	{0x01, 0x007c, 0x0044},
+	{0x01, 0x008c, 0x0045},
+	{0x01, 0x009a, 0x0046},
+	{0x01, 0x00a8, 0x0047},
+	{0x01, 0x00b4, 0x0048},
+	{0x01, 0x00bf, 0x0049},
+	{0x01, 0x00ca, 0x004a},
+	{0x01, 0x00d4, 0x004b},
+	{0x01, 0x00dd, 0x004c},
+	{0x01, 0x00e7, 0x004d},
+	{0x01, 0x00ef, 0x004e},
+	{0x01, 0x00f8, 0x004f},
+	{0x01, 0x00ff, 0x0050},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0060, 0x0057},
+	{0x01, 0x0040, 0x0058},
+	{0x01, 0x0011, 0x0059},
+	{0x01, 0x0001, 0x005a},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x0015, 0x0006},
+	{0x02, 0x100a, 0x0007},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0025, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0001, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0020, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x00, 0x0090, 0x0005},
+	{0x01, 0x00a6, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x2000, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0015, 0x0001},
+	{0x05, 0x00ea, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0023, 0x0001},
+	{0x05, 0x0003, 0x0000},
+	{0x05, 0x0030, 0x0001},
+	{0x05, 0x002b, 0x0000},
+	{0x05, 0x0031, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0032, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0033, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0034, 0x0001},
+	{0x05, 0x0002, 0x0000},
+	{0x05, 0x0050, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0051, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0052, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0054, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x00, 0x0000, 0x0001},
+	{0x00, 0x0000, 0x0002},
+	{0x00, 0x000c, 0x0003},
+	{0x00, 0x0000, 0x0004},
+	{0x00, 0x0090, 0x0005},
+	{0x00, 0x0000, 0x0006},
+	{0x00, 0x0040, 0x0007},
+	{0x00, 0x00c0, 0x0008},
+	{0x00, 0x004a, 0x0009},
+	{0x00, 0x0000, 0x000a},
+	{0x00, 0x0000, 0x000b},
+	{0x00, 0x0001, 0x000c},
+	{0x00, 0x0001, 0x000d},
+	{0x00, 0x0000, 0x000e},
+	{0x00, 0x0002, 0x000f},
+	{0x00, 0x0001, 0x0010},
+	{0x00, 0x0000, 0x0011},
+	{0x00, 0x0000, 0x0012},
+	{0x00, 0x0002, 0x0020},
+	{0x00, 0x0080, 0x0021},
+	{0x00, 0x0001, 0x0022},
+	{0x00, 0x00e0, 0x0023},
+	{0x00, 0x0000, 0x0024},
+	{0x00, 0x00d5, 0x0025},
+	{0x00, 0x0000, 0x0026},
+	{0x00, 0x000b, 0x0027},
+	{0x00, 0x0000, 0x0046},
+	{0x00, 0x0000, 0x0047},
+	{0x00, 0x0000, 0x0048},
+	{0x00, 0x0000, 0x0049},
+	{0x00, 0x0008, 0x004a},
+	{0xff, 0x0000, 0x00d0},
+	{0xff, 0x00d8, 0x00d1},
+	{0xff, 0x0000, 0x00d4},
+	{0xff, 0x0000, 0x00d5},
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0028, 0x0001},
+	{0x01, 0x0000, 0x0002},
+	{0x01, 0x000a, 0x0003},
+	{0x01, 0x0040, 0x0004},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0x00fd, 0x000a},
+	{0x01, 0x0038, 0x000b},
+	{0x01, 0x00d1, 0x000c},
+	{0x01, 0x00f7, 0x000d},
+	{0x01, 0x00ed, 0x000e},
+	{0x01, 0x00d8, 0x000f},
+	{0x01, 0x0038, 0x0010},
+	{0x01, 0x00ff, 0x0015},
+	{0x01, 0x0001, 0x0016},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x0000, 0x001f},
+	{0x01, 0x0000, 0x0020},
+	{0x01, 0x00ff, 0x003e},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0000, 0x0040},
+	{0x01, 0x0035, 0x0041},
+	{0x01, 0x0053, 0x0042},
+	{0x01, 0x0069, 0x0043},
+	{0x01, 0x007c, 0x0044},
+	{0x01, 0x008c, 0x0045},
+	{0x01, 0x009a, 0x0046},
+	{0x01, 0x00a8, 0x0047},
+	{0x01, 0x00b4, 0x0048},
+	{0x01, 0x00bf, 0x0049},
+	{0x01, 0x00ca, 0x004a},
+	{0x01, 0x00d4, 0x004b},
+	{0x01, 0x00dd, 0x004c},
+	{0x01, 0x00e7, 0x004d},
+	{0x01, 0x00ef, 0x004e},
+	{0x01, 0x00f8, 0x004f},
+	{0x01, 0x00ff, 0x0050},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0060, 0x0057},
+	{0x01, 0x0040, 0x0058},
+	{0x01, 0x0011, 0x0059},
+	{0x01, 0x0001, 0x005a},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x0015, 0x0006},
+	{0x02, 0x100a, 0x0007},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0025, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0001, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0020, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x00, 0x0090, 0x0005},
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0xfffd, 0x000a},
+	{0x01, 0x0023, 0x000b},
+	{0x01, 0xffea, 0x000c},
+	{0x01, 0xfff4, 0x000d},
+	{0x01, 0xfffc, 0x000e},
+	{0x01, 0xffe3, 0x000f},
+	{0x01, 0x001f, 0x0010},
+	{0x01, 0x00a8, 0x0001},
+	{0x01, 0x0067, 0x0007},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x00c8, 0x0015},
+	{0x01, 0x0032, 0x0016},
+	{0x01, 0x0000, 0x0011},
+	{0x01, 0x0000, 0x0012},
+	{0x01, 0x0000, 0x0013},
+	{0x01, 0x000a, 0x0003},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xc000, 0x0001},
+	{0x02, 0x0000, 0x0005},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x2000, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0015, 0x0001},
+	{0x05, 0x00ea, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0023, 0x0001},
+	{0x05, 0x0003, 0x0000},
+	{0x05, 0x0030, 0x0001},
+	{0x05, 0x002b, 0x0000},
+	{0x05, 0x0031, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0032, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0033, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0034, 0x0001},
+	{0x05, 0x0002, 0x0000},
+	{0x05, 0x0050, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0051, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0052, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0054, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x00, 0x0000, 0x0001},
+	{0x00, 0x0000, 0x0002},
+	{0x00, 0x000c, 0x0003},
+	{0x00, 0x0000, 0x0004},
+	{0x00, 0x0090, 0x0005},
+	{0x00, 0x0000, 0x0006},
+	{0x00, 0x0040, 0x0007},
+	{0x00, 0x00c0, 0x0008},
+	{0x00, 0x004a, 0x0009},
+	{0x00, 0x0000, 0x000a},
+	{0x00, 0x0000, 0x000b},
+	{0x00, 0x0001, 0x000c},
+	{0x00, 0x0001, 0x000d},
+	{0x00, 0x0000, 0x000e},
+	{0x00, 0x0002, 0x000f},
+	{0x00, 0x0001, 0x0010},
+	{0x00, 0x0000, 0x0011},
+	{0x00, 0x0000, 0x0012},
+	{0x00, 0x0002, 0x0020},
+	{0x00, 0x0080, 0x0021},
+	{0x00, 0x0001, 0x0022},
+	{0x00, 0x00e0, 0x0023},
+	{0x00, 0x0000, 0x0024},
+	{0x00, 0x00d5, 0x0025},
+	{0x00, 0x0000, 0x0026},
+	{0x00, 0x000b, 0x0027},
+	{0x00, 0x0000, 0x0046},
+	{0x00, 0x0000, 0x0047},
+	{0x00, 0x0000, 0x0048},
+	{0x00, 0x0000, 0x0049},
+	{0x00, 0x0008, 0x004a},
+	{0xff, 0x0000, 0x00d0},
+	{0xff, 0x00d8, 0x00d1},
+	{0xff, 0x0000, 0x00d4},
+	{0xff, 0x0000, 0x00d5},
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0028, 0x0001},
+	{0x01, 0x0000, 0x0002},
+	{0x01, 0x000a, 0x0003},
+	{0x01, 0x0040, 0x0004},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0x00fd, 0x000a},
+	{0x01, 0x0038, 0x000b},
+	{0x01, 0x00d1, 0x000c},
+	{0x01, 0x00f7, 0x000d},
+	{0x01, 0x00ed, 0x000e},
+	{0x01, 0x00d8, 0x000f},
+	{0x01, 0x0038, 0x0010},
+	{0x01, 0x00ff, 0x0015},
+	{0x01, 0x0001, 0x0016},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x0000, 0x001f},
+	{0x01, 0x0000, 0x0020},
+	{0x01, 0x00ff, 0x003e},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0000, 0x0040},
+	{0x01, 0x0035, 0x0041},
+	{0x01, 0x0053, 0x0042},
+	{0x01, 0x0069, 0x0043},
+	{0x01, 0x007c, 0x0044},
+	{0x01, 0x008c, 0x0045},
+	{0x01, 0x009a, 0x0046},
+	{0x01, 0x00a8, 0x0047},
+	{0x01, 0x00b4, 0x0048},
+	{0x01, 0x00bf, 0x0049},
+	{0x01, 0x00ca, 0x004a},
+	{0x01, 0x00d4, 0x004b},
+	{0x01, 0x00dd, 0x004c},
+	{0x01, 0x00e7, 0x004d},
+	{0x01, 0x00ef, 0x004e},
+	{0x01, 0x00f8, 0x004f},
+	{0x01, 0x00ff, 0x0050},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0060, 0x0057},
+	{0x01, 0x0040, 0x0058},
+	{0x01, 0x0011, 0x0059},
+	{0x01, 0x0001, 0x005a},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x0015, 0x0006},
+	{0x02, 0x100a, 0x0007},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0025, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0001, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0020, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x00, 0x0090, 0x0005},
+	{0x01, 0x00a6, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x2000, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0015, 0x0001},
+	{0x05, 0x00ea, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0023, 0x0001},
+	{0x05, 0x0003, 0x0000},
+	{0x05, 0x0030, 0x0001},
+	{0x05, 0x002b, 0x0000},
+	{0x05, 0x0031, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0032, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0033, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0034, 0x0001},
+	{0x05, 0x0002, 0x0000},
+	{0x05, 0x0050, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0051, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0052, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0054, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x00, 0x0000, 0x0001},
+	{0x00, 0x0000, 0x0002},
+	{0x00, 0x000c, 0x0003},
+	{0x00, 0x0000, 0x0004},
+	{0x00, 0x0090, 0x0005},
+	{0x00, 0x0000, 0x0006},
+	{0x00, 0x0040, 0x0007},
+	{0x00, 0x00c0, 0x0008},
+	{0x00, 0x004a, 0x0009},
+	{0x00, 0x0000, 0x000a},
+	{0x00, 0x0000, 0x000b},
+	{0x00, 0x0001, 0x000c},
+	{0x00, 0x0001, 0x000d},
+	{0x00, 0x0000, 0x000e},
+	{0x00, 0x0002, 0x000f},
+	{0x00, 0x0001, 0x0010},
+	{0x00, 0x0000, 0x0011},
+	{0x00, 0x0000, 0x0012},
+	{0x00, 0x0002, 0x0020},
+	{0x00, 0x0080, 0x0021},
+	{0x00, 0x0001, 0x0022},
+	{0x00, 0x00e0, 0x0023},
+	{0x00, 0x0000, 0x0024},
+	{0x00, 0x00d5, 0x0025},
+	{0x00, 0x0000, 0x0026},
+	{0x00, 0x000b, 0x0027},
+	{0x00, 0x0000, 0x0046},
+	{0x00, 0x0000, 0x0047},
+	{0x00, 0x0000, 0x0048},
+	{0x00, 0x0000, 0x0049},
+	{0x00, 0x0008, 0x004a},
+	{0xff, 0x0000, 0x00d0},
+	{0xff, 0x00d8, 0x00d1},
+	{0xff, 0x0000, 0x00d4},
+	{0xff, 0x0000, 0x00d5},
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0028, 0x0001},
+	{0x01, 0x0000, 0x0002},
+	{0x01, 0x000a, 0x0003},
+	{0x01, 0x0040, 0x0004},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0x00fd, 0x000a},
+	{0x01, 0x0038, 0x000b},
+	{0x01, 0x00d1, 0x000c},
+	{0x01, 0x00f7, 0x000d},
+	{0x01, 0x00ed, 0x000e},
+	{0x01, 0x00d8, 0x000f},
+	{0x01, 0x0038, 0x0010},
+	{0x01, 0x00ff, 0x0015},
+	{0x01, 0x0001, 0x0016},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x0000, 0x001f},
+	{0x01, 0x0000, 0x0020},
+	{0x01, 0x00ff, 0x003e},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0000, 0x0040},
+	{0x01, 0x0035, 0x0041},
+	{0x01, 0x0053, 0x0042},
+	{0x01, 0x0069, 0x0043},
+	{0x01, 0x007c, 0x0044},
+	{0x01, 0x008c, 0x0045},
+	{0x01, 0x009a, 0x0046},
+	{0x01, 0x00a8, 0x0047},
+	{0x01, 0x00b4, 0x0048},
+	{0x01, 0x00bf, 0x0049},
+	{0x01, 0x00ca, 0x004a},
+	{0x01, 0x00d4, 0x004b},
+	{0x01, 0x00dd, 0x004c},
+	{0x01, 0x00e7, 0x004d},
+	{0x01, 0x00ef, 0x004e},
+	{0x01, 0x00f8, 0x004f},
+	{0x01, 0x00ff, 0x0050},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0060, 0x0057},
+	{0x01, 0x0040, 0x0058},
+	{0x01, 0x0011, 0x0059},
+	{0x01, 0x0001, 0x005a},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x0015, 0x0006},
+	{0x02, 0x100a, 0x0007},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0025, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0001, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0020, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x00, 0x0090, 0x0005},
+	{0x01, 0x00a6, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x000f, 0x0000},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0xfffd, 0x000a},
+	{0x01, 0x0023, 0x000b},
+	{0x01, 0xffea, 0x000c},
+	{0x01, 0xfff4, 0x000d},
+	{0x01, 0xfffc, 0x000e},
+	{0x01, 0xffe3, 0x000f},
+	{0x01, 0x001f, 0x0010},
+	{0x01, 0x00a8, 0x0001},
+	{0x01, 0x0067, 0x0007},
+	{0x01, 0x0042, 0x0051},
+	{0x01, 0x0051, 0x0053},
+	{0x01, 0x000a, 0x0003},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xc000, 0x0001},
+	{0x02, 0x0000, 0x0005},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x2000, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0015, 0x0001},
+	{0x05, 0x00ea, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0023, 0x0001},
+	{0x05, 0x0003, 0x0000},
+	{0x05, 0x0030, 0x0001},
+	{0x05, 0x002b, 0x0000},
+	{0x05, 0x0031, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0032, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0033, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0034, 0x0001},
+	{0x05, 0x0002, 0x0000},
+	{0x05, 0x0050, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0051, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0052, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0054, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x00, 0x0000, 0x0001},
+	{0x00, 0x0000, 0x0002},
+	{0x00, 0x000c, 0x0003},
+	{0x00, 0x0000, 0x0004},
+	{0x00, 0x0090, 0x0005},
+	{0x00, 0x0000, 0x0006},
+	{0x00, 0x0040, 0x0007},
+	{0x00, 0x00c0, 0x0008},
+	{0x00, 0x004a, 0x0009},
+	{0x00, 0x0000, 0x000a},
+	{0x00, 0x0000, 0x000b},
+	{0x00, 0x0001, 0x000c},
+	{0x00, 0x0001, 0x000d},
+	{0x00, 0x0000, 0x000e},
+	{0x00, 0x0002, 0x000f},
+	{0x00, 0x0001, 0x0010},
+	{0x00, 0x0000, 0x0011},
+	{0x00, 0x0000, 0x0012},
+	{0x00, 0x0002, 0x0020},
+	{0x00, 0x0080, 0x0021},
+	{0x00, 0x0001, 0x0022},
+	{0x00, 0x00e0, 0x0023},
+	{0x00, 0x0000, 0x0024},
+	{0x00, 0x00d5, 0x0025},
+	{0x00, 0x0000, 0x0026},
+	{0x00, 0x000b, 0x0027},
+	{0x00, 0x0000, 0x0046},
+	{0x00, 0x0000, 0x0047},
+	{0x00, 0x0000, 0x0048},
+	{0x00, 0x0000, 0x0049},
+	{0x00, 0x0008, 0x004a},
+	{0xff, 0x0000, 0x00d0},
+	{0xff, 0x00d8, 0x00d1},
+	{0xff, 0x0000, 0x00d4},
+	{0xff, 0x0000, 0x00d5},
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0028, 0x0001},
+	{0x01, 0x0000, 0x0002},
+	{0x01, 0x000a, 0x0003},
+	{0x01, 0x0040, 0x0004},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0x00fd, 0x000a},
+	{0x01, 0x0038, 0x000b},
+	{0x01, 0x00d1, 0x000c},
+	{0x01, 0x00f7, 0x000d},
+	{0x01, 0x00ed, 0x000e},
+	{0x01, 0x00d8, 0x000f},
+	{0x01, 0x0038, 0x0010},
+	{0x01, 0x00ff, 0x0015},
+	{0x01, 0x0001, 0x0016},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x0000, 0x001f},
+	{0x01, 0x0000, 0x0020},
+	{0x01, 0x00ff, 0x003e},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0000, 0x0040},
+	{0x01, 0x0035, 0x0041},
+	{0x01, 0x0053, 0x0042},
+	{0x01, 0x0069, 0x0043},
+	{0x01, 0x007c, 0x0044},
+	{0x01, 0x008c, 0x0045},
+	{0x01, 0x009a, 0x0046},
+	{0x01, 0x00a8, 0x0047},
+	{0x01, 0x00b4, 0x0048},
+	{0x01, 0x00bf, 0x0049},
+	{0x01, 0x00ca, 0x004a},
+	{0x01, 0x00d4, 0x004b},
+	{0x01, 0x00dd, 0x004c},
+	{0x01, 0x00e7, 0x004d},
+	{0x01, 0x00ef, 0x004e},
+	{0x01, 0x00f8, 0x004f},
+	{0x01, 0x00ff, 0x0050},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0060, 0x0057},
+	{0x01, 0x0040, 0x0058},
+	{0x01, 0x0011, 0x0059},
+	{0x01, 0x0001, 0x005a},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x0015, 0x0006},
+	{0x02, 0x100a, 0x0007},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0025, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0001, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0020, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x00, 0x0090, 0x0005},
+	{0x01, 0x00a6, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x2000, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0015, 0x0001},
+	{0x05, 0x00ea, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0023, 0x0001},
+	{0x05, 0x0003, 0x0000},
+	{0x05, 0x0030, 0x0001},
+	{0x05, 0x002b, 0x0000},
+	{0x05, 0x0031, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0032, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0033, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0034, 0x0001},
+	{0x05, 0x0002, 0x0000},
+	{0x05, 0x0050, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0051, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0052, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0054, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x00, 0x0000, 0x0001},
+	{0x00, 0x0000, 0x0002},
+	{0x00, 0x000c, 0x0003},
+	{0x00, 0x0000, 0x0004},
+	{0x00, 0x0090, 0x0005},
+	{0x00, 0x0000, 0x0006},
+	{0x00, 0x0040, 0x0007},
+	{0x00, 0x00c0, 0x0008},
+	{0x00, 0x004a, 0x0009},
+	{0x00, 0x0000, 0x000a},
+	{0x00, 0x0000, 0x000b},
+	{0x00, 0x0001, 0x000c},
+	{0x00, 0x0001, 0x000d},
+	{0x00, 0x0000, 0x000e},
+	{0x00, 0x0002, 0x000f},
+	{0x00, 0x0001, 0x0010},
+	{0x00, 0x0000, 0x0011},
+	{0x00, 0x0000, 0x0012},
+	{0x00, 0x0002, 0x0020},
+	{0x00, 0x0080, 0x0021},
+	{0x00, 0x0001, 0x0022},
+	{0x00, 0x00e0, 0x0023},
+	{0x00, 0x0000, 0x0024},
+	{0x00, 0x00d5, 0x0025},
+	{0x00, 0x0000, 0x0026},
+	{0x00, 0x000b, 0x0027},
+	{0x00, 0x0000, 0x0046},
+	{0x00, 0x0000, 0x0047},
+	{0x00, 0x0000, 0x0048},
+	{0x00, 0x0000, 0x0049},
+	{0x00, 0x0008, 0x004a},
+	{0xff, 0x0000, 0x00d0},
+	{0xff, 0x00d8, 0x00d1},
+	{0xff, 0x0000, 0x00d4},
+	{0xff, 0x0000, 0x00d5},
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0028, 0x0001},
+	{0x01, 0x0000, 0x0002},
+	{0x01, 0x000a, 0x0003},
+	{0x01, 0x0040, 0x0004},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0x00fd, 0x000a},
+	{0x01, 0x0038, 0x000b},
+	{0x01, 0x00d1, 0x000c},
+	{0x01, 0x00f7, 0x000d},
+	{0x01, 0x00ed, 0x000e},
+	{0x01, 0x00d8, 0x000f},
+	{0x01, 0x0038, 0x0010},
+	{0x01, 0x00ff, 0x0015},
+	{0x01, 0x0001, 0x0016},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x0000, 0x001f},
+	{0x01, 0x0000, 0x0020},
+	{0x01, 0x00ff, 0x003e},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0000, 0x0040},
+	{0x01, 0x0035, 0x0041},
+	{0x01, 0x0053, 0x0042},
+	{0x01, 0x0069, 0x0043},
+	{0x01, 0x007c, 0x0044},
+	{0x01, 0x008c, 0x0045},
+	{0x01, 0x009a, 0x0046},
+	{0x01, 0x00a8, 0x0047},
+	{0x01, 0x00b4, 0x0048},
+	{0x01, 0x00bf, 0x0049},
+	{0x01, 0x00ca, 0x004a},
+	{0x01, 0x00d4, 0x004b},
+	{0x01, 0x00dd, 0x004c},
+	{0x01, 0x00e7, 0x004d},
+	{0x01, 0x00ef, 0x004e},
+	{0x01, 0x00f8, 0x004f},
+	{0x01, 0x00ff, 0x0050},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0060, 0x0057},
+	{0x01, 0x0040, 0x0058},
+	{0x01, 0x0011, 0x0059},
+	{0x01, 0x0001, 0x005a},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x0015, 0x0006},
+	{0x02, 0x100a, 0x0007},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0025, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0001, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0020, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x00, 0x0090, 0x0005},
+	{0x01, 0x00a6, 0x0000},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x001e, 0x0000},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x0011, 0x0008},
+	{0x01, 0x0032, 0x0009},
+	{0x01, 0xfffd, 0x000a},
+	{0x01, 0x0023, 0x000b},
+	{0x01, 0xffea, 0x000c},
+	{0x01, 0xfff4, 0x000d},
+	{0x01, 0xfffc, 0x000e},
+	{0x01, 0xffe3, 0x000f},
+	{0x01, 0x001f, 0x0010},
+	{0x01, 0x00a8, 0x0001},
+	{0x01, 0x0067, 0x0007},
+	{0x01, 0x0042, 0x0051},
+	{0x01, 0x0051, 0x0053},
+	{0x01, 0x000a, 0x0003},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x0007, 0x0005},
+	{0x01, 0x0042, 0x0051},
+	{0x01, 0x0051, 0x0053},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x002d, 0x0000},
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0001, 0x0056},
+	{0x02, 0xc000, 0x0001},
+	{0x02, 0x0000, 0x0005},
+	{}
+};
+
+/* Unknow camera from Ori Usbid 0x0000:0x0000 */
+/* Based on snoops from Ori Cohen */
+static __u16 spca501c_mysterious_open_data[][3] = {
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+/* DSP Registers */
+	{0x01, 0x0016, 0x0011},	/* RGB offset */
+	{0x01, 0x0000, 0x0012},
+	{0x01, 0x0006, 0x0013},
+	{0x01, 0x0078, 0x0051},
+	{0x01, 0x0040, 0x0052},
+	{0x01, 0x0046, 0x0053},
+	{0x01, 0x0040, 0x0054},
+	{0x00, 0x0025, 0x0000},
+/*	{0x00, 0x0000, 0x0000 }, */
+/* Part 2 */
+/* TG Registers */
+	{0x00, 0x0026, 0x0000},
+	{0x00, 0x0001, 0x0000},
+	{0x00, 0x0027, 0x0000},
+	{0x00, 0x008a, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x2000, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0015, 0x0001},
+	{0x05, 0x00ea, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0023, 0x0001},
+	{0x05, 0x0003, 0x0000},
+	{0x05, 0x0030, 0x0001},
+	{0x05, 0x002b, 0x0000},
+	{0x05, 0x0031, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0032, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0033, 0x0001},
+	{0x05, 0x0023, 0x0000},
+	{0x05, 0x0034, 0x0001},
+	{0x05, 0x0002, 0x0000},
+	{0x05, 0x0050, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0051, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0052, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0054, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{}
+};
+
+/* Based on snoops from Ori Cohen */
+static __u16 spca501c_mysterious_init_data[][3] = {
+/* Part 3 */
+/* TG registers */
+/*	{0x00, 0x0000, 0x0000}, */
+	{0x00, 0x0000, 0x0001},
+	{0x00, 0x0000, 0x0002},
+	{0x00, 0x0006, 0x0003},
+	{0x00, 0x0000, 0x0004},
+	{0x00, 0x0090, 0x0005},
+	{0x00, 0x0000, 0x0006},
+	{0x00, 0x0040, 0x0007},
+	{0x00, 0x00c0, 0x0008},
+	{0x00, 0x004a, 0x0009},
+	{0x00, 0x0000, 0x000a},
+	{0x00, 0x0000, 0x000b},
+	{0x00, 0x0001, 0x000c},
+	{0x00, 0x0001, 0x000d},
+	{0x00, 0x0000, 0x000e},
+	{0x00, 0x0002, 0x000f},
+	{0x00, 0x0001, 0x0010},
+	{0x00, 0x0000, 0x0011},
+	{0x00, 0x0001, 0x0012},
+	{0x00, 0x0002, 0x0020},
+	{0x00, 0x0080, 0x0021},	/* 640 */
+	{0x00, 0x0001, 0x0022},
+	{0x00, 0x00e0, 0x0023},	/* 480 */
+	{0x00, 0x0000, 0x0024},	/* Offset H hight */
+	{0x00, 0x00d3, 0x0025},	/* low */
+	{0x00, 0x0000, 0x0026},	/* Offset V */
+	{0x00, 0x000d, 0x0027},	/* low */
+	{0x00, 0x0000, 0x0046},
+	{0x00, 0x0000, 0x0047},
+	{0x00, 0x0000, 0x0048},
+	{0x00, 0x0000, 0x0049},
+	{0x00, 0x0008, 0x004a},
+/* DSP Registers	 */
+	{0x01, 0x00a6, 0x0000},
+	{0x01, 0x0028, 0x0001},
+	{0x01, 0x0000, 0x0002},
+	{0x01, 0x000a, 0x0003},	/* Level Calc bit7 ->1 Auto */
+	{0x01, 0x0040, 0x0004},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x000f, 0x0008},	/* A11 Color correction coeff */
+	{0x01, 0x002d, 0x0009},	/* A12 */
+	{0x01, 0x0005, 0x000a},	/* A13 */
+	{0x01, 0x0023, 0x000b},	/* A21 */
+	{0x01, 0x00e0, 0x000c},	/* A22 */
+	{0x01, 0x00fd, 0x000d},	/* A23 */
+	{0x01, 0x00f4, 0x000e},	/* A31 */
+	{0x01, 0x00e4, 0x000f},	/* A32 */
+	{0x01, 0x0028, 0x0010},	/* A33 */
+	{0x01, 0x00ff, 0x0015},	/* Reserved */
+	{0x01, 0x0001, 0x0016},	/* Reserved */
+	{0x01, 0x0032, 0x0017},	/* Win1 Start begin */
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x0000, 0x001f},
+	{0x01, 0x0000, 0x0020},	/* Win1 Start end */
+	{0x01, 0x00ff, 0x003e},	/* Reserved begin */
+	{0x01, 0x0002, 0x003f},
+	{0x01, 0x0000, 0x0040},
+	{0x01, 0x0035, 0x0041},
+	{0x01, 0x0053, 0x0042},
+	{0x01, 0x0069, 0x0043},
+	{0x01, 0x007c, 0x0044},
+	{0x01, 0x008c, 0x0045},
+	{0x01, 0x009a, 0x0046},
+	{0x01, 0x00a8, 0x0047},
+	{0x01, 0x00b4, 0x0048},
+	{0x01, 0x00bf, 0x0049},
+	{0x01, 0x00ca, 0x004a},
+	{0x01, 0x00d4, 0x004b},
+	{0x01, 0x00dd, 0x004c},
+	{0x01, 0x00e7, 0x004d},
+	{0x01, 0x00ef, 0x004e},
+	{0x01, 0x00f8, 0x004f},
+	{0x01, 0x00ff, 0x0050},
+	{0x01, 0x0003, 0x0056},	/* Reserved end */
+	{0x01, 0x0060, 0x0057},	/* Edge Gain */
+	{0x01, 0x0040, 0x0058},
+	{0x01, 0x0011, 0x0059},	/* Edge Bandwidth */
+	{0x01, 0x0001, 0x005a},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0x0007, 0x0005},
+	{0x02, 0x0015, 0x0006},
+	{0x02, 0x200a, 0x0007},
+	{0x02, 0xa048, 0x0000},
+	{0x02, 0xc000, 0x0001},
+	{0x02, 0x000f, 0x0005},
+	{0x02, 0xa048, 0x0000},
+	{0x05, 0x0022, 0x0004},
+	{0x05, 0x0025, 0x0001},
+	{0x05, 0x0000, 0x0000},
+/* Part 4             */
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0001, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x05, 0x0021, 0x0001},
+	{0x05, 0x00d2, 0x0000},
+	{0x05, 0x0020, 0x0001},
+	{0x05, 0x0000, 0x0000},
+	{0x00, 0x0090, 0x0005},
+	{0x01, 0x00a6, 0x0000},
+	{0x02, 0x0000, 0x0005},
+	{0x05, 0x0026, 0x0001},
+	{0x05, 0x0001, 0x0000},
+	{0x05, 0x0027, 0x0001},
+	{0x05, 0x004e, 0x0000},
+/* Part 5        	 */
+	{0x01, 0x0003, 0x003f},
+	{0x01, 0x0001, 0x0056},
+	{0x01, 0x000f, 0x0008},
+	{0x01, 0x002d, 0x0009},
+	{0x01, 0x0005, 0x000a},
+	{0x01, 0x0023, 0x000b},
+	{0x01, 0xffe0, 0x000c},
+	{0x01, 0xfffd, 0x000d},
+	{0x01, 0xfff4, 0x000e},
+	{0x01, 0xffe4, 0x000f},
+	{0x01, 0x0028, 0x0010},
+	{0x01, 0x00a8, 0x0001},
+	{0x01, 0x0066, 0x0007},
+	{0x01, 0x0032, 0x0017},
+	{0x01, 0x0023, 0x0018},
+	{0x01, 0x00ce, 0x0019},
+	{0x01, 0x0023, 0x001a},
+	{0x01, 0x0032, 0x001b},
+	{0x01, 0x008d, 0x001c},
+	{0x01, 0x00ce, 0x001d},
+	{0x01, 0x008d, 0x001e},
+	{0x01, 0x00c8, 0x0015},	/* c8 Poids fort Luma */
+	{0x01, 0x0032, 0x0016},	/* 32 */
+	{0x01, 0x0016, 0x0011},	/* R 00 */
+	{0x01, 0x0016, 0x0012},	/* G 00 */
+	{0x01, 0x0016, 0x0013},	/* B 00 */
+	{0x01, 0x000a, 0x0003},
+	{0x02, 0xc002, 0x0001},
+	{0x02, 0x0007, 0x0005},
+	{}
+};
+
+static int reg_write(struct usb_device *dev,
+		     __u16 req, __u16 index, __u16 value)
+{
+	int ret;
+
+	ret = usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			req,
+			USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, NULL, 0, 500);
+	PDEBUG(D_USBO, "reg write: 0x%02x 0x%02x 0x%02x",
+		req, index, value);
+	if (ret < 0)
+		PDEBUG(D_ERR, "reg write: error %d", ret);
+	return ret;
+}
+
+/* returns: negative is error, pos or zero is data */
+static int reg_read(struct usb_device *dev,
+			__u16 req,	/* bRequest */
+			__u16 index,	/* wIndex */
+			__u16 length)	/* wLength (1 or 2 only) */
+{
+	int ret;
+	__u8 buf[2];
+
+	buf[1] = 0;
+	ret = usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			req,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,		/* value */
+			index,
+			buf, length,
+			500);			/* timeout */
+	if (ret < 0) {
+		PDEBUG(D_ERR, "reg_read err %d", ret);
+		return -1;
+	}
+	return (buf[1] << 8) + buf[0];
+}
+
+static int write_vector(struct gspca_dev *gspca_dev,
+				__u16 data[][3])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int ret, i = 0;
+
+	while (data[i][0] != 0 || data[i][1] != 0 || data[i][2] != 0) {
+		ret = reg_write(dev, data[i][0], data[i][2], data[i][1]);
+		if (ret < 0) {
+			PDEBUG(D_ERR,
+				"Reg write failed for 0x%02x,0x%02x,0x%02x",
+				data[i][0], data[i][1], data[i][2]);
+			return ret;
+		}
+		i++;
+	}
+	return 0;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	reg_write(gspca_dev->dev, SPCA501_REG_CCDSP, 0x11, sd->brightness);
+	reg_write(gspca_dev->dev, SPCA501_REG_CCDSP, 0x12, sd->brightness);
+	reg_write(gspca_dev->dev, SPCA501_REG_CCDSP, 0x13, sd->brightness);
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u16 brightness;
+
+	brightness = reg_read(gspca_dev->dev, SPCA501_REG_CCDSP, 0x11, 2);
+	sd->brightness = brightness << 1;
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	reg_write(gspca_dev->dev, 0x00, 0x00,
+				  (sd->contrast >> 8) & 0xff);
+	reg_write(gspca_dev->dev, 0x00, 0x01,
+				  sd->contrast & 0xff);
+}
+
+static void getcontrast(struct gspca_dev *gspca_dev)
+{
+/*	spca50x->contrast = 0xaa01; */
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	reg_write(gspca_dev->dev, SPCA501_REG_CCDSP, 0x0c, sd->colors);
+}
+
+static void getcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = reg_read(gspca_dev->dev, SPCA501_REG_CCDSP, 0x0c, 2);
+/*	sd->hue = (reg_read(gspca_dev->dev, SPCA501_REG_CCDSP, 0x13, */
+/*			2) & 0xFF) << 8; */
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+	__u16 vendor;
+	__u16 product;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+	switch (vendor) {
+	case 0x0000:		/* Unknow Camera */
+/*		switch (product) { */
+/*		case 0x0000: */
+			sd->subtype = MystFromOriUnknownCamera;
+/*			break; */
+/*		} */
+		break;
+	case 0x040a:		/* Kodak cameras */
+/*		switch (product) { */
+/*		case 0x0002: */
+			sd->subtype = KodakDVC325;
+/*			break; */
+/*		} */
+		break;
+	case 0x0497:		/* Smile International */
+/*		switch (product) { */
+/*		case 0xc001: */
+			sd->subtype = SmileIntlCamera;
+/*			break; */
+/*		} */
+		break;
+	case 0x0506:		/* 3COM cameras */
+/*		switch (product) { */
+/*		case 0x00df: */
+			sd->subtype = ThreeComHomeConnectLite;
+/*			break; */
+/*		} */
+		break;
+	case 0x0733:	/* Rebadged ViewQuest (Intel) and ViewQuest cameras */
+		switch (product) {
+		case 0x0401:
+			sd->subtype = IntelCreateAndShare;
+			break;
+		case 0x0402:
+			sd->subtype = ViewQuestM318B;
+			break;
+		}
+		break;
+	case 0x1776:		/* Arowana */
+/*		switch (product) { */
+/*		case 0x501c: */
+			sd->subtype = Arowana300KCMOSCamera;
+/*			break; */
+/*		} */
+		break;
+	}
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	cam->cam_mode = vga_mode;
+	cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	sd->brightness = sd_ctrls[MY_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[MY_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[MY_COLOR].qctrl.default_value;
+
+	switch (sd->subtype) {
+	case Arowana300KCMOSCamera:
+	case SmileIntlCamera:
+		/* Arowana 300k CMOS Camera data */
+		if (write_vector(gspca_dev, spca501c_arowana_init_data))
+			goto error;
+		break;
+	case MystFromOriUnknownCamera:
+		/* UnKnow Ori CMOS Camera data */
+		if (write_vector(gspca_dev, spca501c_mysterious_open_data))
+			goto error;
+		break;
+	default:
+		/* generic spca501 init data */
+		if (write_vector(gspca_dev, spca501_init_data))
+			goto error;
+		break;
+	}
+	return 0;
+error:
+	return -EINVAL;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	PDEBUG(D_STREAM, "SPCA501 init");
+	switch (sd->subtype) {
+	case ThreeComHomeConnectLite:
+		/* Special handling for 3com data */
+		write_vector(gspca_dev, spca501_3com_open_data);
+		break;
+	case Arowana300KCMOSCamera:
+	case SmileIntlCamera:
+		/* Arowana 300k CMOS Camera data */
+		write_vector(gspca_dev, spca501c_arowana_open_data);
+		break;
+	case MystFromOriUnknownCamera:
+		/* UnKnow  CMOS Camera data */
+		write_vector(gspca_dev, spca501c_mysterious_init_data);
+		break;
+	default:
+		/* Generic 501 open data */
+		write_vector(gspca_dev, spca501_open_data);
+	}
+	PDEBUG(D_STREAM, "Initializing SPCA501 finished");
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int mode;
+
+	/* memorize the wanted pixel format */
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+
+	/* Enable ISO packet machine CTRL reg=2,
+	 * index=1 bitmask=0x2 (bit ordinal 1) */
+	reg_write(dev, SPCA50X_REG_USB, 0x6, 0x94);
+	switch (mode) {
+	case 0: /* 640x480 */
+		reg_write(dev, SPCA50X_REG_USB, 0x07, 0x004a);
+		break;
+	case 1: /* 320x240 */
+		reg_write(dev, SPCA50X_REG_USB, 0x07, 0x104a);
+		break;
+	default:
+/*	case 2:  * 160x120 */
+		reg_write(dev, SPCA50X_REG_USB, 0x07, 0x204a);
+		break;
+	}
+	reg_write(dev, SPCA501_REG_CTLRL, 0x01, 0x02);
+
+	/* HDG atleast the Intel CreateAndShare needs to have one of its
+	 * brightness / contrast / color set otherwise it assumes wath seems
+	 * max contrast. Note that strange enough setting any of these is
+	 * enough to fix the max contrast problem, to be sure we set all 3 */
+	setbrightness(gspca_dev);
+	setcontrast(gspca_dev);
+	setcolors(gspca_dev);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	/* Disable ISO packet
+	 * machine CTRL reg=2, index=1 bitmask=0x0 (bit ordinal 1) */
+	reg_write(gspca_dev->dev, SPCA501_REG_CTLRL, 0x01, 0x00);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+/* this function is called at close time */
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+	reg_write(gspca_dev->dev, SPCA501_REG_CTLRL, 0x05, 0x00);
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			__u8 *data,			/* isoc packet */
+			int len)			/* iso packet length */
+{
+	switch (data[0]) {
+	case 0:				/* start of frame */
+		frame = gspca_frame_add(gspca_dev,
+					LAST_PACKET,
+					frame,
+					data, 0);
+		data += SPCA501_OFFSET_DATA;
+		len -= SPCA501_OFFSET_DATA;
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame,
+				data, len);
+		return;
+	case 0xff:			/* drop */
+/*		gspca_dev->last_packet_type = DISCARD_PACKET; */
+		return;
+	}
+	data++;
+	len--;
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame,
+			data, len);
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcontrast(gspca_dev);
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcolors(gspca_dev);
+	*val = sd->colors;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x040a, 0x0002), DVNM("Kodak DVC-325")},
+	{USB_DEVICE(0x0497, 0xc001), DVNM("Smile International")},
+	{USB_DEVICE(0x0506, 0x00df), DVNM("3Com HomeConnect Lite")},
+	{USB_DEVICE(0x0733, 0x0401), DVNM("Intel Create and Share")},
+	{USB_DEVICE(0x0733, 0x0402), DVNM("ViewQuest M318B")},
+	{USB_DEVICE(0x1776, 0x501c), DVNM("Arowana 300K CMOS Camera")},
+	{USB_DEVICE(0x0000, 0x0000), DVNM("MystFromOri Unknow Camera")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/spca505.c b/drivers/media/video/gspca/spca505.c
new file mode 100644
index 000000000000..5b23518d970d
--- /dev/null
+++ b/drivers/media/video/gspca/spca505.c
@@ -0,0 +1,933 @@
+/*
+ * SPCA505 chip based cameras initialization data
+ *
+ * V4L2 by Jean-Francis Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#define MODULE_NAME "spca505"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SPCA505 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;		/* !! must be the first item */
+
+	int buflen;
+	unsigned char tmpbuf[640 * 480 * 3 / 2]; /* YYUV per line */
+	unsigned char tmpbuf2[640 * 480 * 2];	/* YUYV */
+
+	unsigned char brightness;
+
+	char subtype;
+#define IntelPCCameraPro 0
+#define Nxultra 1
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 255,
+		.step    = 1,
+		.default_value = 127,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_YUYV, 160, 120, 5},
+	{V4L2_PIX_FMT_YUYV, 176, 144, 4},
+	{V4L2_PIX_FMT_YUYV, 320, 240, 2},
+	{V4L2_PIX_FMT_YUYV, 352, 288, 1},
+	{V4L2_PIX_FMT_YUYV, 640, 480, 0},
+};
+
+#define SPCA50X_OFFSET_DATA 10
+
+#define SPCA50X_REG_USB 0x02	/* spca505 501 */
+
+#define SPCA50X_USB_CTRL 0x00	/* spca505 */
+#define SPCA50X_CUSB_ENABLE 0x01 /* spca505 */
+#define SPCA50X_REG_GLOBAL 0x03	/* spca505 */
+#define SPCA50X_GMISC0_IDSEL 0x01 /* Global control device ID select spca505 */
+#define SPCA50X_GLOBAL_MISC0 0x00 /* Global control miscellaneous 0 spca505 */
+
+#define SPCA50X_GLOBAL_MISC1 0x01 /* 505 */
+#define SPCA50X_GLOBAL_MISC3 0x03 /* 505 */
+#define SPCA50X_GMISC3_SAA7113RST 0x20	/* Not sure about this one spca505 */
+
+/*
+ * Data to initialize a SPCA505. Common to the CCD and external modes
+ */
+static __u16 spca505_init_data[][3] = {
+	/* line	   bmRequest,value,index */
+	/* 1819 */
+	{SPCA50X_REG_GLOBAL, SPCA50X_GMISC3_SAA7113RST, SPCA50X_GLOBAL_MISC3},
+	/* Sensor reset */
+	/* 1822 */ {SPCA50X_REG_GLOBAL, 0x00, SPCA50X_GLOBAL_MISC3},
+	/* 1825 */ {SPCA50X_REG_GLOBAL, 0x00, SPCA50X_GLOBAL_MISC1},
+	/* Block USB reset */
+	/* 1828 */ {SPCA50X_REG_GLOBAL, SPCA50X_GMISC0_IDSEL,
+		SPCA50X_GLOBAL_MISC0},
+
+	/* 1831 */ {0x5, 0x01, 0x10},
+					/* Maybe power down some stuff */
+	/* 1834 */ {0x5, 0x0f, 0x11},
+
+	/* Setup internal CCD  ? */
+	/* 1837 */ {0x6, 0x10, 0x08},
+	/* 1840 */ {0x6, 0x00, 0x09},
+	/* 1843 */ {0x6, 0x00, 0x0a},
+	/* 1846 */ {0x6, 0x00, 0x0b},
+	/* 1849 */ {0x6, 0x10, 0x0c},
+	/* 1852 */ {0x6, 0x00, 0x0d},
+	/* 1855 */ {0x6, 0x00, 0x0e},
+	/* 1858 */ {0x6, 0x00, 0x0f},
+	/* 1861 */ {0x6, 0x10, 0x10},
+	/* 1864 */ {0x6, 0x02, 0x11},
+	/* 1867 */ {0x6, 0x00, 0x12},
+	/* 1870 */ {0x6, 0x04, 0x13},
+	/* 1873 */ {0x6, 0x02, 0x14},
+	/* 1876 */ {0x6, 0x8a, 0x51},
+	/* 1879 */ {0x6, 0x40, 0x52},
+	/* 1882 */ {0x6, 0xb6, 0x53},
+	/* 1885 */ {0x6, 0x3d, 0x54},
+	{}
+};
+
+/*
+ * Data to initialize the camera using the internal CCD
+ */
+static __u16 spca505_open_data_ccd[][3] = {
+	/* line	   bmRequest,value,index */
+	/* Internal CCD data set */
+	/* 1891 */ {0x3, 0x04, 0x01},
+	/* This could be a reset */
+	/* 1894 */ {0x3, 0x00, 0x01},
+
+	/* Setup compression and image registers. 0x6 and 0x7 seem to be
+	   related to H&V hold, and are resolution mode specific */
+		/* 1897 */ {0x4, 0x10, 0x01},
+		/* DIFF(0x50), was (0x10) */
+	/* 1900 */ {0x4, 0x00, 0x04},
+	/* 1903 */ {0x4, 0x00, 0x05},
+	/* 1906 */ {0x4, 0x20, 0x06},
+	/* 1909 */ {0x4, 0x20, 0x07},
+
+	/* 1912 */ {0x8, 0x0a, 0x00},
+	/* DIFF (0x4a), was (0xa) */
+
+	/* 1915 */ {0x5, 0x00, 0x10},
+	/* 1918 */ {0x5, 0x00, 0x11},
+	/* 1921 */ {0x5, 0x00, 0x00},
+	/* DIFF not written */
+	/* 1924 */ {0x5, 0x00, 0x01},
+	/* DIFF not written */
+	/* 1927 */ {0x5, 0x00, 0x02},
+	/* DIFF not written */
+	/* 1930 */ {0x5, 0x00, 0x03},
+	/* DIFF not written */
+	/* 1933 */ {0x5, 0x00, 0x04},
+	/* DIFF not written */
+		/* 1936 */ {0x5, 0x80, 0x05},
+		/* DIFF not written */
+		/* 1939 */ {0x5, 0xe0, 0x06},
+		/* DIFF not written */
+		/* 1942 */ {0x5, 0x20, 0x07},
+		/* DIFF not written */
+		/* 1945 */ {0x5, 0xa0, 0x08},
+		/* DIFF not written */
+		/* 1948 */ {0x5, 0x0, 0x12},
+		/* DIFF not written */
+	/* 1951 */ {0x5, 0x02, 0x0f},
+	/* DIFF not written */
+		/* 1954 */ {0x5, 0x10, 0x46},
+		/* DIFF not written */
+		/* 1957 */ {0x5, 0x8, 0x4a},
+		/* DIFF not written */
+
+	/* 1960 */ {0x3, 0x08, 0x03},
+	/* DIFF (0x3,0x28,0x3) */
+	/* 1963 */ {0x3, 0x08, 0x01},
+	/* 1966 */ {0x3, 0x0c, 0x03},
+	/* DIFF not written */
+		/* 1969 */ {0x3, 0x21, 0x00},
+		/* DIFF (0x39) */
+
+/* Extra block copied from init to hopefully ensure CCD is in a sane state */
+	/* 1837 */ {0x6, 0x10, 0x08},
+	/* 1840 */ {0x6, 0x00, 0x09},
+	/* 1843 */ {0x6, 0x00, 0x0a},
+	/* 1846 */ {0x6, 0x00, 0x0b},
+	/* 1849 */ {0x6, 0x10, 0x0c},
+	/* 1852 */ {0x6, 0x00, 0x0d},
+	/* 1855 */ {0x6, 0x00, 0x0e},
+	/* 1858 */ {0x6, 0x00, 0x0f},
+	/* 1861 */ {0x6, 0x10, 0x10},
+	/* 1864 */ {0x6, 0x02, 0x11},
+	/* 1867 */ {0x6, 0x00, 0x12},
+	/* 1870 */ {0x6, 0x04, 0x13},
+	/* 1873 */ {0x6, 0x02, 0x14},
+	/* 1876 */ {0x6, 0x8a, 0x51},
+	/* 1879 */ {0x6, 0x40, 0x52},
+	/* 1882 */ {0x6, 0xb6, 0x53},
+	/* 1885 */ {0x6, 0x3d, 0x54},
+	/* End of extra block */
+
+		/* 1972 */ {0x6, 0x3f, 0x1},
+		/* Block skipped */
+	/* 1975 */ {0x6, 0x10, 0x02},
+	/* 1978 */ {0x6, 0x64, 0x07},
+	/* 1981 */ {0x6, 0x10, 0x08},
+	/* 1984 */ {0x6, 0x00, 0x09},
+	/* 1987 */ {0x6, 0x00, 0x0a},
+	/* 1990 */ {0x6, 0x00, 0x0b},
+	/* 1993 */ {0x6, 0x10, 0x0c},
+	/* 1996 */ {0x6, 0x00, 0x0d},
+	/* 1999 */ {0x6, 0x00, 0x0e},
+	/* 2002 */ {0x6, 0x00, 0x0f},
+	/* 2005 */ {0x6, 0x10, 0x10},
+	/* 2008 */ {0x6, 0x02, 0x11},
+	/* 2011 */ {0x6, 0x00, 0x12},
+	/* 2014 */ {0x6, 0x04, 0x13},
+	/* 2017 */ {0x6, 0x02, 0x14},
+	/* 2020 */ {0x6, 0x8a, 0x51},
+	/* 2023 */ {0x6, 0x40, 0x52},
+	/* 2026 */ {0x6, 0xb6, 0x53},
+	/* 2029 */ {0x6, 0x3d, 0x54},
+	/* 2032 */ {0x6, 0x60, 0x57},
+	/* 2035 */ {0x6, 0x20, 0x58},
+	/* 2038 */ {0x6, 0x15, 0x59},
+	/* 2041 */ {0x6, 0x05, 0x5a},
+
+	/* 2044 */ {0x5, 0x01, 0xc0},
+	/* 2047 */ {0x5, 0x10, 0xcb},
+		/* 2050 */ {0x5, 0x80, 0xc1},
+		/* */
+		/* 2053 */ {0x5, 0x0, 0xc2},
+		/* 4 was 0 */
+	/* 2056 */ {0x5, 0x00, 0xca},
+		/* 2059 */ {0x5, 0x80, 0xc1},
+		/*  */
+	/* 2062 */ {0x5, 0x04, 0xc2},
+	/* 2065 */ {0x5, 0x00, 0xca},
+		/* 2068 */ {0x5, 0x0, 0xc1},
+		/*  */
+	/* 2071 */ {0x5, 0x00, 0xc2},
+	/* 2074 */ {0x5, 0x00, 0xca},
+		/* 2077 */ {0x5, 0x40, 0xc1},
+		/* */
+	/* 2080 */ {0x5, 0x17, 0xc2},
+	/* 2083 */ {0x5, 0x00, 0xca},
+		/* 2086 */ {0x5, 0x80, 0xc1},
+		/* */
+	/* 2089 */ {0x5, 0x06, 0xc2},
+	/* 2092 */ {0x5, 0x00, 0xca},
+		/* 2095 */ {0x5, 0x80, 0xc1},
+		/* */
+	/* 2098 */ {0x5, 0x04, 0xc2},
+	/* 2101 */ {0x5, 0x00, 0xca},
+
+	/* 2104 */ {0x3, 0x4c, 0x3},
+	/* 2107 */ {0x3, 0x18, 0x1},
+
+	/* 2110 */ {0x6, 0x70, 0x51},
+	/* 2113 */ {0x6, 0xbe, 0x53},
+	/* 2116 */ {0x6, 0x71, 0x57},
+	/* 2119 */ {0x6, 0x20, 0x58},
+	/* 2122 */ {0x6, 0x05, 0x59},
+	/* 2125 */ {0x6, 0x15, 0x5a},
+
+	/* 2128 */ {0x4, 0x00, 0x08},
+	/* Compress = OFF (0x1 to turn on) */
+	/* 2131 */ {0x4, 0x12, 0x09},
+	/* 2134 */ {0x4, 0x21, 0x0a},
+	/* 2137 */ {0x4, 0x10, 0x0b},
+	/* 2140 */ {0x4, 0x21, 0x0c},
+	/* 2143 */ {0x4, 0x05, 0x00},
+	/* was 5 (Image Type ? ) */
+	/* 2146 */ {0x4, 0x00, 0x01},
+
+	/* 2149 */ {0x6, 0x3f, 0x01},
+
+	/* 2152 */ {0x4, 0x00, 0x04},
+	/* 2155 */ {0x4, 0x00, 0x05},
+	/* 2158 */ {0x4, 0x40, 0x06},
+	/* 2161 */ {0x4, 0x40, 0x07},
+
+	/* 2164 */ {0x6, 0x1c, 0x17},
+	/* 2167 */ {0x6, 0xe2, 0x19},
+	/* 2170 */ {0x6, 0x1c, 0x1b},
+	/* 2173 */ {0x6, 0xe2, 0x1d},
+	/* 2176 */ {0x6, 0xaa, 0x1f},
+	/* 2179 */ {0x6, 0x70, 0x20},
+
+	/* 2182 */ {0x5, 0x01, 0x10},
+	/* 2185 */ {0x5, 0x00, 0x11},
+	/* 2188 */ {0x5, 0x01, 0x00},
+	/* 2191 */ {0x5, 0x05, 0x01},
+		/* 2194 */ {0x5, 0x00, 0xc1},
+		/* */
+	/* 2197 */ {0x5, 0x00, 0xc2},
+	/* 2200 */ {0x5, 0x00, 0xca},
+
+	/* 2203 */ {0x6, 0x70, 0x51},
+	/* 2206 */ {0x6, 0xbe, 0x53},
+	{}
+};
+
+/*
+   Made by Tomasz Zablocki (skalamandra@poczta.onet.pl)
+ * SPCA505b chip based cameras initialization data
+ *
+ */
+/* jfm */
+#define initial_brightness 0x7f	/* 0x0(white)-0xff(black) */
+/* #define initial_brightness 0x0	//0x0(white)-0xff(black) */
+/*
+ * Data to initialize a SPCA505. Common to the CCD and external modes
+ */
+static __u16 spca505b_init_data[][3] = {
+/* start */
+	{0x02, 0x00, 0x00},		/* init */
+	{0x02, 0x00, 0x01},
+	{0x02, 0x00, 0x02},
+	{0x02, 0x00, 0x03},
+	{0x02, 0x00, 0x04},
+	{0x02, 0x00, 0x05},
+	{0x02, 0x00, 0x06},
+	{0x02, 0x00, 0x07},
+	{0x02, 0x00, 0x08},
+	{0x02, 0x00, 0x09},
+	{0x03, 0x00, 0x00},
+	{0x03, 0x00, 0x01},
+	{0x03, 0x00, 0x02},
+	{0x03, 0x00, 0x03},
+	{0x03, 0x00, 0x04},
+	{0x03, 0x00, 0x05},
+	{0x03, 0x00, 0x06},
+	{0x04, 0x00, 0x00},
+	{0x04, 0x00, 0x02},
+	{0x04, 0x00, 0x04},
+	{0x04, 0x00, 0x05},
+	{0x04, 0x00, 0x06},
+	{0x04, 0x00, 0x07},
+	{0x04, 0x00, 0x08},
+	{0x04, 0x00, 0x09},
+	{0x04, 0x00, 0x0a},
+	{0x04, 0x00, 0x0b},
+	{0x04, 0x00, 0x0c},
+	{0x07, 0x00, 0x00},
+	{0x07, 0x00, 0x03},
+	{0x08, 0x00, 0x00},
+	{0x08, 0x00, 0x01},
+	{0x08, 0x00, 0x02},
+	{0x00, 0x01, 0x00},
+	{0x00, 0x01, 0x01},
+	{0x00, 0x01, 0x34},
+	{0x00, 0x01, 0x35},
+	{0x06, 0x18, 0x08},
+	{0x06, 0xfc, 0x09},
+	{0x06, 0xfc, 0x0a},
+	{0x06, 0xfc, 0x0b},
+	{0x06, 0x18, 0x0c},
+	{0x06, 0xfc, 0x0d},
+	{0x06, 0xfc, 0x0e},
+	{0x06, 0xfc, 0x0f},
+	{0x06, 0x18, 0x10},
+	{0x06, 0xfe, 0x12},
+	{0x06, 0x00, 0x11},
+	{0x06, 0x00, 0x14},
+	{0x06, 0x00, 0x13},
+	{0x06, 0x28, 0x51},
+	{0x06, 0xff, 0x53},
+	{0x02, 0x00, 0x08},
+
+	{0x03, 0x00, 0x03},
+	{0x03, 0x10, 0x03},
+	{}
+};
+
+/*
+ * Data to initialize the camera using the internal CCD
+ */
+static __u16 spca505b_open_data_ccd[][3] = {
+
+/* {0x02,0x00,0x00}, */
+	{0x03, 0x04, 0x01},		/* rst */
+	{0x03, 0x00, 0x01},
+	{0x03, 0x00, 0x00},
+	{0x03, 0x21, 0x00},
+	{0x03, 0x00, 0x04},
+	{0x03, 0x00, 0x03},
+	{0x03, 0x18, 0x03},
+	{0x03, 0x08, 0x01},
+	{0x03, 0x1c, 0x03},
+	{0x03, 0x5c, 0x03},
+	{0x03, 0x5c, 0x03},
+	{0x03, 0x18, 0x01},
+
+/* same as 505 */
+	{0x04, 0x10, 0x01},
+	{0x04, 0x00, 0x04},
+	{0x04, 0x00, 0x05},
+	{0x04, 0x20, 0x06},
+	{0x04, 0x20, 0x07},
+
+	{0x08, 0x0a, 0x00},
+
+	{0x05, 0x00, 0x10},
+	{0x05, 0x00, 0x11},
+	{0x05, 0x00, 0x12},
+	{0x05, 0x6f, 0x00},
+	{0x05, initial_brightness >> 6, 0x00},
+	{0x05, initial_brightness << 2, 0x01},
+	{0x05, 0x00, 0x02},
+	{0x05, 0x01, 0x03},
+	{0x05, 0x00, 0x04},
+	{0x05, 0x03, 0x05},
+	{0x05, 0xe0, 0x06},
+	{0x05, 0x20, 0x07},
+	{0x05, 0xa0, 0x08},
+	{0x05, 0x00, 0x12},
+	{0x05, 0x02, 0x0f},
+	{0x05, 128, 0x14},		/* max exposure off (0=on) */
+	{0x05, 0x01, 0xb0},
+	{0x05, 0x01, 0xbf},
+	{0x03, 0x02, 0x06},
+	{0x05, 0x10, 0x46},
+	{0x05, 0x08, 0x4a},
+
+	{0x06, 0x00, 0x01},
+	{0x06, 0x10, 0x02},
+	{0x06, 0x64, 0x07},
+	{0x06, 0x18, 0x08},
+	{0x06, 0xfc, 0x09},
+	{0x06, 0xfc, 0x0a},
+	{0x06, 0xfc, 0x0b},
+	{0x04, 0x00, 0x01},
+	{0x06, 0x18, 0x0c},
+	{0x06, 0xfc, 0x0d},
+	{0x06, 0xfc, 0x0e},
+	{0x06, 0xfc, 0x0f},
+	{0x06, 0x11, 0x10},		/* contrast */
+	{0x06, 0x00, 0x11},
+	{0x06, 0xfe, 0x12},
+	{0x06, 0x00, 0x13},
+	{0x06, 0x00, 0x14},
+	{0x06, 0x9d, 0x51},
+	{0x06, 0x40, 0x52},
+	{0x06, 0x7c, 0x53},
+	{0x06, 0x40, 0x54},
+	{0x06, 0x02, 0x57},
+	{0x06, 0x03, 0x58},
+	{0x06, 0x15, 0x59},
+	{0x06, 0x05, 0x5a},
+	{0x06, 0x03, 0x56},
+	{0x06, 0x02, 0x3f},
+	{0x06, 0x00, 0x40},
+	{0x06, 0x39, 0x41},
+	{0x06, 0x69, 0x42},
+	{0x06, 0x87, 0x43},
+	{0x06, 0x9e, 0x44},
+	{0x06, 0xb1, 0x45},
+	{0x06, 0xbf, 0x46},
+	{0x06, 0xcc, 0x47},
+	{0x06, 0xd5, 0x48},
+	{0x06, 0xdd, 0x49},
+	{0x06, 0xe3, 0x4a},
+	{0x06, 0xe8, 0x4b},
+	{0x06, 0xed, 0x4c},
+	{0x06, 0xf2, 0x4d},
+	{0x06, 0xf7, 0x4e},
+	{0x06, 0xfc, 0x4f},
+	{0x06, 0xff, 0x50},
+
+	{0x05, 0x01, 0xc0},
+	{0x05, 0x10, 0xcb},
+	{0x05, 0x40, 0xc1},
+	{0x05, 0x04, 0xc2},
+	{0x05, 0x00, 0xca},
+	{0x05, 0x40, 0xc1},
+	{0x05, 0x09, 0xc2},
+	{0x05, 0x00, 0xca},
+	{0x05, 0xc0, 0xc1},
+	{0x05, 0x09, 0xc2},
+	{0x05, 0x00, 0xca},
+	{0x05, 0x40, 0xc1},
+	{0x05, 0x59, 0xc2},
+	{0x05, 0x00, 0xca},
+	{0x04, 0x00, 0x01},
+	{0x05, 0x80, 0xc1},
+	{0x05, 0xec, 0xc2},
+	{0x05, 0x0, 0xca},
+
+	{0x06, 0x02, 0x57},
+	{0x06, 0x01, 0x58},
+	{0x06, 0x15, 0x59},
+	{0x06, 0x0a, 0x5a},
+	{0x06, 0x01, 0x57},
+	{0x06, 0x8a, 0x03},
+	{0x06, 0x0a, 0x6c},
+	{0x06, 0x30, 0x01},
+	{0x06, 0x20, 0x02},
+	{0x06, 0x00, 0x03},
+
+	{0x05, 0x8c, 0x25},
+
+	{0x06, 0x4d, 0x51},		/* maybe saturation (4d) */
+	{0x06, 0x84, 0x53},		/* making green (84) */
+	{0x06, 0x00, 0x57},		/* sharpness (1) */
+	{0x06, 0x18, 0x08},
+	{0x06, 0xfc, 0x09},
+	{0x06, 0xfc, 0x0a},
+	{0x06, 0xfc, 0x0b},
+	{0x06, 0x18, 0x0c},		/* maybe hue (18) */
+	{0x06, 0xfc, 0x0d},
+	{0x06, 0xfc, 0x0e},
+	{0x06, 0xfc, 0x0f},
+	{0x06, 0x18, 0x10},		/* maybe contrast (18) */
+
+	{0x05, 0x01, 0x02},
+
+	{0x04, 0x00, 0x08},		/* compression */
+	{0x04, 0x12, 0x09},
+	{0x04, 0x21, 0x0a},
+	{0x04, 0x10, 0x0b},
+	{0x04, 0x21, 0x0c},
+	{0x04, 0x1d, 0x00},		/* imagetype (1d) */
+	{0x04, 0x41, 0x01},		/* hardware snapcontrol */
+
+	{0x04, 0x00, 0x04},
+	{0x04, 0x00, 0x05},
+	{0x04, 0x10, 0x06},
+	{0x04, 0x10, 0x07},
+	{0x04, 0x40, 0x06},
+	{0x04, 0x40, 0x07},
+	{0x04, 0x00, 0x04},
+	{0x04, 0x00, 0x05},
+
+	{0x06, 0x1c, 0x17},
+	{0x06, 0xe2, 0x19},
+	{0x06, 0x1c, 0x1b},
+	{0x06, 0xe2, 0x1d},
+	{0x06, 0x5f, 0x1f},
+	{0x06, 0x32, 0x20},
+
+	{0x05, initial_brightness >> 6, 0x00},
+	{0x05, initial_brightness << 2, 0x01},
+	{0x05, 0x06, 0xc1},
+	{0x05, 0x58, 0xc2},
+	{0x05, 0x0, 0xca},
+	{0x05, 0x0, 0x11},
+	{}
+};
+
+static int reg_write(struct usb_device *dev,
+		     __u16 reg, __u16 index, __u16 value)
+{
+	int ret;
+
+	ret = usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			reg,
+			USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, NULL, 0, 500);
+	PDEBUG(D_PACK, "reg write: 0x%02x,0x%02x:0x%02x, 0x%x",
+		reg, index, value, ret);
+	if (ret < 0)
+		PDEBUG(D_ERR, "reg write: error %d", ret);
+	return ret;
+}
+
+/* returns: negative is error, pos or zero is data */
+static int reg_read(struct usb_device *dev,
+			__u16 reg,	/* bRequest */
+			__u16 index,	/* wIndex */
+			__u16 length)	/* wLength (1 or 2 only) */
+{
+	int ret;
+	unsigned char buf[4];
+
+	buf[1] = 0;
+	ret = usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			reg,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			(__u16) 0,		/* value */
+			(__u16) index,
+			buf,
+			length,
+			500);			/* timeout */
+	if (ret < 0) {
+		PDEBUG(D_ERR, "reg_read err %d", ret);
+		return -1;
+	}
+	return (buf[1] << 8) + buf[0];
+}
+
+static int write_vector(struct gspca_dev *gspca_dev,
+				__u16 data[][3])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int ret, i = 0;
+
+	while (data[i][0] != 0 || data[i][1] != 0 || data[i][2] != 0) {
+		ret = reg_write(dev, data[i][0], data[i][2], data[i][1]);
+		if (ret < 0) {
+			PDEBUG(D_ERR,
+				"Register write failed for 0x%x,0x%x,0x%x",
+				data[i][0], data[i][1], data[i][2]);
+			return ret;
+		}
+		i++;
+	}
+	return 0;
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+	__u16 vendor;
+	__u16 product;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+	switch (vendor) {
+	case 0x041e:		/* Creative cameras */
+/*		switch (product) { */
+/*		case 0x401d:	 * here505b */
+			sd->subtype = Nxultra;
+/*			break; */
+/*		} */
+		break;
+	case 0x0733:	/* Rebadged ViewQuest (Intel) and ViewQuest cameras */
+/*		switch (product) { */
+/*		case 0x0430: */
+/*		fixme: may be UsbGrabberPV321 BRIDGE_SPCA506 SENSOR_SAA7113 */
+			sd->subtype = IntelPCCameraPro;
+/*			break; */
+/*		} */
+		break;
+	}
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	cam->cam_mode = vga_mode;
+	if (sd->subtype != IntelPCCameraPro)
+		cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	else			/* no 640x480 for IntelPCCameraPro */
+		cam->nmodes = sizeof vga_mode / sizeof vga_mode[0] - 1;
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+
+	if (sd->subtype == Nxultra) {
+		if (write_vector(gspca_dev, spca505b_init_data))
+			return -EIO;
+	} else {
+		if (write_vector(gspca_dev, spca505_init_data))
+			return -EIO;
+	}
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int ret;
+
+	PDEBUG(D_STREAM, "Initializing SPCA505");
+	if (sd->subtype == Nxultra)
+		write_vector(gspca_dev, spca505b_open_data_ccd);
+	else
+		write_vector(gspca_dev, spca505_open_data_ccd);
+	ret = reg_read(gspca_dev->dev, 6, 0x16, 2);
+
+	if (ret < 0) {
+		PDEBUG(D_ERR|D_STREAM,
+		       "register read failed for after vector read err = %d",
+		       ret);
+		return -EIO;
+	}
+	PDEBUG(D_STREAM,
+		"After vector read returns : 0x%x should be 0x0101",
+		ret & 0xffff);
+
+	ret = reg_write(gspca_dev->dev, 6, 0x16, 0x0a);
+	if (ret < 0) {
+		PDEBUG(D_ERR, "register write failed for (6,0xa,0x16) err=%d",
+		       ret);
+		return -EIO;
+	}
+	reg_write(gspca_dev->dev, 5, 0xc2, 18);
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int ret;
+
+	/* necessary because without it we can see stream
+	 * only once after loading module */
+	/* stopping usb registers Tomasz change */
+	reg_write(dev, 0x02, 0x0, 0x0);
+	switch (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+	case 0:
+		reg_write(dev, 0x04, 0x00, 0x00);
+		reg_write(dev, 0x04, 0x06, 0x10);
+		reg_write(dev, 0x04, 0x07, 0x10);
+		break;
+	case 1:
+		reg_write(dev, 0x04, 0x00, 0x01);
+		reg_write(dev, 0x04, 0x06, 0x1a);
+		reg_write(dev, 0x04, 0x07, 0x1a);
+		break;
+	case 2:
+		reg_write(dev, 0x04, 0x00, 0x02);
+		reg_write(dev, 0x04, 0x06, 0x1c);
+		reg_write(dev, 0x04, 0x07, 0x1d);
+		break;
+	case 4:
+		reg_write(dev, 0x04, 0x00, 0x04);
+		reg_write(dev, 0x04, 0x06, 0x34);
+		reg_write(dev, 0x04, 0x07, 0x34);
+		break;
+	default:
+/*	case 5: */
+		reg_write(dev, 0x04, 0x00, 0x05);
+		reg_write(dev, 0x04, 0x06, 0x40);
+		reg_write(dev, 0x04, 0x07, 0x40);
+		break;
+	}
+/* Enable ISO packet machine - should we do this here or in ISOC init ? */
+	ret = reg_write(dev, SPCA50X_REG_USB,
+			 SPCA50X_USB_CTRL,
+			 SPCA50X_CUSB_ENABLE);
+
+/*	reg_write(dev, 0x5, 0x0, 0x0); */
+/*	reg_write(dev, 0x5, 0x0, 0x1); */
+/*	reg_write(dev, 0x5, 0x11, 0x2); */
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	/* Disable ISO packet machine */
+	reg_write(gspca_dev->dev, 0x02, 0x00, 0x00);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+/* this function is called at close time */
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+	/* This maybe reset or power control */
+	reg_write(gspca_dev->dev, 0x03, 0x03, 0x20);
+	reg_write(gspca_dev->dev, 0x03, 0x01, 0x0);
+	reg_write(gspca_dev->dev, 0x03, 0x00, 0x1);
+	reg_write(gspca_dev->dev, 0x05, 0x10, 0x1);
+	reg_write(gspca_dev->dev, 0x05, 0x11, 0xf);
+}
+
+/* convert YYUV per line to YUYV (YUV 4:2:2) */
+static void yyuv_decode(unsigned char *out,
+			unsigned char *in,
+			int width,
+			int height)
+{
+	unsigned char *Ui, *Vi, *yi, *yi1;
+	unsigned char *out1;
+	int i, j;
+
+	yi = in;
+	for (i = height / 2; --i >= 0; ) {
+		out1 = out + width * 2;		/* next line */
+		yi1 = yi + width;
+		Ui = yi1 + width;
+		Vi = Ui + width / 2;
+		for (j = width / 2; --j >= 0; ) {
+			*out++ = 128 + *yi++;
+			*out++ = 128 + *Ui;
+			*out++ = 128 + *yi++;
+			*out++ = 128 + *Vi;
+
+			*out1++ = 128 + *yi1++;
+			*out1++ = 128 + *Ui++;
+			*out1++ = 128 + *yi1++;
+			*out1++ = 128 + *Vi++;
+		}
+		yi += width * 2;
+		out = out1;
+	}
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	switch (data[0]) {
+	case 0:				/* start of frame */
+		if (gspca_dev->last_packet_type == FIRST_PACKET) {
+			yyuv_decode(sd->tmpbuf2, sd->tmpbuf,
+					gspca_dev->width,
+					gspca_dev->height);
+			frame = gspca_frame_add(gspca_dev,
+						LAST_PACKET,
+						frame,
+						sd->tmpbuf2,
+						gspca_dev->width
+							* gspca_dev->height
+							* 2);
+		}
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame,
+				data, 0);
+		data += SPCA50X_OFFSET_DATA;
+		len -= SPCA50X_OFFSET_DATA;
+		if (len > 0)
+			memcpy(sd->tmpbuf, data, len);
+		else
+			len = 0;
+		sd->buflen = len;
+		return;
+	case 0xff:			/* drop */
+/*		gspca_dev->last_packet_type = DISCARD_PACKET; */
+		return;
+	}
+	data += 1;
+	len -= 1;
+	memcpy(&sd->tmpbuf[sd->buflen], data, len);
+	sd->buflen += len;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	__u8 brightness = sd->brightness;
+	reg_write(gspca_dev->dev, 5, 0x00, (255 - brightness) >> 6);
+	reg_write(gspca_dev->dev, 5, 0x01, (255 - brightness) << 2);
+
+}
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = 255
+		- ((reg_read(gspca_dev->dev, 5, 0x01, 1) >> 2)
+			+ (reg_read(gspca_dev->dev, 5, 0x0, 1) << 6));
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x041e, 0x401d), DVNM("Creative Webcam NX ULTRA")},
+	{USB_DEVICE(0x0733, 0x0430), DVNM("Intel PC Camera Pro")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof (struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/spca506.c b/drivers/media/video/gspca/spca506.c
new file mode 100644
index 000000000000..614fb3ad7711
--- /dev/null
+++ b/drivers/media/video/gspca/spca506.c
@@ -0,0 +1,830 @@
+/*
+ * SPCA506 chip based cameras function
+ * M Xhaard 15/04/2004 based on different work Mark Taylor and others
+ * and my own snoopy file on a pv-321c donate by a german compagny
+ *                "Firma Frank Gmbh" from  Saarbruecken
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "spca506"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SPCA506 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;		/* !! must be the first item */
+
+	int buflen;
+	unsigned char tmpbuf[640 * 480 * 3];	/* YYUV per line */
+	unsigned char tmpbuf2[640 * 480 * 2];	/* YUYV */
+
+	unsigned char brightness;
+	unsigned char contrast;
+	unsigned char colors;
+	unsigned char hue;
+	char norme;
+	char channel;
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_sethue(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_gethue(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0x80,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0x47,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define SD_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Saturation",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0x40,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+#define SD_HUE 3
+	{
+	    {
+		.id      = V4L2_CID_HUE,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Hue",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0,
+	    },
+	    .set = sd_sethue,
+	    .get = sd_gethue,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_YUYV, 160, 120, 5},
+	{V4L2_PIX_FMT_YUYV, 176, 144, 4},
+	{V4L2_PIX_FMT_YUYV, 320, 240, 2},
+	{V4L2_PIX_FMT_YUYV, 352, 288, 1},
+	{V4L2_PIX_FMT_YUYV, 640, 480, 0},
+};
+
+#define SPCA50X_OFFSET_DATA 10
+
+#define SAA7113_bright 0x0a	/* defaults 0x80 */
+#define SAA7113_contrast 0x0b	/* defaults 0x47 */
+#define SAA7113_saturation 0x0c	/* defaults 0x40 */
+#define SAA7113_hue 0x0d	/* defaults 0x00 */
+#define SAA7113_I2C_BASE_WRITE 0x4a
+
+static void reg_r(struct usb_device *dev,
+		  __u16 req,
+		  __u16 index,
+		  __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			req,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,		/* value */
+			index, buffer, length,
+			500);
+}
+
+static void reg_w(struct usb_device *dev,
+		  __u16 req,
+		  __u16 value,
+		  __u16 index)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			req,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index,
+			NULL, 0, 500);
+}
+
+static void spca506_Initi2c(struct gspca_dev *gspca_dev)
+{
+	reg_w(gspca_dev->dev, 0x07, SAA7113_I2C_BASE_WRITE, 0x0004);
+}
+
+static void spca506_WriteI2c(struct gspca_dev *gspca_dev, __u16 valeur,
+			     __u16 reg)
+{
+	int retry = 60;
+	unsigned char Data[2];
+
+	reg_w(gspca_dev->dev, 0x07, reg, 0x0001);
+	reg_w(gspca_dev->dev, 0x07, valeur, 0x0000);
+	while (retry--) {
+		reg_r(gspca_dev->dev, 0x07, 0x0003, Data, 2);
+		if ((Data[0] | Data[1]) == 0x00)
+			break;
+	}
+}
+
+static int spca506_ReadI2c(struct gspca_dev *gspca_dev, __u16 reg)
+{
+	int retry = 60;
+	unsigned char Data[2];
+	unsigned char value;
+
+	reg_w(gspca_dev->dev, 0x07, SAA7113_I2C_BASE_WRITE, 0x0004);
+	reg_w(gspca_dev->dev, 0x07, reg, 0x0001);
+	reg_w(gspca_dev->dev, 0x07, 0x01, 0x0002);
+	while (--retry) {
+		reg_r(gspca_dev->dev, 0x07, 0x0003, Data, 2);
+		if ((Data[0] | Data[1]) == 0x00)
+			break;
+	}
+	if (retry == 0)
+		return -1;
+	reg_r(gspca_dev->dev, 0x07, 0x0000, &value, 1);
+	return value;
+}
+
+static void spca506_SetNormeInput(struct gspca_dev *gspca_dev,
+				 __u16 norme,
+				 __u16 channel)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+/* fixme: check if channel == 0..3 and 6..9 (8 values) */
+	__u8 setbit0 = 0x00;
+	__u8 setbit1 = 0x00;
+	__u8 videomask = 0x00;
+
+	PDEBUG(D_STREAM, "** Open Set Norme **");
+	spca506_Initi2c(gspca_dev);
+	/* NTSC bit0 -> 1(525 l) PAL SECAM bit0 -> 0 (625 l) */
+	/* Composite channel bit1 -> 1 S-video bit 1 -> 0 */
+	/* and exclude SAA7113 reserved channel set default 0 otherwise */
+	if (norme & V4L2_STD_NTSC)
+		setbit0 = 0x01;
+	if (channel == 4 || channel == 5 || channel > 9)
+		channel = 0;
+	if (channel < 4)
+		setbit1 = 0x02;
+	videomask = (0x48 | setbit0 | setbit1);
+	reg_w(gspca_dev->dev, 0x08, videomask, 0x0000);
+	spca506_WriteI2c(gspca_dev, (0xc0 | (channel & 0x0F)), 0x02);
+
+	if (norme & V4L2_STD_NTSC)
+		spca506_WriteI2c(gspca_dev, 0x33, 0x0e);
+					/* Chrominance Control NTSC N */
+	else if (norme & V4L2_STD_SECAM)
+		spca506_WriteI2c(gspca_dev, 0x53, 0x0e);
+					/* Chrominance Control SECAM */
+	else
+		spca506_WriteI2c(gspca_dev, 0x03, 0x0e);
+					/* Chrominance Control PAL BGHIV */
+
+	sd->norme = norme;
+	sd->channel = channel;
+	PDEBUG(D_STREAM, "Set Video Byte to 0x%2x", videomask);
+	PDEBUG(D_STREAM, "Set Norme: %08x Channel %d", norme, channel);
+}
+
+static void spca506_GetNormeInput(struct gspca_dev *gspca_dev,
+				  __u16 *norme, __u16 *channel)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	/* Read the register is not so good value change so
+	   we use your own copy in spca50x struct */
+	*norme = sd->norme;
+	*channel = sd->channel;
+	PDEBUG(D_STREAM, "Get Norme: %d Channel %d", *norme, *channel);
+}
+
+static void spca506_Setsize(struct gspca_dev *gspca_dev, __u16 code,
+			    __u16 xmult, __u16 ymult)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	PDEBUG(D_STREAM, "** SetSize **");
+	reg_w(dev, 0x04, (0x18 | (code & 0x07)), 0x0000);
+	/* Soft snap 0x40 Hard 0x41 */
+	reg_w(dev, 0x04, 0x41, 0x0001);
+	reg_w(dev, 0x04, 0x00, 0x0002);
+	/* reserved */
+	reg_w(dev, 0x04, 0x00, 0x0003);
+
+	/* reserved */
+	reg_w(dev, 0x04, 0x00, 0x0004);
+	/* reserved */
+	reg_w(dev, 0x04, 0x01, 0x0005);
+	/* reserced */
+	reg_w(dev, 0x04, xmult, 0x0006);
+	/* reserved */
+	reg_w(dev, 0x04, ymult, 0x0007);
+	/* compression 1 */
+	reg_w(dev, 0x04, 0x00, 0x0008);
+	/* T=64 -> 2 */
+	reg_w(dev, 0x04, 0x00, 0x0009);
+	/* threshold2D */
+	reg_w(dev, 0x04, 0x21, 0x000a);
+	/* quantization */
+	reg_w(dev, 0x04, 0x00, 0x000b);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	cam->cam_mode = vga_mode;
+	cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	sd->hue = sd_ctrls[SD_HUE].qctrl.default_value;
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	reg_w(dev, 0x03, 0x00, 0x0004);
+	reg_w(dev, 0x03, 0xFF, 0x0003);
+	reg_w(dev, 0x03, 0x00, 0x0000);
+	reg_w(dev, 0x03, 0x1c, 0x0001);
+	reg_w(dev, 0x03, 0x18, 0x0001);
+	/* Init on PAL and composite input0 */
+	spca506_SetNormeInput(gspca_dev, 0, 0);
+	reg_w(dev, 0x03, 0x1c, 0x0001);
+	reg_w(dev, 0x03, 0x18, 0x0001);
+	reg_w(dev, 0x05, 0x00, 0x0000);
+	reg_w(dev, 0x05, 0xef, 0x0001);
+	reg_w(dev, 0x05, 0x00, 0x00c1);
+	reg_w(dev, 0x05, 0x00, 0x00c2);
+	reg_w(dev, 0x06, 0x18, 0x0002);
+	reg_w(dev, 0x06, 0xf5, 0x0011);
+	reg_w(dev, 0x06, 0x02, 0x0012);
+	reg_w(dev, 0x06, 0xfb, 0x0013);
+	reg_w(dev, 0x06, 0x00, 0x0014);
+	reg_w(dev, 0x06, 0xa4, 0x0051);
+	reg_w(dev, 0x06, 0x40, 0x0052);
+	reg_w(dev, 0x06, 0x71, 0x0053);
+	reg_w(dev, 0x06, 0x40, 0x0054);
+	/************************************************/
+	reg_w(dev, 0x03, 0x00, 0x0004);
+	reg_w(dev, 0x03, 0x00, 0x0003);
+	reg_w(dev, 0x03, 0x00, 0x0004);
+	reg_w(dev, 0x03, 0xFF, 0x0003);
+	reg_w(dev, 0x02, 0x00, 0x0000);
+	reg_w(dev, 0x03, 0x60, 0x0000);
+	reg_w(dev, 0x03, 0x18, 0x0001);
+	/* for a better reading mx :)	  */
+	/*sdca506_WriteI2c(value,register) */
+	spca506_Initi2c(gspca_dev);
+	spca506_WriteI2c(gspca_dev, 0x08, 0x01);
+	spca506_WriteI2c(gspca_dev, 0xc0, 0x02);
+						/* input composite video */
+	spca506_WriteI2c(gspca_dev, 0x33, 0x03);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x04);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x05);
+	spca506_WriteI2c(gspca_dev, 0x0d, 0x06);
+	spca506_WriteI2c(gspca_dev, 0xf0, 0x07);
+	spca506_WriteI2c(gspca_dev, 0x98, 0x08);
+	spca506_WriteI2c(gspca_dev, 0x03, 0x09);
+	spca506_WriteI2c(gspca_dev, 0x80, 0x0a);
+	spca506_WriteI2c(gspca_dev, 0x47, 0x0b);
+	spca506_WriteI2c(gspca_dev, 0x48, 0x0c);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x0d);
+	spca506_WriteI2c(gspca_dev, 0x03, 0x0e);	/* Chroma Pal adjust */
+	spca506_WriteI2c(gspca_dev, 0x2a, 0x0f);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x10);
+	spca506_WriteI2c(gspca_dev, 0x0c, 0x11);
+	spca506_WriteI2c(gspca_dev, 0xb8, 0x12);
+	spca506_WriteI2c(gspca_dev, 0x01, 0x13);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x14);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x15);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x16);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x17);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x18);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x19);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1a);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1b);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1c);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1d);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1e);
+	spca506_WriteI2c(gspca_dev, 0xa1, 0x1f);
+	spca506_WriteI2c(gspca_dev, 0x02, 0x40);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x41);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x42);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x43);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x44);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x45);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x46);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x47);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x48);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x49);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4a);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4b);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4c);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4d);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4e);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4f);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x50);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x51);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x52);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x53);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x54);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x55);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x56);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x57);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x58);
+	spca506_WriteI2c(gspca_dev, 0x54, 0x59);
+	spca506_WriteI2c(gspca_dev, 0x07, 0x5a);
+	spca506_WriteI2c(gspca_dev, 0x83, 0x5b);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5c);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5d);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5e);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5f);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x60);
+	spca506_WriteI2c(gspca_dev, 0x05, 0x61);
+	spca506_WriteI2c(gspca_dev, 0x9f, 0x62);
+	PDEBUG(D_STREAM, "** Close Init *");
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u16 norme;
+	__u16 channel;
+	__u8 Data[2];
+
+	/**************************************/
+	reg_w(dev, 0x03, 0x00, 0x0004);
+	reg_w(dev, 0x03, 0x00, 0x0003);
+	reg_w(dev, 0x03, 0x00, 0x0004);
+	reg_w(dev, 0x03, 0xFF, 0x0003);
+	reg_w(dev, 0x02, 0x00, 0x0000);
+	reg_w(dev, 0x03, 0x60, 0x0000);
+	reg_w(dev, 0x03, 0x18, 0x0001);
+
+	/*sdca506_WriteI2c(value,register) */
+	spca506_Initi2c(gspca_dev);
+	spca506_WriteI2c(gspca_dev, 0x08, 0x01);	/* Increment Delay */
+/*	spca506_WriteI2c(gspca_dev, 0xc0, 0x02); * Analog Input Control 1 */
+	spca506_WriteI2c(gspca_dev, 0x33, 0x03);
+						/* Analog Input Control 2 */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x04);
+						/* Analog Input Control 3 */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x05);
+						/* Analog Input Control 4 */
+	spca506_WriteI2c(gspca_dev, 0x0d, 0x06);
+					/* Horizontal Sync Start 0xe9-0x0d */
+	spca506_WriteI2c(gspca_dev, 0xf0, 0x07);
+					/* Horizontal Sync Stop  0x0d-0xf0 */
+
+	spca506_WriteI2c(gspca_dev, 0x98, 0x08);	/* Sync Control */
+/*		Defaults value			*/
+	spca506_WriteI2c(gspca_dev, 0x03, 0x09);	/* Luminance Control */
+	spca506_WriteI2c(gspca_dev, 0x80, 0x0a);
+						/* Luminance Brightness */
+	spca506_WriteI2c(gspca_dev, 0x47, 0x0b);	/* Luminance Contrast */
+	spca506_WriteI2c(gspca_dev, 0x48, 0x0c);
+						/* Chrominance Saturation */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x0d);
+						/* Chrominance Hue Control */
+	spca506_WriteI2c(gspca_dev, 0x2a, 0x0f);
+						/* Chrominance Gain Control */
+	/**************************************/
+	spca506_WriteI2c(gspca_dev, 0x00, 0x10);
+						/* Format/Delay Control */
+	spca506_WriteI2c(gspca_dev, 0x0c, 0x11);	/* Output Control 1 */
+	spca506_WriteI2c(gspca_dev, 0xb8, 0x12);	/* Output Control 2 */
+	spca506_WriteI2c(gspca_dev, 0x01, 0x13);	/* Output Control 3 */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x14);	/* reserved */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x15);	/* VGATE START */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x16);	/* VGATE STOP */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x17);    /* VGATE Control (MSB) */
+	spca506_WriteI2c(gspca_dev, 0x00, 0x18);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x19);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1a);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1b);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1c);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1d);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x1e);
+	spca506_WriteI2c(gspca_dev, 0xa1, 0x1f);
+	spca506_WriteI2c(gspca_dev, 0x02, 0x40);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x41);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x42);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x43);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x44);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x45);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x46);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x47);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x48);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x49);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4a);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4b);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4c);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4d);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4e);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x4f);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x50);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x51);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x52);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x53);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x54);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x55);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x56);
+	spca506_WriteI2c(gspca_dev, 0xff, 0x57);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x58);
+	spca506_WriteI2c(gspca_dev, 0x54, 0x59);
+	spca506_WriteI2c(gspca_dev, 0x07, 0x5a);
+	spca506_WriteI2c(gspca_dev, 0x83, 0x5b);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5c);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5d);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5e);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x5f);
+	spca506_WriteI2c(gspca_dev, 0x00, 0x60);
+	spca506_WriteI2c(gspca_dev, 0x05, 0x61);
+	spca506_WriteI2c(gspca_dev, 0x9f, 0x62);
+	/**************************************/
+	reg_w(dev, 0x05, 0x00, 0x0003);
+	reg_w(dev, 0x05, 0x00, 0x0004);
+	reg_w(dev, 0x03, 0x10, 0x0001);
+	reg_w(dev, 0x03, 0x78, 0x0000);
+	switch (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+	case 0:
+		spca506_Setsize(gspca_dev, 0, 0x10, 0x10);
+		break;
+	case 1:
+		spca506_Setsize(gspca_dev, 1, 0x1a, 0x1a);
+		break;
+	case 2:
+		spca506_Setsize(gspca_dev, 2, 0x1c, 0x1c);
+		break;
+	case 4:
+		spca506_Setsize(gspca_dev, 4, 0x34, 0x34);
+		break;
+	default:
+/*	case 5: */
+		spca506_Setsize(gspca_dev, 5, 0x40, 0x40);
+		break;
+	}
+
+	/* compress setting and size */
+	/* set i2c luma */
+	reg_w(dev, 0x02, 0x01, 0x0000);
+	reg_w(dev, 0x03, 0x12, 0x000);
+	reg_r(dev, 0x04, 0x0001, Data, 2);
+	PDEBUG(D_STREAM, "webcam started");
+	spca506_GetNormeInput(gspca_dev, &norme, &channel);
+	spca506_SetNormeInput(gspca_dev, norme, channel);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	reg_w(dev, 0x02, 0x00, 0x0000);
+	reg_w(dev, 0x03, 0x00, 0x0004);
+	reg_w(dev, 0x03, 0x00, 0x0003);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+/* convert YYUV per line to YUYV (YUV 4:2:2) */
+static void yyuv_decode(unsigned char *out,
+			unsigned char *in,
+			int width,
+			int height)
+{
+	unsigned char *Ui, *Vi, *yi, *yi1;
+	unsigned char *out1;
+	int i, j;
+
+	yi = in;
+	for (i = height / 2; --i >= 0; ) {
+		out1 = out + width * 2;		/* next line */
+		yi1 = yi + width;
+		Ui = yi1 + width;
+		Vi = Ui + width / 2;
+		for (j = width / 2; --j >= 0; ) {
+			*out++ = 128 + *yi++;
+			*out++ = 128 + *Ui;
+			*out++ = 128 + *yi++;
+			*out++ = 128 + *Vi;
+
+			*out1++ = 128 + *yi1++;
+			*out1++ = 128 + *Ui++;
+			*out1++ = 128 + *yi1++;
+			*out1++ = 128 + *Vi++;
+		}
+		yi += width * 2;
+		out = out1;
+	}
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	switch (data[0]) {
+	case 0:				/* start of frame */
+		if (gspca_dev->last_packet_type == FIRST_PACKET) {
+			yyuv_decode(sd->tmpbuf2, sd->tmpbuf,
+					gspca_dev->width,
+					gspca_dev->height);
+			frame = gspca_frame_add(gspca_dev,
+						LAST_PACKET,
+						frame,
+						sd->tmpbuf2,
+						gspca_dev->width
+							* gspca_dev->height
+							* 2);
+		}
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame,
+				data, 0);
+		data += SPCA50X_OFFSET_DATA;
+		len -= SPCA50X_OFFSET_DATA;
+		if (len > 0)
+			memcpy(sd->tmpbuf, data, len);
+		else
+			len = 0;
+		sd->buflen = len;
+		return;
+	case 0xff:			/* drop */
+/*		gspca_dev->last_packet_type = DISCARD_PACKET; */
+		return;
+	}
+	data += 1;
+	len -= 1;
+	memcpy(&sd->tmpbuf[sd->buflen], data, len);
+	sd->buflen += len;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	spca506_Initi2c(gspca_dev);
+	spca506_WriteI2c(gspca_dev, sd->brightness, SAA7113_bright);
+	spca506_WriteI2c(gspca_dev, 0x01, 0x09);
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = spca506_ReadI2c(gspca_dev, SAA7113_bright);
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	spca506_Initi2c(gspca_dev);
+	spca506_WriteI2c(gspca_dev, sd->contrast, SAA7113_contrast);
+	spca506_WriteI2c(gspca_dev, 0x01, 0x09);
+}
+
+static void getcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = spca506_ReadI2c(gspca_dev, SAA7113_contrast);
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	spca506_Initi2c(gspca_dev);
+	spca506_WriteI2c(gspca_dev, sd->colors, SAA7113_saturation);
+	spca506_WriteI2c(gspca_dev, 0x01, 0x09);
+}
+
+static void getcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = spca506_ReadI2c(gspca_dev, SAA7113_saturation);
+}
+
+static void sethue(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	spca506_Initi2c(gspca_dev);
+	spca506_WriteI2c(gspca_dev, sd->hue, SAA7113_hue);
+	spca506_WriteI2c(gspca_dev, 0x01, 0x09);
+}
+
+static void gethue(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->hue = spca506_ReadI2c(gspca_dev, SAA7113_hue);
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcontrast(gspca_dev);
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcolors(gspca_dev);
+	*val = sd->colors;
+	return 0;
+}
+
+static int sd_sethue(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->hue = val;
+	if (gspca_dev->streaming)
+		sethue(gspca_dev);
+	return 0;
+}
+
+static int sd_gethue(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	gethue(gspca_dev);
+	*val = sd->hue;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x06e1, 0xa190), DVNM("ADS Instant VCD")},
+/*	{USB_DEVICE(0x0733, 0x0430), DVNM("UsbGrabber PV321c")}, */
+	{USB_DEVICE(0x0734, 0x043b), DVNM("3DeMon USB Capture aka")},
+	{USB_DEVICE(0x99fa, 0x8988), DVNM("Grandtec V.cap")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/spca508.c b/drivers/media/video/gspca/spca508.c
new file mode 100644
index 000000000000..566adf41f59f
--- /dev/null
+++ b/drivers/media/video/gspca/spca508.c
@@ -0,0 +1,1774 @@
+/*
+ * SPCA508 chip based cameras subdriver
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "spca508"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SPCA508 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;		/* !! must be the first item */
+
+	int buflen;
+	unsigned char tmpbuf[352 * 288 * 3 / 2]; /* YUVY per line */
+	unsigned char tmpbuf2[352 * 288 * 2];	/* YUYV */
+
+	unsigned char brightness;
+
+	char subtype;
+#define CreativeVista 0
+#define HamaUSBSightcam 1
+#define HamaUSBSightcam2 2
+#define IntelEasyPCCamera 3
+#define MicroInnovationIC200 4
+#define ViewQuestVQ110 5
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0x80,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+};
+
+static struct cam_mode sif_mode[] = {
+	{V4L2_PIX_FMT_YUYV, 160, 120, 3},
+	{V4L2_PIX_FMT_YUYV, 176, 144, 2},
+	{V4L2_PIX_FMT_YUYV, 320, 240, 1},
+	{V4L2_PIX_FMT_YUYV, 352, 288, 0},
+};
+
+/* Frame packet header offsets for the spca508 */
+#define SPCA508_OFFSET_TYPE 1
+#define SPCA508_OFFSET_COMPRESS 2
+#define SPCA508_OFFSET_FRAMSEQ 8
+#define SPCA508_OFFSET_WIN1LUM 11
+#define SPCA508_OFFSET_DATA 37
+
+#define SPCA508_SNAPBIT 0x20
+#define SPCA508_SNAPCTRL 0x40
+/*************** I2c ****************/
+#define SPCA508_INDEX_I2C_BASE 0x8800
+
+/*
+ * Initialization data: this is the first set-up data written to the
+ * device (before the open data).
+ */
+static __u16 spca508_init_data[][3] =
+#define IGN(x)			/* nothing */
+{
+	/*  line   URB      value, index */
+	/* 44274  1804 */ {0x0000, 0x870b},
+
+	/* 44299  1805 */ {0x0020, 0x8112},
+	/* Video drop enable, ISO streaming disable */
+	/* 44324  1806 */ {0x0003, 0x8111},
+	/* Reset compression & memory */
+	/* 44349  1807 */ {0x0000, 0x8110},
+	/* Disable all outputs */
+	/* 44372  1808 */ /* READ {0x0000, 0x8114} -> 0000: 00  */
+	/* 44398  1809 */ {0x0000, 0x8114},
+	/* SW GPIO data */
+	/* 44423  1810 */ {0x0008, 0x8110},
+	/* Enable charge pump output */
+	/* 44527  1811 */ {0x0002, 0x8116},
+	/* 200 kHz pump clock */
+	/* 44555  1812 */
+		/* UNKNOWN DIRECTION (URB_FUNCTION_SELECT_INTERFACE:) */
+	/* 44590  1813 */ {0x0003, 0x8111},
+	/* Reset compression & memory */
+	/* 44615  1814 */ {0x0000, 0x8111},
+	/* Normal mode (not reset) */
+	/* 44640  1815 */ {0x0098, 0x8110},
+	/* Enable charge pump output, sync.serial,external 2x clock */
+	/* 44665  1816 */ {0x000d, 0x8114},
+	/* SW GPIO data */
+	/* 44690  1817 */ {0x0002, 0x8116},
+	/* 200 kHz pump clock */
+	/* 44715  1818 */ {0x0020, 0x8112},
+	/* Video drop enable, ISO streaming disable */
+/* --------------------------------------- */
+	/* 44740  1819 */ {0x000f, 0x8402},
+	/* memory bank */
+	/* 44765  1820 */ {0x0000, 0x8403},
+	/* ... address */
+/* --------------------------------------- */
+/* 0x88__ is Synchronous Serial Interface. */
+/* TBD: This table could be expressed more compactly */
+/* using spca508_write_i2c_vector(). */
+/* TBD: Should see if the values in spca50x_i2c_data */
+/* would work with the VQ110 instead of the values */
+/* below. */
+	/* 44790  1821 */ {0x00c0, 0x8804},
+	/* SSI slave addr */
+	/* 44815  1822 */ {0x0008, 0x8802},
+	/* 375 Khz SSI clock */
+	/* 44838  1823 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 44862  1824 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 44888  1825 */ {0x0008, 0x8802},
+	/* 375 Khz SSI clock */
+	/* 44913  1826 */ {0x0012, 0x8801},
+	/* SSI reg addr */
+	/* 44938  1827 */ {0x0080, 0x8800},
+	/* SSI data to write */
+	/* 44961  1828 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 44985  1829 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45009  1830 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 45035  1831 */ {0x0008, 0x8802},
+	/* 375 Khz SSI clock */
+	/* 45060  1832 */ {0x0012, 0x8801},
+	/* SSI reg addr */
+	/* 45085  1833 */ {0x0000, 0x8800},
+	/* SSI data to write */
+	/* 45108  1834 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45132  1835 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45156  1836 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 45182  1837 */ {0x0008, 0x8802},
+	/* 375 Khz SSI clock */
+	/* 45207  1838 */ {0x0011, 0x8801},
+	/* SSI reg addr */
+	/* 45232  1839 */ {0x0040, 0x8800},
+	/* SSI data to write */
+	/* 45255  1840 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45279  1841 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45303  1842 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 45329  1843 */ {0x0008, 0x8802},
+	/* 45354  1844 */ {0x0013, 0x8801},
+	/* 45379  1845 */ {0x0000, 0x8800},
+	/* 45402  1846 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45426  1847 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45450  1848 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 45476  1849 */ {0x0008, 0x8802},
+	/* 45501  1850 */ {0x0014, 0x8801},
+	/* 45526  1851 */ {0x0000, 0x8800},
+	/* 45549  1852 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45573  1853 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45597  1854 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 45623  1855 */ {0x0008, 0x8802},
+	/* 45648  1856 */ {0x0015, 0x8801},
+	/* 45673  1857 */ {0x0001, 0x8800},
+	/* 45696  1858 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45720  1859 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45744  1860 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 45770  1861 */ {0x0008, 0x8802},
+	/* 45795  1862 */ {0x0016, 0x8801},
+	/* 45820  1863 */ {0x0003, 0x8800},
+	/* 45843  1864 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45867  1865 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 45891  1866 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 45917  1867 */ {0x0008, 0x8802},
+	/* 45942  1868 */ {0x0017, 0x8801},
+	/* 45967  1869 */ {0x0036, 0x8800},
+	/* 45990  1870 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46014  1871 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46038  1872 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 46064  1873 */ {0x0008, 0x8802},
+	/* 46089  1874 */ {0x0018, 0x8801},
+	/* 46114  1875 */ {0x00ec, 0x8800},
+	/* 46137  1876 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46161  1877 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46185  1878 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 46211  1879 */ {0x0008, 0x8802},
+	/* 46236  1880 */ {0x001a, 0x8801},
+	/* 46261  1881 */ {0x0094, 0x8800},
+	/* 46284  1882 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46308  1883 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46332  1884 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 46358  1885 */ {0x0008, 0x8802},
+	/* 46383  1886 */ {0x001b, 0x8801},
+	/* 46408  1887 */ {0x0000, 0x8800},
+	/* 46431  1888 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46455  1889 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46479  1890 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 46505  1891 */ {0x0008, 0x8802},
+	/* 46530  1892 */ {0x0027, 0x8801},
+	/* 46555  1893 */ {0x00a2, 0x8800},
+	/* 46578  1894 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46602  1895 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46626  1896 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 46652  1897 */ {0x0008, 0x8802},
+	/* 46677  1898 */ {0x0028, 0x8801},
+	/* 46702  1899 */ {0x0040, 0x8800},
+	/* 46725  1900 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46749  1901 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46773  1902 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 46799  1903 */ {0x0008, 0x8802},
+	/* 46824  1904 */ {0x002a, 0x8801},
+	/* 46849  1905 */ {0x0084, 0x8800},
+	/* 46872  1906 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 46896  1907 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00 */
+	/* 46920  1908 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 46946  1909 */ {0x0008, 0x8802},
+	/* 46971  1910 */ {0x002b, 0x8801},
+	/* 46996  1911 */ {0x00a8, 0x8800},
+	/* 47019  1912 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47043  1913 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47067  1914 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 47093  1915 */ {0x0008, 0x8802},
+	/* 47118  1916 */ {0x002c, 0x8801},
+	/* 47143  1917 */ {0x00fe, 0x8800},
+	/* 47166  1918 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47190  1919 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47214  1920 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 47240  1921 */ {0x0008, 0x8802},
+	/* 47265  1922 */ {0x002d, 0x8801},
+	/* 47290  1923 */ {0x0003, 0x8800},
+	/* 47313  1924 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47337  1925 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47361  1926 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 47387  1927 */ {0x0008, 0x8802},
+	/* 47412  1928 */ {0x0038, 0x8801},
+	/* 47437  1929 */ {0x0083, 0x8800},
+	/* 47460  1930 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47484  1931 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47508  1932 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 47534  1933 */ {0x0008, 0x8802},
+	/* 47559  1934 */ {0x0033, 0x8801},
+	/* 47584  1935 */ {0x0081, 0x8800},
+	/* 47607  1936 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47631  1937 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47655  1938 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 47681  1939 */ {0x0008, 0x8802},
+	/* 47706  1940 */ {0x0034, 0x8801},
+	/* 47731  1941 */ {0x004a, 0x8800},
+	/* 47754  1942 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47778  1943 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47802  1944 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 47828  1945 */ {0x0008, 0x8802},
+	/* 47853  1946 */ {0x0039, 0x8801},
+	/* 47878  1947 */ {0x0000, 0x8800},
+	/* 47901  1948 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47925  1949 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 47949  1950 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 47975  1951 */ {0x0008, 0x8802},
+	/* 48000  1952 */ {0x0010, 0x8801},
+	/* 48025  1953 */ {0x00a8, 0x8800},
+	/* 48048  1954 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48072  1955 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48096  1956 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 48122  1957 */ {0x0008, 0x8802},
+	/* 48147  1958 */ {0x0006, 0x8801},
+	/* 48172  1959 */ {0x0058, 0x8800},
+	/* 48195  1960 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48219  1961 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00 */
+	/* 48243  1962 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 48269  1963 */ {0x0008, 0x8802},
+	/* 48294  1964 */ {0x0000, 0x8801},
+	/* 48319  1965 */ {0x0004, 0x8800},
+	/* 48342  1966 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48366  1967 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48390  1968 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 48416  1969 */ {0x0008, 0x8802},
+	/* 48441  1970 */ {0x0040, 0x8801},
+	/* 48466  1971 */ {0x0080, 0x8800},
+	/* 48489  1972 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48513  1973 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48537  1974 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 48563  1975 */ {0x0008, 0x8802},
+	/* 48588  1976 */ {0x0041, 0x8801},
+	/* 48613  1977 */ {0x000c, 0x8800},
+	/* 48636  1978 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48660  1979 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48684  1980 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 48710  1981 */ {0x0008, 0x8802},
+	/* 48735  1982 */ {0x0042, 0x8801},
+	/* 48760  1983 */ {0x000c, 0x8800},
+	/* 48783  1984 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48807  1985 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48831  1986 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 48857  1987 */ {0x0008, 0x8802},
+	/* 48882  1988 */ {0x0043, 0x8801},
+	/* 48907  1989 */ {0x0028, 0x8800},
+	/* 48930  1990 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48954  1991 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 48978  1992 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 49004  1993 */ {0x0008, 0x8802},
+	/* 49029  1994 */ {0x0044, 0x8801},
+	/* 49054  1995 */ {0x0080, 0x8800},
+	/* 49077  1996 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49101  1997 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49125  1998 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 49151  1999 */ {0x0008, 0x8802},
+	/* 49176  2000 */ {0x0045, 0x8801},
+	/* 49201  2001 */ {0x0020, 0x8800},
+	/* 49224  2002 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49248  2003 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49272  2004 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 49298  2005 */ {0x0008, 0x8802},
+	/* 49323  2006 */ {0x0046, 0x8801},
+	/* 49348  2007 */ {0x0020, 0x8800},
+	/* 49371  2008 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49395  2009 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49419  2010 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 49445  2011 */ {0x0008, 0x8802},
+	/* 49470  2012 */ {0x0047, 0x8801},
+	/* 49495  2013 */ {0x0080, 0x8800},
+	/* 49518  2014 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49542  2015 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49566  2016 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 49592  2017 */ {0x0008, 0x8802},
+	/* 49617  2018 */ {0x0048, 0x8801},
+	/* 49642  2019 */ {0x004c, 0x8800},
+	/* 49665  2020 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49689  2021 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49713  2022 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 49739  2023 */ {0x0008, 0x8802},
+	/* 49764  2024 */ {0x0049, 0x8801},
+	/* 49789  2025 */ {0x0084, 0x8800},
+	/* 49812  2026 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49836  2027 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49860  2028 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 49886  2029 */ {0x0008, 0x8802},
+	/* 49911  2030 */ {0x004a, 0x8801},
+	/* 49936  2031 */ {0x0084, 0x8800},
+	/* 49959  2032 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 49983  2033 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 50007  2034 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 50033  2035 */ {0x0008, 0x8802},
+	/* 50058  2036 */ {0x004b, 0x8801},
+	/* 50083  2037 */ {0x0084, 0x8800},
+	/* 50106  2038 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* --------------------------------------- */
+	/* 50132  2039 */ {0x0012, 0x8700},
+	/* Clock speed 48Mhz/(2+2)/2= 6 Mhz */
+	/* 50157  2040 */ {0x0000, 0x8701},
+	/* CKx1 clock delay adj */
+	/* 50182  2041 */ {0x0000, 0x8701},
+	/* CKx1 clock delay adj */
+	/* 50207  2042 */ {0x0001, 0x870c},
+	/* CKOx2 output */
+	/* --------------------------------------- */
+	/* 50232  2043 */ {0x0080, 0x8600},
+	/* Line memory read counter (L) */
+	/* 50257  2044 */ {0x0001, 0x8606},
+	/* reserved */
+	/* 50282  2045 */ {0x0064, 0x8607},
+	/* Line memory read counter (H) 0x6480=25,728 */
+	/* 50307  2046 */ {0x002a, 0x8601},
+	/* CDSP sharp interpolation mode,
+	 *			line sel for color sep, edge enhance enab */
+	/* 50332  2047 */ {0x0000, 0x8602},
+	/* optical black level for user settng = 0 */
+	/* 50357  2048 */ {0x0080, 0x8600},
+	/* Line memory read counter (L) */
+	/* 50382  2049 */ {0x000a, 0x8603},
+	/* optical black level calc mode: auto; optical black offset = 10 */
+	/* 50407  2050 */ {0x00df, 0x865b},
+	/* Horiz offset for valid pixels (L)=0xdf */
+	/* 50432  2051 */ {0x0012, 0x865c},
+	/* Vert offset for valid lines (L)=0x12 */
+
+/* The following two lines seem to be the "wrong" resolution. */
+/* But perhaps these indicate the actual size of the sensor */
+/* rather than the size of the current video mode. */
+	/* 50457  2052 */ {0x0058, 0x865d},
+	/* Horiz valid pixels (*4) (L) = 352 */
+	/* 50482  2053 */ {0x0048, 0x865e},
+	/* Vert valid lines (*4) (L) = 288 */
+
+	/* 50507  2054 */ {0x0015, 0x8608},
+	/* A11 Coef ... */
+	/* 50532  2055 */ {0x0030, 0x8609},
+	/* 50557  2056 */ {0x00fb, 0x860a},
+	/* 50582  2057 */ {0x003e, 0x860b},
+	/* 50607  2058 */ {0x00ce, 0x860c},
+	/* 50632  2059 */ {0x00f4, 0x860d},
+	/* 50657  2060 */ {0x00eb, 0x860e},
+	/* 50682  2061 */ {0x00dc, 0x860f},
+	/* 50707  2062 */ {0x0039, 0x8610},
+	/* 50732  2063 */ {0x0001, 0x8611},
+	/* R offset for white balance ... */
+	/* 50757  2064 */ {0x0000, 0x8612},
+	/* 50782  2065 */ {0x0001, 0x8613},
+	/* 50807  2066 */ {0x0000, 0x8614},
+	/* 50832  2067 */ {0x005b, 0x8651},
+	/* R gain for white balance ... */
+	/* 50857  2068 */ {0x0040, 0x8652},
+	/* 50882  2069 */ {0x0060, 0x8653},
+	/* 50907  2070 */ {0x0040, 0x8654},
+	/* 50932  2071 */ {0x0000, 0x8655},
+	/* 50957  2072 */ {0x0001, 0x863f},
+	/* Fixed gamma correction enable, USB control,
+	 *			 lum filter disable, lum noise clip disable */
+	/* 50982  2073 */ {0x00a1, 0x8656},
+	/* Window1 size 256x256, Windows2 size 64x64,
+	 *		 gamma look-up disable, new edge enhancement enable */
+	/* 51007  2074 */ {0x0018, 0x8657},
+	/* Edge gain high thresh */
+	/* 51032  2075 */ {0x0020, 0x8658},
+	/* Edge gain low thresh */
+	/* 51057  2076 */ {0x000a, 0x8659},
+	/* Edge bandwidth high threshold */
+	/* 51082  2077 */ {0x0005, 0x865a},
+	/* Edge bandwidth low threshold */
+	/* -------------------------------- */
+	/* 51107  2078 */ {0x0030, 0x8112},
+	/* Video drop enable, ISO streaming enable */
+	/* 51130  2079 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 51154  2080 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 51180  2081 */ {0xa908, 0x8802},
+	/* 51205  2082 */ {0x0034, 0x8801},
+	/* SSI reg addr */
+	/* 51230  2083 */ {0x00ca, 0x8800},
+	/* SSI data to write */
+	/* 51253  2084 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 51277  2085 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 51301  2086 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 51327  2087 */ {0x1f08, 0x8802},
+	/* 51352  2088 */ {0x0006, 0x8801},
+	/* 51377  2089 */ {0x0080, 0x8800},
+	/* 51400  2090 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+
+/* ----- Read back coefs we wrote earlier. */
+	/* 51424  2091 */ /* READ { 0, 0x0000, 0x8608 } -> 0000: 15  */
+	/* 51448  2092 */ /* READ { 0, 0x0000, 0x8609 } -> 0000: 30  */
+	/* 51472  2093 */ /* READ { 0, 0x0000, 0x860a } -> 0000: fb  */
+	/* 51496  2094 */ /* READ { 0, 0x0000, 0x860b } -> 0000: 3e  */
+	/* 51520  2095 */ /* READ { 0, 0x0000, 0x860c } -> 0000: ce  */
+	/* 51544  2096 */ /* READ { 0, 0x0000, 0x860d } -> 0000: f4  */
+	/* 51568  2097 */ /* READ { 0, 0x0000, 0x860e } -> 0000: eb  */
+	/* 51592  2098 */ /* READ { 0, 0x0000, 0x860f } -> 0000: dc  */
+	/* 51616  2099 */ /* READ { 0, 0x0000, 0x8610 } -> 0000: 39  */
+	/* 51640  2100 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 51664  2101 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 08  */
+	/* 51690  2102 */ {0xb008, 0x8802},
+	/* 51715  2103 */ {0x0006, 0x8801},
+	/* 51740  2104 */ {0x007d, 0x8800},
+	/* 51763  2105 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+
+
+	/* This chunk is seemingly redundant with */
+	/* earlier commands (A11 Coef...), but if I disable it, */
+	/* the image appears too dark.  Maybe there was some kind of */
+	/* reset since the earlier commands, so this is necessary again. */
+	/* 51789  2106 */ {0x0015, 0x8608},
+	/* 51814  2107 */ {0x0030, 0x8609},
+	/* 51839  2108 */ {0xfffb, 0x860a},
+	/* 51864  2109 */ {0x003e, 0x860b},
+	/* 51889  2110 */ {0xffce, 0x860c},
+	/* 51914  2111 */ {0xfff4, 0x860d},
+	/* 51939  2112 */ {0xffeb, 0x860e},
+	/* 51964  2113 */ {0xffdc, 0x860f},
+	/* 51989  2114 */ {0x0039, 0x8610},
+	/* 52014  2115 */ {0x0018, 0x8657},
+
+	/* 52039  2116 */ {0x0000, 0x8508},
+	/* Disable compression. */
+	/* Previous line was:
+	 * 52039  2116 *  { 0, 0x0021, 0x8508 },  * Enable compression. */
+	/* 52064  2117 */ {0x0032, 0x850b},
+	/* compression stuff */
+	/* 52089  2118 */ {0x0003, 0x8509},
+	/* compression stuff */
+	/* 52114  2119 */ {0x0011, 0x850a},
+	/* compression stuff */
+	/* 52139  2120 */ {0x0021, 0x850d},
+	/* compression stuff */
+	/* 52164  2121 */ {0x0010, 0x850c},
+	/* compression stuff */
+	/* 52189  2122 */ {0x0003, 0x8500},
+	/* *** Video mode: 160x120 */
+	/* 52214  2123 */ {0x0001, 0x8501},
+	/* Hardware-dominated snap control */
+	/* 52239  2124 */ {0x0061, 0x8656},
+	/* Window1 size 128x128, Windows2 size 128x128,
+	 *		gamma look-up disable, new edge enhancement enable */
+	/* 52264  2125 */ {0x0018, 0x8617},
+	/* Window1 start X (*2) */
+	/* 52289  2126 */ {0x0008, 0x8618},
+	/* Window1 start Y (*2) */
+	/* 52314  2127 */ {0x0061, 0x8656},
+	/* Window1 size 128x128, Windows2 size 128x128,
+	 *		gamma look-up disable, new edge enhancement enable */
+	/* 52339  2128 */ {0x0058, 0x8619},
+	/* Window2 start X (*2) */
+	/* 52364  2129 */ {0x0008, 0x861a},
+	/* Window2 start Y (*2) */
+	/* 52389  2130 */ {0x00ff, 0x8615},
+	/* High lum thresh for white balance */
+	/* 52414  2131 */ {0x0000, 0x8616},
+	/* Low lum thresh for white balance */
+	/* 52439  2132 */ {0x0012, 0x8700},
+	/* Clock speed 48Mhz/(2+2)/2= 6 Mhz */
+	/* 52464  2133 */ {0x0012, 0x8700},
+	/* Clock speed 48Mhz/(2+2)/2= 6 Mhz */
+	/* 52487  2134 */ /* READ { 0, 0x0000, 0x8656 } -> 0000: 61  */
+	/* 52513  2135 */ {0x0028, 0x8802},
+	/* 375 Khz SSI clock, SSI r/w sync with VSYNC */
+	/* 52536  2136 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 52560  2137 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 28  */
+	/* 52586  2138 */ {0x1f28, 0x8802},
+	/* 375 Khz SSI clock, SSI r/w sync with VSYNC */
+	/* 52611  2139 */ {0x0010, 0x8801},
+	/* SSI reg addr */
+	/* 52636  2140 */ {0x003e, 0x8800},
+	/* SSI data to write */
+	/* 52659  2141 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 52685  2142 */ {0x0028, 0x8802},
+	/* 52708  2143 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 52732  2144 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 28  */
+	/* 52758  2145 */ {0x1f28, 0x8802},
+	/* 52783  2146 */ {0x0000, 0x8801},
+	/* 52808  2147 */ {0x001f, 0x8800},
+	/* 52831  2148 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 52857  2149 */ {0x0001, 0x8602},
+	/* optical black level for user settning = 1 */
+
+	/* Original: */
+	/* 52882  2150 */ {0x0023, 0x8700},
+	/* Clock speed 48Mhz/(3+2)/4= 2.4 Mhz */
+	/* 52907  2151 */ {0x000f, 0x8602},
+	/* optical black level for user settning = 15 */
+
+	/* 52932  2152 */ {0x0028, 0x8802},
+	/* 52955  2153 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 52979  2154 */ /* READ { 0, 0x0001, 0x8802 } -> 0000: 28  */
+	/* 53005  2155 */ {0x1f28, 0x8802},
+	/* 53030  2156 */ {0x0010, 0x8801},
+	/* 53055  2157 */ {0x007b, 0x8800},
+	/* 53078  2158 */ /* READ { 0, 0x0001, 0x8803 } -> 0000: 00  */
+	/* 53104  2159 */ {0x002f, 0x8651},
+	/* R gain for white balance ... */
+	/* 53129  2160 */ {0x0080, 0x8653},
+	/* 53152  2161 */ /* READ { 0, 0x0000, 0x8655 } -> 0000: 00  */
+	/* 53178  2162 */ {0x0000, 0x8655},
+
+	/* 53203  2163 */ {0x0030, 0x8112},
+	/* Video drop enable, ISO streaming enable */
+	/* 53228  2164 */ {0x0020, 0x8112},
+	/* Video drop enable, ISO streaming disable */
+	/* 53252  2165 */
+	     /* UNKNOWN DIRECTION (URB_FUNCTION_SELECT_INTERFACE: (ALT=0) ) */
+	{0, 0}
+};
+
+
+/*
+ * Initialization data for Intel EasyPC Camera CS110
+ */
+static __u16 spca508cs110_init_data[][3] = {
+	{0x0000, 0x870b}, /* Reset CTL3 */
+	{0x0003, 0x8111}, /* Soft Reset compression, memory, TG & CDSP */
+	{0x0000, 0x8111}, /* Normal operation on reset */
+	{0x0090, 0x8110},
+		 /* External Clock 2x & Synchronous Serial Interface Output */
+	{0x0020, 0x8112}, /* Video Drop packet enable */
+	{0x0000, 0x8114}, /* Software GPIO output data */
+	{0x0001, 0x8114},
+	{0x0001, 0x8114},
+	{0x0001, 0x8114},
+	{0x0003, 0x8114},
+
+	/* Initial sequence Synchronous Serial Interface */
+	{0x000f, 0x8402}, /* Memory bank Address */
+	{0x0000, 0x8403}, /* Memory bank Address */
+	{0x00ba, 0x8804}, /* SSI Slave address */
+	{0x0010, 0x8802}, /* 93.75kHz SSI Clock Two DataByte */
+	{0x0010, 0x8802}, /* 93.75kHz SSI Clock two DataByte */
+
+	{0x0001, 0x8801},
+	{0x000a, 0x8805},/* a - NWG: Dunno what this is about */
+	{0x0000, 0x8800},
+	{0x0010, 0x8802},
+
+	{0x0002, 0x8801},
+	{0x0000, 0x8805},
+	{0x0000, 0x8800},
+	{0x0010, 0x8802},
+
+	{0x0003, 0x8801},
+	{0x0027, 0x8805},
+	{0x0001, 0x8800},
+	{0x0010, 0x8802},
+
+	{0x0004, 0x8801},
+	{0x0065, 0x8805},
+	{0x0001, 0x8800},
+	{0x0010, 0x8802},
+
+	{0x0005, 0x8801},
+	{0x0003, 0x8805},
+	{0x0000, 0x8800},
+	{0x0010, 0x8802},
+
+	{0x0006, 0x8801},
+	{0x001c, 0x8805},
+	{0x0000, 0x8800},
+	{0x0010, 0x8802},
+
+	{0x0007, 0x8801},
+	{0x002a, 0x8805},
+	{0x0000, 0x8800},
+	{0x0010, 0x8802},
+
+	{0x0002, 0x8704}, /* External input CKIx1 */
+	{0x0001, 0x8606}, /* 1 Line memory Read Counter (H) Result: (d)410 */
+	{0x009a, 0x8600}, /* Line memory Read Counter (L) */
+	{0x0001, 0x865b}, /* 1 Horizontal Offset for Valid Pixel(L) */
+	{0x0003, 0x865c}, /* 3 Vertical Offset for Valid Lines(L) */
+	{0x0058, 0x865d}, /* 58 Horizontal Valid Pixel Window(L) */
+
+	{0x0006, 0x8660}, /* Nibble data + input order */
+
+	{0x000a, 0x8602}, /* Optical black level set to 0x0a */
+/* 1945 */ {0x0000, 0x8603}, /* Optical black level Offset */
+
+/* 1962 *  {0, 0x0000, 0x8611},  * 0 R  Offset for white Balance */
+/* 1963 *  {0, 0x0000, 0x8612},  * 1 Gr Offset for white Balance */
+/* 1964 *  {0, 0x0000, 0x8613},  * 1f B  Offset for white Balance */
+/* 1965 *  {0, 0x0000, 0x8614},  * f0 Gb Offset for white Balance */
+
+	{0x0040, 0x8651}, /* 2b BLUE gain for white balance  good at all 60 */
+	{0x0030, 0x8652}, /* 41 Gr Gain for white Balance (L) */
+	{0x0035, 0x8653}, /* 26 RED gain for white balance */
+	{0x0035, 0x8654}, /* 40Gb Gain for white Balance (L) */
+	{0x0041, 0x863f},
+	      /* Fixed Gamma correction enabled (makes colours look better) */
+
+/* 2422 */ {0x0000, 0x8655},
+	/* High bits for white balance*****brightness control*** */
+	{}
+};
+
+static __u16 spca508_sightcam_init_data[][3] = {
+/* This line seems to setup the frame/canvas */
+	/*368  */ {0x000f, 0x8402},
+
+/* Theese 6 lines are needed to startup the webcam */
+	/*398  */ {0x0090, 0x8110},
+	/*399  */ {0x0001, 0x8114},
+	/*400  */ {0x0001, 0x8114},
+	/*401  */ {0x0001, 0x8114},
+	/*402  */ {0x0003, 0x8114},
+	/*403  */ {0x0080, 0x8804},
+
+/* This part seems to make the pictures darker? (autobrightness?) */
+	/*436  */ {0x0001, 0x8801},
+	/*437  */ {0x0004, 0x8800},
+	/*439  */ {0x0003, 0x8801},
+	/*440  */ {0x00e0, 0x8800},
+	/*442  */ {0x0004, 0x8801},
+	/*443  */ {0x00b4, 0x8800},
+	/*445  */ {0x0005, 0x8801},
+	/*446  */ {0x0000, 0x8800},
+
+	/*448  */ {0x0006, 0x8801},
+	/*449  */ {0x00e0, 0x8800},
+	/*451  */ {0x0007, 0x8801},
+	/*452  */ {0x000c, 0x8800},
+
+/* This section is just needed, it probably
+ * does something like the previous section,
+ * but the cam won't start if it's not included.
+ */
+	/*484  */ {0x0014, 0x8801},
+	/*485  */ {0x0008, 0x8800},
+	/*487  */ {0x0015, 0x8801},
+	/*488  */ {0x0067, 0x8800},
+	/*490  */ {0x0016, 0x8801},
+	/*491  */ {0x0000, 0x8800},
+	/*493  */ {0x0017, 0x8801},
+	/*494  */ {0x0020, 0x8800},
+	/*496  */ {0x0018, 0x8801},
+	/*497  */ {0x0044, 0x8800},
+
+/* Makes the picture darker - and the
+ * cam won't start if not included
+ */
+	/*505  */ {0x001e, 0x8801},
+	/*506  */ {0x00ea, 0x8800},
+	/*508  */ {0x001f, 0x8801},
+	/*509  */ {0x0001, 0x8800},
+	/*511  */ {0x0003, 0x8801},
+	/*512  */ {0x00e0, 0x8800},
+
+/* seems to place the colors ontop of each other #1 */
+	/*517  */ {0x0006, 0x8704},
+	/*518  */ {0x0001, 0x870c},
+	/*519  */ {0x0016, 0x8600},
+	/*520  */ {0x0002, 0x8606},
+
+/* if not included the pictures becomes _very_ dark */
+	/*521  */ {0x0064, 0x8607},
+	/*522  */ {0x003a, 0x8601},
+	/*523  */ {0x0000, 0x8602},
+
+/* seems to place the colors ontop of each other #2 */
+	/*524  */ {0x0016, 0x8600},
+	/*525  */ {0x0018, 0x8617},
+	/*526  */ {0x0008, 0x8618},
+	/*527  */ {0x00a1, 0x8656},
+
+/* webcam won't start if not included */
+	/*528  */ {0x0007, 0x865b},
+	/*529  */ {0x0001, 0x865c},
+	/*530  */ {0x0058, 0x865d},
+	/*531  */ {0x0048, 0x865e},
+
+/* adjusts the colors */
+	/*541  */ {0x0049, 0x8651},
+	/*542  */ {0x0040, 0x8652},
+	/*543  */ {0x004c, 0x8653},
+	/*544  */ {0x0040, 0x8654},
+
+	{0, 0}
+};
+
+static __u16 spca508_sightcam2_init_data[][3] = {
+/* 35 */ {0x0020, 0x8112},
+
+/* 36 */ {0x000f, 0x8402},
+/* 37 */ {0x0000, 0x8403},
+
+/* 38 */ {0x0008, 0x8201},
+/* 39 */ {0x0008, 0x8200},
+/* 40 */ {0x0001, 0x8200},
+/* 43 */ {0x0009, 0x8201},
+/* 44 */ {0x0008, 0x8200},
+/* 45 */ {0x0001, 0x8200},
+/* 48 */ {0x000a, 0x8201},
+/* 49 */ {0x0008, 0x8200},
+/* 50 */ {0x0001, 0x8200},
+/* 53 */ {0x000b, 0x8201},
+/* 54 */ {0x0008, 0x8200},
+/* 55 */ {0x0001, 0x8200},
+/* 58 */ {0x000c, 0x8201},
+/* 59 */ {0x0008, 0x8200},
+/* 60 */ {0x0001, 0x8200},
+/* 63 */ {0x000d, 0x8201},
+/* 64 */ {0x0008, 0x8200},
+/* 65 */ {0x0001, 0x8200},
+/* 68 */ {0x000e, 0x8201},
+/* 69 */ {0x0008, 0x8200},
+/* 70 */ {0x0001, 0x8200},
+/* 73 */ {0x0007, 0x8201},
+/* 74 */ {0x0008, 0x8200},
+/* 75 */ {0x0001, 0x8200},
+/* 78 */ {0x000f, 0x8201},
+/* 79 */ {0x0008, 0x8200},
+/* 80 */ {0x0001, 0x8200},
+
+/* 84 */ {0x0018, 0x8660},
+/* 85 */ {0x0010, 0x8201},
+
+/* 86 */ {0x0008, 0x8200},
+/* 87 */ {0x0001, 0x8200},
+/* 90 */ {0x0011, 0x8201},
+/* 91 */ {0x0008, 0x8200},
+/* 92 */ {0x0001, 0x8200},
+
+/* 95 */ {0x0000, 0x86b0},
+/* 96 */ {0x0034, 0x86b1},
+/* 97 */ {0x0000, 0x86b2},
+/* 98 */ {0x0049, 0x86b3},
+/* 99 */ {0x0000, 0x86b4},
+/* 100 */ {0x0000, 0x86b4},
+
+/* 101 */ {0x0012, 0x8201},
+/* 102 */ {0x0008, 0x8200},
+/* 103 */ {0x0001, 0x8200},
+/* 106 */ {0x0013, 0x8201},
+/* 107 */ {0x0008, 0x8200},
+/* 108 */ {0x0001, 0x8200},
+
+/* 111 */ {0x0001, 0x86b0},
+/* 112 */ {0x00aa, 0x86b1},
+/* 113 */ {0x0000, 0x86b2},
+/* 114 */ {0x00e4, 0x86b3},
+/* 115 */ {0x0000, 0x86b4},
+/* 116 */ {0x0000, 0x86b4},
+
+/* 118 */ {0x0018, 0x8660},
+
+/* 119 */ {0x0090, 0x8110},
+/* 120 */ {0x0001, 0x8114},
+/* 121 */ {0x0001, 0x8114},
+/* 122 */ {0x0001, 0x8114},
+/* 123 */ {0x0003, 0x8114},
+
+/* 124 */ {0x0080, 0x8804},
+/* 157 */ {0x0003, 0x8801},
+/* 158 */ {0x0012, 0x8800},
+/* 160 */ {0x0004, 0x8801},
+/* 161 */ {0x0005, 0x8800},
+/* 163 */ {0x0005, 0x8801},
+/* 164 */ {0x0000, 0x8800},
+/* 166 */ {0x0006, 0x8801},
+/* 167 */ {0x0000, 0x8800},
+/* 169 */ {0x0007, 0x8801},
+/* 170 */ {0x0000, 0x8800},
+/* 172 */ {0x0008, 0x8801},
+/* 173 */ {0x0005, 0x8800},
+/* 175 */ {0x000a, 0x8700},
+/* 176 */ {0x000e, 0x8801},
+/* 177 */ {0x0004, 0x8800},
+/* 179 */ {0x0005, 0x8801},
+/* 180 */ {0x0047, 0x8800},
+/* 182 */ {0x0006, 0x8801},
+/* 183 */ {0x0000, 0x8800},
+/* 185 */ {0x0007, 0x8801},
+/* 186 */ {0x00c0, 0x8800},
+/* 188 */ {0x0008, 0x8801},
+/* 189 */ {0x0003, 0x8800},
+/* 191 */ {0x0013, 0x8801},
+/* 192 */ {0x0001, 0x8800},
+/* 194 */ {0x0009, 0x8801},
+/* 195 */ {0x0000, 0x8800},
+/* 197 */ {0x000a, 0x8801},
+/* 198 */ {0x0000, 0x8800},
+/* 200 */ {0x000b, 0x8801},
+/* 201 */ {0x0000, 0x8800},
+/* 203 */ {0x000c, 0x8801},
+/* 204 */ {0x0000, 0x8800},
+/* 206 */ {0x000e, 0x8801},
+/* 207 */ {0x0004, 0x8800},
+/* 209 */ {0x000f, 0x8801},
+/* 210 */ {0x0000, 0x8800},
+/* 212 */ {0x0010, 0x8801},
+/* 213 */ {0x0006, 0x8800},
+/* 215 */ {0x0011, 0x8801},
+/* 216 */ {0x0006, 0x8800},
+/* 218 */ {0x0012, 0x8801},
+/* 219 */ {0x0000, 0x8800},
+/* 221 */ {0x0013, 0x8801},
+/* 222 */ {0x0001, 0x8800},
+
+/* 224 */ {0x000a, 0x8700},
+/* 225 */ {0x0000, 0x8702},
+/* 226 */ {0x0000, 0x8703},
+/* 227 */ {0x00c2, 0x8704},
+/* 228 */ {0x0001, 0x870c},
+
+/* 229 */ {0x0044, 0x8600},
+/* 230 */ {0x0002, 0x8606},
+/* 231 */ {0x0064, 0x8607},
+/* 232 */ {0x003a, 0x8601},
+/* 233 */ {0x0008, 0x8602},
+/* 234 */ {0x0044, 0x8600},
+/* 235 */ {0x0018, 0x8617},
+/* 236 */ {0x0008, 0x8618},
+/* 237 */ {0x00a1, 0x8656},
+/* 238 */ {0x0004, 0x865b},
+/* 239 */ {0x0002, 0x865c},
+/* 240 */ {0x0058, 0x865d},
+/* 241 */ {0x0048, 0x865e},
+/* 242 */ {0x0012, 0x8608},
+/* 243 */ {0x002c, 0x8609},
+/* 244 */ {0x0002, 0x860a},
+/* 245 */ {0x002c, 0x860b},
+/* 246 */ {0x00db, 0x860c},
+/* 247 */ {0x00f9, 0x860d},
+/* 248 */ {0x00f1, 0x860e},
+/* 249 */ {0x00e3, 0x860f},
+/* 250 */ {0x002c, 0x8610},
+/* 251 */ {0x006c, 0x8651},
+/* 252 */ {0x0041, 0x8652},
+/* 253 */ {0x0059, 0x8653},
+/* 254 */ {0x0040, 0x8654},
+/* 255 */ {0x00fa, 0x8611},
+/* 256 */ {0x00ff, 0x8612},
+/* 257 */ {0x00f8, 0x8613},
+/* 258 */ {0x0000, 0x8614},
+/* 259 */ {0x0001, 0x863f},
+/* 260 */ {0x0000, 0x8640},
+/* 261 */ {0x0026, 0x8641},
+/* 262 */ {0x0045, 0x8642},
+/* 263 */ {0x0060, 0x8643},
+/* 264 */ {0x0075, 0x8644},
+/* 265 */ {0x0088, 0x8645},
+/* 266 */ {0x009b, 0x8646},
+/* 267 */ {0x00b0, 0x8647},
+/* 268 */ {0x00c5, 0x8648},
+/* 269 */ {0x00d2, 0x8649},
+/* 270 */ {0x00dc, 0x864a},
+/* 271 */ {0x00e5, 0x864b},
+/* 272 */ {0x00eb, 0x864c},
+/* 273 */ {0x00f0, 0x864d},
+/* 274 */ {0x00f6, 0x864e},
+/* 275 */ {0x00fa, 0x864f},
+/* 276 */ {0x00ff, 0x8650},
+/* 277 */ {0x0060, 0x8657},
+/* 278 */ {0x0010, 0x8658},
+/* 279 */ {0x0018, 0x8659},
+/* 280 */ {0x0005, 0x865a},
+/* 281 */ {0x0018, 0x8660},
+/* 282 */ {0x0003, 0x8509},
+/* 283 */ {0x0011, 0x850a},
+/* 284 */ {0x0032, 0x850b},
+/* 285 */ {0x0010, 0x850c},
+/* 286 */ {0x0021, 0x850d},
+/* 287 */ {0x0001, 0x8500},
+/* 288 */ {0x0000, 0x8508},
+/* 289 */ {0x0012, 0x8608},
+/* 290 */ {0x002c, 0x8609},
+/* 291 */ {0x0002, 0x860a},
+/* 292 */ {0x0039, 0x860b},
+/* 293 */ {0x00d0, 0x860c},
+/* 294 */ {0x00f7, 0x860d},
+/* 295 */ {0x00ed, 0x860e},
+/* 296 */ {0x00db, 0x860f},
+/* 297 */ {0x0039, 0x8610},
+/* 298 */ {0x0012, 0x8657},
+/* 299 */ {0x000c, 0x8619},
+/* 300 */ {0x0004, 0x861a},
+/* 301 */ {0x00a1, 0x8656},
+/* 302 */ {0x00c8, 0x8615},
+/* 303 */ {0x0032, 0x8616},
+
+/* 306 */ {0x0030, 0x8112},
+/* 313 */ {0x0020, 0x8112},
+/* 314 */ {0x0020, 0x8112},
+/* 315 */ {0x000f, 0x8402},
+/* 316 */ {0x0000, 0x8403},
+
+/* 317 */ {0x0090, 0x8110},
+/* 318 */ {0x0001, 0x8114},
+/* 319 */ {0x0001, 0x8114},
+/* 320 */ {0x0001, 0x8114},
+/* 321 */ {0x0003, 0x8114},
+/* 322 */ {0x0080, 0x8804},
+
+/* 355 */ {0x0003, 0x8801},
+/* 356 */ {0x0012, 0x8800},
+/* 358 */ {0x0004, 0x8801},
+/* 359 */ {0x0005, 0x8800},
+/* 361 */ {0x0005, 0x8801},
+/* 362 */ {0x0047, 0x8800},
+/* 364 */ {0x0006, 0x8801},
+/* 365 */ {0x0000, 0x8800},
+/* 367 */ {0x0007, 0x8801},
+/* 368 */ {0x00c0, 0x8800},
+/* 370 */ {0x0008, 0x8801},
+/* 371 */ {0x0003, 0x8800},
+/* 373 */ {0x000a, 0x8700},
+/* 374 */ {0x000e, 0x8801},
+/* 375 */ {0x0004, 0x8800},
+/* 377 */ {0x0005, 0x8801},
+/* 378 */ {0x0047, 0x8800},
+/* 380 */ {0x0006, 0x8801},
+/* 381 */ {0x0000, 0x8800},
+/* 383 */ {0x0007, 0x8801},
+/* 384 */ {0x00c0, 0x8800},
+/* 386 */ {0x0008, 0x8801},
+/* 387 */ {0x0003, 0x8800},
+/* 389 */ {0x0013, 0x8801},
+/* 390 */ {0x0001, 0x8800},
+/* 392 */ {0x0009, 0x8801},
+/* 393 */ {0x0000, 0x8800},
+/* 395 */ {0x000a, 0x8801},
+/* 396 */ {0x0000, 0x8800},
+/* 398 */ {0x000b, 0x8801},
+/* 399 */ {0x0000, 0x8800},
+/* 401 */ {0x000c, 0x8801},
+/* 402 */ {0x0000, 0x8800},
+/* 404 */ {0x000e, 0x8801},
+/* 405 */ {0x0004, 0x8800},
+/* 407 */ {0x000f, 0x8801},
+/* 408 */ {0x0000, 0x8800},
+/* 410 */ {0x0010, 0x8801},
+/* 411 */ {0x0006, 0x8800},
+/* 413 */ {0x0011, 0x8801},
+/* 414 */ {0x0006, 0x8800},
+/* 416 */ {0x0012, 0x8801},
+/* 417 */ {0x0000, 0x8800},
+/* 419 */ {0x0013, 0x8801},
+/* 420 */ {0x0001, 0x8800},
+/* 422 */ {0x000a, 0x8700},
+/* 423 */ {0x0000, 0x8702},
+/* 424 */ {0x0000, 0x8703},
+/* 425 */ {0x00c2, 0x8704},
+/* 426 */ {0x0001, 0x870c},
+/* 427 */ {0x0044, 0x8600},
+/* 428 */ {0x0002, 0x8606},
+/* 429 */ {0x0064, 0x8607},
+/* 430 */ {0x003a, 0x8601},
+/* 431 */ {0x0008, 0x8602},
+/* 432 */ {0x0044, 0x8600},
+/* 433 */ {0x0018, 0x8617},
+/* 434 */ {0x0008, 0x8618},
+/* 435 */ {0x00a1, 0x8656},
+/* 436 */ {0x0004, 0x865b},
+/* 437 */ {0x0002, 0x865c},
+/* 438 */ {0x0058, 0x865d},
+/* 439 */ {0x0048, 0x865e},
+/* 440 */ {0x0012, 0x8608},
+/* 441 */ {0x002c, 0x8609},
+/* 442 */ {0x0002, 0x860a},
+/* 443 */ {0x002c, 0x860b},
+/* 444 */ {0x00db, 0x860c},
+/* 445 */ {0x00f9, 0x860d},
+/* 446 */ {0x00f1, 0x860e},
+/* 447 */ {0x00e3, 0x860f},
+/* 448 */ {0x002c, 0x8610},
+/* 449 */ {0x006c, 0x8651},
+/* 450 */ {0x0041, 0x8652},
+/* 451 */ {0x0059, 0x8653},
+/* 452 */ {0x0040, 0x8654},
+/* 453 */ {0x00fa, 0x8611},
+/* 454 */ {0x00ff, 0x8612},
+/* 455 */ {0x00f8, 0x8613},
+/* 456 */ {0x0000, 0x8614},
+/* 457 */ {0x0001, 0x863f},
+/* 458 */ {0x0000, 0x8640},
+/* 459 */ {0x0026, 0x8641},
+/* 460 */ {0x0045, 0x8642},
+/* 461 */ {0x0060, 0x8643},
+/* 462 */ {0x0075, 0x8644},
+/* 463 */ {0x0088, 0x8645},
+/* 464 */ {0x009b, 0x8646},
+/* 465 */ {0x00b0, 0x8647},
+/* 466 */ {0x00c5, 0x8648},
+/* 467 */ {0x00d2, 0x8649},
+/* 468 */ {0x00dc, 0x864a},
+/* 469 */ {0x00e5, 0x864b},
+/* 470 */ {0x00eb, 0x864c},
+/* 471 */ {0x00f0, 0x864d},
+/* 472 */ {0x00f6, 0x864e},
+/* 473 */ {0x00fa, 0x864f},
+/* 474 */ {0x00ff, 0x8650},
+/* 475 */ {0x0060, 0x8657},
+/* 476 */ {0x0010, 0x8658},
+/* 477 */ {0x0018, 0x8659},
+/* 478 */ {0x0005, 0x865a},
+/* 479 */ {0x0018, 0x8660},
+/* 480 */ {0x0003, 0x8509},
+/* 481 */ {0x0011, 0x850a},
+/* 482 */ {0x0032, 0x850b},
+/* 483 */ {0x0010, 0x850c},
+/* 484 */ {0x0021, 0x850d},
+/* 485 */ {0x0001, 0x8500},
+/* 486 */ {0x0000, 0x8508},
+
+/* 487 */ {0x0012, 0x8608},
+/* 488 */ {0x002c, 0x8609},
+/* 489 */ {0x0002, 0x860a},
+/* 490 */ {0x0039, 0x860b},
+/* 491 */ {0x00d0, 0x860c},
+/* 492 */ {0x00f7, 0x860d},
+/* 493 */ {0x00ed, 0x860e},
+/* 494 */ {0x00db, 0x860f},
+/* 495 */ {0x0039, 0x8610},
+/* 496 */ {0x0012, 0x8657},
+/* 497 */ {0x0064, 0x8619},
+
+/* This line starts it all, it is not needed here */
+/* since it has been build into the driver */
+/* jfm: don't start now */
+/* 590  *  {0x0030, 0x8112}, */
+	{}
+};
+
+/*
+ * Initialization data for Creative Webcam Vista
+ */
+static __u16 spca508_vista_init_data[][3] = {
+	{0x0008, 0x8200},	/* Clear register */
+	{0x0000, 0x870b},	/* Reset CTL3 */
+	{0x0020, 0x8112},	/* Video Drop packet enable */
+	{0x0003, 0x8111},  /* Soft Reset compression, memory, TG & CDSP */
+	{0x0000, 0x8110},	/* Disable everything */
+	{0x0000, 0x8114},	/* Software GPIO output data */
+	{0x0000, 0x8114},
+
+	{0x0003, 0x8111},
+	{0x0000, 0x8111},
+	{0x0090, 0x8110},  /* Enable: SSI output, External 2X clock output */
+	{0x0020, 0x8112},
+	{0x0000, 0x8114},
+	{0x0001, 0x8114},
+	{0x0001, 0x8114},
+	{0x0001, 0x8114},
+	{0x0003, 0x8114},
+
+	{0x000f, 0x8402},	/* Memory bank Address */
+	{0x0000, 0x8403},	/* Memory bank Address */
+	{0x00ba, 0x8804},	/* SSI Slave address */
+	{0x0010, 0x8802},	/* 93.75kHz SSI Clock Two DataByte */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},	/* Will write 2 bytes (DATA1+DATA2) */
+	{0x0020, 0x8801},	/* Register address for SSI read/write */
+	{0x0044, 0x8805},	/* DATA2 */
+	{0x0004, 0x8800},	/* DATA1 -> write triggered */
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0009, 0x8801},
+	{0x0042, 0x8805},
+	{0x0001, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x003c, 0x8801},
+	{0x0001, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0001, 0x8801},
+	{0x000a, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0002, 0x8801},
+	{0x0000, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0003, 0x8801},
+	{0x0027, 0x8805},
+	{0x0001, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0004, 0x8801},
+	{0x0065, 0x8805},
+	{0x0001, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0005, 0x8801},
+	{0x0003, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0006, 0x8801},
+	{0x001c, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0007, 0x8801},
+	{0x002a, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x000e, 0x8801},
+	{0x0000, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0028, 0x8801},
+	{0x002e, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0039, 0x8801},
+	{0x0013, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x003b, 0x8801},
+	{0x000c, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0035, 0x8801},
+	{0x0028, 0x8805},
+	{0x0000, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+	/* READ { 0, 0x0001, 0x8802 } ->
+		0000: 10  */
+	{0x0010, 0x8802},
+	{0x0009, 0x8801},
+	{0x0042, 0x8805},
+	{0x0001, 0x8800},
+	/* READ { 0, 0x0001, 0x8803 } ->
+		0000: 00  */
+
+	{0x0050, 0x8703},
+	{0x0002, 0x8704},	/* External input CKIx1 */
+	{0x0001, 0x870C},	/* Select CKOx2 output */
+	{0x009A, 0x8600},	/* Line memory Read Counter (L) */
+	{0x0001, 0x8606},  /* 1 Line memory Read Counter (H) Result: (d)410 */
+	{0x0023, 0x8601},
+	{0x0010, 0x8602},
+	{0x000A, 0x8603},
+	{0x009A, 0x8600},
+	{0x0001, 0x865B},	/* 1 Horizontal Offset for Valid Pixel(L) */
+	{0x0003, 0x865C},	/* Vertical offset for valid lines (L) */
+	{0x0058, 0x865D},	/* Horizontal valid pixels window (L) */
+	{0x0048, 0x865E},	/* Vertical valid lines window (L) */
+	{0x0000, 0x865F},
+
+	{0x0006, 0x8660},
+		    /* Enable nibble data input, select nibble input order */
+
+	{0x0013, 0x8608},	/* A11 Coeficients for color correction */
+	{0x0028, 0x8609},
+		    /* Note: these values are confirmed at the end of array */
+	{0x0005, 0x860A},	/* ... */
+	{0x0025, 0x860B},
+	{0x00E1, 0x860C},
+	{0x00FA, 0x860D},
+	{0x00F4, 0x860E},
+	{0x00E8, 0x860F},
+	{0x0025, 0x8610},	/* A33 Coef. */
+	{0x00FC, 0x8611},	/* White balance offset: R */
+	{0x0001, 0x8612},	/* White balance offset: Gr */
+	{0x00FE, 0x8613},	/* White balance offset: B */
+	{0x0000, 0x8614},	/* White balance offset: Gb */
+
+	{0x0064, 0x8651},	/* R gain for white balance (L) */
+	{0x0040, 0x8652},	/* Gr gain for white balance (L) */
+	{0x0066, 0x8653},	/* B gain for white balance (L) */
+	{0x0040, 0x8654},	/* Gb gain for white balance (L) */
+	{0x0001, 0x863F},	/* Enable fixed gamma correction */
+
+	{0x00A1, 0x8656},	/* Size - Window1: 256x256, Window2: 128x128 */
+	/* UV division: UV no change, Enable New edge enhancement */
+	{0x0018, 0x8657},	/* Edge gain high threshold */
+	{0x0020, 0x8658},	/* Edge gain low threshold */
+	{0x000A, 0x8659},	/* Edge bandwidth high threshold */
+	{0x0005, 0x865A},	/* Edge bandwidth low threshold */
+	{0x0064, 0x8607},	/* UV filter enable */
+
+	{0x0016, 0x8660},
+	{0x0000, 0x86B0},	/* Bad pixels compensation address */
+	{0x00DC, 0x86B1},	/* X coord for bad pixels compensation (L) */
+	{0x0000, 0x86B2},
+	{0x0009, 0x86B3},	/* Y coord for bad pixels compensation (L) */
+	{0x0000, 0x86B4},
+
+	{0x0001, 0x86B0},
+	{0x00F5, 0x86B1},
+	{0x0000, 0x86B2},
+	{0x00C6, 0x86B3},
+	{0x0000, 0x86B4},
+
+	{0x0002, 0x86B0},
+	{0x001C, 0x86B1},
+	{0x0001, 0x86B2},
+	{0x00D7, 0x86B3},
+	{0x0000, 0x86B4},
+
+	{0x0003, 0x86B0},
+	{0x001C, 0x86B1},
+	{0x0001, 0x86B2},
+	{0x00D8, 0x86B3},
+	{0x0000, 0x86B4},
+
+	{0x0004, 0x86B0},
+	{0x001D, 0x86B1},
+	{0x0001, 0x86B2},
+	{0x00D8, 0x86B3},
+	{0x0000, 0x86B4},
+	{0x001E, 0x8660},
+
+	/* READ { 0, 0x0000, 0x8608 } ->
+		0000: 13  */
+	/* READ { 0, 0x0000, 0x8609 } ->
+		0000: 28  */
+	/* READ { 0, 0x0000, 0x8610 } ->
+		0000: 05  */
+	/* READ { 0, 0x0000, 0x8611 } ->
+		0000: 25  */
+	/* READ { 0, 0x0000, 0x8612 } ->
+		0000: e1  */
+	/* READ { 0, 0x0000, 0x8613 } ->
+		0000: fa  */
+	/* READ { 0, 0x0000, 0x8614 } ->
+		0000: f4  */
+	/* READ { 0, 0x0000, 0x8615 } ->
+		0000: e8  */
+	/* READ { 0, 0x0000, 0x8616 } ->
+		0000: 25  */
+	{}
+};
+
+static int reg_write(struct usb_device *dev,
+			__u16 index, __u16 value)
+{
+	int ret;
+
+	ret = usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			0,		/* request */
+			USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, NULL, 0, 500);
+	PDEBUG(D_USBO, "reg write i:0x%04x = 0x%02x",
+		index, value);
+	if (ret < 0)
+		PDEBUG(D_ERR|D_USBO, "reg write: error %d", ret);
+	return ret;
+}
+
+/* read 1 byte */
+/* returns: negative is error, pos or zero is data */
+static int reg_read(struct usb_device *dev,
+			__u16 index)	/* wIndex */
+{
+	int ret;
+	__u8 data;
+
+	ret = usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,			/* register */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			(__u16) 0,		/* value */
+			index,
+			&data, 1,
+			500);			/* timeout */
+	PDEBUG(D_USBI, "reg read i:%04x --> %02x", index, data);
+	if (ret < 0) {
+		PDEBUG(D_ERR|D_USBI, "reg_read err %d", ret);
+		return ret;
+	}
+	return data;
+}
+
+static int write_vector(struct gspca_dev *gspca_dev,
+				__u16 data[][3])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int ret, i = 0;
+
+	while (data[i][1] != 0) {
+		ret = reg_write(dev, data[i][1], data[i][0]);
+		if (ret < 0)
+			return ret;
+		i++;
+	}
+	return 0;
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	struct cam *cam;
+	__u16 vendor;
+	__u16 product;
+	int data1, data2;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+	switch (vendor) {
+	case 0x041e:		/* Creative cameras */
+/*		switch (product) { */
+/*		case 0x4018: */
+			sd->subtype = CreativeVista;
+/*			break; */
+/*		} */
+		break;
+	case 0x0461:		/* MicroInnovation */
+/*		switch (product) { */
+/*		case 0x0815: */
+			sd->subtype = MicroInnovationIC200;
+/*			break; */
+/*		} */
+		break;
+	case 0x0733:	/* Rebadged ViewQuest (Intel) and ViewQuest cameras */
+/*		switch (product) { */
+/*		case 0x110: */
+			sd->subtype = ViewQuestVQ110;
+/*			break; */
+/*		} */
+		break;
+	case 0x0af9:		/* Hama cameras */
+		switch (product) {
+		case 0x0010:
+			sd->subtype = HamaUSBSightcam;
+			break;
+		case 0x0011:
+			sd->subtype = HamaUSBSightcam2;
+			break;
+		}
+		break;
+	case 0x8086:		/* Intel */
+/*		switch (product) { */
+/*		case 0x0110: */
+			sd->subtype = IntelEasyPCCamera;
+/*			break; */
+/*		} */
+		break;
+	}
+
+	/* Read from global register the USB product and vendor IDs, just to */
+	/* prove that we can communicate with the device.  This works, which */
+	/* confirms at we are communicating properly and that the device */
+	/* is a 508. */
+	data1 = reg_read(dev, 0x8104);
+	data2 = reg_read(dev, 0x8105);
+	PDEBUG(D_PROBE,
+		"Read from GLOBAL: USB Vendor ID 0x%02x%02x", data2, data1);
+
+	data1 = reg_read(dev, 0x8106);
+	data2 = reg_read(dev, 0x8107);
+	PDEBUG(D_PROBE,
+		"Read from GLOBAL: USB Product ID 0x%02x%02x", data2, data1);
+
+	data1 = reg_read(dev, 0x8621);
+	PDEBUG(D_PROBE,
+		"Read from GLOBAL: Window 1 average luminance %d", data1);
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	cam->cam_mode = sif_mode;
+	cam->nmodes = sizeof sif_mode / sizeof sif_mode[0];
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+
+	switch (sd->subtype) {
+	case ViewQuestVQ110:
+		if (write_vector(gspca_dev, spca508_init_data))
+			return -1;
+		break;
+	default:
+/*	case MicroInnovationIC200: */
+/*	case IntelEasyPCCamera: */
+		if (write_vector(gspca_dev, spca508cs110_init_data))
+			return -1;
+		break;
+	case HamaUSBSightcam:
+		if (write_vector(gspca_dev, spca508_sightcam_init_data))
+			return -1;
+		break;
+	case HamaUSBSightcam2:
+		if (write_vector(gspca_dev, spca508_sightcam2_init_data))
+			return -1;
+		break;
+	case CreativeVista:
+		if (write_vector(gspca_dev, spca508_vista_init_data))
+			return -1;
+		break;
+	}
+	return 0;			/* success */
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+/*	write_vector(gspca_dev, spca508_open_data); */
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	int mode;
+
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+	reg_write(gspca_dev->dev, 0x8500, mode);
+	switch (mode) {
+	case 0:
+	case 1:
+		reg_write(gspca_dev->dev, 0x8700, 0x28);	/* clock */
+		break;
+	default:
+/*	case 2: */
+/*	case 3: */
+		reg_write(gspca_dev->dev, 0x8700, 0x23);	/* clock */
+		break;
+	}
+	reg_write(gspca_dev->dev, 0x8112, 0x10 | 0x20);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	/* Video ISO disable, Video Drop Packet enable: */
+	reg_write(gspca_dev->dev, 0x8112, 0x20);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+/* this function is called at close time */
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+/* convert YUVY per line to YUYV (YUV 4:2:2) */
+static void yuvy_decode(unsigned char *out,
+			unsigned char *in,
+			int width,
+			int height)
+{
+	unsigned char *Ui, *Vi, *yi, *yi1;
+	unsigned char *out1;
+	int i, j;
+
+	yi = in;
+	for (i = height / 2; --i >= 0; ) {
+		out1 = out + width * 2;		/* next line */
+		Ui = yi + width;
+		Vi = Ui + width / 2;
+		yi1 = Vi + width / 2;
+		for (j = width / 2; --j >= 0; ) {
+			*out++ = 128 + *yi++;
+			*out++ = 128 + *Ui;
+			*out++ = 128 + *yi++;
+			*out++ = 128 + *Vi;
+
+			*out1++ = 128 + *yi1++;
+			*out1++ = 128 + *Ui++;
+			*out1++ = 128 + *yi1++;
+			*out1++ = 128 + *Vi++;
+		}
+		yi += width * 2;
+		out = out1;
+	}
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	switch (data[0]) {
+	case 0:				/* start of frame */
+		if (gspca_dev->last_packet_type == FIRST_PACKET) {
+			yuvy_decode(sd->tmpbuf2, sd->tmpbuf,
+					gspca_dev->width,
+					gspca_dev->height);
+			frame = gspca_frame_add(gspca_dev,
+						LAST_PACKET,
+						frame,
+						sd->tmpbuf2,
+						gspca_dev->width
+							* gspca_dev->height
+							* 2);
+		}
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame,
+				data, 0);
+		data += SPCA508_OFFSET_DATA;
+		len -= SPCA508_OFFSET_DATA;
+		if (len > 0)
+			memcpy(sd->tmpbuf, data, len);
+		else
+			len = 0;
+		sd->buflen = len;
+		return;
+	case 0xff:			/* drop */
+/*		gspca_dev->last_packet_type = DISCARD_PACKET; */
+		return;
+	}
+	data += 1;
+	len -= 1;
+	memcpy(&sd->tmpbuf[sd->buflen], data, len);
+	sd->buflen += len;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 brightness = sd->brightness;
+
+/* MX seem contrast */
+	reg_write(gspca_dev->dev, 0x8651, brightness);
+	reg_write(gspca_dev->dev, 0x8652, brightness);
+	reg_write(gspca_dev->dev, 0x8653, brightness);
+	reg_write(gspca_dev->dev, 0x8654, brightness);
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = reg_read(gspca_dev->dev, 0x8651);
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x041e, 0x4018), DVNM("Creative Webcam Vista (PD1100)")},
+	{USB_DEVICE(0x0461, 0x0815), DVNM("Micro Innovation IC200")},
+	{USB_DEVICE(0x0733, 0x0110), DVNM("ViewQuest VQ110")},
+	{USB_DEVICE(0x0af9, 0x0010), DVNM("Hama USB Sightcam 100")},
+	{USB_DEVICE(0x0af9, 0x0011), DVNM("Hama USB Sightcam 100")},
+	{USB_DEVICE(0x8086, 0x0110), DVNM("Intel Easy PC Camera")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/spca561.c b/drivers/media/video/gspca/spca561.c
new file mode 100644
index 000000000000..a94e6270115e
--- /dev/null
+++ b/drivers/media/video/gspca/spca561.c
@@ -0,0 +1,1025 @@
+/*
+ * Sunplus spca561 subdriver
+ *
+ * Copyright (C) 2004 Michel Xhaard mxhaard@magic.fr
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "spca561"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SPCA561 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned short contrast;
+	__u8 brightness;
+	__u8 autogain;
+
+	__u8 chip_revision;
+	signed char ag_cnt;
+#define AG_CNT_START 13
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	 {
+	  .id = V4L2_CID_BRIGHTNESS,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Brightness",
+	  .minimum = 0,
+	  .maximum = 63,
+	  .step = 1,
+	  .default_value = 32,
+	  },
+	 .set = sd_setbrightness,
+	 .get = sd_getbrightness,
+	 },
+#define SD_CONTRAST 1
+	{
+	 {
+	  .id = V4L2_CID_CONTRAST,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Contrast",
+	  .minimum = 0,
+	  .maximum = 0x3fff,
+	  .step = 1,
+	  .default_value = 0x2000,
+	  },
+	 .set = sd_setcontrast,
+	 .get = sd_getcontrast,
+	 },
+#define SD_AUTOGAIN 2
+	{
+	 {
+	  .id = V4L2_CID_AUTOGAIN,
+	  .type = V4L2_CTRL_TYPE_BOOLEAN,
+	  .name = "Auto Gain",
+	  .minimum = 0,
+	  .maximum = 1,
+	  .step = 1,
+	  .default_value = 1,
+	  },
+	 .set = sd_setautogain,
+	 .get = sd_getautogain,
+	 },
+};
+
+static struct cam_mode sif_mode[] = {
+	{V4L2_PIX_FMT_SPCA561, 160, 120, 3},
+	{V4L2_PIX_FMT_SPCA561, 176, 144, 2},
+	{V4L2_PIX_FMT_SPCA561, 320, 240, 1},
+	{V4L2_PIX_FMT_SPCA561, 352, 288, 0},
+};
+
+/*
+ * Initialization data
+ * I'm not very sure how to split initialization from open data
+ * chunks. For now, we'll consider everything as initialization
+ */
+/* Frame packet header offsets for the spca561 */
+#define SPCA561_OFFSET_SNAP 1
+#define SPCA561_OFFSET_TYPE 2
+#define SPCA561_OFFSET_COMPRESS 3
+#define SPCA561_OFFSET_FRAMSEQ   4
+#define SPCA561_OFFSET_GPIO 5
+#define SPCA561_OFFSET_USBBUFF 6
+#define SPCA561_OFFSET_WIN2GRAVE 7
+#define SPCA561_OFFSET_WIN2RAVE 8
+#define SPCA561_OFFSET_WIN2BAVE 9
+#define SPCA561_OFFSET_WIN2GBAVE 10
+#define SPCA561_OFFSET_WIN1GRAVE 11
+#define SPCA561_OFFSET_WIN1RAVE 12
+#define SPCA561_OFFSET_WIN1BAVE 13
+#define SPCA561_OFFSET_WIN1GBAVE 14
+#define SPCA561_OFFSET_FREQ 15
+#define SPCA561_OFFSET_VSYNC 16
+#define SPCA561_OFFSET_DATA 1
+#define SPCA561_INDEX_I2C_BASE 0x8800
+#define SPCA561_SNAPBIT 0x20
+#define SPCA561_SNAPCTRL 0x40
+enum {
+	Rev072A = 0,
+	Rev012A,
+};
+
+static void reg_w_val(struct usb_device *dev, __u16 index, __u16 value)
+{
+	int ret;
+
+	ret = usb_control_msg(dev, usb_sndctrlpipe(dev, 0),
+			      0,		/* request */
+			      USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			      value, index, NULL, 0, 500);
+	PDEBUG(D_USBO, "reg write: 0x%02x:0x%02x", index, value);
+	if (ret < 0)
+		PDEBUG(D_ERR, "reg write: error %d", ret);
+}
+
+static void write_vector(struct gspca_dev *gspca_dev, __u16 data[][2])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int i;
+
+	i = 0;
+	while (data[i][1] != 0) {
+		reg_w_val(dev, data[i][1], data[i][0]);
+		i++;
+	}
+}
+
+static void reg_r(struct usb_device *dev,
+		  __u16 index, __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev, usb_rcvctrlpipe(dev, 0),
+			0,			/* request */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,			/* value */
+			index, buffer, length, 500);
+}
+
+static void reg_w_buf(struct usb_device *dev,
+		      __u16 index, __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev, usb_sndctrlpipe(dev, 0),
+			0,			/* request */
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,			/* value */
+			index, buffer, length, 500);
+}
+
+static void i2c_init(struct gspca_dev *gspca_dev, __u8 mode)
+{
+	reg_w_val(gspca_dev->dev, 0x92, 0x8804);
+	reg_w_val(gspca_dev->dev, mode, 0x8802);
+}
+
+static void i2c_write(struct gspca_dev *gspca_dev, __u16 valeur, __u16 reg)
+{
+	int retry = 60;
+	__u8 DataLow;
+	__u8 DataHight;
+	__u8 Data;
+
+	DataLow = valeur;
+	DataHight = valeur >> 8;
+	reg_w_val(gspca_dev->dev, reg, 0x8801);
+	reg_w_val(gspca_dev->dev, DataLow, 0x8805);
+	reg_w_val(gspca_dev->dev, DataHight, 0x8800);
+	while (retry--) {
+		reg_r(gspca_dev->dev, 0x8803, &Data, 1);
+		if (!Data)
+			break;
+	}
+}
+
+static int i2c_read(struct gspca_dev *gspca_dev, __u16 reg, __u8 mode)
+{
+	int retry = 60;
+	__u8 value;
+	__u8 vallsb;
+	__u8 Data;
+
+	reg_w_val(gspca_dev->dev, 0x92, 0x8804);
+	reg_w_val(gspca_dev->dev, reg, 0x8801);
+	reg_w_val(gspca_dev->dev, (mode | 0x01), 0x8802);
+	while (retry--) {
+		reg_r(gspca_dev->dev, 0x8803, &Data, 1);
+		if (!Data)
+			break;
+	}
+	if (retry == 0)
+		return -1;
+	reg_r(gspca_dev->dev, 0x8800, &value, 1);
+	reg_r(gspca_dev->dev, 0x8805, &vallsb, 1);
+	return ((int) value << 8) | vallsb;
+}
+
+static __u16 spca561_init_data[][2] = {
+	{0x0000, 0x8114},	/* Software GPIO output data */
+	{0x0001, 0x8114},	/* Software GPIO output data */
+	{0x0000, 0x8112},	/* Some kind of reset */
+	{0x0003, 0x8701},	/* PCLK clock delay adjustment */
+	{0x0001, 0x8703},	/* HSYNC from cmos inverted */
+	{0x0011, 0x8118},	/* Enable and conf sensor */
+	{0x0001, 0x8118},	/* Conf sensor */
+	{0x0092, 0x8804},	/* I know nothing about these */
+	{0x0010, 0x8802},	/* 0x88xx registers, so I won't */
+	/***************/
+	{0x000d, 0x8805},	/* sensor default setting */
+	{0x0001, 0x8801},	/* 1 <- 0x0d */
+	{0x0000, 0x8800},
+	{0x0018, 0x8805},
+	{0x0002, 0x8801},	/* 2 <- 0x18 */
+	{0x0000, 0x8800},
+	{0x0065, 0x8805},
+	{0x0004, 0x8801},	/* 4 <- 0x01 0x65 */
+	{0x0001, 0x8800},
+	{0x0021, 0x8805},
+	{0x0005, 0x8801},	/* 5 <- 0x21 */
+	{0x0000, 0x8800},
+	{0x00aa, 0x8805},
+	{0x0007, 0x8801},	/* 7 <- 0xaa */
+	{0x0000, 0x8800},
+	{0x0004, 0x8805},
+	{0x0020, 0x8801},	/* 0x20 <- 0x15 0x04 */
+	{0x0015, 0x8800},
+	{0x0002, 0x8805},
+	{0x0039, 0x8801},	/* 0x39 <- 0x02 */
+	{0x0000, 0x8800},
+	{0x0010, 0x8805},
+	{0x0035, 0x8801},	/* 0x35 <- 0x10 */
+	{0x0000, 0x8800},
+	{0x0049, 0x8805},
+	{0x0009, 0x8801},	/* 0x09 <- 0x10 0x49 */
+	{0x0010, 0x8800},
+	{0x000b, 0x8805},
+	{0x0028, 0x8801},	/* 0x28 <- 0x0b */
+	{0x0000, 0x8800},
+	{0x000f, 0x8805},
+	{0x003b, 0x8801},	/* 0x3b <- 0x0f */
+	{0x0000, 0x8800},
+	{0x0000, 0x8805},
+	{0x003c, 0x8801},	/* 0x3c <- 0x00 */
+	{0x0000, 0x8800},
+	/***************/
+	{0x0018, 0x8601},	/* Pixel/line selection for color separation */
+	{0x0000, 0x8602},	/* Optical black level for user setting */
+	{0x0060, 0x8604},	/* Optical black horizontal offset */
+	{0x0002, 0x8605},	/* Optical black vertical offset */
+	{0x0000, 0x8603},	/* Non-automatic optical black level */
+	{0x0002, 0x865b},	/* Horizontal offset for valid pixels */
+	{0x0000, 0x865f},	/* Vertical valid pixels window (x2) */
+	{0x00b0, 0x865d},	/* Horizontal valid pixels window (x2) */
+	{0x0090, 0x865e},	/* Vertical valid lines window (x2) */
+	{0x00e0, 0x8406},	/* Memory buffer threshold */
+	{0x0000, 0x8660},	/* Compensation memory stuff */
+	{0x0002, 0x8201},	/* Output address for r/w serial EEPROM */
+	{0x0008, 0x8200},	/* Clear valid bit for serial EEPROM */
+	{0x0001, 0x8200},	/* OprMode to be executed by hardware */
+	{0x0007, 0x8201},	/* Output address for r/w serial EEPROM */
+	{0x0008, 0x8200},	/* Clear valid bit for serial EEPROM */
+	{0x0001, 0x8200},	/* OprMode to be executed by hardware */
+	{0x0010, 0x8660},	/* Compensation memory stuff */
+	{0x0018, 0x8660},	/* Compensation memory stuff */
+
+	{0x0004, 0x8611},	/* R offset for white balance */
+	{0x0004, 0x8612},	/* Gr offset for white balance */
+	{0x0007, 0x8613},	/* B offset for white balance */
+	{0x0000, 0x8614},	/* Gb offset for white balance */
+	{0x008c, 0x8651},	/* R gain for white balance */
+	{0x008c, 0x8652},	/* Gr gain for white balance */
+	{0x00b5, 0x8653},	/* B gain for white balance */
+	{0x008c, 0x8654},	/* Gb gain for white balance */
+	{0x0002, 0x8502},	/* Maximum average bit rate stuff */
+
+	{0x0011, 0x8802},
+	{0x0087, 0x8700},	/* Set master clock (96Mhz????) */
+	{0x0081, 0x8702},	/* Master clock output enable */
+
+	{0x0000, 0x8500},	/* Set image type (352x288 no compression) */
+	/* Originally was 0x0010 (352x288 compression) */
+
+	{0x0002, 0x865b},	/* Horizontal offset for valid pixels */
+	{0x0003, 0x865c},	/* Vertical offset for valid lines */
+	/***************//* sensor active */
+	{0x0003, 0x8801},	/* 0x03 <- 0x01 0x21 //289 */
+	{0x0021, 0x8805},
+	{0x0001, 0x8800},
+	{0x0004, 0x8801},	/* 0x04 <- 0x01 0x65 //357 */
+	{0x0065, 0x8805},
+	{0x0001, 0x8800},
+	{0x0005, 0x8801},	/* 0x05 <- 0x2f */
+	{0x002f, 0x8805},
+	{0x0000, 0x8800},
+	{0x0006, 0x8801},	/* 0x06 <- 0 */
+	{0x0000, 0x8805},
+	{0x0000, 0x8800},
+	{0x000a, 0x8801},	/* 0x0a <- 2 */
+	{0x0002, 0x8805},
+	{0x0000, 0x8800},
+	{0x0009, 0x8801},	/* 0x09 <- 0x1061 */
+	{0x0061, 0x8805},
+	{0x0010, 0x8800},
+	{0x0035, 0x8801},	/* 0x35 <-0x14 */
+	{0x0014, 0x8805},
+	{0x0000, 0x8800},
+	{0x0030, 0x8112},	/* ISO and drop packet enable */
+	{0x0000, 0x8112},	/* Some kind of reset ???? */
+	{0x0009, 0x8118},	/* Enable sensor and set standby */
+	{0x0000, 0x8114},	/* Software GPIO output data */
+	{0x0000, 0x8114},	/* Software GPIO output data */
+	{0x0001, 0x8114},	/* Software GPIO output data */
+	{0x0000, 0x8112},	/* Some kind of reset ??? */
+	{0x0003, 0x8701},
+	{0x0001, 0x8703},
+	{0x0011, 0x8118},
+	{0x0001, 0x8118},
+	/***************/
+	{0x0092, 0x8804},
+	{0x0010, 0x8802},
+	{0x000d, 0x8805},
+	{0x0001, 0x8801},
+	{0x0000, 0x8800},
+	{0x0018, 0x8805},
+	{0x0002, 0x8801},
+	{0x0000, 0x8800},
+	{0x0065, 0x8805},
+	{0x0004, 0x8801},
+	{0x0001, 0x8800},
+	{0x0021, 0x8805},
+	{0x0005, 0x8801},
+	{0x0000, 0x8800},
+	{0x00aa, 0x8805},
+	{0x0007, 0x8801},	/* mode 0xaa */
+	{0x0000, 0x8800},
+	{0x0004, 0x8805},
+	{0x0020, 0x8801},
+	{0x0015, 0x8800},	/* mode 0x0415 */
+	{0x0002, 0x8805},
+	{0x0039, 0x8801},
+	{0x0000, 0x8800},
+	{0x0010, 0x8805},
+	{0x0035, 0x8801},
+	{0x0000, 0x8800},
+	{0x0049, 0x8805},
+	{0x0009, 0x8801},
+	{0x0010, 0x8800},
+	{0x000b, 0x8805},
+	{0x0028, 0x8801},
+	{0x0000, 0x8800},
+	{0x000f, 0x8805},
+	{0x003b, 0x8801},
+	{0x0000, 0x8800},
+	{0x0000, 0x8805},
+	{0x003c, 0x8801},
+	{0x0000, 0x8800},
+	{0x0002, 0x8502},
+	{0x0039, 0x8801},
+	{0x0000, 0x8805},
+	{0x0000, 0x8800},
+
+	{0x0087, 0x8700},	/* overwrite by start */
+	{0x0081, 0x8702},
+	{0x0000, 0x8500},
+/*	{0x0010, 0x8500},  -- Previous line was this */
+	{0x0002, 0x865b},
+	{0x0003, 0x865c},
+	/***************/
+	{0x0003, 0x8801},	/* 0x121-> 289 */
+	{0x0021, 0x8805},
+	{0x0001, 0x8800},
+	{0x0004, 0x8801},	/* 0x165 -> 357 */
+	{0x0065, 0x8805},
+	{0x0001, 0x8800},
+	{0x0005, 0x8801},	/* 0x2f //blanking control colonne */
+	{0x002f, 0x8805},
+	{0x0000, 0x8800},
+	{0x0006, 0x8801},	/* 0x00 //blanking mode row */
+	{0x0000, 0x8805},
+	{0x0000, 0x8800},
+	{0x000a, 0x8801},	/* 0x01 //0x02 */
+	{0x0001, 0x8805},
+	{0x0000, 0x8800},
+	{0x0009, 0x8801},	/* 0x1061 - setexposure times && pixel clock
+				 * 0001 0 | 000 0110 0001 */
+	{0x0061, 0x8805},	/* 61 31 */
+	{0x0008, 0x8800},	/* 08 */
+	{0x0035, 0x8801},	/* 0x14 - set gain general */
+	{0x001f, 0x8805},	/* 0x14 */
+	{0x0000, 0x8800},
+	{0x0030, 0x8112},
+	{}
+};
+
+static void sensor_reset(struct gspca_dev *gspca_dev)
+{
+	reg_w_val(gspca_dev->dev, 0x8631, 0xc8);
+	reg_w_val(gspca_dev->dev, 0x8634, 0xc8);
+	reg_w_val(gspca_dev->dev, 0x8112, 0x00);
+	reg_w_val(gspca_dev->dev, 0x8114, 0x00);
+	reg_w_val(gspca_dev->dev, 0x8118, 0x21);
+	i2c_init(gspca_dev, 0x14);
+	i2c_write(gspca_dev, 1, 0x0d);
+	i2c_write(gspca_dev, 0, 0x0d);
+}
+
+/******************** QC Express etch2 stuff ********************/
+static __u16 Pb100_1map8300[][2] = {
+	/* reg, value */
+	{0x8320, 0x3304},
+
+	{0x8303, 0x0125},	/* image area */
+	{0x8304, 0x0169},
+	{0x8328, 0x000b},
+	{0x833c, 0x0001},
+
+	{0x832f, 0x0419},
+	{0x8307, 0x00aa},
+	{0x8301, 0x0003},
+	{0x8302, 0x000e},
+	{}
+};
+static __u16 Pb100_2map8300[][2] = {
+	/* reg, value */
+	{0x8339, 0x0000},
+	{0x8307, 0x00aa},
+	{}
+};
+
+static __u16 spca561_161rev12A_data1[][2] = {
+	{0x21, 0x8118},
+	{0x01, 0x8114},
+	{0x00, 0x8112},
+	{0x92, 0x8804},
+	{0x04, 0x8802},		/* windows uses 08 */
+	{}
+};
+static __u16 spca561_161rev12A_data2[][2] = {
+	{0x21, 0x8118},
+	{0x10, 0x8500},
+	{0x07, 0x8601},
+	{0x07, 0x8602},
+	{0x04, 0x8501},
+	{0x21, 0x8118},
+
+	{0x07, 0x8201},		/* windows uses 02 */
+	{0x08, 0x8200},
+	{0x01, 0x8200},
+
+	{0x00, 0x8114},
+	{0x01, 0x8114},		/* windows uses 00 */
+
+	{0x90, 0x8604},
+	{0x00, 0x8605},
+	{0xb0, 0x8603},
+
+	/* sensor gains */
+	{0x00, 0x8610},		/* *red */
+	{0x00, 0x8611},		/* 3f   *green */
+	{0x00, 0x8612},		/* green *blue */
+	{0x00, 0x8613},		/* blue *green */
+	{0x35, 0x8614},		/* green *red */
+	{0x35, 0x8615},		/* 40   *green */
+	{0x35, 0x8616},		/* 7a   *blue */
+	{0x35, 0x8617},		/* 40   *green */
+
+	{0x0c, 0x8620},		/* 0c */
+	{0xc8, 0x8631},		/* c8 */
+	{0xc8, 0x8634},		/* c8 */
+	{0x23, 0x8635},		/* 23 */
+	{0x1f, 0x8636},		/* 1f */
+	{0xdd, 0x8637},		/* dd */
+	{0xe1, 0x8638},		/* e1 */
+	{0x1d, 0x8639},		/* 1d */
+	{0x21, 0x863a},		/* 21 */
+	{0xe3, 0x863b},		/* e3 */
+	{0xdf, 0x863c},		/* df */
+	{0xf0, 0x8505},
+	{0x32, 0x850a},
+	{}
+};
+
+static void sensor_mapwrite(struct gspca_dev *gspca_dev,
+			    __u16 sensormap[][2])
+{
+	int i = 0;
+	__u8 usbval[2];
+
+	while (sensormap[i][0]) {
+		usbval[0] = sensormap[i][1];
+		usbval[1] = sensormap[i][1] >> 8;
+		reg_w_buf(gspca_dev->dev, sensormap[i][0], usbval, 2);
+		i++;
+	}
+}
+static void init_161rev12A(struct gspca_dev *gspca_dev)
+{
+	sensor_reset(gspca_dev);
+	write_vector(gspca_dev, spca561_161rev12A_data1);
+	sensor_mapwrite(gspca_dev, Pb100_1map8300);
+	write_vector(gspca_dev, spca561_161rev12A_data2);
+	sensor_mapwrite(gspca_dev, Pb100_2map8300);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+		     const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	struct cam *cam;
+	__u16 vendor, product;
+	__u8 data1, data2;
+
+	/* Read frm global register the USB product and vendor IDs, just to
+	 * prove that we can communicate with the device.  This works, which
+	 * confirms at we are communicating properly and that the device
+	 * is a 561. */
+	reg_r(dev, 0x8104, &data1, 1);
+	reg_r(dev, 0x8105, &data2, 1);
+	vendor = (data2 << 8) | data1;
+	reg_r(dev, 0x8106, &data1, 1);
+	reg_r(dev, 0x8107, &data2, 1);
+	product = (data2 << 8) | data1;
+	if (vendor != id->idVendor || product != id->idProduct) {
+		PDEBUG(D_PROBE, "Bad vendor / product from device");
+		return -EINVAL;
+	}
+	switch (product) {
+	case 0x0928:
+	case 0x0929:
+	case 0x092a:
+	case 0x092b:
+	case 0x092c:
+	case 0x092d:
+	case 0x092e:
+	case 0x092f:
+	case 0x403b:
+		sd->chip_revision = Rev012A;
+		break;
+	default:
+/*	case 0x0561:
+	case 0x0815:			* ?? in spca508.c
+	case 0x401a:
+	case 0x7004:
+	case 0x7e50:
+	case 0xa001:
+	case 0xcdee: */
+		sd->chip_revision = Rev072A;
+		break;
+	}
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+	gspca_dev->nbalt = 7 + 1;	/* choose alternate 7 first */
+	cam->cam_mode = sif_mode;
+	cam->nmodes = sizeof sif_mode / sizeof sif_mode[0];
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->autogain = sd_ctrls[SD_AUTOGAIN].qctrl.default_value;
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	switch (sd->chip_revision) {
+	case Rev072A:
+		PDEBUG(D_STREAM, "Chip revision id: 072a");
+		write_vector(gspca_dev, spca561_init_data);
+		break;
+	default:
+/*	case Rev012A: */
+		PDEBUG(D_STREAM, "Chip revision id: 012a");
+		init_161rev12A(gspca_dev);
+		break;
+	}
+	return 0;
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 lowb;
+	int expotimes;
+
+	switch (sd->chip_revision) {
+	case Rev072A:
+		lowb = sd->contrast >> 8;
+		reg_w_val(dev, lowb, 0x8651);
+		reg_w_val(dev, lowb, 0x8652);
+		reg_w_val(dev, lowb, 0x8653);
+		reg_w_val(dev, lowb, 0x8654);
+		break;
+	case Rev012A: {
+		__u8 Reg8391[] =
+			{ 0x00, 0x00, 0x00, 0x00, 0x0c, 0x00, 0x00, 0x00 };
+
+		/* Write camera sensor settings */
+		expotimes = (sd->contrast >> 5) & 0x07ff;
+		Reg8391[0] = expotimes & 0xff;	/* exposure */
+		Reg8391[1] = 0x18 | (expotimes >> 8);
+		Reg8391[2] = sd->brightness;	/* gain */
+		reg_w_buf(dev, 0x8391, Reg8391, 8);
+		reg_w_buf(dev, 0x8390, Reg8391, 8);
+		break;
+	    }
+	}
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int Clck;
+	__u8 Reg8307[] = { 0xaa, 0x00 };
+	int mode;
+
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+	switch (sd->chip_revision) {
+	case Rev072A:
+		switch (mode) {
+		default:
+/*		case 0:
+		case 1: */
+			Clck = 0x25;
+			break;
+		case 2:
+			Clck = 0x22;
+			break;
+		case 3:
+			Clck = 0x21;
+			break;
+		}
+		reg_w_val(dev, 0x8500, mode);	/* mode */
+		reg_w_val(dev, 0x8700, Clck);	/* 0x27 clock */
+		reg_w_val(dev, 0x8112, 0x10 | 0x20);
+		break;
+	default:
+/*	case Rev012A: */
+		switch (mode) {
+		case 0:
+		case 1:
+			Clck = 0x8a;
+			break;
+		case 2:
+			Clck = 0x85;
+			break;
+		default:
+			Clck = 0x83;
+			break;
+		}
+		if (mode <= 1) {
+			/* Use compression on 320x240 and above */
+			reg_w_val(dev, 0x8500, 0x10 | mode);
+		} else {
+			/* I couldn't get the compression to work below 320x240
+			 * Fortunately at these resolutions the bandwidth
+			 * is sufficient to push raw frames at ~20fps */
+			reg_w_val(dev, 0x8500, mode);
+		}		/* -- qq@kuku.eu.org */
+		reg_w_buf(dev, 0x8307, Reg8307, 2);
+		reg_w_val(dev, 0x8700, Clck);	/* 0x8f 0x85 0x27 clock */
+		reg_w_val(dev, 0x8112, 0x1e | 0x20);
+		reg_w_val(dev, 0x850b, 0x03);
+		setcontrast(gspca_dev);
+		break;
+	}
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	reg_w_val(gspca_dev->dev, 0x8112, 0x20);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+/* this function is called at close time */
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+	reg_w_val(gspca_dev->dev, 0x8114, 0);
+}
+
+static void setautogain(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int expotimes = 0;
+	int pixelclk = 0;
+	int gainG = 0;
+	__u8 R, Gr, Gb, B;
+	int y;
+	__u8 luma_mean = 110;
+	__u8 luma_delta = 20;
+	__u8 spring = 4;
+
+	switch (sd->chip_revision) {
+	case Rev072A:
+		reg_r(gspca_dev->dev, 0x8621, &Gr, 1);
+		reg_r(gspca_dev->dev, 0x8622, &R, 1);
+		reg_r(gspca_dev->dev, 0x8623, &B, 1);
+		reg_r(gspca_dev->dev, 0x8624, &Gb, 1);
+		y = (77 * R + 75 * (Gr + Gb) + 29 * B) >> 8;
+		/* u= (128*B-(43*(Gr+Gb+R))) >> 8; */
+		/* v= (128*R-(53*(Gr+Gb))-21*B) >> 8; */
+		/* PDEBUG(D_CONF,"reading Y %d U %d V %d ",y,u,v); */
+
+		if (y < luma_mean - luma_delta ||
+		    y > luma_mean + luma_delta) {
+			expotimes = i2c_read(gspca_dev, 0x09, 0x10);
+			pixelclk = 0x0800;
+			expotimes = expotimes & 0x07ff;
+			/* PDEBUG(D_PACK,
+				"Exposition Times 0x%03X Clock 0x%04X ",
+				expotimes,pixelclk); */
+			gainG = i2c_read(gspca_dev, 0x35, 0x10);
+			/* PDEBUG(D_PACK,
+				"reading Gain register %d", gainG); */
+
+			expotimes += (luma_mean - y) >> spring;
+			gainG += (luma_mean - y) / 50;
+			/* PDEBUG(D_PACK,
+				"compute expotimes %d gain %d",
+				expotimes,gainG); */
+
+			if (gainG > 0x3f)
+				gainG = 0x3f;
+			else if (gainG < 4)
+				gainG = 3;
+			i2c_write(gspca_dev, gainG, 0x35);
+
+			if (expotimes >= 0x0256)
+				expotimes = 0x0256;
+			else if (expotimes < 4)
+				expotimes = 3;
+			i2c_write(gspca_dev, expotimes | pixelclk, 0x09);
+		}
+		break;
+	case Rev012A:
+		/* sensor registers is access and memory mapped to 0x8300 */
+		/* readind all 0x83xx block the sensor */
+		/*
+		 * The data from the header seem wrong where is the luma
+		 * and chroma mean value
+		 * at the moment set exposure in contrast set
+		 */
+		break;
+	}
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame, /* target */
+			__u8 *data,		/* isoc packet */
+			int len)		/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	switch (data[0]) {
+	case 0:		/* start of frame */
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					data, 0);
+		if (sd->ag_cnt >= 0) {
+			if (--sd->ag_cnt < 0) {
+				sd->ag_cnt = AG_CNT_START;
+				setautogain(gspca_dev);
+			}
+		}
+		data += SPCA561_OFFSET_DATA;
+		len -= SPCA561_OFFSET_DATA;
+		if (data[1] & 0x10) {
+			/* compressed bayer */
+			gspca_frame_add(gspca_dev, FIRST_PACKET,
+					frame, data, len);
+		} else {
+			/*fixme: which format?*/
+			data += 20;
+			len -= 20;
+			gspca_frame_add(gspca_dev, FIRST_PACKET,
+						frame, data, len);
+		}
+		return;
+	case 0xff:		/* drop */
+/*		gspca_dev->last_packet_type = DISCARD_PACKET; */
+		return;
+	}
+	data++;
+	len--;
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 value;
+
+	switch (sd->chip_revision) {
+	case Rev072A:
+		value = sd->brightness;
+		reg_w_val(gspca_dev->dev, value, 0x8611);
+		reg_w_val(gspca_dev->dev, value, 0x8612);
+		reg_w_val(gspca_dev->dev, value, 0x8613);
+		reg_w_val(gspca_dev->dev, value, 0x8614);
+		break;
+	default:
+/*	case Rev012A: */
+		setcontrast(gspca_dev);
+		break;
+	}
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 value;
+	__u16 tot;
+
+	switch (sd->chip_revision) {
+	case Rev072A:
+		tot = 0;
+		reg_r(gspca_dev->dev, 0x8611, &value, 1);
+		tot += value;
+		reg_r(gspca_dev->dev, 0x8612, &value, 1);
+		tot += value;
+		reg_r(gspca_dev->dev, 0x8613, &value, 1);
+		tot += value;
+		reg_r(gspca_dev->dev, 0x8614, &value, 1);
+		tot += value;
+		sd->brightness = tot >> 2;
+		break;
+	default:
+/*	case Rev012A: */
+		/* no way to read sensor settings */
+		break;
+	}
+}
+
+static void getcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 value;
+	__u16 tot;
+
+	switch (sd->chip_revision) {
+	case Rev072A:
+		tot = 0;
+		reg_r(gspca_dev->dev, 0x8651, &value, 1);
+		tot += value;
+		reg_r(gspca_dev->dev, 0x8652, &value, 1);
+		tot += value;
+		reg_r(gspca_dev->dev, 0x8653, &value, 1);
+		tot += value;
+		reg_r(gspca_dev->dev, 0x8654, &value, 1);
+		tot += value;
+		sd->contrast = tot << 6;
+		break;
+	default:
+/*	case Rev012A: */
+		/* no way to read sensor settings */
+		break;
+	}
+	PDEBUG(D_CONF, "get contrast %d", sd->contrast);
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcontrast(gspca_dev);
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->autogain = val;
+	if (val)
+		sd->ag_cnt = AG_CNT_START;
+	else
+		sd->ag_cnt = -1;
+	return 0;
+}
+
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autogain;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x041e, 0x401a), DVNM("Creative Webcam Vista (PD1100)")},
+	{USB_DEVICE(0x041e, 0x403b),  DVNM("Creative Webcam Vista (VF0010)")},
+	{USB_DEVICE(0x0458, 0x7004), DVNM("Genius VideoCAM Express V2")},
+	{USB_DEVICE(0x046d, 0x0928), DVNM("Logitech QC Express Etch2")},
+	{USB_DEVICE(0x046d, 0x0929), DVNM("Labtec Webcam Elch2")},
+	{USB_DEVICE(0x046d, 0x092a), DVNM("Logitech QC for Notebook")},
+	{USB_DEVICE(0x046d, 0x092b), DVNM("Labtec Webcam Plus")},
+	{USB_DEVICE(0x046d, 0x092c), DVNM("Logitech QC chat Elch2")},
+	{USB_DEVICE(0x046d, 0x092d), DVNM("Logitech QC Elch2")},
+	{USB_DEVICE(0x046d, 0x092e), DVNM("Logitech QC Elch2")},
+	{USB_DEVICE(0x046d, 0x092f), DVNM("Logitech QC Elch2")},
+	{USB_DEVICE(0x04fc, 0x0561), DVNM("Flexcam 100")},
+	{USB_DEVICE(0x060b, 0xa001), DVNM("Maxell Compact Pc PM3")},
+	{USB_DEVICE(0x10fd, 0x7e50), DVNM("FlyCam Usb 100")},
+	{USB_DEVICE(0xabcd, 0xcdee), DVNM("Petcam")},
+	{}
+};
+
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+		    const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+			       THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/stk014.c b/drivers/media/video/gspca/stk014.c
index d8c203e99cd3..6832fe0f3403 100644
--- a/drivers/media/video/gspca/stk014.c
+++ b/drivers/media/video/gspca/stk014.c
@@ -23,8 +23,8 @@
 #include "gspca.h"
 #include "jpeg.h"
 
-#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(0, 2, 7)
-static const char version[] = "0.2.7";
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
 
 MODULE_AUTHOR("Jean-Francois Moine <http://moinejf.free.fr>");
 MODULE_DESCRIPTION("Syntek DV4000 (STK014) USB Camera Driver");
@@ -37,10 +37,10 @@ struct sd {
 	unsigned char brightness;
 	unsigned char contrast;
 	unsigned char colors;
+	unsigned char lightfreq;
 };
 
 /* global parameters */
-static int lightfreq = 50;
 static int sd_quant = 7;		/* <= 4 KO - 7: good (enough!) */
 
 /* V4L2 controls supported by the driver */
@@ -50,6 +50,8 @@ static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
 static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val);
 
 static struct ctrl sd_ctrls[] = {
 #define SD_BRIGHTNESS 0
@@ -94,6 +96,20 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setcolors,
 	    .get = sd_getcolors,
 	},
+#define SD_FREQ 3
+	{
+	    {
+		.id	 = V4L2_CID_POWER_LINE_FREQUENCY,
+		.type    = V4L2_CTRL_TYPE_MENU,
+		.name    = "Light frequency filter",
+		.minimum = 1,
+		.maximum = 2,	/* 0: 0, 1: 50Hz, 2:60Hz */
+		.step    = 1,
+		.default_value = 1,
+	    },
+	    .set = sd_setfreq,
+	    .get = sd_getfreq,
+	},
 };
 
 static struct cam_mode vga_mode[] = {
@@ -102,11 +118,11 @@ static struct cam_mode vga_mode[] = {
 };
 
 /* -- read a register -- */
-static int reg_read(struct gspca_dev *gspca_dev,
+static int reg_r(struct gspca_dev *gspca_dev,
 			__u16 index, __u8 *buf)
 {
-	int ret;
 	struct usb_device *dev = gspca_dev->dev;
+	int ret;
 
 	ret = usb_control_msg(dev, usb_rcvctrlpipe(dev, 0),
 			0x00,
@@ -116,12 +132,12 @@ static int reg_read(struct gspca_dev *gspca_dev,
 			buf, 1,
 			500);
 	if (ret < 0)
-		PDEBUG(D_ERR, "reg_read err %d", ret);
+		PDEBUG(D_ERR, "reg_r err %d", ret);
 	return ret;
 }
 
 /* -- write a register -- */
-static int reg_write(struct gspca_dev *gspca_dev,
+static int reg_w(struct gspca_dev *gspca_dev,
 			__u16 index, __u16 value)
 {
 	struct usb_device *dev = gspca_dev->dev;
@@ -136,7 +152,7 @@ static int reg_write(struct gspca_dev *gspca_dev,
 			0,
 			500);
 	if (ret < 0)
-		PDEBUG(D_ERR, "reg_write err %d", ret);
+		PDEBUG(D_ERR, "reg_w err %d", ret);
 	return ret;
 }
 
@@ -149,15 +165,15 @@ static int rcv_val(struct gspca_dev *gspca_dev,
 	int alen, ret;
 	unsigned char bulk_buf[4];
 
-	reg_write(gspca_dev, 0x634, (ads >> 16) & 0xff);
-	reg_write(gspca_dev, 0x635, (ads >> 8) & 0xff);
-	reg_write(gspca_dev, 0x636, ads & 0xff);
-	reg_write(gspca_dev, 0x637, 0);
-	reg_write(gspca_dev, 0x638, len & 0xff);
-	reg_write(gspca_dev, 0x639, len >> 8);
-	reg_write(gspca_dev, 0x63a, 0);
-	reg_write(gspca_dev, 0x63b, 0);
-	reg_write(gspca_dev, 0x630, 5);
+	reg_w(gspca_dev, 0x634, (ads >> 16) & 0xff);
+	reg_w(gspca_dev, 0x635, (ads >> 8) & 0xff);
+	reg_w(gspca_dev, 0x636, ads & 0xff);
+	reg_w(gspca_dev, 0x637, 0);
+	reg_w(gspca_dev, 0x638, len & 0xff);
+	reg_w(gspca_dev, 0x639, len >> 8);
+	reg_w(gspca_dev, 0x63a, 0);
+	reg_w(gspca_dev, 0x63b, 0);
+	reg_w(gspca_dev, 0x630, 5);
 	if (len > sizeof bulk_buf)
 		return -1;
 	ret = usb_bulk_msg(dev,
@@ -180,26 +196,26 @@ static int snd_val(struct gspca_dev *gspca_dev,
 	unsigned char bulk_buf[4];
 
 	if (ads == 0x003f08) {
-		ret = reg_read(gspca_dev, 0x0704, &value);
+		ret = reg_r(gspca_dev, 0x0704, &value);
 		if (ret < 0)
 			goto ko;
-		ret = reg_read(gspca_dev, 0x0705, &seq);
+		ret = reg_r(gspca_dev, 0x0705, &seq);
 		if (ret < 0)
 			goto ko;
-		ret = reg_read(gspca_dev, 0x0650, &value);
+		ret = reg_r(gspca_dev, 0x0650, &value);
 		if (ret < 0)
 			goto ko;
-		reg_write(gspca_dev, 0x654, seq);
+		reg_w(gspca_dev, 0x654, seq);
 	} else
-		reg_write(gspca_dev, 0x654, (ads >> 16) & 0xff);
-	reg_write(gspca_dev, 0x655, (ads >> 8) & 0xff);
-	reg_write(gspca_dev, 0x656, ads & 0xff);
-	reg_write(gspca_dev, 0x657, 0);
-	reg_write(gspca_dev, 0x658, 0x04);	/* size */
-	reg_write(gspca_dev, 0x659, 0);
-	reg_write(gspca_dev, 0x65a, 0);
-	reg_write(gspca_dev, 0x65b, 0);
-	reg_write(gspca_dev, 0x650, 5);
+		reg_w(gspca_dev, 0x654, (ads >> 16) & 0xff);
+	reg_w(gspca_dev, 0x655, (ads >> 8) & 0xff);
+	reg_w(gspca_dev, 0x656, ads & 0xff);
+	reg_w(gspca_dev, 0x657, 0);
+	reg_w(gspca_dev, 0x658, 0x04);	/* size */
+	reg_w(gspca_dev, 0x659, 0);
+	reg_w(gspca_dev, 0x65a, 0);
+	reg_w(gspca_dev, 0x65b, 0);
+	reg_w(gspca_dev, 0x650, 5);
 	bulk_buf[0] = (val >> 24) & 0xff;
 	bulk_buf[1] = (val >> 16) & 0xff;
 	bulk_buf[2] = (val >> 8) & 0xff;
@@ -215,7 +231,7 @@ static int snd_val(struct gspca_dev *gspca_dev,
 	if (ads == 0x003f08) {
 		seq += 4;
 		seq &= 0x3f;
-		reg_write(gspca_dev, 0x705, seq);
+		reg_w(gspca_dev, 0x705, seq);
 	}
 	return ret;
 ko:
@@ -235,7 +251,6 @@ static void setbrightness(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 	int parval;
 
-	PDEBUG(D_CONF, "brightness: %d", sd->brightness);
 	parval = 0x06000000		/* whiteness */
 		+ (sd->brightness << 16);
 	set_par(gspca_dev, parval);
@@ -246,7 +261,6 @@ static void setcontrast(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 	int parval;
 
-	PDEBUG(D_CONF, "contrast: %d", sd->contrast);
 	parval = 0x07000000		/* contrast */
 		+ (sd->contrast << 16);
 	set_par(gspca_dev, parval);
@@ -257,13 +271,20 @@ static void setcolors(struct gspca_dev *gspca_dev)
 	struct sd *sd = (struct sd *) gspca_dev;
 	int parval;
 
-	PDEBUG(D_CONF, "saturation: %d",
-		sd->colors);
 	parval = 0x08000000		/* saturation */
 		+ (sd->colors << 16);
 	set_par(gspca_dev, parval);
 }
 
+static void setfreq(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	set_par(gspca_dev, sd->lightfreq == 1
+			? 0x33640000		/* 50 Hz */
+			: 0x33780000);		/* 60 Hz */
+}
+
 /* this function is called at probe time */
 static int sd_config(struct gspca_dev *gspca_dev,
 			const struct usb_device_id *id)
@@ -278,6 +299,7 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
 	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
 	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	sd->lightfreq = sd_ctrls[SD_FREQ].qctrl.default_value;
 	return 0;
 }
 
@@ -289,7 +311,7 @@ static int sd_open(struct gspca_dev *gspca_dev)
 
 	/* check if the device responds */
 	usb_set_interface(gspca_dev->dev, gspca_dev->iface, 1);
-	ret = reg_read(gspca_dev, 0x0740, &value);
+	ret = reg_r(gspca_dev, 0x0740, &value);
 	if (ret < 0)
 		return ret;
 	if (value != 0xff) {
@@ -320,21 +342,24 @@ static void sd_start(struct gspca_dev *gspca_dev)
 	ret = usb_set_interface(gspca_dev->dev,
 					gspca_dev->iface,
 					gspca_dev->alt);
-	if (ret < 0)
+	if (ret < 0) {
+		PDEBUG(D_ERR|D_STREAM, "set intf %d %d failed",
+			gspca_dev->iface, gspca_dev->alt);
 		goto out;
-	ret = reg_read(gspca_dev, 0x0630, &dum);
+	}
+	ret = reg_r(gspca_dev, 0x0630, &dum);
 	if (ret < 0)
 		goto out;
 	rcv_val(gspca_dev, 0x000020, 4);	/* << (value ff ff ff ff) */
-	ret = reg_read(gspca_dev, 0x0650, &dum);
+	ret = reg_r(gspca_dev, 0x0650, &dum);
 	if (ret < 0)
 		goto out;
 	snd_val(gspca_dev, 0x000020, 0xffffffff);
-	reg_write(gspca_dev, 0x0620, 0);
-	reg_write(gspca_dev, 0x0630, 0);
-	reg_write(gspca_dev, 0x0640, 0);
-	reg_write(gspca_dev, 0x0650, 0);
-	reg_write(gspca_dev, 0x0660, 0);
+	reg_w(gspca_dev, 0x0620, 0);
+	reg_w(gspca_dev, 0x0630, 0);
+	reg_w(gspca_dev, 0x0640, 0);
+	reg_w(gspca_dev, 0x0650, 0);
+	reg_w(gspca_dev, 0x0660, 0);
 	setbrightness(gspca_dev);		/* whiteness */
 	setcontrast(gspca_dev);			/* contrast */
 	setcolors(gspca_dev);			/* saturation */
@@ -342,9 +367,7 @@ static void sd_start(struct gspca_dev *gspca_dev)
 	set_par(gspca_dev, 0x0a800000);		/* Green ? */
 	set_par(gspca_dev, 0x0b800000);		/* Blue ? */
 	set_par(gspca_dev, 0x0d030000);		/* Gamma ? */
-	set_par(gspca_dev, lightfreq == 60
-			? 0x33780000		/* 60 Hz */
-			: 0x33640000);		/* 50 Hz */
+	setfreq(gspca_dev);			/* light frequency */
 
 	/* start the video flow */
 	set_par(gspca_dev, 0x01000000);
@@ -363,15 +386,15 @@ static void sd_stopN(struct gspca_dev *gspca_dev)
 	set_par(gspca_dev, 0x02000000);
 	set_par(gspca_dev, 0x02000000);
 	usb_set_interface(dev, gspca_dev->iface, 1);
-	reg_read(gspca_dev, 0x0630, &value);
+	reg_r(gspca_dev, 0x0630, &value);
 	rcv_val(gspca_dev, 0x000020, 4);	/* << (value ff ff ff ff) */
-	reg_read(gspca_dev, 0x0650, &value);
+	reg_r(gspca_dev, 0x0650, &value);
 	snd_val(gspca_dev, 0x000020, 0xffffffff);
-	reg_write(gspca_dev, 0x0620, 0);
-	reg_write(gspca_dev, 0x0630, 0);
-	reg_write(gspca_dev, 0x0640, 0);
-	reg_write(gspca_dev, 0x0650, 0);
-	reg_write(gspca_dev, 0x0660, 0);
+	reg_w(gspca_dev, 0x0620, 0);
+	reg_w(gspca_dev, 0x0630, 0);
+	reg_w(gspca_dev, 0x0640, 0);
+	reg_w(gspca_dev, 0x0650, 0);
+	reg_w(gspca_dev, 0x0660, 0);
 	PDEBUG(D_STREAM, "camera stopped");
 }
 
@@ -470,6 +493,42 @@ static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
 	return 0;
 }
 
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->lightfreq = val;
+	if (gspca_dev->streaming)
+		setfreq(gspca_dev);
+	return 0;
+}
+
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->lightfreq;
+	return 0;
+}
+
+static int sd_querymenu(struct gspca_dev *gspca_dev,
+			struct v4l2_querymenu *menu)
+{
+	switch (menu->id) {
+	case V4L2_CID_POWER_LINE_FREQUENCY:
+		switch (menu->index) {
+		case 1:		/* V4L2_CID_POWER_LINE_FREQUENCY_50HZ */
+			strcpy(menu->name, "50 Hz");
+			return 0;
+		case 2:		/* V4L2_CID_POWER_LINE_FREQUENCY_60HZ */
+			strcpy(menu->name, "60 Hz");
+			return 0;
+		}
+		break;
+	}
+	return -EINVAL;
+}
+
 /* sub-driver description */
 static struct sd_desc sd_desc = {
 	.name = MODULE_NAME,
@@ -482,6 +541,7 @@ static struct sd_desc sd_desc = {
 	.stop0 = sd_stop0,
 	.close = sd_close,
 	.pkt_scan = sd_pkt_scan,
+	.querymenu = sd_querymenu,
 };
 
 /* -- module initialisation -- */
@@ -524,7 +584,5 @@ static void __exit sd_mod_exit(void)
 module_init(sd_mod_init);
 module_exit(sd_mod_exit);
 
-module_param(lightfreq, int, 0644);
-MODULE_PARM_DESC(lightfreq, "Light frequency 50 or 60 Hz");
 module_param_named(quant, sd_quant, int, 0644);
 MODULE_PARM_DESC(quant, "Quantization index (0..8)");
diff --git a/drivers/media/video/gspca/sunplus.c b/drivers/media/video/gspca/sunplus.c
new file mode 100644
index 000000000000..52d1b32523b1
--- /dev/null
+++ b/drivers/media/video/gspca/sunplus.c
@@ -0,0 +1,1638 @@
+/*
+ *		Sunplus spca504(abc) spca533 spca536 library
+ *		Copyright (C) 2005 Michel Xhaard mxhaard@magic.fr
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "sunplus"
+
+#include "gspca.h"
+#include "jpeg.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/SPCA5xx USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned char packet[ISO_MAX_SIZE + 128];
+				/* !! no more than 128 ff in an ISO packet */
+
+	unsigned char brightness;
+	unsigned char contrast;
+	unsigned char colors;
+	unsigned char autogain;
+
+	char qindex;
+	char bridge;
+#define BRIDGE_SPCA504 0
+#define BRIDGE_SPCA504B 1
+#define BRIDGE_SPCA504C 2
+#define BRIDGE_SPCA533 3
+#define BRIDGE_SPCA536 4
+	char subtype;
+#define AiptekMiniPenCam13 1
+#define LogitechClickSmart420 2
+#define LogitechClickSmart820 3
+#define MegapixV4 4
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	    {
+		.id      = V4L2_CID_BRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Brightness",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0,
+	    },
+	    .set = sd_setbrightness,
+	    .get = sd_getbrightness,
+	},
+#define SD_CONTRAST 1
+	{
+	    {
+		.id      = V4L2_CID_CONTRAST,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Contrast",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0x20,
+	    },
+	    .set = sd_setcontrast,
+	    .get = sd_getcontrast,
+	},
+#define SD_COLOR 2
+	{
+	    {
+		.id      = V4L2_CID_SATURATION,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Color",
+		.minimum = 0,
+		.maximum = 0xff,
+		.step    = 1,
+		.default_value = 0x1a,
+	    },
+	    .set = sd_setcolors,
+	    .get = sd_getcolors,
+	},
+#define SD_AUTOGAIN 3
+	{
+	    {
+		.id      = V4L2_CID_AUTOGAIN,
+		.type    = V4L2_CTRL_TYPE_BOOLEAN,
+		.name    = "Auto Gain",
+		.minimum = 0,
+		.maximum = 1,
+		.step    = 1,
+		.default_value = 1,
+	    },
+	    .set = sd_setautogain,
+	    .get = sd_getautogain,
+	},
+};
+
+static struct cam_mode vga_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 320, 240, 2},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 1},
+};
+
+static struct cam_mode custom_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 320, 240, 2},
+	{V4L2_PIX_FMT_JPEG, 464, 480, 1},
+};
+
+static struct cam_mode vga_mode2[] = {
+	{V4L2_PIX_FMT_JPEG, 176, 144, 4},
+	{V4L2_PIX_FMT_JPEG, 320, 240, 3},
+	{V4L2_PIX_FMT_JPEG, 352, 288, 2},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 1},
+};
+
+#define SPCA50X_OFFSET_DATA 10
+#define SPCA504_PCCAM600_OFFSET_SNAPSHOT 3
+#define SPCA504_PCCAM600_OFFSET_COMPRESS 4
+#define SPCA504_PCCAM600_OFFSET_MODE	 5
+#define SPCA504_PCCAM600_OFFSET_DATA	 14
+ /* Frame packet header offsets for the spca533 */
+#define SPCA533_OFFSET_DATA      16
+#define SPCA533_OFFSET_FRAMSEQ	15
+/* Frame packet header offsets for the spca536 */
+#define SPCA536_OFFSET_DATA      4
+#define SPCA536_OFFSET_FRAMSEQ	 1
+
+/* Initialisation data for the Creative PC-CAM 600 */
+static __u16 spca504_pccam600_init_data[][3] = {
+/*	{0xa0, 0x0000, 0x0503},  * capture mode */
+	{0x00, 0x0000, 0x2000},
+	{0x00, 0x0013, 0x2301},
+	{0x00, 0x0003, 0x2000},
+	{0x00, 0x0001, 0x21ac},
+	{0x00, 0x0001, 0x21a6},
+	{0x00, 0x0000, 0x21a7},	/* brightness */
+	{0x00, 0x0020, 0x21a8},	/* contrast */
+	{0x00, 0x0001, 0x21ac},	/* sat/hue */
+	{0x00, 0x0000, 0x21ad},	/* hue */
+	{0x00, 0x001a, 0x21ae},	/* saturation */
+	{0x00, 0x0002, 0x21a3},	/* gamma */
+	{0x30, 0x0154, 0x0008},
+	{0x30, 0x0004, 0x0006},
+	{0x30, 0x0258, 0x0009},
+	{0x30, 0x0004, 0x0000},
+	{0x30, 0x0093, 0x0004},
+	{0x30, 0x0066, 0x0005},
+	{0x00, 0x0000, 0x2000},
+	{0x00, 0x0013, 0x2301},
+	{0x00, 0x0003, 0x2000},
+	{0x00, 0x0013, 0x2301},
+	{0x00, 0x0003, 0x2000},
+	{}
+};
+
+/* Creative PC-CAM 600 specific open data, sent before using the
+ * generic initialisation data from spca504_open_data.
+ */
+static __u16 spca504_pccam600_open_data[][3] = {
+	{0x00, 0x0001, 0x2501},
+	{0x20, 0x0500, 0x0001},	/* snapshot mode */
+	{0x00, 0x0003, 0x2880},
+	{0x00, 0x0001, 0x2881},
+	{}
+};
+
+/* Initialisation data for the logitech clicksmart 420 */
+static __u16 spca504A_clicksmart420_init_data[][3] = {
+/*	{0xa0, 0x0000, 0x0503},  * capture mode */
+	{0x00, 0x0000, 0x2000},
+	{0x00, 0x0013, 0x2301},
+	{0x00, 0x0003, 0x2000},
+	{0x00, 0x0001, 0x21ac},
+	{0x00, 0x0001, 0x21a6},
+	{0x00, 0x0000, 0x21a7},	/* brightness */
+	{0x00, 0x0020, 0x21a8},	/* contrast */
+	{0x00, 0x0001, 0x21ac},	/* sat/hue */
+	{0x00, 0x0000, 0x21ad},	/* hue */
+	{0x00, 0x001a, 0x21ae},	/* saturation */
+	{0x00, 0x0002, 0x21a3},	/* gamma */
+	{0x30, 0x0004, 0x000a},
+	{0xb0, 0x0001, 0x0000},
+
+
+	{0x0a1, 0x0080, 0x0001},
+	{0x30, 0x0049, 0x0000},
+	{0x30, 0x0060, 0x0005},
+	{0x0c, 0x0004, 0x0000},
+	{0x00, 0x0000, 0x0000},
+	{0x00, 0x0000, 0x2000},
+	{0x00, 0x0013, 0x2301},
+	{0x00, 0x0003, 0x2000},
+	{0x00, 0x0000, 0x2000},
+
+	{}
+};
+
+/* clicksmart 420 open data ? */
+static __u16 spca504A_clicksmart420_open_data[][3] = {
+	{0x00, 0x0001, 0x2501},
+	{0x20, 0x0502, 0x0000},
+	{0x06, 0x0000, 0x0000},
+	{0x00, 0x0004, 0x2880},
+	{0x00, 0x0001, 0x2881},
+/* look like setting a qTable */
+	{0x00, 0x0006, 0x2800},
+	{0x00, 0x0004, 0x2801},
+	{0x00, 0x0004, 0x2802},
+	{0x00, 0x0006, 0x2803},
+	{0x00, 0x000a, 0x2804},
+	{0x00, 0x0010, 0x2805},
+	{0x00, 0x0014, 0x2806},
+	{0x00, 0x0018, 0x2807},
+	{0x00, 0x0005, 0x2808},
+	{0x00, 0x0005, 0x2809},
+	{0x00, 0x0006, 0x280a},
+	{0x00, 0x0008, 0x280b},
+	{0x00, 0x000a, 0x280c},
+	{0x00, 0x0017, 0x280d},
+	{0x00, 0x0018, 0x280e},
+	{0x00, 0x0016, 0x280f},
+
+	{0x00, 0x0006, 0x2810},
+	{0x00, 0x0005, 0x2811},
+	{0x00, 0x0006, 0x2812},
+	{0x00, 0x000a, 0x2813},
+	{0x00, 0x0010, 0x2814},
+	{0x00, 0x0017, 0x2815},
+	{0x00, 0x001c, 0x2816},
+	{0x00, 0x0016, 0x2817},
+	{0x00, 0x0006, 0x2818},
+	{0x00, 0x0007, 0x2819},
+	{0x00, 0x0009, 0x281a},
+	{0x00, 0x000c, 0x281b},
+	{0x00, 0x0014, 0x281c},
+	{0x00, 0x0023, 0x281d},
+	{0x00, 0x0020, 0x281e},
+	{0x00, 0x0019, 0x281f},
+
+	{0x00, 0x0007, 0x2820},
+	{0x00, 0x0009, 0x2821},
+	{0x00, 0x000f, 0x2822},
+	{0x00, 0x0016, 0x2823},
+	{0x00, 0x001b, 0x2824},
+	{0x00, 0x002c, 0x2825},
+	{0x00, 0x0029, 0x2826},
+	{0x00, 0x001f, 0x2827},
+	{0x00, 0x000a, 0x2828},
+	{0x00, 0x000e, 0x2829},
+	{0x00, 0x0016, 0x282a},
+	{0x00, 0x001a, 0x282b},
+	{0x00, 0x0020, 0x282c},
+	{0x00, 0x002a, 0x282d},
+	{0x00, 0x002d, 0x282e},
+	{0x00, 0x0025, 0x282f},
+
+	{0x00, 0x0014, 0x2830},
+	{0x00, 0x001a, 0x2831},
+	{0x00, 0x001f, 0x2832},
+	{0x00, 0x0023, 0x2833},
+	{0x00, 0x0029, 0x2834},
+	{0x00, 0x0030, 0x2835},
+	{0x00, 0x0030, 0x2836},
+	{0x00, 0x0028, 0x2837},
+	{0x00, 0x001d, 0x2838},
+	{0x00, 0x0025, 0x2839},
+	{0x00, 0x0026, 0x283a},
+	{0x00, 0x0027, 0x283b},
+	{0x00, 0x002d, 0x283c},
+	{0x00, 0x0028, 0x283d},
+	{0x00, 0x0029, 0x283e},
+	{0x00, 0x0028, 0x283f},
+
+	{0x00, 0x0007, 0x2840},
+	{0x00, 0x0007, 0x2841},
+	{0x00, 0x000a, 0x2842},
+	{0x00, 0x0013, 0x2843},
+	{0x00, 0x0028, 0x2844},
+	{0x00, 0x0028, 0x2845},
+	{0x00, 0x0028, 0x2846},
+	{0x00, 0x0028, 0x2847},
+	{0x00, 0x0007, 0x2848},
+	{0x00, 0x0008, 0x2849},
+	{0x00, 0x000a, 0x284a},
+	{0x00, 0x001a, 0x284b},
+	{0x00, 0x0028, 0x284c},
+	{0x00, 0x0028, 0x284d},
+	{0x00, 0x0028, 0x284e},
+	{0x00, 0x0028, 0x284f},
+
+	{0x00, 0x000a, 0x2850},
+	{0x00, 0x000a, 0x2851},
+	{0x00, 0x0016, 0x2852},
+	{0x00, 0x0028, 0x2853},
+	{0x00, 0x0028, 0x2854},
+	{0x00, 0x0028, 0x2855},
+	{0x00, 0x0028, 0x2856},
+	{0x00, 0x0028, 0x2857},
+	{0x00, 0x0013, 0x2858},
+	{0x00, 0x001a, 0x2859},
+	{0x00, 0x0028, 0x285a},
+	{0x00, 0x0028, 0x285b},
+	{0x00, 0x0028, 0x285c},
+	{0x00, 0x0028, 0x285d},
+	{0x00, 0x0028, 0x285e},
+	{0x00, 0x0028, 0x285f},
+
+	{0x00, 0x0028, 0x2860},
+	{0x00, 0x0028, 0x2861},
+	{0x00, 0x0028, 0x2862},
+	{0x00, 0x0028, 0x2863},
+	{0x00, 0x0028, 0x2864},
+	{0x00, 0x0028, 0x2865},
+	{0x00, 0x0028, 0x2866},
+	{0x00, 0x0028, 0x2867},
+	{0x00, 0x0028, 0x2868},
+	{0x00, 0x0028, 0x2869},
+	{0x00, 0x0028, 0x286a},
+	{0x00, 0x0028, 0x286b},
+	{0x00, 0x0028, 0x286c},
+	{0x00, 0x0028, 0x286d},
+	{0x00, 0x0028, 0x286e},
+	{0x00, 0x0028, 0x286f},
+
+	{0x00, 0x0028, 0x2870},
+	{0x00, 0x0028, 0x2871},
+	{0x00, 0x0028, 0x2872},
+	{0x00, 0x0028, 0x2873},
+	{0x00, 0x0028, 0x2874},
+	{0x00, 0x0028, 0x2875},
+	{0x00, 0x0028, 0x2876},
+	{0x00, 0x0028, 0x2877},
+	{0x00, 0x0028, 0x2878},
+	{0x00, 0x0028, 0x2879},
+	{0x00, 0x0028, 0x287a},
+	{0x00, 0x0028, 0x287b},
+	{0x00, 0x0028, 0x287c},
+	{0x00, 0x0028, 0x287d},
+	{0x00, 0x0028, 0x287e},
+	{0x00, 0x0028, 0x287f},
+
+	{0xa0, 0x0000, 0x0503},
+	{}
+};
+
+static unsigned char qtable_creative_pccam[2][64] = {
+	{				/* Q-table Y-components */
+	 0x05, 0x03, 0x03, 0x05, 0x07, 0x0c, 0x0f, 0x12,
+	 0x04, 0x04, 0x04, 0x06, 0x08, 0x11, 0x12, 0x11,
+	 0x04, 0x04, 0x05, 0x07, 0x0c, 0x11, 0x15, 0x11,
+	 0x04, 0x05, 0x07, 0x09, 0x0f, 0x1a, 0x18, 0x13,
+	 0x05, 0x07, 0x0b, 0x11, 0x14, 0x21, 0x1f, 0x17,
+	 0x07, 0x0b, 0x11, 0x13, 0x18, 0x1f, 0x22, 0x1c,
+	 0x0f, 0x13, 0x17, 0x1a, 0x1f, 0x24, 0x24, 0x1e,
+	 0x16, 0x1c, 0x1d, 0x1d, 0x22, 0x1e, 0x1f, 0x1e},
+	{				/* Q-table C-components */
+	 0x05, 0x05, 0x07, 0x0e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x05, 0x06, 0x08, 0x14, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x07, 0x08, 0x11, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x0e, 0x14, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e}
+};
+
+/* FIXME: This Q-table is identical to the Creative PC-CAM one,
+ *		except for one byte. Possibly a typo?
+ *		NWG: 18/05/2003.
+ */
+static unsigned char qtable_spca504_default[2][64] = {
+	{				/* Q-table Y-components */
+	 0x05, 0x03, 0x03, 0x05, 0x07, 0x0c, 0x0f, 0x12,
+	 0x04, 0x04, 0x04, 0x06, 0x08, 0x11, 0x12, 0x11,
+	 0x04, 0x04, 0x05, 0x07, 0x0c, 0x11, 0x15, 0x11,
+	 0x04, 0x05, 0x07, 0x09, 0x0f, 0x1a, 0x18, 0x13,
+	 0x05, 0x07, 0x0b, 0x11, 0x14, 0x21, 0x1f, 0x17,
+	 0x07, 0x0b, 0x11, 0x13, 0x18, 0x1f, 0x22, 0x1c,
+	 0x0f, 0x13, 0x17, 0x1a, 0x1f, 0x24, 0x24, 0x1e,
+	 0x16, 0x1c, 0x1d, 0x1d, 0x1d /* 0x22 */ , 0x1e, 0x1f, 0x1e,
+	 },
+	{				/* Q-table C-components */
+	 0x05, 0x05, 0x07, 0x0e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x05, 0x06, 0x08, 0x14, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x07, 0x08, 0x11, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x0e, 0x14, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e,
+	 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e, 0x1e}
+};
+
+static void spca5xxRegRead(struct usb_device *dev,
+			   __u16 req,
+			   __u16 index,
+			   __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			req,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,		/* value */
+			index, buffer, length,
+			500);
+}
+
+static void spca5xxRegWrite(struct usb_device *dev,
+			    __u16 req,
+			    __u16 value,
+			    __u16 index,
+			    __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			req,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, buffer, length,
+			500);
+}
+
+static int reg_write(struct usb_device *dev,
+		     __u16 req, __u16 index, __u16 value)
+{
+	int ret;
+
+	ret = usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			req,
+			USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, NULL, 0, 500);
+	PDEBUG(D_PACK, "reg write: 0x%02x,0x%02x:0x%02x, 0x%x",
+		reg, index, value, ret);
+	if (ret < 0)
+		PDEBUG(D_ERR, "reg write: error %d", ret);
+	return ret;
+}
+
+static int reg_read_info(struct usb_device *dev,
+			__u16 value)	/* wValue */
+{
+	int ret;
+	__u8 data;
+
+	ret = usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0x20,			/* request */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value,
+			0,			/* index */
+			&data, 1,
+			500);			/* timeout */
+	if (ret < 0) {
+		PDEBUG(D_ERR, "reg_read_info err %d", ret);
+		return 0;
+	}
+	return data;
+}
+
+/* returns: negative is error, pos or zero is data */
+static int reg_read(struct usb_device *dev,
+			__u16 req,	/* bRequest */
+			__u16 index,	/* wIndex */
+			__u16 length)	/* wLength (1 or 2 only) */
+{
+	int ret;
+	__u8 buf[2];
+
+	buf[1] = 0;
+	ret = usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			req,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,		/* value */
+			index,
+			buf, length,
+			500);
+	if (ret < 0) {
+		PDEBUG(D_ERR, "reg_read err %d", ret);
+		return -1;
+	}
+	return (buf[1] << 8) + buf[0];
+}
+
+static int write_vector(struct gspca_dev *gspca_dev,
+				__u16 data[][3])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int ret, i = 0;
+
+	while (data[i][0] != 0 || data[i][1] != 0 || data[i][2] != 0) {
+		ret = reg_write(dev, data[i][0], data[i][2], data[i][1]);
+		if (ret < 0) {
+			PDEBUG(D_ERR,
+				"Register write failed for 0x%x,0x%x,0x%x",
+				data[i][0], data[i][1], data[i][2]);
+			return ret;
+		}
+		i++;
+	}
+	return 0;
+}
+
+static int spca50x_setup_qtable(struct gspca_dev *gspca_dev,
+				unsigned int request,
+				unsigned int ybase,
+				unsigned int cbase,
+				unsigned char qtable[2][64])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int i, err;
+
+	/* loop over y components */
+	for (i = 0; i < 64; i++) {
+		err = reg_write(dev, request, ybase + i, qtable[0][i]);
+		if (err < 0)
+			return err;
+	}
+
+	/* loop over c components */
+	for (i = 0; i < 64; i++) {
+		err = reg_write(dev, request, cbase + i, qtable[1][i]);
+		if (err < 0)
+			return err;
+	}
+	return 0;
+}
+
+static void spca504_acknowledged_command(struct gspca_dev *gspca_dev,
+			     __u16 req, __u16 idx, __u16 val)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 notdone;
+
+	reg_write(dev, req, idx, val);
+	notdone = reg_read(dev, 0x01, 0x0001, 1);
+	reg_write(dev, req, idx, val);
+
+	PDEBUG(D_FRAM, "before wait 0x%x", notdone);
+
+	msleep(200);
+	notdone = reg_read(dev, 0x01, 0x0001, 1);
+	PDEBUG(D_FRAM, "after wait 0x%x", notdone);
+}
+
+static void spca504A_acknowledged_command(struct gspca_dev *gspca_dev,
+			__u16 req,
+			__u16 idx, __u16 val, __u8 stat, __u8 count)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 status;
+	__u8 endcode;
+
+	reg_write(dev, req, idx, val);
+	status = reg_read(dev, 0x01, 0x0001, 1);
+	endcode = stat;
+	PDEBUG(D_FRAM, "Status 0x%x Need 0x%x", status, stat);
+	if (!count)
+		return;
+	count = 200;
+	while (--count > 0) {
+		msleep(10);
+		/* gsmart mini2 write a each wait setting 1 ms is enought */
+/*		reg_write(dev, req, idx, val); */
+		status = reg_read(dev, 0x01, 0x0001, 1);
+		if (status == endcode) {
+			PDEBUG(D_FRAM, "status 0x%x after wait 0x%x",
+				status, 200 - count);
+				break;
+		}
+	}
+}
+
+static int spca504B_PollingDataReady(struct usb_device *dev)
+{
+	__u8 DataReady;
+	int count = 10;
+
+	while (--count > 0) {
+		spca5xxRegRead(dev, 0x21, 0, &DataReady, 1);
+		if ((DataReady & 0x01) == 0)
+			break;
+		msleep(10);
+	}
+	return DataReady;
+}
+
+static void spca504B_WaitCmdStatus(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 DataReady;
+	int count = 50;
+
+	while (--count > 0) {
+		spca5xxRegRead(dev, 0x21, 1, &DataReady, 1);
+
+		if (DataReady) {
+			DataReady = 0;
+			spca5xxRegWrite(dev, 0x21, 0, 1, &DataReady, 1);
+			spca5xxRegRead(dev, 0x21, 1, &DataReady, 1);
+			spca504B_PollingDataReady(dev);
+			break;
+		}
+		msleep(10);
+	}
+}
+
+static void spca50x_GetFirmware(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 FW[5];
+	__u8 ProductInfo[64];
+
+	spca5xxRegRead(dev, 0x20, 0, FW, 5);
+	PDEBUG(D_STREAM, "FirmWare : %d %d %d %d %d ",
+		FW[0], FW[1], FW[2], FW[3], FW[4]);
+	spca5xxRegRead(dev, 0x23, 0, ProductInfo, 64);
+	spca5xxRegRead(dev, 0x23, 1, ProductInfo, 64);
+}
+
+static void spca504B_SetSizeType(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 Size;
+	__u8 Type;
+	int rc;
+
+	Size = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+	Type = 0;
+	switch (sd->bridge) {
+	case BRIDGE_SPCA533:
+		spca5xxRegWrite(dev, 0x31, 0, 0, NULL, 0);
+		spca504B_WaitCmdStatus(gspca_dev);
+		rc = spca504B_PollingDataReady(dev);
+		spca50x_GetFirmware(gspca_dev);
+		Type = 2;
+		spca5xxRegWrite(dev, 0x24, 0, 8, &Type, 1);
+		spca5xxRegRead(dev, 0x24, 8, &Type, 1);
+
+		spca5xxRegWrite(dev, 0x25, 0, 4, &Size, 1);
+		spca5xxRegRead(dev, 0x25, 4, &Size, 1);
+		rc = spca504B_PollingDataReady(dev);
+
+		/* Init the cam width height with some values get on init ? */
+		spca5xxRegWrite(dev, 0x31, 0, 4, NULL, 0);
+		spca504B_WaitCmdStatus(gspca_dev);
+		rc = spca504B_PollingDataReady(dev);
+		break;
+	default:
+/* case BRIDGE_SPCA504B: */
+/* case BRIDGE_SPCA536: */
+		Type = 6;
+		spca5xxRegWrite(dev, 0x25, 0, 4, &Size, 1);
+		spca5xxRegRead(dev, 0x25, 4, &Size, 1);
+		spca5xxRegWrite(dev, 0x27, 0, 0, &Type, 1);
+		spca5xxRegRead(dev, 0x27, 0, &Type, 1);
+		rc = spca504B_PollingDataReady(dev);
+		break;
+	case BRIDGE_SPCA504:
+		Size += 3;
+		if (sd->subtype == AiptekMiniPenCam13) {
+			/* spca504a aiptek */
+			spca504A_acknowledged_command(gspca_dev,
+						0x08, Size, 0,
+						0x80 | (Size & 0x0f), 1);
+			spca504A_acknowledged_command(gspca_dev,
+							1, 3, 0, 0x9f, 0);
+		} else {
+			spca504_acknowledged_command(gspca_dev, 0x08, Size, 0);
+		}
+		break;
+	case BRIDGE_SPCA504C:
+		/* capture mode */
+		reg_write(dev, 0xa0, (0x0500 | (Size & 0x0f)), 0x0);
+		reg_write(dev, 0x20, 0x01, 0x0500 | (Size & 0x0f));
+		break;
+	}
+}
+
+static void spca504_wait_status(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int cnt;
+
+	cnt = 256;
+	while (--cnt > 0) {
+		/* With this we get the status, when return 0 it's all ok */
+		if (reg_read(dev, 0x06, 0x00, 1) == 0)
+			return;
+		msleep(10);
+	}
+}
+
+static void spca504B_setQtable(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 Data = 3;
+
+	spca5xxRegWrite(dev, 0x26, 0, 0, &Data, 1);
+	spca5xxRegRead(dev, 0x26, 0, &Data, 1);
+	spca504B_PollingDataReady(dev);
+}
+
+static void sp5xx_initContBrigHueRegisters(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int pollreg = 1;
+
+	switch (sd->bridge) {
+	case BRIDGE_SPCA504:
+	case BRIDGE_SPCA504C:
+		pollreg = 0;
+		/* fall thru */
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA504B: */
+		spca5xxRegWrite(dev, 0, 0, 0x21a7, NULL, 0);	/* brightness */
+		spca5xxRegWrite(dev, 0, 0x20, 0x21a8, NULL, 0);	/* contrast */
+		spca5xxRegWrite(dev, 0, 0, 0x21ad, NULL, 0);	/* hue */
+		spca5xxRegWrite(dev, 0, 1, 0x21ac, NULL, 0);	/* sat/hue */
+		spca5xxRegWrite(dev, 0, 0x20, 0x21ae, NULL, 0);	/* saturation */
+		spca5xxRegWrite(dev, 0, 0, 0x21a3, NULL, 0);	/* gamma */
+		break;
+	case BRIDGE_SPCA536:
+		spca5xxRegWrite(dev, 0, 0, 0x20f0, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0x21, 0x20f1, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0x40, 0x20f5, NULL, 0);
+		spca5xxRegWrite(dev, 0, 1, 0x20f4, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0x40, 0x20f6, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0, 0x2089, NULL, 0);
+		break;
+	}
+	if (pollreg)
+		spca504B_PollingDataReady(dev);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	struct cam *cam;
+	__u16 vendor;
+	__u16 product;
+	__u8 fw;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+	switch (vendor) {
+	case 0x041e:		/* Creative cameras */
+/*		switch (product) { */
+/*		case 0x400b: */
+/*		case 0x4012: */
+/*		case 0x4013: */
+/*			sd->bridge = BRIDGE_SPCA504C; */
+/*			break; */
+/*		} */
+		break;
+	case 0x0458:		/* Genius KYE cameras */
+/*		switch (product) { */
+/*		case 0x7006: */
+			sd->bridge = BRIDGE_SPCA504B;
+/*			break; */
+/*		} */
+		break;
+	case 0x046d:		/* Logitech Labtec */
+		switch (product) {
+		case 0x0905:
+			sd->subtype = LogitechClickSmart820;
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x0960:
+			sd->subtype = LogitechClickSmart420;
+			sd->bridge = BRIDGE_SPCA504C;
+			break;
+		}
+		break;
+	case 0x0471:				/* Philips */
+/*		switch (product) { */
+/*		case 0x0322: */
+			sd->bridge = BRIDGE_SPCA504B;
+/*			break; */
+/*		} */
+		break;
+	case 0x04a5:		/* Benq */
+		switch (product) {
+		case 0x3003:
+			sd->bridge = BRIDGE_SPCA504B;
+			break;
+		case 0x3008:
+		case 0x300a:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		}
+		break;
+	case 0x04f1:		/* JVC */
+/*		switch (product) { */
+/*		case 0x1001: */
+			sd->bridge = BRIDGE_SPCA504B;
+/*			break; */
+/*		} */
+		break;
+	case 0x04fc:		/* SunPlus */
+		switch (product) {
+		case 0x500c:
+			sd->bridge = BRIDGE_SPCA504B;
+			break;
+		case 0x504a:
+/* try to get the firmware as some cam answer 2.0.1.2.2
+ * and should be a spca504b then overwrite that setting */
+			spca5xxRegRead(dev, 0x20, 0, &fw, 1);
+			if (fw == 1) {
+				sd->subtype = AiptekMiniPenCam13;
+				sd->bridge = BRIDGE_SPCA504;
+			} else if (fw == 2) {
+				sd->bridge = BRIDGE_SPCA504B;
+			} else
+				return -ENODEV;
+			break;
+		case 0x504b:
+			sd->bridge = BRIDGE_SPCA504B;
+			break;
+		case 0x5330:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x5360:
+			sd->bridge = BRIDGE_SPCA536;
+			break;
+		case 0xffff:
+			sd->bridge = BRIDGE_SPCA504B;
+			break;
+		}
+		break;
+	case 0x052b:		/* ?? Megapix */
+/*		switch (product) { */
+/*		case 0x1513: */
+			sd->subtype = MegapixV4;
+			sd->bridge = BRIDGE_SPCA533;
+/*			break; */
+/*		} */
+		break;
+	case 0x0546:		/* Polaroid */
+		switch (product) {
+		case 0x3155:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x3191:
+		case 0x3273:
+			sd->bridge = BRIDGE_SPCA504B;
+			break;
+		}
+		break;
+	case 0x055f:		/* Mustek cameras */
+		switch (product) {
+		case 0xc211:
+			sd->bridge = BRIDGE_SPCA536;
+			break;
+		case 0xc230:
+		case 0xc232:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0xc360:
+			sd->bridge = BRIDGE_SPCA536;
+			break;
+		case 0xc420:
+			sd->bridge = BRIDGE_SPCA504;
+			break;
+		case 0xc430:
+		case 0xc440:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0xc520:
+			sd->bridge = BRIDGE_SPCA504;
+			break;
+		case 0xc530:
+		case 0xc540:
+		case 0xc630:
+		case 0xc650:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		}
+		break;
+	case 0x05da:		/* Digital Dream cameras */
+/*		switch (product) { */
+/*		case 0x1018: */
+			sd->bridge = BRIDGE_SPCA504B;
+/*			break; */
+/*		} */
+		break;
+	case 0x06d6:		/* Trust */
+/*		switch (product) { */
+/*		case 0x0031: */
+			sd->bridge = BRIDGE_SPCA533;	/* SPCA533A */
+/*			break; */
+/*		} */
+		break;
+	case 0x0733:	/* Rebadged ViewQuest (Intel) and ViewQuest cameras */
+		switch (product) {
+		case 0x1311:
+		case 0x1314:
+		case 0x2211:
+		case 0x2221:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x3261:
+		case 0x3281:
+			sd->bridge = BRIDGE_SPCA536;
+			break;
+		}
+		break;
+	case 0x08ca:		/* Aiptek */
+		switch (product) {
+		case 0x0104:
+		case 0x0106:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x2008:
+			sd->bridge = BRIDGE_SPCA504B;
+			break;
+		case 0x2010:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x2016:
+		case 0x2018:
+			sd->bridge = BRIDGE_SPCA504B;
+			break;
+		case 0x2020:
+		case 0x2022:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x2024:
+			sd->bridge = BRIDGE_SPCA536;
+			break;
+		case 0x2028:
+			sd->bridge = BRIDGE_SPCA533;
+			break;
+		case 0x2040:
+		case 0x2042:
+		case 0x2060:
+			sd->bridge = BRIDGE_SPCA536;
+			break;
+		}
+		break;
+	case 0x0d64:		/* SunPlus */
+/*		switch (product) { */
+/*		case 0x0303: */
+			sd->bridge = BRIDGE_SPCA536;
+/*			break; */
+/*		} */
+		break;
+	}
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA536: */
+		cam->cam_mode = vga_mode;
+		cam->nmodes = sizeof vga_mode / sizeof vga_mode[0];
+		break;
+	case BRIDGE_SPCA533:
+		cam->cam_mode = custom_mode;
+		cam->nmodes = sizeof custom_mode / sizeof custom_mode[0];
+		break;
+	case BRIDGE_SPCA504C:
+		cam->cam_mode = vga_mode2;
+		cam->nmodes = sizeof vga_mode2 / sizeof vga_mode2[0];
+		break;
+	}
+	sd->qindex = 5;			/* set the quantization table */
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int rc;
+	__u8 Data;
+	__u8 i;
+	__u8 info[6];
+	int err_code;
+
+	switch (sd->bridge) {
+	case BRIDGE_SPCA504B:
+		spca5xxRegWrite(dev, 0x1d, 0, 0, NULL, 0);
+		spca5xxRegWrite(dev, 0, 1, 0x2306, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0, 0x0d04, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0, 0x2000, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0x13, 0x2301, NULL, 0);
+		spca5xxRegWrite(dev, 0, 0, 0x2306, NULL, 0);
+		/* fall thru */
+	case BRIDGE_SPCA533:
+		rc = spca504B_PollingDataReady(dev);
+		spca50x_GetFirmware(gspca_dev);
+		break;
+	case BRIDGE_SPCA536:
+		spca50x_GetFirmware(gspca_dev);
+		spca5xxRegRead(dev, 0x00, 0x5002, &Data, 1);
+		Data = 0;
+		spca5xxRegWrite(dev, 0x24, 0, 0, &Data, 1);
+		spca5xxRegRead(dev, 0x24, 0, &Data, 1);
+		rc = spca504B_PollingDataReady(dev);
+		spca5xxRegWrite(dev, 0x34, 0, 0, NULL, 0);
+		spca504B_WaitCmdStatus(gspca_dev);
+		break;
+	case BRIDGE_SPCA504C:	/* pccam600 */
+		PDEBUG(D_STREAM, "Opening SPCA504 (PC-CAM 600)");
+		reg_write(dev, 0xe0, 0x0000, 0x0000);
+		reg_write(dev, 0xe0, 0x0000, 0x0001);	/* reset */
+		spca504_wait_status(gspca_dev);
+		if (sd->subtype == LogitechClickSmart420)
+			write_vector(gspca_dev,
+					spca504A_clicksmart420_open_data);
+		else
+			write_vector(gspca_dev, spca504_pccam600_open_data);
+		err_code = spca50x_setup_qtable(gspca_dev,
+						0x00, 0x2800,
+						0x2840, qtable_creative_pccam);
+		if (err_code < 0) {
+			PDEBUG(D_ERR|D_STREAM, "spca50x_setup_qtable failed");
+			return err_code;
+		}
+		break;
+	default:
+/*	case BRIDGE_SPCA504: */
+		PDEBUG(D_STREAM, "Opening SPCA504");
+		if (sd->subtype == AiptekMiniPenCam13) {
+			/*****************************/
+			for (i = 0; i < 6; i++)
+				info[i] = reg_read_info(dev, i);
+			PDEBUG(D_STREAM,
+				"Read info: %d %d %d %d %d %d."
+				" Should be 1,0,2,2,0,0",
+				info[0], info[1], info[2],
+				info[3], info[4], info[5]);
+			/* spca504a aiptek */
+			/* Set AE AWB Banding Type 3-> 50Hz 2-> 60Hz */
+			spca504A_acknowledged_command(gspca_dev, 0x24,
+							8, 3, 0x9e, 1);
+			/* Twice sequencial need status 0xff->0x9e->0x9d */
+			spca504A_acknowledged_command(gspca_dev, 0x24,
+							8, 3, 0x9e, 0);
+
+			spca504A_acknowledged_command(gspca_dev, 0x24,
+							0, 0, 0x9d, 1);
+			/******************************/
+			/* spca504a aiptek */
+			spca504A_acknowledged_command(gspca_dev, 0x08,
+							6, 0, 0x86, 1);
+/*			reg_write (dev, 0, 0x2000, 0); */
+/*			reg_write (dev, 0, 0x2883, 1); */
+/*			spca504A_acknowledged_command (gspca_dev, 0x08,
+							6, 0, 0x86, 1); */
+/*			spca504A_acknowledged_command (gspca_dev, 0x24,
+							0, 0, 0x9D, 1); */
+			reg_write(dev, 0x0, 0x270c, 0x5); /* L92 sno1t.txt */
+			reg_write(dev, 0x0, 0x2310, 0x5);
+			spca504A_acknowledged_command(gspca_dev, 0x01,
+							0x0f, 0, 0xff, 0);
+		}
+		/* setup qtable */
+		reg_write(dev, 0, 0x2000, 0);
+		reg_write(dev, 0, 0x2883, 1);
+		err_code = spca50x_setup_qtable(gspca_dev,
+						0x00, 0x2800,
+						0x2840,
+						qtable_spca504_default);
+		if (err_code < 0) {
+			PDEBUG(D_ERR, "spca50x_setup_qtable failed");
+			return err_code;
+		}
+		break;
+	}
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	int rc;
+	int enable;
+	__u8 i;
+	__u8 info[6];
+
+	if (sd->bridge == BRIDGE_SPCA504B)
+		spca504B_setQtable(gspca_dev);
+	spca504B_SetSizeType(gspca_dev);
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA536: */
+		if (sd->subtype == MegapixV4 ||
+		    sd->subtype == LogitechClickSmart820) {
+			spca5xxRegWrite(dev, 0xf0, 0, 0, NULL, 0);
+			spca504B_WaitCmdStatus(gspca_dev);
+			spca5xxRegRead(dev, 0xf0, 4, NULL, 0);
+			spca504B_WaitCmdStatus(gspca_dev);
+		} else {
+			spca5xxRegWrite(dev, 0x31, 0, 4, NULL, 0);
+			spca504B_WaitCmdStatus(gspca_dev);
+			rc = spca504B_PollingDataReady(dev);
+		}
+		break;
+	case BRIDGE_SPCA504:
+		if (sd->subtype == AiptekMiniPenCam13) {
+			for (i = 0; i < 6; i++)
+				info[i] = reg_read_info(dev, i);
+			PDEBUG(D_STREAM,
+				"Read info: %d %d %d %d %d %d."
+				" Should be 1,0,2,2,0,0",
+				info[0], info[1], info[2],
+				info[3], info[4], info[5]);
+			/* spca504a aiptek */
+			/* Set AE AWB Banding Type 3-> 50Hz 2-> 60Hz */
+			spca504A_acknowledged_command(gspca_dev, 0x24,
+							8, 3, 0x9e, 1);
+			/* Twice sequencial need status 0xff->0x9e->0x9d */
+			spca504A_acknowledged_command(gspca_dev, 0x24,
+							8, 3, 0x9e, 0);
+			spca504A_acknowledged_command(gspca_dev, 0x24,
+							0, 0, 0x9d, 1);
+		} else {
+			spca504_acknowledged_command(gspca_dev, 0x24, 8, 3);
+			for (i = 0; i < 6; i++)
+				info[i] = reg_read_info(dev, i);
+			PDEBUG(D_STREAM,
+				"Read info: %d %d %d %d %d %d."
+				" Should be 1,0,2,2,0,0",
+				info[0], info[1], info[2],
+				info[3], info[4], info[5]);
+			spca504_acknowledged_command(gspca_dev, 0x24, 8, 3);
+			spca504_acknowledged_command(gspca_dev, 0x24, 0, 0);
+		}
+		spca504B_SetSizeType(gspca_dev);
+		reg_write(dev, 0x0, 0x270c, 0x5);	/* L92 sno1t.txt */
+		reg_write(dev, 0x0, 0x2310, 0x5);
+		break;
+	case BRIDGE_SPCA504C:
+		if (sd->subtype == LogitechClickSmart420) {
+			write_vector(gspca_dev,
+					spca504A_clicksmart420_init_data);
+		} else {
+			write_vector(gspca_dev, spca504_pccam600_init_data);
+		}
+		enable = (sd->autogain ? 0x4 : 0x1);
+		reg_write(dev, 0x0c, 0x0000, enable);	/* auto exposure */
+		reg_write(dev, 0xb0, 0x0000, enable);	/* auto whiteness */
+
+		/* set default exposure compensation and whiteness balance */
+		reg_write(dev, 0x30, 0x0001, 800);	/* ~ 20 fps */
+		reg_write(dev, 0x30, 0x0002, 1600);
+		spca504B_SetSizeType(gspca_dev);
+		break;
+	}
+	sp5xx_initContBrigHueRegisters(gspca_dev);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA536: */
+/*	case BRIDGE_SPCA504B: */
+		spca5xxRegWrite(dev, 0x31, 0, 0, NULL, 0);
+		spca504B_WaitCmdStatus(gspca_dev);
+		spca504B_PollingDataReady(dev);
+		break;
+	case BRIDGE_SPCA504:
+	case BRIDGE_SPCA504C:
+		reg_write(dev, 0x00, 0x2000, 0x0000);
+
+		if (sd->subtype == AiptekMiniPenCam13) {
+			/* spca504a aiptek */
+/*			spca504A_acknowledged_command(gspca_dev, 0x08,
+							 6, 0, 0x86, 1); */
+			spca504A_acknowledged_command(gspca_dev, 0x24,
+							0x00, 0x00, 0x9d, 1);
+			spca504A_acknowledged_command(gspca_dev, 0x01,
+							0x0f, 0x00, 0xff, 1);
+		} else {
+			spca504_acknowledged_command(gspca_dev, 0x24, 0, 0);
+			reg_write(dev, 0x01, 0x000f, 0x0);
+		}
+		break;
+	}
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int i, sof = 0;
+	unsigned char *s, *d;
+	static unsigned char ffd9[] = {0xff, 0xd9};
+
+/* frames are jpeg 4.1.1 without 0xff escape */
+	switch (sd->bridge) {
+	case BRIDGE_SPCA533:
+		if (data[0] == 0xff) {
+			if (data[1] != 0x01) {	/* drop packet */
+/*				gspca_dev->last_packet_type = DISCARD_PACKET; */
+				return;
+			}
+			sof = 1;
+			data += SPCA533_OFFSET_DATA;
+			len -= SPCA533_OFFSET_DATA;
+		} else {
+			data += 1;
+			len -= 1;
+		}
+		break;
+	case BRIDGE_SPCA536:
+		if (data[0] == 0xff) {
+			sof = 1;
+			data += SPCA536_OFFSET_DATA;
+			len -= SPCA536_OFFSET_DATA;
+		} else {
+			data += 2;
+			len -= 2;
+		}
+		break;
+	default:
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA504B: */
+		switch (data[0]) {
+		case 0xfe:			/* start of frame */
+			sof = 1;
+			data += SPCA50X_OFFSET_DATA;
+			len -= SPCA50X_OFFSET_DATA;
+			break;
+		case 0xff:			/* drop packet */
+/*			gspca_dev->last_packet_type = DISCARD_PACKET; */
+			return;
+		default:
+			data += 1;
+			len -= 1;
+			break;
+		}
+		break;
+	case BRIDGE_SPCA504C:
+		switch (data[0]) {
+		case 0xfe:			/* start of frame */
+			sof = 1;
+			data += SPCA504_PCCAM600_OFFSET_DATA;
+			len -= SPCA504_PCCAM600_OFFSET_DATA;
+			break;
+		case 0xff:			/* drop packet */
+/*			gspca_dev->last_packet_type = DISCARD_PACKET; */
+			return;
+		default:
+			data += 1;
+			len -= 1;
+			break;
+		}
+		break;
+	}
+	if (sof) {		/* start of frame */
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					ffd9, 2);
+
+		/* put the JPEG header in the new frame */
+		jpeg_put_header(gspca_dev, frame,
+				((struct sd *) gspca_dev)->qindex,
+				0x22);
+	}
+
+	/* add 0x00 after 0xff */
+	for (i = len; --i >= 0; )
+		if (data[i] == 0xff)
+			break;
+	if (i < 0) {			/* no 0xff */
+		gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+		return;
+	}
+	s = data;
+	d = sd->packet;
+	for (i = 0; i < len; i++) {
+		*d++ = *s++;
+		if (s[-1] == 0xff)
+			*d++ = 0x00;
+	}
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame,
+			sd->packet, d - sd->packet);
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA504C: */
+		reg_write(dev, 0x0, 0x21a7, sd->brightness);
+		break;
+	case BRIDGE_SPCA536:
+		reg_write(dev, 0x0, 0x20f0, sd->brightness);
+		break;
+	}
+}
+
+static void getbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u16 brightness = 0;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA504C: */
+		brightness = reg_read(dev, 0x0, 0x21a7, 2);
+		break;
+	case BRIDGE_SPCA536:
+		brightness = reg_read(dev, 0x0, 0x20f0, 2);
+		break;
+	}
+	sd->brightness = ((brightness & 0xff) - 128) % 255;
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA504C: */
+		reg_write(dev, 0x0, 0x21a8, sd->contrast);
+		break;
+	case BRIDGE_SPCA536:
+		reg_write(dev, 0x0, 0x20f1, sd->contrast);
+		break;
+	}
+}
+
+static void getcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA504C: */
+		sd->contrast = reg_read(dev, 0x0, 0x21a8, 2);
+		break;
+	case BRIDGE_SPCA536:
+		sd->contrast = reg_read(dev, 0x0, 0x20f1, 2);
+		break;
+	}
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA504C: */
+		reg_write(dev, 0x0, 0x21ae, sd->colors);
+		break;
+	case BRIDGE_SPCA536:
+		reg_write(dev, 0x0, 0x20f6, sd->colors);
+		break;
+	}
+}
+
+static void getcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	switch (sd->bridge) {
+	default:
+/*	case BRIDGE_SPCA533: */
+/*	case BRIDGE_SPCA504B: */
+/*	case BRIDGE_SPCA504: */
+/*	case BRIDGE_SPCA504C: */
+		sd->colors = reg_read(dev, 0x0, 0x21ae, 2) >> 1;
+		break;
+	case BRIDGE_SPCA536:
+		sd->colors = reg_read(dev, 0x0, 0x20f6, 2) >> 1;
+		break;
+	}
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getbrightness(gspca_dev);
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcontrast(gspca_dev);
+	*val = sd->contrast;
+	return 0;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	getcolors(gspca_dev);
+	*val = sd->colors;
+	return 0;
+}
+
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->autogain = val;
+	return 0;
+}
+
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autogain;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x041e, 0x400b), DVNM("Creative PC-CAM 600")},
+	{USB_DEVICE(0x041e, 0x4012), DVNM("PC-Cam350")},
+	{USB_DEVICE(0x041e, 0x4013), DVNM("Creative Pccam750")},
+	{USB_DEVICE(0x0458, 0x7006), DVNM("Genius Dsc 1.3 Smart")},
+	{USB_DEVICE(0x046d, 0x0905), DVNM("Logitech ClickSmart 820")},
+	{USB_DEVICE(0x046d, 0x0960), DVNM("Logitech ClickSmart 420")},
+	{USB_DEVICE(0x0471, 0x0322), DVNM("Philips DMVC1300K")},
+	{USB_DEVICE(0x04a5, 0x3003), DVNM("Benq DC 1300")},
+	{USB_DEVICE(0x04a5, 0x3008), DVNM("Benq DC 1500")},
+	{USB_DEVICE(0x04a5, 0x300a), DVNM("Benq DC3410")},
+	{USB_DEVICE(0x04f1, 0x1001), DVNM("JVC GC A50")},
+	{USB_DEVICE(0x04fc, 0x500c), DVNM("Sunplus CA500C")},
+	{USB_DEVICE(0x04fc, 0x504a), DVNM("Aiptek Mini PenCam 1.3")},
+	{USB_DEVICE(0x04fc, 0x504b), DVNM("Maxell MaxPocket LE 1.3")},
+	{USB_DEVICE(0x04fc, 0x5330), DVNM("Digitrex 2110")},
+	{USB_DEVICE(0x04fc, 0x5360), DVNM("Sunplus Generic")},
+	{USB_DEVICE(0x04fc, 0xffff), DVNM("Pure DigitalDakota")},
+	{USB_DEVICE(0x052b, 0x1513), DVNM("Megapix V4")},
+	{USB_DEVICE(0x0546, 0x3155), DVNM("Polaroid PDC3070")},
+	{USB_DEVICE(0x0546, 0x3191), DVNM("Polaroid Ion 80")},
+	{USB_DEVICE(0x0546, 0x3273), DVNM("Polaroid PDC2030")},
+	{USB_DEVICE(0x055f, 0xc211), DVNM("Kowa Bs888e Microcamera")},
+	{USB_DEVICE(0x055f, 0xc230), DVNM("Mustek Digicam 330K")},
+	{USB_DEVICE(0x055f, 0xc232), DVNM("Mustek MDC3500")},
+	{USB_DEVICE(0x055f, 0xc360), DVNM("Mustek DV4000 Mpeg4 ")},
+	{USB_DEVICE(0x055f, 0xc420), DVNM("Mustek gSmart Mini 2")},
+	{USB_DEVICE(0x055f, 0xc430), DVNM("Mustek Gsmart LCD 2")},
+	{USB_DEVICE(0x055f, 0xc440), DVNM("Mustek DV 3000")},
+	{USB_DEVICE(0x055f, 0xc520), DVNM("Mustek gSmart Mini 3")},
+	{USB_DEVICE(0x055f, 0xc530), DVNM("Mustek Gsmart LCD 3")},
+	{USB_DEVICE(0x055f, 0xc540), DVNM("Gsmart D30")},
+	{USB_DEVICE(0x055f, 0xc630), DVNM("Mustek MDC4000")},
+	{USB_DEVICE(0x055f, 0xc650), DVNM("Mustek MDC5500Z")},
+	{USB_DEVICE(0x05da, 0x1018), DVNM("Digital Dream Enigma 1.3")},
+	{USB_DEVICE(0x06d6, 0x0031), DVNM("Trust 610 LCD PowerC@m Zoom")},
+	{USB_DEVICE(0x0733, 0x1311), DVNM("Digital Dream Epsilon 1.3")},
+	{USB_DEVICE(0x0733, 0x1314), DVNM("Mercury 2.1MEG Deluxe Classic Cam")},
+	{USB_DEVICE(0x0733, 0x2211), DVNM("Jenoptik jdc 21 LCD")},
+	{USB_DEVICE(0x0733, 0x2221), DVNM("Mercury Digital Pro 3.1p")},
+	{USB_DEVICE(0x0733, 0x3261), DVNM("Concord 3045 spca536a")},
+	{USB_DEVICE(0x0733, 0x3281), DVNM("Cyberpix S550V")},
+	{USB_DEVICE(0x08ca, 0x0104), DVNM("Aiptek PocketDVII 1.3")},
+	{USB_DEVICE(0x08ca, 0x0106), DVNM("Aiptek Pocket DV3100+")},
+	{USB_DEVICE(0x08ca, 0x2008), DVNM("Aiptek Mini PenCam 2 M")},
+	{USB_DEVICE(0x08ca, 0x2010), DVNM("Aiptek PocketCam 3M")},
+	{USB_DEVICE(0x08ca, 0x2016), DVNM("Aiptek PocketCam 2 Mega")},
+	{USB_DEVICE(0x08ca, 0x2018), DVNM("Aiptek Pencam SD 2M")},
+	{USB_DEVICE(0x08ca, 0x2020), DVNM("Aiptek Slim 3000F")},
+	{USB_DEVICE(0x08ca, 0x2022), DVNM("Aiptek Slim 3200")},
+	{USB_DEVICE(0x08ca, 0x2024), DVNM("Aiptek DV3500 Mpeg4 ")},
+	{USB_DEVICE(0x08ca, 0x2028), DVNM("Aiptek PocketCam4M")},
+	{USB_DEVICE(0x08ca, 0x2040), DVNM("Aiptek PocketDV4100M")},
+	{USB_DEVICE(0x08ca, 0x2042), DVNM("Aiptek PocketDV5100")},
+	{USB_DEVICE(0x08ca, 0x2060), DVNM("Aiptek PocketDV5300")},
+	{USB_DEVICE(0x0d64, 0x0303), DVNM("Sunplus FashionCam DXG")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/t613.c b/drivers/media/video/gspca/t613.c
new file mode 100644
index 000000000000..c22b301ebae0
--- /dev/null
+++ b/drivers/media/video/gspca/t613.c
@@ -0,0 +1,1013 @@
+/*
+ *Notes: * t613  + tas5130A
+ *	* Focus to light do not balance well as in win.
+ *	  Quality in win is not good, but its kinda better.
+ *	 * Fix some "extraneous bytes", most of apps will show the image anyway
+ *	 * Gamma table, is there, but its really doing something?
+ *	 * 7~8 Fps, its ok, max on win its 10.
+ *			Costantino Leandro
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "t613"
+#include "gspca.h"
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+struct control_menu_info {
+	int value;
+	char name[32];
+};
+
+#define MAX_GAMMA 0x10		/* 0 to 15 */
+
+/* From LUVCVIEW */
+#define V4L2_CID_EFFECTS (V4L2_CID_PRIVATE_BASE + 3)
+
+MODULE_AUTHOR("Leandro Costantino <le_costantino@pixartargentina.com.ar>");
+MODULE_DESCRIPTION("GSPCA/T613 (JPEG Compliance) USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned char brightness;
+	unsigned char contrast;
+	unsigned char colors;
+	unsigned char autogain;
+	unsigned char gamma;
+	unsigned char sharpness;
+	unsigned char freq;
+	unsigned char whitebalance;
+	unsigned char mirror;
+	unsigned char effect;
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setlowlight(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getlowlight(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setgamma(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getgamma(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setsharpness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getsharpness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setwhitebalance(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getwhitebalance(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setflip(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getflip(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_seteffect(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_geteffect(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_querymenu(struct gspca_dev *gspca_dev,
+			struct v4l2_querymenu *menu);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	 {
+	  .id = V4L2_CID_BRIGHTNESS,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Brightness",
+	  .minimum = 0,
+	  .maximum = 0x0f,
+	  .step = 1,
+	  .default_value = 0x09,
+	  },
+	 .set = sd_setbrightness,
+	 .get = sd_getbrightness,
+	 },
+#define SD_CONTRAST 1
+	{
+	 {
+	  .id = V4L2_CID_CONTRAST,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Contrast",
+	  .minimum = 0,
+	  .maximum = 0x0d,
+	  .step = 1,
+	  .default_value = 0x07,
+	  },
+	 .set = sd_setcontrast,
+	 .get = sd_getcontrast,
+	 },
+#define SD_COLOR 2
+	{
+	 {
+	  .id = V4L2_CID_SATURATION,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Color",
+	  .minimum = 0,
+	  .maximum = 0x0f,
+	  .step = 1,
+	  .default_value = 0x05,
+	  },
+	 .set = sd_setcolors,
+	 .get = sd_getcolors,
+	 },
+#define SD_GAMMA 3
+	{
+	 {
+	  .id = V4L2_CID_GAMMA,	/* (gamma on win) */
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Gamma (Untested)",
+	  .minimum = 0,
+	  .maximum = MAX_GAMMA,
+	  .step = 1,
+	  .default_value = 0x09,
+	  },
+	 .set = sd_setgamma,
+	 .get = sd_getgamma,
+	 },
+#define SD_AUTOGAIN 4
+	{
+	 {
+	  .id = V4L2_CID_GAIN,	/* here, i activate only the lowlight,
+				 * some apps dont bring up the
+				 * backligth_compensation control) */
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Low Light",
+	  .minimum = 0,
+	  .maximum = 1,
+	  .step = 1,
+	  .default_value = 0x01,
+	  },
+	 .set = sd_setlowlight,
+	 .get = sd_getlowlight,
+	 },
+#define SD_MIRROR 5
+	{
+	 {
+	  .id = V4L2_CID_HFLIP,
+	  .type = V4L2_CTRL_TYPE_BOOLEAN,
+	  .name = "Mirror Image",
+	  .minimum = 0,
+	  .maximum = 1,
+	  .step = 1,
+	  .default_value = 0,
+	  },
+	 .set = sd_setflip,
+	 .get = sd_getflip
+	},
+#define SD_LIGHTFREQ 6
+	{
+	 {
+	  .id = V4L2_CID_POWER_LINE_FREQUENCY,
+	  .type = V4L2_CTRL_TYPE_MENU,
+	  .name = "Light Frequency Filter",
+	  .minimum = 1,		/* 1 -> 0x50, 2->0x60 */
+	  .maximum = 2,
+	  .step = 1,
+	  .default_value = 1,
+	  },
+	 .set = sd_setfreq,
+	 .get = sd_getfreq},
+
+#define SD_WHITE_BALANCE 7
+	{
+	 {
+	  .id = V4L2_CID_WHITE_BALANCE_TEMPERATURE,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "White Balance",
+	  .minimum = 0,
+	  .maximum = 1,
+	  .step = 1,
+	  .default_value = 1,
+	  },
+	 .set = sd_setwhitebalance,
+	 .get = sd_getwhitebalance
+	},
+#define SD_SHARPNESS 8		/* (aka definition on win) */
+	{
+	 {
+	  .id = V4L2_CID_SHARPNESS,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Sharpness",
+	  .minimum = 0,
+	  .maximum = MAX_GAMMA,	/* 0 to 16 */
+	  .step = 1,
+	  .default_value = 0x06,
+	  },
+	 .set = sd_setsharpness,
+	 .get = sd_getsharpness,
+	 },
+#define SD_EFFECTS 9
+	{
+	 {
+	  .id = V4L2_CID_EFFECTS,
+	  .type = V4L2_CTRL_TYPE_MENU,
+	  .name = "Webcam Effects",
+	  .minimum = 0,
+	  .maximum = 4,
+	  .step = 1,
+	  .default_value = 0,
+	  },
+	 .set = sd_seteffect,
+	 .get = sd_geteffect
+	},
+};
+
+static struct control_menu_info effects_control[] = {
+	{0, "Normal"},
+	{1, "Emboss"},		/* disabled */
+	{2, "Monochrome"},
+	{3, "Sepia"},
+	{4, "Sketch"},
+	{5, "Sun Effect"},	/* disabled */
+	{6, "Negative"},
+};
+
+#define NUM_EFFECTS_CONTROL \
+			(sizeof(effects_control)/sizeof(effects_control[0]))
+
+static struct cam_mode vga_mode_t16[] = {
+	{V4L2_PIX_FMT_JPEG, 160, 120, 4},
+	{V4L2_PIX_FMT_JPEG, 176, 144, 3},
+	{V4L2_PIX_FMT_JPEG, 320, 240, 2},
+	{V4L2_PIX_FMT_JPEG, 352, 288, 1},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 0},
+};
+
+#define T16_OFFSET_DATA 631
+#define MAX_EFFECTS 7
+/* easily done by soft, this table could be removed,
+ * i keep it here just in case */
+unsigned char effects_table[MAX_EFFECTS][6] = {
+	{0xa8, 0xe8, 0xc6, 0xd2, 0xc0, 0x00},	/* Normal */
+	{0xa8, 0xc8, 0xc6, 0x52, 0xc0, 0x04},	/* Repujar */
+	{0xa8, 0xe8, 0xc6, 0xd2, 0xc0, 0x20},	/* Monochrome */
+	{0xa8, 0xe8, 0xc6, 0xd2, 0xc0, 0x80},	/* Sepia */
+	{0xa8, 0xc8, 0xc6, 0x52, 0xc0, 0x02},	/* Croquis */
+	{0xa8, 0xc8, 0xc6, 0xd2, 0xc0, 0x10},	/* Sun Effect */
+	{0xa8, 0xc8, 0xc6, 0xd2, 0xc0, 0x40},	/* Negative */
+};
+
+unsigned char gamma_table[MAX_GAMMA][34] = {
+	{0x90, 0x00, 0x91, 0x3e, 0x92, 0x69, 0x93, 0x85,
+	 0x94, 0x95, 0x95, 0xa1, 0x96, 0xae, 0x97, 0xb9,
+	 0x98, 0xc2, 0x99, 0xcb, 0x9a, 0xd4, 0x9b, 0xdb,
+	 0x9c, 0xe3, 0x9d, 0xea, 0x9e, 0xf1, 0x9f, 0xf8,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x33, 0x92, 0x5A, 0x93, 0x75,
+	 0x94, 0x85, 0x95, 0x93, 0x96, 0xA1, 0x97, 0xAD,
+	 0x98, 0xB7, 0x99, 0xC2, 0x9A, 0xCB, 0x9B, 0xD4,
+	 0x9C, 0xDE, 0x9D, 0xE7, 0x9E, 0xF0, 0x9F, 0xF7,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x2F, 0x92, 0x51, 0x93, 0x6B,
+	 0x94, 0x7C, 0x95, 0x8A, 0x96, 0x99, 0x97, 0xA6,
+	 0x98, 0xB1, 0x99, 0xBC, 0x9A, 0xC6, 0x9B, 0xD0,
+	 0x9C, 0xDB, 0x9D, 0xE4, 0x9E, 0xED, 0x9F, 0xF6,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x29, 0x92, 0x48, 0x93, 0x60,
+	 0x94, 0x72, 0x95, 0x81, 0x96, 0x90, 0x97, 0x9E,
+	 0x98, 0xAA, 0x99, 0xB5, 0x9A, 0xBF, 0x9B, 0xCB,
+	 0x9C, 0xD6, 0x9D, 0xE1, 0x9E, 0xEB, 0x9F, 0xF5,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x23, 0x92, 0x3F, 0x93, 0x55,
+	 0x94, 0x68, 0x95, 0x77, 0x96, 0x86, 0x97, 0x95,
+	 0x98, 0xA2, 0x99, 0xAD, 0x9A, 0xB9, 0x9B, 0xC6,
+	 0x9C, 0xD2, 0x9D, 0xDE, 0x9E, 0xE9, 0x9F, 0xF4,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x1B, 0x92, 0x33, 0x93, 0x48,
+	 0x94, 0x59, 0x95, 0x69, 0x96, 0x79, 0x97, 0x87,
+	 0x98, 0x96, 0x99, 0xA3, 0x9A, 0xB1, 0x9B, 0xBE,
+	 0x9C, 0xCC, 0x9D, 0xDA, 0x9E, 0xE7, 0x9F, 0xF3,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x02, 0x92, 0x10, 0x93, 0x20,
+	 0x94, 0x32, 0x95, 0x40, 0x96, 0x57, 0x97, 0x67,
+	 0x98, 0x77, 0x99, 0x88, 0x9a, 0x99, 0x9b, 0xaa,
+	 0x9c, 0xbb, 0x9d, 0xcc, 0x9e, 0xdd, 0x9f, 0xee,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x02, 0x92, 0x14, 0x93, 0x26,
+	 0x94, 0x38, 0x95, 0x4A, 0x96, 0x60, 0x97, 0x70,
+	 0x98, 0x80, 0x99, 0x90, 0x9A, 0xA0, 0x9B, 0xB0,
+	 0x9C, 0xC0, 0x9D, 0xD0, 0x9E, 0xE0, 0x9F, 0xF0,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x10, 0x92, 0x22, 0x93, 0x35,
+	 0x94, 0x47, 0x95, 0x5A, 0x96, 0x69, 0x97, 0x79,
+	 0x98, 0x88, 0x99, 0x97, 0x9A, 0xA7, 0x9B, 0xB6,
+	 0x9C, 0xC4, 0x9D, 0xD3, 0x9E, 0xE0, 0x9F, 0xF0,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x10, 0x92, 0x26, 0x93, 0x40,
+	 0x94, 0x54, 0x95, 0x65, 0x96, 0x75, 0x97, 0x84,
+	 0x98, 0x93, 0x99, 0xa1, 0x9a, 0xb0, 0x9b, 0xbd,
+	 0x9c, 0xca, 0x9d, 0xd6, 0x9e, 0xe0, 0x9f, 0xf0,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x18, 0x92, 0x2B, 0x93, 0x44,
+	 0x94, 0x60, 0x95, 0x70, 0x96, 0x80, 0x97, 0x8E,
+	 0x98, 0x9C, 0x99, 0xAA, 0x9A, 0xB7, 0x9B, 0xC4,
+	 0x9C, 0xD0, 0x9D, 0xD8, 0x9E, 0xE2, 0x9F, 0xF0,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x1A, 0x92, 0x34, 0x93, 0x52,
+	 0x94, 0x66, 0x95, 0x7E, 0x96, 0x8D, 0x97, 0x9B,
+	 0x98, 0xA8, 0x99, 0xB4, 0x9A, 0xC0, 0x9B, 0xCB,
+	 0x9C, 0xD6, 0x9D, 0xE1, 0x9E, 0xEB, 0x9F, 0xF5,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x3F, 0x92, 0x5A, 0x93, 0x6E,
+	 0x94, 0x7F, 0x95, 0x8E, 0x96, 0x9C, 0x97, 0xA8,
+	 0x98, 0xB4, 0x99, 0xBF, 0x9A, 0xC9, 0x9B, 0xD3,
+	 0x9C, 0xDC, 0x9D, 0xE5, 0x9E, 0xEE, 0x9F, 0xF6,
+	 0xA0, 0xFF},
+	{0x90, 0x00, 0x91, 0x54, 0x92, 0x6F, 0x93, 0x83,
+	 0x94, 0x93, 0x95, 0xA0, 0x96, 0xAD, 0x97, 0xB7,
+	 0x98, 0xC2, 0x99, 0xCB, 0x9A, 0xD4, 0x9B, 0xDC,
+	 0x9C, 0xE4, 0x9D, 0xEB, 0x9E, 0xF2, 0x9F, 0xF9,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x6E, 0x92, 0x88, 0x93, 0x9A,
+	 0x94, 0xA8, 0x95, 0xB3, 0x96, 0xBD, 0x97, 0xC6,
+	 0x98, 0xCF, 0x99, 0xD6, 0x9A, 0xDD, 0x9B, 0xE3,
+	 0x9C, 0xE9, 0x9D, 0xEF, 0x9E, 0xF4, 0x9F, 0xFA,
+	 0xa0, 0xff},
+	{0x90, 0x00, 0x91, 0x93, 0x92, 0xA8, 0x93, 0xB7,
+	 0x94, 0xC1, 0x95, 0xCA, 0x96, 0xD2, 0x97, 0xD8,
+	 0x98, 0xDE, 0x99, 0xE3, 0x9A, 0xE8, 0x9B, 0xED,
+	 0x9C, 0xF1, 0x9D, 0xF5, 0x9E, 0xF8, 0x9F, 0xFC,
+	 0xA0, 0xFF}
+};
+
+static __u8 tas5130a_sensor_init[][8] = {
+	{0x62, 0x08, 0x63, 0x70, 0x64, 0x1d, 0x60, 0x09},
+	{0x62, 0x20, 0x63, 0x01, 0x64, 0x02, 0x60, 0x09},
+	{0x62, 0x07, 0x63, 0x03, 0x64, 0x00, 0x60, 0x09},
+	{0x62, 0x07, 0x63, 0x03, 0x64, 0x00, 0x60, 0x09},
+	{},
+};
+
+static void t16RegRead(struct usb_device *dev,
+		       __u16 index, __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			0,		/* request */
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,		/* value */
+			index, buffer, length, 500);
+}
+
+static void t16RegWrite(struct usb_device *dev,
+			__u16 value,
+			__u16 index, __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			0,		/* request */
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, buffer, length, 500);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+		     const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x01;
+
+	cam->cam_mode = vga_mode_t16;
+	cam->nmodes = ARRAY_SIZE(vga_mode_t16);
+
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	sd->colors = sd_ctrls[SD_COLOR].qctrl.default_value;
+	sd->gamma = sd_ctrls[SD_GAMMA].qctrl.default_value;
+	sd->mirror = sd_ctrls[SD_MIRROR].qctrl.default_value;
+	sd->freq = sd_ctrls[SD_LIGHTFREQ].qctrl.default_value;
+	sd->whitebalance = sd_ctrls[SD_WHITE_BALANCE].qctrl.default_value;
+	sd->sharpness = sd_ctrls[SD_SHARPNESS].qctrl.default_value;
+	sd->effect = sd_ctrls[SD_EFFECTS].qctrl.default_value;
+	return 0;
+}
+
+static int init_default_parameters(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	/* some of this registers are not really neded, because
+	 * they are overriden by setbrigthness, setcontrast, etc,
+	 * but wont hurt anyway, and can help someone with similar webcam
+	 * to see the initial parameters.*/
+	int i = 0;
+	__u8 test_byte;
+
+	static unsigned char read_indexs[] =
+		{ 0x06, 0x07, 0x0a, 0x0b, 0x66, 0x80, 0x81, 0x8e, 0x8f, 0xa5,
+		  0xa6, 0xa8, 0xbb, 0xbc, 0xc6, 0x00, 0x00 };
+	static unsigned char n1[6] =
+			{0x08, 0x03, 0x09, 0x03, 0x12, 0x04};
+	static unsigned char n2[2] =
+			{0x08, 0x00};
+	static unsigned char nset[6] =
+		{ 0x61, 0x68, 0x62, 0xff, 0x60, 0x07 };
+	static unsigned char n3[6] =
+			{0x61, 0x68, 0x65, 0x0a, 0x60, 0x04};
+	static unsigned char n4[0x46] =
+		{0x09, 0x01, 0x12, 0x04, 0x66, 0x8a, 0x80, 0x3c,
+		 0x81, 0x22, 0x84, 0x50, 0x8a, 0x78, 0x8b, 0x68,
+		 0x8c, 0x88, 0x8e, 0x33, 0x8f, 0x24, 0xaa, 0xb1,
+		 0xa2, 0x60, 0xa5, 0x30, 0xa6, 0x3a, 0xa8, 0xe8,
+		 0xae, 0x05, 0xb1, 0x00, 0xbb, 0x04, 0xbc, 0x48,
+		 0xbe, 0x36, 0xc6, 0x88, 0xe9, 0x00, 0xc5, 0xc0,
+		 0x65, 0x0a, 0xbb, 0x86, 0xaf, 0x58, 0xb0, 0x68,
+		 0x87, 0x40, 0x89, 0x2b, 0x8d, 0xff, 0x83, 0x40,
+		 0xac, 0x84, 0xad, 0x86, 0xaf, 0x46};
+	static unsigned char nset4[18] = {
+		0xe0, 0x60, 0xe1, 0xa8, 0xe2, 0xe0, 0xe3, 0x60, 0xe4, 0xa8,
+		0xe5, 0xe0, 0xe6, 0x60, 0xe7, 0xa8,
+		0xe8, 0xe0
+	};
+	/* ojo puede ser 0xe6 en vez de 0xe9 */
+	static unsigned char nset2[20] = {
+		0xd0, 0xbb, 0xd1, 0x28, 0xd2, 0x10, 0xd3, 0x10, 0xd4, 0xbb,
+		0xd5, 0x28, 0xd6, 0x1e, 0xd7, 0x27,
+		0xd8, 0xc8, 0xd9, 0xfc
+	};
+	static unsigned char missing[8] =
+		{ 0x87, 0x20, 0x88, 0x20, 0x89, 0x20, 0x80, 0x38 };
+	static unsigned char nset3[18] = {
+		0xc7, 0x60, 0xc8, 0xa8, 0xc9, 0xe0, 0xca, 0x60, 0xcb, 0xa8,
+		0xcc, 0xe0, 0xcd, 0x60, 0xce, 0xa8,
+		0xcf, 0xe0
+	};
+	static unsigned char nset5[4] =
+		{ 0x8f, 0x24, 0xc3, 0x00 };	/* bright */
+	static unsigned char nset6[34] = {
+		0x90, 0x00, 0x91, 0x1c, 0x92, 0x30, 0x93, 0x43, 0x94, 0x54,
+		0x95, 0x65, 0x96, 0x75, 0x97, 0x84,
+		0x98, 0x93, 0x99, 0xa1, 0x9a, 0xb0, 0x9b, 0xbd, 0x9c, 0xca,
+		0x9d, 0xd8, 0x9e, 0xe5, 0x9f, 0xf2,
+		0xa0, 0xff
+	};			/* Gamma */
+	static unsigned char nset7[4] =
+			{ 0x66, 0xca, 0xa8, 0xf8 };	/* 50/60 Hz */
+	static unsigned char nset9[4] =
+			{ 0x0b, 0x04, 0x0a, 0x78 };
+	static unsigned char nset8[6] =
+			{ 0xa8, 0xf0, 0xc6, 0x88, 0xc0, 0x00 };
+	static unsigned char nset10[6] =
+			{ 0x0c, 0x03, 0xab, 0x10, 0x81, 0x20 };
+
+	t16RegWrite(dev, 0x01, 0x0000, n1, 0x06);
+	t16RegWrite(dev, 0x01, 0x0000, nset, 0x06);
+	t16RegRead(dev, 0x0063, &test_byte, 1);
+	t16RegWrite(dev, 0x01, 0x0000, n2, 0x02);
+
+	while (read_indexs[i] != 0x00) {
+		t16RegRead(dev, read_indexs[i], &test_byte, 1);
+		PDEBUG(D_CONF, "Reg 0x%x => 0x%x", read_indexs[i],
+		       test_byte);
+		i++;
+	}
+
+	t16RegWrite(dev, 0x01, 0x0000, n3, 0x06);
+	t16RegWrite(dev, 0x01, 0x0000, n4, 0x46);
+	t16RegRead(dev, 0x0080, &test_byte, 1);
+	t16RegWrite(dev, 0x00, 0x2c80, 0x00, 0x0);
+	t16RegWrite(dev, 0x01, 0x0000, nset2, 0x14);
+	t16RegWrite(dev, 0x01, 0x0000, nset3, 0x12);
+	t16RegWrite(dev, 0x01, 0x0000, nset4, 0x12);
+	t16RegWrite(dev, 0x00, 0x3880, 0x00, 0x0);
+	t16RegWrite(dev, 0x00, 0x3880, 0x00, 0x0);
+	t16RegWrite(dev, 0x00, 0x338e, 0x00, 0x0);
+	t16RegWrite(dev, 0x01, 00, nset5, 0x04);
+	t16RegWrite(dev, 0x00, 0x00a9, 0x00, 0x0);
+	t16RegWrite(dev, 0x01, 00, nset6, 0x22);
+	t16RegWrite(dev, 0x00, 0x86bb, 0x00, 0x0);
+	t16RegWrite(dev, 0x00, 0x4aa6, 0x00, 0x0);
+
+	t16RegWrite(dev, 0x01, 00, missing, 0x08);
+
+	t16RegWrite(dev, 0x00, 0x2087, 0x00, 0x0);
+	t16RegWrite(dev, 0x00, 0x2088, 0x00, 0x0);
+	t16RegWrite(dev, 0x00, 0x2089, 0x00, 0x0);
+
+	t16RegWrite(dev, 0x01, 00, nset7, 0x4);
+	t16RegWrite(dev, 0x01, 00, nset10, 0x06);
+	t16RegWrite(dev, 0x01, 00, nset8, 0x06);
+	t16RegWrite(dev, 0x01, 00, nset9, 0x04);
+
+	t16RegWrite(dev, 0x00, 0x2880, 0x00, 0x0);
+	t16RegWrite(dev, 0x01, 0x0000, nset2, 0x14);
+	t16RegWrite(dev, 0x01, 0x0000, nset3, 0x12);
+	t16RegWrite(dev, 0x01, 0x0000, nset4, 0x12);
+
+	return 0;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	unsigned int brightness;
+	unsigned char set6[4] = { 0x8f, 0x26, 0xc3, 0x80 };
+	brightness = sd->brightness;
+
+	if (brightness < 7) {
+		set6[3] = 0x70 - (brightness * 0xa);
+	} else {
+		set6[1] = 0x24;
+		set6[3] = 0x00 + ((brightness - 7) * 0xa);
+	}
+
+	t16RegWrite(dev, 0x01, 0x0000, set6, 4);
+}
+
+static void setflip(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	unsigned char flipcmd[8] =
+	    { 0x62, 0x07, 0x63, 0x03, 0x64, 0x00, 0x60, 0x09 };
+
+	if (sd->mirror == 1)
+		flipcmd[3] = 0x01;
+
+	t16RegWrite(dev, 0x01, 0x0000, flipcmd, 8);
+}
+
+static void seteffect(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	t16RegWrite(dev, 0x01, 0x0000, effects_table[sd->effect], 0x06);
+	if (sd->effect == 1 || sd->effect == 5) {
+		PDEBUG(D_CONF,
+		       "This effect have been disabled for webcam \"safety\"");
+		return;
+	}
+
+	if (sd->effect == 1 || sd->effect == 4)
+		t16RegWrite(dev, 0x00, 0x4aa6, 0x00, 0x00);
+	else
+		t16RegWrite(dev, 0x00, 0xfaa6, 0x00, 0x00);
+}
+
+static void setwhitebalance(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	unsigned char white_balance[8] =
+	    { 0x87, 0x20, 0x88, 0x20, 0x89, 0x20, 0x80, 0x38 };
+
+	if (sd->whitebalance == 1)
+		white_balance[7] = 0x3c;
+
+	t16RegWrite(dev, 0x01, 0x0000, white_balance, 8);
+}
+
+static void setlightfreq(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 freq[4] = { 0x66, 0x40, 0xa8, 0xe8 };
+
+	if (sd->freq == 2)	/* 60hz */
+		freq[1] = 0x00;
+
+	t16RegWrite(dev, 0x1, 0x0000, freq, 0x4);
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	unsigned int contrast = sd->contrast;
+	__u16 reg_to_write = 0x00;
+
+	if (contrast < 7)
+		reg_to_write = 0x8ea9 - (0x200 * contrast);
+	else
+		reg_to_write = (0x00a9 + ((contrast - 7) * 0x200));
+
+	t16RegWrite(dev, 0x00, reg_to_write, 0x00, 0);
+
+}
+
+static void setcolors(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u16 reg_to_write = 0x00;
+
+	reg_to_write = 0xc0bb + sd->colors * 0x100;
+	t16RegWrite(dev, 0x00, reg_to_write, 0x00, 0);
+}
+
+static void setgamma(struct gspca_dev *gspca_dev)
+{
+}
+
+static void setsharpness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	__u16 reg_to_write = 0x00;
+
+	reg_to_write = 0x0aa6 + 0x1000 * sd->sharpness;
+
+	t16RegWrite(dev, 0x00, reg_to_write, 0x00, 0x00);
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	*val = sd->brightness;
+	return *val;
+}
+
+static int sd_setwhitebalance(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->whitebalance = val;
+	if (gspca_dev->streaming)
+		setwhitebalance(gspca_dev);
+	return 0;
+}
+
+static int sd_getwhitebalance(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->whitebalance;
+	return *val;
+}
+
+
+static int sd_setflip(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->mirror = val;
+	if (gspca_dev->streaming)
+		setflip(gspca_dev);
+	return 0;
+}
+
+static int sd_getflip(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->mirror;
+	return *val;
+}
+
+static int sd_seteffect(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->effect = val;
+	if (gspca_dev->streaming)
+		seteffect(gspca_dev);
+	return 0;
+}
+
+static int sd_geteffect(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->effect;
+	return *val;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->contrast;
+	return *val;
+}
+
+static int sd_setcolors(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->colors = val;
+	if (gspca_dev->streaming)
+		setcolors(gspca_dev);
+	return 0;
+}
+
+static int sd_getcolors(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->colors;
+	return 0;
+}
+
+static int sd_setgamma(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->gamma = val;
+	if (gspca_dev->streaming)
+		setgamma(gspca_dev);
+	return 0;
+}
+
+static int sd_getgamma(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	*val = sd->gamma;
+	return 0;
+}
+
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->freq = val;
+	if (gspca_dev->streaming)
+		setlightfreq(gspca_dev);
+	return 0;
+}
+
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->freq;
+	return 0;
+}
+
+static int sd_setsharpness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->sharpness = val;
+	if (gspca_dev->streaming)
+		setsharpness(gspca_dev);
+	return 0;
+}
+
+static int sd_getsharpness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->sharpness;
+	return 0;
+}
+
+/* Low Light set  here......*/
+static int sd_setlowlight(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+
+	sd->autogain = val;
+	if (val != 0)
+		t16RegWrite(dev, 0x00, 0xf48e, 0x00, 0);
+	else
+		t16RegWrite(dev, 0x00, 0xb48e, 0x00, 0);
+	return 0;
+}
+
+static int sd_getlowlight(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autogain;
+	return 0;
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int mode;
+	__u8 test_byte;
+
+	static __u8 t1[] = { 0x66, 0x00, 0xa8, 0xe8 };
+	__u8 t2[] = { 0x07, 0x00, 0x0d, 0x60, 0x0e, 0x80 };
+	static __u8 t3[] =
+		{ 0xb3, 0x07, 0xb4, 0x00, 0xb5, 0x88, 0xb6, 0x02, 0xb7, 0x06,
+		  0xb8, 0x00, 0xb9, 0xe7, 0xba, 0x01 };
+	static __u8 t4[] = { 0x0b, 0x04, 0x0a, 0x40 };
+
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode]. mode;
+	switch (mode) {
+	case 1:		/* 352x288 */
+		t2[1] = 0x40;
+		break;
+	case 2:		/* 320x240 */
+		t2[1] = 0x10;
+		break;
+	case 3:		/* 176x144 */
+		t2[1] = 0x50;
+		break;
+	case 4:		/* 160x120 */
+		t2[1] = 0x20;
+		break;
+	default:	/* 640x480 (0x00) */
+		break;
+	}
+
+	t16RegWrite(dev, 0x01, 0x0000, tas5130a_sensor_init[0], 0x8);
+	t16RegWrite(dev, 0x01, 0x0000, tas5130a_sensor_init[1], 0x8);
+	t16RegWrite(dev, 0x01, 0x0000, tas5130a_sensor_init[2], 0x8);
+	t16RegWrite(dev, 0x01, 0x0000, tas5130a_sensor_init[3], 0x8);
+	t16RegWrite(dev, 0x00, 0x3c80, 0x00, 0x00);
+		/* just in case and to keep sync with logs  (for mine) */
+	t16RegWrite(dev, 0x01, 0x0000, tas5130a_sensor_init[3], 0x8);
+	t16RegWrite(dev, 0x00, 0x3c80, 0x00, 0x00);
+		/* just in case and to keep sync with logs  (for mine) */
+	t16RegWrite(dev, 0x01, 0x0000, t1, 4);
+	t16RegWrite(dev, 0x01, 0x0000, t2, 6);
+	t16RegRead(dev, 0x0012, &test_byte, 0x1);
+	t16RegWrite(dev, 0x01, 0x0000, t3, 0x10);
+	t16RegWrite(dev, 0x00, 0x0013, 0x00, 0x00);
+	t16RegWrite(dev, 0x01, 0x0000, t4, 0x4);
+	/* restart on each start, just in case, sometimes regs goes wrong
+	 * when using controls from app */
+	setbrightness(gspca_dev);
+	setcontrast(gspca_dev);
+	setcolors(gspca_dev);
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso packet length */
+{
+	int sof = 0;
+	static unsigned char ffd9[] = { 0xff, 0xd9 };
+
+	if (data[0] == 0x5a) {
+		/* Control Packet, after this came the header again,
+		 * but extra bytes came in the packet before this,
+		 * sometimes an EOF arrives, sometimes not... */
+		return;
+	}
+
+	if (data[len - 1] == 0xff && data[len] == 0xd9) {
+		/* Just in case, i have seen packets with the marker,
+		 * other's do not include it... */
+		data += 2;
+		len -= 4;
+	} else if (data[2] == 0xff && data[3] == 0xd8) {
+		sof = 1;
+		data += 2;
+		len -= 2;
+	} else {
+		data += 2;
+		len -= 2;
+	}
+
+	if (sof) {
+		/* extra bytes....., could be processed too but would be
+		 * a waste of time, right now leave the application and
+		 * libjpeg do it for ourserlves.. */
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					ffd9, 2);
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame, data, len);
+		return;
+	}
+
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+}
+
+static int sd_querymenu(struct gspca_dev *gspca_dev,
+			struct v4l2_querymenu *menu)
+{
+	memset(menu->name, 0, sizeof menu->name);
+
+	switch (menu->id) {
+	case V4L2_CID_POWER_LINE_FREQUENCY:
+		switch (menu->index) {
+		case 1:		/* V4L2_CID_POWER_LINE_FREQUENCY_50HZ */
+			strcpy(menu->name, "50 Hz");
+			return 0;
+		case 2:		/* V4L2_CID_POWER_LINE_FREQUENCY_60HZ */
+			strcpy(menu->name, "60 Hz");
+			return 0;
+		}
+		break;
+	case V4L2_CID_EFFECTS:
+		if (menu->index < 0 || menu->index >= NUM_EFFECTS_CONTROL)
+			return -EINVAL;
+		strncpy((char *) menu->name,
+			effects_control[menu->index].name, 32);
+		break;
+	}
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	init_default_parameters(gspca_dev);
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+	.querymenu = sd_querymenu,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x17a1, 0x0128), DVNM("XPX Webcam")},
+	{}
+};
+
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+		    const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+			       THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/tv8532.c b/drivers/media/video/gspca/tv8532.c
new file mode 100644
index 000000000000..6218441ba1f0
--- /dev/null
+++ b/drivers/media/video/gspca/tv8532.c
@@ -0,0 +1,709 @@
+/*
+ * Quickcam cameras initialization data
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+#define MODULE_NAME "tv8532"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("TV8532 USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	int buflen;			/* current length of tmpbuf */
+	__u8 tmpbuf[352 * 288 + 10 * 288];	/* no protection... */
+	__u8 tmpbuf2[352 * 288];		/* no protection... */
+
+	unsigned short brightness;
+	unsigned short contrast;
+
+	char packet;
+	char synchro;
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_BRIGHTNESS 0
+	{
+	 {
+	  .id = V4L2_CID_BRIGHTNESS,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Brightness",
+	  .minimum = 1,
+	  .maximum = 0x2ff,
+	  .step = 1,
+	  .default_value = 0x18f,
+	  },
+	 .set = sd_setbrightness,
+	 .get = sd_getbrightness,
+	 },
+#define SD_CONTRAST 1
+	{
+	 {
+	  .id = V4L2_CID_CONTRAST,
+	  .type = V4L2_CTRL_TYPE_INTEGER,
+	  .name = "Contrast",
+	  .minimum = 0,
+	  .maximum = 0xffff,
+	  .step = 1,
+	  .default_value = 0x7fff,
+	  },
+	 .set = sd_setcontrast,
+	 .get = sd_getcontrast,
+	 },
+};
+
+static struct cam_mode sif_mode[] = {
+	{V4L2_PIX_FMT_SBGGR8, 176, 144, 1},
+	{V4L2_PIX_FMT_SBGGR8, 352, 288, 0},
+};
+
+/*
+ * Initialization data: this is the first set-up data written to the
+ * device (before the open data).
+ */
+#define TESTCLK 0x10		/* reg 0x2c -> 0x12 //10 */
+#define TESTCOMP 0x90		/* reg 0x28 -> 0x80 */
+#define TESTLINE 0x81		/* reg 0x29 -> 0x81 */
+#define QCIFLINE 0x41		/* reg 0x29 -> 0x81 */
+#define TESTPTL 0x14		/* reg 0x2D -> 0x14 */
+#define TESTPTH 0x01		/* reg 0x2E -> 0x01 */
+#define TESTPTBL 0x12		/* reg 0x2F -> 0x0a */
+#define TESTPTBH 0x01		/* reg 0x30 -> 0x01 */
+#define ADWIDTHL 0xe8		/* reg 0x0c -> 0xe8 */
+#define ADWIDTHH 0x03		/* reg 0x0d -> 0x03 */
+#define ADHEIGHL 0x90		/* reg 0x0e -> 0x91 //93 */
+#define ADHEIGHH 0x01		/* reg 0x0f -> 0x01 */
+#define EXPOL 0x8f		/* reg 0x1c -> 0x8f */
+#define EXPOH 0x01		/* reg 0x1d -> 0x01 */
+#define ADCBEGINL 0x44		/* reg 0x10 -> 0x46 //47 */
+#define ADCBEGINH 0x00		/* reg 0x11 -> 0x00 */
+#define ADRBEGINL 0x0a		/* reg 0x14 -> 0x0b //0x0c */
+#define ADRBEGINH 0x00		/* reg 0x15 -> 0x00 */
+#define TV8532_CMD_UPDATE 0x84
+
+#define TV8532_EEprom_Add 0x03
+#define TV8532_EEprom_DataL 0x04
+#define TV8532_EEprom_DataM 0x05
+#define TV8532_EEprom_DataH 0x06
+#define TV8532_EEprom_TableLength 0x07
+#define TV8532_EEprom_Write 0x08
+#define TV8532_PART_CTRL 0x00
+#define TV8532_CTRL 0x01
+#define TV8532_CMD_EEprom_Open 0x30
+#define TV8532_CMD_EEprom_Close 0x29
+#define TV8532_UDP_UPDATE 0x31
+#define TV8532_GPIO 0x39
+#define TV8532_GPIO_OE 0x3B
+#define TV8532_REQ_RegWrite 0x02
+#define TV8532_REQ_RegRead 0x03
+
+#define TV8532_ADWIDTH_L 0x0C
+#define TV8532_ADWIDTH_H 0x0D
+#define TV8532_ADHEIGHT_L 0x0E
+#define TV8532_ADHEIGHT_H 0x0F
+#define TV8532_EXPOSURE 0x1C
+#define TV8532_QUANT_COMP 0x28
+#define TV8532_MODE_PACKET 0x29
+#define TV8532_SETCLK 0x2C
+#define TV8532_POINT_L 0x2D
+#define TV8532_POINT_H 0x2E
+#define TV8532_POINTB_L 0x2F
+#define TV8532_POINTB_H 0x30
+#define TV8532_BUDGET_L 0x2A
+#define TV8532_BUDGET_H 0x2B
+#define TV8532_VID_L 0x34
+#define TV8532_VID_H 0x35
+#define TV8532_PID_L 0x36
+#define TV8532_PID_H 0x37
+#define TV8532_DeviceID 0x83
+#define TV8532_AD_SLOPE 0x91
+#define TV8532_AD_BITCTRL 0x94
+#define TV8532_AD_COLBEGIN_L 0x10
+#define TV8532_AD_COLBEGIN_H 0x11
+#define TV8532_AD_ROWBEGIN_L 0x14
+#define TV8532_AD_ROWBEGIN_H 0x15
+
+static __u32 tv_8532_eeprom_data[] = {
+/*	add		dataL	   dataM	dataH */
+	0x00010001, 0x01018011, 0x02050014, 0x0305001c,
+	0x040d001e, 0x0505001f, 0x06050519, 0x0705011b,
+	0x0805091e, 0x090d892e, 0x0a05892f, 0x0b050dd9,
+	0x0c0509f1, 0
+};
+
+static void reg_r(struct usb_device *dev,
+		  __u16 index, __u8 *buffer)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			TV8532_REQ_RegRead,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,	/* value */
+			index, buffer, sizeof(__u8),
+			500);
+}
+
+static void reg_w(struct usb_device *dev,
+		  __u16 index, __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			TV8532_REQ_RegWrite,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			0,	/* value */
+			index, buffer, length, 500);
+}
+
+static void tv_8532WriteEEprom(struct gspca_dev *gspca_dev)
+{
+	int i = 0;
+	__u8 reg, data0, data1, data2, datacmd;
+	struct usb_device *dev = gspca_dev->dev;
+
+	datacmd = 0xb0;;
+	reg_w(dev, TV8532_GPIO, &datacmd, 1);
+	datacmd = TV8532_CMD_EEprom_Open;
+	reg_w(dev, TV8532_CTRL, &datacmd,
+			1);
+/*	msleep(1); */
+	while (tv_8532_eeprom_data[i]) {
+		reg = (tv_8532_eeprom_data[i] & 0xff000000) >> 24;
+		reg_w(dev, TV8532_EEprom_Add, &reg, 1);
+		/* msleep(1); */
+		data0 = (tv_8532_eeprom_data[i] & 0x000000ff);
+		reg_w(dev, TV8532_EEprom_DataL, &data0, 1);
+		/* msleep(1); */
+		data1 = (tv_8532_eeprom_data[i] & 0x0000FF00) >> 8;
+		reg_w(dev, TV8532_EEprom_DataM, &data1, 1);
+		/* msleep(1); */
+		data2 = (tv_8532_eeprom_data[i] & 0x00FF0000) >> 16;
+		reg_w(dev, TV8532_EEprom_DataH, &data2, 1);
+		/* msleep(1); */
+		datacmd = 0;
+		reg_w(dev, TV8532_EEprom_Write, &datacmd, 1);
+		/* msleep(10); */
+		i++;
+	}
+	datacmd = i;
+	reg_w(dev, TV8532_EEprom_TableLength, &datacmd, 1);
+/*	msleep(1); */
+	datacmd = TV8532_CMD_EEprom_Close;
+	reg_w(dev, TV8532_CTRL, &datacmd, 1);
+	msleep(10);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+		     const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct cam *cam;
+
+	tv_8532WriteEEprom(gspca_dev);
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 1;
+	cam->cam_mode = sif_mode;
+	cam->nmodes = sizeof sif_mode / sizeof sif_mode[0];
+
+	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
+	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
+	return 0;
+}
+
+static void tv_8532ReadRegisters(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 data = 0;
+/*	__u16 vid, pid; */
+
+	reg_r(dev, 0x0001, &data);
+	PDEBUG(D_USBI, "register 0x01-> %x", data);
+	reg_r(dev, 0x0002, &data);
+	PDEBUG(D_USBI, "register 0x02-> %x", data);
+	reg_r(dev, TV8532_ADWIDTH_L, &data);
+	reg_r(dev, TV8532_ADWIDTH_H, &data);
+	reg_r(dev, TV8532_QUANT_COMP, &data);
+	reg_r(dev, TV8532_MODE_PACKET, &data);
+	reg_r(dev, TV8532_SETCLK, &data);
+	reg_r(dev, TV8532_POINT_L, &data);
+	reg_r(dev, TV8532_POINT_H, &data);
+	reg_r(dev, TV8532_POINTB_L, &data);
+	reg_r(dev, TV8532_POINTB_H, &data);
+	reg_r(dev, TV8532_BUDGET_L, &data);
+	reg_r(dev, TV8532_BUDGET_H, &data);
+	reg_r(dev, TV8532_VID_L, &data);
+	reg_r(dev, TV8532_VID_H, &data);
+	reg_r(dev, TV8532_PID_L, &data);
+	reg_r(dev, TV8532_PID_H, &data);
+	reg_r(dev, TV8532_DeviceID, &data);
+	reg_r(dev, TV8532_AD_COLBEGIN_L, &data);
+	reg_r(dev, TV8532_AD_COLBEGIN_H, &data);
+	reg_r(dev, TV8532_AD_ROWBEGIN_L, &data);
+	reg_r(dev, TV8532_AD_ROWBEGIN_H, &data);
+}
+
+static void tv_8532_setReg(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 data = 0;
+	__u8 value[2] = { 0, 0 };
+
+	data = ADCBEGINL;
+	reg_w(dev, TV8532_AD_COLBEGIN_L, &data, 1);	/* 0x10 */
+	data = ADCBEGINH;	/* also digital gain */
+	reg_w(dev, TV8532_AD_COLBEGIN_H, &data, 1);
+	data = TV8532_CMD_UPDATE;
+	reg_w(dev, TV8532_PART_CTRL, &data, 1);	/* 0x00<-0x84 */
+
+	data = 0x0a;
+	reg_w(dev, TV8532_GPIO_OE, &data, 1);
+	/******************************************************/
+	data = ADHEIGHL;
+	reg_w(dev, TV8532_ADHEIGHT_L, &data, 1);	/* 0e */
+	data = ADHEIGHH;
+	reg_w(dev, TV8532_ADHEIGHT_H, &data, 1);	/* 0f */
+	value[0] = EXPOL;
+	value[1] = EXPOH;	/* 350d 0x014c; */
+	reg_w(dev, TV8532_EXPOSURE, value, 2);	/* 1c */
+	data = ADCBEGINL;
+	reg_w(dev, TV8532_AD_COLBEGIN_L, &data, 1);	/* 0x10 */
+	data = ADCBEGINH;	/* also digital gain */
+	reg_w(dev, TV8532_AD_COLBEGIN_H, &data, 1);
+	data = ADRBEGINL;
+	reg_w(dev, TV8532_AD_ROWBEGIN_L, &data, 1);	/* 0x14 */
+
+	data = 0x00;
+	reg_w(dev, TV8532_AD_SLOPE, &data, 1);	/* 0x91 */
+	data = 0x02;
+	reg_w(dev, TV8532_AD_BITCTRL, &data, 1);	/* 0x94 */
+
+
+	data = TV8532_CMD_EEprom_Close;
+	reg_w(dev, TV8532_CTRL, &data, 1);	/* 0x01 */
+
+	data = 0x00;
+	reg_w(dev, TV8532_AD_SLOPE, &data, 1);	/* 0x91 */
+	data = TV8532_CMD_UPDATE;
+	reg_w(dev, TV8532_PART_CTRL, &data, 1);	/* 0x00<-0x84 */
+}
+
+static void tv_8532_PollReg(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 data = 0;
+	int i;
+
+	/* strange polling from tgc */
+	for (i = 0; i < 10; i++) {
+		data = TESTCLK;	/* 0x48; //0x08; */
+		reg_w(dev, TV8532_SETCLK, &data, 1);	/* 0x2c */
+		data = TV8532_CMD_UPDATE;
+		reg_w(dev, TV8532_PART_CTRL, &data, 1);
+		data = 0x01;
+		reg_w(dev, TV8532_UDP_UPDATE, &data, 1);	/* 0x31 */
+	}
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 data = 0;
+	__u8 dataStart = 0;
+	__u8 value[2] = { 0, 0 };
+
+	data = 0x32;
+	reg_w(dev, TV8532_AD_SLOPE, &data, 1);
+	data = 0;
+	reg_w(dev, TV8532_AD_BITCTRL, &data, 1);
+	tv_8532ReadRegisters(gspca_dev);
+	data = 0x0b;
+	reg_w(dev, TV8532_GPIO_OE, &data, 1);
+	value[0] = ADHEIGHL;
+	value[1] = ADHEIGHH;	/* 401d 0x0169; */
+	reg_w(dev, TV8532_ADHEIGHT_L, value, 2);	/* 0e */
+	value[0] = EXPOL;
+	value[1] = EXPOH;	/* 350d 0x014c; */
+	reg_w(dev, TV8532_EXPOSURE, value, 2);	/* 1c */
+	data = ADWIDTHL;	/* 0x20; */
+	reg_w(dev, TV8532_ADWIDTH_L, &data, 1);	/* 0x0c */
+	data = ADWIDTHH;
+	reg_w(dev, TV8532_ADWIDTH_H, &data, 1);	/* 0x0d */
+
+	/*******************************************************************/
+	data = TESTCOMP;	/* 0x72 compressed mode */
+	reg_w(dev, TV8532_QUANT_COMP, &data, 1);	/* 0x28 */
+	data = TESTLINE;	/* 0x84; // CIF | 4 packet */
+	reg_w(dev, TV8532_MODE_PACKET, &data, 1);	/* 0x29 */
+
+	/************************************************/
+	data = TESTCLK;		/* 0x48; //0x08; */
+	reg_w(dev, TV8532_SETCLK, &data, 1);	/* 0x2c */
+	data = TESTPTL;		/* 0x38; */
+	reg_w(dev, TV8532_POINT_L, &data, 1);	/* 0x2d */
+	data = TESTPTH;		/* 0x04; */
+	reg_w(dev, TV8532_POINT_H, &data, 1);	/* 0x2e */
+	dataStart = TESTPTBL;	/* 0x04; */
+	reg_w(dev, TV8532_POINTB_L, &dataStart, 1);	/* 0x2f */
+	data = TESTPTBH;	/* 0x04; */
+	reg_w(dev, TV8532_POINTB_H, &data, 1);	/* 0x30 */
+	data = TV8532_CMD_UPDATE;
+	reg_w(dev, TV8532_PART_CTRL, &data, 1);	/* 0x00<-0x84 */
+	/*************************************************/
+	data = 0x01;
+	reg_w(dev, TV8532_UDP_UPDATE, &data, 1);	/* 0x31 */
+	msleep(200);
+	data = 0x00;
+	reg_w(dev, TV8532_UDP_UPDATE, &data, 1);	/* 0x31 */
+	/*************************************************/
+	tv_8532_setReg(gspca_dev);
+	/*************************************************/
+	data = 0x0b;
+	reg_w(dev, TV8532_GPIO_OE, &data,
+			1);
+	/*************************************************/
+	tv_8532_setReg(gspca_dev);
+	/*************************************************/
+	tv_8532_PollReg(gspca_dev);
+	return 0;
+}
+
+static void setbrightness(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	__u8 value[2];
+	__u8 data;
+	int brightness = sd->brightness;
+
+	value[1] = (brightness >> 8) & 0xff;
+	value[0] = (brightness) & 0xff;
+	reg_w(gspca_dev->dev, TV8532_EXPOSURE, value, 2);	/* 1c */
+	data = TV8532_CMD_UPDATE;
+	reg_w(gspca_dev->dev, TV8532_PART_CTRL, &data, 1);
+}
+
+/* -- start the camera -- */
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 data = 0;
+	__u8 dataStart = 0;
+	__u8 value[2];
+
+	data = 0x32;
+	reg_w(dev, TV8532_AD_SLOPE, &data, 1);
+	data = 0;
+	reg_w(dev, TV8532_AD_BITCTRL, &data, 1);
+	tv_8532ReadRegisters(gspca_dev);
+	data = 0x0b;
+	reg_w(dev, TV8532_GPIO_OE, &data, 1);
+	value[0] = ADHEIGHL;
+	value[1] = ADHEIGHH;	/* 401d 0x0169; */
+	reg_w(dev, TV8532_ADHEIGHT_L, value, 2);	/* 0e */
+/*	value[0] = EXPOL; value[1] =EXPOH; 		 * 350d 0x014c; */
+/*	reg_w(dev,TV8532_REQ_RegWrite,0,TV8532_EXPOSURE,value,2);  * 1c */
+	setbrightness(gspca_dev);
+
+	data = ADWIDTHL;	/* 0x20; */
+	reg_w(dev, TV8532_ADWIDTH_L, &data, 1);	/* 0x0c */
+	data = ADWIDTHH;
+	reg_w(dev, TV8532_ADWIDTH_H, &data, 1);	/* 0x0d */
+
+	/************************************************/
+	data = TESTCOMP;	/* 0x72 compressed mode */
+	reg_w(dev, TV8532_QUANT_COMP, &data, 1);	/* 0x28 */
+	if (gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode) {
+		/* 176x144 */
+		data = QCIFLINE;	/* 0x84; // CIF | 4 packet */
+		reg_w(dev, TV8532_MODE_PACKET, &data, 1);	/* 0x29 */
+	} else {
+		/* 352x288 */
+		data = TESTLINE;	/* 0x84; // CIF | 4 packet */
+		reg_w(dev, TV8532_MODE_PACKET, &data, 1);	/* 0x29 */
+	}
+	/************************************************/
+	data = TESTCLK;		/* 0x48; //0x08; */
+	reg_w(dev, TV8532_SETCLK, &data, 1);	/* 0x2c */
+	data = TESTPTL;		/* 0x38; */
+	reg_w(dev, TV8532_POINT_L, &data, 1);	/* 0x2d */
+	data = TESTPTH;		/* 0x04; */
+	reg_w(dev, TV8532_POINT_H, &data, 1);	/* 0x2e */
+	dataStart = TESTPTBL;	/* 0x04; */
+	reg_w(dev, TV8532_POINTB_L, &dataStart, 1);	/* 0x2f */
+	data = TESTPTBH;	/* 0x04; */
+	reg_w(dev, TV8532_POINTB_H, &data, 1);	/* 0x30 */
+	data = TV8532_CMD_UPDATE;
+	reg_w(dev, TV8532_PART_CTRL, &data, 1);	/* 0x00<-0x84 */
+	/************************************************/
+	data = 0x01;
+	reg_w(dev, TV8532_UDP_UPDATE, &data, 1);	/* 0x31 */
+	msleep(200);
+	data = 0x00;
+	reg_w(dev, TV8532_UDP_UPDATE, &data, 1);	/* 0x31 */
+	/************************************************/
+	tv_8532_setReg(gspca_dev);
+	/************************************************/
+	data = 0x0b;
+	reg_w(dev, TV8532_GPIO_OE, &data, 1);
+	/************************************************/
+	tv_8532_setReg(gspca_dev);
+	/************************************************/
+	tv_8532_PollReg(gspca_dev);
+	data = 0x00;
+	reg_w(dev, TV8532_UDP_UPDATE, &data, 1);	/* 0x31 */
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	__u8 data;
+
+	data = 0x0b;
+	reg_w(dev, TV8532_GPIO_OE, &data, 1);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+}
+
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+}
+
+static void tv8532_preprocess(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+/* we should received a whole frame with header and EOL marker
+ * in gspca_dev->tmpbuf and return a GBRG pattern in gspca_dev->tmpbuf2
+ * sequence 2bytes header the Alternate pixels bayer GB 4 bytes
+ * Alternate pixels bayer RG 4 bytes EOL */
+	int width = gspca_dev->width;
+	int height = gspca_dev->height;
+	unsigned char *dst = sd->tmpbuf2;
+	unsigned char *data = sd->tmpbuf;
+	int i;
+
+	/* precompute where is the good bayer line */
+	if (((data[3] + data[width + 7]) >> 1)
+	    + (data[4] >> 2)
+	    + (data[width + 6] >> 1) >= ((data[2] + data[width + 6]) >> 1)
+	    + (data[3] >> 2)
+	    + (data[width + 5] >> 1))
+		data += 3;
+	else
+		data += 2;
+	for (i = 0; i < height / 2; i++) {
+		memcpy(dst, data, width);
+		data += width + 3;
+		dst += width;
+		memcpy(dst, data, width);
+		data += width + 7;
+		dst += width;
+	}
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			__u8 *data,			/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	if (data[0] != 0x80) {
+		sd->packet++;
+		if (sd->buflen + len > sizeof sd->tmpbuf) {
+			if (gspca_dev->last_packet_type != DISCARD_PACKET) {
+				PDEBUG(D_PACK, "buffer overflow");
+				gspca_dev->last_packet_type = DISCARD_PACKET;
+			}
+			return;
+		}
+		memcpy(&sd->tmpbuf[sd->buflen], data, len);
+		sd->buflen += len;
+		return;
+	}
+
+	/* here we detect 0x80 */
+	/* counter is limited so we need few header for a frame :) */
+
+	/* header 0x80 0x80 0x80 0x80 0x80 */
+	/* packet  00   63  127  145  00   */
+	/* sof     0     1   1    0    0   */
+
+	/* update sequence */
+	if (sd->packet == 63 || sd->packet == 127)
+		sd->synchro = 1;
+
+	/* is there a frame start ? */
+	if (sd->packet >= (gspca_dev->height >> 1) - 1) {
+		PDEBUG(D_PACK, "SOF > %d packet %d", sd->synchro,
+		       sd->packet);
+		if (!sd->synchro) {	/* start of frame */
+			if (gspca_dev->last_packet_type == FIRST_PACKET) {
+				tv8532_preprocess(gspca_dev);
+				frame = gspca_frame_add(gspca_dev,
+							LAST_PACKET,
+							frame, sd->tmpbuf2,
+							gspca_dev->width *
+							    gspca_dev->width);
+			}
+			gspca_frame_add(gspca_dev, FIRST_PACKET,
+					frame, data, 0);
+			memcpy(sd->tmpbuf, data, len);
+			sd->buflen = len;
+			sd->packet = 0;
+			return;
+		}
+		if (gspca_dev->last_packet_type != DISCARD_PACKET) {
+			PDEBUG(D_PACK,
+			       "Warning wrong TV8532 frame detection %d",
+			       sd->packet);
+			gspca_dev->last_packet_type = DISCARD_PACKET;
+		}
+		return;
+	}
+
+	if (!sd->synchro) {
+		/* Drop packet frame corrupt */
+		PDEBUG(D_PACK, "DROP SOF %d packet %d",
+		       sd->synchro, sd->packet);
+		sd->packet = 0;
+		gspca_dev->last_packet_type = DISCARD_PACKET;
+		return;
+	}
+	sd->synchro = 1;
+	sd->packet++;
+	memcpy(&sd->tmpbuf[sd->buflen], data, len);
+	sd->buflen += len;
+}
+
+static void setcontrast(struct gspca_dev *gspca_dev)
+{
+}
+
+static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->brightness = val;
+	if (gspca_dev->streaming)
+		setbrightness(gspca_dev);
+	return 0;
+}
+
+static int sd_getbrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->brightness;
+	return 0;
+}
+
+static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->contrast = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
+static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->contrast;
+	return 0;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x046d, 0x0920), DVNM("QC Express")},
+	{USB_DEVICE(0x046d, 0x0921), DVNM("Labtec Webcam")},
+	{USB_DEVICE(0x0545, 0x808b), DVNM("Veo Stingray")},
+	{USB_DEVICE(0x0545, 0x8333), DVNM("Veo Stingray")},
+	{USB_DEVICE(0x0923, 0x010f), DVNM("ICM532 cams")},
+	{}
+};
+
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+		    const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+			       THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/vc032x.c b/drivers/media/video/gspca/vc032x.c
new file mode 100644
index 000000000000..8b5f6d17d2a5
--- /dev/null
+++ b/drivers/media/video/gspca/vc032x.c
@@ -0,0 +1,1816 @@
+/*
+ *		Z-star vc0321 library
+ *		Copyright (C) 2006 Koninski Artur takeshi87@o2.pl
+ *		Copyright (C) 2006 Michel Xhaard
+ *
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#define MODULE_NAME "vc032x"
+
+#include "gspca.h"
+
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
+
+MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>");
+MODULE_DESCRIPTION("GSPCA/VC032X USB Camera Driver");
+MODULE_LICENSE("GPL");
+
+/* specific webcam descriptor */
+struct sd {
+	struct gspca_dev gspca_dev;	/* !! must be the first item */
+
+	unsigned char autogain;
+	unsigned char lightfreq;
+
+	char qindex;
+	char bridge;
+#define BRIDGE_VC0321 0
+#define BRIDGE_VC0323 1
+	char sensor;
+#define SENSOR_HV7131R 0
+#define SENSOR_MI1320 1
+#define SENSOR_MI1310_SOC 2
+#define SENSOR_OV7660 3
+#define SENSOR_OV7670 4
+#define SENSOR_PO3130NC 5
+};
+
+/* V4L2 controls supported by the driver */
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val);
+
+static struct ctrl sd_ctrls[] = {
+#define SD_AUTOGAIN 0
+	{
+	    {
+		.id      = V4L2_CID_AUTOGAIN,
+		.type    = V4L2_CTRL_TYPE_BOOLEAN,
+		.name    = "Auto Gain",
+		.minimum = 0,
+		.maximum = 1,
+		.step    = 1,
+		.default_value = 1,
+	    },
+	    .set = sd_setautogain,
+	    .get = sd_getautogain,
+	},
+#define SD_FREQ 1
+	{
+	    {
+		.id	 = V4L2_CID_POWER_LINE_FREQUENCY,
+		.type    = V4L2_CTRL_TYPE_MENU,
+		.name    = "Light frequency filter",
+		.minimum = 0,
+		.maximum = 2,	/* 0: 0, 1: 50Hz, 2:60Hz */
+		.step    = 1,
+		.default_value = 1,
+	    },
+	    .set = sd_setfreq,
+	    .get = sd_getfreq,
+	},
+};
+
+static struct cam_mode vc0321_mode[] = {
+	{V4L2_PIX_FMT_YUYV, 320, 240, 1},
+	{V4L2_PIX_FMT_YUYV, 640, 480, 0},
+};
+static struct cam_mode vc0323_mode[] = {
+	{V4L2_PIX_FMT_JPEG, 320, 240, 1},
+	{V4L2_PIX_FMT_JPEG, 640, 480, 0},
+};
+
+static __u8 mi1310_socinitVGA_JPG[][4] = {
+	{0xb0, 0x03, 0x19, 0xcc},
+	{0xb0, 0x04, 0x02, 0xcc},
+	{0xb3, 0x00, 0x64, 0xcc},
+	{0xb3, 0x00, 0x65, 0xcc},
+	{0xb3, 0x05, 0x00, 0xcc},
+	{0xb3, 0x06, 0x00, 0xcc},
+	{0xb3, 0x08, 0x01, 0xcc},
+	{0xb3, 0x09, 0x0c, 0xcc},
+	{0xb3, 0x34, 0x02, 0xcc},
+	{0xb3, 0x35, 0xdd, 0xcc},
+	{0xb3, 0x02, 0x00, 0xcc},
+	{0xb3, 0x03, 0x0a, 0xcc},
+	{0xb3, 0x04, 0x05, 0xcc},
+	{0xb3, 0x20, 0x00, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},
+	{0xb3, 0x22, 0x03, 0xcc},
+	{0xb3, 0x23, 0xc0, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},
+	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x04, 0xcc},
+	{0xb3, 0x17, 0xff, 0xcc},
+	{0xb3, 0x00, 0x65, 0xcc},
+	{0xb8, 0x00, 0x00, 0xcc},
+	{0xbc, 0x00, 0xd0, 0xcc},
+	{0xbc, 0x01, 0x01, 0xcc},
+	{0xf0, 0x00, 0x02, 0xbb},
+	{0xc8, 0x9f, 0x0b, 0xbb},
+	{0x5b, 0x00, 0x01, 0xbb},
+	{0x2f, 0xde, 0x20, 0xbb},
+	{0xf0, 0x00, 0x00, 0xbb},
+	{0x20, 0x03, 0x02, 0xbb},
+	{0xf0, 0x00, 0x01, 0xbb},
+	{0x05, 0x00, 0x07, 0xbb},
+	{0x34, 0x00, 0x00, 0xbb},
+	{0x35, 0xff, 0x00, 0xbb},
+	{0xdc, 0x07, 0x02, 0xbb},
+	{0xdd, 0x3c, 0x18, 0xbb},
+	{0xde, 0x92, 0x6d, 0xbb},
+	{0xdf, 0xcd, 0xb1, 0xbb},
+	{0xe0, 0xff, 0xe7, 0xbb},
+	{0x06, 0xf0, 0x0d, 0xbb},
+	{0x06, 0x70, 0x0e, 0xbb},
+	{0x4c, 0x00, 0x01, 0xbb},
+	{0x4d, 0x00, 0x01, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},
+	{0x2e, 0x0c, 0x55, 0xbb},
+	{0x21, 0xb6, 0x6e, 0xbb},
+	{0x36, 0x30, 0x10, 0xbb},
+	{0x37, 0x00, 0xc1, 0xbb},
+	{0xf0, 0x00, 0x00, 0xbb},
+	{0x07, 0x00, 0x84, 0xbb},
+	{0x08, 0x02, 0x4a, 0xbb},
+	{0x05, 0x01, 0x10, 0xbb},
+	{0x06, 0x00, 0x39, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},
+	{0x58, 0x02, 0x67, 0xbb},
+	{0x57, 0x02, 0x00, 0xbb},
+	{0x5a, 0x02, 0x67, 0xbb},
+	{0x59, 0x02, 0x00, 0xbb},
+	{0x5c, 0x12, 0x0d, 0xbb},
+	{0x5d, 0x16, 0x11, 0xbb},
+	{0x39, 0x06, 0x18, 0xbb},
+	{0x3a, 0x06, 0x18, 0xbb},
+	{0x3b, 0x06, 0x18, 0xbb},
+	{0x3c, 0x06, 0x18, 0xbb},
+	{0x64, 0x7b, 0x5b, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},
+	{0x36, 0x30, 0x10, 0xbb},
+	{0x37, 0x00, 0xc0, 0xbb},
+	{0xbc, 0x0e, 0x00, 0xcc},
+	{0xbc, 0x0f, 0x05, 0xcc},
+	{0xbc, 0x10, 0xc0, 0xcc},
+	{0xbc, 0x11, 0x03, 0xcc},
+	{0xb6, 0x00, 0x00, 0xcc},
+	{0xb6, 0x03, 0x02, 0xcc},
+	{0xb6, 0x02, 0x80, 0xcc},
+	{0xb6, 0x05, 0x01, 0xcc},
+	{0xb6, 0x04, 0xe0, 0xcc},
+	{0xb6, 0x12, 0xf8, 0xcc},
+	{0xb6, 0x13, 0x25, 0xcc},
+	{0xb6, 0x18, 0x02, 0xcc},
+	{0xb6, 0x17, 0x58, 0xcc},
+	{0xb6, 0x16, 0x00, 0xcc},
+	{0xb6, 0x22, 0x12, 0xcc},
+	{0xb6, 0x23, 0x0b, 0xcc},
+	{0xbf, 0xc0, 0x39, 0xcc},
+	{0xbf, 0xc1, 0x04, 0xcc},
+	{0xbf, 0xcc, 0x00, 0xcc},
+	{0xbc, 0x02, 0x18, 0xcc},
+	{0xbc, 0x03, 0x50, 0xcc},
+	{0xbc, 0x04, 0x18, 0xcc},
+	{0xbc, 0x05, 0x00, 0xcc},
+	{0xbc, 0x06, 0x00, 0xcc},
+	{0xbc, 0x08, 0x30, 0xcc},
+	{0xbc, 0x09, 0x40, 0xcc},
+	{0xbc, 0x0a, 0x10, 0xcc},
+	{0xbc, 0x0b, 0x00, 0xcc},
+	{0xbc, 0x0c, 0x00, 0xcc},
+	{0xb3, 0x5c, 0x01, 0xcc},
+	{0xf0, 0x00, 0x01, 0xbb},
+	{0x80, 0x00, 0x03, 0xbb},
+	{0x81, 0xc7, 0x14, 0xbb},
+	{0x82, 0xeb, 0xe8, 0xbb},
+	{0x83, 0xfe, 0xf4, 0xbb},
+	{0x84, 0xcd, 0x10, 0xbb},
+	{0x85, 0xf3, 0xee, 0xbb},
+	{0x86, 0xff, 0xf1, 0xbb},
+	{0x87, 0xcd, 0x10, 0xbb},
+	{0x88, 0xf3, 0xee, 0xbb},
+	{0x89, 0x01, 0xf1, 0xbb},
+	{0x8a, 0xe5, 0x17, 0xbb},
+	{0x8b, 0xe8, 0xe2, 0xbb},
+	{0x8c, 0xf7, 0xed, 0xbb},
+	{0x8d, 0x00, 0xff, 0xbb},
+	{0x8e, 0xec, 0x10, 0xbb},
+	{0x8f, 0xf0, 0xed, 0xbb},
+	{0x90, 0xf9, 0xf2, 0xbb},
+	{0x91, 0x00, 0x00, 0xbb},
+	{0x92, 0xe9, 0x0d, 0xbb},
+	{0x93, 0xf4, 0xf2, 0xbb},
+	{0x94, 0xfb, 0xf5, 0xbb},
+	{0x95, 0x00, 0xff, 0xbb},
+	{0xb6, 0x0f, 0x08, 0xbb},
+	{0xb7, 0x3d, 0x16, 0xbb},
+	{0xb8, 0x0c, 0x04, 0xbb},
+	{0xb9, 0x1c, 0x07, 0xbb},
+	{0xba, 0x0a, 0x03, 0xbb},
+	{0xbb, 0x1b, 0x09, 0xbb},
+	{0xbc, 0x17, 0x0d, 0xbb},
+	{0xbd, 0x23, 0x1d, 0xbb},
+	{0xbe, 0x00, 0x28, 0xbb},
+	{0xbf, 0x11, 0x09, 0xbb},
+	{0xc0, 0x16, 0x15, 0xbb},
+	{0xc1, 0x00, 0x1b, 0xbb},
+	{0xc2, 0x0e, 0x07, 0xbb},
+	{0xc3, 0x14, 0x10, 0xbb},
+	{0xc4, 0x00, 0x17, 0xbb},
+	{0x06, 0x74, 0x8e, 0xbb},
+	{0xf0, 0x00, 0x01, 0xbb},
+	{0x06, 0xf4, 0x8e, 0xbb},
+	{0x00, 0x00, 0x50, 0xdd},
+	{0x06, 0x74, 0x8e, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},
+	{0x24, 0x50, 0x20, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},
+	{0x34, 0x0c, 0x50, 0xbb},
+	{0xb3, 0x01, 0x41, 0xcc},
+	{0xf0, 0x00, 0x00, 0xbb},
+	{0x03, 0x03, 0xc0, 0xbb},
+	{},
+};
+static __u8 mi1310_socinitQVGA_JPG[][4] = {
+	{0xb0, 0x03, 0x19, 0xcc},	{0xb0, 0x04, 0x02, 0xcc},
+	{0xb3, 0x00, 0x64, 0xcc},	{0xb3, 0x00, 0x65, 0xcc},
+	{0xb3, 0x05, 0x00, 0xcc},	{0xb3, 0x06, 0x00, 0xcc},
+	{0xb3, 0x08, 0x01, 0xcc},	{0xb3, 0x09, 0x0c, 0xcc},
+	{0xb3, 0x34, 0x02, 0xcc},	{0xb3, 0x35, 0xdd, 0xcc},
+	{0xb3, 0x02, 0x00, 0xcc},	{0xb3, 0x03, 0x0a, 0xcc},
+	{0xb3, 0x04, 0x05, 0xcc},	{0xb3, 0x20, 0x00, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},	{0xb3, 0x22, 0x03, 0xcc},
+	{0xb3, 0x23, 0xc0, 0xcc},	{0xb3, 0x14, 0x00, 0xcc},
+	{0xb3, 0x15, 0x00, 0xcc},	{0xb3, 0x16, 0x04, 0xcc},
+	{0xb3, 0x17, 0xff, 0xcc},	{0xb3, 0x00, 0x65, 0xcc},
+	{0xb8, 0x00, 0x00, 0xcc},	{0xbc, 0x00, 0xf0, 0xcc},
+	{0xbc, 0x01, 0x01, 0xcc},	{0xf0, 0x00, 0x02, 0xbb},
+	{0xc8, 0x9f, 0x0b, 0xbb},	{0x5b, 0x00, 0x01, 0xbb},
+	{0x2f, 0xde, 0x20, 0xbb},	{0xf0, 0x00, 0x00, 0xbb},
+	{0x20, 0x03, 0x02, 0xbb},	{0xf0, 0x00, 0x01, 0xbb},
+	{0x05, 0x00, 0x07, 0xbb},	{0x34, 0x00, 0x00, 0xbb},
+	{0x35, 0xff, 0x00, 0xbb},	{0xdc, 0x07, 0x02, 0xbb},
+	{0xdd, 0x3c, 0x18, 0xbb},	{0xde, 0x92, 0x6d, 0xbb},
+	{0xdf, 0xcd, 0xb1, 0xbb},	{0xe0, 0xff, 0xe7, 0xbb},
+	{0x06, 0xf0, 0x0d, 0xbb},	{0x06, 0x70, 0x0e, 0xbb},
+	{0x4c, 0x00, 0x01, 0xbb},	{0x4d, 0x00, 0x01, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x2e, 0x0c, 0x55, 0xbb},
+	{0x21, 0xb6, 0x6e, 0xbb},	{0x36, 0x30, 0x10, 0xbb},
+	{0x37, 0x00, 0xc1, 0xbb},	{0xf0, 0x00, 0x00, 0xbb},
+	{0x07, 0x00, 0x84, 0xbb},	{0x08, 0x02, 0x4a, 0xbb},
+	{0x05, 0x01, 0x10, 0xbb},	{0x06, 0x00, 0x39, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x58, 0x02, 0x67, 0xbb},
+	{0x57, 0x02, 0x00, 0xbb},	{0x5a, 0x02, 0x67, 0xbb},
+	{0x59, 0x02, 0x00, 0xbb},	{0x5c, 0x12, 0x0d, 0xbb},
+	{0x5d, 0x16, 0x11, 0xbb},	{0x39, 0x06, 0x18, 0xbb},
+	{0x3a, 0x06, 0x18, 0xbb},	{0x3b, 0x06, 0x18, 0xbb},
+	{0x3c, 0x06, 0x18, 0xbb},	{0x64, 0x7b, 0x5b, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x36, 0x30, 0x10, 0xbb},
+	{0x37, 0x00, 0xc0, 0xbb},	{0xbc, 0x0e, 0x00, 0xcc},
+	{0xbc, 0x0f, 0x05, 0xcc},	{0xbc, 0x10, 0xc0, 0xcc},
+	{0xbc, 0x11, 0x03, 0xcc},	{0xb6, 0x00, 0x00, 0xcc},
+	{0xb6, 0x03, 0x01, 0xcc},	{0xb6, 0x02, 0x40, 0xcc},
+	{0xb6, 0x05, 0x00, 0xcc},	{0xb6, 0x04, 0xf0, 0xcc},
+	{0xb6, 0x12, 0xf8, 0xcc},	{0xb6, 0x13, 0x25, 0xcc},
+	{0xb6, 0x18, 0x00, 0xcc},	{0xb6, 0x17, 0x96, 0xcc},
+	{0xb6, 0x16, 0x00, 0xcc},	{0xb6, 0x22, 0x12, 0xcc},
+	{0xb6, 0x23, 0x0b, 0xcc},	{0xbf, 0xc0, 0x39, 0xcc},
+	{0xbf, 0xc1, 0x04, 0xcc},	{0xbf, 0xcc, 0x00, 0xcc},
+	{0xb3, 0x5c, 0x01, 0xcc},	{0xf0, 0x00, 0x01, 0xbb},
+	{0x80, 0x00, 0x03, 0xbb},	{0x81, 0xc7, 0x14, 0xbb},
+	{0x82, 0xeb, 0xe8, 0xbb},	{0x83, 0xfe, 0xf4, 0xbb},
+	{0x84, 0xcd, 0x10, 0xbb},	{0x85, 0xf3, 0xee, 0xbb},
+	{0x86, 0xff, 0xf1, 0xbb},	{0x87, 0xcd, 0x10, 0xbb},
+	{0x88, 0xf3, 0xee, 0xbb},	{0x89, 0x01, 0xf1, 0xbb},
+	{0x8a, 0xe5, 0x17, 0xbb},	{0x8b, 0xe8, 0xe2, 0xbb},
+	{0x8c, 0xf7, 0xed, 0xbb},	{0x8d, 0x00, 0xff, 0xbb},
+	{0x8e, 0xec, 0x10, 0xbb},	{0x8f, 0xf0, 0xed, 0xbb},
+	{0x90, 0xf9, 0xf2, 0xbb},	{0x91, 0x00, 0x00, 0xbb},
+	{0x92, 0xe9, 0x0d, 0xbb},	{0x93, 0xf4, 0xf2, 0xbb},
+	{0x94, 0xfb, 0xf5, 0xbb},	{0x95, 0x00, 0xff, 0xbb},
+	{0xb6, 0x0f, 0x08, 0xbb},	{0xb7, 0x3d, 0x16, 0xbb},
+	{0xb8, 0x0c, 0x04, 0xbb},	{0xb9, 0x1c, 0x07, 0xbb},
+	{0xba, 0x0a, 0x03, 0xbb},	{0xbb, 0x1b, 0x09, 0xbb},
+	{0xbc, 0x17, 0x0d, 0xbb},	{0xbd, 0x23, 0x1d, 0xbb},
+	{0xbe, 0x00, 0x28, 0xbb},	{0xbf, 0x11, 0x09, 0xbb},
+	{0xc0, 0x16, 0x15, 0xbb},	{0xc1, 0x00, 0x1b, 0xbb},
+	{0xc2, 0x0e, 0x07, 0xbb},	{0xc3, 0x14, 0x10, 0xbb},
+	{0xc4, 0x00, 0x17, 0xbb},	{0x06, 0x74, 0x8e, 0xbb},
+	{0xf0, 0x00, 0x01, 0xbb},	{0x06, 0xf4, 0x8e, 0xbb},
+	{0x00, 0x00, 0x50, 0xdd},	{0x06, 0x74, 0x8e, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x24, 0x50, 0x20, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x34, 0x0c, 0x50, 0xbb},
+	{0xb3, 0x01, 0x41, 0xcc},	{0xf0, 0x00, 0x00, 0xbb},
+	{0x03, 0x03, 0xc0, 0xbb},
+	{},
+};
+
+static __u8 mi1320_gamma[17] = {
+	0x00, 0x13, 0x38, 0x59, 0x79, 0x92, 0xa7, 0xb9, 0xc8,
+	0xd4, 0xdf, 0xe7, 0xee, 0xf4, 0xf9, 0xfc, 0xff
+};
+static __u8 mi1320_matrix[9] = {
+	0x54, 0xda, 0x06, 0xf1, 0x50, 0xf4, 0xf7, 0xea, 0x52
+};
+static __u8 mi1320_initVGA_data[][4] = {
+	{0xb3, 0x01, 0x01, 0xcc},	{0x00, 0x00, 0x33, 0xdd},
+	{0xb0, 0x03, 0x19, 0xcc},	{0x00, 0x00, 0x33, 0xdd},
+	{0xb0, 0x04, 0x02, 0xcc},	{0x00, 0x00, 0x33, 0xdd},
+	{0xb3, 0x00, 0x64, 0xcc},	{0xb3, 0x00, 0x65, 0xcc},
+	{0xb0, 0x16, 0x03, 0xcc},	{0xb3, 0x05, 0x00, 0xcc},
+	{0xb3, 0x06, 0x00, 0xcc},	{0xb3, 0x08, 0x01, 0xcc},
+	{0xb3, 0x09, 0x0c, 0xcc},	{0xb3, 0x34, 0x02, 0xcc},
+	{0xb3, 0x35, 0xc8, 0xcc},	{0xb3, 0x02, 0x00, 0xcc},
+	{0xb3, 0x03, 0x0a, 0xcc},	{0xb3, 0x04, 0x05, 0xcc},
+	{0xb3, 0x20, 0x00, 0xcc},	{0xb3, 0x21, 0x00, 0xcc},
+	{0xb3, 0x22, 0x03, 0xcc},	{0xb3, 0x23, 0xc0, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x04, 0xcc},	{0xb3, 0x17, 0xff, 0xcc},
+	{0xb3, 0x00, 0x67, 0xcc},	{0xbc, 0x00, 0xd0, 0xcc},
+	{0xbc, 0x01, 0x01, 0xcc},	{0xf0, 0x00, 0x00, 0xbb},
+	{0x0d, 0x00, 0x09, 0xbb},	{0x00, 0x01, 0x00, 0xdd},
+	{0x0d, 0x00, 0x08, 0xbb},	{0xf0, 0x00, 0x01, 0xbb},
+	{0xa1, 0x05, 0x00, 0xbb},	{0xa4, 0x03, 0xc0, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x00, 0x00, 0x10, 0xdd},
+	{0xc8, 0x9f, 0x0b, 0xbb},	{0x00, 0x00, 0x10, 0xdd},
+	{0xf0, 0x00, 0x00, 0xbb},	{0x00, 0x00, 0x10, 0xdd},
+	{0x20, 0x01, 0x00, 0xbb},	{0x00, 0x00, 0x10, 0xdd},
+	{0xf0, 0x00, 0x01, 0xbb},	{0x9d, 0x3c, 0xa0, 0xbb},
+	{0x47, 0x30, 0x30, 0xbb},	{0xf0, 0x00, 0x00, 0xbb},
+	{0x0a, 0x80, 0x11, 0xbb},	{0x35, 0x00, 0x22, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x9d, 0xc5, 0x05, 0xbb},
+	{0xdc, 0x0f, 0xfc, 0xbb},	{0xf0, 0x00, 0x01, 0xbb},
+	{0x06, 0x74, 0x0e, 0xbb},	{0x80, 0x00, 0x06, 0xbb},
+	{0x81, 0x04, 0x00, 0xbb},	{0x82, 0x01, 0x02, 0xbb},
+	{0x83, 0x03, 0x02, 0xbb},	{0x84, 0x05, 0x00, 0xbb},
+	{0x85, 0x01, 0x00, 0xbb},	{0x86, 0x03, 0x02, 0xbb},
+	{0x87, 0x05, 0x00, 0xbb},	{0x88, 0x01, 0x00, 0xbb},
+	{0x89, 0x02, 0x02, 0xbb},	{0x8a, 0xfd, 0x04, 0xbb},
+	{0x8b, 0xfc, 0xfd, 0xbb},	{0x8c, 0xff, 0xfd, 0xbb},
+	{0x8d, 0x00, 0x00, 0xbb},	{0x8e, 0xfe, 0x05, 0xbb},
+	{0x8f, 0xfc, 0xfd, 0xbb},	{0x90, 0xfe, 0xfd, 0xbb},
+	{0x91, 0x00, 0x00, 0xbb},	{0x92, 0xfe, 0x03, 0xbb},
+	{0x93, 0xfd, 0xfe, 0xbb},	{0x94, 0xff, 0xfd, 0xbb},
+	{0x95, 0x00, 0x00, 0xbb},	{0xb6, 0x07, 0x05, 0xbb},
+	{0xb7, 0x13, 0x06, 0xbb},	{0xb8, 0x08, 0x06, 0xbb},
+	{0xb9, 0x14, 0x08, 0xbb},	{0xba, 0x06, 0x05, 0xbb},
+	{0xbb, 0x13, 0x06, 0xbb},	{0xbc, 0x03, 0x01, 0xbb},
+	{0xbd, 0x03, 0x04, 0xbb},	{0xbe, 0x00, 0x02, 0xbb},
+	{0xbf, 0x03, 0x01, 0xbb},	{0xc0, 0x02, 0x04, 0xbb},
+	{0xc1, 0x00, 0x04, 0xbb},	{0xc2, 0x02, 0x01, 0xbb},
+	{0xc3, 0x01, 0x03, 0xbb},	{0xc4, 0x00, 0x04, 0xbb},
+	{0xf0, 0x00, 0x00, 0xbb},	{0x05, 0x01, 0x13, 0xbb},
+	{0x06, 0x00, 0x11, 0xbb},	{0x07, 0x00, 0x85, 0xbb},
+	{0x08, 0x00, 0x27, 0xbb},	{0x20, 0x01, 0x03, 0xbb},
+	{0x21, 0x80, 0x00, 0xbb},	{0x22, 0x0d, 0x0f, 0xbb},
+	{0x24, 0x80, 0x00, 0xbb},	{0x59, 0x00, 0xff, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x39, 0x03, 0x0d, 0xbb},
+	{0x3a, 0x06, 0x1b, 0xbb},	{0x3b, 0x00, 0x95, 0xbb},
+	{0x3c, 0x04, 0xdb, 0xbb},	{0x57, 0x02, 0x00, 0xbb},
+	{0x58, 0x02, 0x66, 0xbb},	{0x59, 0x00, 0xff, 0xbb},
+	{0x5a, 0x01, 0x33, 0xbb},	{0x5c, 0x12, 0x0d, 0xbb},
+	{0x5d, 0x16, 0x11, 0xbb},	{0x64, 0x5e, 0x1c, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x2f, 0xd1, 0x00, 0xbb},
+	{0x5b, 0x00, 0x01, 0xbb},	{0xf0, 0x00, 0x02, 0xbb},
+	{0x36, 0x68, 0x10, 0xbb},	{0x00, 0x00, 0x30, 0xdd},
+	{0x37, 0x82, 0x00, 0xbb},	{0xbc, 0x0e, 0x00, 0xcc},
+	{0xbc, 0x0f, 0x05, 0xcc},	{0xbc, 0x10, 0xc0, 0xcc},
+	{0xbc, 0x11, 0x03, 0xcc},	{0xb6, 0x00, 0x00, 0xcc},
+	{0xb6, 0x03, 0x05, 0xcc},	{0xb6, 0x02, 0x00, 0xcc},
+	{0xb6, 0x05, 0x04, 0xcc},	{0xb6, 0x04, 0x00, 0xcc},
+	{0xb6, 0x12, 0xf8, 0xcc},	{0xb6, 0x13, 0x29, 0xcc},
+	{0xb6, 0x18, 0x0a, 0xcc},	{0xb6, 0x17, 0x00, 0xcc},
+	{0xb6, 0x16, 0x00, 0xcc},	{0xb6, 0x22, 0x12, 0xcc},
+	{0xb6, 0x23, 0x0b, 0xcc},	{0xbf, 0xc0, 0x26, 0xcc},
+	{0xbf, 0xc1, 0x02, 0xcc},	{0xbf, 0xcc, 0x04, 0xcc},
+	{0xbc, 0x02, 0x18, 0xcc},	{0xbc, 0x03, 0x50, 0xcc},
+	{0xbc, 0x04, 0x18, 0xcc},	{0xbc, 0x05, 0x00, 0xcc},
+	{0xbc, 0x06, 0x00, 0xcc},	{0xbc, 0x08, 0x30, 0xcc},
+	{0xbc, 0x09, 0x40, 0xcc},	{0xbc, 0x0a, 0x10, 0xcc},
+	{0xbc, 0x0b, 0x00, 0xcc},	{0xbc, 0x0c, 0x00, 0xcc},
+	{0xb3, 0x5c, 0x01, 0xcc},	{0xb3, 0x01, 0x41, 0xcc},
+	{}
+};
+static __u8 mi1320_initQVGA_data[][4] = {
+	{0xb3, 0x01, 0x01, 0xcc},	{0x00, 0x00, 0x33, 0xdd},
+	{0xb0, 0x03, 0x19, 0xcc},	{0x00, 0x00, 0x33, 0xdd},
+	{0xb0, 0x04, 0x02, 0xcc},	{0x00, 0x00, 0x33, 0xdd},
+	{0xb3, 0x00, 0x64, 0xcc},	{0xb3, 0x00, 0x65, 0xcc},
+	{0xb0, 0x16, 0x03, 0xcc},	{0xb3, 0x05, 0x01, 0xcc},
+	{0xb3, 0x06, 0x01, 0xcc},	{0xb3, 0x08, 0x01, 0xcc},
+	{0xb3, 0x09, 0x0c, 0xcc},	{0xb3, 0x34, 0x02, 0xcc},
+	{0xb3, 0x35, 0xc8, 0xcc},	{0xb3, 0x02, 0x00, 0xcc},
+	{0xb3, 0x03, 0x0a, 0xcc},	{0xb3, 0x04, 0x05, 0xcc},
+	{0xb3, 0x20, 0x00, 0xcc},	{0xb3, 0x21, 0x00, 0xcc},
+	{0xb3, 0x22, 0x01, 0xcc},	{0xb3, 0x23, 0xe0, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x02, 0xcc},	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x00, 0x65, 0xcc},	{0xb8, 0x00, 0x00, 0xcc},
+	{0xbc, 0x00, 0xd0, 0xcc},	{0xbc, 0x01, 0x01, 0xcc},
+	{0xf0, 0x00, 0x00, 0xbb},	{0x0d, 0x00, 0x09, 0xbb},
+	{0x00, 0x01, 0x00, 0xdd},	{0x0d, 0x00, 0x08, 0xbb},
+	{0xf0, 0x00, 0x00, 0xbb},	{0x02, 0x00, 0x64, 0xbb},
+	{0x05, 0x01, 0x78, 0xbb},	{0x06, 0x00, 0x11, 0xbb},
+	{0x07, 0x01, 0x42, 0xbb},	{0x08, 0x00, 0x11, 0xbb},
+	{0x20, 0x01, 0x00, 0xbb},	{0x21, 0x80, 0x00, 0xbb},
+	{0x22, 0x0d, 0x0f, 0xbb},	{0x24, 0x80, 0x00, 0xbb},
+	{0x59, 0x00, 0xff, 0xbb},	{0xf0, 0x00, 0x01, 0xbb},
+	{0x9d, 0x3c, 0xa0, 0xbb},	{0x47, 0x30, 0x30, 0xbb},
+	{0xf0, 0x00, 0x00, 0xbb},	{0x0a, 0x80, 0x11, 0xbb},
+	{0x35, 0x00, 0x22, 0xbb},	{0xf0, 0x00, 0x02, 0xbb},
+	{0x9d, 0xc5, 0x05, 0xbb},	{0xdc, 0x0f, 0xfc, 0xbb},
+	{0xf0, 0x00, 0x01, 0xbb},	{0x06, 0x74, 0x0e, 0xbb},
+	{0x80, 0x00, 0x06, 0xbb},	{0x81, 0x04, 0x00, 0xbb},
+	{0x82, 0x01, 0x02, 0xbb},	{0x83, 0x03, 0x02, 0xbb},
+	{0x84, 0x05, 0x00, 0xbb},	{0x85, 0x01, 0x00, 0xbb},
+	{0x86, 0x03, 0x02, 0xbb},	{0x87, 0x05, 0x00, 0xbb},
+	{0x88, 0x01, 0x00, 0xbb},	{0x89, 0x02, 0x02, 0xbb},
+	{0x8a, 0xfd, 0x04, 0xbb},	{0x8b, 0xfc, 0xfd, 0xbb},
+	{0x8c, 0xff, 0xfd, 0xbb},	{0x8d, 0x00, 0x00, 0xbb},
+	{0x8e, 0xfe, 0x05, 0xbb},	{0x8f, 0xfc, 0xfd, 0xbb},
+	{0x90, 0xfe, 0xfd, 0xbb},	{0x91, 0x00, 0x00, 0xbb},
+	{0x92, 0xfe, 0x03, 0xbb},	{0x93, 0xfd, 0xfe, 0xbb},
+	{0x94, 0xff, 0xfd, 0xbb},	{0x95, 0x00, 0x00, 0xbb},
+	{0xb6, 0x07, 0x05, 0xbb},	{0xb7, 0x13, 0x06, 0xbb},
+	{0xb8, 0x08, 0x06, 0xbb},	{0xb9, 0x14, 0x08, 0xbb},
+	{0xba, 0x06, 0x05, 0xbb},	{0xbb, 0x13, 0x06, 0xbb},
+	{0xbc, 0x03, 0x01, 0xbb},	{0xbd, 0x03, 0x04, 0xbb},
+	{0xbe, 0x00, 0x02, 0xbb},	{0xbf, 0x03, 0x01, 0xbb},
+	{0xc0, 0x02, 0x04, 0xbb},	{0xc1, 0x00, 0x04, 0xbb},
+	{0xc2, 0x02, 0x01, 0xbb},	{0xc3, 0x01, 0x03, 0xbb},
+	{0xc4, 0x00, 0x04, 0xbb},	{0xf0, 0x00, 0x02, 0xbb},
+	{0xc8, 0x00, 0x00, 0xbb},	{0x2e, 0x00, 0x00, 0xbb},
+	{0x2e, 0x0c, 0x5b, 0xbb},	{0x2f, 0xd1, 0x00, 0xbb},
+	{0x39, 0x03, 0xca, 0xbb},	{0x3a, 0x06, 0x80, 0xbb},
+	{0x3b, 0x01, 0x52, 0xbb},	{0x3c, 0x05, 0x40, 0xbb},
+	{0x57, 0x01, 0x9c, 0xbb},	{0x58, 0x01, 0xee, 0xbb},
+	{0x59, 0x00, 0xf0, 0xbb},	{0x5a, 0x01, 0x20, 0xbb},
+	{0x5c, 0x1d, 0x17, 0xbb},	{0x5d, 0x22, 0x1c, 0xbb},
+	{0x64, 0x1e, 0x1c, 0xbb},	{0x5b, 0x00, 0x01, 0xbb},
+	{0xf0, 0x00, 0x02, 0xbb},	{0x36, 0x68, 0x10, 0xbb},
+	{0x00, 0x00, 0x30, 0xdd},	{0x37, 0x81, 0x00, 0xbb},
+	{0xbc, 0x02, 0x18, 0xcc},	{0xbc, 0x03, 0x50, 0xcc},
+	{0xbc, 0x04, 0x18, 0xcc},	{0xbc, 0x05, 0x00, 0xcc},
+	{0xbc, 0x06, 0x00, 0xcc},	{0xbc, 0x08, 0x30, 0xcc},
+	{0xbc, 0x09, 0x40, 0xcc},	{0xbc, 0x0a, 0x10, 0xcc},
+	{0xbc, 0x0b, 0x00, 0xcc},	{0xbc, 0x0c, 0x00, 0xcc},
+	{0xbf, 0xc0, 0x26, 0xcc},	{0xbf, 0xc1, 0x02, 0xcc},
+	{0xbf, 0xcc, 0x04, 0xcc},	{0xb3, 0x5c, 0x01, 0xcc},
+	{0xb3, 0x01, 0x41, 0xcc},
+	{}
+};
+
+static __u8 po3130_gamma[17] = {
+	0x00, 0x13, 0x38, 0x59, 0x79, 0x92, 0xa7, 0xb9, 0xc8,
+	0xd4, 0xdf, 0xe7, 0xee, 0xf4, 0xf9, 0xfc, 0xff
+};
+static __u8 po3130_matrix[9] = {
+	0x5f, 0xec, 0xf5, 0xf1, 0x5a, 0xf5, 0xf1, 0xec, 0x63
+};
+
+static __u8 po3130_initVGA_data[][4] = {
+	{0xb0, 0x4d, 0x00, 0xcc},	{0xb3, 0x01, 0x01, 0xcc},
+	{0x00, 0x00, 0x50, 0xdd},	{0xb0, 0x03, 0x01, 0xcc},
+	{0xb3, 0x00, 0x04, 0xcc},	{0xb3, 0x00, 0x24, 0xcc},
+	{0xb3, 0x00, 0x25, 0xcc},	{0xb3, 0x08, 0x01, 0xcc},
+	{0xb3, 0x09, 0x0c, 0xcc},	{0xb3, 0x05, 0x00, 0xcc},
+	{0xb3, 0x06, 0x01, 0xcc},	{0xb3, 0x03, 0x1a, 0xcc},
+	{0xb3, 0x04, 0x15, 0xcc},	{0xb3, 0x20, 0x00, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},	{0xb3, 0x22, 0x01, 0xcc},
+	{0xb3, 0x23, 0xe8, 0xcc},	{0xb8, 0x08, 0xe8, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x02, 0xcc},	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x34, 0x01, 0xcc},	{0xb3, 0x35, 0xf6, 0xcc},
+	{0xb3, 0x00, 0x27, 0xcc},	{0xbc, 0x00, 0x71, 0xcc},
+	{0xb8, 0x00, 0x21, 0xcc},	{0xb8, 0x27, 0x20, 0xcc},
+	{0xb8, 0x01, 0x79, 0xcc},	{0xb8, 0x81, 0x09, 0xcc},
+	{0xb8, 0x2c, 0x50, 0xcc},	{0xb8, 0x2d, 0xf8, 0xcc},
+	{0xb8, 0x2e, 0xf8, 0xcc},	{0xb8, 0x2f, 0xf8, 0xcc},
+	{0xb8, 0x30, 0x50, 0xcc},	{0xb8, 0x31, 0xf8, 0xcc},
+	{0xb8, 0x32, 0xf8, 0xcc},	{0xb8, 0x33, 0xf8, 0xcc},
+	{0xb8, 0x34, 0x50, 0xcc},	{0xb8, 0x35, 0x00, 0xcc},
+	{0xb8, 0x36, 0x00, 0xcc},	{0xb8, 0x37, 0x00, 0xcc},
+	{0x00, 0x1e, 0xc6, 0xaa},	{0x00, 0x20, 0x44, 0xaa},
+	{0x00, 0xad, 0x02, 0xaa},	{0x00, 0xae, 0x2c, 0xaa},
+	{0x00, 0x12, 0x08, 0xaa},	{0x00, 0x17, 0x41, 0xaa},
+	{0x00, 0x19, 0x41, 0xaa},	{0x00, 0x1e, 0x06, 0xaa},
+	{0x00, 0x21, 0x00, 0xaa},	{0x00, 0x36, 0xc0, 0xaa},
+	{0x00, 0x37, 0xc8, 0xaa},	{0x00, 0x3b, 0x36, 0xaa},
+	{0x00, 0x4b, 0xfe, 0xaa},	{0x00, 0x51, 0x1c, 0xaa},
+	{0x00, 0x52, 0x01, 0xaa},	{0x00, 0x55, 0x0a, 0xaa},
+	{0x00, 0x59, 0x02, 0xaa},	{0x00, 0x5a, 0x04, 0xaa},
+	{0x00, 0x5c, 0x10, 0xaa},	{0x00, 0x5d, 0x10, 0xaa},
+	{0x00, 0x5e, 0x10, 0xaa},	{0x00, 0x5f, 0x10, 0xaa},
+	{0x00, 0x61, 0x00, 0xaa},	{0x00, 0x62, 0x18, 0xaa},
+	{0x00, 0x63, 0x30, 0xaa},	{0x00, 0x70, 0x68, 0xaa},
+	{0x00, 0x80, 0x71, 0xaa},	{0x00, 0x81, 0x08, 0xaa},
+	{0x00, 0x82, 0x00, 0xaa},	{0x00, 0x83, 0x55, 0xaa},
+	{0x00, 0x84, 0x06, 0xaa},	{0x00, 0x85, 0x06, 0xaa},
+	{0x00, 0x86, 0x13, 0xaa},	{0x00, 0x87, 0x18, 0xaa},
+	{0x00, 0xaa, 0x3f, 0xaa},	{0x00, 0xab, 0x44, 0xaa},
+	{0x00, 0xb0, 0x68, 0xaa},	{0x00, 0xb5, 0x10, 0xaa},
+	{0x00, 0xb8, 0x20, 0xaa},	{0x00, 0xb9, 0xa0, 0xaa},
+	{0x00, 0xbc, 0x04, 0xaa},	{0x00, 0x8b, 0x40, 0xaa},
+	{0x00, 0x8c, 0x91, 0xaa},	{0x00, 0x8d, 0x8f, 0xaa},
+	{0x00, 0x8e, 0x91, 0xaa},	{0x00, 0x8f, 0x43, 0xaa},
+	{0x00, 0x90, 0x92, 0xaa},	{0x00, 0x91, 0x89, 0xaa},
+	{0x00, 0x92, 0x9d, 0xaa},	{0x00, 0x93, 0x46, 0xaa},
+	{0x00, 0xd6, 0x22, 0xaa},	{0x00, 0x73, 0x00, 0xaa},
+	{0x00, 0x74, 0x10, 0xaa},	{0x00, 0x75, 0x20, 0xaa},
+	{0x00, 0x76, 0x2b, 0xaa},	{0x00, 0x77, 0x36, 0xaa},
+	{0x00, 0x78, 0x49, 0xaa},	{0x00, 0x79, 0x5a, 0xaa},
+	{0x00, 0x7a, 0x7f, 0xaa},	{0x00, 0x7b, 0x9b, 0xaa},
+	{0x00, 0x7c, 0xba, 0xaa},	{0x00, 0x7d, 0xd4, 0xaa},
+	{0x00, 0x7e, 0xea, 0xaa},	{0x00, 0xd6, 0x62, 0xaa},
+	{0x00, 0x73, 0x00, 0xaa},	{0x00, 0x74, 0x10, 0xaa},
+	{0x00, 0x75, 0x20, 0xaa},	{0x00, 0x76, 0x2b, 0xaa},
+	{0x00, 0x77, 0x36, 0xaa},	{0x00, 0x78, 0x49, 0xaa},
+	{0x00, 0x79, 0x5a, 0xaa},	{0x00, 0x7a, 0x7f, 0xaa},
+	{0x00, 0x7b, 0x9b, 0xaa},	{0x00, 0x7c, 0xba, 0xaa},
+	{0x00, 0x7d, 0xd4, 0xaa},	{0x00, 0x7e, 0xea, 0xaa},
+	{0x00, 0xd6, 0xa2, 0xaa},	{0x00, 0x73, 0x00, 0xaa},
+	{0x00, 0x74, 0x10, 0xaa},	{0x00, 0x75, 0x20, 0xaa},
+	{0x00, 0x76, 0x2b, 0xaa},	{0x00, 0x77, 0x36, 0xaa},
+	{0x00, 0x78, 0x49, 0xaa},	{0x00, 0x79, 0x5a, 0xaa},
+	{0x00, 0x7a, 0x7f, 0xaa},	{0x00, 0x7b, 0x9b, 0xaa},
+	{0x00, 0x7c, 0xba, 0xaa},	{0x00, 0x7d, 0xd4, 0xaa},
+	{0x00, 0x7e, 0xea, 0xaa},
+	{0x00, 0x4c, 0x07, 0xaa},
+	{0x00, 0x4b, 0xe0, 0xaa},	{0x00, 0x4e, 0x77, 0xaa},
+	{0x00, 0x59, 0x02, 0xaa},	{0x00, 0x4d, 0x0a, 0xaa},
+/*	{0x00, 0xd1, 0x00, 0xaa},	{0x00, 0x20, 0xc4, 0xaa},
+	{0xb8, 0x8e, 0x00, 0xcc},	{0xb8, 0x8f, 0xff, 0xcc}, */
+	{0x00, 0xd1, 0x3c, 0xaa},	{0x00, 0x20, 0xc4, 0xaa},
+	{0xb8, 0x8e, 0x00, 0xcc},	{0xb8, 0x8f, 0xff, 0xcc},
+	{0xb8, 0xfe, 0x00, 0xcc},	{0xb8, 0xff, 0x28, 0xcc},
+	{0xb9, 0x00, 0x28, 0xcc},	{0xb9, 0x01, 0x28, 0xcc},
+	{0xb9, 0x02, 0x28, 0xcc},	{0xb9, 0x03, 0x00, 0xcc},
+	{0xb9, 0x04, 0x00, 0xcc},	{0xb9, 0x05, 0x3c, 0xcc},
+	{0xb9, 0x06, 0x3c, 0xcc},	{0xb9, 0x07, 0x3c, 0xcc},
+	{0xb9, 0x08, 0x3c, 0xcc},	{0x00, 0x05, 0x00, 0xaa},
+	{0xb3, 0x5c, 0x00, 0xcc},	{0xb3, 0x01, 0x41, 0xcc},
+	{}
+};
+static __u8 po3130_rundata[][4] = {
+	{0x00, 0x47, 0x45, 0xaa},	{0x00, 0x48, 0x9b, 0xaa},
+	{0x00, 0x49, 0x3a, 0xaa},	{0x00, 0x4a, 0x01, 0xaa},
+	{0x00, 0x44, 0x40, 0xaa},
+/*	{0x00, 0xd5, 0x7c, 0xaa}, */
+	{0x00, 0xad, 0x04, 0xaa},	{0x00, 0xae, 0x00, 0xaa},
+	{0x00, 0xb0, 0x78, 0xaa},	{0x00, 0x98, 0x02, 0xaa},
+	{0x00, 0x94, 0x25, 0xaa},	{0x00, 0x95, 0x25, 0xaa},
+	{0x00, 0x59, 0x68, 0xaa},	{0x00, 0x44, 0x20, 0xaa},
+	{0x00, 0x17, 0x50, 0xaa},	{0x00, 0x19, 0x50, 0xaa},
+	{0x00, 0xd1, 0x3c, 0xaa},	{0x00, 0xd1, 0x3c, 0xaa},
+	{0x00, 0x1e, 0x06, 0xaa},	{0x00, 0x1e, 0x06, 0xaa},
+	{}
+};
+
+static __u8 po3130_initQVGA_data[][4] = {
+	{0xb0, 0x4d, 0x00, 0xcc},	{0xb3, 0x01, 0x01, 0xcc},
+	{0x00, 0x00, 0x50, 0xdd},	{0xb0, 0x03, 0x09, 0xcc},
+	{0xb3, 0x00, 0x04, 0xcc},	{0xb3, 0x00, 0x24, 0xcc},
+	{0xb3, 0x00, 0x25, 0xcc},	{0xb3, 0x08, 0x01, 0xcc},
+	{0xb3, 0x09, 0x0c, 0xcc},	{0xb3, 0x05, 0x00, 0xcc},
+	{0xb3, 0x06, 0x01, 0xcc},	{0xb3, 0x03, 0x1a, 0xcc},
+	{0xb3, 0x04, 0x15, 0xcc},	{0xb3, 0x20, 0x00, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},	{0xb3, 0x22, 0x01, 0xcc},
+	{0xb3, 0x23, 0xe0, 0xcc},	{0xb8, 0x08, 0xe0, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x02, 0xcc},	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x34, 0x01, 0xcc},	{0xb3, 0x35, 0xf6, 0xcc},
+	{0xb3, 0x00, 0x27, 0xcc},	{0xbc, 0x00, 0xd1, 0xcc},
+	{0xb8, 0x00, 0x21, 0xcc},	{0xb8, 0x27, 0x20, 0xcc},
+	{0xb8, 0x01, 0x79, 0xcc},	{0xb8, 0x81, 0x09, 0xcc},
+	{0xb8, 0x2c, 0x50, 0xcc},	{0xb8, 0x2d, 0xf8, 0xcc},
+	{0xb8, 0x2e, 0xf8, 0xcc},	{0xb8, 0x2f, 0xf8, 0xcc},
+	{0xb8, 0x30, 0x50, 0xcc},	{0xb8, 0x31, 0xf8, 0xcc},
+	{0xb8, 0x32, 0xf8, 0xcc},	{0xb8, 0x33, 0xf8, 0xcc},
+	{0xb8, 0x34, 0x50, 0xcc},	{0xb8, 0x35, 0x00, 0xcc},
+	{0xb8, 0x36, 0x00, 0xcc},	{0xb8, 0x37, 0x00, 0xcc},
+	{0x00, 0x1e, 0xc6, 0xaa},	{0x00, 0x20, 0x44, 0xaa},
+	{0x00, 0xad, 0x02, 0xaa},	{0x00, 0xae, 0x2c, 0xaa},
+	{0x00, 0x12, 0x08, 0xaa},	{0x00, 0x17, 0x41, 0xaa},
+	{0x00, 0x19, 0x41, 0xaa},	{0x00, 0x1e, 0x06, 0xaa},
+	{0x00, 0x21, 0x00, 0xaa},	{0x00, 0x36, 0xc0, 0xaa},
+	{0x00, 0x37, 0xc8, 0xaa},	{0x00, 0x3b, 0x36, 0xaa},
+	{0x00, 0x4b, 0xfe, 0xaa},	{0x00, 0x51, 0x1c, 0xaa},
+	{0x00, 0x52, 0x01, 0xaa},	{0x00, 0x55, 0x0a, 0xaa},
+	{0x00, 0x59, 0x6f, 0xaa},	{0x00, 0x5a, 0x04, 0xaa},
+	{0x00, 0x5c, 0x10, 0xaa},	{0x00, 0x5d, 0x10, 0xaa},
+	{0x00, 0x5e, 0x10, 0xaa},	{0x00, 0x5f, 0x10, 0xaa},
+	{0x00, 0x61, 0x00, 0xaa},	{0x00, 0x62, 0x18, 0xaa},
+	{0x00, 0x63, 0x30, 0xaa},	{0x00, 0x70, 0x68, 0xaa},
+	{0x00, 0x80, 0x71, 0xaa},	{0x00, 0x81, 0x08, 0xaa},
+	{0x00, 0x82, 0x00, 0xaa},	{0x00, 0x83, 0x55, 0xaa},
+	{0x00, 0x84, 0x06, 0xaa},	{0x00, 0x85, 0x06, 0xaa},
+	{0x00, 0x86, 0x13, 0xaa},	{0x00, 0x87, 0x18, 0xaa},
+	{0x00, 0xaa, 0x3f, 0xaa},	{0x00, 0xab, 0x44, 0xaa},
+	{0x00, 0xb0, 0x68, 0xaa},	{0x00, 0xb5, 0x10, 0xaa},
+	{0x00, 0xb8, 0x20, 0xaa},	{0x00, 0xb9, 0xa0, 0xaa},
+	{0x00, 0xbc, 0x04, 0xaa},	{0x00, 0x8b, 0x40, 0xaa},
+	{0x00, 0x8c, 0x91, 0xaa},	{0x00, 0x8d, 0x8f, 0xaa},
+	{0x00, 0x8e, 0x91, 0xaa},	{0x00, 0x8f, 0x43, 0xaa},
+	{0x00, 0x90, 0x92, 0xaa},	{0x00, 0x91, 0x89, 0xaa},
+	{0x00, 0x92, 0x9d, 0xaa},	{0x00, 0x93, 0x46, 0xaa},
+	{0x00, 0xd6, 0x22, 0xaa},	{0x00, 0x73, 0x00, 0xaa},
+	{0x00, 0x74, 0x10, 0xaa},	{0x00, 0x75, 0x20, 0xaa},
+	{0x00, 0x76, 0x2b, 0xaa},	{0x00, 0x77, 0x36, 0xaa},
+	{0x00, 0x78, 0x49, 0xaa},	{0x00, 0x79, 0x5a, 0xaa},
+	{0x00, 0x7a, 0x7f, 0xaa},	{0x00, 0x7b, 0x9b, 0xaa},
+	{0x00, 0x7c, 0xba, 0xaa},	{0x00, 0x7d, 0xd4, 0xaa},
+	{0x00, 0x7e, 0xea, 0xaa},	{0x00, 0xd6, 0x62, 0xaa},
+	{0x00, 0x73, 0x00, 0xaa},	{0x00, 0x74, 0x10, 0xaa},
+	{0x00, 0x75, 0x20, 0xaa},	{0x00, 0x76, 0x2b, 0xaa},
+	{0x00, 0x77, 0x36, 0xaa},	{0x00, 0x78, 0x49, 0xaa},
+	{0x00, 0x79, 0x5a, 0xaa},	{0x00, 0x7a, 0x7f, 0xaa},
+	{0x00, 0x7b, 0x9b, 0xaa},	{0x00, 0x7c, 0xba, 0xaa},
+	{0x00, 0x7d, 0xd4, 0xaa},	{0x00, 0x7e, 0xea, 0xaa},
+	{0x00, 0xd6, 0xa2, 0xaa},	{0x00, 0x73, 0x00, 0xaa},
+	{0x00, 0x74, 0x10, 0xaa},	{0x00, 0x75, 0x20, 0xaa},
+	{0x00, 0x76, 0x2b, 0xaa},	{0x00, 0x77, 0x36, 0xaa},
+	{0x00, 0x78, 0x49, 0xaa},	{0x00, 0x79, 0x5a, 0xaa},
+	{0x00, 0x7a, 0x7f, 0xaa},	{0x00, 0x7b, 0x9b, 0xaa},
+	{0x00, 0x7c, 0xba, 0xaa},	{0x00, 0x7d, 0xd4, 0xaa},
+	{0x00, 0x7e, 0xea, 0xaa},	{0x00, 0x4c, 0x07, 0xaa},
+	{0x00, 0x4b, 0xe0, 0xaa},	{0x00, 0x4e, 0x77, 0xaa},
+	{0x00, 0x59, 0x66, 0xaa},	{0x00, 0x4d, 0x0a, 0xaa},
+	{0x00, 0xd1, 0x00, 0xaa},	{0x00, 0x20, 0xc4, 0xaa},
+	{0xb8, 0x8e, 0x00, 0xcc},	{0xb8, 0x8f, 0xff, 0xcc},
+	{0xb8, 0xfe, 0x00, 0xcc},	{0xb8, 0xff, 0x28, 0xcc},
+	{0xb9, 0x00, 0x28, 0xcc},	{0xb9, 0x01, 0x28, 0xcc},
+	{0xb9, 0x02, 0x28, 0xcc},	{0xb9, 0x03, 0x00, 0xcc},
+	{0xb9, 0x04, 0x00, 0xcc},	{0xb9, 0x05, 0x3c, 0xcc},
+	{0xb9, 0x06, 0x3c, 0xcc},	{0xb9, 0x07, 0x3c, 0xcc},
+	{0xb9, 0x08, 0x3c, 0xcc},	{0xbc, 0x02, 0x18, 0xcc},
+	{0xbc, 0x03, 0x50, 0xcc},	{0xbc, 0x04, 0x18, 0xcc},
+	{0xbc, 0x05, 0x00, 0xcc},	{0xbc, 0x06, 0x00, 0xcc},
+	{0xbc, 0x08, 0x30, 0xcc},	{0xbc, 0x09, 0x40, 0xcc},
+	{0xbc, 0x0a, 0x10, 0xcc},	{0xbc, 0x0b, 0x00, 0xcc},
+	{0xbc, 0x0c, 0x00, 0xcc},	{0x00, 0x05, 0x00, 0xaa},
+	{0xb3, 0x5c, 0x00, 0xcc},	{0xb3, 0x01, 0x41, 0xcc},
+	{}
+};
+
+static __u8 hv7131r_gamma[17] = {
+/*	0x00, 0x13, 0x38, 0x59, 0x79, 0x92, 0xa7, 0xb9, 0xc8,
+ *	0xd4, 0xdf, 0xe7, 0xee, 0xf4, 0xf9, 0xfc, 0xff */
+	0x04, 0x1a, 0x36, 0x55, 0x6f, 0x87, 0x9d, 0xb0, 0xc1,
+	0xcf, 0xda, 0xe4, 0xec, 0xf3, 0xf8, 0xfd, 0xff
+};
+static __u8 hv7131r_matrix[9] = {
+	0x5f, 0xec, 0xf5, 0xf1, 0x5a, 0xf5, 0xf1, 0xec, 0x63
+};
+static __u8 hv7131r_initVGA_data[][4] = {
+	{0xb0, 0x4d, 0x00, 0xcc},	{0xb3, 0x01, 0x01, 0xcc},
+	{0x00, 0x00, 0x50, 0xdd},	{0xb0, 0x03, 0x01, 0xcc},
+	{0xb3, 0x00, 0x24, 0xcc},
+	{0xb3, 0x00, 0x25, 0xcc},	{0xb3, 0x08, 0x01, 0xcc},
+	{0xb3, 0x09, 0x0c, 0xcc},	{0xb3, 0x05, 0x00, 0xcc},
+	{0xb3, 0x06, 0x01, 0xcc},
+	{0xb3, 0x01, 0x45, 0xcc},	{0xb3, 0x03, 0x0b, 0xcc},
+	{0xb3, 0x04, 0x05, 0xcc},	{0xb3, 0x20, 0x00, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},
+	{0xb3, 0x22, 0x01, 0xcc},	{0xb3, 0x23, 0xe0, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x02, 0xcc},
+	{0xb3, 0x17, 0x7f, 0xcc},	{0xb3, 0x34, 0x01, 0xcc},
+	{0xb3, 0x35, 0x91, 0xcc},	{0xb3, 0x00, 0x27, 0xcc},
+	{0xbc, 0x00, 0x73, 0xcc},
+	{0xb8, 0x00, 0x23, 0xcc},	{0x00, 0x01, 0x0c, 0xaa},
+	{0x00, 0x14, 0x01, 0xaa},	{0x00, 0x15, 0xe6, 0xaa},
+	{0x00, 0x16, 0x02, 0xaa},
+	{0x00, 0x17, 0x86, 0xaa},	{0x00, 0x23, 0x00, 0xaa},
+	{0x00, 0x25, 0x09, 0xaa},	{0x00, 0x26, 0x27, 0xaa},
+	{0x00, 0x27, 0xc0, 0xaa},
+	{0xb8, 0x2c, 0x60, 0xcc},	{0xb8, 0x2d, 0xf8, 0xcc},
+	{0xb8, 0x2e, 0xf8, 0xcc},	{0xb8, 0x2f, 0xf8, 0xcc},
+	{0xb8, 0x30, 0x50, 0xcc},
+	{0xb8, 0x31, 0xf8, 0xcc},	{0xb8, 0x32, 0xf8, 0xcc},
+	{0xb8, 0x33, 0xf8, 0xcc},	{0xb8, 0x34, 0x65, 0xcc},
+	{0xb8, 0x35, 0x00, 0xcc},
+	{0xb8, 0x36, 0x00, 0xcc},	{0xb8, 0x37, 0x00, 0xcc},
+	{0xb8, 0x27, 0x20, 0xcc},	{0xb8, 0x01, 0x7d, 0xcc},
+	{0xb8, 0x81, 0x09, 0xcc},
+	{0xb3, 0x01, 0x41, 0xcc},	{0xb8, 0xfe, 0x00, 0xcc},
+	{0xb8, 0xff, 0x28, 0xcc},	{0xb9, 0x00, 0x28, 0xcc},
+	{0xb9, 0x01, 0x28, 0xcc},
+	{0xb9, 0x02, 0x28, 0xcc},	{0xb9, 0x03, 0x00, 0xcc},
+	{0xb9, 0x04, 0x00, 0xcc},	{0xb9, 0x05, 0x3c, 0xcc},
+	{0xb9, 0x06, 0x3c, 0xcc},
+	{0xb9, 0x07, 0x3c, 0xcc},	{0xb9, 0x08, 0x3c, 0xcc},
+	{0xb8, 0x8e, 0x00, 0xcc},	{0xb8, 0x8f, 0xff, 0xcc},
+	{0x00, 0x30, 0x18, 0xaa},
+	{}
+};
+
+static __u8 hv7131r_initQVGA_data[][4] = {
+	{0xb0, 0x4d, 0x00, 0xcc},	{0xb3, 0x01, 0x01, 0xcc},
+	{0x00, 0x00, 0x50, 0xdd},	{0xb0, 0x03, 0x01, 0xcc},
+	{0xb3, 0x00, 0x24, 0xcc},
+	{0xb3, 0x00, 0x25, 0xcc},	{0xb3, 0x08, 0x01, 0xcc},
+	{0xb3, 0x09, 0x0c, 0xcc},	{0xb3, 0x05, 0x00, 0xcc},
+	{0xb3, 0x06, 0x01, 0xcc},
+	{0xb3, 0x03, 0x0b, 0xcc},	{0xb3, 0x04, 0x05, 0xcc},
+	{0xb3, 0x20, 0x00, 0xcc},	{0xb3, 0x21, 0x00, 0xcc},
+	{0xb3, 0x22, 0x01, 0xcc},
+	{0xb3, 0x23, 0xe0, 0xcc},	{0xb3, 0x14, 0x00, 0xcc},
+	{0xb3, 0x15, 0x00, 0xcc},	{0xb3, 0x16, 0x02, 0xcc},
+	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x34, 0x01, 0xcc},	{0xb3, 0x35, 0x91, 0xcc},
+	{0xb3, 0x00, 0x27, 0xcc},	{0xbc, 0x00, 0xd1, 0xcc},
+	{0xb8, 0x00, 0x21, 0xcc},
+	{0x00, 0x01, 0x0c, 0xaa},	{0x00, 0x14, 0x01, 0xaa},
+	{0x00, 0x15, 0xe6, 0xaa},	{0x00, 0x16, 0x02, 0xaa},
+	{0x00, 0x17, 0x86, 0xaa},
+	{0x00, 0x23, 0x00, 0xaa},	{0x00, 0x25, 0x01, 0xaa},
+	{0x00, 0x26, 0xd4, 0xaa},	{0x00, 0x27, 0xc0, 0xaa},
+	{0xbc, 0x02, 0x08, 0xcc},
+	{0xbc, 0x03, 0x70, 0xcc},	{0xbc, 0x04, 0x08, 0xcc},
+	{0xbc, 0x05, 0x00, 0xcc},	{0xbc, 0x06, 0x00, 0xcc},
+	{0xbc, 0x08, 0x3c, 0xcc},
+	{0xbc, 0x09, 0x40, 0xcc},	{0xbc, 0x0a, 0x04, 0xcc},
+	{0xbc, 0x0b, 0x00, 0xcc},	{0xbc, 0x0c, 0x00, 0xcc},
+	{0xb8, 0xfe, 0x02, 0xcc},
+	{0xb8, 0xff, 0x07, 0xcc},	{0xb9, 0x00, 0x14, 0xcc},
+	{0xb9, 0x01, 0x14, 0xcc},	{0xb9, 0x02, 0x14, 0xcc},
+	{0xb9, 0x03, 0x00, 0xcc},
+	{0xb9, 0x04, 0x02, 0xcc},	{0xb9, 0x05, 0x05, 0xcc},
+	{0xb9, 0x06, 0x0f, 0xcc},	{0xb9, 0x07, 0x0f, 0xcc},
+	{0xb9, 0x08, 0x0f, 0xcc},
+	{0xb8, 0x2c, 0x60, 0xcc},	{0xb8, 0x2d, 0xf8, 0xcc},
+	{0xb8, 0x2e, 0xf8, 0xcc},	{0xb8, 0x2f, 0xf8, 0xcc},
+	{0xb8, 0x30, 0x50, 0xcc},
+	{0xb8, 0x31, 0xf8, 0xcc},	{0xb8, 0x32, 0xf8, 0xcc},
+	{0xb8, 0x33, 0xf8, 0xcc},
+	{0xb8, 0x34, 0x65, 0xcc},	{0xb8, 0x35, 0x00, 0xcc},
+	{0xb8, 0x36, 0x00, 0xcc},	{0xb8, 0x37, 0x00, 0xcc},
+	{0xb8, 0x27, 0x20, 0xcc},
+	{0xb8, 0x01, 0x7d, 0xcc},	{0xb8, 0x81, 0x09, 0xcc},
+	{0xb3, 0x01, 0x41, 0xcc},	{0xb8, 0xfe, 0x00, 0xcc},
+	{0xb8, 0xff, 0x28, 0xcc},
+	{0xb9, 0x00, 0x28, 0xcc},	{0xb9, 0x01, 0x28, 0xcc},
+	{0xb9, 0x02, 0x28, 0xcc},	{0xb9, 0x03, 0x00, 0xcc},
+	{0xb9, 0x04, 0x00, 0xcc},
+	{0xb9, 0x05, 0x3c, 0xcc},	{0xb9, 0x06, 0x3c, 0xcc},
+	{0xb9, 0x07, 0x3c, 0xcc},	{0xb9, 0x08, 0x3c, 0xcc},
+	{0xb8, 0x8e, 0x00, 0xcc},
+	{0xb8, 0x8f, 0xff, 0xcc},	{0x00, 0x30, 0x18, 0xaa},
+	{}
+};
+
+static __u8 ov7660_gamma[17] = {
+	0x00, 0x13, 0x38, 0x59, 0x79, 0x92, 0xa7, 0xb9, 0xc8,
+	0xd4, 0xdf, 0xe7, 0xee, 0xf4, 0xf9, 0xfc, 0xff
+};
+static __u8 ov7660_matrix[9] = {
+	0x5a, 0xf0, 0xf6, 0xf3, 0x57, 0xf6, 0xf3, 0xef, 0x62
+};
+static __u8 ov7660_initVGA_data[][4] = {
+	{0xb0, 0x4d, 0x00, 0xcc},	{0xb3, 0x01, 0x01, 0xcc},
+	{0x00, 0x00, 0x50, 0xdd},
+	{0xb0, 0x03, 0x01, 0xcc},
+	{0xb3, 0x00, 0x21, 0xcc},	{0xb3, 0x00, 0x26, 0xcc},
+	{0xb3, 0x05, 0x01, 0xcc},
+	{0xb3, 0x06, 0x03, 0xcc},
+	{0xb3, 0x03, 0x1f, 0xcc},	{0xb3, 0x04, 0x05, 0xcc},
+	{0xb3, 0x05, 0x00, 0xcc},
+	{0xb3, 0x06, 0x01, 0xcc},
+	{0xb3, 0x15, 0x00, 0xcc},/* 0xb315  <-0 href startl */
+	{0xb3, 0x16, 0x02, 0xcc},	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},
+	{0xb3, 0x23, 0xe0, 0xcc},	{0xb3, 0x1d, 0x01, 0xcc},
+	{0xb3, 0x1f, 0x02, 0xcc},
+	{0xb3, 0x34, 0x01, 0xcc},
+	{0xb3, 0x35, 0xa1, 0xcc},	{0xb3, 0x00, 0x26, 0xcc},
+	{0xb8, 0x00, 0x33, 0xcc}, /* 13 */
+	{0xb8, 0x01, 0x7d, 0xcc},
+	{0xbc, 0x00, 0x73, 0xcc},	{0xb8, 0x81, 0x09, 0xcc},
+	{0xb8, 0x27, 0x20, 0xcc},
+	{0xb8, 0x8f, 0x50, 0xcc},
+	{0x00, 0x01, 0x80, 0xaa},	{0x00, 0x02, 0x80, 0xaa},
+	{0x00, 0x12, 0x80, 0xaa},
+	{0x00, 0x12, 0x05, 0xaa},
+	{0x00, 0x1e, 0x01, 0xaa},
+	{0x00, 0x3d, 0x40, 0xaa}, /* 0x3d <-40 gamma 01 */
+	{0x00, 0x41, 0x00, 0xaa}, /* edge 00 */
+	{0x00, 0x0d, 0x48, 0xaa},	{0x00, 0x0e, 0x04, 0xaa},
+	{0x00, 0x13, 0xa7, 0xaa},
+	{0x00, 0x40, 0xc1, 0xaa},	{0x00, 0x35, 0x00, 0xaa},
+	{0x00, 0x36, 0x00, 0xaa},
+	{0x00, 0x3c, 0x68, 0xaa},	{0x00, 0x1b, 0x05, 0xaa},
+	{0x00, 0x39, 0x43, 0xaa},
+	{0x00, 0x8d, 0xcf, 0xaa},
+	{0x00, 0x8b, 0xcc, 0xaa},	{0x00, 0x8c, 0xcc, 0xaa},
+	{0x00, 0x0f, 0x62, 0xaa},
+	{0x00, 0x35, 0x84, 0xaa},
+	{0x00, 0x3b, 0x08, 0xaa}, /* 0 * Nightframe 1/4 + 50Hz -> 0xC8 */
+	{0x00, 0x3a, 0x00, 0xaa}, /* mx change yuyv format 00, 04, 01; 08, 0c*/
+	{0x00, 0x14, 0x2a, 0xaa}, /* agc ampli */
+	{0x00, 0x9e, 0x40, 0xaa},	{0xb8, 0x8f, 0x50, 0xcc},
+	{0x00, 0x01, 0x80, 0xaa},
+	{0x00, 0x02, 0x80, 0xaa},
+	{0xb8, 0xfe, 0x00, 0xcc},	{0xb8, 0xff, 0x28, 0xcc},
+	{0xb9, 0x00, 0x28, 0xcc},
+	{0xb9, 0x01, 0x28, 0xcc},	{0xb9, 0x02, 0x28, 0xcc},
+	{0xb9, 0x03, 0x00, 0xcc},
+	{0xb9, 0x04, 0x00, 0xcc},
+	{0xb9, 0x05, 0x3c, 0xcc},	{0xb9, 0x06, 0x3c, 0xcc},
+	{0xb9, 0x07, 0x3c, 0xcc},
+	{0xb9, 0x08, 0x3c, 0xcc},
+
+	{0xb8, 0x8e, 0x00, 0xcc},	{0xb8, 0x8f, 0xff, 0xcc},
+
+	{0x00, 0x29, 0x3c, 0xaa},	{0xb3, 0x01, 0x45, 0xcc},
+	{}
+};
+static __u8 ov7660_initQVGA_data[][4] = {
+	{0xb0, 0x4d, 0x00, 0xcc},	{0xb3, 0x01, 0x01, 0xcc},
+	{0x00, 0x00, 0x50, 0xdd},	{0xb0, 0x03, 0x01, 0xcc},
+	{0xb3, 0x00, 0x21, 0xcc},	{0xb3, 0x00, 0x26, 0xcc},
+	{0xb3, 0x05, 0x01, 0xcc},	{0xb3, 0x06, 0x03, 0xcc},
+	{0xb3, 0x03, 0x1f, 0xcc},	{0xb3, 0x04, 0x05, 0xcc},
+	{0xb3, 0x05, 0x00, 0xcc},	{0xb3, 0x06, 0x01, 0xcc},
+	{0xb3, 0x15, 0x00, 0xcc},/* 0xb315  <-0 href startl */
+	{0xb3, 0x16, 0x02, 0xcc},	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},
+	{0xb3, 0x23, 0xe0, 0xcc},	{0xb3, 0x1d, 0x01, 0xcc},
+	{0xb3, 0x1f, 0x02, 0xcc},	{0xb3, 0x34, 0x01, 0xcc},
+	{0xb3, 0x35, 0xa1, 0xcc},	{0xb3, 0x00, 0x26, 0xcc},
+	{0xb8, 0x00, 0x33, 0xcc}, /* 13 */
+	{0xb8, 0x01, 0x7d, 0xcc},
+/* sizer */
+	{0xbc, 0x00, 0xd3, 0xcc},
+	{0xb8, 0x81, 0x09, 0xcc},	{0xb8, 0x81, 0x09, 0xcc},
+	{0xb8, 0x27, 0x20, 0xcc},	{0xb8, 0x8f, 0x50, 0xcc},
+	{0x00, 0x01, 0x80, 0xaa},	{0x00, 0x02, 0x80, 0xaa},
+	{0x00, 0x12, 0x80, 0xaa},	{0x00, 0x12, 0x05, 0xaa},
+	{0x00, 0x1e, 0x01, 0xaa},
+	{0x00, 0x3d, 0x40, 0xaa}, /* 0x3d <-40 gamma 01 */
+	{0x00, 0x41, 0x00, 0xaa}, /* edge 00 */
+	{0x00, 0x0d, 0x48, 0xaa},	{0x00, 0x0e, 0x04, 0xaa},
+	{0x00, 0x13, 0xa7, 0xaa},
+	{0x00, 0x40, 0xc1, 0xaa},	{0x00, 0x35, 0x00, 0xaa},
+	{0x00, 0x36, 0x00, 0xaa},
+	{0x00, 0x3c, 0x68, 0xaa},	{0x00, 0x1b, 0x05, 0xaa},
+	{0x00, 0x39, 0x43, 0xaa},	{0x00, 0x8d, 0xcf, 0xaa},
+	{0x00, 0x8b, 0xcc, 0xaa},	{0x00, 0x8c, 0xcc, 0xaa},
+	{0x00, 0x0f, 0x62, 0xaa},	{0x00, 0x35, 0x84, 0xaa},
+	{0x00, 0x3b, 0x08, 0xaa}, /* 0  * Nightframe 1/4 + 50Hz -> 0xC8 */
+	{0x00, 0x3a, 0x00, 0xaa}, /* mx change yuyv format 00, 04, 01; 08, 0c*/
+	{0x00, 0x14, 0x2a, 0xaa}, /* agc ampli */
+	{0x00, 0x9e, 0x40, 0xaa},	{0xb8, 0x8f, 0x50, 0xcc},
+	{0x00, 0x01, 0x80, 0xaa},
+	{0x00, 0x02, 0x80, 0xaa},
+/* sizer filters */
+	{0xbc, 0x02, 0x08, 0xcc},
+	{0xbc, 0x03, 0x70, 0xcc},
+	{0xb8, 0x35, 0x00, 0xcc},
+	{0xb8, 0x36, 0x00, 0xcc},
+	{0xb8, 0x37, 0x00, 0xcc},
+	{0xbc, 0x04, 0x08, 0xcc},
+	{0xbc, 0x05, 0x00, 0xcc},
+	{0xbc, 0x06, 0x00, 0xcc},
+	{0xbc, 0x08, 0x3c, 0xcc},
+	{0xbc, 0x09, 0x40, 0xcc},
+	{0xbc, 0x0a, 0x04, 0xcc},
+	{0xbc, 0x0b, 0x00, 0xcc},
+	{0xbc, 0x0c, 0x00, 0xcc},
+/* */
+	{0xb8, 0xfe, 0x00, 0xcc},
+	{0xb8, 0xff, 0x28, 0xcc},
+/* */
+	{0xb9, 0x00, 0x28, 0xcc},	{0xb9, 0x01, 0x28, 0xcc},
+	{0xb9, 0x02, 0x28, 0xcc},	{0xb9, 0x03, 0x00, 0xcc},
+	{0xb9, 0x04, 0x00, 0xcc},	{0xb9, 0x05, 0x3c, 0xcc},
+	{0xb9, 0x06, 0x3c, 0xcc},	{0xb9, 0x07, 0x3c, 0xcc},
+	{0xb9, 0x08, 0x3c, 0xcc},
+/* */
+	{0xb8, 0x8e, 0x00, 0xcc},
+	{0xb8, 0x8f, 0xff, 0xcc}, /* ff */
+	{0x00, 0x29, 0x3c, 0xaa},
+	{0xb3, 0x01, 0x45, 0xcc}, /* 45 */
+	{0x00, 0x00, 0x00, 0x00}
+};
+
+static __u8 ov7660_50HZ[][4] = {
+	{0x00, 0x3b, 0x08, 0xaa},
+	{0x00, 0x9d, 0x40, 0xaa},
+	{0x00, 0x13, 0xa7, 0xaa},
+	{0x00, 0x00, 0x00, 0x00}
+};
+
+static __u8 ov7660_60HZ[][4] = {
+	{0x00, 0x3b, 0x00, 0xaa},
+	{0x00, 0x9e, 0x40, 0xaa},
+	{0x00, 0x13, 0xa7, 0xaa},
+	{}
+};
+
+static __u8 ov7660_NoFliker[][4] = {
+	{0x00, 0x13, 0x87, 0xaa},
+	{}
+};
+
+static __u8 ov7670_initVGA_JPG[][4] = {
+	{0xb3, 0x01, 0x05, 0xcc},
+	{0x00, 0x00, 0x30, 0xdd},	{0xb0, 0x03, 0x19, 0xcc},
+	{0x00, 0x00, 0x10, 0xdd},
+	{0xb0, 0x04, 0x02, 0xcc},	{0x00, 0x00, 0x10, 0xdd},
+	{0xb3, 0x00, 0x66, 0xcc},	{0xb3, 0x00, 0x67, 0xcc},
+	{0xb3, 0x35, 0xa1, 0xcc},	{0xb3, 0x34, 0x01, 0xcc},
+	{0xb3, 0x05, 0x01, 0xcc},	{0xb3, 0x06, 0x01, 0xcc},
+	{0xb3, 0x08, 0x01, 0xcc},	{0xb3, 0x09, 0x0c, 0xcc},
+	{0xb3, 0x02, 0x02, 0xcc},	{0xb3, 0x03, 0x1f, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x02, 0xcc},	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x04, 0x05, 0xcc},	{0xb3, 0x20, 0x00, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},	{0xb3, 0x22, 0x01, 0xcc},
+	{0xb3, 0x23, 0xe0, 0xcc},	{0xbc, 0x00, 0x41, 0xcc},
+	{0xbc, 0x01, 0x01, 0xcc},	{0x00, 0x12, 0x80, 0xaa},
+	{0x00, 0x00, 0x20, 0xdd},	{0x00, 0x12, 0x00, 0xaa},
+	{0x00, 0x11, 0x40, 0xaa},	{0x00, 0x6b, 0x0a, 0xaa},
+	{0x00, 0x3a, 0x04, 0xaa},	{0x00, 0x40, 0xc0, 0xaa},
+	{0x00, 0x8c, 0x00, 0xaa},	{0x00, 0x7a, 0x29, 0xaa},
+	{0x00, 0x7b, 0x0e, 0xaa},	{0x00, 0x7c, 0x1a, 0xaa},
+	{0x00, 0x7d, 0x31, 0xaa},	{0x00, 0x7e, 0x53, 0xaa},
+	{0x00, 0x7f, 0x60, 0xaa},	{0x00, 0x80, 0x6b, 0xaa},
+	{0x00, 0x81, 0x73, 0xaa},	{0x00, 0x82, 0x7b, 0xaa},
+	{0x00, 0x83, 0x82, 0xaa},	{0x00, 0x84, 0x89, 0xaa},
+	{0x00, 0x85, 0x96, 0xaa},	{0x00, 0x86, 0xa1, 0xaa},
+	{0x00, 0x87, 0xb7, 0xaa},	{0x00, 0x88, 0xcc, 0xaa},
+	{0x00, 0x89, 0xe1, 0xaa},	{0x00, 0x13, 0xe0, 0xaa},
+	{0x00, 0x00, 0x00, 0xaa},	{0x00, 0x10, 0x00, 0xaa},
+	{0x00, 0x0d, 0x40, 0xaa},	{0x00, 0x14, 0x28, 0xaa},
+	{0x00, 0xa5, 0x05, 0xaa},	{0x00, 0xab, 0x07, 0xaa},
+	{0x00, 0x24, 0x95, 0xaa},	{0x00, 0x25, 0x33, 0xaa},
+	{0x00, 0x26, 0xe3, 0xaa},	{0x00, 0x9f, 0x88, 0xaa},
+	{0x00, 0xa0, 0x78, 0xaa},	{0x00, 0x55, 0x90, 0xaa},
+	{0x00, 0xa1, 0x03, 0xaa},	{0x00, 0xa6, 0xe0, 0xaa},
+	{0x00, 0xa7, 0xd8, 0xaa},	{0x00, 0xa8, 0xf0, 0xaa},
+	{0x00, 0xa9, 0x90, 0xaa},	{0x00, 0xaa, 0x14, 0xaa},
+	{0x00, 0x13, 0xe5, 0xaa},	{0x00, 0x0e, 0x61, 0xaa},
+	{0x00, 0x0f, 0x4b, 0xaa},	{0x00, 0x16, 0x02, 0xaa},
+	{0x00, 0x1e, 0x07, 0xaa},	{0x00, 0x21, 0x02, 0xaa},
+	{0x00, 0x22, 0x91, 0xaa},	{0x00, 0x29, 0x07, 0xaa},
+	{0x00, 0x33, 0x0b, 0xaa},	{0x00, 0x35, 0x0b, 0xaa},
+	{0x00, 0x37, 0x1d, 0xaa},	{0x00, 0x38, 0x71, 0xaa},
+	{0x00, 0x39, 0x2a, 0xaa},	{0x00, 0x3c, 0x78, 0xaa},
+	{0x00, 0x4d, 0x40, 0xaa},	{0x00, 0x4e, 0x20, 0xaa},
+	{0x00, 0x74, 0x19, 0xaa},	{0x00, 0x8d, 0x4f, 0xaa},
+	{0x00, 0x8e, 0x00, 0xaa},	{0x00, 0x8f, 0x00, 0xaa},
+	{0x00, 0x90, 0x00, 0xaa},	{0x00, 0x91, 0x00, 0xaa},
+	{0x00, 0x96, 0x00, 0xaa},	{0x00, 0x9a, 0x80, 0xaa},
+	{0x00, 0xb0, 0x84, 0xaa},	{0x00, 0xb1, 0x0c, 0xaa},
+	{0x00, 0xb2, 0x0e, 0xaa},	{0x00, 0xb3, 0x82, 0xaa},
+	{0x00, 0xb8, 0x0a, 0xaa},	{0x00, 0x43, 0x14, 0xaa},
+	{0x00, 0x44, 0xf0, 0xaa},	{0x00, 0x45, 0x45, 0xaa},
+	{0x00, 0x46, 0x63, 0xaa},	{0x00, 0x47, 0x2d, 0xaa},
+	{0x00, 0x48, 0x46, 0xaa},	{0x00, 0x59, 0x88, 0xaa},
+	{0x00, 0x5a, 0xa0, 0xaa},	{0x00, 0x5b, 0xc6, 0xaa},
+	{0x00, 0x5c, 0x7d, 0xaa},	{0x00, 0x5d, 0x5f, 0xaa},
+	{0x00, 0x5e, 0x19, 0xaa},	{0x00, 0x6c, 0x0a, 0xaa},
+	{0x00, 0x6d, 0x55, 0xaa},	{0x00, 0x6e, 0x11, 0xaa},
+	{0x00, 0x6f, 0x9e, 0xaa},	{0x00, 0x69, 0x00, 0xaa},
+	{0x00, 0x6a, 0x40, 0xaa},	{0x00, 0x01, 0x40, 0xaa},
+	{0x00, 0x02, 0x40, 0xaa},	{0x00, 0x13, 0xe7, 0xaa},
+	{0x00, 0x5f, 0xf0, 0xaa},	{0x00, 0x60, 0xf0, 0xaa},
+	{0x00, 0x61, 0xf0, 0xaa},	{0x00, 0x27, 0xa0, 0xaa},
+	{0x00, 0x28, 0x80, 0xaa},	{0x00, 0x2c, 0x90, 0xaa},
+	{0x00, 0x4f, 0x66, 0xaa},	{0x00, 0x50, 0x66, 0xaa},
+	{0x00, 0x51, 0x00, 0xaa},	{0x00, 0x52, 0x22, 0xaa},
+	{0x00, 0x53, 0x5e, 0xaa},	{0x00, 0x54, 0x80, 0xaa},
+	{0x00, 0x58, 0x9e, 0xaa},	{0x00, 0x41, 0x08, 0xaa},
+	{0x00, 0x3f, 0x00, 0xaa},	{0x00, 0x75, 0x85, 0xaa},
+	{0x00, 0x76, 0xe1, 0xaa},	{0x00, 0x4c, 0x00, 0xaa},
+	{0x00, 0x77, 0x0a, 0xaa},	{0x00, 0x3d, 0x88, 0xaa},
+	{0x00, 0x4b, 0x09, 0xaa},	{0x00, 0xc9, 0x60, 0xaa},
+	{0x00, 0x41, 0x38, 0xaa},	{0x00, 0x62, 0x30, 0xaa},
+	{0x00, 0x63, 0x30, 0xaa},	{0x00, 0x64, 0x08, 0xaa},
+	{0x00, 0x94, 0x07, 0xaa},	{0x00, 0x95, 0x0b, 0xaa},
+	{0x00, 0x65, 0x00, 0xaa},	{0x00, 0x66, 0x05, 0xaa},
+	{0x00, 0x56, 0x50, 0xaa},	{0x00, 0x34, 0x11, 0xaa},
+	{0x00, 0xa4, 0x88, 0xaa},	{0x00, 0x96, 0x00, 0xaa},
+	{0x00, 0x97, 0x30, 0xaa},	{0x00, 0x98, 0x20, 0xaa},
+	{0x00, 0x99, 0x30, 0xaa},	{0x00, 0x9a, 0x84, 0xaa},
+	{0x00, 0x9b, 0x29, 0xaa},	{0x00, 0x9c, 0x03, 0xaa},
+	{0x00, 0x78, 0x04, 0xaa},	{0x00, 0x79, 0x01, 0xaa},
+	{0x00, 0xc8, 0xf0, 0xaa},	{0x00, 0x79, 0x0f, 0xaa},
+	{0x00, 0xc8, 0x00, 0xaa},	{0x00, 0x79, 0x10, 0xaa},
+	{0x00, 0xc8, 0x7e, 0xaa},	{0x00, 0x79, 0x0a, 0xaa},
+	{0x00, 0xc8, 0x80, 0xaa},	{0x00, 0x79, 0x0b, 0xaa},
+	{0x00, 0xc8, 0x01, 0xaa},	{0x00, 0x79, 0x0c, 0xaa},
+	{0x00, 0xc8, 0x0f, 0xaa},	{0x00, 0x79, 0x0d, 0xaa},
+	{0x00, 0xc8, 0x20, 0xaa},	{0x00, 0x79, 0x09, 0xaa},
+	{0x00, 0xc8, 0x80, 0xaa},	{0x00, 0x79, 0x02, 0xaa},
+	{0x00, 0xc8, 0xc0, 0xaa},	{0x00, 0x79, 0x03, 0xaa},
+	{0x00, 0xc8, 0x40, 0xaa},	{0x00, 0x79, 0x05, 0xaa},
+	{0x00, 0xc8, 0x30, 0xaa},	{0x00, 0x79, 0x26, 0xaa},
+	{0x00, 0x11, 0x40, 0xaa},	{0x00, 0x3a, 0x04, 0xaa},
+	{0x00, 0x12, 0x00, 0xaa},	{0x00, 0x40, 0xc0, 0xaa},
+	{0x00, 0x8c, 0x00, 0xaa},	{0x00, 0x17, 0x14, 0xaa},
+	{0x00, 0x18, 0x02, 0xaa},	{0x00, 0x32, 0x92, 0xaa},
+	{0x00, 0x19, 0x02, 0xaa},	{0x00, 0x1a, 0x7a, 0xaa},
+	{0x00, 0x03, 0x0a, 0xaa},	{0x00, 0x0c, 0x00, 0xaa},
+	{0x00, 0x3e, 0x00, 0xaa},	{0x00, 0x70, 0x3a, 0xaa},
+	{0x00, 0x71, 0x35, 0xaa},	{0x00, 0x72, 0x11, 0xaa},
+	{0x00, 0x73, 0xf0, 0xaa},	{0x00, 0xa2, 0x02, 0xaa},
+	{0x00, 0xb1, 0x00, 0xaa},	{0x00, 0xb1, 0x0c, 0xaa},
+	{0x00, 0x1e, 0x37, 0xaa},	{0x00, 0xaa, 0x14, 0xaa},
+	{0x00, 0x24, 0x80, 0xaa},	{0x00, 0x25, 0x74, 0xaa},
+	{0x00, 0x26, 0xd3, 0xaa},	{0x00, 0x0d, 0x00, 0xaa},
+	{0x00, 0x14, 0x18, 0xaa},	{0x00, 0x9d, 0x99, 0xaa},
+	{0x00, 0x9e, 0x7f, 0xaa},	{0x00, 0x64, 0x08, 0xaa},
+	{0x00, 0x94, 0x07, 0xaa},	{0x00, 0x95, 0x06, 0xaa},
+	{0x00, 0x66, 0x05, 0xaa},	{0x00, 0x41, 0x08, 0xaa},
+	{0x00, 0x3f, 0x00, 0xaa},	{0x00, 0x75, 0x07, 0xaa},
+	{0x00, 0x76, 0xe1, 0xaa},	{0x00, 0x4c, 0x00, 0xaa},
+	{0x00, 0x77, 0x00, 0xaa},	{0x00, 0x3d, 0xc2, 0xaa},
+	{0x00, 0x4b, 0x09, 0xaa},	{0x00, 0xc9, 0x60, 0xaa},
+	{0x00, 0x41, 0x38, 0xaa},	{0xb6, 0x00, 0x00, 0xcc},
+	{0xb6, 0x03, 0x02, 0xcc},	{0xb6, 0x02, 0x80, 0xcc},
+	{0xb6, 0x05, 0x01, 0xcc},	{0xb6, 0x04, 0xe0, 0xcc},
+	{0xb6, 0x12, 0xf8, 0xcc},	{0xb6, 0x13, 0x13, 0xcc},
+	{0xb6, 0x18, 0x02, 0xcc},	{0xb6, 0x17, 0x58, 0xcc},
+	{0xb6, 0x16, 0x00, 0xcc},	{0xb6, 0x22, 0x12, 0xcc},
+	{0xb6, 0x23, 0x0b, 0xcc},	{0xbf, 0xc0, 0x39, 0xcc},
+	{0xbf, 0xc1, 0x04, 0xcc},	{0xbf, 0xcc, 0x00, 0xcc},
+	{0xb3, 0x5c, 0x01, 0xcc},	{0xb3, 0x01, 0x45, 0xcc},
+	{0x00, 0x77, 0x05, 0xaa},
+	{},
+};
+
+static __u8 ov7670_initQVGA_JPG[][4] = {
+	{0xb3, 0x01, 0x05, 0xcc},	{0x00, 0x00, 0x30, 0xdd},
+	{0xb0, 0x03, 0x19, 0xcc},	{0x00, 0x00, 0x10, 0xdd},
+	{0xb0, 0x04, 0x02, 0xcc},	{0x00, 0x00, 0x10, 0xdd},
+	{0xb3, 0x00, 0x66, 0xcc},	{0xb3, 0x00, 0x67, 0xcc},
+	{0xb3, 0x35, 0xa1, 0xcc},	{0xb3, 0x34, 0x01, 0xcc},
+	{0xb3, 0x05, 0x01, 0xcc},	{0xb3, 0x06, 0x01, 0xcc},
+	{0xb3, 0x08, 0x01, 0xcc},	{0xb3, 0x09, 0x0c, 0xcc},
+	{0xb3, 0x02, 0x02, 0xcc},	{0xb3, 0x03, 0x1f, 0xcc},
+	{0xb3, 0x14, 0x00, 0xcc},	{0xb3, 0x15, 0x00, 0xcc},
+	{0xb3, 0x16, 0x02, 0xcc},	{0xb3, 0x17, 0x7f, 0xcc},
+	{0xb3, 0x04, 0x05, 0xcc},	{0xb3, 0x20, 0x00, 0xcc},
+	{0xb3, 0x21, 0x00, 0xcc},	{0xb3, 0x22, 0x01, 0xcc},
+	{0xb3, 0x23, 0xe0, 0xcc},	{0xbc, 0x00, 0xd1, 0xcc},
+	{0xbc, 0x01, 0x01, 0xcc},	{0x00, 0x12, 0x80, 0xaa},
+	{0x00, 0x00, 0x20, 0xdd},	{0x00, 0x12, 0x00, 0xaa},
+	{0x00, 0x11, 0x40, 0xaa},	{0x00, 0x6b, 0x0a, 0xaa},
+	{0x00, 0x3a, 0x04, 0xaa},	{0x00, 0x40, 0xc0, 0xaa},
+	{0x00, 0x8c, 0x00, 0xaa},	{0x00, 0x7a, 0x29, 0xaa},
+	{0x00, 0x7b, 0x0e, 0xaa},	{0x00, 0x7c, 0x1a, 0xaa},
+	{0x00, 0x7d, 0x31, 0xaa},	{0x00, 0x7e, 0x53, 0xaa},
+	{0x00, 0x7f, 0x60, 0xaa},	{0x00, 0x80, 0x6b, 0xaa},
+	{0x00, 0x81, 0x73, 0xaa},	{0x00, 0x82, 0x7b, 0xaa},
+	{0x00, 0x83, 0x82, 0xaa},	{0x00, 0x84, 0x89, 0xaa},
+	{0x00, 0x85, 0x96, 0xaa},	{0x00, 0x86, 0xa1, 0xaa},
+	{0x00, 0x87, 0xb7, 0xaa},	{0x00, 0x88, 0xcc, 0xaa},
+	{0x00, 0x89, 0xe1, 0xaa},	{0x00, 0x13, 0xe0, 0xaa},
+	{0x00, 0x00, 0x00, 0xaa},	{0x00, 0x10, 0x00, 0xaa},
+	{0x00, 0x0d, 0x40, 0xaa},	{0x00, 0x14, 0x28, 0xaa},
+	{0x00, 0xa5, 0x05, 0xaa},	{0x00, 0xab, 0x07, 0xaa},
+	{0x00, 0x24, 0x95, 0xaa},	{0x00, 0x25, 0x33, 0xaa},
+	{0x00, 0x26, 0xe3, 0xaa},	{0x00, 0x9f, 0x88, 0xaa},
+	{0x00, 0xa0, 0x78, 0xaa},	{0x00, 0x55, 0x90, 0xaa},
+	{0x00, 0xa1, 0x03, 0xaa},	{0x00, 0xa6, 0xe0, 0xaa},
+	{0x00, 0xa7, 0xd8, 0xaa},	{0x00, 0xa8, 0xf0, 0xaa},
+	{0x00, 0xa9, 0x90, 0xaa},	{0x00, 0xaa, 0x14, 0xaa},
+	{0x00, 0x13, 0xe5, 0xaa},	{0x00, 0x0e, 0x61, 0xaa},
+	{0x00, 0x0f, 0x4b, 0xaa},	{0x00, 0x16, 0x02, 0xaa},
+	{0x00, 0x1e, 0x07, 0xaa},	{0x00, 0x21, 0x02, 0xaa},
+	{0x00, 0x22, 0x91, 0xaa},	{0x00, 0x29, 0x07, 0xaa},
+	{0x00, 0x33, 0x0b, 0xaa},	{0x00, 0x35, 0x0b, 0xaa},
+	{0x00, 0x37, 0x1d, 0xaa},	{0x00, 0x38, 0x71, 0xaa},
+	{0x00, 0x39, 0x2a, 0xaa},	{0x00, 0x3c, 0x78, 0xaa},
+	{0x00, 0x4d, 0x40, 0xaa},	{0x00, 0x4e, 0x20, 0xaa},
+	{0x00, 0x74, 0x19, 0xaa},	{0x00, 0x8d, 0x4f, 0xaa},
+	{0x00, 0x8e, 0x00, 0xaa},	{0x00, 0x8f, 0x00, 0xaa},
+	{0x00, 0x90, 0x00, 0xaa},	{0x00, 0x91, 0x00, 0xaa},
+	{0x00, 0x96, 0x00, 0xaa},	{0x00, 0x9a, 0x80, 0xaa},
+	{0x00, 0xb0, 0x84, 0xaa},	{0x00, 0xb1, 0x0c, 0xaa},
+	{0x00, 0xb2, 0x0e, 0xaa},	{0x00, 0xb3, 0x82, 0xaa},
+	{0x00, 0xb8, 0x0a, 0xaa},	{0x00, 0x43, 0x14, 0xaa},
+	{0x00, 0x44, 0xf0, 0xaa},	{0x00, 0x45, 0x45, 0xaa},
+	{0x00, 0x46, 0x63, 0xaa},	{0x00, 0x47, 0x2d, 0xaa},
+	{0x00, 0x48, 0x46, 0xaa},	{0x00, 0x59, 0x88, 0xaa},
+	{0x00, 0x5a, 0xa0, 0xaa},	{0x00, 0x5b, 0xc6, 0xaa},
+	{0x00, 0x5c, 0x7d, 0xaa},	{0x00, 0x5d, 0x5f, 0xaa},
+	{0x00, 0x5e, 0x19, 0xaa},	{0x00, 0x6c, 0x0a, 0xaa},
+	{0x00, 0x6d, 0x55, 0xaa},	{0x00, 0x6e, 0x11, 0xaa},
+	{0x00, 0x6f, 0x9e, 0xaa},	{0x00, 0x69, 0x00, 0xaa},
+	{0x00, 0x6a, 0x40, 0xaa},	{0x00, 0x01, 0x40, 0xaa},
+	{0x00, 0x02, 0x40, 0xaa},	{0x00, 0x13, 0xe7, 0xaa},
+	{0x00, 0x5f, 0xf0, 0xaa},	{0x00, 0x60, 0xf0, 0xaa},
+	{0x00, 0x61, 0xf0, 0xaa},	{0x00, 0x27, 0xa0, 0xaa},
+	{0x00, 0x28, 0x80, 0xaa},	{0x00, 0x2c, 0x90, 0xaa},
+	{0x00, 0x4f, 0x66, 0xaa},	{0x00, 0x50, 0x66, 0xaa},
+	{0x00, 0x51, 0x00, 0xaa},	{0x00, 0x52, 0x22, 0xaa},
+	{0x00, 0x53, 0x5e, 0xaa},	{0x00, 0x54, 0x80, 0xaa},
+	{0x00, 0x58, 0x9e, 0xaa},	{0x00, 0x41, 0x08, 0xaa},
+	{0x00, 0x3f, 0x00, 0xaa},	{0x00, 0x75, 0x85, 0xaa},
+	{0x00, 0x76, 0xe1, 0xaa},	{0x00, 0x4c, 0x00, 0xaa},
+	{0x00, 0x77, 0x0a, 0xaa},	{0x00, 0x3d, 0x88, 0xaa},
+	{0x00, 0x4b, 0x09, 0xaa},	{0x00, 0xc9, 0x60, 0xaa},
+	{0x00, 0x41, 0x38, 0xaa},	{0x00, 0x62, 0x30, 0xaa},
+	{0x00, 0x63, 0x30, 0xaa},	{0x00, 0x64, 0x08, 0xaa},
+	{0x00, 0x94, 0x07, 0xaa},	{0x00, 0x95, 0x0b, 0xaa},
+	{0x00, 0x65, 0x00, 0xaa},	{0x00, 0x66, 0x05, 0xaa},
+	{0x00, 0x56, 0x50, 0xaa},	{0x00, 0x34, 0x11, 0xaa},
+	{0x00, 0xa4, 0x88, 0xaa},	{0x00, 0x96, 0x00, 0xaa},
+	{0x00, 0x97, 0x30, 0xaa},	{0x00, 0x98, 0x20, 0xaa},
+	{0x00, 0x99, 0x30, 0xaa},	{0x00, 0x9a, 0x84, 0xaa},
+	{0x00, 0x9b, 0x29, 0xaa},	{0x00, 0x9c, 0x03, 0xaa},
+	{0x00, 0x78, 0x04, 0xaa},	{0x00, 0x79, 0x01, 0xaa},
+	{0x00, 0xc8, 0xf0, 0xaa},	{0x00, 0x79, 0x0f, 0xaa},
+	{0x00, 0xc8, 0x00, 0xaa},	{0x00, 0x79, 0x10, 0xaa},
+	{0x00, 0xc8, 0x7e, 0xaa},	{0x00, 0x79, 0x0a, 0xaa},
+	{0x00, 0xc8, 0x80, 0xaa},	{0x00, 0x79, 0x0b, 0xaa},
+	{0x00, 0xc8, 0x01, 0xaa},	{0x00, 0x79, 0x0c, 0xaa},
+	{0x00, 0xc8, 0x0f, 0xaa},	{0x00, 0x79, 0x0d, 0xaa},
+	{0x00, 0xc8, 0x20, 0xaa},	{0x00, 0x79, 0x09, 0xaa},
+	{0x00, 0xc8, 0x80, 0xaa},	{0x00, 0x79, 0x02, 0xaa},
+	{0x00, 0xc8, 0xc0, 0xaa},	{0x00, 0x79, 0x03, 0xaa},
+	{0x00, 0xc8, 0x40, 0xaa},	{0x00, 0x79, 0x05, 0xaa},
+	{0x00, 0xc8, 0x30, 0xaa},	{0x00, 0x79, 0x26, 0xaa},
+	{0x00, 0x11, 0x40, 0xaa},	{0x00, 0x3a, 0x04, 0xaa},
+	{0x00, 0x12, 0x00, 0xaa},	{0x00, 0x40, 0xc0, 0xaa},
+	{0x00, 0x8c, 0x00, 0xaa},	{0x00, 0x17, 0x14, 0xaa},
+	{0x00, 0x18, 0x02, 0xaa},	{0x00, 0x32, 0x92, 0xaa},
+	{0x00, 0x19, 0x02, 0xaa},	{0x00, 0x1a, 0x7a, 0xaa},
+	{0x00, 0x03, 0x0a, 0xaa},	{0x00, 0x0c, 0x00, 0xaa},
+	{0x00, 0x3e, 0x00, 0xaa},	{0x00, 0x70, 0x3a, 0xaa},
+	{0x00, 0x71, 0x35, 0xaa},	{0x00, 0x72, 0x11, 0xaa},
+	{0x00, 0x73, 0xf0, 0xaa},	{0x00, 0xa2, 0x02, 0xaa},
+	{0x00, 0xb1, 0x00, 0xaa},	{0x00, 0xb1, 0x0c, 0xaa},
+	{0x00, 0x1e, 0x37, 0xaa},	{0x00, 0xaa, 0x14, 0xaa},
+	{0x00, 0x24, 0x80, 0xaa},	{0x00, 0x25, 0x74, 0xaa},
+	{0x00, 0x26, 0xd3, 0xaa},	{0x00, 0x0d, 0x00, 0xaa},
+	{0x00, 0x14, 0x18, 0xaa},	{0x00, 0x9d, 0x99, 0xaa},
+	{0x00, 0x9e, 0x7f, 0xaa},	{0x00, 0x64, 0x08, 0xaa},
+	{0x00, 0x94, 0x07, 0xaa},	{0x00, 0x95, 0x06, 0xaa},
+	{0x00, 0x66, 0x05, 0xaa},	{0x00, 0x41, 0x08, 0xaa},
+	{0x00, 0x3f, 0x00, 0xaa},	{0x00, 0x75, 0x07, 0xaa},
+	{0x00, 0x76, 0xe1, 0xaa},	{0x00, 0x4c, 0x00, 0xaa},
+	{0x00, 0x77, 0x00, 0xaa},	{0x00, 0x3d, 0xc2, 0xaa},
+	{0x00, 0x4b, 0x09, 0xaa},	{0x00, 0xc9, 0x60, 0xaa},
+	{0x00, 0x41, 0x38, 0xaa},	{0xb6, 0x00, 0x00, 0xcc},
+	{0xb6, 0x03, 0x01, 0xcc},	{0xb6, 0x02, 0x40, 0xcc},
+	{0xb6, 0x05, 0x00, 0xcc},	{0xb6, 0x04, 0xf0, 0xcc},
+	{0xb6, 0x12, 0xf8, 0xcc},	{0xb6, 0x13, 0x21, 0xcc},
+	{0xb6, 0x18, 0x00, 0xcc},	{0xb6, 0x17, 0x96, 0xcc},
+	{0xb6, 0x16, 0x00, 0xcc},	{0xb6, 0x22, 0x12, 0xcc},
+	{0xb6, 0x23, 0x0b, 0xcc},	{0xbf, 0xc0, 0x39, 0xcc},
+	{0xbf, 0xc1, 0x04, 0xcc},	{0xbf, 0xcc, 0x00, 0xcc},
+	{0xbc, 0x02, 0x18, 0xcc},	{0xbc, 0x03, 0x50, 0xcc},
+	{0xbc, 0x04, 0x18, 0xcc},	{0xbc, 0x05, 0x00, 0xcc},
+	{0xbc, 0x06, 0x00, 0xcc},	{0xbc, 0x08, 0x30, 0xcc},
+	{0xbc, 0x09, 0x40, 0xcc},	{0xbc, 0x0a, 0x10, 0xcc},
+	{0xbc, 0x0b, 0x00, 0xcc},	{0xbc, 0x0c, 0x00, 0xcc},
+	{0xb3, 0x5c, 0x01, 0xcc},	{0xb3, 0x01, 0x45, 0xcc},
+	{0x00, 0x77, 0x05, 0xaa },
+	{},
+};
+
+struct sensor_info {
+	int sensorId;
+	__u8 I2cAdd;
+	__u8 IdAdd;
+	__u16 VpId;
+	__u8 m1;
+	__u8 m2;
+	__u8 op;
+	};
+
+static struct sensor_info sensor_info_data[] = {
+/*      sensorId,         I2cAdd,	IdAdd,  VpId,  m1,    m2,  op */
+	{SENSOR_HV7131R,    0x80 | 0x11, 0x00, 0x0209, 0x24, 0x25, 0x01},
+	{SENSOR_OV7660,     0x80 | 0x21, 0x0a, 0x7660, 0x26, 0x26, 0x05},
+	{SENSOR_PO3130NC,   0x80 | 0x76, 0x00, 0x3130, 0x24, 0x25, 0x01},
+	{SENSOR_MI1320,     0x80 | 0xc8, 0x00, 0x148c, 0x64, 0x65, 0x01},
+	{SENSOR_OV7670,     0x80 | 0x21, 0x0a, 0x7673, 0x66, 0x67, 0x05},
+	{SENSOR_MI1310_SOC, 0x80 | 0x5d, 0x00, 0x143a, 0x24, 0x25, 0x01},
+};
+
+static void reg_r(struct usb_device *dev,
+			   __u16 req,
+			   __u16 index,
+			   __u8 *buffer, __u16 length)
+{
+	usb_control_msg(dev,
+			usb_rcvctrlpipe(dev, 0),
+			req,
+			USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			1,			/* value */
+			index, buffer, length,
+			500);
+}
+
+static void reg_w(struct usb_device *dev,
+			    __u16 req,
+			    __u16 value,
+			    __u16 index)
+{
+	usb_control_msg(dev,
+			usb_sndctrlpipe(dev, 0),
+			req,
+			USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+			value, index, NULL, 0,
+			500);
+}
+
+static void vc032x_read_sensor_register(struct usb_device *dev,
+				__u16 address, __u16 *value)
+{
+	__u8 ldata, mdata, hdata;
+	__u8 tmpvalue = 0;
+	int retry = 50;
+	ldata = 0;
+	mdata = 0;
+	hdata = 0;
+	*value = 0;
+
+	reg_r(dev, 0xa1, 0xb33f, &tmpvalue, 1);
+	/*PDEBUG(D_PROBE, " I2c Bus Busy Wait  0x%02X ", tmpvalue); */
+	if (!(tmpvalue & 0x02)) {
+		PDEBUG(D_ERR, "I2c Bus Busy Wait %d", tmpvalue & 0x02);
+		return;
+	}
+	reg_w(dev, 0xa0, address, 0xb33a);
+	reg_w(dev, 0xa0, 0x02, 0xb339);
+
+	tmpvalue = 0;
+	reg_r(dev, 0xa1, 0xb33b, &tmpvalue, 1);
+	while (retry-- && tmpvalue) {
+		reg_r(dev, 0xa1, 0xb33b, &tmpvalue, 1);
+/*		PDEBUG(D_PROBE, "Read again 0xb33b %d", tmpvalue); */
+		msleep(1);
+	}
+	reg_r(dev, 0xa1, 0xb33e, &hdata, 1);
+	reg_r(dev, 0xa1, 0xb33d, &mdata, 1);
+	reg_r(dev, 0xa1, 0xb33c, &ldata, 1);
+	PDEBUG(D_PROBE, "Read Sensor h (0x%02X) m (0x%02X) l (0x%02X)",
+		hdata, mdata, ldata);
+	tmpvalue = 0;
+	reg_r(dev, 0xa1, 0xb334, &tmpvalue, 1);
+	if (tmpvalue == 0x02)
+		*value = (ldata << 8) + mdata;
+	else
+		*value = ldata;
+}
+static int vc032x_probe_sensor(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int i;
+	__u8 data;
+	__u16 value;
+	struct sensor_info *ptsensor_info;
+
+	reg_r(dev, 0xa1, 0xbfcf, &data, 1);
+	PDEBUG(D_PROBE, "check sensor header %d", data);
+	for (i = 0; i < ARRAY_SIZE(sensor_info_data); i++) {
+		ptsensor_info = &sensor_info_data[i];
+		reg_w(dev, 0xa0, 0x02, 0xb334);
+		reg_w(dev, 0xa0, ptsensor_info->m1, 0xb300);
+		reg_w(dev, 0xa0, ptsensor_info->m2, 0xb300);
+		reg_w(dev, 0xa0, 0x01, 0xb308);
+		reg_w(dev, 0xa0, 0x0c, 0xb309);
+		reg_w(dev, 0xa0, ptsensor_info->I2cAdd, 0xb335);
+/*		PDEBUG(D_PROBE,
+			"check sensor VC032X -> %d Add -> ox%02X!",
+			i, ptsensor_info->I2cAdd); */
+		reg_w(dev, 0xa0, ptsensor_info->op, 0xb301);
+		vc032x_read_sensor_register(dev, ptsensor_info->IdAdd, &value);
+		if (value == ptsensor_info->VpId) {
+/*			PDEBUG(D_PROBE, "find sensor VC032X -> ox%04X!",
+				ptsensor_info->VpId); */
+			return ptsensor_info->sensorId;
+		}
+	}
+	return -1;
+}
+
+static __u8 i2c_write(struct usb_device *dev,
+				__u8 reg, __u8 *val, __u8 size)
+{
+	__u8 retbyte;
+
+	if (size > 3 || size < 1)
+		return -EINVAL;
+	reg_r(dev, 0xa1, 0xb33f, &retbyte, 1);
+	reg_w(dev, 0xa0, size , 0xb334);
+	reg_w(dev, 0xa0, reg , 0xb33a);
+	switch (size) {
+	case 1:
+		reg_w(dev, 0xa0, val[0] , 0xb336);
+		break;
+	case 2:
+		reg_w(dev, 0xa0, val[0] , 0xb336);
+		reg_w(dev, 0xa0, val[1] , 0xb337);
+		break;
+	case 3:
+		reg_w(dev, 0xa0, val[0] , 0xb336);
+		reg_w(dev, 0xa0, val[1] , 0xb337);
+		reg_w(dev, 0xa0, val[2] , 0xb338);
+		break;
+	default:
+		reg_w(dev, 0xa0, 0x01, 0xb334);
+		return -EINVAL;
+	}
+	reg_w(dev, 0xa0, 0x01, 0xb339);
+	reg_r(dev, 0xa1, 0xb33b, &retbyte, 1);
+	return retbyte == 0;
+}
+
+static void put_tab_to_reg(struct gspca_dev *gspca_dev,
+			__u8 *tab, __u8 tabsize, __u16 addr)
+{
+	int j;
+	__u16 ad = addr;
+
+	for (j = 0; j < tabsize; j++)
+		reg_w(gspca_dev->dev, 0xa0, tab[j], ad++);
+}
+
+static void usb_exchange(struct gspca_dev *gspca_dev,
+				__u8 data[][4])
+{
+	struct usb_device *dev = gspca_dev->dev;
+	int i = 0;
+
+	for (;;) {
+		switch (data[i][3]) {
+		default:
+			return;
+		case 0xcc:			/* normal write */
+			reg_w(dev, 0xa0, data[i][2],
+					((data[i][0])<<8) | data[i][1]);
+			break;
+		case 0xaa:			/* i2c op */
+			i2c_write(dev, data[i][1], &data[i][2], 1);
+			break;
+		case 0xbb:			/* i2c op */
+			i2c_write(dev, data[i][0], &data[i][1], 2);
+			break;
+		case 0xdd:
+			msleep(data[i][2] + 10);
+			break;
+		}
+		i++;
+	}
+	/*not reached*/
+}
+
+/*
+ "GammaT"=hex:04,17,31,4f,6a,83,99,ad,bf,ce,da,e5,ee,f5,fb,ff,ff
+ "MatrixT"=hex:60,f9,e5,e7,50,05,f3,e6,66
+ */
+
+static void vc0321_reset(struct gspca_dev *gspca_dev)
+{
+	reg_w(gspca_dev->dev, 0xa0, 0x00, 0xb04d);
+	reg_w(gspca_dev->dev, 0xa0, 0x01, 0xb301);
+	msleep(100);
+	reg_w(gspca_dev->dev, 0xa0, 0x01, 0xb003);
+	msleep(100);
+}
+
+/* this function is called at probe time */
+static int sd_config(struct gspca_dev *gspca_dev,
+			const struct usb_device_id *id)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	struct usb_device *dev = gspca_dev->dev;
+	struct cam *cam;
+	__u8 tmp2[4];
+	int sensor;
+	__u16 vendor;
+	__u16 product;
+
+	vendor = id->idVendor;
+	product = id->idProduct;
+	switch (vendor) {
+	case 0x046d:		/* Logitech Labtec */
+/*		switch (product) { */
+/*		case 0x0892: */
+/*		case 0x0896: */
+			sd->bridge = BRIDGE_VC0321;
+/*			break; */
+/*		} */
+		break;
+	case 0x0ac8:		/* Vimicro z-star */
+		switch (product) {
+		case 0x0321:
+		case 0x0328:
+		case 0xc001:
+		case 0xc002:
+			sd->bridge = BRIDGE_VC0321;
+			break;
+		case 0x0323:
+			sd->bridge = BRIDGE_VC0323;
+			break;
+		}
+		break;
+	case 0x17ef:		/* Lenovo */
+/*		switch (product) { */
+/*		case 0x4802:	 * Lenovo MI1310_SOC */
+			sd->bridge = BRIDGE_VC0323;
+/*			break; */
+/*		} */
+		break;
+	}
+
+	cam = &gspca_dev->cam;
+	cam->dev_name = (char *) id->driver_info;
+	cam->epaddr = 0x02;
+	if (sd->bridge == BRIDGE_VC0321) {
+		cam->cam_mode = vc0321_mode;
+		cam->nmodes = sizeof vc0321_mode / sizeof vc0321_mode[0];
+	} else {
+		cam->cam_mode = vc0323_mode;
+		cam->nmodes = sizeof vc0323_mode / sizeof vc0323_mode[0];
+	}
+
+	vc0321_reset(gspca_dev);
+	sensor = vc032x_probe_sensor(gspca_dev);
+	switch (sensor) {
+	case -1:
+		PDEBUG(D_PROBE, "Unknown sensor...");
+		return -EINVAL;
+	case SENSOR_HV7131R:
+		PDEBUG(D_PROBE, "Find Sensor HV7131R");
+		sd->sensor = SENSOR_HV7131R;
+		break;
+	case SENSOR_MI1310_SOC:
+		PDEBUG(D_PROBE, "Find Sensor MI1310_SOC");
+		sd->sensor = SENSOR_MI1310_SOC;
+		break;
+	case SENSOR_MI1320:
+		PDEBUG(D_PROBE, "Find Sensor MI1320");
+		sd->sensor = SENSOR_MI1320;
+		break;
+	case SENSOR_OV7660:
+		PDEBUG(D_PROBE, "Find Sensor OV7660");
+		sd->sensor = SENSOR_OV7660;
+		break;
+	case SENSOR_OV7670:
+		PDEBUG(D_PROBE, "Find Sensor OV7670");
+		sd->sensor = SENSOR_OV7670;
+		break;
+	case SENSOR_PO3130NC:
+		PDEBUG(D_PROBE, "Find Sensor PO3130NC");
+		sd->sensor = SENSOR_PO3130NC;
+		break;
+	}
+
+	sd->qindex = 7;
+	sd->autogain = sd_ctrls[SD_AUTOGAIN].qctrl.default_value;
+
+	if (sd->bridge == BRIDGE_VC0321) {
+		reg_r(dev, 0x8a, 0, tmp2, 3);
+		reg_w(dev, 0x87, 0x00, 0x0f0f);
+
+		reg_r(dev, 0x8b, 0, tmp2, 3);
+		reg_w(dev, 0x88, 0x00, 0x0202);
+	}
+	return 0;
+}
+
+/* this function is called at open time */
+static int sd_open(struct gspca_dev *gspca_dev)
+{
+	return 0;
+}
+
+static void setquality(struct gspca_dev *gspca_dev)
+{
+}
+
+static void setautogain(struct gspca_dev *gspca_dev)
+{
+}
+
+static void setlightfreq(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	static __u8 (*ov7660_freq_tb[3])[4] =
+		{ov7660_NoFliker, ov7660_50HZ, ov7660_60HZ};
+
+	if (sd->sensor != SENSOR_OV7660)
+		return;
+	usb_exchange(gspca_dev, ov7660_freq_tb[sd->lightfreq]);
+}
+
+static void sd_start(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+/*	__u8 tmp2; */
+	__u8 *GammaT = NULL;
+	__u8 *MatrixT = NULL;
+	int mode;
+
+	/* Assume start use the good resolution from gspca_dev->mode */
+	if (sd->bridge == BRIDGE_VC0321) {
+		reg_w(gspca_dev->dev, 0xa0, 0xff, 0xbfec);
+		reg_w(gspca_dev->dev, 0xa0, 0xff, 0xbfed);
+		reg_w(gspca_dev->dev, 0xa0, 0xff, 0xbfee);
+		reg_w(gspca_dev->dev, 0xa0, 0xff, 0xbfef);
+	}
+
+	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
+	switch (sd->sensor) {
+	case SENSOR_HV7131R:
+		GammaT = hv7131r_gamma;
+		MatrixT = hv7131r_matrix;
+		if (mode) {
+			/* 320x240 */
+			usb_exchange(gspca_dev, hv7131r_initQVGA_data);
+		} else {
+			/* 640x480 */
+			usb_exchange(gspca_dev, hv7131r_initVGA_data);
+		}
+		break;
+	case SENSOR_OV7660:
+		GammaT = ov7660_gamma;
+		MatrixT = ov7660_matrix;
+		if (mode) {
+			/* 320x240 */
+			usb_exchange(gspca_dev, ov7660_initQVGA_data);
+		} else {
+			/* 640x480 */
+			usb_exchange(gspca_dev, ov7660_initVGA_data);
+		}
+		break;
+	case SENSOR_OV7670:
+		/*GammaT = ov7660_gamma; */
+		/*MatrixT = ov7660_matrix; */
+		if (mode) {
+			/* 320x240 */
+			usb_exchange(gspca_dev, ov7670_initQVGA_JPG);
+		} else {
+			/* 640x480 */
+			usb_exchange(gspca_dev, ov7670_initVGA_JPG);
+		}
+		break;
+	case SENSOR_MI1310_SOC:
+		if (mode) {
+			/* 320x240 */
+			usb_exchange(gspca_dev, mi1310_socinitQVGA_JPG);
+		} else {
+			/* 640x480 */
+			usb_exchange(gspca_dev, mi1310_socinitVGA_JPG);
+		}
+		break;
+	case SENSOR_MI1320:
+		GammaT = mi1320_gamma;
+		MatrixT = mi1320_matrix;
+		if (mode) {
+			/* 320x240 */
+			usb_exchange(gspca_dev, mi1320_initQVGA_data);
+		} else {
+			/* 640x480 */
+			usb_exchange(gspca_dev, mi1320_initVGA_data);
+		}
+		break;
+	case SENSOR_PO3130NC:
+		GammaT = po3130_gamma;
+		MatrixT = po3130_matrix;
+		if (mode) {
+			/* 320x240 */
+			usb_exchange(gspca_dev, po3130_initQVGA_data);
+		} else {
+			/* 640x480 */
+			usb_exchange(gspca_dev, po3130_initVGA_data);
+		}
+		usb_exchange(gspca_dev, po3130_rundata);
+		break;
+	default:
+		PDEBUG(D_PROBE, "Damned !! no sensor found Bye");
+		return;
+	}
+	if (GammaT && MatrixT) {
+		put_tab_to_reg(gspca_dev, GammaT, 17, 0xb84a);
+		put_tab_to_reg(gspca_dev, GammaT, 17, 0xb85b);
+		put_tab_to_reg(gspca_dev, GammaT, 17, 0xb86c);
+		put_tab_to_reg(gspca_dev, MatrixT, 9, 0xb82c);
+
+		/* Seem SHARPNESS */
+		/*
+		reg_w(gspca_dev->dev, 0xa0, 0x80, 0xb80a);
+		reg_w(gspca_dev->dev, 0xa0, 0xff, 0xb80b);
+		reg_w(gspca_dev->dev, 0xa0, 0xff, 0xb80e);
+		*/
+		/* all 0x40 ??? do nothing
+		reg_w(gspca_dev->dev, 0xa0, 0x40, 0xb822);
+		reg_w(gspca_dev->dev, 0xa0, 0x40, 0xb823);
+		reg_w(gspca_dev->dev, 0xa0, 0x40, 0xb824);
+		*/
+		/* Only works for HV7131R ??
+		reg_r (gspca_dev->dev, 0xa1, 0xb881, &tmp2, 1);
+		reg_w(gspca_dev->dev, 0xa0, 0xfe01, 0xb881);
+		reg_w(gspca_dev->dev, 0xa0, 0x79, 0xb801);
+		*/
+		/* only hv7131r et ov7660
+		reg_w(gspca_dev->dev, 0xa0, 0x20, 0xb827);
+		reg_w(gspca_dev->dev, 0xa0, 0xff, 0xb826); * ISP_GAIN 80
+		reg_w(gspca_dev->dev, 0xa0, 0x23, 0xb800); * ISP CTRL_BAS
+		*/
+		/* set the led on 0x0892 0x0896 */
+		reg_w(gspca_dev->dev, 0x89, 0xffff, 0xfdff);
+		msleep(100);
+		setquality(gspca_dev);
+		setautogain(gspca_dev);
+		setlightfreq(gspca_dev);
+	}
+}
+
+static void sd_stopN(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	reg_w(dev, 0x89, 0xffff, 0xffff);
+	reg_w(dev, 0xa0, 0x01, 0xb301);
+	reg_w(dev, 0xa0, 0x09, 0xb003);
+}
+
+static void sd_stop0(struct gspca_dev *gspca_dev)
+{
+	struct usb_device *dev = gspca_dev->dev;
+
+	reg_w(dev, 0x89, 0xffff, 0xffff);
+}
+
+/* this function is called at close time */
+static void sd_close(struct gspca_dev *gspca_dev)
+{
+/*	struct usb_device *dev = gspca_dev->dev;
+	__u8 buffread;
+
+	reg_w(dev, 0x89, 0xffff, 0xffff);
+	reg_w(dev, 0xa0, 0x01, 0xb301);
+	reg_w(dev, 0xa0, 0x09, 0xb303);
+	reg_w(dev, 0x89, 0xffff, 0xffff);
+*/
+}
+
+static void sd_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			unsigned char *data,		/* isoc packet */
+			int len)			/* iso pkt length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	if (data[0] == 0xff && data[1] == 0xd8) {
+		PDEBUG(D_PACK,
+			"vc032x header packet found len %d", len);
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+						data, 0);
+		if (sd->bridge == BRIDGE_VC0321) {
+#define VCHDRSZ 46
+			data += VCHDRSZ;
+			len -= VCHDRSZ;
+#undef VCHDRSZ
+		}
+		gspca_frame_add(gspca_dev, FIRST_PACKET, frame,
+				data, len);
+		return;
+	}
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
+}
+
+static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->autogain = val;
+	if (gspca_dev->streaming)
+		setautogain(gspca_dev);
+	return 0;
+}
+
+static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autogain;
+	return 0;
+}
+
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->lightfreq = val;
+	if (gspca_dev->streaming)
+		setlightfreq(gspca_dev);
+	return 0;
+}
+
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->lightfreq;
+	return 0;
+}
+
+static int sd_querymenu(struct gspca_dev *gspca_dev,
+			struct v4l2_querymenu *menu)
+{
+	switch (menu->id) {
+	case V4L2_CID_POWER_LINE_FREQUENCY:
+		switch (menu->index) {
+		case 0:		/* V4L2_CID_POWER_LINE_FREQUENCY_DISABLED */
+			strcpy(menu->name, "NoFliker");
+			return 0;
+		case 1:		/* V4L2_CID_POWER_LINE_FREQUENCY_50HZ */
+			strcpy(menu->name, "50 Hz");
+			return 0;
+		case 2:		/* V4L2_CID_POWER_LINE_FREQUENCY_60HZ */
+			strcpy(menu->name, "60 Hz");
+			return 0;
+		}
+		break;
+	}
+	return -EINVAL;
+}
+
+/* sub-driver description */
+static struct sd_desc sd_desc = {
+	.name = MODULE_NAME,
+	.ctrls = sd_ctrls,
+	.nctrls = ARRAY_SIZE(sd_ctrls),
+	.config = sd_config,
+	.open = sd_open,
+	.start = sd_start,
+	.stopN = sd_stopN,
+	.stop0 = sd_stop0,
+	.close = sd_close,
+	.pkt_scan = sd_pkt_scan,
+	.querymenu = sd_querymenu,
+};
+
+/* -- module initialisation -- */
+#define DVNM(name) .driver_info = (kernel_ulong_t) name
+static __devinitdata struct usb_device_id device_table[] = {
+	{USB_DEVICE(0x046d, 0x0892), DVNM("Logitech Orbicam")},
+	{USB_DEVICE(0x046d, 0x0896), DVNM("Logitech Orbicam")},
+	{USB_DEVICE(0x0ac8, 0x0321), DVNM("Vimicro generic vc0321")},
+	{USB_DEVICE(0x0ac8, 0x0323), DVNM("Vimicro Vc0323")},
+	{USB_DEVICE(0x0ac8, 0x0328), DVNM("A4Tech PK-130MG")},
+	{USB_DEVICE(0x0ac8, 0xc001), DVNM("Sony embedded vimicro")},
+	{USB_DEVICE(0x0ac8, 0xc002), DVNM("Sony embedded vimicro")},
+	{USB_DEVICE(0x17ef, 0x4802), DVNM("Lenovo Vc0323+MI1310_SOC")},
+	{}
+};
+MODULE_DEVICE_TABLE(usb, device_table);
+
+/* -- device connect -- */
+static int sd_probe(struct usb_interface *intf,
+			const struct usb_device_id *id)
+{
+	return gspca_dev_probe(intf, id, &sd_desc, sizeof(struct sd),
+				THIS_MODULE);
+}
+
+static struct usb_driver sd_driver = {
+	.name = MODULE_NAME,
+	.id_table = device_table,
+	.probe = sd_probe,
+	.disconnect = gspca_disconnect,
+};
+
+/* -- module insert / remove -- */
+static int __init sd_mod_init(void)
+{
+	if (usb_register(&sd_driver) < 0)
+		return -1;
+	PDEBUG(D_PROBE, "v%s registered", version);
+	return 0;
+}
+static void __exit sd_mod_exit(void)
+{
+	usb_deregister(&sd_driver);
+	PDEBUG(D_PROBE, "deregistered");
+}
+
+module_init(sd_mod_init);
+module_exit(sd_mod_exit);
diff --git a/drivers/media/video/gspca/zc3xx.c b/drivers/media/video/gspca/zc3xx.c
index 03cc7fc58dbc..b767f32511bb 100644
--- a/drivers/media/video/gspca/zc3xx.c
+++ b/drivers/media/video/gspca/zc3xx.c
@@ -3,7 +3,7 @@
  *	Copyright (C) 2004 2005 2006 Michel Xhaard
  *		mxhaard@magic.fr
  *
- * V4L2 by Jean-François Moine <http://moinejf.free.fr>
+ * V4L2 by Jean-Francois Moine <http://moinejf.free.fr>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -24,15 +24,14 @@
 
 #include "gspca.h"
 
-#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(0, 2, 13)
-static const char version[] = "0.2.13";
+#define DRIVER_VERSION_NUMBER	KERNEL_VERSION(2, 1, 0)
+static const char version[] = "2.1.0";
 
 MODULE_AUTHOR("Michel Xhaard <mxhaard@users.sourceforge.net>, "
 		"Serge A. Suchkov <Serge.A.S@tochka.ru>");
 MODULE_DESCRIPTION("GSPCA ZC03xx/VC3xx USB Camera Driver");
 MODULE_LICENSE("GPL");
 
-static int lightfreq = 50;
 static int force_sensor = -1;
 
 #include "jpeg.h"
@@ -41,10 +40,12 @@ static int force_sensor = -1;
 struct sd {
 	struct gspca_dev gspca_dev;	/* !! must be the first item */
 
-	unsigned char brightness;
-	unsigned char contrast;
-	unsigned char autogain;
-	unsigned char gamma;
+	__u8 brightness;
+	__u8 contrast;
+	__u8 gamma;
+	__u8 autogain;
+	__u8 lightfreq;
+	__u8 sharpness;
 
 	char qindex;
 	char sensor;			/* Type of image sensor chip */
@@ -61,14 +62,13 @@ struct sd {
 #define SENSOR_OV7620 9
 /*#define SENSOR_OV7648 9 - same values */
 #define SENSOR_OV7630C 10
-/*#define SENSOR_free 11 */
-#define SENSOR_PAS106 12
-#define SENSOR_PB0330 13
-#define SENSOR_PO2030 14
-#define SENSOR_TAS5130CK 15
-#define SENSOR_TAS5130CXX 16
-#define SENSOR_TAS5130C_VF0250 17
-#define SENSOR_MAX 18
+#define SENSOR_PAS106 11
+#define SENSOR_PB0330 12
+#define SENSOR_PO2030 13
+#define SENSOR_TAS5130CK 14
+#define SENSOR_TAS5130CXX 15
+#define SENSOR_TAS5130C_VF0250 16
+#define SENSOR_MAX 17
 	unsigned short chip_revision;
 };
 
@@ -79,7 +79,12 @@ static int sd_setcontrast(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getcontrast(struct gspca_dev *gspca_dev, __s32 *val);
 static int sd_setautogain(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setgamma(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getgamma(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setsharpness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getsharpness(struct gspca_dev *gspca_dev, __s32 *val);
 
 static struct ctrl sd_ctrls[] = {
 #define SD_BRIGHTNESS 0
@@ -110,7 +115,21 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setcontrast,
 	    .get = sd_getcontrast,
 	},
-#define SD_AUTOGAIN 2
+#define SD_GAMMA 2
+	{
+	    {
+		.id      = V4L2_CID_GAMMA,
+		.type    = V4L2_CTRL_TYPE_INTEGER,
+		.name    = "Gamma",
+		.minimum = 1,
+		.maximum = 6,
+		.step    = 1,
+		.default_value = 4,
+	    },
+	    .set = sd_setgamma,
+	    .get = sd_getgamma,
+	},
+#define SD_AUTOGAIN 3
 	{
 	    {
 		.id      = V4L2_CID_AUTOGAIN,
@@ -124,19 +143,33 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setautogain,
 	    .get = sd_getautogain,
 	},
-#define SD_GAMMA 3
+#define SD_FREQ 4
 	{
 	    {
-		.id      = V4L2_CID_GAMMA,
+		.id	 = V4L2_CID_POWER_LINE_FREQUENCY,
+		.type    = V4L2_CTRL_TYPE_MENU,
+		.name    = "Light frequency filter",
+		.minimum = 0,
+		.maximum = 2,	/* 0: 0, 1: 50Hz, 2:60Hz */
+		.step    = 1,
+		.default_value = 1,
+	    },
+	    .set = sd_setfreq,
+	    .get = sd_getfreq,
+	},
+#define SD_SHARPNESS 5
+	{
+	    {
+		.id	 = V4L2_CID_SHARPNESS,
 		.type    = V4L2_CTRL_TYPE_INTEGER,
-		.name    = "Gamma",
-		.minimum = 1,
-		.maximum = 6,
+		.name    = "Sharpness",
+		.minimum = 0,
+		.maximum = 3,
 		.step    = 1,
-		.default_value = 4,
+		.default_value = 2,
 	    },
-	    .set = sd_setcontrast,
-	    .get = sd_getgamma,
+	    .set = sd_setsharpness,
+	    .get = sd_getsharpness,
 	},
 };
 
@@ -211,11 +244,11 @@ static struct usb_action cs2102_Initial[] = {
 	{0xa1, 0x01, 0x0002},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* 00 */
-	{0xa0, 0x08, 0x01c6},	/* clock ? */
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x24, 0x0120},	/* gamma 5 */
 	{0xa0, 0x44, 0x0121},
 	{0xa0, 0x64, 0x0122},
@@ -284,7 +317,7 @@ static struct usb_action cs2102_Initial[] = {
 	{0xa0, 0x40, 0x0116},
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action cs2102_InitialScale[] = {
@@ -341,11 +374,11 @@ static struct usb_action cs2102_InitialScale[] = {
 	{0xa1, 0x01, 0x0002},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* 00 */
-	{0xa0, 0x08, 0x01c6},	/* clock ? */
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x24, 0x0120},	/* gamma 5 */
 	{0xa0, 0x44, 0x0121},
 	{0xa0, 0x64, 0x0122},
@@ -414,7 +447,7 @@ static struct usb_action cs2102_InitialScale[] = {
 	{0xa0, 0x40, 0x0116},
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
-	{0, 0, 0}
+	{}
 };
 static struct usb_action cs2102_50HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -439,7 +472,7 @@ static struct usb_action cs2102_50HZ[] = {
 	{0xa0, 0x8c, 0x001d}, /* 00,1d,8c,cc */
 	{0xa0, 0xb0, 0x001e}, /* 00,1e,b0,cc */
 	{0xa0, 0xd0, 0x001f}, /* 00,1f,d0,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action cs2102_50HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -464,7 +497,7 @@ static struct usb_action cs2102_50HZScale[] = {
 	{0xa0, 0x93, 0x001d}, /* 00,1d,93,cc */
 	{0xa0, 0xb0, 0x001e}, /* 00,1e,b0,cc */
 	{0xa0, 0xd0, 0x001f}, /* 00,1f,d0,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action cs2102_60HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -489,7 +522,7 @@ static struct usb_action cs2102_60HZ[] = {
 	{0xa0, 0x5d, 0x001d}, /* 00,1d,5d,cc */
 	{0xa0, 0x90, 0x001e}, /* 00,1e,90,cc */
 	{0xa0, 0xd0, 0x00c8}, /* 00,c8,d0,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action cs2102_60HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -514,7 +547,7 @@ static struct usb_action cs2102_60HZScale[] = {
 	{0xa0, 0xb7, 0x001d}, /* 00,1d,b7,cc */
 	{0xa0, 0xd0, 0x001e}, /* 00,1e,d0,cc */
 	{0xa0, 0xe8, 0x001f}, /* 00,1f,e8,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action cs2102_NoFliker[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -539,7 +572,7 @@ static struct usb_action cs2102_NoFliker[] = {
 	{0xa0, 0x59, 0x001d}, /* 00,1d,59,cc */
 	{0xa0, 0x90, 0x001e}, /* 00,1e,90,cc */
 	{0xa0, 0xc8, 0x001f}, /* 00,1f,c8,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action cs2102_NoFlikerScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -564,7 +597,7 @@ static struct usb_action cs2102_NoFlikerScale[] = {
 	{0xa0, 0x59, 0x001d}, /* 00,1d,59,cc */
 	{0xa0, 0x90, 0x001e}, /* 00,1e,90,cc */
 	{0xa0, 0xc8, 0x001f}, /* 00,1f,c8,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* CS2102_KOCOM */
@@ -676,8 +709,8 @@ static struct usb_action cs2102K_Initial[] = {
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x4c, 0x0118},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x13, 0x0120},	/* gamma 4 */
 	{0xa0, 0x38, 0x0121},
 	{0xa0, 0x59, 0x0122},
@@ -824,7 +857,7 @@ static struct usb_action cs2102K_Initial[] = {
 	{0xa0, 0x00, 0x01a7},
 	{0xa0, 0x04, 0x01a7},
 	{0xa0, 0x00, 0x01a7},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action cs2102K_InitialScale[] = {
@@ -936,8 +969,8 @@ static struct usb_action cs2102K_InitialScale[] = {
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x4c, 0x0118},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x13, 0x0120},	/* gamma 4 */
 	{0xa0, 0x38, 0x0121},
 	{0xa0, 0x59, 0x0122},
@@ -1137,8 +1170,8 @@ static struct usb_action cs2102K_InitialScale[] = {
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x4c, 0x0118},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x13, 0x0120},	/* gamma 4 */
 	{0xa0, 0x38, 0x0121},
 	{0xa0, 0x59, 0x0122},
@@ -1401,7 +1434,7 @@ static struct usb_action cs2102K_InitialScale[] = {
 	{0xa0, 0x04, 0x01a7},
 	{0xa0, 0x00, 0x01a7},
 	{0xa0, 0x04, 0x01a7},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action gc0305_Initial[] = {		/* 640x480 */
@@ -1466,7 +1499,7 @@ static struct usb_action gc0305_Initial[] = {		/* 640x480 */
 	{0xa0, 0x40, 0x0117},	/* 01,17,40,cc */
 	{0xa0, 0x52, 0x0118},	/* 01,18,52,cc */
 	{0xa0, 0x03, 0x0113},	/* 01,13,03,cc */
-	{0,0,0}
+	{}
 };
 static struct usb_action gc0305_InitialScale[] = {	/* 320x240 */
 	{0xa0, 0x01, 0x0000},	/* 00,00,01,cc */
@@ -1529,7 +1562,7 @@ static struct usb_action gc0305_InitialScale[] = {	/* 320x240 */
 	{0xa0, 0x40, 0x0117},	/* 01,17,40,cc */
 	{0xa0, 0x52, 0x0118},	/* 01,18,52,cc */
 	{0xa0, 0x03, 0x0113},	/* 01,13,03,cc */
-	{0,0,0}
+	{}
 };
 static struct usb_action gc0305_50HZ[] = {
 	{0xaa, 0x82, 0x0000},	/* 00,82,00,aa */
@@ -1552,7 +1585,7 @@ static struct usb_action gc0305_50HZ[] = {
 	{0xa0, 0x60, 0x011d},	/* 01,1d,60,cc */
 	{0xa0, 0x42, 0x0180},	/* 01,80,42,cc */
 /*	{0xa0, 0x85, 0x018d},	 * 01,8d,85,cc *	 * if 640x480 */
-	{0,0,0}
+	{}
 };
 static struct usb_action gc0305_60HZ[] = {
 	{0xaa, 0x82, 0x0000},	/* 00,82,00,aa */
@@ -1575,7 +1608,7 @@ static struct usb_action gc0305_60HZ[] = {
 	{0xa0, 0x60, 0x011d},	/* 01,1d,60,cc */
 	{0xa0, 0x42, 0x0180},	/* 01,80,42,cc */
 	{0xa0, 0x80, 0x018d},	/* 01,8d,80,cc */
-	{0,0,0}
+	{}
 };
 
 static struct usb_action gc0305_NoFliker[] = {
@@ -1598,7 +1631,7 @@ static struct usb_action gc0305_NoFliker[] = {
 	{0xa0, 0x60, 0x011d},	/* 01,1d,60,cc */
 	{0xa0, 0x03, 0x0180},	/* 01,80,03,cc */
 	{0xa0, 0x80, 0x018d},	/* 01,8d,80,cc */
-	{0,0,0}
+	{}
 };
 
 /* play poker with registers at your own risk !! */
@@ -1647,11 +1680,11 @@ static struct usb_action hdcs2020xx_Initial[] = {
 	{0xa1, 0x01, 0x0002},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x04, 0x01c6},
+	{0xa0, 0x04, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x07, 0x01cb},
+	{0xa0, 0x07, 0x01cb},	/* sharpness- */
 	{0xa0, 0x11, 0x0120},	/* gamma ~4 */
 	{0xa0, 0x37, 0x0121},
 	{0xa0, 0x58, 0x0122},
@@ -1744,7 +1777,7 @@ static struct usb_action hdcs2020xx_Initial[] = {
 	{0xa1, 0x01, 0x0118},
 /*	{0xa0, 0x02, 0x0008}, */
 	{0xa0, 0x00, 0x0007},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action hdcs2020xx_InitialScale[] = {
@@ -1792,11 +1825,11 @@ static struct usb_action hdcs2020xx_InitialScale[] = {
 	{0xa1, 0x01, 0x0002},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x04, 0x01c6},
+	{0xa0, 0x04, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x07, 0x01cb},
+	{0xa0, 0x07, 0x01cb},	/* sharpness- */
 	{0xa0, 0x11, 0x0120},	/* gamma ~4*/
 	{0xa0, 0x37, 0x0121},
 	{0xa0, 0x58, 0x0122},
@@ -1887,7 +1920,7 @@ static struct usb_action hdcs2020xx_InitialScale[] = {
 /*	{0xa0, 0x02, 0x0008}, */
 	{0xa0, 0x00, 0x0007},
 /*	{0xa0, 0x18, 0x00fe}, */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action hdcs2020xb_Initial[] = {
 	{0xa0, 0x01, 0x0000},
@@ -1942,11 +1975,11 @@ static struct usb_action hdcs2020xb_Initial[] = {
 	{0xa0, 0x40, 0x0118},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x13, 0x0120},	/* gamma 4 */
 	{0xa0, 0x38, 0x0121},
 	{0xa0, 0x59, 0x0122},
@@ -2019,7 +2052,7 @@ static struct usb_action hdcs2020xb_Initial[] = {
 	{0xa0, 0x40, 0x0116},
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
-	{0, 0, 0}
+	{}
 };
 static struct usb_action hdcs2020xb_InitialScale[] = {
 	{0xa0, 0x01, 0x0000},
@@ -2072,11 +2105,11 @@ static struct usb_action hdcs2020xb_InitialScale[] = {
 	{0xa0, 0x40, 0x0118},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x13, 0x0120},	/* gamma 4 */
 	{0xa0, 0x38, 0x0121},
 	{0xa0, 0x59, 0x0122},
@@ -2147,7 +2180,7 @@ static struct usb_action hdcs2020xb_InitialScale[] = {
 	{0xa0, 0x40, 0x0116},
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
-	{0, 0, 0}
+	{}
 };
 static struct usb_action hdcs2020b_50HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -2168,7 +2201,7 @@ static struct usb_action hdcs2020b_50HZ[] = {
 	{0xa0, 0x05, 0x001d}, /* 00,1d,05,cc */
 	{0xa0, 0x1a, 0x001e}, /* 00,1e,1a,cc */
 	{0xa0, 0x2f, 0x001f}, /* 00,1f,2f,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action hdcs2020b_60HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -2189,7 +2222,7 @@ static struct usb_action hdcs2020b_60HZ[] = {
 	{0xa0, 0x04, 0x001d}, /* 00,1d,04,cc */
 	{0xa0, 0x18, 0x001e}, /* 00,1e,18,cc */
 	{0xa0, 0x2c, 0x001f}, /* 00,1f,2c,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action hdcs2020b_NoFliker[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -2210,7 +2243,7 @@ static struct usb_action hdcs2020b_NoFliker[] = {
 	{0xa0, 0x04, 0x001d}, /* 00,1d,04,cc */
 	{0xa0, 0x17, 0x001e}, /* 00,1e,17,cc */
 	{0xa0, 0x2a, 0x001f}, /* 00,1f,2a,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action hv7131bxx_Initial[] = {
@@ -2266,11 +2299,11 @@ static struct usb_action hv7131bxx_Initial[] = {
 
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x50, 0x010a},	/* matrix */
 	{0xa0, 0xf8, 0x010b},
@@ -2318,7 +2351,7 @@ static struct usb_action hv7131bxx_Initial[] = {
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
 /*	{0xa0, 0x02, 0x0008}, */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action hv7131bxx_InitialScale[] = {
@@ -2373,11 +2406,11 @@ static struct usb_action hv7131bxx_InitialScale[] = {
 	{0xa1, 0x01, 0x0096},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x50, 0x010a},	/* matrix */
 	{0xa0, 0xf8, 0x010b},
@@ -2424,7 +2457,7 @@ static struct usb_action hv7131bxx_InitialScale[] = {
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
 /*	{0xa0, 0x02, 0x0008}, */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action hv7131cxx_Initial[] = {
@@ -2478,11 +2511,11 @@ static struct usb_action hv7131cxx_Initial[] = {
 
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x60, 0x010a},	/* matrix */
 	{0xa0, 0xf0, 0x010b},
@@ -2518,7 +2551,7 @@ static struct usb_action hv7131cxx_Initial[] = {
 	{0xa0, 0x40, 0x0180},
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action hv7131cxx_InitialScale[] = {
@@ -2577,11 +2610,11 @@ static struct usb_action hv7131cxx_InitialScale[] = {
 
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x60, 0x010a},	/* matrix */
 	{0xa0, 0xf0, 0x010b},
@@ -2619,7 +2652,7 @@ static struct usb_action hv7131cxx_InitialScale[] = {
 	{0xa0, 0x40, 0x0180},
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action icm105axx_Initial[] = {
@@ -2743,11 +2776,11 @@ static struct usb_action icm105axx_Initial[] = {
 	{0xa1, 0x01, 0x0008},
 
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x52, 0x010a},	/* matrix */
 	{0xa0, 0xf7, 0x010b},
 	{0xa0, 0xf7, 0x010c},
@@ -2796,7 +2829,7 @@ static struct usb_action icm105axx_Initial[] = {
 	{0xa0, 0x40, 0x0116},
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action icm105axx_InitialScale[] = {
@@ -2924,11 +2957,11 @@ static struct usb_action icm105axx_InitialScale[] = {
 	{0xa1, 0x01, 0x0008},
 
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x52, 0x010a},	/* matrix */
 	{0xa0, 0xf7, 0x010b},
@@ -2976,7 +3009,7 @@ static struct usb_action icm105axx_InitialScale[] = {
 	{0xa0, 0x40, 0x0116},
 	{0xa0, 0x40, 0x0117},
 	{0xa0, 0x40, 0x0118},
-	{0, 0, 0}
+	{}
 };
 static struct usb_action icm105a_50HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -3007,7 +3040,7 @@ static struct usb_action icm105a_50HZ[] = {
 	{0xa0, 0xd8, 0x001e}, /* 00,1e,d8,cc */
 	{0xa0, 0xea, 0x001f}, /* 00,1f,ea,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action icm105a_50HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -3040,7 +3073,7 @@ static struct usb_action icm105a_50HZScale[] = {
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x00, 0x01a7}, /* 01,a7,00,cc */
 	{0xa0, 0xc0, 0x01a8}, /* 01,a8,c0,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action icm105a_60HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -3071,7 +3104,7 @@ static struct usb_action icm105a_60HZ[] = {
 	{0xa0, 0xd4, 0x001e}, /* 00,1e,d4,cc */
 	{0xa0, 0xe8, 0x001f}, /* 00,1f,e8,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action icm105a_60HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -3104,7 +3137,7 @@ static struct usb_action icm105a_60HZScale[] = {
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x00, 0x01a7}, /* 01,a7,00,cc */
 	{0xa0, 0xc0, 0x01a8}, /* 01,a8,c0,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action icm105a_NoFliker[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -3135,7 +3168,7 @@ static struct usb_action icm105a_NoFliker[] = {
 	{0xa0, 0xd4, 0x001e}, /* 00,1e,d4,cc */
 	{0xa0, 0xe8, 0x001f}, /* 00,1f,e8,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action icm105a_NoFlikerScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -3168,7 +3201,7 @@ static struct usb_action icm105a_NoFlikerScale[] = {
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x00, 0x01a7}, /* 01,a7,00,cc */
 	{0xa0, 0xc0, 0x01a8}, /* 01,a8,c0,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_InitialScale[] = {
@@ -3288,7 +3321,7 @@ static struct usb_action MC501CB_InitialScale[] = {
 	{0xaa, 0x36, 0x001d}, /* 00,36,1D,aa */
 	{0xaa, 0x37, 0x004c}, /* 00,37,4C,aa */
 	{0xaa, 0x3b, 0x001d}, /* 00,3B,1D,aa */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_Initial[] = {	 /* 320x240 */
@@ -3407,7 +3440,7 @@ static struct usb_action MC501CB_Initial[] = {	 /* 320x240 */
 	{0xaa, 0x36, 0x001d}, /* 00,36,1D,aa */
 	{0xaa, 0x37, 0x004c}, /* 00,37,4C,aa */
 	{0xaa, 0x3b, 0x001d}, /* 00,3B,1D,aa */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_50HZ[] = {
@@ -3424,7 +3457,7 @@ static struct usb_action MC501CB_50HZ[] = {
 	{0xaa, 0x36, 0x003a}, /* 00,36,3A,aa */
 	{0xaa, 0x37, 0x0098}, /* 00,37,98,aa */
 	{0xaa, 0x3b, 0x003a}, /* 00,3B,3A,aa */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_50HZScale[] = {
@@ -3441,7 +3474,7 @@ static struct usb_action MC501CB_50HZScale[] = {
 	{0xaa, 0x36, 0x0018}, /* 00,36,18,aa */
 	{0xaa, 0x37, 0x006a}, /* 00,37,6A,aa */
 	{0xaa, 0x3d, 0x0018}, /* 00,3D,18,aa */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_60HZ[] = {
@@ -3458,7 +3491,7 @@ static struct usb_action MC501CB_60HZ[] = {
 	{0xaa, 0x36, 0x0030}, /* 00,36,30,aa */
 	{0xaa, 0x37, 0x00d4}, /* 00,37,D4,aa */
 	{0xaa, 0x3d, 0x0030}, /* 00,3D,30,aa */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_60HZScale[] = {
@@ -3475,7 +3508,7 @@ static struct usb_action MC501CB_60HZScale[] = {
 	{0xaa, 0x36, 0x0018}, /* 00,36,18,aa */
 	{0xaa, 0x37, 0x006a}, /* 00,37,6A,aa */
 	{0xaa, 0x3d, 0x0018}, /* 00,3D,18,aa */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_NoFliker[] = {
@@ -3492,7 +3525,7 @@ static struct usb_action MC501CB_NoFliker[] = {
 	{0xaa, 0x36, 0x0030}, /* 00,36,30,aa */
 	{0xaa, 0x37, 0x00d4}, /* 00,37,D4,aa */
 	{0xaa, 0x3d, 0x0030}, /* 00,3D,30,aa */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action MC501CB_NoFlikerScale[] = {
@@ -3504,7 +3537,7 @@ static struct usb_action MC501CB_NoFlikerScale[] = {
 	{0xaa, 0x3e, 0x00d4}, /* 00,3E,D4,aa */
 	{0xaa, 0x3b, 0x0030}, /* 00,3B,30,aa */
 	{0xaa, 0x3c, 0x00d4}, /* 00,3C,D4,aa */
-	{0, 0, 0}
+	{}
 };
 
 /* from zs211.inf - HKR,%OV7620%,Initial - 640x480 */
@@ -3575,7 +3608,7 @@ static struct usb_action OV7620_mode0[] = {
 	{0xa0, 0x40, 0x011d}, /* 01,1d,40,cc */
 	{0xa0, 0x02, 0x0180}, /* 01,80,02,cc */
 	{0xa0, 0x50, 0x01a8}, /* 01,a8,50,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* from zs211.inf - HKR,%OV7620%,InitialScale - 320x240 */
@@ -3646,7 +3679,7 @@ static struct usb_action OV7620_mode1[] = {
 	{0xa0, 0x50, 0x011d}, /* 01,1d,50,cc */
 	{0xa0, 0x02, 0x0180}, /* 01,80,02,cc */
 	{0xa0, 0x50, 0x01a8}, /* 01,a8,50,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* from zs211.inf - HKR,%OV7620%\AE,50HZ */
@@ -3665,7 +3698,7 @@ static struct usb_action OV7620_50HZ[] = {
 	{0xaa, 0x10, 0x0082},	/* 00,10,82,aa */
 	{0xaa, 0x76, 0x0003},	/* 00,76,03,aa */
 /*	{0xa0, 0x40, 0x0002},	 * 00,02,40,cc - if mode0 (640x480) */
-	{0, 0, 0}
+	{}
 };
 
 /* from zs211.inf - HKR,%OV7620%\AE,60HZ */
@@ -3687,7 +3720,7 @@ static struct usb_action OV7620_60HZ[] = {
 /* ?? in gspca v1, it was
 	{0xa0, 0x00, 0x0039},  * 00,00,00,dd *
 	{0xa1, 0x01, 0x0037},		*/
-	{0, 0, 0}
+	{}
 };
 
 /* from zs211.inf - HKR,%OV7620%\AE,NoFliker */
@@ -3707,7 +3740,7 @@ static struct usb_action OV7620_NoFliker[] = {
 /* ?? was
 	{0xa0, 0x00, 0x0039},  * 00,00,00,dd *
 	{0xa1, 0x01, 0x0037},		*/
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action ov7630c_Initial[] = {
@@ -3795,14 +3828,11 @@ static struct usb_action ov7630c_Initial[] = {
 /* 0x03, */
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
-/* 0x05, */
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
-/* 0x07, */
 	{0xa1, 0x01, 0x01c9},
-/* 0x0f, */
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x01, 0x0120},	/* gamma 2 ?*/
 	{0xa0, 0x0c, 0x0121},
 	{0xa0, 0x1f, 0x0122},
@@ -3867,7 +3897,7 @@ static struct usb_action ov7630c_Initial[] = {
 	{0xaa, 0x13, 0x0083},	/* 40 */
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action ov7630c_InitialScale[] = {
@@ -3954,14 +3984,11 @@ static struct usb_action ov7630c_InitialScale[] = {
 
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
-
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
-
 	{0xa1, 0x01, 0x01c9},
-
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x16, 0x0120},	/* gamma ~4 */
 	{0xa0, 0x3a, 0x0121},
 	{0xa0, 0x5b, 0x0122},
@@ -4027,7 +4054,7 @@ static struct usb_action ov7630c_InitialScale[] = {
 
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action pas106b_Initial_com[] = {
@@ -4041,7 +4068,7 @@ static struct usb_action pas106b_Initial_com[] = {
 	{0xa0, 0x03, 0x003a},
 	{0xa0, 0x0c, 0x003b},
 	{0xa0, 0x04, 0x0038},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action pas106b_Initial[] = {	/* 176x144 */
@@ -4099,10 +4126,8 @@ static struct usb_action pas106b_Initial[] = {	/* 176x144 */
 	{0xa0, 0x08, 0x0301},	/* EEPROMAccess */
 /* JPEG control */
 	{0xa0, 0x03, 0x0008},	/* ClockSetting */
-/* Unknown */
-	{0xa0, 0x08, 0x01c6},
-/* Sharpness */
-	{0xa0, 0x0f, 0x01cb},	/* Sharpness05 */
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 /* Other registers */
 	{0xa0, 0x0d, 0x0100},	/* OperationMode */
 /* Auto exposure and white balance */
@@ -4113,9 +4138,8 @@ static struct usb_action pas106b_Initial[] = {	/* 176x144 */
 	{0xa0, 0x08, 0x0301},	/* EEPROMAccess */
 /* JPEG control */
 	{0xa0, 0x03, 0x0008},	/* ClockSetting */
-/* Sharpness */
-	{0xa0, 0x08, 0x01c6},	/* Sharpness00 */
-	{0xa0, 0x0f, 0x01cb},	/* Sharpness05 */
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x58, 0x010a},	/* matrix */
 	{0xa0, 0xf4, 0x010b},
@@ -4162,7 +4186,7 @@ static struct usb_action pas106b_Initial[] = {	/* 176x144 */
 	{0xa0, 0x40, 0x0116},	/* RGain */
 	{0xa0, 0x40, 0x0117},	/* GGain */
 	{0xa0, 0x40, 0x0118},	/* BGain */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action pas106b_InitialScale[] = {	/* 352x288 */
@@ -4221,10 +4245,8 @@ static struct usb_action pas106b_InitialScale[] = {	/* 352x288 */
 	{0xa0, 0x08, 0x0301},	/* EEPROMAccess */
 /* JPEG control */
 	{0xa0, 0x03, 0x0008},	/* ClockSetting */
-/* Unknown */
-	{0xa0, 0x08, 0x01c6},
-/* Sharpness */
-	{0xa0, 0x0f, 0x01cb},	/* Sharpness05 */
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 /* Other registers */
 	{0xa0, 0x0d, 0x0100},	/* OperationMode */
 /* Auto exposure and white balance */
@@ -4235,9 +4257,8 @@ static struct usb_action pas106b_InitialScale[] = {	/* 352x288 */
 	{0xa0, 0x08, 0x0301},	/* EEPROMAccess */
 /* JPEG control */
 	{0xa0, 0x03, 0x0008},	/* ClockSetting */
-/* Sharpness */
-	{0xa0, 0x08, 0x01c6},	/* Sharpness00 */
-	{0xa0, 0x0f, 0x01cb},	/* Sharpness05 */
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x58, 0x010a},	/* matrix */
 	{0xa0, 0xf4, 0x010b},
@@ -4289,7 +4310,7 @@ static struct usb_action pas106b_InitialScale[] = {	/* 352x288 */
 
 	{0xa0, 0x00, 0x0007},	/* AutoCorrectEnable */
 	{0xa0, 0xff, 0x0018},	/* Frame adjust */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pas106b_50HZ[] = {
 	{0xa0, 0x00, 0x0190}, /* 01,90,00,cc */
@@ -4305,7 +4326,7 @@ static struct usb_action pas106b_50HZ[] = {
 	{0xaa, 0x05, 0x0002}, /* 00,05,02,aa */
 	{0xaa, 0x07, 0x001c}, /* 00,07,1c,aa */
 	{0xa0, 0x04, 0x01a9}, /* 01,a9,04,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pas106b_60HZ[] = {
 	{0xa0, 0x00, 0x0190}, /* 01,90,00,cc */
@@ -4321,7 +4342,7 @@ static struct usb_action pas106b_60HZ[] = {
 	{0xaa, 0x05, 0x0001}, /* 00,05,01,aa */
 	{0xaa, 0x07, 0x00c4}, /* 00,07,c4,aa */
 	{0xa0, 0x04, 0x01a9}, /* 01,a9,04,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pas106b_NoFliker[] = {
 	{0xa0, 0x00, 0x0190}, /* 01,90,00,cc */
@@ -4337,10 +4358,9 @@ static struct usb_action pas106b_NoFliker[] = {
 	{0xaa, 0x05, 0x0001}, /* 00,05,01,aa */
 	{0xaa, 0x07, 0x0030}, /* 00,07,30,aa */
 	{0xa0, 0x00, 0x01a9}, /* 01,a9,00,cc */
-	{0, 0, 0}
+	{}
 };
 
-/* Aurelien setting from snoop */
 static struct usb_action pb03303x_Initial[] = {
 	{0xa0, 0x01, 0x0000},
 	{0xa0, 0x03, 0x0008},
@@ -4411,11 +4431,11 @@ static struct usb_action pb03303x_Initial[] = {
 
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x13, 0x0120},	/* gamma 4 */
 	{0xa0, 0x38, 0x0121},
 	{0xa0, 0x59, 0x0122},
@@ -4484,7 +4504,7 @@ static struct usb_action pb03303x_Initial[] = {
 	{0xa0, 0x40, 0x0180},
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action pb03303x_InitialScale[] = {
@@ -4559,11 +4579,11 @@ static struct usb_action pb03303x_InitialScale[] = {
 
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x13, 0x0120},	/* gamma 4 */
 	{0xa0, 0x38, 0x0121},
@@ -4633,7 +4653,7 @@ static struct usb_action pb03303x_InitialScale[] = {
 	{0xa0, 0x40, 0x0180},
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pb0330xx_Initial[] = {
 	{0xa1, 0x01, 0x0008},
@@ -4701,11 +4721,11 @@ static struct usb_action pb0330xx_Initial[] = {
 	{0xa0, 0x50, 0x0112},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x50, 0x010a},	/* matrix */
 	{0xa0, 0xf8, 0x010b},
@@ -4747,7 +4767,7 @@ static struct usb_action pb0330xx_Initial[] = {
 	{0xa1, 0x01, 0x0007},
 /*	{0xa0, 0x30, 0x0007}, */
 /*	{0xa0, 0x00, 0x0007}, */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action pb0330xx_InitialScale[] = {
@@ -4816,11 +4836,11 @@ static struct usb_action pb0330xx_InitialScale[] = {
 	{0xa0, 0x50, 0x0112},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x50, 0x010a},	/* matrix */
 	{0xa0, 0xf8, 0x010b},
@@ -4861,7 +4881,7 @@ static struct usb_action pb0330xx_InitialScale[] = {
 	{0xa1, 0x01, 0x0007},
 /*	{0xa0, 0x30, 0x0007}, */
 /*	{0xa0, 0x00, 0x0007}, */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pb0330_50HZ[] = {
 	{0xa0, 0x00, 0x0190}, /* 01,90,00,cc */
@@ -4877,7 +4897,7 @@ static struct usb_action pb0330_50HZ[] = {
 	{0xa0, 0x68, 0x001d}, /* 00,1d,68,cc */
 	{0xa0, 0x90, 0x001e}, /* 00,1e,90,cc */
 	{0xa0, 0xc8, 0x001f}, /* 00,1f,c8,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pb0330_50HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -4894,7 +4914,7 @@ static struct usb_action pb0330_50HZScale[] = {
 	{0xa0, 0xe5, 0x001d}, /* 00,1d,e5,cc */
 	{0xa0, 0xf0, 0x001e}, /* 00,1e,f0,cc */
 	{0xa0, 0xf8, 0x001f}, /* 00,1f,f8,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pb0330_60HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -4911,7 +4931,7 @@ static struct usb_action pb0330_60HZ[] = {
 	{0xa0, 0x43, 0x001d}, /* 00,1d,43,cc */
 	{0xa0, 0x50, 0x001e}, /* 00,1e,50,cc */
 	{0xa0, 0x90, 0x001f}, /* 00,1f,90,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pb0330_60HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -4928,7 +4948,7 @@ static struct usb_action pb0330_60HZScale[] = {
 	{0xa0, 0x41, 0x001d}, /* 00,1d,41,cc */
 	{0xa0, 0x50, 0x001e}, /* 00,1e,50,cc */
 	{0xa0, 0x90, 0x001f}, /* 00,1f,90,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pb0330_NoFliker[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -4945,7 +4965,7 @@ static struct usb_action pb0330_NoFliker[] = {
 	{0xa0, 0x09, 0x001d}, /* 00,1d,09,cc */
 	{0xa0, 0x40, 0x001e}, /* 00,1e,40,cc */
 	{0xa0, 0x90, 0x001f}, /* 00,1f,90,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action pb0330_NoFlikerScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -4962,7 +4982,7 @@ static struct usb_action pb0330_NoFlikerScale[] = {
 	{0xa0, 0x09, 0x001d}, /* 00,1d,09,cc */
 	{0xa0, 0x40, 0x001e}, /* 00,1e,40,cc */
 	{0xa0, 0x90, 0x001f}, /* 00,1f,90,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* from oem9.inf - HKR,%PO2030%,Initial - 640x480 - (close to CS2102) */
@@ -5039,7 +5059,7 @@ static struct usb_action PO2030_mode0[] = {
 	{0xa0, 0x08, 0x0301}, /* 03,01,08,cc */
 	{0xa0, 0x7a, 0x0116}, /* 01,16,7a,cc */
 	{0xa0, 0x4a, 0x0118}, /* 01,18,4a,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* from oem9.inf - HKR,%PO2030%,InitialScale - 320x240 */
@@ -5116,7 +5136,7 @@ static struct usb_action PO2030_mode1[] = {
 	{0xa0, 0x08, 0x0301}, /* 03,01,08,cc */
 	{0xa0, 0x7a, 0x0116}, /* 01,16,7a,cc */
 	{0xa0, 0x4a, 0x0118}, /* 01,18,4a,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action PO2030_50HZ[] = {
@@ -5138,7 +5158,7 @@ static struct usb_action PO2030_50HZ[] = {
 	{0xa0, 0x88, 0x018d}, /* 01,8d,88,cc */
 	{0xa0, 0x58, 0x011d}, /* 01,1d,58,cc */
 	{0xa0, 0x42, 0x0180}, /* 01,80,42,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action PO2030_60HZ[] = {
@@ -5160,7 +5180,7 @@ static struct usb_action PO2030_60HZ[] = {
 	{0xa0, 0x88, 0x018d}, /* 01,8d,88,cc */	/* win: 01,8d,80 */
 	{0xa0, 0x58, 0x011d}, /* 01,1d,58,cc */
 	{0xa0, 0x42, 0x0180}, /* 01,80,42,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action PO2030_NoFliker[] = {
@@ -5171,7 +5191,7 @@ static struct usb_action PO2030_NoFliker[] = {
 	{0xaa, 0x1c, 0x0078}, /* 00,1c,78,aa */
 	{0xaa, 0x46, 0x0000}, /* 00,46,00,aa */
 	{0xaa, 0x15, 0x0000}, /* 00,15,00,aa */
-	{0, 0, 0}
+	{}
 };
 
 /* TEST */
@@ -5302,8 +5322,8 @@ static struct usb_action tas5130CK_Initial[] = {
 	{0xa0, 0x03, 0x0111},
 	{0xa0, 0x51, 0x0112},
 	{0xa0, 0x03, 0x0008},
-	{0xa0, 0x08, 0x01c6},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x38, 0x0120},	/* gamma > 5 */
 	{0xa0, 0x51, 0x0121},
 	{0xa0, 0x6e, 0x0122},
@@ -5375,7 +5395,7 @@ static struct usb_action tas5130CK_Initial[] = {
 	{0xa0, 0x15, 0x01ae},
 	{0xa0, 0x40, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action tas5130CK_InitialScale[] = {
@@ -5505,8 +5525,8 @@ static struct usb_action tas5130CK_InitialScale[] = {
 	{0xa0, 0x03, 0x0111},
 	{0xa0, 0x51, 0x0112},
 	{0xa0, 0x03, 0x0008},
-	{0xa0, 0x08, 0x01c6},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 	{0xa0, 0x38, 0x0120},	/* gamma > 5 */
 	{0xa0, 0x51, 0x0121},
 	{0xa0, 0x6e, 0x0122},
@@ -5583,7 +5603,7 @@ static struct usb_action tas5130CK_InitialScale[] = {
 	{0xa0, 0x02, 0x0008},
 	{0xa0, 0x00, 0x0007},
 	{0xa0, 0x03, 0x0008},
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action tas5130cxx_Initial[] = {
@@ -5625,11 +5645,11 @@ static struct usb_action tas5130cxx_Initial[] = {
 	{0xa1, 0x01, 0x0002},
 	{0xa1, 0x01, 0x0008},
 	{0xa0, 0x03, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x68, 0x010a},	/* matrix */
 	{0xa0, 0xec, 0x010b},
@@ -5673,7 +5693,7 @@ static struct usb_action tas5130cxx_Initial[] = {
 	{0xa0, 0x40, 0x0180},
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 static struct usb_action tas5130cxx_InitialScale[] = {
 	{0xa0, 0x01, 0x0000},
@@ -5718,11 +5738,11 @@ static struct usb_action tas5130cxx_InitialScale[] = {
 
 	{0xa0, 0x03, 0x0008},
 	{0xa1, 0x01, 0x0008},	/* clock ? */
-	{0xa0, 0x08, 0x01c6},
+	{0xa0, 0x08, 0x01c6},	/* sharpness+ */
 	{0xa1, 0x01, 0x01c8},
 	{0xa1, 0x01, 0x01c9},
 	{0xa1, 0x01, 0x01ca},
-	{0xa0, 0x0f, 0x01cb},
+	{0xa0, 0x0f, 0x01cb},	/* sharpness- */
 
 	{0xa0, 0x68, 0x010a},	/* matrix */
 	{0xa0, 0xec, 0x010b},
@@ -5763,7 +5783,7 @@ static struct usb_action tas5130cxx_InitialScale[] = {
 	{0xa0, 0x40, 0x0180},
 	{0xa1, 0x01, 0x0180},
 	{0xa0, 0x42, 0x0180},
-	{0, 0, 0}
+	{}
 };
 static struct usb_action tas5130cxx_50HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -5786,7 +5806,7 @@ static struct usb_action tas5130cxx_50HZ[] = {
 	{0xa0, 0xea, 0x001f}, /* 00,1f,ea,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x03, 0x009f}, /* 00,9f,03,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action tas5130cxx_50HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -5809,7 +5829,7 @@ static struct usb_action tas5130cxx_50HZScale[] = {
 	{0xa0, 0xf8, 0x001f}, /* 00,1f,f8,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x03, 0x009f}, /* 00,9f,03,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action tas5130cxx_60HZ[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -5832,7 +5852,7 @@ static struct usb_action tas5130cxx_60HZ[] = {
 	{0xa0, 0xe0, 0x001f}, /* 00,1f,e0,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x03, 0x009f}, /* 00,9f,03,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action tas5130cxx_60HZScale[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -5855,7 +5875,7 @@ static struct usb_action tas5130cxx_60HZScale[] = {
 	{0xa0, 0xe0, 0x001f}, /* 00,1f,e0,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x03, 0x009f}, /* 00,9f,03,cc */
-	{0, 0, 0}
+	{}
 };
 static struct usb_action tas5130cxx_NoFliker[] = {
 	{0xa0, 0x00, 0x0019}, /* 00,19,00,cc */
@@ -5878,7 +5898,7 @@ static struct usb_action tas5130cxx_NoFliker[] = {
 	{0xa0, 0xe0, 0x001f}, /* 00,1f,e0,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x02, 0x009f}, /* 00,9f,02,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action tas5130cxx_NoFlikerScale[] = {
@@ -5902,7 +5922,7 @@ static struct usb_action tas5130cxx_NoFlikerScale[] = {
 	{0xa0, 0xe0, 0x001f}, /* 00,1f,e0,cc */
 	{0xa0, 0xff, 0x0020}, /* 00,20,ff,cc */
 	{0xa0, 0x02, 0x009f}, /* 00,9f,02,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action tas5130c_vf0250_Initial[] = {
@@ -5966,7 +5986,7 @@ static struct usb_action tas5130c_vf0250_Initial[] = {
 	{0xa0, 0x60, 0x01a8},		/* 01,a8,60,cc, */
 	{0xa0, 0x61, 0x0116},		/* 01,16,61,cc, */
 	{0xa0, 0x65, 0x0118},		/* 01,18,65,cc */
-	{0, 0, 0}
+	{}
 };
 
 static struct usb_action tas5130c_vf0250_InitialScale[] = {
@@ -6030,7 +6050,7 @@ static struct usb_action tas5130c_vf0250_InitialScale[] = {
 	{0xa0, 0x60, 0x01a8},		/* 01,a8,60,cc, */
 	{0xa0, 0x61, 0x0116},		/* 01,16,61,cc, */
 	{0xa0, 0x65, 0x0118},		/* 01,18,65,cc */
-	{0, 0, 0}
+	{}
 };
 /* "50HZ" light frequency banding filter */
 static struct usb_action tas5130c_vf0250_50HZ[] = {
@@ -6054,7 +6074,7 @@ static struct usb_action tas5130c_vf0250_50HZ[] = {
 	{0xa0, 0x58, 0x011d},		/* 01,1d,58,cc, */
 	{0xa0, 0x42, 0x0180},		/* 01,80,42,cc, */
 	{0xa0, 0x78, 0x018d},		/* 01,8d,78,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* "50HZScale" light frequency banding filter */
@@ -6079,7 +6099,7 @@ static struct usb_action tas5130c_vf0250_50HZScale[] = {
 	{0xa0, 0x58, 0x011d},		/* 01,1d,58,cc, */
 	{0xa0, 0x42, 0x0180},		/* 01,80,42,cc, */
 	{0xa0, 0x78, 0x018d},		/* 01,8d,78,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* "60HZ" light frequency banding filter */
@@ -6104,7 +6124,7 @@ static struct usb_action tas5130c_vf0250_60HZ[] = {
 	{0xa0, 0x58, 0x011d},		/* 01,1d,58,cc, */
 	{0xa0, 0x42, 0x0180},		/* 01,80,42,cc, */
 	{0xa0, 0x78, 0x018d},		/* 01,8d,78,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* "60HZScale" light frequency banding ilter */
@@ -6129,7 +6149,7 @@ static struct usb_action tas5130c_vf0250_60HZScale[] = {
 	{0xa0, 0x58, 0x011d},		/* 01,d,58,cc, */
 	{0xa0, 0x42, 0x0180},		/* 01,80,42,cc, */
 	{0xa0, 0x78, 0x018d},		/* 01,d,78,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* "NoFliker" light frequency banding flter */
@@ -6152,7 +6172,7 @@ static struct usb_action tas5130c_vf0250_NoFliker[] = {
 	{0xa0, 0xff, 0x0020},		/* 00,20,ff,cc, */
 	{0xa0, 0x58, 0x011d},		/* 01,1d,58,cc, */
 	{0xa0, 0x03, 0x0180},		/* 01,80,03,cc */
-	{0, 0, 0}
+	{}
 };
 
 /* "NoFlikerScale" light frequency banding filter */
@@ -6175,7 +6195,7 @@ static struct usb_action tas5130c_vf0250_NoFlikerScale[] = {
 	{0xa0, 0xff, 0x0020},		/* 00,20,ff,cc, */
 	{0xa0, 0x58, 0x011d},		/* 01,1d,58,cc, */
 	{0xa0, 0x03, 0x0180},		/* 01,80,03,cc */
-	{0, 0, 0}
+	{}
 };
 
 static void reg_r_i(struct usb_device *dev,
@@ -6325,7 +6345,7 @@ static void setbrightness(struct gspca_dev *gspca_dev)
 	case SENSOR_PO2030:
 		return;
 	}
-/*fixme: is it really 011d 018d for all other sensors? */
+/*fixme: is it really write to 011d and 018d for all other sensors? */
 	brightness = sd->brightness;
 	reg_w(gspca_dev->dev, brightness, 0x011d);
 	if (brightness < 0x70)
@@ -6348,20 +6368,7 @@ static void setsharpness(struct gspca_dev *gspca_dev)
 		{0x10, 0x1e}
 	};
 
-	switch (sd->sensor) {
-	case SENSOR_GC0305:
-		sharpness = 3;
-		break;
-	case SENSOR_OV7620:
-		sharpness = 2;
-		break;
-	case SENSOR_PO2030:
-		sharpness = 0;
-		break;
-	default:
-		return;
-	}
-/*fixme: sharpness set by V4L2_CID_SATURATION?*/
+	sharpness = sd->sharpness;
 	reg_w(dev, sharpness_tb[sharpness][0], 0x01c6);
 	reg_r(dev, 0x01c8, &retbyte);
 	reg_r(dev, 0x01c9, &retbyte);
@@ -6411,7 +6418,7 @@ static void setcontrast(struct gspca_dev *gspca_dev)
 	static __u8 Tgradient_5[16] =
 		{0x37, 0x26, 0x20, 0x1a, 0x14, 0x10, 0x0e, 0x0b,
 		 0x09, 0x07, 0x06, 0x05, 0x04, 0x03, 0x02, 0x02};
-	static __u8 Tgamma_6[16] =		/* ?? was gama 5 */
+	static __u8 Tgamma_6[16] =		/* ?? was gamma 5 */
 		{0x24, 0x44, 0x64, 0x84, 0x9d, 0xb2, 0xc4, 0xd3,
 		 0xe0, 0xeb, 0xf4, 0xff, 0xff, 0xff, 0xff, 0xff};
 	static __u8 Tgradient_6[16] =
@@ -6425,7 +6432,7 @@ static void setcontrast(struct gspca_dev *gspca_dev)
 		0, Tgradient_1, Tgradient_2,
 		Tgradient_3, Tgradient_4, Tgradient_5, Tgradient_6
 	};
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 	__u8 v[16];
 #endif
 
@@ -6443,7 +6450,7 @@ static void setcontrast(struct gspca_dev *gspca_dev)
 		else if (g <= 0)
 			g = 1;
 		reg_w(dev, g, 0x0120 + i);	/* gamma */
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 		if (gspca_debug & D_CONF)
 			v[i] = g;
 #endif
@@ -6463,7 +6470,7 @@ static void setcontrast(struct gspca_dev *gspca_dev)
 				g = 1;
 		}
 		reg_w(dev, g, 0x0130 + i);	/* gradient */
-#ifdef GSPCA_DEBUG
+#ifdef VIDEO_ADV_DEBUG
 		if (gspca_debug & D_CONF)
 			v[i] = g;
 #endif
@@ -6488,7 +6495,7 @@ static void setquality(struct gspca_dev *gspca_dev)
 		return;
 	}
 /*fixme: is it really 0008 0007 0018 for all other sensors? */
-	quality = sd->qindex & 0x0f;
+	quality = sd->qindex;
 	reg_w(dev, quality, 0x0008);
 	frxt = 0x30;
 	reg_w(dev, frxt, 0x0007);
@@ -6525,25 +6532,25 @@ static int setlightfreq(struct gspca_dev *gspca_dev)
 	struct usb_action *zc3_freq;
 	static struct usb_action *freq_tb[SENSOR_MAX][6] = {
 /* SENSOR_CS2102 0 */
-		{cs2102_50HZ, cs2102_50HZScale,
-		 cs2102_60HZ, cs2102_60HZScale,
-		 cs2102_NoFliker, cs2102_NoFlikerScale},
+		{cs2102_NoFliker, cs2102_NoFlikerScale,
+		 cs2102_50HZ, cs2102_50HZScale,
+		 cs2102_60HZ, cs2102_60HZScale},
 /* SENSOR_CS2102K 1 */
-		{cs2102_50HZ, cs2102_50HZScale,
-		 cs2102_60HZ, cs2102_60HZScale,
-		 cs2102_NoFliker, cs2102_NoFlikerScale},
+		{cs2102_NoFliker, cs2102_NoFlikerScale,
+		 cs2102_50HZ, cs2102_50HZScale,
+		 cs2102_60HZ, cs2102_60HZScale},
 /* SENSOR_GC0305 2 */
-		{gc0305_50HZ, gc0305_50HZ,
-		 gc0305_60HZ, gc0305_60HZ,
-		 gc0305_NoFliker, gc0305_NoFliker},
+		{gc0305_NoFliker, gc0305_NoFliker,
+		 gc0305_50HZ, gc0305_50HZ,
+		 gc0305_60HZ, gc0305_60HZ},
 /* SENSOR_HDCS2020 3 */
 		{0, 0,
 		 0, 0,
 		 0, 0},
 /* SENSOR_HDCS2020b 4 */
-		{hdcs2020b_50HZ, hdcs2020b_50HZ,
-		 hdcs2020b_60HZ, hdcs2020b_60HZ,
-		 hdcs2020b_NoFliker, hdcs2020b_NoFliker},
+		{hdcs2020b_NoFliker, hdcs2020b_NoFliker,
+		 hdcs2020b_50HZ, hdcs2020b_50HZ,
+		 hdcs2020b_60HZ, hdcs2020b_60HZ},
 /* SENSOR_HV7131B 5 */
 		{0, 0,
 		 0, 0,
@@ -6553,66 +6560,48 @@ static int setlightfreq(struct gspca_dev *gspca_dev)
 		 0, 0,
 		 0, 0},
 /* SENSOR_ICM105A 7 */
-		{icm105a_50HZ, icm105a_50HZScale,
-		 icm105a_60HZ, icm105a_60HZScale,
-		 icm105a_NoFliker, icm105a_NoFlikerScale},
+		{icm105a_NoFliker, icm105a_NoFlikerScale,
+		 icm105a_50HZ, icm105a_50HZScale,
+		 icm105a_60HZ, icm105a_60HZScale},
 /* SENSOR_MC501CB 8 */
-		{MC501CB_50HZ, MC501CB_50HZScale,
-		 MC501CB_60HZ, MC501CB_60HZScale,
-		 MC501CB_NoFliker, MC501CB_NoFlikerScale},
+		{MC501CB_NoFliker, MC501CB_NoFlikerScale,
+		 MC501CB_50HZ, MC501CB_50HZScale,
+		 MC501CB_60HZ, MC501CB_60HZScale},
 /* SENSOR_OV7620 9 */
-		{OV7620_50HZ, OV7620_50HZ,
-		 OV7620_60HZ, OV7620_60HZ,
-		 OV7620_NoFliker, OV7620_NoFliker},
+		{OV7620_NoFliker, OV7620_NoFliker,
+		 OV7620_50HZ, OV7620_50HZ,
+		 OV7620_60HZ, OV7620_60HZ},
 /* SENSOR_OV7630C 10 */
 		{0, 0,
 		 0, 0,
 		 0, 0},
-/* SENSOR_free 11 */
-		{0, 0,
-		 0, 0,
-		 0, 0},
-/* SENSOR_PAS106 12 */
-		{pas106b_50HZ, pas106b_50HZ,
-		 pas106b_60HZ, pas106b_60HZ,
-		 pas106b_NoFliker, pas106b_NoFliker},
-/* SENSOR_PB0330 13 */
-		{pb0330_50HZ, pb0330_50HZScale,
-		 pb0330_60HZ, pb0330_60HZScale,
-		 pb0330_NoFliker, pb0330_NoFlikerScale},
-/* SENSOR_PO2030 14 */
-		{PO2030_50HZ, PO2030_50HZ,
-		 PO2030_60HZ, PO2030_60HZ,
-		 PO2030_NoFliker, PO2030_NoFliker},
-/* SENSOR_TAS5130CK 15 */
-		{tas5130cxx_50HZ, tas5130cxx_50HZScale,
-		 tas5130cxx_60HZ, tas5130cxx_60HZScale,
-		 tas5130cxx_NoFliker, tas5130cxx_NoFlikerScale},
-/* SENSOR_TAS5130CXX 16 */
-		{tas5130cxx_50HZ, tas5130cxx_50HZScale,
-		 tas5130cxx_60HZ, tas5130cxx_60HZScale,
-		 tas5130cxx_NoFliker, tas5130cxx_NoFlikerScale},
-/* SENSOR_TAS5130C_VF0250 17 */
-		{tas5130c_vf0250_50HZ, tas5130c_vf0250_50HZScale,
-		 tas5130c_vf0250_60HZ, tas5130c_vf0250_60HZScale,
-		 tas5130c_vf0250_NoFliker, tas5130c_vf0250_NoFlikerScale},
+/* SENSOR_PAS106 11 */
+		{pas106b_NoFliker, pas106b_NoFliker,
+		 pas106b_50HZ, pas106b_50HZ,
+		 pas106b_60HZ, pas106b_60HZ},
+/* SENSOR_PB0330 12 */
+		{pb0330_NoFliker, pb0330_NoFlikerScale,
+		 pb0330_50HZ, pb0330_50HZScale,
+		 pb0330_60HZ, pb0330_60HZScale},
+/* SENSOR_PO2030 13 */
+		{PO2030_NoFliker, PO2030_NoFliker,
+		 PO2030_50HZ, PO2030_50HZ,
+		 PO2030_60HZ, PO2030_60HZ},
+/* SENSOR_TAS5130CK 14 */
+		{tas5130cxx_NoFliker, tas5130cxx_NoFlikerScale,
+		 tas5130cxx_50HZ, tas5130cxx_50HZScale,
+		 tas5130cxx_60HZ, tas5130cxx_60HZScale},
+/* SENSOR_TAS5130CXX 15 */
+		{tas5130cxx_NoFliker, tas5130cxx_NoFlikerScale,
+		 tas5130cxx_50HZ, tas5130cxx_50HZScale,
+		 tas5130cxx_60HZ, tas5130cxx_60HZScale},
+/* SENSOR_TAS5130C_VF0250 16 */
+		{tas5130c_vf0250_NoFliker, tas5130c_vf0250_NoFlikerScale,
+		 tas5130c_vf0250_50HZ, tas5130c_vf0250_50HZScale,
+		 tas5130c_vf0250_60HZ, tas5130c_vf0250_60HZScale},
 	};
 
-	switch (lightfreq) {
-	case 50:
-		i = 0;
-		break;
-	case 60:
-		i = 2;
-		break;
-	default:
-		PDEBUG(D_ERR, "Invalid light freq value %d", lightfreq);
-		lightfreq = 0;		/* set to default filter value */
-		/* fall thru */
-	case 0:
-		i = 4;
-		break;
-	}
+	i = sd->lightfreq * 2;
 	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
 	if (!mode)
 		i++;			/* 640x480 */
@@ -6622,13 +6611,13 @@ static int setlightfreq(struct gspca_dev *gspca_dev)
 		switch (sd->sensor) {
 		case SENSOR_GC0305:
 			if (mode			/* if 320x240 */
-			    && lightfreq == 50)
+			    && sd->lightfreq == 1)	/* and 50Hz */
 				reg_w(gspca_dev->dev, 0x85, 0x018d);
 						/* win: 0x80, 0x018d */
 			break;
 		case SENSOR_OV7620:
 			if (!mode) {			/* if 640x480 */
-				if (lightfreq != 0)	/* 50 or 60 Hz */
+				if (sd->lightfreq != 0)	/* and 50 or 60 Hz */
 					reg_w(gspca_dev->dev, 0x40, 0x0002);
 				else
 					reg_w(gspca_dev->dev, 0x44, 0x0002);
@@ -6653,9 +6642,9 @@ static void setautogain(struct gspca_dev *gspca_dev)
 
 static void send_unknown(struct usb_device *dev, int sensor)
 {
+	reg_w(dev, 0x01, 0x0000);		/* led off */
 	switch (sensor) {
 	case SENSOR_PAS106:
-		reg_w(dev, 0x01, 0x0000);
 		reg_w(dev, 0x03, 0x003a);
 		reg_w(dev, 0x0c, 0x003b);
 		reg_w(dev, 0x08, 0x0038);
@@ -6664,7 +6653,6 @@ static void send_unknown(struct usb_device *dev, int sensor)
 	case SENSOR_OV7620:
 	case SENSOR_PB0330:
 	case SENSOR_PO2030:
-		reg_w(dev, 0x01, 0x0000);
 		reg_w(dev, 0x0d, 0x003a);
 		reg_w(dev, 0x02, 0x003b);
 		reg_w(dev, 0x00, 0x0038);
@@ -6817,7 +6805,7 @@ static int vga_3wr_probe(struct gspca_dev *gspca_dev)
 
 /*fixme: lack of 8b=b3 (11,12)-> 10, 8b=e0 (14,15,16)-> 12 found in gspcav1*/
 	reg_w(dev, 0x02, 0x0010);
-	reg_r(dev, 0x0010, &retbyte);
+	reg_r(dev, 0x10, &retbyte);
 	reg_w(dev, 0x01, 0x0000);
 	reg_w(dev, 0x00, 0x0010);
 	reg_w(dev, 0x01, 0x0001);
@@ -6964,7 +6952,7 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	int sensor;
 	__u8 bsensor;
 	int vga = 1;		/* 1: vga, 0: sif */
-	static unsigned char gamma[SENSOR_MAX] = {
+	static __u8 gamma[SENSOR_MAX] = {
 		5,	/* SENSOR_CS2102 0 */
 		5,	/* SENSOR_CS2102K 1 */
 		4,	/* SENSOR_GC0305 2 */
@@ -6976,16 +6964,16 @@ static int sd_config(struct gspca_dev *gspca_dev,
 		4,	/* SENSOR_MC501CB 8 */
 		3,	/* SENSOR_OV7620 9 */
 		4,	/* SENSOR_OV7630C 10 */
-		4,	/* SENSOR_free 11 */
-		4,	/* SENSOR_PAS106 12 */
-		4,	/* SENSOR_PB0330 13 */
-		4,	/* SENSOR_PO2030 14 */
-		4,	/* SENSOR_TAS5130CK 15 */
-		4,	/* SENSOR_TAS5130CXX 16 */
-		3,	/* SENSOR_TAS5130C_VF0250 17 */
+		4,	/* SENSOR_PAS106 11 */
+		4,	/* SENSOR_PB0330 12 */
+		4,	/* SENSOR_PO2030 13 */
+		4,	/* SENSOR_TAS5130CK 14 */
+		4,	/* SENSOR_TAS5130CXX 15 */
+		3,	/* SENSOR_TAS5130C_VF0250 16 */
 	};
 
 	/* define some sensors from the vendor/product */
+	sd->sharpness = 2;
 	switch (id->idVendor) {
 	case 0x041e:				/* Creative */
 		switch (id->idProduct) {
@@ -7055,8 +7043,9 @@ static int sd_config(struct gspca_dev *gspca_dev,
 			sd->sensor = SENSOR_ICM105A;
 			break;
 		case 0x0e:
-			PDEBUG(D_PROBE, "Find Sensor PAS202BCB");
+			PDEBUG(D_PROBE, "Find Sensor HDCS2020");
 			sd->sensor = SENSOR_HDCS2020;
+			sd->sharpness = 1;
 			break;
 		case 0x0f:
 			PDEBUG(D_PROBE, "Find Sensor PAS106");
@@ -7097,6 +7086,7 @@ static int sd_config(struct gspca_dev *gspca_dev,
 		case 0x2030:
 			PDEBUG(D_PROBE, "Find Sensor PO2030");
 			sd->sensor = SENSOR_PO2030;
+			sd->sharpness = 0;		/* from win traces */
 			break;
 		case 0x7620:
 			PDEBUG(D_PROBE, "Find Sensor OV7620");
@@ -7134,13 +7124,13 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	sd->qindex = 1;
 	sd->brightness = sd_ctrls[SD_BRIGHTNESS].qctrl.default_value;
 	sd->contrast = sd_ctrls[SD_CONTRAST].qctrl.default_value;
-	sd->autogain = sd_ctrls[SD_AUTOGAIN].qctrl.default_value;
 	sd->gamma = gamma[(int) sd->sensor];
+	sd->autogain = sd_ctrls[SD_AUTOGAIN].qctrl.default_value;
+	sd->lightfreq = sd_ctrls[SD_FREQ].qctrl.default_value;
+	sd->sharpness = sd_ctrls[SD_SHARPNESS].qctrl.default_value;
 
 	/* switch the led off */
-/*fixme: other sensors? */
-	if (sensor == 0x06 || sensor == 0x11)
-		reg_w(gspca_dev->dev, 0x01, 0x0000);
+	reg_w(gspca_dev->dev, 0x01, 0x0000);
 	return 0;
 }
 
@@ -7170,15 +7160,14 @@ static void sd_start(struct gspca_dev *gspca_dev)
 		{MC501CB_InitialScale, MC501CB_Initial},	/* 9 */
 		{OV7620_mode0, OV7620_mode1},			/* 9 */
 		{ov7630c_InitialScale, ov7630c_Initial},	/* 10 */
-		{0, 0},						/* 11 */
-		{pas106b_InitialScale, pas106b_Initial},	/* 12 */
-		{pb0330xx_InitialScale, pb0330xx_Initial},	/* 13 */
+		{pas106b_InitialScale, pas106b_Initial},	/* 11 */
+		{pb0330xx_InitialScale, pb0330xx_Initial},	/* 12 */
 /* or		{pb03303x_InitialScale, pb03303x_Initial}, */
-		{PO2030_mode0, PO2030_mode1},			/* 14 */
-		{tas5130CK_InitialScale, tas5130CK_Initial},	/* 15 */
-		{tas5130cxx_InitialScale, tas5130cxx_Initial},	/* 16 */
+		{PO2030_mode0, PO2030_mode1},			/* 13 */
+		{tas5130CK_InitialScale, tas5130CK_Initial},	/* 14 */
+		{tas5130cxx_InitialScale, tas5130cxx_Initial},	/* 15 */
 		{tas5130c_vf0250_InitialScale, tas5130c_vf0250_Initial},
-								/* 17 */
+								/* 16 */
 	};
 
 	mode = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].mode;
@@ -7324,7 +7313,7 @@ static void sd_close(struct gspca_dev *gspca_dev)
 
 static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 			struct gspca_frame *frame,
-			unsigned char *data,
+			__u8 *data,
 			int len)
 {
 
@@ -7401,6 +7390,16 @@ static int sd_getautogain(struct gspca_dev *gspca_dev, __s32 *val)
 	return 0;
 }
 
+static int sd_setgamma(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->gamma = val;
+	if (gspca_dev->streaming)
+		setcontrast(gspca_dev);
+	return 0;
+}
+
 static int sd_getgamma(struct gspca_dev *gspca_dev, __s32 *val)
 {
 	struct sd *sd = (struct sd *) gspca_dev;
@@ -7409,6 +7408,63 @@ static int sd_getgamma(struct gspca_dev *gspca_dev, __s32 *val)
 	return 0;
 }
 
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->lightfreq = val;
+	if (gspca_dev->streaming)
+		setlightfreq(gspca_dev);
+	return 0;
+}
+
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->lightfreq;
+	return 0;
+}
+
+static int sd_setsharpness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->sharpness = val;
+	if (gspca_dev->streaming)
+		setsharpness(gspca_dev);
+	return 0;
+}
+
+static int sd_getsharpness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->sharpness;
+	return 0;
+}
+
+static int sd_querymenu(struct gspca_dev *gspca_dev,
+			struct v4l2_querymenu *menu)
+{
+	switch (menu->id) {
+	case V4L2_CID_POWER_LINE_FREQUENCY:
+		switch (menu->index) {
+		case 0:		/* V4L2_CID_POWER_LINE_FREQUENCY_DISABLED */
+			strcpy(menu->name, "NoFliker");
+			return 0;
+		case 1:		/* V4L2_CID_POWER_LINE_FREQUENCY_50HZ */
+			strcpy(menu->name, "50 Hz");
+			return 0;
+		case 2:		/* V4L2_CID_POWER_LINE_FREQUENCY_60HZ */
+			strcpy(menu->name, "60 Hz");
+			return 0;
+		}
+		break;
+	}
+	return -EINVAL;
+}
+
 static struct sd_desc sd_desc = {
 	.name = MODULE_NAME,
 	.ctrls = sd_ctrls,
@@ -7420,6 +7476,7 @@ static struct sd_desc sd_desc = {
 	.stop0 = sd_stop0,
 	.close = sd_close,
 	.pkt_scan = sd_pkt_scan,
+	.querymenu = sd_querymenu,
 };
 
 #define DVNM(name) .driver_info = (kernel_ulong_t) name
@@ -7514,10 +7571,6 @@ static void __exit sd_mod_exit(void)
 module_init(sd_mod_init);
 module_exit(sd_mod_exit);
 
-module_param(lightfreq, int, 0644);
-MODULE_PARM_DESC(lightfreq,
-	"Light frequency banding filter: 50, 60 Hz or"
-	" 0 to NoFliker (default=50)");
 module_param(force_sensor, int, 0644);
 MODULE_PARM_DESC(force_sensor,
 	"Force sensor. Only for experts!!!");
diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 9385c823a978..43a6c81a53ef 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -324,6 +324,8 @@ struct v4l2_pix_format
 #define V4L2_PIX_FMT_PWC1     v4l2_fourcc('P','W','C','1') /* pwc older webcam */
 #define V4L2_PIX_FMT_PWC2     v4l2_fourcc('P','W','C','2') /* pwc newer webcam */
 #define V4L2_PIX_FMT_ET61X251 v4l2_fourcc('E','6','2','5') /* ET61X251 compression */
+#define V4L2_PIX_FMT_SPCA501  v4l2_fourcc('S','5','0','1') /* YUYV per line */
+#define V4L2_PIX_FMT_SPCA561  v4l2_fourcc('S','5','6','1') /* compressed BGGR bayer */
 
 /*
  *	F O R M A T   E N U M E R A T I O N
-- 
cgit v1.2.3


From 54ab92ca05550550bcec2462de2605f35d079b66 Mon Sep 17 00:00:00 2001
From: Hans de Goede <j.w.r.degoede@hhs.nl>
Date: Thu, 3 Jul 2008 11:20:58 -0300
Subject: V4L/DVB (8194): gspca: Fix the format of the low resolution mode of
 spca561.

The low (half) res modes of the spca561 are not spca561 compressed, but are
raw bayer, this patches fixes this and adds a PIX_FMT define for the GBRG
bayer format used by the spca561 in low res mode.

Signed-off-by: Hans de Goede <j.w.r.degoede@hhs.nl>
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/gspca/spca561.c | 6 +++---
 include/linux/videodev2.h           | 1 +
 2 files changed, 4 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/media/video/gspca/spca561.c b/drivers/media/video/gspca/spca561.c
index a94e6270115e..dbd1648a655f 100644
--- a/drivers/media/video/gspca/spca561.c
+++ b/drivers/media/video/gspca/spca561.c
@@ -98,8 +98,8 @@ static struct ctrl sd_ctrls[] = {
 };
 
 static struct cam_mode sif_mode[] = {
-	{V4L2_PIX_FMT_SPCA561, 160, 120, 3},
-	{V4L2_PIX_FMT_SPCA561, 176, 144, 2},
+	{V4L2_PIX_FMT_SGBRG8, 160, 120, 3},
+	{V4L2_PIX_FMT_SGBRG8, 176, 144, 2},
 	{V4L2_PIX_FMT_SPCA561, 320, 240, 1},
 	{V4L2_PIX_FMT_SPCA561, 352, 288, 0},
 };
@@ -808,7 +808,7 @@ static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 			gspca_frame_add(gspca_dev, FIRST_PACKET,
 					frame, data, len);
 		} else {
-			/*fixme: which format?*/
+			/* raw bayer (with a header, which we skip) */
 			data += 20;
 			len -= 20;
 			gspca_frame_add(gspca_dev, FIRST_PACKET,
diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 43a6c81a53ef..577f15634791 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -310,6 +310,7 @@ struct v4l2_pix_format
 
 /* see http://www.siliconimaging.com/RGB%20Bayer.htm */
 #define V4L2_PIX_FMT_SBGGR8  v4l2_fourcc('B','A','8','1') /*  8  BGBG.. GRGR.. */
+#define V4L2_PIX_FMT_SGBRG8  v4l2_fourcc('G','B','R','G') /*  8  GBGB.. RGRG.. */
 #define V4L2_PIX_FMT_SBGGR16 v4l2_fourcc('B','Y','R','2') /* 16  BGBG.. GRGR.. */
 
 /* compressed formats */
-- 
cgit v1.2.3


From ab8f12cf8e3d33adec382585f9f60f8686a372c3 Mon Sep 17 00:00:00 2001
From: Hans de Goede <j.w.r.degoede@hhs.nl>
Date: Fri, 4 Jul 2008 18:29:32 -0300
Subject: V4L/DVB (8197): gspca: pac207 frames no more decoded in the
 subdriver.

videodev2: New pixfmt
pac207:   Remove the specific decoding.
main:     get_buff_size operation added for the subdriver.

Signed-off-by: Hans de Goede <j.w.r.degoede@hhs.nl>
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/gspca/gspca.c  |   7 +-
 drivers/media/video/gspca/gspca.h  |   2 +
 drivers/media/video/gspca/pac207.c | 353 +++++--------------------------------
 include/linux/videodev2.h          |   3 +-
 4 files changed, 53 insertions(+), 312 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/media/video/gspca/gspca.c b/drivers/media/video/gspca/gspca.c
index f6d216316b11..708270a46e1f 100644
--- a/drivers/media/video/gspca/gspca.c
+++ b/drivers/media/video/gspca/gspca.c
@@ -321,6 +321,7 @@ static int gspca_is_compressed(__u32 format)
 	case V4L2_PIX_FMT_MJPEG:
 	case V4L2_PIX_FMT_JPEG:
 	case V4L2_PIX_FMT_SPCA561:
+	case V4L2_PIX_FMT_PAC207:
 		return 1;
 	}
 	return 0;
@@ -380,7 +381,8 @@ static __u32 get_v4l2_depth(__u32 pixfmt)
 	case V4L2_PIX_FMT_JPEG:
 	case V4L2_PIX_FMT_SBGGR8:	/* 'BA81' Bayer */
 	case V4L2_PIX_FMT_SN9C10X:	/* 'S910' SN9C10x compression */
-	case V4L2_PIX_FMT_SPCA561:	/* 'S561' compressed BGGR bayer */
+	case V4L2_PIX_FMT_SPCA561:	/* 'S561' compressed GBRG bayer */
+	case V4L2_PIX_FMT_PAC207:	/* 'P207' compressed BGGR bayer */
 		return 8;
 	}
 	PDEBUG(D_ERR|D_CONF, "Unknown pixel format %c%c%c%c",
@@ -395,6 +397,9 @@ static int gspca_get_buff_size(struct gspca_dev *gspca_dev, int mode)
 {
 	unsigned int size;
 
+	if (gspca_dev->sd_desc->get_buff_size)
+		return gspca_dev->sd_desc->get_buff_size(gspca_dev, mode);
+
 	size =  gspca_dev->cam.cam_mode[mode].width *
 		gspca_dev->cam.cam_mode[mode].height *
 		get_v4l2_depth(gspca_dev->cam.cam_mode[mode].pixfmt) / 8;
diff --git a/drivers/media/video/gspca/gspca.h b/drivers/media/video/gspca/gspca.h
index 0d23e0a90c77..05e8ee06071b 100644
--- a/drivers/media/video/gspca/gspca.h
+++ b/drivers/media/video/gspca/gspca.h
@@ -80,6 +80,7 @@ typedef void (*cam_pkt_op) (struct gspca_dev *gspca_dev,
 				struct gspca_frame *frame,
 				__u8 *data,
 				int len);
+typedef int (*cam_get_buff_size_op) (struct gspca_dev *gspca_dev, int mode);
 
 struct ctrl {
 	struct v4l2_queryctrl qctrl;
@@ -106,6 +107,7 @@ struct sd_desc {
 	cam_jpg_op get_jcomp;
 	cam_jpg_op set_jcomp;
 	cam_qmnu_op querymenu;
+	cam_get_buff_size_op get_buff_size; /* optional */
 };
 
 /* packet types when moving from iso buf to frame buf */
diff --git a/drivers/media/video/gspca/pac207.c b/drivers/media/video/gspca/pac207.c
index 7e067456e62b..0135ba599e81 100644
--- a/drivers/media/video/gspca/pac207.c
+++ b/drivers/media/video/gspca/pac207.c
@@ -58,34 +58,10 @@ MODULE_LICENSE("GPL");
    before doing any other adjustments */
 #define PAC207_AUTOGAIN_IGNORE_FRAMES	3
 
-enum pac207_line_state {
-	LINE_HEADER1,
-	LINE_HEADER2,
-	LINE_UNCOMPRESSED,
-	LINE_COMPRESSED,
-};
-
-struct pac207_decoder_state {
-	/* generic state */
-	u16 line_read;
-	u16 line_marker;
-	u8 line_state;
-	u8 header_read;
-	/* compression state */
-	u16 processed_bytes;
-	u8 remaining_bits;
-	s8 no_remaining_bits;
-	u8 get_abs;
-	u8 discard_byte;
-	u8 line_decode_buf[352];
-};
-
 /* specific webcam descriptor */
 struct sd {
 	struct gspca_dev gspca_dev;		/* !! must be the first item */
 
-	struct pac207_decoder_state decoder_state;
-
 	u8 mode;
 
 	u8 brightness;
@@ -94,6 +70,7 @@ struct sd {
 	u8 gain;
 
 	u8 sof_read;
+	u8 header_read;
 	u8 autogain_ignore_frames;
 
 	atomic_t avg_lum;
@@ -173,8 +150,8 @@ static struct ctrl sd_ctrls[] = {
 };
 
 static struct cam_mode sif_mode[] = {
-	{V4L2_PIX_FMT_SBGGR8, 176, 144, 1},
-	{V4L2_PIX_FMT_SBGGR8, 352, 288, 0},
+	{V4L2_PIX_FMT_PAC207, 176, 144, 1},
+	{V4L2_PIX_FMT_PAC207, 352, 288, 0},
 };
 
 static const __u8 pac207_sensor_init[][8] = {
@@ -361,68 +338,17 @@ static void sd_close(struct gspca_dev *gspca_dev)
 {
 }
 
-/* -- convert pixart frames to Bayer -- */
-/* Sonix decompressor struct B.S.(2004) */
-static struct {
-	u8 is_abs;
-	u8 len;
-	s8 val;
-} table[256];
-
-void init_pixart_decoder(void)
+static int sd_get_buff_size_op(struct gspca_dev *gspca_dev, int mode)
 {
-	int i, is_abs, val, len;
-
-	for (i = 0; i < 256; i++) {
-		is_abs = 0;
-		val = 0;
-		len = 0;
-		if ((i & 0xC0) == 0) {
-			/* code 00 */
-			val = 0;
-			len = 2;
-		} else if ((i & 0xC0) == 0x40) {
-			/* code 01 */
-			val = -5;
-			len = 2;
-		} else if ((i & 0xC0) == 0x80) {
-			/* code 10 */
-			val = 5;
-			len = 2;
-		} else if ((i & 0xF0) == 0xC0) {
-			/* code 1100 */
-			val = -10;
-			len = 4;
-		} else if ((i & 0xF0) == 0xD0) {
-			/* code 1101 */
-			val = 10;
-			len = 4;
-		} else if ((i & 0xF8) == 0xE0) {
-			/* code 11100 */
-			val = -15;
-			len = 5;
-		} else if ((i & 0xF8) == 0xE8) {
-			/* code 11101 */
-			val = 15;
-			len = 5;
-		} else if ((i & 0xFC) == 0xF0) {
-			/* code 111100 */
-			val = -20;
-			len = 6;
-		} else if ((i & 0xFC) == 0xF4) {
-			/* code 111101 */
-			val = 20;
-			len = 6;
-		} else if ((i & 0xF8) == 0xF8) {
-			/* code 11111xxxxxx */
-			is_abs = 1;
-			val = 0;
-			len = 5;
-		}
-		table[i].is_abs = is_abs;
-		table[i].val = val;
-		table[i].len = len;
+	switch (gspca_dev->cam.cam_mode[mode].width) {
+	case 176: /* 176x144 */
+		/* uncompressed, add 2 bytes / line for line header */
+		return (176 + 2) * 144;
+	case 352: /* 352x288 */
+		/* compressed */
+		return 352 * 288 / 2;
 	}
+	return -EIO; /* should never happen */
 }
 
 /* auto gain and exposure algorithm based on the knee algorithm described here:
@@ -517,245 +443,52 @@ static unsigned char *pac207_find_sof(struct gspca_dev *gspca_dev,
 	return NULL;
 }
 
-static int pac207_decompress_row(struct gspca_dev *gspca_dev,
-				struct gspca_frame *f,
-				__u8 *cdata,
-				int len)
-{
-	struct sd *sd = (struct sd *) gspca_dev;
-	struct pac207_decoder_state *decoder_state = &sd->decoder_state;
-	unsigned char *outp = decoder_state->line_decode_buf +
-				decoder_state->line_read;
-	int val, bitlen, bitpos = -decoder_state->no_remaining_bits;
-	u8 code;
-
-	/* first two pixels are stored as raw 8-bit */
-	while (decoder_state->line_read < 2) {
-		*outp++ = *cdata++;
-		decoder_state->line_read++;
-		len--;
-		if (len == 0)
-			return 0;
-	}
-
-	while (decoder_state->line_read < gspca_dev->width) {
-		if (bitpos < 0) {
-			code = decoder_state->remaining_bits << (8 + bitpos) |
-				cdata[0] >> -bitpos;
-		} else {
-			u8 *addr = cdata + bitpos / 8;
-			code = addr[0] << (bitpos & 7) |
-				addr[1] >> (8 - (bitpos & 7));
-		}
-
-		bitlen = decoder_state->get_abs ?
-				6 : table[code].len;
-
-		/* Stop decompressing if we're out of input data */
-		if ((bitpos + bitlen) > (len * 8))
-			break;
-
-		if (decoder_state->get_abs) {
-			*outp++ = code & 0xFC;
-			decoder_state->line_read++;
-			decoder_state->get_abs = 0;
-		} else {
-			if (table[code].is_abs) {
-				decoder_state->get_abs = 1;
-			} else {
-				/* relative to left pixel */
-				val = outp[-2] +
-					table[code].val;
-				if (val > 0xff)
-					val = 0xff;
-				else if (val < 0)
-					val = 0;
-				*outp++ = val;
-				decoder_state->line_read++;
-			}
-		}
-		bitpos += bitlen;
-	}
-
-	if (decoder_state->line_read == gspca_dev->width) {
-		int compressed_line_len;
-
-		gspca_frame_add(gspca_dev, INTER_PACKET, f,
-				decoder_state->line_decode_buf,
-				gspca_dev->width);
-
-		/* completely decompressed line, round pos to nearest word */
-		compressed_line_len = ((decoder_state->processed_bytes * 8 +
-			bitpos + 15) / 16) * 2;
-
-		len -= compressed_line_len - decoder_state->processed_bytes;
-		if (len < 0) {
-			decoder_state->discard_byte = 1;
-			len = 0;
-		}
-	} else {
-		decoder_state->processed_bytes += len;
-		decoder_state->remaining_bits = cdata[bitpos/8];
-		decoder_state->no_remaining_bits = (8 - bitpos) & 7;
-		len = 0;
-	}
-
-	return len;
-}
-
-static void pac207_decode_line_init(struct gspca_dev *gspca_dev)
-{
-	struct sd *sd = (struct sd *) gspca_dev;
-	struct pac207_decoder_state *decoder_state = &sd->decoder_state;
-
-	decoder_state->line_read = 0;
-	decoder_state->line_state = LINE_HEADER1;
-	decoder_state->processed_bytes = 0;
-	decoder_state->no_remaining_bits = 0;
-	decoder_state->get_abs = 0;
-}
-
-static void pac207_decode_frame_init(struct gspca_dev *gspca_dev)
-{
-	struct sd *sd = (struct sd *) gspca_dev;
-	struct pac207_decoder_state *decoder_state = &sd->decoder_state;
-
-	decoder_state->header_read = 0;
-	decoder_state->discard_byte = 0;
-
-	pac207_decode_line_init(gspca_dev);
-}
-
-static int pac207_decode_frame_data(struct gspca_dev *gspca_dev,
-	struct gspca_frame *f, unsigned char *data, int len)
-{
-	struct sd *sd = (struct sd *) gspca_dev;
-	struct pac207_decoder_state *decoder_state = &sd->decoder_state;
-	int needed = 0;
-
-	/* first 11 bytes after sof marker: frame header */
-	if (decoder_state->header_read < 11) {
-		/* get average lumination from frame header (byte 5) */
-		if (decoder_state->header_read < 5) {
-			needed = 5 - decoder_state->header_read;
-			if (len >= needed)
-				atomic_set(&sd->avg_lum, data[needed-1]);
-		}
-		/* skip the rest of the header */
-		needed = 11 - decoder_state->header_read;
-		if (len <= needed) {
-			decoder_state->header_read += len;
-			return 0;
-		}
-		data += needed;
-		len -= needed;
-		decoder_state->header_read = 11;
-	}
-
-	while (len) {
-		if (decoder_state->discard_byte) {
-			data++;
-			len--;
-			decoder_state->discard_byte = 0;
-			continue;
-		}
-
-		switch (decoder_state->line_state) {
-		case LINE_HEADER1:
-			decoder_state->line_marker = data[0] << 8;
-			decoder_state->line_state = LINE_HEADER2;
-			needed = 1;
-			break;
-		case LINE_HEADER2:
-			decoder_state->line_marker |= data[0];
-			switch (decoder_state->line_marker) {
-			case 0x0ff0:
-				decoder_state->line_state = LINE_UNCOMPRESSED;
-				break;
-			case 0x1ee1:
-				decoder_state->line_state = LINE_COMPRESSED;
-				break;
-			default:
-				PDEBUG(D_STREAM,
-					"Error unknown line-header %04X",
-					(int) decoder_state->line_marker);
-				gspca_dev->last_packet_type = DISCARD_PACKET;
-				return 0;
-			}
-			needed = 1;
-			break;
-		case LINE_UNCOMPRESSED:
-			needed = gspca_dev->width - decoder_state->line_read;
-			if (needed > len)
-				needed = len;
-			gspca_frame_add(gspca_dev, INTER_PACKET, f, data,
-				needed);
-			decoder_state->line_read += needed;
-			break;
-		case LINE_COMPRESSED:
-			needed = len -
-				pac207_decompress_row(gspca_dev, f, data, len);
-			break;
-		}
-
-		data += needed;
-		len -= needed;
-
-		if (decoder_state->line_read == gspca_dev->width) {
-			if ((f->data_end - f->data) ==
-				(gspca_dev->width * gspca_dev->height)) {
-				/* eureka we've got a frame */
-				return 1;
-			}
-			pac207_decode_line_init(gspca_dev);
-		}
-	}
-
-	return 0;
-}
-
 static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 			struct gspca_frame *frame,
 			__u8 *data,
 			int len)
 {
+	struct sd *sd = (struct sd *) gspca_dev;
 	unsigned char *sof;
-	int n;
 
 	sof = pac207_find_sof(gspca_dev, data, len);
-
 	if (sof) {
+		int n;
+
 		/* finish decoding current frame */
-		if (gspca_dev->last_packet_type == INTER_PACKET) {
-			n = sof - data;
-			if (n > sizeof(pac207_sof_marker))
-				n -= sizeof(pac207_sof_marker);
-			else
-				n = 0;
-			n = pac207_decode_frame_data(gspca_dev, frame,
-							data, n);
-			if (n)
-				frame = gspca_frame_add(gspca_dev,
-						LAST_PACKET,
-						frame,
-						NULL,
-						0);
-			else
-				PDEBUG(D_STREAM, "Incomplete frame");
-		}
-		pac207_decode_frame_init(gspca_dev);
+		n = sof - data;
+		if (n > sizeof pac207_sof_marker)
+			n -= sizeof pac207_sof_marker;
+		else
+			n = 0;
+		frame = gspca_frame_add(gspca_dev, LAST_PACKET, frame,
+					data, n);
+		sd->header_read = 0;
 		gspca_frame_add(gspca_dev, FIRST_PACKET, frame, NULL, 0);
 		len -= sof - data;
 		data = sof;
 	}
+	if (sd->header_read < 11) {
+		int needed;
 
-	if (gspca_dev->last_packet_type == DISCARD_PACKET)
-		return;
+		/* get average lumination from frame header (byte 5) */
+		if (sd->header_read < 5) {
+			needed = 5 - sd->header_read;
+			if (len >= needed)
+				atomic_set(&sd->avg_lum, data[needed - 1]);
+		}
+		/* skip the rest of the header */
+		needed = 11 - sd->header_read;
+		if (len <= needed) {
+			sd->header_read += len;
+			return;
+		}
+		data += needed;
+		len -= needed;
+		sd->header_read = 11;
+	}
 
-	n = pac207_decode_frame_data(gspca_dev, frame, data, len);
-	if (n)
-		gspca_frame_add(gspca_dev, LAST_PACKET,
-				frame, NULL, 0);
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, data, len);
 }
 
 static void setbrightness(struct gspca_dev *gspca_dev)
@@ -891,6 +624,7 @@ static const struct sd_desc sd_desc = {
 	.close = sd_close,
 	.dq_callback = pac207_do_auto_gain,
 	.pkt_scan = sd_pkt_scan,
+	.get_buff_size = sd_get_buff_size_op,
 };
 
 /* -- module initialisation -- */
@@ -927,7 +661,6 @@ static struct usb_driver sd_driver = {
 /* -- module insert / remove -- */
 static int __init sd_mod_init(void)
 {
-	init_pixart_decoder();
 	if (usb_register(&sd_driver) < 0)
 		return -1;
 	PDEBUG(D_PROBE, "v%s registered", version);
diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 577f15634791..2e66a95e8d32 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -326,7 +326,8 @@ struct v4l2_pix_format
 #define V4L2_PIX_FMT_PWC2     v4l2_fourcc('P','W','C','2') /* pwc newer webcam */
 #define V4L2_PIX_FMT_ET61X251 v4l2_fourcc('E','6','2','5') /* ET61X251 compression */
 #define V4L2_PIX_FMT_SPCA501  v4l2_fourcc('S','5','0','1') /* YUYV per line */
-#define V4L2_PIX_FMT_SPCA561  v4l2_fourcc('S','5','6','1') /* compressed BGGR bayer */
+#define V4L2_PIX_FMT_SPCA561  v4l2_fourcc('S','5','6','1') /* compressed GBRG bayer */
+#define V4L2_PIX_FMT_PAC207   v4l2_fourcc('P','2','0','7') /* compressed BGGR bayer */
 
 /*
  *	F O R M A T   E N U M E R A T I O N
-- 
cgit v1.2.3


From be99af6679174e5d0e9f36fc8c18318a8ce34bca Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Thu, 10 Jul 2008 11:04:10 -0300
Subject: V4L/DVB (8245): ovcamchip: Delete stray I2C bus ID

I2C_HW_SMBUS_OVFX2 is referenced in ovcamchip_core.c, but no bus uses
this driver ID, so we can remove the reference. As far as I can see,
the Cypress FX2 webcam is handled by a different driver (dvb-usb).

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/ovcamchip/ovcamchip_core.c | 1 -
 include/linux/i2c-id.h                         | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/media/video/ovcamchip/ovcamchip_core.c b/drivers/media/video/ovcamchip/ovcamchip_core.c
index 8063e33f1c85..065c2454113e 100644
--- a/drivers/media/video/ovcamchip/ovcamchip_core.c
+++ b/drivers/media/video/ovcamchip/ovcamchip_core.c
@@ -297,7 +297,6 @@ static int ovcamchip_attach(struct i2c_adapter *adap)
 	switch (adap->id) {
 	case I2C_HW_SMBUS_OV511:
 	case I2C_HW_SMBUS_OV518:
-	case I2C_HW_SMBUS_OVFX2:
 	case I2C_HW_SMBUS_W9968CF:
 		PDEBUG(1, "Adapter ID 0x%06x accepted", adap->id);
 		break;
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index ef13b7c66df3..9083847157a3 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -158,7 +158,6 @@
 #define I2C_HW_SMBUS_W9968CF	0x04000d
 #define I2C_HW_SMBUS_OV511	0x04000e /* OV511(+) USB 1.1 webcam ICs */
 #define I2C_HW_SMBUS_OV518	0x04000f /* OV518(+) USB 1.1 webcam ICs */
-#define I2C_HW_SMBUS_OVFX2	0x040011 /* Cypress/OmniVision FX2 webcam */
 #define I2C_HW_SMBUS_CAFE	0x040012 /* Marvell 88ALP01 "CAFE" cam  */
 #define I2C_HW_SMBUS_ALI1563	0x040013
 
-- 
cgit v1.2.3


From 5a367dfb739831d54caf226ce0bc4c75ef264d8d Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Thu, 10 Jul 2008 11:23:37 -0300
Subject: V4L/DVB (8246): tvaudio: Stop I2C driver ID abuse

The tvaudio driver is using "official" I2C device IDs for internal
purpose. There must be some historical reason behind this but anyway,
it shouldn't do that. As the stored values are never used, the easiest
way to fix the problem is simply to remove them altogether.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
---
 drivers/media/video/tvaudio.c | 13 -------------
 include/linux/i2c-id.h        |  7 -------
 2 files changed, 20 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c
index c77914d99d15..463680b13289 100644
--- a/drivers/media/video/tvaudio.c
+++ b/drivers/media/video/tvaudio.c
@@ -69,7 +69,6 @@ typedef struct AUDIOCMD {
 /* chip description */
 struct CHIPDESC {
 	char       *name;             /* chip name         */
-	int        id;                /* ID */
 	int        addr_lo, addr_hi;  /* i2c address range */
 	int        registers;         /* # of registers    */
 
@@ -1256,7 +1255,6 @@ module_param(ta8874z, int, 0444);
 static struct CHIPDESC chiplist[] = {
 	{
 		.name       = "tda9840",
-		.id         = I2C_DRIVERID_TDA9840,
 		.insmodopt  = &tda9840,
 		.addr_lo    = I2C_ADDR_TDA9840 >> 1,
 		.addr_hi    = I2C_ADDR_TDA9840 >> 1,
@@ -1272,7 +1270,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tda9873h",
-		.id         = I2C_DRIVERID_TDA9873,
 		.checkit    = tda9873_checkit,
 		.insmodopt  = &tda9873,
 		.addr_lo    = I2C_ADDR_TDA985x_L >> 1,
@@ -1293,7 +1290,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tda9874h/a",
-		.id         = I2C_DRIVERID_TDA9874,
 		.checkit    = tda9874a_checkit,
 		.initialize = tda9874a_initialize,
 		.insmodopt  = &tda9874a,
@@ -1306,7 +1302,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tda9850",
-		.id         = I2C_DRIVERID_TDA9850,
 		.insmodopt  = &tda9850,
 		.addr_lo    = I2C_ADDR_TDA985x_L >> 1,
 		.addr_hi    = I2C_ADDR_TDA985x_H >> 1,
@@ -1319,7 +1314,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tda9855",
-		.id         = I2C_DRIVERID_TDA9855,
 		.insmodopt  = &tda9855,
 		.addr_lo    = I2C_ADDR_TDA985x_L >> 1,
 		.addr_hi    = I2C_ADDR_TDA985x_H >> 1,
@@ -1344,7 +1338,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tea6300",
-		.id         = I2C_DRIVERID_TEA6300,
 		.insmodopt  = &tea6300,
 		.addr_lo    = I2C_ADDR_TEA6300 >> 1,
 		.addr_hi    = I2C_ADDR_TEA6300 >> 1,
@@ -1365,7 +1358,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tea6320",
-		.id         = I2C_DRIVERID_TEA6300,
 		.initialize = tea6320_initialize,
 		.insmodopt  = &tea6320,
 		.addr_lo    = I2C_ADDR_TEA6300 >> 1,
@@ -1387,7 +1379,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tea6420",
-		.id         = I2C_DRIVERID_TEA6420,
 		.insmodopt  = &tea6420,
 		.addr_lo    = I2C_ADDR_TEA6420 >> 1,
 		.addr_hi    = I2C_ADDR_TEA6420 >> 1,
@@ -1400,7 +1391,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "tda8425",
-		.id         = I2C_DRIVERID_TDA8425,
 		.insmodopt  = &tda8425,
 		.addr_lo    = I2C_ADDR_TDA8425 >> 1,
 		.addr_hi    = I2C_ADDR_TDA8425 >> 1,
@@ -1424,7 +1414,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "pic16c54 (PV951)",
-		.id         = I2C_DRIVERID_PIC16C54_PV9,
 		.insmodopt  = &pic16c54,
 		.addr_lo    = I2C_ADDR_PIC16C54 >> 1,
 		.addr_hi    = I2C_ADDR_PIC16C54>> 1,
@@ -1440,8 +1429,6 @@ static struct CHIPDESC chiplist[] = {
 	},
 	{
 		.name       = "ta8874z",
-		.id         = -1,
-		/*.id         = I2C_DRIVERID_TA8874Z, */
 		.checkit    = ta8874z_checkit,
 		.insmodopt  = &ta8874z,
 		.addr_lo    = I2C_ADDR_TDA9840 >> 1,
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 9083847157a3..4862398e05bf 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -33,15 +33,11 @@
 
 #define I2C_DRIVERID_MSP3400	 1
 #define I2C_DRIVERID_TUNER	 2
-#define I2C_DRIVERID_TDA8425	 4	/* stereo sound processor	*/
 #define I2C_DRIVERID_TEA6420	 5	/* audio matrix switch		*/
 #define I2C_DRIVERID_TEA6415C	 6	/* video matrix switch		*/
 #define I2C_DRIVERID_TDA9840	 7	/* stereo sound processor	*/
 #define I2C_DRIVERID_SAA7111A	 8	/* video input processor	*/
 #define I2C_DRIVERID_SAA7185B	13	/* video encoder		*/
-#define I2C_DRIVERID_TEA6300	18	/* audio mixer			*/
-#define I2C_DRIVERID_TDA9850	20	/* audio mixer			*/
-#define I2C_DRIVERID_TDA9855	21	/* audio mixer			*/
 #define I2C_DRIVERID_SAA7110	22	/* video decoder		*/
 #define I2C_DRIVERID_MGATVO	23	/* Matrox TVOut			*/
 #define I2C_DRIVERID_SAA5249	24	/* SAA5249 and compatibles	*/
@@ -50,9 +46,7 @@
 #define I2C_DRIVERID_TDA7432	27	/* Stereo sound processor	*/
 #define I2C_DRIVERID_TVMIXER    28      /* Mixer driver for tv cards    */
 #define I2C_DRIVERID_TVAUDIO    29      /* Generic TV sound driver      */
-#define I2C_DRIVERID_TDA9873    31      /* TV sound decoder chip        */
 #define I2C_DRIVERID_TDA9875    32      /* TV sound decoder chip        */
-#define I2C_DRIVERID_PIC16C54_PV9 33    /* Audio mux/ir receiver        */
 #define I2C_DRIVERID_BT819	40     /* video decoder			*/
 #define I2C_DRIVERID_BT856	41     /* video encoder			*/
 #define I2C_DRIVERID_VPX3220	42     /* video decoder+vbi/vtxt	*/
@@ -63,7 +57,6 @@
 #define I2C_DRIVERID_INDYCAM	58	/* SGI IndyCam			*/
 #define I2C_DRIVERID_OVCAMCHIP	61	/* OmniVision CMOS image sens.	*/
 #define I2C_DRIVERID_MAX6900	63	/* MAX6900 real-time clock	*/
-#define I2C_DRIVERID_TDA9874	66	/* TV sound decoder		*/
 #define I2C_DRIVERID_SAA6752HS	67	/* MPEG2 encoder		*/
 #define I2C_DRIVERID_TVEEPROM	68	/* TV EEPROM			*/
 #define I2C_DRIVERID_WM8775	69	/* wm8775 audio processor	*/
-- 
cgit v1.2.3


From 6f2384c4bdd4be3dc1e5d22ed5e6f0c3076fda60 Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <sameo@openedhand.com>
Date: Fri, 20 Jun 2008 11:02:19 +0200
Subject: mfd: asic3 gpiolib support

ASIC3 is, among other things, a GPIO extender. We should thus have it
supporting the current gpiolib API.

Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/mfd/asic3.c       | 224 +++++++++++++++++++++++++++++++---------------
 include/linux/mfd/asic3.h |  24 +++--
 2 files changed, 164 insertions(+), 84 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index ef8a492766a7..c70e7a5a5a90 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -9,7 +9,7 @@
  *
  * Copyright 2001 Compaq Computer Corporation.
  * Copyright 2004-2005 Phil Blundell
- * Copyright 2007 OpenedHand Ltd.
+ * Copyright 2007-2008 OpenedHand Ltd.
  *
  * Authors: Phil Blundell <pb@handhelds.org>,
  *	    Samuel Ortiz <sameo@openedhand.com>
@@ -19,12 +19,26 @@
 #include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/irq.h>
+#include <linux/gpio.h>
 #include <linux/io.h>
 #include <linux/spinlock.h>
 #include <linux/platform_device.h>
 
 #include <linux/mfd/asic3.h>
 
+struct asic3 {
+	void __iomem *mapping;
+	unsigned int bus_shift;
+	unsigned int irq_nr;
+	unsigned int irq_base;
+	spinlock_t lock;
+	u16 irq_bothedge[4];
+	struct gpio_chip gpio;
+	struct device *dev;
+};
+
+static int asic3_gpio_get(struct gpio_chip *chip, unsigned offset);
+
 static inline void asic3_write_register(struct asic3 *asic,
 				 unsigned int reg, u32 value)
 {
@@ -251,7 +265,7 @@ static int asic3_gpio_irq_type(unsigned int irq, unsigned int type)
 		edge &= ~bit;
 	} else if (type == IRQT_BOTHEDGE) {
 		trigger |= bit;
-		if (asic3_gpio_get_value(asic, irq - asic->irq_base))
+		if (asic3_gpio_get(&asic->gpio, irq - asic->irq_base))
 			edge &= ~bit;
 		else
 			edge |= bit;
@@ -350,6 +364,107 @@ static void asic3_irq_remove(struct platform_device *pdev)
 }
 
 /* GPIOs */
+static int asic3_gpio_direction(struct gpio_chip *chip,
+				unsigned offset, int out)
+{
+	u32 mask = ASIC3_GPIO_TO_MASK(offset), out_reg;
+	unsigned int gpio_base;
+	unsigned long flags;
+	struct asic3 *asic;
+
+	asic = container_of(chip, struct asic3, gpio);
+	gpio_base = ASIC3_GPIO_TO_BASE(offset);
+
+	if (gpio_base > ASIC3_GPIO_D_Base) {
+		printk(KERN_ERR "Invalid base (0x%x) for gpio %d\n",
+		       gpio_base, offset);
+		return -EINVAL;
+	}
+
+	spin_lock_irqsave(&asic->lock, flags);
+
+	out_reg = asic3_read_register(asic, gpio_base + ASIC3_GPIO_Direction);
+
+	/* Input is 0, Output is 1 */
+	if (out)
+		out_reg |= mask;
+	else
+		out_reg &= ~mask;
+
+	asic3_write_register(asic, gpio_base + ASIC3_GPIO_Direction, out_reg);
+
+	spin_unlock_irqrestore(&asic->lock, flags);
+
+	return 0;
+
+}
+
+static int asic3_gpio_direction_input(struct gpio_chip *chip,
+				      unsigned offset)
+{
+	return asic3_gpio_direction(chip, offset, 0);
+}
+
+static int asic3_gpio_direction_output(struct gpio_chip *chip,
+				       unsigned offset, int value)
+{
+	return asic3_gpio_direction(chip, offset, 1);
+}
+
+static int asic3_gpio_get(struct gpio_chip *chip,
+			  unsigned offset)
+{
+	unsigned int gpio_base;
+	u32 mask = ASIC3_GPIO_TO_MASK(offset);
+	struct asic3 *asic;
+
+	asic = container_of(chip, struct asic3, gpio);
+	gpio_base = ASIC3_GPIO_TO_BASE(offset);
+
+	if (gpio_base > ASIC3_GPIO_D_Base) {
+		printk(KERN_ERR "Invalid base (0x%x) for gpio %d\n",
+		       gpio_base, offset);
+		return -EINVAL;
+	}
+
+	return asic3_read_register(asic, gpio_base + ASIC3_GPIO_Status) & mask;
+}
+
+static void asic3_gpio_set(struct gpio_chip *chip,
+			   unsigned offset, int value)
+{
+	u32 mask, out_reg;
+	unsigned int gpio_base;
+	unsigned long flags;
+	struct asic3 *asic;
+
+	asic = container_of(chip, struct asic3, gpio);
+	gpio_base = ASIC3_GPIO_TO_BASE(offset);
+
+	if (gpio_base > ASIC3_GPIO_D_Base) {
+		printk(KERN_ERR "Invalid base (0x%x) for gpio %d\n",
+		       gpio_base, offset);
+		return;
+	}
+
+	mask = ASIC3_GPIO_TO_MASK(offset);
+
+	spin_lock_irqsave(&asic->lock, flags);
+
+	out_reg = asic3_read_register(asic, gpio_base + ASIC3_GPIO_Out);
+
+	if (value)
+		out_reg |= mask;
+	else
+		out_reg &= ~mask;
+
+	asic3_write_register(asic, gpio_base + ASIC3_GPIO_Out, out_reg);
+
+	spin_unlock_irqrestore(&asic->lock, flags);
+
+	return;
+}
+
 static inline u32 asic3_get_gpio(struct asic3 *asic, unsigned int base,
 				 unsigned int function)
 {
@@ -368,15 +483,6 @@ static void asic3_set_gpio(struct asic3 *asic, unsigned int base,
 	spin_unlock_irqrestore(&asic->lock, flags);
 }
 
-#define asic3_get_gpio_a(asic, fn) \
-	asic3_get_gpio(asic, ASIC3_GPIO_A_Base, ASIC3_GPIO_##fn)
-#define asic3_get_gpio_b(asic, fn) \
-	asic3_get_gpio(asic, ASIC3_GPIO_B_Base, ASIC3_GPIO_##fn)
-#define asic3_get_gpio_c(asic, fn) \
-	asic3_get_gpio(asic, ASIC3_GPIO_C_Base, ASIC3_GPIO_##fn)
-#define asic3_get_gpio_d(asic, fn) \
-	asic3_get_gpio(asic, ASIC3_GPIO_D_Base, ASIC3_GPIO_##fn)
-
 #define asic3_set_gpio_a(asic, fn, bits, val) \
 	asic3_set_gpio(asic, ASIC3_GPIO_A_Base, ASIC3_GPIO_##fn, bits, val)
 #define asic3_set_gpio_b(asic, fn, bits, val) \
@@ -394,54 +500,6 @@ static void asic3_set_gpio(struct asic3 *asic, unsigned int base,
 	     asic3_set_gpio_d((asic), fn, (bits), (pdata)->gpio_d.field); \
 	} while (0)
 
-int asic3_gpio_get_value(struct asic3 *asic, unsigned gpio)
-{
-	u32 mask = ASIC3_GPIO_bit(gpio);
-
-	switch (gpio >> 4) {
-	case ASIC3_GPIO_BANK_A:
-		return asic3_get_gpio_a(asic, Status) & mask;
-	case ASIC3_GPIO_BANK_B:
-		return asic3_get_gpio_b(asic, Status) & mask;
-	case ASIC3_GPIO_BANK_C:
-		return asic3_get_gpio_c(asic, Status) & mask;
-	case ASIC3_GPIO_BANK_D:
-		return asic3_get_gpio_d(asic, Status) & mask;
-	default:
-		printk(KERN_ERR "%s: invalid GPIO value 0x%x",
-		       __func__, gpio);
-		return -EINVAL;
-	}
-}
-EXPORT_SYMBOL(asic3_gpio_get_value);
-
-void asic3_gpio_set_value(struct asic3 *asic, unsigned gpio, int val)
-{
-	u32 mask = ASIC3_GPIO_bit(gpio);
-	u32 bitval = 0;
-	if (val)
-		bitval = mask;
-
-	switch (gpio >> 4) {
-	case ASIC3_GPIO_BANK_A:
-		asic3_set_gpio_a(asic, Out, mask, bitval);
-		return;
-	case ASIC3_GPIO_BANK_B:
-		asic3_set_gpio_b(asic, Out, mask, bitval);
-		return;
-	case ASIC3_GPIO_BANK_C:
-		asic3_set_gpio_c(asic, Out, mask, bitval);
-		return;
-	case ASIC3_GPIO_BANK_D:
-		asic3_set_gpio_d(asic, Out, mask, bitval);
-		return;
-	default:
-		printk(KERN_ERR "%s: invalid GPIO value 0x%x",
-		       __func__, gpio);
-		return;
-	}
-}
-EXPORT_SYMBOL(asic3_gpio_set_value);
 
 static int asic3_gpio_probe(struct platform_device *pdev)
 {
@@ -472,12 +530,14 @@ static int asic3_gpio_probe(struct platform_device *pdev)
 				     alt_function);
 	}
 
-	return 0;
+	return gpiochip_add(&asic->gpio);
 }
 
-static void asic3_gpio_remove(struct platform_device *pdev)
+static int asic3_gpio_remove(struct platform_device *pdev)
 {
-	return;
+	struct asic3 *asic = platform_get_drvdata(pdev);
+
+	return gpiochip_remove(&asic->gpio);
 }
 
 
@@ -488,11 +548,13 @@ static int asic3_probe(struct platform_device *pdev)
 	struct asic3 *asic;
 	struct resource *mem;
 	unsigned long clksel;
-	int ret;
+	int ret = 0;
 
 	asic = kzalloc(sizeof(struct asic3), GFP_KERNEL);
-	if (!asic)
+	if (asic == NULL) {
+		printk(KERN_ERR "kzalloc failed\n");
 		return -ENOMEM;
+	}
 
 	spin_lock_init(&asic->lock);
 	platform_set_drvdata(pdev, asic);
@@ -502,14 +564,15 @@ static int asic3_probe(struct platform_device *pdev)
 	if (!mem) {
 		ret = -ENOMEM;
 		printk(KERN_ERR "asic3: no MEM resource\n");
-		goto err_out_1;
+		goto out_free;
 	}
 
+
 	asic->mapping = ioremap(mem->start, PAGE_SIZE);
 	if (!asic->mapping) {
 		ret = -ENOMEM;
 		printk(KERN_ERR "asic3: couldn't ioremap\n");
-		goto err_out_1;
+		goto out_free;
 	}
 
 	asic->irq_base = pdata->irq_base;
@@ -525,9 +588,21 @@ static int asic3_probe(struct platform_device *pdev)
 	ret = asic3_irq_probe(pdev);
 	if (ret < 0) {
 		printk(KERN_ERR "asic3: couldn't probe IRQs\n");
-		goto err_out_2;
+		goto out_unmap;
+	}
+
+	asic->gpio.base = pdata->gpio_base;
+	asic->gpio.ngpio = ASIC3_NUM_GPIOS;
+	asic->gpio.get = asic3_gpio_get;
+	asic->gpio.set = asic3_gpio_set;
+	asic->gpio.direction_input = asic3_gpio_direction_input;
+	asic->gpio.direction_output = asic3_gpio_direction_output;
+
+	ret = asic3_gpio_probe(pdev);
+	if (ret < 0) {
+		printk(KERN_ERR "GPIO probe failed\n");
+		goto out_irq;
 	}
-	asic3_gpio_probe(pdev);
 
 	if (pdata->children) {
 		int i;
@@ -541,9 +616,13 @@ static int asic3_probe(struct platform_device *pdev)
 
 	return 0;
 
- err_out_2:
+ out_irq:
+	asic3_irq_remove(pdev);
+
+ out_unmap:
 	iounmap(asic->mapping);
- err_out_1:
+
+ out_free:
 	kfree(asic);
 
 	return ret;
@@ -551,9 +630,12 @@ static int asic3_probe(struct platform_device *pdev)
 
 static int asic3_remove(struct platform_device *pdev)
 {
+	int ret;
 	struct asic3 *asic = platform_get_drvdata(pdev);
 
-	asic3_gpio_remove(pdev);
+	ret = asic3_gpio_remove(pdev);
+	if (ret < 0)
+		return ret;
 	asic3_irq_remove(pdev);
 
 	asic3_write_register(asic, ASIC3_OFFSET(CLOCK, SEL), 0);
diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index 4ab2162db13b..06ef8165f406 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -16,16 +16,6 @@
 
 #include <linux/types.h>
 
-struct asic3 {
-	void __iomem *mapping;
-	unsigned int bus_shift;
-	unsigned int irq_nr;
-	unsigned int irq_base;
-	spinlock_t lock;
-	u16 irq_bothedge[4];
-	struct device *dev;
-};
-
 struct asic3_platform_data {
 	struct {
 		u32 dir;
@@ -41,18 +31,19 @@ struct asic3_platform_data {
 
 	unsigned int irq_base;
 
+	unsigned int gpio_base;
+
 	struct platform_device **children;
 	unsigned int n_children;
 };
 
-int asic3_gpio_get_value(struct asic3 *asic, unsigned gpio);
-void asic3_gpio_set_value(struct asic3 *asic, unsigned gpio, int val);
-
 #define ASIC3_NUM_GPIO_BANKS	4
 #define ASIC3_GPIOS_PER_BANK	16
 #define ASIC3_NUM_GPIOS		64
 #define ASIC3_NR_IRQS		ASIC3_NUM_GPIOS + 6
 
+#define ASIC3_TO_GPIO(gpio) (NR_BUILTIN_GPIO + (gpio))
+
 #define ASIC3_GPIO_BANK_A	0
 #define ASIC3_GPIO_BANK_B	1
 #define ASIC3_GPIO_BANK_C	2
@@ -73,6 +64,13 @@ void asic3_gpio_set_value(struct asic3 *asic, unsigned gpio, int val);
 #define ASIC3_GPIO_C_Base      0x0200
 #define ASIC3_GPIO_D_Base      0x0300
 
+#define ASIC3_GPIO_TO_BANK(gpio) ((gpio) >> 4)
+#define ASIC3_GPIO_TO_BIT(gpio)  ((gpio) - \
+				  (ASIC3_GPIOS_PER_BANK * ((gpio) >> 4)))
+#define ASIC3_GPIO_TO_MASK(gpio) (1 << ASIC3_GPIO_TO_BIT(gpio))
+#define ASIC3_GPIO_TO_BASE(gpio) (ASIC3_GPIO_A_Base + (((gpio) >> 4) * 0x0100))
+#define ASIC3_BANK_TO_BASE(bank) (ASIC3_GPIO_A_Base + ((bank) * 0x100))
+
 #define ASIC3_GPIO_Mask          0x00    /* R/W 0:don't mask */
 #define ASIC3_GPIO_Direction     0x04    /* R/W 0:input */
 #define ASIC3_GPIO_Out           0x08    /* R/W 0:output low */
-- 
cgit v1.2.3


From 1effe5bc6cfbac4506d7944d68dadbd29ad62645 Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <sameo@openedhand.com>
Date: Fri, 20 Jun 2008 11:07:39 +0200
Subject: mfd: asic3 children platform data removal

Platform devices should be dynamically allocated, and each supported
device should have its own platform data.
For now we just remove this buggy code.

Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/mfd/asic3.c       | 8 --------
 include/linux/mfd/asic3.h | 3 ---
 2 files changed, 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index c70e7a5a5a90..dfee0a2ba167 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -604,14 +604,6 @@ static int asic3_probe(struct platform_device *pdev)
 		goto out_irq;
 	}
 
-	if (pdata->children) {
-		int i;
-		for (i = 0; i < pdata->n_children; i++) {
-			pdata->children[i]->dev.parent = &pdev->dev;
-			platform_device_register(pdata->children[i]);
-		}
-	}
-
 	printk(KERN_INFO "ASIC3 Core driver\n");
 
 	return 0;
diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index 06ef8165f406..b1c365800ab5 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -32,9 +32,6 @@ struct asic3_platform_data {
 	unsigned int irq_base;
 
 	unsigned int gpio_base;
-
-	struct platform_device **children;
-	unsigned int n_children;
 };
 
 #define ASIC3_NUM_GPIO_BANKS	4
-- 
cgit v1.2.3


From 3b26bf17226f66bfd6cd4e36ac81f83fe994043a Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <sameo@openedhand.com>
Date: Fri, 20 Jun 2008 11:09:51 +0200
Subject: mfd: New asic3 gpio configuration code

The ASIC3 GPIO configuration code is a bit obscure and hardly readable.
This patch changes it so that it is now more readable and understandable,
by being more explicit.

Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/mfd/asic3.c       | 99 ++++++++++++++++++++---------------------------
 include/linux/mfd/asic3.h | 34 +++++++++++-----
 2 files changed, 67 insertions(+), 66 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index dfee0a2ba167..36b46ded1bff 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -465,69 +465,54 @@ static void asic3_gpio_set(struct gpio_chip *chip,
 	return;
 }
 
-static inline u32 asic3_get_gpio(struct asic3 *asic, unsigned int base,
-				 unsigned int function)
+static int asic3_gpio_probe(struct platform_device *pdev,
+			    u16 *gpio_config, int num)
 {
-	return asic3_read_register(asic, base + function);
-}
-
-static void asic3_set_gpio(struct asic3 *asic, unsigned int base,
-			   unsigned int function, u32 bits, u32 val)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&asic->lock, flags);
-	val |= (asic3_read_register(asic, base + function) & ~bits);
-
-	asic3_write_register(asic, base + function, val);
-	spin_unlock_irqrestore(&asic->lock, flags);
-}
-
-#define asic3_set_gpio_a(asic, fn, bits, val) \
-	asic3_set_gpio(asic, ASIC3_GPIO_A_Base, ASIC3_GPIO_##fn, bits, val)
-#define asic3_set_gpio_b(asic, fn, bits, val) \
-	asic3_set_gpio(asic, ASIC3_GPIO_B_Base, ASIC3_GPIO_##fn, bits, val)
-#define asic3_set_gpio_c(asic, fn, bits, val) \
-	asic3_set_gpio(asic, ASIC3_GPIO_C_Base, ASIC3_GPIO_##fn, bits, val)
-#define asic3_set_gpio_d(asic, fn, bits, val) \
-	asic3_set_gpio(asic, ASIC3_GPIO_D_Base, ASIC3_GPIO_##fn, bits, val)
-
-#define asic3_set_gpio_banks(asic, fn, bits, pdata, field) 		  \
-	do {								  \
-	     asic3_set_gpio_a((asic), fn, (bits), (pdata)->gpio_a.field); \
-	     asic3_set_gpio_b((asic), fn, (bits), (pdata)->gpio_b.field); \
-	     asic3_set_gpio_c((asic), fn, (bits), (pdata)->gpio_c.field); \
-	     asic3_set_gpio_d((asic), fn, (bits), (pdata)->gpio_d.field); \
-	} while (0)
-
-
-static int asic3_gpio_probe(struct platform_device *pdev)
-{
-	struct asic3_platform_data *pdata = pdev->dev.platform_data;
 	struct asic3 *asic = platform_get_drvdata(pdev);
+	u16 alt_reg[ASIC3_NUM_GPIO_BANKS];
+	u16 out_reg[ASIC3_NUM_GPIO_BANKS];
+	u16 dir_reg[ASIC3_NUM_GPIO_BANKS];
+	int i;
 
+	memset(alt_reg, 0, ASIC3_NUM_GPIO_BANKS);
+	memset(out_reg, 0, ASIC3_NUM_GPIO_BANKS);
+	memset(dir_reg, 0, ASIC3_NUM_GPIO_BANKS);
+
+	/* Enable all GPIOs */
 	asic3_write_register(asic, ASIC3_GPIO_OFFSET(A, Mask), 0xffff);
 	asic3_write_register(asic, ASIC3_GPIO_OFFSET(B, Mask), 0xffff);
 	asic3_write_register(asic, ASIC3_GPIO_OFFSET(C, Mask), 0xffff);
 	asic3_write_register(asic, ASIC3_GPIO_OFFSET(D, Mask), 0xffff);
 
-	asic3_set_gpio_a(asic, SleepMask, 0xffff, 0xffff);
-	asic3_set_gpio_b(asic, SleepMask, 0xffff, 0xffff);
-	asic3_set_gpio_c(asic, SleepMask, 0xffff, 0xffff);
-	asic3_set_gpio_d(asic, SleepMask, 0xffff, 0xffff);
-
-	if (pdata) {
-		asic3_set_gpio_banks(asic, Out, 0xffff, pdata, init);
-		asic3_set_gpio_banks(asic, Direction, 0xffff, pdata, dir);
-		asic3_set_gpio_banks(asic, SleepMask, 0xffff, pdata,
-				     sleep_mask);
-		asic3_set_gpio_banks(asic, SleepOut, 0xffff, pdata, sleep_out);
-		asic3_set_gpio_banks(asic, BattFaultOut, 0xffff, pdata,
-				     batt_fault_out);
-		asic3_set_gpio_banks(asic, SleepConf, 0xffff, pdata,
-				     sleep_conf);
-		asic3_set_gpio_banks(asic, AltFunction, 0xffff, pdata,
-				     alt_function);
+	for (i = 0; i < num; i++) {
+		u8 alt, pin, dir, init, bank_num, bit_num;
+		u16 config = gpio_config[i];
+
+		pin = ASIC3_CONFIG_GPIO_PIN(config);
+		alt = ASIC3_CONFIG_GPIO_ALT(config);
+		dir = ASIC3_CONFIG_GPIO_DIR(config);
+		init = ASIC3_CONFIG_GPIO_INIT(config);
+
+		bank_num = ASIC3_GPIO_TO_BANK(pin);
+		bit_num = ASIC3_GPIO_TO_BIT(pin);
+
+		alt_reg[bank_num] |= (alt << bit_num);
+		out_reg[bank_num] |= (init << bit_num);
+		dir_reg[bank_num] |= (dir << bit_num);
+	}
+
+	for (i = 0; i < ASIC3_NUM_GPIO_BANKS; i++) {
+		asic3_write_register(asic,
+				     ASIC3_BANK_TO_BASE(i) +
+				     ASIC3_GPIO_Direction,
+				     dir_reg[i]);
+		asic3_write_register(asic,
+				     ASIC3_BANK_TO_BASE(i) + ASIC3_GPIO_Out,
+				     out_reg[i]);
+		asic3_write_register(asic,
+				     ASIC3_BANK_TO_BASE(i) +
+				     ASIC3_GPIO_AltFunction,
+				     alt_reg[i]);
 	}
 
 	return gpiochip_add(&asic->gpio);
@@ -598,7 +583,9 @@ static int asic3_probe(struct platform_device *pdev)
 	asic->gpio.direction_input = asic3_gpio_direction_input;
 	asic->gpio.direction_output = asic3_gpio_direction_output;
 
-	ret = asic3_gpio_probe(pdev);
+	ret = asic3_gpio_probe(pdev,
+			       pdata->gpio_config,
+			       pdata->gpio_config_num);
 	if (ret < 0) {
 		printk(KERN_ERR "GPIO probe failed\n");
 		goto out_irq;
diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index b1c365800ab5..7e47cfb0c440 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -8,7 +8,7 @@
  * published by the Free Software Foundation.
  *
  * Copyright 2001 Compaq Computer Corporation.
- * Copyright 2007 OpendHand.
+ * Copyright 2007-2008 OpenedHand Ltd.
  */
 
 #ifndef __ASIC3_H__
@@ -17,15 +17,8 @@
 #include <linux/types.h>
 
 struct asic3_platform_data {
-	struct {
-		u32 dir;
-		u32 init;
-		u32 sleep_mask;
-		u32 sleep_out;
-		u32 batt_fault_out;
-		u32 sleep_conf;
-		u32 alt_function;
-	} gpio_a, gpio_b, gpio_c, gpio_d;
+	u16 *gpio_config;
+	unsigned int gpio_config_num;
 
 	unsigned int bus_shift;
 
@@ -86,6 +79,27 @@ struct asic3_platform_data {
 					  */
 #define ASIC3_GPIO_Status        0x30    /* R   Pin status */
 
+/*
+ * ASIC3 GPIO config
+ *
+ * Bits 0..6   gpio number
+ * Bits 7..13  Alternate function
+ * Bit  14     Direction
+ * Bit  15     Initial value
+ *
+ */
+#define ASIC3_CONFIG_GPIO_PIN(config) ((config) & 0x7f)
+#define ASIC3_CONFIG_GPIO_ALT(config)  (((config) & (0x7f << 7)) >> 7)
+#define ASIC3_CONFIG_GPIO_DIR(config)  ((config & (1 << 14)) >> 14)
+#define ASIC3_CONFIG_GPIO_INIT(config) ((config & (1 << 15)) >> 15)
+#define ASIC3_CONFIG_GPIO(gpio, alt, dir, init) (((gpio) & 0x7f) \
+	| (((alt) & 0x7f) << 7) | (((dir) & 0x1) << 14) \
+	| (((init) & 0x1) << 15))
+#define ASIC3_CONFIG_GPIO_DEFAULT(gpio, dir, init) \
+	ASIC3_CONFIG_GPIO((gpio), 0, (dir), (init))
+#define ASIC3_CONFIG_GPIO_DEFAULT_OUT(gpio, init) \
+	ASIC3_CONFIG_GPIO((gpio), 0, 1, (init))
+
 #define ASIC3_SPI_Base		      0x0400
 #define ASIC3_SPI_Control               0x0000
 #define ASIC3_SPI_TxData                0x0004
-- 
cgit v1.2.3


From 3b8139f8b1457af7b5295d97050b3f9a2545a17a Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <sameo@openedhand.com>
Date: Fri, 20 Jun 2008 11:12:21 +0200
Subject: mfd: Use uppercase only for asic3 macros and defines

Let's be consistent and use uppercase only, for both macro and defines.

Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/mfd/asic3.c       | 88 +++++++++++++++++++++++------------------------
 include/linux/mfd/asic3.h | 56 +++++++++++++++---------------
 2 files changed, 72 insertions(+), 72 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index 1924eb01cc8c..9931581b08a1 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -55,8 +55,8 @@ static inline u32 asic3_read_register(struct asic3 *asic,
 
 /* IRQs */
 #define MAX_ASIC_ISR_LOOPS    20
-#define ASIC3_GPIO_Base_INCR \
-	(ASIC3_GPIO_B_Base - ASIC3_GPIO_A_Base)
+#define ASIC3_GPIO_BASE_INCR \
+	(ASIC3_GPIO_B_BASE - ASIC3_GPIO_A_BASE)
 
 static void asic3_irq_flip_edge(struct asic3 *asic,
 				u32 base, int bit)
@@ -66,10 +66,10 @@ static void asic3_irq_flip_edge(struct asic3 *asic,
 
 	spin_lock_irqsave(&asic->lock, flags);
 	edge = asic3_read_register(asic,
-				   base + ASIC3_GPIO_EdgeTrigger);
+				   base + ASIC3_GPIO_EDGE_TRIGGER);
 	edge ^= bit;
 	asic3_write_register(asic,
-			     base + ASIC3_GPIO_EdgeTrigger, edge);
+			     base + ASIC3_GPIO_EDGE_TRIGGER, edge);
 	spin_unlock_irqrestore(&asic->lock, flags);
 }
 
@@ -89,7 +89,7 @@ static void asic3_irq_demux(unsigned int irq, struct irq_desc *desc)
 
 		spin_lock_irqsave(&asic->lock, flags);
 		status = asic3_read_register(asic,
-					     ASIC3_OFFSET(INTR, PIntStat));
+					     ASIC3_OFFSET(INTR, P_INT_STAT));
 		spin_unlock_irqrestore(&asic->lock, flags);
 
 		/* Check all ten register bits */
@@ -101,17 +101,17 @@ static void asic3_irq_demux(unsigned int irq, struct irq_desc *desc)
 			if (status & (1 << bank)) {
 				unsigned long base, istat;
 
-				base = ASIC3_GPIO_A_Base
-				       + bank * ASIC3_GPIO_Base_INCR;
+				base = ASIC3_GPIO_A_BASE
+				       + bank * ASIC3_GPIO_BASE_INCR;
 
 				spin_lock_irqsave(&asic->lock, flags);
 				istat = asic3_read_register(asic,
 							    base +
-							    ASIC3_GPIO_IntStatus);
+							    ASIC3_GPIO_INT_STATUS);
 				/* Clearing IntStatus */
 				asic3_write_register(asic,
 						     base +
-						     ASIC3_GPIO_IntStatus, 0);
+						     ASIC3_GPIO_INT_STATUS, 0);
 				spin_unlock_irqrestore(&asic->lock, flags);
 
 				for (i = 0; i < ASIC3_GPIOS_PER_BANK; i++) {
@@ -154,7 +154,7 @@ static inline int asic3_irq_to_bank(struct asic3 *asic, int irq)
 
 	n = (irq - asic->irq_base) >> 4;
 
-	return (n * (ASIC3_GPIO_B_Base - ASIC3_GPIO_A_Base));
+	return (n * (ASIC3_GPIO_B_BASE - ASIC3_GPIO_A_BASE));
 }
 
 static inline int asic3_irq_to_index(struct asic3 *asic, int irq)
@@ -172,9 +172,9 @@ static void asic3_mask_gpio_irq(unsigned int irq)
 	index = asic3_irq_to_index(asic, irq);
 
 	spin_lock_irqsave(&asic->lock, flags);
-	val = asic3_read_register(asic, bank + ASIC3_GPIO_Mask);
+	val = asic3_read_register(asic, bank + ASIC3_GPIO_MASK);
 	val |= 1 << index;
-	asic3_write_register(asic, bank + ASIC3_GPIO_Mask, val);
+	asic3_write_register(asic, bank + ASIC3_GPIO_MASK, val);
 	spin_unlock_irqrestore(&asic->lock, flags);
 }
 
@@ -186,15 +186,15 @@ static void asic3_mask_irq(unsigned int irq)
 
 	spin_lock_irqsave(&asic->lock, flags);
 	regval = asic3_read_register(asic,
-				     ASIC3_INTR_Base +
-				     ASIC3_INTR_IntMask);
+				     ASIC3_INTR_BASE +
+				     ASIC3_INTR_INT_MASK);
 
 	regval &= ~(ASIC3_INTMASK_MASK0 <<
 		    (irq - (asic->irq_base + ASIC3_NUM_GPIOS)));
 
 	asic3_write_register(asic,
-			     ASIC3_INTR_Base +
-			     ASIC3_INTR_IntMask,
+			     ASIC3_INTR_BASE +
+			     ASIC3_INTR_INT_MASK,
 			     regval);
 	spin_unlock_irqrestore(&asic->lock, flags);
 }
@@ -209,9 +209,9 @@ static void asic3_unmask_gpio_irq(unsigned int irq)
 	index = asic3_irq_to_index(asic, irq);
 
 	spin_lock_irqsave(&asic->lock, flags);
-	val = asic3_read_register(asic, bank + ASIC3_GPIO_Mask);
+	val = asic3_read_register(asic, bank + ASIC3_GPIO_MASK);
 	val &= ~(1 << index);
-	asic3_write_register(asic, bank + ASIC3_GPIO_Mask, val);
+	asic3_write_register(asic, bank + ASIC3_GPIO_MASK, val);
 	spin_unlock_irqrestore(&asic->lock, flags);
 }
 
@@ -223,15 +223,15 @@ static void asic3_unmask_irq(unsigned int irq)
 
 	spin_lock_irqsave(&asic->lock, flags);
 	regval = asic3_read_register(asic,
-				     ASIC3_INTR_Base +
-				     ASIC3_INTR_IntMask);
+				     ASIC3_INTR_BASE +
+				     ASIC3_INTR_INT_MASK);
 
 	regval |= (ASIC3_INTMASK_MASK0 <<
 		   (irq - (asic->irq_base + ASIC3_NUM_GPIOS)));
 
 	asic3_write_register(asic,
-			     ASIC3_INTR_Base +
-			     ASIC3_INTR_IntMask,
+			     ASIC3_INTR_BASE +
+			     ASIC3_INTR_INT_MASK,
 			     regval);
 	spin_unlock_irqrestore(&asic->lock, flags);
 }
@@ -249,11 +249,11 @@ static int asic3_gpio_irq_type(unsigned int irq, unsigned int type)
 
 	spin_lock_irqsave(&asic->lock, flags);
 	level = asic3_read_register(asic,
-				    bank + ASIC3_GPIO_LevelTrigger);
+				    bank + ASIC3_GPIO_LEVEL_TRIGGER);
 	edge = asic3_read_register(asic,
-				   bank + ASIC3_GPIO_EdgeTrigger);
+				   bank + ASIC3_GPIO_EDGE_TRIGGER);
 	trigger = asic3_read_register(asic,
-				      bank + ASIC3_GPIO_TriggerType);
+				      bank + ASIC3_GPIO_TRIGGER_TYPE);
 	asic->irq_bothedge[(irq - asic->irq_base) >> 4] &= ~bit;
 
 	if (type == IRQT_RISING) {
@@ -283,11 +283,11 @@ static int asic3_gpio_irq_type(unsigned int irq, unsigned int type)
 		 */
 		dev_notice(asic->dev, "irq type not changed\n");
 	}
-	asic3_write_register(asic, bank + ASIC3_GPIO_LevelTrigger,
+	asic3_write_register(asic, bank + ASIC3_GPIO_LEVEL_TRIGGER,
 			     level);
-	asic3_write_register(asic, bank + ASIC3_GPIO_EdgeTrigger,
+	asic3_write_register(asic, bank + ASIC3_GPIO_EDGE_TRIGGER,
 			     edge);
-	asic3_write_register(asic, bank + ASIC3_GPIO_TriggerType,
+	asic3_write_register(asic, bank + ASIC3_GPIO_TRIGGER_TYPE,
 			     trigger);
 	spin_unlock_irqrestore(&asic->lock, flags);
 	return 0;
@@ -336,7 +336,7 @@ static int asic3_irq_probe(struct platform_device *pdev)
 		set_irq_flags(irq, IRQF_VALID | IRQF_PROBE);
 	}
 
-	asic3_write_register(asic, ASIC3_OFFSET(INTR, IntMask),
+	asic3_write_register(asic, ASIC3_OFFSET(INTR, INT_MASK),
 			     ASIC3_INTMASK_GINTMASK);
 
 	set_irq_chained_handler(asic->irq_nr, asic3_irq_demux);
@@ -374,7 +374,7 @@ static int asic3_gpio_direction(struct gpio_chip *chip,
 	asic = container_of(chip, struct asic3, gpio);
 	gpio_base = ASIC3_GPIO_TO_BASE(offset);
 
-	if (gpio_base > ASIC3_GPIO_D_Base) {
+	if (gpio_base > ASIC3_GPIO_D_BASE) {
 		dev_err(asic->dev, "Invalid base (0x%x) for gpio %d\n",
 			gpio_base, offset);
 		return -EINVAL;
@@ -382,7 +382,7 @@ static int asic3_gpio_direction(struct gpio_chip *chip,
 
 	spin_lock_irqsave(&asic->lock, flags);
 
-	out_reg = asic3_read_register(asic, gpio_base + ASIC3_GPIO_Direction);
+	out_reg = asic3_read_register(asic, gpio_base + ASIC3_GPIO_DIRECTION);
 
 	/* Input is 0, Output is 1 */
 	if (out)
@@ -390,7 +390,7 @@ static int asic3_gpio_direction(struct gpio_chip *chip,
 	else
 		out_reg &= ~mask;
 
-	asic3_write_register(asic, gpio_base + ASIC3_GPIO_Direction, out_reg);
+	asic3_write_register(asic, gpio_base + ASIC3_GPIO_DIRECTION, out_reg);
 
 	spin_unlock_irqrestore(&asic->lock, flags);
 
@@ -420,13 +420,13 @@ static int asic3_gpio_get(struct gpio_chip *chip,
 	asic = container_of(chip, struct asic3, gpio);
 	gpio_base = ASIC3_GPIO_TO_BASE(offset);
 
-	if (gpio_base > ASIC3_GPIO_D_Base) {
+	if (gpio_base > ASIC3_GPIO_D_BASE) {
 		dev_err(asic->dev, "Invalid base (0x%x) for gpio %d\n",
 			gpio_base, offset);
 		return -EINVAL;
 	}
 
-	return asic3_read_register(asic, gpio_base + ASIC3_GPIO_Status) & mask;
+	return asic3_read_register(asic, gpio_base + ASIC3_GPIO_STATUS) & mask;
 }
 
 static void asic3_gpio_set(struct gpio_chip *chip,
@@ -440,7 +440,7 @@ static void asic3_gpio_set(struct gpio_chip *chip,
 	asic = container_of(chip, struct asic3, gpio);
 	gpio_base = ASIC3_GPIO_TO_BASE(offset);
 
-	if (gpio_base > ASIC3_GPIO_D_Base) {
+	if (gpio_base > ASIC3_GPIO_D_BASE) {
 		dev_err(asic->dev, "Invalid base (0x%x) for gpio %d\n",
 			gpio_base, offset);
 		return;
@@ -450,14 +450,14 @@ static void asic3_gpio_set(struct gpio_chip *chip,
 
 	spin_lock_irqsave(&asic->lock, flags);
 
-	out_reg = asic3_read_register(asic, gpio_base + ASIC3_GPIO_Out);
+	out_reg = asic3_read_register(asic, gpio_base + ASIC3_GPIO_OUT);
 
 	if (value)
 		out_reg |= mask;
 	else
 		out_reg &= ~mask;
 
-	asic3_write_register(asic, gpio_base + ASIC3_GPIO_Out, out_reg);
+	asic3_write_register(asic, gpio_base + ASIC3_GPIO_OUT, out_reg);
 
 	spin_unlock_irqrestore(&asic->lock, flags);
 
@@ -478,10 +478,10 @@ static int asic3_gpio_probe(struct platform_device *pdev,
 	memset(dir_reg, 0, ASIC3_NUM_GPIO_BANKS);
 
 	/* Enable all GPIOs */
-	asic3_write_register(asic, ASIC3_GPIO_OFFSET(A, Mask), 0xffff);
-	asic3_write_register(asic, ASIC3_GPIO_OFFSET(B, Mask), 0xffff);
-	asic3_write_register(asic, ASIC3_GPIO_OFFSET(C, Mask), 0xffff);
-	asic3_write_register(asic, ASIC3_GPIO_OFFSET(D, Mask), 0xffff);
+	asic3_write_register(asic, ASIC3_GPIO_OFFSET(A, MASK), 0xffff);
+	asic3_write_register(asic, ASIC3_GPIO_OFFSET(B, MASK), 0xffff);
+	asic3_write_register(asic, ASIC3_GPIO_OFFSET(C, MASK), 0xffff);
+	asic3_write_register(asic, ASIC3_GPIO_OFFSET(D, MASK), 0xffff);
 
 	for (i = 0; i < num; i++) {
 		u8 alt, pin, dir, init, bank_num, bit_num;
@@ -503,14 +503,14 @@ static int asic3_gpio_probe(struct platform_device *pdev,
 	for (i = 0; i < ASIC3_NUM_GPIO_BANKS; i++) {
 		asic3_write_register(asic,
 				     ASIC3_BANK_TO_BASE(i) +
-				     ASIC3_GPIO_Direction,
+				     ASIC3_GPIO_DIRECTION,
 				     dir_reg[i]);
 		asic3_write_register(asic,
-				     ASIC3_BANK_TO_BASE(i) + ASIC3_GPIO_Out,
+				     ASIC3_BANK_TO_BASE(i) + ASIC3_GPIO_OUT,
 				     out_reg[i]);
 		asic3_write_register(asic,
 				     ASIC3_BANK_TO_BASE(i) +
-				     ASIC3_GPIO_AltFunction,
+				     ASIC3_GPIO_ALT_FUNCTION,
 				     alt_reg[i]);
 	}
 
diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index 7e47cfb0c440..6461a2ae69a2 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -45,39 +45,39 @@ struct asic3_platform_data {
 /* All offsets below are specified with this address bus shift */
 #define ASIC3_DEFAULT_ADDR_SHIFT 2
 
-#define ASIC3_OFFSET(base, reg) (ASIC3_##base##_Base + ASIC3_##base##_##reg)
+#define ASIC3_OFFSET(base, reg) (ASIC3_##base##_BASE + ASIC3_##base##_##reg)
 #define ASIC3_GPIO_OFFSET(base, reg) \
-	(ASIC3_GPIO_##base##_Base + ASIC3_GPIO_##reg)
+	(ASIC3_GPIO_##base##_BASE + ASIC3_GPIO_##reg)
 
-#define ASIC3_GPIO_A_Base      0x0000
-#define ASIC3_GPIO_B_Base      0x0100
-#define ASIC3_GPIO_C_Base      0x0200
-#define ASIC3_GPIO_D_Base      0x0300
+#define ASIC3_GPIO_A_BASE      0x0000
+#define ASIC3_GPIO_B_BASE      0x0100
+#define ASIC3_GPIO_C_BASE      0x0200
+#define ASIC3_GPIO_D_BASE      0x0300
 
 #define ASIC3_GPIO_TO_BANK(gpio) ((gpio) >> 4)
 #define ASIC3_GPIO_TO_BIT(gpio)  ((gpio) - \
 				  (ASIC3_GPIOS_PER_BANK * ((gpio) >> 4)))
 #define ASIC3_GPIO_TO_MASK(gpio) (1 << ASIC3_GPIO_TO_BIT(gpio))
-#define ASIC3_GPIO_TO_BASE(gpio) (ASIC3_GPIO_A_Base + (((gpio) >> 4) * 0x0100))
-#define ASIC3_BANK_TO_BASE(bank) (ASIC3_GPIO_A_Base + ((bank) * 0x100))
-
-#define ASIC3_GPIO_Mask          0x00    /* R/W 0:don't mask */
-#define ASIC3_GPIO_Direction     0x04    /* R/W 0:input */
-#define ASIC3_GPIO_Out           0x08    /* R/W 0:output low */
-#define ASIC3_GPIO_TriggerType   0x0c    /* R/W 0:level */
-#define ASIC3_GPIO_EdgeTrigger   0x10    /* R/W 0:falling */
-#define ASIC3_GPIO_LevelTrigger  0x14    /* R/W 0:low level detect */
-#define ASIC3_GPIO_SleepMask     0x18    /* R/W 0:don't mask in sleep mode */
-#define ASIC3_GPIO_SleepOut      0x1c    /* R/W level 0:low in sleep mode */
-#define ASIC3_GPIO_BattFaultOut  0x20    /* R/W level 0:low in batt_fault */
-#define ASIC3_GPIO_IntStatus     0x24    /* R/W 0:none, 1:detect */
-#define ASIC3_GPIO_AltFunction   0x28	 /* R/W 1:LED register control */
-#define ASIC3_GPIO_SleepConf     0x2c    /*
+#define ASIC3_GPIO_TO_BASE(gpio) (ASIC3_GPIO_A_BASE + (((gpio) >> 4) * 0x0100))
+#define ASIC3_BANK_TO_BASE(bank) (ASIC3_GPIO_A_BASE + ((bank) * 0x100))
+
+#define ASIC3_GPIO_MASK          0x00    /* R/W 0:don't mask */
+#define ASIC3_GPIO_DIRECTION     0x04    /* R/W 0:input */
+#define ASIC3_GPIO_OUT           0x08    /* R/W 0:output low */
+#define ASIC3_GPIO_TRIGGER_TYPE  0x0c    /* R/W 0:level */
+#define ASIC3_GPIO_EDGE_TRIGGER  0x10    /* R/W 0:falling */
+#define ASIC3_GPIO_LEVEL_TRIGGER 0x14    /* R/W 0:low level detect */
+#define ASIC3_GPIO_SLEEP_MASK    0x18    /* R/W 0:don't mask in sleep mode */
+#define ASIC3_GPIO_SLEEP_OUT     0x1c    /* R/W level 0:low in sleep mode */
+#define ASIC3_GPIO_BAT_FAULT_OUT 0x20    /* R/W level 0:low in batt_fault */
+#define ASIC3_GPIO_INT_STATUS    0x24    /* R/W 0:none, 1:detect */
+#define ASIC3_GPIO_ALT_FUNCTION  0x28	 /* R/W 1:LED register control */
+#define ASIC3_GPIO_SLEEP_CONF    0x2c    /*
 					  * R/W bit 1: autosleep
 					  * 0: disable gposlpout in normal mode,
 					  * enable gposlpout in sleep mode.
 					  */
-#define ASIC3_GPIO_Status        0x30    /* R   Pin status */
+#define ASIC3_GPIO_STATUS        0x30    /* R   Pin status */
 
 /*
  * ASIC3 GPIO config
@@ -137,7 +137,7 @@ struct asic3_platform_data {
 #define LED_AUTOSTOP	(1 << 5) /* LED ON/OFF auto stop 0:disable, 1:enable */
 #define LED_ALWAYS	(1 << 6) /* LED Interrupt Mask 0:No mask, 1:mask */
 
-#define ASIC3_CLOCK_Base		0x0A00
+#define ASIC3_CLOCK_BASE	   0x0A00
 #define ASIC3_CLOCK_CDEX           0x00
 #define ASIC3_CLOCK_SEL            0x04
 
@@ -168,12 +168,12 @@ struct asic3_platform_data {
 #define CLOCK_SEL_CX            (1 << 2)
 
 
-#define ASIC3_INTR_Base		0x0B00
+#define ASIC3_INTR_BASE		0x0B00
 
-#define ASIC3_INTR_IntMask       0x00  /* Interrupt mask control */
-#define ASIC3_INTR_PIntStat      0x04  /* Peripheral interrupt status */
-#define ASIC3_INTR_IntCPS        0x08  /* Interrupt timer clock pre-scale */
-#define ASIC3_INTR_IntTBS        0x0c  /* Interrupt timer set */
+#define ASIC3_INTR_INT_MASK       0x00  /* Interrupt mask control */
+#define ASIC3_INTR_P_INT_STAT     0x04  /* Peripheral interrupt status */
+#define ASIC3_INTR_INT_CPS        0x08  /* Interrupt timer clock pre-scale */
+#define ASIC3_INTR_INT_TBS        0x0c  /* Interrupt timer set */
 
 #define ASIC3_INTMASK_GINTMASK    (1 << 0)  /* Global INTs mask 1:enable */
 #define ASIC3_INTMASK_GINTEL      (1 << 1)  /* 1: rising edge, 0: hi level */
-- 
cgit v1.2.3


From 4a67b528e0be5d855b1a7bb71ec769d954765f6c Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 3 Jul 2008 12:27:32 +0200
Subject: mfd: add ASIC3_CONFIG_GPIO templates

As ASIC3 GPIO alternate function configuration is expected to be similar
for several devices, it is convenient to define descriptive macros. This
patch is inspired by the PXA MFP configuration, the alternate functions
were observed on hx4700 and blueangel.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 include/linux/mfd/asic3.h | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index 6461a2ae69a2..0eae3083da43 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -100,6 +100,35 @@ struct asic3_platform_data {
 #define ASIC3_CONFIG_GPIO_DEFAULT_OUT(gpio, init) \
 	ASIC3_CONFIG_GPIO((gpio), 0, 1, (init))
 
+/*
+ * Alternate functions
+ */
+#define ASIC3_GPIOA11_PWM0		ASIC3_CONFIG_GPIO(11, 1, 1, 0)
+#define ASIC3_GPIOA12_PWM1		ASIC3_CONFIG_GPIO(12, 1, 1, 0)
+#define ASIC3_GPIOA15_CONTROL_CX	ASIC3_CONFIG_GPIO(15, 1, 1, 0)
+#define ASIC3_GPIOC0_LED0		ASIC3_CONFIG_GPIO(32, 1, 1, 0)
+#define ASIC3_GPIOC1_LED1		ASIC3_CONFIG_GPIO(33, 1, 1, 0)
+#define ASIC3_GPIOC2_LED2		ASIC3_CONFIG_GPIO(34, 1, 1, 0)
+#define ASIC3_GPIOC3_SPI_RXD		ASIC3_CONFIG_GPIO(35, 1, 0, 0)
+#define ASIC3_GPIOC4_CF_nCD		ASIC3_CONFIG_GPIO(36, 1, 0, 0)
+#define ASIC3_GPIOC4_SPI_TXD		ASIC3_CONFIG_GPIO(36, 1, 1, 0)
+#define ASIC3_GPIOC5_SPI_CLK		ASIC3_CONFIG_GPIO(37, 1, 1, 0)
+#define ASIC3_GPIOC5_nCIOW		ASIC3_CONFIG_GPIO(37, 1, 1, 0)
+#define ASIC3_GPIOC6_nCIOR		ASIC3_CONFIG_GPIO(38, 1, 1, 0)
+#define ASIC3_GPIOC7_nPCE_1		ASIC3_CONFIG_GPIO(39, 1, 0, 0)
+#define ASIC3_GPIOC8_nPCE_2		ASIC3_CONFIG_GPIO(40, 1, 0, 0)
+#define ASIC3_GPIOC9_nPOE		ASIC3_CONFIG_GPIO(41, 1, 0, 0)
+#define ASIC3_GPIOC10_nPWE		ASIC3_CONFIG_GPIO(42, 1, 0, 0)
+#define ASIC3_GPIOC11_PSKTSEL		ASIC3_CONFIG_GPIO(43, 1, 0, 0)
+#define ASIC3_GPIOC12_nPREG		ASIC3_CONFIG_GPIO(44, 1, 0, 0)
+#define ASIC3_GPIOC13_nPWAIT		ASIC3_CONFIG_GPIO(45, 1, 1, 0)
+#define ASIC3_GPIOC14_nPIOIS16		ASIC3_CONFIG_GPIO(46, 1, 1, 0)
+#define ASIC3_GPIOC15_nPIOR		ASIC3_CONFIG_GPIO(47, 1, 0, 0)
+#define ASIC3_GPIOD11_nCIOIS16		ASIC3_CONFIG_GPIO(59, 1, 0, 0)
+#define ASIC3_GPIOD12_nCWAIT		ASIC3_CONFIG_GPIO(60, 1, 0, 0)
+#define ASIC3_GPIOD15_nPIOW		ASIC3_CONFIG_GPIO(63, 1, 0, 0)
+
+
 #define ASIC3_SPI_Base		      0x0400
 #define ASIC3_SPI_Control               0x0000
 #define ASIC3_SPI_TxData                0x0004
-- 
cgit v1.2.3


From 279cac484e55317456900fe3567c7cb5bd46fd5f Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 10 Jul 2008 02:16:27 +0200
Subject: mfd: remove DS1WM register definitions from asic3.h

There is a dedicated ds1wm driver, no need to duplicate this
information here.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 include/linux/mfd/asic3.h | 38 +++-----------------------------------
 1 file changed, 3 insertions(+), 35 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index 0eae3083da43..8f8c46c41f44 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -265,44 +265,12 @@ struct asic3_platform_data {
 #define ASIC3_EXTCF_CF_SLEEP             (1 << 15) /* CF sleep mode control */
 
 /*********************************************
- *  The Onewire interface registers
- *
- *  OWM_CMD
- *  OWM_DAT
- *  OWM_INTR
- *  OWM_INTEN
- *  OWM_CLKDIV
+ *  The Onewire interface (DS1WM) is handled
+ *  by the ds1wm driver.
  *
  *********************************************/
 
-#define ASIC3_OWM_Base		0xC00
-
-#define ASIC3_OWM_CMD         0x00
-#define ASIC3_OWM_DAT         0x04
-#define ASIC3_OWM_INTR        0x08
-#define ASIC3_OWM_INTEN       0x0C
-#define ASIC3_OWM_CLKDIV      0x10
-
-#define ASIC3_OWM_CMD_ONEWR         (1 << 0)
-#define ASIC3_OWM_CMD_SRA           (1 << 1)
-#define ASIC3_OWM_CMD_DQO           (1 << 2)
-#define ASIC3_OWM_CMD_DQI           (1 << 3)
-
-#define ASIC3_OWM_INTR_PD          (1 << 0)
-#define ASIC3_OWM_INTR_PDR         (1 << 1)
-#define ASIC3_OWM_INTR_TBE         (1 << 2)
-#define ASIC3_OWM_INTR_TEMP        (1 << 3)
-#define ASIC3_OWM_INTR_RBF         (1 << 4)
-
-#define ASIC3_OWM_INTEN_EPD        (1 << 0)
-#define ASIC3_OWM_INTEN_IAS        (1 << 1)
-#define ASIC3_OWM_INTEN_ETBE       (1 << 2)
-#define ASIC3_OWM_INTEN_ETMT       (1 << 3)
-#define ASIC3_OWM_INTEN_ERBF       (1 << 4)
-
-#define ASIC3_OWM_CLKDIV_PRE       (3 << 0) /* two bits wide at bit 0 */
-#define ASIC3_OWM_CLKDIV_DIV       (7 << 2) /* 3 bits wide at bit 2 */
-
+#define ASIC3_OWM_BASE		0xC00
 
 /*****************************************************************************
  *  The SD configuration registers are at a completely different location
-- 
cgit v1.2.3


From 99cdb0c8c5e0e43652d25951a85bac82a1231591 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 10 Jul 2008 02:17:02 +0200
Subject: mfd: let asic3 use mem resource instead of bus_shift

The bus_shift parameter in platform_data is not needed
as we can tell the driver with the IOMEM_RESOURCE whether
the ASIC is located on a 16bit or 32bit memory bus.

The htc-egpio driver uses a more descriptive bus_width parameter,
but for drivers where the register map size fixed, we don't even
need this.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/asic3.c       | 11 +++++------
 include/linux/mfd/asic3.h |  5 ++---
 2 files changed, 7 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index f2fb233b0ff3..3b870e7fb3e1 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -313,6 +313,7 @@ static int __init asic3_irq_probe(struct platform_device *pdev)
 	struct asic3 *asic = platform_get_drvdata(pdev);
 	unsigned long clksel = 0;
 	unsigned int irq, irq_base;
+	int map_size;
 
 	asic->irq_nr = platform_get_irq(pdev, 0);
 	if (asic->irq_nr < 0)
@@ -551,8 +552,8 @@ static int __init asic3_probe(struct platform_device *pdev)
 		goto out_free;
 	}
 
-
-	asic->mapping = ioremap(mem->start, PAGE_SIZE);
+	map_size = mem->end - mem->start + 1;
+	asic->mapping = ioremap(mem->start, map_size);
 	if (!asic->mapping) {
 		ret = -ENOMEM;
 		dev_err(asic->dev, "Couldn't ioremap\n");
@@ -561,10 +562,8 @@ static int __init asic3_probe(struct platform_device *pdev)
 
 	asic->irq_base = pdata->irq_base;
 
-	if (pdata && pdata->bus_shift)
-		asic->bus_shift = 2 - pdata->bus_shift;
-	else
-		asic->bus_shift = 0;
+	/* calculate bus shift from mem resource */
+	asic->bus_shift = 2 - (map_size >> 12);
 
 	clksel = 0;
 	asic3_write_register(asic, ASIC3_OFFSET(CLOCK, SEL), clksel);
diff --git a/include/linux/mfd/asic3.h b/include/linux/mfd/asic3.h
index 8f8c46c41f44..322cd6deb9f0 100644
--- a/include/linux/mfd/asic3.h
+++ b/include/linux/mfd/asic3.h
@@ -20,8 +20,6 @@ struct asic3_platform_data {
 	u16 *gpio_config;
 	unsigned int gpio_config_num;
 
-	unsigned int bus_shift;
-
 	unsigned int irq_base;
 
 	unsigned int gpio_base;
@@ -498,6 +496,7 @@ struct asic3_platform_data {
 #define ASIC3_SDIO_CTRL_LEDCtrl              0x7C
 #define ASIC3_SDIO_CTRL_SoftwareReset        0x1C0
 
-#define ASIC3_MAP_SIZE	     		     0x2000
+#define ASIC3_MAP_SIZE_32BIT	0x2000
+#define ASIC3_MAP_SIZE_16BIT	0x1000
 
 #endif /* __ASIC3_H__ */
-- 
cgit v1.2.3


From 6bb0e3a59a089e23eecc0af3b6f6012b2a9affba Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Wed, 16 Jul 2008 21:52:36 +0100
Subject: Subject: [PATCH 1/2] serial: Add flush_buffer() operation to uart_ops

Serial drivers using DMA (like the atmel_serial driver) tend to get very
confused when the xmit buffer is flushed and nobody told them.  They
also tend to spew a lot of garbage since the DMA engine keeps running
after the buffer is flushed and possibly refilled with unrelated data.

This patch adds a new flush_buffer operation to the uart_ops struct,
along with a call to it from uart_flush_buffer() right after the xmit
buffer has been cleared. The driver can implement this in order to
syncronize its internal DMA state with the xmit buffer when the buffer
is flushed.

Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/serial/driver  | 11 +++++++++++
 drivers/serial/serial_core.c |  2 ++
 include/linux/serial_core.h  |  1 +
 3 files changed, 14 insertions(+)

(limited to 'include/linux')

diff --git a/Documentation/serial/driver b/Documentation/serial/driver
index 88ad615dd338..77ba0afbe4db 100644
--- a/Documentation/serial/driver
+++ b/Documentation/serial/driver
@@ -186,6 +186,17 @@ hardware.
 	Locking: port_sem taken.
 	Interrupts: caller dependent.
 
+  flush_buffer(port)
+	Flush any write buffers, reset any DMA state and stop any
+	ongoing DMA transfers.
+
+	This will be called whenever the port->info->xmit circular
+	buffer is cleared.
+
+	Locking: port->lock taken.
+	Interrupts: locally disabled.
+	This call must not sleep
+
   set_termios(port,termios,oldtermios)
 	Change the port parameters, including word length, parity, stop
 	bits.  Update read_status_mask and ignore_status_mask to indicate
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 42d2e108b679..9884bc9eecb1 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -573,6 +573,8 @@ static void uart_flush_buffer(struct tty_struct *tty)
 
 	spin_lock_irqsave(&port->lock, flags);
 	uart_circ_clear(&state->info->xmit);
+	if (port->ops->flush_buffer)
+		port->ops->flush_buffer(port);
 	spin_unlock_irqrestore(&port->lock, flags);
 	tty_wakeup(tty);
 }
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index d8f31de632c5..1d2faa6592ae 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -190,6 +190,7 @@ struct uart_ops {
 	void		(*break_ctl)(struct uart_port *, int ctl);
 	int		(*startup)(struct uart_port *);
 	void		(*shutdown)(struct uart_port *);
+	void		(*flush_buffer)(struct uart_port *);
 	void		(*set_termios)(struct uart_port *, struct ktermios *new,
 				       struct ktermios *old);
 	void		(*set_ldisc)(struct uart_port *);
-- 
cgit v1.2.3


From a352def21a642133758b868c71bee12ab34ad5c5 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:53:12 +0100
Subject: tty: Ldisc revamp

Move the line disciplines towards a conventional ->ops arrangement.  For
the moment the actual 'tty_ldisc' struct in the tty is kept as part of
the tty struct but this can then be changed if it turns out that when it
all settles down we want to refcount ldiscs separately to the tty.

Pull the ldisc code out of /proc and put it with our ldisc code.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/bluetooth/hci_ldisc.c      |   6 +-
 drivers/char/cyclades.c            |   3 +-
 drivers/char/epca.c                |   4 +-
 drivers/char/ip2/i2lib.c           |   4 +-
 drivers/char/ip2/ip2main.c         |   7 +-
 drivers/char/n_hdlc.c              |   6 +-
 drivers/char/n_r3964.c             |   2 +-
 drivers/char/n_tty.c               |   2 +-
 drivers/char/pcmcia/synclink_cs.c  |   4 +-
 drivers/char/pty.c                 |  10 +-
 drivers/char/selection.c           |   3 +-
 drivers/char/synclink.c            |   4 +-
 drivers/char/synclink_gt.c         |   4 +-
 drivers/char/synclinkmp.c          |   4 +-
 drivers/char/tty_io.c              | 336 +++++++++++++++++++++++--------------
 drivers/char/tty_ioctl.c           |  16 +-
 drivers/input/serio/serport.c      |   2 +-
 drivers/isdn/capi/capi.c           |   4 +-
 drivers/isdn/gigaset/ser-gigaset.c |   2 +-
 drivers/net/hamradio/6pack.c       |   2 +-
 drivers/net/hamradio/mkiss.c       |   2 +-
 drivers/net/irda/irtty-sir.c       |   2 +-
 drivers/net/ppp_async.c            |   2 +-
 drivers/net/ppp_synctty.c          |   2 +-
 drivers/net/slip.c                 |   2 +-
 drivers/net/wan/pc300_tty.c        |   4 +-
 drivers/net/wan/x25_asy.c          |   2 +-
 fs/proc/proc_tty.c                 |  48 ------
 include/linux/tty.h                |   9 +-
 include/linux/tty_ldisc.h          |   7 +-
 net/bluetooth/rfcomm/tty.c         |  13 +-
 net/irda/ircomm/ircomm_tty.c       |  14 +-
 32 files changed, 286 insertions(+), 246 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c
index e5cd856a2fea..69df187d74ce 100644
--- a/drivers/bluetooth/hci_ldisc.c
+++ b/drivers/bluetooth/hci_ldisc.c
@@ -282,8 +282,8 @@ static int hci_uart_tty_open(struct tty_struct *tty)
 	/* FIXME: why is this needed. Note don't use ldisc_ref here as the
 	   open path is before the ldisc is referencable */
 
-	if (tty->ldisc.flush_buffer)
-		tty->ldisc.flush_buffer(tty);
+	if (tty->ldisc.ops->flush_buffer)
+		tty->ldisc.ops->flush_buffer(tty);
 	tty_driver_flush_buffer(tty);
 
 	return 0;
@@ -514,7 +514,7 @@ static unsigned int hci_uart_tty_poll(struct tty_struct *tty,
 
 static int __init hci_uart_init(void)
 {
-	static struct tty_ldisc hci_uart_ldisc;
+	static struct tty_ldisc_ops hci_uart_ldisc;
 	int err;
 
 	BT_INFO("HCI UART driver ver %s", VERSION);
diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c
index 6bff9d87dc57..a957dbcc5a46 100644
--- a/drivers/char/cyclades.c
+++ b/drivers/char/cyclades.c
@@ -5246,7 +5246,8 @@ cyclades_get_proc_info(char *buf, char **start, off_t offset, int length,
 					HZ, info->idle_stats.recv_bytes,
 					(cur_jifs - info->idle_stats.recv_idle)/
 					HZ, info->idle_stats.overruns,
-					(long)info->tty->ldisc.num);
+					/* FIXME: double check locking */
+					(long)info->tty->ldisc.ops->num);
 			else
 				size = sprintf(buf + len, "%3d %8lu %10lu %8lu "
 					"%10lu %8lu %9lu %6ld\n",
diff --git a/drivers/char/epca.c b/drivers/char/epca.c
index 60a4df7dac12..aa8e19f44b45 100644
--- a/drivers/char/epca.c
+++ b/drivers/char/epca.c
@@ -2262,8 +2262,8 @@ static int pc_ioctl(struct tty_struct *tty, struct file *file,
 			tty_wait_until_sent(tty, 0);
 		} else {
 			/* ldisc lock already held in ioctl */
-			if (tty->ldisc.flush_buffer)
-				tty->ldisc.flush_buffer(tty);
+			if (tty->ldisc.ops->flush_buffer)
+				tty->ldisc.ops->flush_buffer(tty);
 		}
 		unlock_kernel();
 		/* Fall Thru */
diff --git a/drivers/char/ip2/i2lib.c b/drivers/char/ip2/i2lib.c
index 938879cc7bcc..0061e18aff60 100644
--- a/drivers/char/ip2/i2lib.c
+++ b/drivers/char/ip2/i2lib.c
@@ -868,11 +868,11 @@ i2Input(i2ChanStrPtr pCh)
 		amountToMove = count;
 	}
 	// Move the first block
-	pCh->pTTY->ldisc.receive_buf( pCh->pTTY, 
+	pCh->pTTY->ldisc.ops->receive_buf( pCh->pTTY,
 		 &(pCh->Ibuf[stripIndex]), NULL, amountToMove );
 	// If we needed to wrap, do the second data move
 	if (count > amountToMove) {
-		pCh->pTTY->ldisc.receive_buf( pCh->pTTY, 
+		pCh->pTTY->ldisc.ops->receive_buf( pCh->pTTY,
 		 pCh->Ibuf, NULL, count - amountToMove );
 	}
 	// Bump and wrap the stripIndex all at once by the amount of data read. This
diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c
index 9a2394cda943..5dc74404058f 100644
--- a/drivers/char/ip2/ip2main.c
+++ b/drivers/char/ip2/ip2main.c
@@ -1289,11 +1289,12 @@ static void do_input(struct work_struct *work)
 // code duplicated from n_tty (ldisc)
 static inline void  isig(int sig, struct tty_struct *tty, int flush)
 {
+	/* FIXME: This is completely bogus */
 	if (tty->pgrp)
 		kill_pgrp(tty->pgrp, sig, 1);
 	if (flush || !L_NOFLSH(tty)) {
-		if ( tty->ldisc.flush_buffer )  
-			tty->ldisc.flush_buffer(tty);
+		if ( tty->ldisc.ops->flush_buffer )  
+			tty->ldisc.ops->flush_buffer(tty);
 		i2InputFlush( tty->driver_data );
 	}
 }
@@ -1342,7 +1343,7 @@ static void do_status(struct work_struct *work)
 		}
 		tmp = pCh->pTTY->real_raw;
 		pCh->pTTY->real_raw = 0;
-		pCh->pTTY->ldisc.receive_buf( pCh->pTTY, &brkc, &brkf, 1 );
+		pCh->pTTY->ldisc->ops.receive_buf( pCh->pTTY, &brkc, &brkf, 1 );
 		pCh->pTTY->real_raw = tmp;
 	}
 #endif /* NEVER_HAPPENS_AS_SETUP_XXX */
diff --git a/drivers/char/n_hdlc.c b/drivers/char/n_hdlc.c
index a35bfd7ee80e..ed4e03333ab4 100644
--- a/drivers/char/n_hdlc.c
+++ b/drivers/char/n_hdlc.c
@@ -199,7 +199,7 @@ static void n_hdlc_tty_wakeup(struct tty_struct *tty);
 #define tty2n_hdlc(tty)	((struct n_hdlc *) ((tty)->disc_data))
 #define n_hdlc2tty(n_hdlc)	((n_hdlc)->tty)
 
-static struct tty_ldisc n_hdlc_ldisc = {
+static struct tty_ldisc_ops n_hdlc_ldisc = {
 	.owner		= THIS_MODULE,
 	.magic		= TTY_LDISC_MAGIC,
 	.name		= "hdlc",
@@ -342,8 +342,8 @@ static int n_hdlc_tty_open (struct tty_struct *tty)
 #endif
 	
 	/* Flush any pending characters in the driver and discipline. */
-	if (tty->ldisc.flush_buffer)
-		tty->ldisc.flush_buffer(tty);
+	if (tty->ldisc.ops->flush_buffer)
+		tty->ldisc.ops->flush_buffer(tty);
 
 	tty_driver_flush_buffer(tty);
 		
diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c
index 902169062332..ae377aa473ba 100644
--- a/drivers/char/n_r3964.c
+++ b/drivers/char/n_r3964.c
@@ -143,7 +143,7 @@ static unsigned int r3964_poll(struct tty_struct *tty, struct file *file,
 static void r3964_receive_buf(struct tty_struct *tty, const unsigned char *cp,
 		char *fp, int count);
 
-static struct tty_ldisc tty_ldisc_N_R3964 = {
+static struct tty_ldisc_ops tty_ldisc_N_R3964 = {
 	.owner = THIS_MODULE,
 	.magic = TTY_LDISC_MAGIC,
 	.name = "R3964",
diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c
index 8096389b0dc2..708c2b1dbe51 100644
--- a/drivers/char/n_tty.c
+++ b/drivers/char/n_tty.c
@@ -1573,7 +1573,7 @@ static unsigned int normal_poll(struct tty_struct *tty, struct file *file,
 	return mask;
 }
 
-struct tty_ldisc tty_ldisc_N_TTY = {
+struct tty_ldisc_ops tty_ldisc_N_TTY = {
 	.magic           = TTY_LDISC_MAGIC,
 	.name            = "n_tty",
 	.open            = n_tty_open,
diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 1dd0e992c83d..95743e2682fe 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -514,8 +514,8 @@ static void ldisc_receive_buf(struct tty_struct *tty,
 		return;
 	ld = tty_ldisc_ref(tty);
 	if (ld) {
-		if (ld->receive_buf)
-			ld->receive_buf(tty, data, flags, count);
+		if (ld->ops->receive_buf)
+			ld->ops->receive_buf(tty, data, flags, count);
 		tty_ldisc_deref(ld);
 	}
 }
diff --git a/drivers/char/pty.c b/drivers/char/pty.c
index 0a05c038ae6f..76b27932d229 100644
--- a/drivers/char/pty.c
+++ b/drivers/char/pty.c
@@ -111,7 +111,7 @@ static int pty_write(struct tty_struct * tty, const unsigned char *buf, int coun
 	c = to->receive_room;
 	if (c > count)
 		c = count;
-	to->ldisc.receive_buf(to, buf, NULL, c);
+	to->ldisc.ops->receive_buf(to, buf, NULL, c);
 	
 	return c;
 }
@@ -149,11 +149,11 @@ static int pty_chars_in_buffer(struct tty_struct *tty)
 	int count;
 
 	/* We should get the line discipline lock for "tty->link" */
-	if (!to || !to->ldisc.chars_in_buffer)
+	if (!to || !to->ldisc.ops->chars_in_buffer)
 		return 0;
 
 	/* The ldisc must report 0 if no characters available to be read */
-	count = to->ldisc.chars_in_buffer(to);
+	count = to->ldisc.ops->chars_in_buffer(to);
 
 	if (tty->driver->subtype == PTY_TYPE_SLAVE) return count;
 
@@ -186,8 +186,8 @@ static void pty_flush_buffer(struct tty_struct *tty)
 	if (!to)
 		return;
 	
-	if (to->ldisc.flush_buffer)
-		to->ldisc.flush_buffer(to);
+	if (to->ldisc.ops->flush_buffer)
+		to->ldisc.ops->flush_buffer(to);
 	
 	if (to->packet) {
 		spin_lock_irqsave(&tty->ctrl_lock, flags);
diff --git a/drivers/char/selection.c b/drivers/char/selection.c
index d63f5ccc29e6..2978a49a172b 100644
--- a/drivers/char/selection.c
+++ b/drivers/char/selection.c
@@ -327,7 +327,8 @@ int paste_selection(struct tty_struct *tty)
 		}
 		count = sel_buffer_lth - pasted;
 		count = min(count, tty->receive_room);
-		tty->ldisc.receive_buf(tty, sel_buffer + pasted, NULL, count);
+		tty->ldisc.ops->receive_buf(tty, sel_buffer + pasted,
+								NULL, count);
 		pasted += count;
 	}
 	remove_wait_queue(&vc->paste_wait, &wait);
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index ac5080df2565..5e4b2e638d0c 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -975,8 +975,8 @@ static void ldisc_receive_buf(struct tty_struct *tty,
 		return;
 	ld = tty_ldisc_ref(tty);
 	if (ld) {
-		if (ld->receive_buf)
-			ld->receive_buf(tty, data, flags, count);
+		if (ld->ops->receive_buf)
+			ld->ops->receive_buf(tty, data, flags, count);
 		tty_ldisc_deref(ld);
 	}
 }
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 55c1653be00c..e473778cd6fa 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -641,8 +641,8 @@ static void ldisc_receive_buf(struct tty_struct *tty,
 		return;
 	ld = tty_ldisc_ref(tty);
 	if (ld) {
-		if (ld->receive_buf)
-			ld->receive_buf(tty, data, flags, count);
+		if (ld->ops->receive_buf)
+			ld->ops->receive_buf(tty, data, flags, count);
 		tty_ldisc_deref(ld);
 	}
 }
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index bec54866e0bb..5341b5aaf8bc 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -712,8 +712,8 @@ static void ldisc_receive_buf(struct tty_struct *tty,
 		return;
 	ld = tty_ldisc_ref(tty);
 	if (ld) {
-		if (ld->receive_buf)
-			ld->receive_buf(tty, data, flags, count);
+		if (ld->ops->receive_buf)
+			ld->ops->receive_buf(tty, data, flags, count);
 		tty_ldisc_deref(ld);
 	}
 }
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 047a17339f83..54c4ada460ee 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -95,8 +95,9 @@
 #include <linux/wait.h>
 #include <linux/bitops.h>
 #include <linux/delay.h>
+#include <linux/seq_file.h>
 
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <asm/system.h>
 
 #include <linux/kbd_kern.h>
@@ -682,7 +683,7 @@ static void tty_set_termios_ldisc(struct tty_struct *tty, int num)
 static DEFINE_SPINLOCK(tty_ldisc_lock);
 static DECLARE_WAIT_QUEUE_HEAD(tty_ldisc_wait);
 /* Line disc dispatch table */
-static struct tty_ldisc tty_ldiscs[NR_LDISCS];
+static struct tty_ldisc_ops *tty_ldiscs[NR_LDISCS];
 
 /**
  *	tty_register_ldisc	-	install a line discipline
@@ -697,7 +698,7 @@ static struct tty_ldisc tty_ldiscs[NR_LDISCS];
  *		takes tty_ldisc_lock to guard against ldisc races
  */
 
-int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc)
+int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc)
 {
 	unsigned long flags;
 	int ret = 0;
@@ -706,10 +707,9 @@ int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc)
 		return -EINVAL;
 
 	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	tty_ldiscs[disc] = *new_ldisc;
-	tty_ldiscs[disc].num = disc;
-	tty_ldiscs[disc].flags |= LDISC_FLAG_DEFINED;
-	tty_ldiscs[disc].refcount = 0;
+	tty_ldiscs[disc] = new_ldisc;
+	new_ldisc->num = disc;
+	new_ldisc->refcount = 0;
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 
 	return ret;
@@ -737,19 +737,56 @@ int tty_unregister_ldisc(int disc)
 		return -EINVAL;
 
 	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	if (tty_ldiscs[disc].refcount)
+	if (tty_ldiscs[disc]->refcount)
 		ret = -EBUSY;
 	else
-		tty_ldiscs[disc].flags &= ~LDISC_FLAG_DEFINED;
+		tty_ldiscs[disc] = NULL;
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 
 	return ret;
 }
 EXPORT_SYMBOL(tty_unregister_ldisc);
 
+
+/**
+ *	tty_ldisc_try_get	-	try and reference an ldisc
+ *	@disc: ldisc number
+ *	@ld: tty ldisc structure to complete
+ *
+ *	Attempt to open and lock a line discipline into place. Return
+ *	the line discipline refcounted and assigned in ld. On an error
+ *	report the error code back
+ */
+
+static int tty_ldisc_try_get(int disc, struct tty_ldisc *ld)
+{
+	unsigned long flags;
+	struct tty_ldisc_ops *ldops;
+	int err = -EINVAL;
+	
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	ld->ops = NULL;
+	ldops = tty_ldiscs[disc];
+	/* Check the entry is defined */
+	if (ldops) {
+		/* If the module is being unloaded we can't use it */
+		if (!try_module_get(ldops->owner))
+			err = -EAGAIN;
+		else {
+			/* lock it */
+			ldops->refcount++;
+			ld->ops = ldops;
+			err = 0;
+		}
+	}
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+	return err;
+}
+
 /**
  *	tty_ldisc_get		-	take a reference to an ldisc
  *	@disc: ldisc number
+ *	@ld: tty line discipline structure to use
  *
  *	Takes a reference to a line discipline. Deals with refcounts and
  *	module locking counts. Returns NULL if the discipline is not available.
@@ -760,32 +797,20 @@ EXPORT_SYMBOL(tty_unregister_ldisc);
  *		takes tty_ldisc_lock to guard against ldisc races
  */
 
-struct tty_ldisc *tty_ldisc_get(int disc)
+static int tty_ldisc_get(int disc, struct tty_ldisc *ld)
 {
-	unsigned long flags;
-	struct tty_ldisc *ld;
+	int err;
 
 	if (disc < N_TTY || disc >= NR_LDISCS)
-		return NULL;
-
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-
-	ld = &tty_ldiscs[disc];
-	/* Check the entry is defined */
-	if (ld->flags & LDISC_FLAG_DEFINED) {
-		/* If the module is being unloaded we can't use it */
-		if (!try_module_get(ld->owner))
-			ld = NULL;
-		else /* lock it */
-			ld->refcount++;
-	} else
-		ld = NULL;
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-	return ld;
+		return -EINVAL;
+	err = tty_ldisc_try_get(disc, ld);
+	if (err == -EAGAIN) {
+		request_module("tty-ldisc-%d", disc);
+		err = tty_ldisc_try_get(disc, ld);
+	}
+	return err;
 }
 
-EXPORT_SYMBOL_GPL(tty_ldisc_get);
-
 /**
  *	tty_ldisc_put		-	drop ldisc reference
  *	@disc: ldisc number
@@ -797,22 +822,67 @@ EXPORT_SYMBOL_GPL(tty_ldisc_get);
  *		takes tty_ldisc_lock to guard against ldisc races
  */
 
-void tty_ldisc_put(int disc)
+static void tty_ldisc_put(struct tty_ldisc_ops *ld)
 {
-	struct tty_ldisc *ld;
 	unsigned long flags;
+	int disc = ld->num;
 
 	BUG_ON(disc < N_TTY || disc >= NR_LDISCS);
 
 	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	ld = &tty_ldiscs[disc];
+	ld = tty_ldiscs[disc];
 	BUG_ON(ld->refcount == 0);
 	ld->refcount--;
 	module_put(ld->owner);
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 }
 
-EXPORT_SYMBOL_GPL(tty_ldisc_put);
+static void * tty_ldiscs_seq_start(struct seq_file *m, loff_t *pos)
+{
+	return (*pos < NR_LDISCS) ? pos : NULL;
+}
+
+static void * tty_ldiscs_seq_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	(*pos)++;
+	return (*pos < NR_LDISCS) ? pos : NULL;
+}
+
+static void tty_ldiscs_seq_stop(struct seq_file *m, void *v)
+{
+}
+
+static int tty_ldiscs_seq_show(struct seq_file *m, void *v)
+{
+	int i = *(loff_t *)v;
+	struct tty_ldisc ld;
+	
+	if (tty_ldisc_get(i, &ld) < 0)
+		return 0;
+	seq_printf(m, "%-10s %2d\n", ld.ops->name ? ld.ops->name : "???", i);
+	tty_ldisc_put(ld.ops);
+	return 0;
+}
+
+static const struct seq_operations tty_ldiscs_seq_ops = {
+	.start	= tty_ldiscs_seq_start,
+	.next	= tty_ldiscs_seq_next,
+	.stop	= tty_ldiscs_seq_stop,
+	.show	= tty_ldiscs_seq_show,
+};
+
+static int proc_tty_ldiscs_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &tty_ldiscs_seq_ops);
+}
+
+const struct file_operations tty_ldiscs_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= proc_tty_ldiscs_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
 
 /**
  *	tty_ldisc_assign	-	set ldisc on a tty
@@ -829,8 +899,8 @@ EXPORT_SYMBOL_GPL(tty_ldisc_put);
 
 static void tty_ldisc_assign(struct tty_struct *tty, struct tty_ldisc *ld)
 {
+	ld->refcount = 0;
 	tty->ldisc = *ld;
-	tty->ldisc.refcount = 0;
 }
 
 /**
@@ -953,6 +1023,41 @@ static void tty_ldisc_enable(struct tty_struct *tty)
 	wake_up(&tty_ldisc_wait);
 }
 
+/**
+ *	tty_ldisc_restore	-	helper for tty ldisc change
+ *	@tty: tty to recover
+ *	@old: previous ldisc
+ *
+ *	Restore the previous line discipline or N_TTY when a line discipline
+ *	change fails due to an open error
+ */
+
+static void tty_ldisc_restore(struct tty_struct *tty, struct tty_ldisc *old)
+{
+	char buf[64];
+	struct tty_ldisc new_ldisc;
+
+	/* There is an outstanding reference here so this is safe */
+	tty_ldisc_get(old->ops->num, old);
+	tty_ldisc_assign(tty, old);
+	tty_set_termios_ldisc(tty, old->ops->num);
+	if (old->ops->open && (old->ops->open(tty) < 0)) {
+		tty_ldisc_put(old->ops);
+		/* This driver is always present */
+		if (tty_ldisc_get(N_TTY, &new_ldisc) < 0)
+			panic("n_tty: get");
+		tty_ldisc_assign(tty, &new_ldisc);
+		tty_set_termios_ldisc(tty, N_TTY);
+		if (new_ldisc.ops->open) {
+			int r = new_ldisc.ops->open(tty);
+				if (r < 0)
+				panic("Couldn't open N_TTY ldisc for "
+				      "%s --- error %d.",
+				      tty_name(tty, buf), r);
+		}
+	}
+}
+
 /**
  *	tty_set_ldisc		-	set line discipline
  *	@tty: the terminal to set
@@ -967,28 +1072,18 @@ static void tty_ldisc_enable(struct tty_struct *tty)
 
 static int tty_set_ldisc(struct tty_struct *tty, int ldisc)
 {
-	int retval = 0;
-	struct tty_ldisc o_ldisc;
-	char buf[64];
+	int retval;
+	struct tty_ldisc o_ldisc, new_ldisc;
 	int work;
 	unsigned long flags;
-	struct tty_ldisc *ld;
 	struct tty_struct *o_tty;
 
-	if ((ldisc < N_TTY) || (ldisc >= NR_LDISCS))
-		return -EINVAL;
-
 restart:
-
-	ld = tty_ldisc_get(ldisc);
-	/* Eduardo Blanco <ejbs@cs.cs.com.uy> */
-	/* Cyrus Durgin <cider@speakeasy.org> */
-	if (ld == NULL) {
-		request_module("tty-ldisc-%d", ldisc);
-		ld = tty_ldisc_get(ldisc);
-	}
-	if (ld == NULL)
-		return -EINVAL;
+	/* This is a bit ugly for now but means we can break the 'ldisc
+	   is part of the tty struct' assumption later */
+	retval = tty_ldisc_get(ldisc, &new_ldisc);
+	if (retval)
+		return retval;
 
 	/*
 	 *	Problem: What do we do if this blocks ?
@@ -996,8 +1091,8 @@ restart:
 
 	tty_wait_until_sent(tty, 0);
 
-	if (tty->ldisc.num == ldisc) {
-		tty_ldisc_put(ldisc);
+	if (tty->ldisc.ops->num == ldisc) {
+		tty_ldisc_put(new_ldisc.ops);
 		return 0;
 	}
 
@@ -1024,7 +1119,7 @@ restart:
 			/* Free the new ldisc we grabbed. Must drop the lock
 			   first. */
 			spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-			tty_ldisc_put(ldisc);
+			tty_ldisc_put(o_ldisc.ops);
 			/*
 			 * There are several reasons we may be busy, including
 			 * random momentary I/O traffic. We must therefore
@@ -1038,7 +1133,7 @@ restart:
 		}
 		if (o_tty && o_tty->ldisc.refcount) {
 			spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-			tty_ldisc_put(ldisc);
+			tty_ldisc_put(o_tty->ldisc.ops);
 			if (wait_event_interruptible(tty_ldisc_wait, o_tty->ldisc.refcount == 0) < 0)
 				return -ERESTARTSYS;
 			goto restart;
@@ -1049,8 +1144,9 @@ restart:
 	 *	another ldisc change
 	 */
 	if (!test_bit(TTY_LDISC, &tty->flags)) {
+		struct tty_ldisc *ld;
 		spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-		tty_ldisc_put(ldisc);
+		tty_ldisc_put(new_ldisc.ops);
 		ld = tty_ldisc_ref_wait(tty);
 		tty_ldisc_deref(ld);
 		goto restart;
@@ -1060,7 +1156,7 @@ restart:
 	if (o_tty)
 		clear_bit(TTY_LDISC, &o_tty->flags);
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-
+	
 	/*
 	 *	From this point on we know nobody has an ldisc
 	 *	usage reference, nor can they obtain one until
@@ -1070,45 +1166,30 @@ restart:
 	work = cancel_delayed_work(&tty->buf.work);
 	/*
 	 * Wait for ->hangup_work and ->buf.work handlers to terminate
+	 * MUST NOT hold locks here.
 	 */
 	flush_scheduled_work();
 	/* Shutdown the current discipline. */
-	if (tty->ldisc.close)
-		(tty->ldisc.close)(tty);
+	if (o_ldisc.ops->close)
+		(o_ldisc.ops->close)(tty);
 
 	/* Now set up the new line discipline. */
-	tty_ldisc_assign(tty, ld);
+	tty_ldisc_assign(tty, &new_ldisc);
 	tty_set_termios_ldisc(tty, ldisc);
-	if (tty->ldisc.open)
-		retval = (tty->ldisc.open)(tty);
+	if (new_ldisc.ops->open)
+		retval = (new_ldisc.ops->open)(tty);
 	if (retval < 0) {
-		tty_ldisc_put(ldisc);
-		/* There is an outstanding reference here so this is safe */
-		tty_ldisc_assign(tty, tty_ldisc_get(o_ldisc.num));
-		tty_set_termios_ldisc(tty, tty->ldisc.num);
-		if (tty->ldisc.open && (tty->ldisc.open(tty) < 0)) {
-			tty_ldisc_put(o_ldisc.num);
-			/* This driver is always present */
-			tty_ldisc_assign(tty, tty_ldisc_get(N_TTY));
-			tty_set_termios_ldisc(tty, N_TTY);
-			if (tty->ldisc.open) {
-				int r = tty->ldisc.open(tty);
-
-				if (r < 0)
-					panic("Couldn't open N_TTY ldisc for "
-					      "%s --- error %d.",
-					      tty_name(tty, buf), r);
-			}
-		}
+		tty_ldisc_put(new_ldisc.ops);
+		tty_ldisc_restore(tty, &o_ldisc);
 	}
 	/* At this point we hold a reference to the new ldisc and a
 	   a reference to the old ldisc. If we ended up flipping back
 	   to the existing ldisc we have two references to it */
 
-	if (tty->ldisc.num != o_ldisc.num && tty->ops->set_ldisc)
+	if (tty->ldisc.ops->num != o_ldisc.ops->num && tty->ops->set_ldisc)
 		tty->ops->set_ldisc(tty);
 
-	tty_ldisc_put(o_ldisc.num);
+	tty_ldisc_put(o_ldisc.ops);
 
 	/*
 	 *	Allow ldisc referencing to occur as soon as the driver
@@ -1335,8 +1416,8 @@ void tty_wakeup(struct tty_struct *tty)
 	if (test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags)) {
 		ld = tty_ldisc_ref(tty);
 		if (ld) {
-			if (ld->write_wakeup)
-				ld->write_wakeup(tty);
+			if (ld->ops->write_wakeup)
+				ld->ops->write_wakeup(tty);
 			tty_ldisc_deref(ld);
 		}
 	}
@@ -1357,8 +1438,8 @@ void tty_ldisc_flush(struct tty_struct *tty)
 {
 	struct tty_ldisc *ld = tty_ldisc_ref(tty);
 	if (ld) {
-		if (ld->flush_buffer)
-			ld->flush_buffer(tty);
+		if (ld->ops->flush_buffer)
+			ld->ops->flush_buffer(tty);
 		tty_ldisc_deref(ld);
 	}
 	tty_buffer_flush(tty);
@@ -1386,7 +1467,7 @@ static void tty_reset_termios(struct tty_struct *tty)
  *	do_tty_hangup		-	actual handler for hangup events
  *	@work: tty device
  *
- *	This can be called by the "eventd" kernel thread.  That is process
+k *	This can be called by the "eventd" kernel thread.  That is process
  *	synchronous but doesn't hold any locks, so we need to make sure we
  *	have the appropriate locks for what we're doing.
  *
@@ -1449,14 +1530,14 @@ static void do_tty_hangup(struct work_struct *work)
 	ld = tty_ldisc_ref(tty);
 	if (ld != NULL) {
 		/* We may have no line discipline at this point */
-		if (ld->flush_buffer)
-			ld->flush_buffer(tty);
+		if (ld->ops->flush_buffer)
+			ld->ops->flush_buffer(tty);
 		tty_driver_flush_buffer(tty);
 		if ((test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags)) &&
-		    ld->write_wakeup)
-			ld->write_wakeup(tty);
-		if (ld->hangup)
-			ld->hangup(tty);
+		    ld->ops->write_wakeup)
+			ld->ops->write_wakeup(tty);
+		if (ld->ops->hangup)
+			ld->ops->hangup(tty);
 	}
 	/*
 	 * FIXME: Once we trust the LDISC code better we can wait here for
@@ -1825,8 +1906,8 @@ static ssize_t tty_read(struct file *file, char __user *buf, size_t count,
 	/* We want to wait for the line discipline to sort out in this
 	   situation */
 	ld = tty_ldisc_ref_wait(tty);
-	if (ld->read)
-		i = (ld->read)(tty, file, buf, count);
+	if (ld->ops->read)
+		i = (ld->ops->read)(tty, file, buf, count);
 	else
 		i = -EIO;
 	tty_ldisc_deref(ld);
@@ -1978,10 +2059,10 @@ static ssize_t tty_write(struct file *file, const char __user *buf,
 		printk(KERN_ERR "tty driver %s lacks a write_room method.\n",
 			tty->driver->name);
 	ld = tty_ldisc_ref_wait(tty);
-	if (!ld->write)
+	if (!ld->ops->write)
 		ret = -EIO;
 	else
-		ret = do_tty_write(ld->write, tty, file, buf, count);
+		ret = do_tty_write(ld->ops->write, tty, file, buf, count);
 	tty_ldisc_deref(ld);
 	return ret;
 }
@@ -2076,6 +2157,7 @@ static int init_dev(struct tty_driver *driver, int idx,
 	struct ktermios *tp, **tp_loc, *o_tp, **o_tp_loc;
 	struct ktermios *ltp, **ltp_loc, *o_ltp, **o_ltp_loc;
 	int retval = 0;
+	struct tty_ldisc *ld;
 
 	/* check whether we're reopening an existing tty */
 	if (driver->flags & TTY_DRIVER_DEVPTS_MEM) {
@@ -2224,17 +2306,19 @@ static int init_dev(struct tty_driver *driver, int idx,
 	 * If we fail here just call release_tty to clean up.  No need
 	 * to decrement the use counts, as release_tty doesn't care.
 	 */
+	 
+	ld = &tty->ldisc;
 
-	if (tty->ldisc.open) {
-		retval = (tty->ldisc.open)(tty);
+	if (ld->ops->open) {
+		retval = (ld->ops->open)(tty);
 		if (retval)
 			goto release_mem_out;
 	}
-	if (o_tty && o_tty->ldisc.open) {
-		retval = (o_tty->ldisc.open)(o_tty);
+	if (o_tty && o_tty->ldisc.ops->open) {
+		retval = (o_tty->ldisc.ops->open)(o_tty);
 		if (retval) {
-			if (tty->ldisc.close)
-				(tty->ldisc.close)(tty);
+			if (ld->ops->close)
+				(ld->ops->close)(tty);
 			goto release_mem_out;
 		}
 		tty_ldisc_enable(o_tty);
@@ -2378,6 +2462,7 @@ static void release_tty(struct tty_struct *tty, int idx)
 static void release_dev(struct file *filp)
 {
 	struct tty_struct *tty, *o_tty;
+	struct tty_ldisc ld;
 	int	pty_master, tty_closing, o_tty_closing, do_sleep;
 	int	devpts;
 	int	idx;
@@ -2611,26 +2696,27 @@ static void release_dev(struct file *filp)
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 	/*
 	 * Shutdown the current line discipline, and reset it to N_TTY.
-	 * N.B. why reset ldisc when we're releasing the memory??
 	 *
 	 * FIXME: this MUST get fixed for the new reflocking
 	 */
-	if (tty->ldisc.close)
-		(tty->ldisc.close)(tty);
-	tty_ldisc_put(tty->ldisc.num);
+	if (tty->ldisc.ops->close)
+		(tty->ldisc.ops->close)(tty);
+	tty_ldisc_put(tty->ldisc.ops);
 
 	/*
 	 *	Switch the line discipline back
 	 */
-	tty_ldisc_assign(tty, tty_ldisc_get(N_TTY));
+	WARN_ON(tty_ldisc_get(N_TTY, &ld));
+	tty_ldisc_assign(tty, &ld);
 	tty_set_termios_ldisc(tty, N_TTY);
 	if (o_tty) {
 		/* FIXME: could o_tty be in setldisc here ? */
 		clear_bit(TTY_LDISC, &o_tty->flags);
-		if (o_tty->ldisc.close)
-			(o_tty->ldisc.close)(o_tty);
-		tty_ldisc_put(o_tty->ldisc.num);
-		tty_ldisc_assign(o_tty, tty_ldisc_get(N_TTY));
+		if (o_tty->ldisc.ops->close)
+			(o_tty->ldisc.ops->close)(o_tty);
+		tty_ldisc_put(o_tty->ldisc.ops);
+		WARN_ON(tty_ldisc_get(N_TTY, &ld));
+		tty_ldisc_assign(o_tty, &ld);
 		tty_set_termios_ldisc(o_tty, N_TTY);
 	}
 	/*
@@ -2899,8 +2985,8 @@ static unsigned int tty_poll(struct file *filp, poll_table *wait)
 		return 0;
 
 	ld = tty_ldisc_ref_wait(tty);
-	if (ld->poll)
-		ret = (ld->poll)(tty, filp, wait);
+	if (ld->ops->poll)
+		ret = (ld->ops->poll)(tty, filp, wait);
 	tty_ldisc_deref(ld);
 	return ret;
 }
@@ -2974,7 +3060,7 @@ static int tiocsti(struct tty_struct *tty, char __user *p)
 	if (get_user(ch, p))
 		return -EFAULT;
 	ld = tty_ldisc_ref_wait(tty);
-	ld->receive_buf(tty, &ch, &mbz, 1);
+	ld->ops->receive_buf(tty, &ch, &mbz, 1);
 	tty_ldisc_deref(ld);
 	return 0;
 }
@@ -3528,7 +3614,7 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case TIOCGSID:
 		return tiocgsid(tty, real_tty, p);
 	case TIOCGETD:
-		return put_user(tty->ldisc.num, (int __user *)p);
+		return put_user(tty->ldisc.ops->num, (int __user *)p);
 	case TIOCSETD:
 		return tiocsetd(tty, p);
 #ifdef CONFIG_VT
@@ -3581,8 +3667,8 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	}
 	ld = tty_ldisc_ref_wait(tty);
 	retval = -EINVAL;
-	if (ld->ioctl) {
-		retval = ld->ioctl(tty, file, cmd, arg);
+	if (ld->ops->ioctl) {
+		retval = ld->ops->ioctl(tty, file, cmd, arg);
 		if (retval == -ENOIOCTLCMD)
 			retval = -EINVAL;
 	}
@@ -3609,8 +3695,8 @@ static long tty_compat_ioctl(struct file *file, unsigned int cmd,
 	}
 
 	ld = tty_ldisc_ref_wait(tty);
-	if (ld->compat_ioctl)
-		retval = ld->compat_ioctl(tty, file, cmd, arg);
+	if (ld->ops->compat_ioctl)
+		retval = ld->ops->compat_ioctl(tty, file, cmd, arg);
 	tty_ldisc_deref(ld);
 
 	return retval;
@@ -3782,7 +3868,8 @@ static void flush_to_ldisc(struct work_struct *work)
 			flag_buf = head->flag_buf_ptr + head->read;
 			head->read += count;
 			spin_unlock_irqrestore(&tty->buf.lock, flags);
-			disc->receive_buf(tty, char_buf, flag_buf, count);
+			disc->ops->receive_buf(tty, char_buf,
+							flag_buf, count);
 			spin_lock_irqsave(&tty->buf.lock, flags);
 		}
 		/* Restore the queue head */
@@ -3843,9 +3930,12 @@ EXPORT_SYMBOL(tty_flip_buffer_push);
 
 static void initialize_tty_struct(struct tty_struct *tty)
 {
+	struct tty_ldisc ld;
 	memset(tty, 0, sizeof(struct tty_struct));
 	tty->magic = TTY_MAGIC;
-	tty_ldisc_assign(tty, tty_ldisc_get(N_TTY));
+	if (tty_ldisc_get(N_TTY, &ld) < 0)
+		panic("n_tty: init_tty");
+	tty_ldisc_assign(tty, &ld);
 	tty->session = NULL;
 	tty->pgrp = NULL;
 	tty->overrun_time = jiffies;
diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c
index 8f81139d6194..ea9fc5d03b99 100644
--- a/drivers/char/tty_ioctl.c
+++ b/drivers/char/tty_ioctl.c
@@ -491,8 +491,8 @@ static void change_termios(struct tty_struct *tty, struct ktermios *new_termios)
 
 	ld = tty_ldisc_ref(tty);
 	if (ld != NULL) {
-		if (ld->set_termios)
-			(ld->set_termios)(tty, &old_termios);
+		if (ld->ops->set_termios)
+			(ld->ops->set_termios)(tty, &old_termios);
 		tty_ldisc_deref(ld);
 	}
 	mutex_unlock(&tty->termios_mutex);
@@ -552,8 +552,8 @@ static int set_termios(struct tty_struct *tty, void __user *arg, int opt)
 	ld = tty_ldisc_ref(tty);
 
 	if (ld != NULL) {
-		if ((opt & TERMIOS_FLUSH) && ld->flush_buffer)
-			ld->flush_buffer(tty);
+		if ((opt & TERMIOS_FLUSH) && ld->ops->flush_buffer)
+			ld->ops->flush_buffer(tty);
 		tty_ldisc_deref(ld);
 	}
 
@@ -959,12 +959,12 @@ int tty_perform_flush(struct tty_struct *tty, unsigned long arg)
 	ld = tty_ldisc_ref(tty);
 	switch (arg) {
 	case TCIFLUSH:
-		if (ld && ld->flush_buffer)
-			ld->flush_buffer(tty);
+		if (ld && ld->ops->flush_buffer)
+			ld->ops->flush_buffer(tty);
 		break;
 	case TCIOFLUSH:
-		if (ld && ld->flush_buffer)
-			ld->flush_buffer(tty);
+		if (ld && ld->ops->flush_buffer)
+			ld->ops->flush_buffer(tty);
 		/* fall through */
 	case TCOFLUSH:
 		tty_driver_flush_buffer(tty);
diff --git a/drivers/input/serio/serport.c b/drivers/input/serio/serport.c
index 7ff71ba7b7c9..b9694b6445d0 100644
--- a/drivers/input/serio/serport.c
+++ b/drivers/input/serio/serport.c
@@ -216,7 +216,7 @@ static void serport_ldisc_write_wakeup(struct tty_struct * tty)
  * The line discipline structure.
  */
 
-static struct tty_ldisc serport_ldisc = {
+static struct tty_ldisc_ops serport_ldisc = {
 	.owner =	THIS_MODULE,
 	.name =		"input",
 	.open =		serport_ldisc_open,
diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c
index 2095153582f1..8a35029caca0 100644
--- a/drivers/isdn/capi/capi.c
+++ b/drivers/isdn/capi/capi.c
@@ -466,7 +466,7 @@ static int handle_recv_skb(struct capiminor *mp, struct sk_buff *skb)
 	ld = tty_ldisc_ref(mp->tty);
 	if (ld == NULL)
 		return -1;
-	if (ld->receive_buf == NULL) {
+	if (ld->ops->receive_buf == NULL) {
 #if defined(_DEBUG_DATAFLOW) || defined(_DEBUG_TTYFUNCS)
 		printk(KERN_DEBUG "capi: ldisc has no receive_buf function\n");
 #endif
@@ -501,7 +501,7 @@ static int handle_recv_skb(struct capiminor *mp, struct sk_buff *skb)
 	printk(KERN_DEBUG "capi: DATA_B3_RESP %u len=%d => ldisc\n",
 				datahandle, skb->len);
 #endif
-	ld->receive_buf(mp->tty, skb->data, NULL, skb->len);
+	ld->ops->receive_buf(mp->tty, skb->data, NULL, skb->len);
 	kfree_skb(skb);
 	tty_ldisc_deref(ld);
 	return 0;
diff --git a/drivers/isdn/gigaset/ser-gigaset.c b/drivers/isdn/gigaset/ser-gigaset.c
index 45d1ee93cd39..5e89fa177816 100644
--- a/drivers/isdn/gigaset/ser-gigaset.c
+++ b/drivers/isdn/gigaset/ser-gigaset.c
@@ -766,7 +766,7 @@ gigaset_tty_wakeup(struct tty_struct *tty)
 	cs_put(cs);
 }
 
-static struct tty_ldisc gigaset_ldisc = {
+static struct tty_ldisc_ops gigaset_ldisc = {
 	.owner		= THIS_MODULE,
 	.magic		= TTY_LDISC_MAGIC,
 	.name		= "ser_gigaset",
diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c
index 9d5721287d6f..19dd0a61749c 100644
--- a/drivers/net/hamradio/6pack.c
+++ b/drivers/net/hamradio/6pack.c
@@ -783,7 +783,7 @@ static int sixpack_ioctl(struct tty_struct *tty, struct file *file,
 	return err;
 }
 
-static struct tty_ldisc sp_ldisc = {
+static struct tty_ldisc_ops sp_ldisc = {
 	.owner		= THIS_MODULE,
 	.magic		= TTY_LDISC_MAGIC,
 	.name		= "6pack",
diff --git a/drivers/net/hamradio/mkiss.c b/drivers/net/hamradio/mkiss.c
index 65166035aca0..c6ca47599fd4 100644
--- a/drivers/net/hamradio/mkiss.c
+++ b/drivers/net/hamradio/mkiss.c
@@ -969,7 +969,7 @@ out:
 	mkiss_put(ax);
 }
 
-static struct tty_ldisc ax_ldisc = {
+static struct tty_ldisc_ops ax_ldisc = {
 	.owner		= THIS_MODULE,
 	.magic		= TTY_LDISC_MAGIC,
 	.name		= "mkiss",
diff --git a/drivers/net/irda/irtty-sir.c b/drivers/net/irda/irtty-sir.c
index e6f40b7f9041..9e33196f9459 100644
--- a/drivers/net/irda/irtty-sir.c
+++ b/drivers/net/irda/irtty-sir.c
@@ -533,7 +533,7 @@ static void irtty_close(struct tty_struct *tty)
 
 /* ------------------------------------------------------- */
 
-static struct tty_ldisc irda_ldisc = {
+static struct tty_ldisc_ops irda_ldisc = {
 	.magic		= TTY_LDISC_MAGIC,
  	.name		= "irda",
 	.flags		= 0,
diff --git a/drivers/net/ppp_async.c b/drivers/net/ppp_async.c
index f1a52def1241..451bdb57d6fc 100644
--- a/drivers/net/ppp_async.c
+++ b/drivers/net/ppp_async.c
@@ -378,7 +378,7 @@ ppp_asynctty_wakeup(struct tty_struct *tty)
 }
 
 
-static struct tty_ldisc ppp_ldisc = {
+static struct tty_ldisc_ops ppp_ldisc = {
 	.owner  = THIS_MODULE,
 	.magic	= TTY_LDISC_MAGIC,
 	.name	= "ppp",
diff --git a/drivers/net/ppp_synctty.c b/drivers/net/ppp_synctty.c
index b8f0369a71e7..801d8f99d471 100644
--- a/drivers/net/ppp_synctty.c
+++ b/drivers/net/ppp_synctty.c
@@ -418,7 +418,7 @@ ppp_sync_wakeup(struct tty_struct *tty)
 }
 
 
-static struct tty_ldisc ppp_sync_ldisc = {
+static struct tty_ldisc_ops ppp_sync_ldisc = {
 	.owner	= THIS_MODULE,
 	.magic	= TTY_LDISC_MAGIC,
 	.name	= "pppsync",
diff --git a/drivers/net/slip.c b/drivers/net/slip.c
index 84af68fdb6c2..1d58991d395b 100644
--- a/drivers/net/slip.c
+++ b/drivers/net/slip.c
@@ -1301,7 +1301,7 @@ static int sl_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 #endif
 /* VSV changes end */
 
-static struct tty_ldisc	sl_ldisc = {
+static struct tty_ldisc_ops sl_ldisc = {
 	.owner 		= THIS_MODULE,
 	.magic 		= TTY_LDISC_MAGIC,
 	.name 		= "slip",
diff --git a/drivers/net/wan/pc300_tty.c b/drivers/net/wan/pc300_tty.c
index e03eef2f2282..c2c10c632260 100644
--- a/drivers/net/wan/pc300_tty.c
+++ b/drivers/net/wan/pc300_tty.c
@@ -688,9 +688,9 @@ static void cpc_tty_rx_work(struct work_struct *work)
 				if (cpc_tty->tty) {
 					ld = tty_ldisc_ref(cpc_tty->tty);
 					if (ld) {
-						if (ld->receive_buf) {
+						if (ld->ops->receive_buf) {
 							CPC_TTY_DBG("%s: call line disc. receive_buf\n",cpc_tty->name);
-							ld->receive_buf(cpc_tty->tty, (char *)(buf->data), &flags, buf->size);
+							ld->ops->receive_buf(cpc_tty->tty, (char *)(buf->data), &flags, buf->size);
 						}
 						tty_ldisc_deref(ld);
 					}
diff --git a/drivers/net/wan/x25_asy.c b/drivers/net/wan/x25_asy.c
index 069f8bb0a99f..2a6c7a60756f 100644
--- a/drivers/net/wan/x25_asy.c
+++ b/drivers/net/wan/x25_asy.c
@@ -754,7 +754,7 @@ static void x25_asy_setup(struct net_device *dev)
 	dev->flags		= IFF_NOARP;
 }
 
-static struct tty_ldisc x25_ldisc = {
+static struct tty_ldisc_ops x25_ldisc = {
 	.owner		= THIS_MODULE,
 	.magic		= TTY_LDISC_MAGIC,
 	.name		= "X.25",
diff --git a/fs/proc/proc_tty.c b/fs/proc/proc_tty.c
index 21f490f5d65c..d153946d6d15 100644
--- a/fs/proc/proc_tty.c
+++ b/fs/proc/proc_tty.c
@@ -136,54 +136,6 @@ static const struct file_operations proc_tty_drivers_operations = {
 	.release	= seq_release,
 };
 
-static void * tty_ldiscs_seq_start(struct seq_file *m, loff_t *pos)
-{
-	return (*pos < NR_LDISCS) ? pos : NULL;
-}
-
-static void * tty_ldiscs_seq_next(struct seq_file *m, void *v, loff_t *pos)
-{
-	(*pos)++;
-	return (*pos < NR_LDISCS) ? pos : NULL;
-}
-
-static void tty_ldiscs_seq_stop(struct seq_file *m, void *v)
-{
-}
-
-static int tty_ldiscs_seq_show(struct seq_file *m, void *v)
-{
-	int i = *(loff_t *)v;
-	struct tty_ldisc *ld;
-	
-	ld = tty_ldisc_get(i);
-	if (ld == NULL)
-		return 0;
-	seq_printf(m, "%-10s %2d\n", ld->name ? ld->name : "???", i);
-	tty_ldisc_put(i);
-	return 0;
-}
-
-static const struct seq_operations tty_ldiscs_seq_ops = {
-	.start	= tty_ldiscs_seq_start,
-	.next	= tty_ldiscs_seq_next,
-	.stop	= tty_ldiscs_seq_stop,
-	.show	= tty_ldiscs_seq_show,
-};
-
-static int proc_tty_ldiscs_open(struct inode *inode, struct file *file)
-{
-	return seq_open(file, &tty_ldiscs_seq_ops);
-}
-
-static const struct file_operations tty_ldiscs_proc_fops = {
-	.owner		= THIS_MODULE,
-	.open		= proc_tty_ldiscs_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= seq_release,
-};
-
 /*
  * This function is called by tty_register_driver() to handle
  * registering the driver's /proc handler into /proc/tty/driver/<foo>
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 324a3b231d40..013711ea7385 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -185,6 +185,7 @@ struct tty_struct {
 	struct tty_driver *driver;
 	const struct tty_operations *ops;
 	int index;
+	/* The ldisc objects are protected by tty_ldisc_lock at the moment */
 	struct tty_ldisc ldisc;
 	struct mutex termios_mutex;
 	spinlock_t ctrl_lock;
@@ -289,7 +290,7 @@ extern void tty_wait_until_sent(struct tty_struct * tty, long timeout);
 extern int tty_check_change(struct tty_struct * tty);
 extern void stop_tty(struct tty_struct * tty);
 extern void start_tty(struct tty_struct * tty);
-extern int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc);
+extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
 extern int tty_register_driver(struct tty_driver *driver);
 extern int tty_unregister_driver(struct tty_driver *driver);
@@ -330,9 +331,7 @@ extern int tty_termios_hw_change(struct ktermios *a, struct ktermios *b);
 extern struct tty_ldisc *tty_ldisc_ref(struct tty_struct *);
 extern void tty_ldisc_deref(struct tty_ldisc *);
 extern struct tty_ldisc *tty_ldisc_ref_wait(struct tty_struct *);
-
-extern struct tty_ldisc *tty_ldisc_get(int);
-extern void tty_ldisc_put(int);
+extern const struct file_operations tty_ldiscs_proc_fops;
 
 extern void tty_wakeup(struct tty_struct *tty);
 extern void tty_ldisc_flush(struct tty_struct *tty);
@@ -354,7 +353,7 @@ extern int tty_write_lock(struct tty_struct *tty, int ndelay);
 
 
 /* n_tty.c */
-extern struct tty_ldisc tty_ldisc_N_TTY;
+extern struct tty_ldisc_ops tty_ldisc_N_TTY;
 
 /* tty_audit.c */
 #ifdef CONFIG_AUDIT
diff --git a/include/linux/tty_ldisc.h b/include/linux/tty_ldisc.h
index 6226504d9108..40f38d896777 100644
--- a/include/linux/tty_ldisc.h
+++ b/include/linux/tty_ldisc.h
@@ -104,7 +104,7 @@
 #include <linux/fs.h>
 #include <linux/wait.h>
 
-struct tty_ldisc {
+struct tty_ldisc_ops {
 	int	magic;
 	char	*name;
 	int	num;
@@ -142,6 +142,11 @@ struct tty_ldisc {
 	int refcount;
 };
 
+struct tty_ldisc {
+	struct tty_ldisc_ops *ops;
+	int refcount;
+};
+
 #define TTY_LDISC_MAGIC	0x5403
 
 #define LDISC_FLAG_DEFINED	0x00000001
diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c
index c9191871c1e0..0a387f2eb7a9 100644
--- a/net/bluetooth/rfcomm/tty.c
+++ b/net/bluetooth/rfcomm/tty.c
@@ -617,14 +617,7 @@ static void rfcomm_tty_wakeup(unsigned long arg)
 		return;
 
 	BT_DBG("dev %p tty %p", dev, tty);
-
-	if (test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags) && tty->ldisc.write_wakeup)
-		(tty->ldisc.write_wakeup)(tty);
-
-	wake_up_interruptible(&tty->write_wait);
-#ifdef SERIAL_HAVE_POLL_WAIT
-	wake_up_interruptible(&tty->poll_wait);
-#endif
+	tty_wakeup(tty);
 }
 
 static int rfcomm_tty_open(struct tty_struct *tty, struct file *filp)
@@ -1005,9 +998,7 @@ static void rfcomm_tty_flush_buffer(struct tty_struct *tty)
 		return;
 
 	skb_queue_purge(&dev->dlc->tx_queue);
-
-	if (test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags) && tty->ldisc.write_wakeup)
-		tty->ldisc.write_wakeup(tty);
+	tty_wakeup(tty);
 }
 
 static void rfcomm_tty_send_xchar(struct tty_struct *tty, char ch)
diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c
index 76c3057d0179..e4e2caeb9d82 100644
--- a/net/irda/ircomm/ircomm_tty.c
+++ b/net/irda/ircomm/ircomm_tty.c
@@ -650,12 +650,7 @@ static void ircomm_tty_do_softint(struct work_struct *work)
 	}
 
 	/* Check if user (still) wants to be waken up */
-	if ((tty->flags & (1 << TTY_DO_WRITE_WAKEUP)) &&
-	    tty->ldisc.write_wakeup)
-	{
-		(tty->ldisc.write_wakeup)(tty);
-	}
-	wake_up_interruptible(&tty->write_wait);
+	tty_wakeup(tty);
 }
 
 /*
@@ -1141,6 +1136,7 @@ static int ircomm_tty_data_indication(void *instance, void *sap,
 				      struct sk_buff *skb)
 {
 	struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) instance;
+	struct tty_ldisc *ld;
 
 	IRDA_DEBUG(2, "%s()\n", __func__ );
 
@@ -1173,7 +1169,11 @@ static int ircomm_tty_data_indication(void *instance, void *sap,
 	 * involve the flip buffers, since we are not running in an interrupt
 	 * handler
 	 */
-	self->tty->ldisc.receive_buf(self->tty, skb->data, NULL, skb->len);
+
+	ld = tty_ldisc_ref(self->tty);
+	if (ld)
+		ld->ops->receive_buf(self->tty, skb->data, NULL, skb->len);
+	tty_ldisc_deref(ld);
 
 	/* No need to kfree_skb - see ircomm_ttp_data_indication() */
 
-- 
cgit v1.2.3


From 6f67048cd010afe19d79d821f16055d9c704c6f0 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:53:41 +0100
Subject: tty: Introduce a tty_port common structure

Every tty driver has its own concept of a port structure and because
they all differ we cannot extract commonality.  Begin fixing this by
creating a structure drivers can elect to use so that over time we can
push fields into this and create commonality and then introduce common
methods.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tty_io.c | 34 ++++++++++++++++++++++++++++++++++
 include/linux/tty.h   | 30 +++++++++++++++++++++++++++++-
 2 files changed, 63 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 54c4ada460ee..739c9c59fc62 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -2088,6 +2088,40 @@ ssize_t redirected_tty_write(struct file *file, const char __user *buf,
 	return tty_write(file, buf, count, ppos);
 }
 
+void tty_port_init(struct tty_port *port)
+{
+	memset(port, 0, sizeof(*port));
+	init_waitqueue_head(&port->open_wait);
+	init_waitqueue_head(&port->close_wait);
+	mutex_init(&port->mutex);
+}
+EXPORT_SYMBOL(tty_port_init);
+
+int tty_port_alloc_xmit_buf(struct tty_port *port)
+{
+	/* We may sleep in get_zeroed_page() */
+	mutex_lock(&port->mutex);
+	if (port->xmit_buf == NULL)
+		port->xmit_buf = (unsigned char *)get_zeroed_page(GFP_KERNEL);
+	mutex_unlock(&port->mutex);
+	if (port->xmit_buf == NULL)
+		return -ENOMEM;
+	return 0;
+}
+EXPORT_SYMBOL(tty_port_alloc_xmit_buf);
+
+void tty_port_free_xmit_buf(struct tty_port *port)
+{
+	mutex_lock(&port->mutex);
+	if (port->xmit_buf != NULL) {
+		free_page((unsigned long)port->xmit_buf);
+		port->xmit_buf = NULL;
+	}
+	mutex_unlock(&port->mutex);
+}
+EXPORT_SYMBOL(tty_port_free_xmit_buf);
+
+
 static char ptychar[] = "pqrstuvwxyzabcde";
 
 /**
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 013711ea7385..d7c695b65c0e 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -166,6 +166,29 @@ struct tty_bufhead {
 
 struct device;
 struct signal_struct;
+
+/*
+ * Port level information. Each device keeps its own port level information
+ * so provide a common structure for those ports wanting to use common support
+ * routines.
+ *
+ * The tty port has a different lifetime to the tty so must be kept apart.
+ * In addition be careful as tty -> port mappings are valid for the life
+ * of the tty object but in many cases port -> tty mappings are valid only
+ * until a hangup so don't use the wrong path.
+ */
+ 
+struct tty_port {
+	struct tty_struct	*tty;		/* Back pointer */
+	int			blocked_open;	/* Waiting to open */
+	int			count;		/* Usage count */
+	wait_queue_head_t	open_wait;	/* Open waiters */
+	wait_queue_head_t	close_wait;	/* Close waiters */
+	unsigned long		flags;		/* TTY flags ASY_*/
+	struct mutex		mutex;		/* Locking */
+	unsigned char		*xmit_buf;	/* Optional buffer */
+};
+
 /*
  * Where all of the state associated with a tty is kept while the tty
  * is open.  Since the termios state should be kept even if the tty
@@ -214,7 +237,7 @@ struct tty_struct {
 	struct list_head tty_files;
 
 #define N_TTY_BUF_SIZE 4096
-	
+
 	/*
 	 * The following is data for the N_TTY line discipline.  For
 	 * historical reasons, this is included in the tty structure.
@@ -242,6 +265,7 @@ struct tty_struct {
 	spinlock_t read_lock;
 	/* If the tty has a pending do_SAK, queue it here - akpm */
 	struct work_struct SAK_work;
+	struct tty_port *port;
 };
 
 /* tty magic number */
@@ -350,6 +374,10 @@ extern void tty_write_unlock(struct tty_struct *tty);
 extern int tty_write_lock(struct tty_struct *tty, int ndelay);
 #define tty_is_writelocked(tty)  (mutex_is_locked(&tty->atomic_write_lock))
 
+extern void tty_port_init(struct tty_port *port);
+extern int tty_port_alloc_xmit_buf(struct tty_port *port);
+extern void tty_port_free_xmit_buf(struct tty_port *port);
+
 
 
 /* n_tty.c */
-- 
cgit v1.2.3


From df4f4dd429870f435f8d5d9d561db029a29f063b Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:53:50 +0100
Subject: serial: use tty_port

Switch the serial_core based drivers to use the new tty_port structure.
We can't quite use all of it yet because of the dynamically allocated
extras in the serial_core layer.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250.c        |  2 +-
 drivers/serial/jsm/jsm_neo.c |  2 +-
 drivers/serial/jsm/jsm_tty.c |  8 ++---
 drivers/serial/serial_core.c | 80 +++++++++++++++++++++++---------------------
 include/linux/serial_core.h  | 26 +++++++-------
 5 files changed, 60 insertions(+), 58 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index ac4f20cbfe41..ce948b66bbd4 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -1287,7 +1287,7 @@ static void serial8250_enable_ms(struct uart_port *port)
 static void
 receive_chars(struct uart_8250_port *up, unsigned int *status)
 {
-	struct tty_struct *tty = up->port.info->tty;
+	struct tty_struct *tty = up->port.info->port.tty;
 	unsigned char ch, lsr = *status;
 	int max_count = 256;
 	char flag;
diff --git a/drivers/serial/jsm/jsm_neo.c b/drivers/serial/jsm/jsm_neo.c
index b2d6f5b1a7c2..b7584ca55ade 100644
--- a/drivers/serial/jsm/jsm_neo.c
+++ b/drivers/serial/jsm/jsm_neo.c
@@ -998,7 +998,7 @@ static void neo_param(struct jsm_channel *ch)
 			{     50, B50     },
 		};
 
-		cflag = C_BAUD(ch->uart_port.info->tty);
+		cflag = C_BAUD(ch->uart_port.info->port.tty);
 		baud = 9600;
 		for (i = 0; i < ARRAY_SIZE(baud_rates); i++) {
 			if (baud_rates[i].cflag == cflag) {
diff --git a/drivers/serial/jsm/jsm_tty.c b/drivers/serial/jsm/jsm_tty.c
index 94ec66372508..a697914ae3d0 100644
--- a/drivers/serial/jsm/jsm_tty.c
+++ b/drivers/serial/jsm/jsm_tty.c
@@ -145,7 +145,7 @@ static void jsm_tty_send_xchar(struct uart_port *port, char ch)
 	struct ktermios *termios;
 
 	spin_lock_irqsave(&port->lock, lock_flags);
-	termios = port->info->tty->termios;
+	termios = port->info->port.tty->termios;
 	if (ch == termios->c_cc[VSTART])
 		channel->ch_bd->bd_ops->send_start_character(channel);
 
@@ -239,7 +239,7 @@ static int jsm_tty_open(struct uart_port *port)
 	channel->ch_cached_lsr = 0;
 	channel->ch_stops_sent = 0;
 
-	termios = port->info->tty->termios;
+	termios = port->info->port.tty->termios;
 	channel->ch_c_cflag	= termios->c_cflag;
 	channel->ch_c_iflag	= termios->c_iflag;
 	channel->ch_c_oflag	= termios->c_oflag;
@@ -272,7 +272,7 @@ static void jsm_tty_close(struct uart_port *port)
 	jsm_printk(CLOSE, INFO, &channel->ch_bd->pci_dev, "start\n");
 
 	bd = channel->ch_bd;
-	ts = channel->uart_port.info->tty->termios;
+	ts = channel->uart_port.info->port.tty->termios;
 
 	channel->ch_flags &= ~(CH_STOPI);
 
@@ -515,7 +515,7 @@ void jsm_input(struct jsm_channel *ch)
 	if (!ch)
 		return;
 
-	tp = ch->uart_port.info->tty;
+	tp = ch->uart_port.info->port.tty;
 
 	bd = ch->ch_bd;
 	if(!bd)
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 9884bc9eecb1..0bce1fe2c62a 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -50,7 +50,7 @@ static struct lock_class_key port_lock_key;
 
 #define HIGH_BITS_OFFSET	((sizeof(long)-sizeof(int))*8)
 
-#define uart_users(state)	((state)->count + ((state)->info ? (state)->info->blocked_open : 0))
+#define uart_users(state)	((state)->count + ((state)->info ? (state)->info->port.blocked_open : 0))
 
 #ifdef CONFIG_SERIAL_CORE_CONSOLE
 #define uart_console(port)	((port)->cons && (port)->cons->index == (port)->line)
@@ -113,7 +113,7 @@ static void uart_start(struct tty_struct *tty)
 static void uart_tasklet_action(unsigned long data)
 {
 	struct uart_state *state = (struct uart_state *)data;
-	tty_wakeup(state->info->tty);
+	tty_wakeup(state->info->port.tty);
 }
 
 static inline void
@@ -135,7 +135,7 @@ uart_update_mctrl(struct uart_port *port, unsigned int set, unsigned int clear)
 
 /*
  * Startup the port.  This will be called once per open.  All calls
- * will be serialised by the per-port semaphore.
+ * will be serialised by the per-port mutex.
  */
 static int uart_startup(struct uart_state *state, int init_hw)
 {
@@ -152,7 +152,7 @@ static int uart_startup(struct uart_state *state, int init_hw)
 	 * once we have successfully opened the port.  Also set
 	 * up the tty->alt_speed kludge
 	 */
-	set_bit(TTY_IO_ERROR, &info->tty->flags);
+	set_bit(TTY_IO_ERROR, &info->port.tty->flags);
 
 	if (port->type == PORT_UNKNOWN)
 		return 0;
@@ -162,6 +162,7 @@ static int uart_startup(struct uart_state *state, int init_hw)
 	 * buffer.
 	 */
 	if (!info->xmit.buf) {
+		/* This is protected by the per port mutex */
 		page = get_zeroed_page(GFP_KERNEL);
 		if (!page)
 			return -ENOMEM;
@@ -182,20 +183,20 @@ static int uart_startup(struct uart_state *state, int init_hw)
 			 * Setup the RTS and DTR signals once the
 			 * port is open and ready to respond.
 			 */
-			if (info->tty->termios->c_cflag & CBAUD)
+			if (info->port.tty->termios->c_cflag & CBAUD)
 				uart_set_mctrl(port, TIOCM_RTS | TIOCM_DTR);
 		}
 
 		if (info->flags & UIF_CTS_FLOW) {
 			spin_lock_irq(&port->lock);
 			if (!(port->ops->get_mctrl(port) & TIOCM_CTS))
-				info->tty->hw_stopped = 1;
+				info->port.tty->hw_stopped = 1;
 			spin_unlock_irq(&port->lock);
 		}
 
 		info->flags |= UIF_INITIALIZED;
 
-		clear_bit(TTY_IO_ERROR, &info->tty->flags);
+		clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
 	}
 
 	if (retval && capable(CAP_SYS_ADMIN))
@@ -217,8 +218,8 @@ static void uart_shutdown(struct uart_state *state)
 	/*
 	 * Set the TTY IO error marker
 	 */
-	if (info->tty)
-		set_bit(TTY_IO_ERROR, &info->tty->flags);
+	if (info->port.tty)
+		set_bit(TTY_IO_ERROR, &info->port.tty->flags);
 
 	if (info->flags & UIF_INITIALIZED) {
 		info->flags &= ~UIF_INITIALIZED;
@@ -226,7 +227,7 @@ static void uart_shutdown(struct uart_state *state)
 		/*
 		 * Turn off DTR and RTS early.
 		 */
-		if (!info->tty || (info->tty->termios->c_cflag & HUPCL))
+		if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL))
 			uart_clear_mctrl(port, TIOCM_DTR | TIOCM_RTS);
 
 		/*
@@ -426,7 +427,7 @@ EXPORT_SYMBOL(uart_get_divisor);
 static void
 uart_change_speed(struct uart_state *state, struct ktermios *old_termios)
 {
-	struct tty_struct *tty = state->info->tty;
+	struct tty_struct *tty = state->info->port.tty;
 	struct uart_port *port = state->port;
 	struct ktermios *termios;
 
@@ -836,8 +837,8 @@ static int uart_set_info(struct uart_state *state,
 	state->closing_wait    = closing_wait;
 	if (new_serial.xmit_fifo_size)
 		port->fifosize = new_serial.xmit_fifo_size;
-	if (state->info->tty)
-		state->info->tty->low_latency =
+	if (state->info->port.tty)
+		state->info->port.tty->low_latency =
 			(port->flags & UPF_LOW_LATENCY) ? 1 : 0;
 
  check_and_exit:
@@ -857,7 +858,7 @@ static int uart_set_info(struct uart_state *state,
 				printk(KERN_NOTICE
 				       "%s sets custom speed on %s. This "
 				       "is deprecated.\n", current->comm,
-				       tty_name(state->info->tty, buf));
+				       tty_name(state->info->port.tty, buf));
 			}
 			uart_change_speed(state, NULL);
 		}
@@ -889,7 +890,7 @@ static int uart_get_lsr_info(struct uart_state *state,
 	 */
 	if (port->x_char ||
 	    ((uart_circ_chars_pending(&state->info->xmit) > 0) &&
-	     !state->info->tty->stopped && !state->info->tty->hw_stopped))
+	     !state->info->port.tty->stopped && !state->info->port.tty->hw_stopped))
 		result &= ~TIOCSER_TEMT;
 
 	return put_user(result, value);
@@ -1239,7 +1240,7 @@ static void uart_set_termios(struct tty_struct *tty,
 	 */
 	if (!(old_termios->c_cflag & CLOCAL) &&
 	    (tty->termios->c_cflag & CLOCAL))
-		wake_up_interruptible(&state->info->open_wait);
+		wake_up_interruptible(&state->info->port.open_wait);
 #endif
 }
 
@@ -1320,9 +1321,9 @@ static void uart_close(struct tty_struct *tty, struct file *filp)
 	tty_ldisc_flush(tty);
 
 	tty->closing = 0;
-	state->info->tty = NULL;
+	state->info->port.tty = NULL;
 
-	if (state->info->blocked_open) {
+	if (state->info->port.blocked_open) {
 		if (state->close_delay)
 			msleep_interruptible(state->close_delay);
 	} else if (!uart_console(port)) {
@@ -1333,7 +1334,7 @@ static void uart_close(struct tty_struct *tty, struct file *filp)
 	 * Wake up anyone trying to open this port.
 	 */
 	state->info->flags &= ~UIF_NORMAL_ACTIVE;
-	wake_up_interruptible(&state->info->open_wait);
+	wake_up_interruptible(&state->info->port.open_wait);
 
  done:
 	mutex_unlock(&state->mutex);
@@ -1417,8 +1418,8 @@ static void uart_hangup(struct tty_struct *tty)
 		uart_shutdown(state);
 		state->count = 0;
 		state->info->flags &= ~UIF_NORMAL_ACTIVE;
-		state->info->tty = NULL;
-		wake_up_interruptible(&state->info->open_wait);
+		state->info->port.tty = NULL;
+		wake_up_interruptible(&state->info->port.open_wait);
 		wake_up_interruptible(&state->info->delta_msr_wait);
 	}
 	mutex_unlock(&state->mutex);
@@ -1432,7 +1433,7 @@ static void uart_hangup(struct tty_struct *tty)
  */
 static void uart_update_termios(struct uart_state *state)
 {
-	struct tty_struct *tty = state->info->tty;
+	struct tty_struct *tty = state->info->port.tty;
 	struct uart_port *port = state->port;
 
 	if (uart_console(port) && port->cons->cflag) {
@@ -1471,17 +1472,17 @@ uart_block_til_ready(struct file *filp, struct uart_state *state)
 	struct uart_port *port = state->port;
 	unsigned int mctrl;
 
-	info->blocked_open++;
+	info->port.blocked_open++;
 	state->count--;
 
-	add_wait_queue(&info->open_wait, &wait);
+	add_wait_queue(&info->port.open_wait, &wait);
 	while (1) {
 		set_current_state(TASK_INTERRUPTIBLE);
 
 		/*
 		 * If we have been hung up, tell userspace/restart open.
 		 */
-		if (tty_hung_up_p(filp) || info->tty == NULL)
+		if (tty_hung_up_p(filp) || info->port.tty == NULL)
 			break;
 
 		/*
@@ -1500,8 +1501,8 @@ uart_block_til_ready(struct file *filp, struct uart_state *state)
 		 * have set TTY_IO_ERROR for a non-existant port.
 		 */
 		if ((filp->f_flags & O_NONBLOCK) ||
-		    (info->tty->termios->c_cflag & CLOCAL) ||
-		    (info->tty->flags & (1 << TTY_IO_ERROR)))
+		    (info->port.tty->termios->c_cflag & CLOCAL) ||
+		    (info->port.tty->flags & (1 << TTY_IO_ERROR)))
 			break;
 
 		/*
@@ -1509,7 +1510,7 @@ uart_block_til_ready(struct file *filp, struct uart_state *state)
 		 * not set RTS here - we want to make sure we catch
 		 * the data from the modem.
 		 */
-		if (info->tty->termios->c_cflag & CBAUD)
+		if (info->port.tty->termios->c_cflag & CBAUD)
 			uart_set_mctrl(port, TIOCM_DTR);
 
 		/*
@@ -1531,15 +1532,15 @@ uart_block_til_ready(struct file *filp, struct uart_state *state)
 			break;
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->open_wait, &wait);
+	remove_wait_queue(&info->port.open_wait, &wait);
 
 	state->count++;
-	info->blocked_open--;
+	info->port.blocked_open--;
 
 	if (signal_pending(current))
 		return -ERESTARTSYS;
 
-	if (!info->tty || tty_hung_up_p(filp))
+	if (!info->port.tty || tty_hung_up_p(filp))
 		return -EAGAIN;
 
 	return 0;
@@ -1562,10 +1563,13 @@ static struct uart_state *uart_get(struct uart_driver *drv, int line)
 		goto err_unlock;
 	}
 
+	/* BKL: RACE HERE - LEAK */
+	/* We should move this into the uart_state structure and kill off
+	   this whole complexity */
 	if (!state->info) {
 		state->info = kzalloc(sizeof(struct uart_info), GFP_KERNEL);
 		if (state->info) {
-			init_waitqueue_head(&state->info->open_wait);
+			init_waitqueue_head(&state->info->port.open_wait);
 			init_waitqueue_head(&state->info->delta_msr_wait);
 
 			/*
@@ -1622,7 +1626,7 @@ static int uart_open(struct tty_struct *tty, struct file *filp)
 	 * be re-entered while allocating the info structure, or while we
 	 * request any IRQs that the driver may need.  This also has the nice
 	 * side-effect that it delays the action of uart_hangup, so we can
-	 * guarantee that info->tty will always contain something reasonable.
+	 * guarantee that info->port.tty will always contain something reasonable.
 	 */
 	state = uart_get(drv, line);
 	if (IS_ERR(state)) {
@@ -1638,7 +1642,7 @@ static int uart_open(struct tty_struct *tty, struct file *filp)
 	tty->driver_data = state;
 	tty->low_latency = (state->port->flags & UPF_LOW_LATENCY) ? 1 : 0;
 	tty->alt_speed = 0;
-	state->info->tty = tty;
+	state->info->port.tty = tty;
 
 	/*
 	 * If the port is in the middle of closing, bail out now.
@@ -2101,8 +2105,8 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port)
 		/*
 		 * If that's unset, use the tty termios setting.
 		 */
-		if (state->info && state->info->tty && termios.c_cflag == 0)
-			termios = *state->info->tty->termios;
+		if (state->info && state->info->port.tty && termios.c_cflag == 0)
+			termios = *state->info->port.tty->termios;
 
 		uart_change_pm(state, 0);
 		port->ops->set_termios(port, &termios, NULL);
@@ -2521,8 +2525,8 @@ int uart_remove_one_port(struct uart_driver *drv, struct uart_port *port)
 	tty_unregister_device(drv->tty_driver, port->line);
 
 	info = state->info;
-	if (info && info->tty)
-		tty_vhangup(info->tty);
+	if (info && info->port.tty)
+		tty_vhangup(info->port.tty);
 
 	/*
 	 * All users of this port should now be disconnected from
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 1d2faa6592ae..f3a1c0e45021 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -344,13 +344,15 @@ typedef unsigned int __bitwise__ uif_t;
  * stuff here.
  */
 struct uart_info {
-	struct tty_struct	*tty;
+	struct tty_port		port;
 	struct circ_buf		xmit;
 	uif_t			flags;
 
 /*
  * Definitions for info->flags.  These are _private_ to serial_core, and
  * are specific to this structure.  They may be queried by low level drivers.
+ *
+ * FIXME: use the ASY_ definitions
  */
 #define UIF_CHECK_CD		((__force uif_t) (1 << 25))
 #define UIF_CTS_FLOW		((__force uif_t) (1 << 26))
@@ -358,11 +360,7 @@ struct uart_info {
 #define UIF_INITIALIZED		((__force uif_t) (1 << 31))
 #define UIF_SUSPENDED		((__force uif_t) (1 << 30))
 
-	int			blocked_open;
-
 	struct tasklet_struct	tlet;
-
-	wait_queue_head_t	open_wait;
 	wait_queue_head_t	delta_msr_wait;
 };
 
@@ -439,8 +437,8 @@ int uart_resume_port(struct uart_driver *reg, struct uart_port *port);
 #define uart_circ_chars_free(circ)	\
 	(CIRC_SPACE((circ)->head, (circ)->tail, UART_XMIT_SIZE))
 
-#define uart_tx_stopped(port)		\
-	((port)->info->tty->stopped || (port)->info->tty->hw_stopped)
+#define uart_tx_stopped(portp)		\
+	((portp)->info->port.tty->stopped || (portp)->info->port.tty->hw_stopped)
 
 /*
  * The following are helper functions for the low level drivers.
@@ -451,7 +449,7 @@ uart_handle_sysrq_char(struct uart_port *port, unsigned int ch)
 #ifdef SUPPORT_SYSRQ
 	if (port->sysrq) {
 		if (ch && time_before(jiffies, port->sysrq)) {
-			handle_sysrq(ch, port->info ? port->info->tty : NULL);
+			handle_sysrq(ch, port->info ? port->info->port.tty : NULL);
 			port->sysrq = 0;
 			return 1;
 		}
@@ -480,7 +478,7 @@ static inline int uart_handle_break(struct uart_port *port)
 	}
 #endif
 	if (port->flags & UPF_SAK)
-		do_SAK(info->tty);
+		do_SAK(info->port.tty);
 	return 0;
 }
 
@@ -503,9 +501,9 @@ uart_handle_dcd_change(struct uart_port *port, unsigned int status)
 
 	if (info->flags & UIF_CHECK_CD) {
 		if (status)
-			wake_up_interruptible(&info->open_wait);
-		else if (info->tty)
-			tty_hangup(info->tty);
+			wake_up_interruptible(&info->port.open_wait);
+		else if (info->port.tty)
+			tty_hangup(info->port.tty);
 	}
 }
 
@@ -518,7 +516,7 @@ static inline void
 uart_handle_cts_change(struct uart_port *port, unsigned int status)
 {
 	struct uart_info *info = port->info;
-	struct tty_struct *tty = info->tty;
+	struct tty_struct *tty = info->port.tty;
 
 	port->icount.cts++;
 
@@ -544,7 +542,7 @@ static inline void
 uart_insert_char(struct uart_port *port, unsigned int status,
 		 unsigned int overrun, unsigned int ch, unsigned int flag)
 {
-	struct tty_struct *tty = port->info->tty;
+	struct tty_struct *tty = port->info->port.tty;
 
 	if ((status & port->ignore_status_mask & ~overrun) == 0)
 		tty_insert_flip_char(tty, ch, flag);
-- 
cgit v1.2.3


From 7a4d29f426f17479395980ded8fa5e3bdd6d94e4 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:54:54 +0100
Subject: tty.h: clean up

Coding style clean up and white space tidy

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tty.h | 165 +++++++++++++++++++++++++++-------------------------
 1 file changed, 85 insertions(+), 80 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tty.h b/include/linux/tty.h
index d7c695b65c0e..46008e86b191 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -71,7 +71,8 @@ struct tty_bufhead {
 	struct tty_buffer *head;	/* Queue head */
 	struct tty_buffer *tail;	/* Active buffer */
 	struct tty_buffer *free;	/* Free queue head */
-	int memory_used;		/* Buffer space used excluding free queue */
+	int memory_used;		/* Buffer space used excluding
+								free queue */
 };
 /*
  * When a break, frame error, or parity error happens, these codes are
@@ -101,68 +102,68 @@ struct tty_bufhead {
 #define LNEXT_CHAR(tty)	((tty)->termios->c_cc[VLNEXT])
 #define EOL2_CHAR(tty) ((tty)->termios->c_cc[VEOL2])
 
-#define _I_FLAG(tty,f)	((tty)->termios->c_iflag & (f))
-#define _O_FLAG(tty,f)	((tty)->termios->c_oflag & (f))
-#define _C_FLAG(tty,f)	((tty)->termios->c_cflag & (f))
-#define _L_FLAG(tty,f)	((tty)->termios->c_lflag & (f))
-
-#define I_IGNBRK(tty)	_I_FLAG((tty),IGNBRK)
-#define I_BRKINT(tty)	_I_FLAG((tty),BRKINT)
-#define I_IGNPAR(tty)	_I_FLAG((tty),IGNPAR)
-#define I_PARMRK(tty)	_I_FLAG((tty),PARMRK)
-#define I_INPCK(tty)	_I_FLAG((tty),INPCK)
-#define I_ISTRIP(tty)	_I_FLAG((tty),ISTRIP)
-#define I_INLCR(tty)	_I_FLAG((tty),INLCR)
-#define I_IGNCR(tty)	_I_FLAG((tty),IGNCR)
-#define I_ICRNL(tty)	_I_FLAG((tty),ICRNL)
-#define I_IUCLC(tty)	_I_FLAG((tty),IUCLC)
-#define I_IXON(tty)	_I_FLAG((tty),IXON)
-#define I_IXANY(tty)	_I_FLAG((tty),IXANY)
-#define I_IXOFF(tty)	_I_FLAG((tty),IXOFF)
-#define I_IMAXBEL(tty)	_I_FLAG((tty),IMAXBEL)
-#define I_IUTF8(tty)	_I_FLAG((tty),IUTF8)
-
-#define O_OPOST(tty)	_O_FLAG((tty),OPOST)
-#define O_OLCUC(tty)	_O_FLAG((tty),OLCUC)
-#define O_ONLCR(tty)	_O_FLAG((tty),ONLCR)
-#define O_OCRNL(tty)	_O_FLAG((tty),OCRNL)
-#define O_ONOCR(tty)	_O_FLAG((tty),ONOCR)
-#define O_ONLRET(tty)	_O_FLAG((tty),ONLRET)
-#define O_OFILL(tty)	_O_FLAG((tty),OFILL)
-#define O_OFDEL(tty)	_O_FLAG((tty),OFDEL)
-#define O_NLDLY(tty)	_O_FLAG((tty),NLDLY)
-#define O_CRDLY(tty)	_O_FLAG((tty),CRDLY)
-#define O_TABDLY(tty)	_O_FLAG((tty),TABDLY)
-#define O_BSDLY(tty)	_O_FLAG((tty),BSDLY)
-#define O_VTDLY(tty)	_O_FLAG((tty),VTDLY)
-#define O_FFDLY(tty)	_O_FLAG((tty),FFDLY)
-
-#define C_BAUD(tty)	_C_FLAG((tty),CBAUD)
-#define C_CSIZE(tty)	_C_FLAG((tty),CSIZE)
-#define C_CSTOPB(tty)	_C_FLAG((tty),CSTOPB)
-#define C_CREAD(tty)	_C_FLAG((tty),CREAD)
-#define C_PARENB(tty)	_C_FLAG((tty),PARENB)
-#define C_PARODD(tty)	_C_FLAG((tty),PARODD)
-#define C_HUPCL(tty)	_C_FLAG((tty),HUPCL)
-#define C_CLOCAL(tty)	_C_FLAG((tty),CLOCAL)
-#define C_CIBAUD(tty)	_C_FLAG((tty),CIBAUD)
-#define C_CRTSCTS(tty)	_C_FLAG((tty),CRTSCTS)
-
-#define L_ISIG(tty)	_L_FLAG((tty),ISIG)
-#define L_ICANON(tty)	_L_FLAG((tty),ICANON)
-#define L_XCASE(tty)	_L_FLAG((tty),XCASE)
-#define L_ECHO(tty)	_L_FLAG((tty),ECHO)
-#define L_ECHOE(tty)	_L_FLAG((tty),ECHOE)
-#define L_ECHOK(tty)	_L_FLAG((tty),ECHOK)
-#define L_ECHONL(tty)	_L_FLAG((tty),ECHONL)
-#define L_NOFLSH(tty)	_L_FLAG((tty),NOFLSH)
-#define L_TOSTOP(tty)	_L_FLAG((tty),TOSTOP)
-#define L_ECHOCTL(tty)	_L_FLAG((tty),ECHOCTL)
-#define L_ECHOPRT(tty)	_L_FLAG((tty),ECHOPRT)
-#define L_ECHOKE(tty)	_L_FLAG((tty),ECHOKE)
-#define L_FLUSHO(tty)	_L_FLAG((tty),FLUSHO)
-#define L_PENDIN(tty)	_L_FLAG((tty),PENDIN)
-#define L_IEXTEN(tty)	_L_FLAG((tty),IEXTEN)
+#define _I_FLAG(tty, f)	((tty)->termios->c_iflag & (f))
+#define _O_FLAG(tty, f)	((tty)->termios->c_oflag & (f))
+#define _C_FLAG(tty, f)	((tty)->termios->c_cflag & (f))
+#define _L_FLAG(tty, f)	((tty)->termios->c_lflag & (f))
+
+#define I_IGNBRK(tty)	_I_FLAG((tty), IGNBRK)
+#define I_BRKINT(tty)	_I_FLAG((tty), BRKINT)
+#define I_IGNPAR(tty)	_I_FLAG((tty), IGNPAR)
+#define I_PARMRK(tty)	_I_FLAG((tty), PARMRK)
+#define I_INPCK(tty)	_I_FLAG((tty), INPCK)
+#define I_ISTRIP(tty)	_I_FLAG((tty), ISTRIP)
+#define I_INLCR(tty)	_I_FLAG((tty), INLCR)
+#define I_IGNCR(tty)	_I_FLAG((tty), IGNCR)
+#define I_ICRNL(tty)	_I_FLAG((tty), ICRNL)
+#define I_IUCLC(tty)	_I_FLAG((tty), IUCLC)
+#define I_IXON(tty)	_I_FLAG((tty), IXON)
+#define I_IXANY(tty)	_I_FLAG((tty), IXANY)
+#define I_IXOFF(tty)	_I_FLAG((tty), IXOFF)
+#define I_IMAXBEL(tty)	_I_FLAG((tty), IMAXBEL)
+#define I_IUTF8(tty)	_I_FLAG((tty), IUTF8)
+
+#define O_OPOST(tty)	_O_FLAG((tty), OPOST)
+#define O_OLCUC(tty)	_O_FLAG((tty), OLCUC)
+#define O_ONLCR(tty)	_O_FLAG((tty), ONLCR)
+#define O_OCRNL(tty)	_O_FLAG((tty), OCRNL)
+#define O_ONOCR(tty)	_O_FLAG((tty), ONOCR)
+#define O_ONLRET(tty)	_O_FLAG((tty), ONLRET)
+#define O_OFILL(tty)	_O_FLAG((tty), OFILL)
+#define O_OFDEL(tty)	_O_FLAG((tty), OFDEL)
+#define O_NLDLY(tty)	_O_FLAG((tty), NLDLY)
+#define O_CRDLY(tty)	_O_FLAG((tty), CRDLY)
+#define O_TABDLY(tty)	_O_FLAG((tty), TABDLY)
+#define O_BSDLY(tty)	_O_FLAG((tty), BSDLY)
+#define O_VTDLY(tty)	_O_FLAG((tty), VTDLY)
+#define O_FFDLY(tty)	_O_FLAG((tty), FFDLY)
+
+#define C_BAUD(tty)	_C_FLAG((tty), CBAUD)
+#define C_CSIZE(tty)	_C_FLAG((tty), CSIZE)
+#define C_CSTOPB(tty)	_C_FLAG((tty), CSTOPB)
+#define C_CREAD(tty)	_C_FLAG((tty), CREAD)
+#define C_PARENB(tty)	_C_FLAG((tty), PARENB)
+#define C_PARODD(tty)	_C_FLAG((tty), PARODD)
+#define C_HUPCL(tty)	_C_FLAG((tty), HUPCL)
+#define C_CLOCAL(tty)	_C_FLAG((tty), CLOCAL)
+#define C_CIBAUD(tty)	_C_FLAG((tty), CIBAUD)
+#define C_CRTSCTS(tty)	_C_FLAG((tty), CRTSCTS)
+
+#define L_ISIG(tty)	_L_FLAG((tty), ISIG)
+#define L_ICANON(tty)	_L_FLAG((tty), ICANON)
+#define L_XCASE(tty)	_L_FLAG((tty), XCASE)
+#define L_ECHO(tty)	_L_FLAG((tty), ECHO)
+#define L_ECHOE(tty)	_L_FLAG((tty), ECHOE)
+#define L_ECHOK(tty)	_L_FLAG((tty), ECHOK)
+#define L_ECHONL(tty)	_L_FLAG((tty), ECHONL)
+#define L_NOFLSH(tty)	_L_FLAG((tty), NOFLSH)
+#define L_TOSTOP(tty)	_L_FLAG((tty), TOSTOP)
+#define L_ECHOCTL(tty)	_L_FLAG((tty), ECHOCTL)
+#define L_ECHOPRT(tty)	_L_FLAG((tty), ECHOPRT)
+#define L_ECHOKE(tty)	_L_FLAG((tty), ECHOKE)
+#define L_FLUSHO(tty)	_L_FLAG((tty), FLUSHO)
+#define L_PENDIN(tty)	_L_FLAG((tty), PENDIN)
+#define L_IEXTEN(tty)	_L_FLAG((tty), IEXTEN)
 
 struct device;
 struct signal_struct;
@@ -177,7 +178,7 @@ struct signal_struct;
  * of the tty object but in many cases port -> tty mappings are valid only
  * until a hangup so don't use the wrong path.
  */
- 
+
 struct tty_port {
 	struct tty_struct	*tty;		/* Back pointer */
 	int			blocked_open;	/* Waiting to open */
@@ -273,14 +274,14 @@ struct tty_struct {
 
 /*
  * These bits are used in the flags field of the tty structure.
- * 
+ *
  * So that interrupts won't be able to mess up the queues,
  * copy_to_cooked must be atomic with respect to itself, as must
  * tty->write.  Thus, you must use the inline functions set_bit() and
  * clear_bit() to make things atomic.
  */
 #define TTY_THROTTLED 		0	/* Call unthrottle() at threshold min */
-#define TTY_IO_ERROR 		1	/* Canse an I/O error (may be no ldisc too) */
+#define TTY_IO_ERROR 		1	/* Cause an I/O error (may be no ldisc too) */
 #define TTY_OTHER_CLOSED 	2	/* Other side (if any) has closed */
 #define TTY_EXCLUSIVE 		3	/* Exclusive open mode */
 #define TTY_DEBUG 		4	/* Debugging */
@@ -310,10 +311,10 @@ extern int vcs_init(void);
 extern int tty_paranoia_check(struct tty_struct *tty, struct inode *inode,
 			      const char *routine);
 extern char *tty_name(struct tty_struct *tty, char *buf);
-extern void tty_wait_until_sent(struct tty_struct * tty, long timeout);
-extern int tty_check_change(struct tty_struct * tty);
-extern void stop_tty(struct tty_struct * tty);
-extern void start_tty(struct tty_struct * tty);
+extern void tty_wait_until_sent(struct tty_struct *tty, long timeout);
+extern int tty_check_change(struct tty_struct *tty);
+extern void stop_tty(struct tty_struct *tty);
+extern void start_tty(struct tty_struct *tty);
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
 extern int tty_register_driver(struct tty_driver *driver);
@@ -335,10 +336,10 @@ extern int is_current_pgrp_orphaned(void);
 extern struct pid *tty_get_pgrp(struct tty_struct *tty);
 extern int is_ignored(int sig);
 extern int tty_signal(int sig, struct tty_struct *tty);
-extern void tty_hangup(struct tty_struct * tty);
-extern void tty_vhangup(struct tty_struct * tty);
+extern void tty_hangup(struct tty_struct *tty);
+extern void tty_vhangup(struct tty_struct *tty);
 extern void tty_unhangup(struct file *filp);
-extern int tty_hung_up_p(struct file * filp);
+extern int tty_hung_up_p(struct file *filp);
 extern void do_SAK(struct tty_struct *tty);
 extern void __do_SAK(struct tty_struct *tty);
 extern void disassociate_ctty(int priv);
@@ -347,8 +348,10 @@ extern void tty_flip_buffer_push(struct tty_struct *tty);
 extern speed_t tty_get_baud_rate(struct tty_struct *tty);
 extern speed_t tty_termios_baud_rate(struct ktermios *termios);
 extern speed_t tty_termios_input_baud_rate(struct ktermios *termios);
-extern void tty_termios_encode_baud_rate(struct ktermios *termios, speed_t ibaud, speed_t obaud);
-extern void tty_encode_baud_rate(struct tty_struct *tty, speed_t ibaud, speed_t obaud);
+extern void tty_termios_encode_baud_rate(struct ktermios *termios,
+						speed_t ibaud, speed_t obaud);
+extern void tty_encode_baud_rate(struct tty_struct *tty,
+						speed_t ibaud, speed_t obaud);
 extern void tty_termios_copy_hw(struct ktermios *new, struct ktermios *old);
 extern int tty_termios_hw_change(struct ktermios *a, struct ktermios *b);
 
@@ -390,7 +393,8 @@ extern void tty_audit_add_data(struct tty_struct *tty, unsigned char *data,
 extern void tty_audit_exit(void);
 extern void tty_audit_fork(struct signal_struct *sig);
 extern void tty_audit_push(struct tty_struct *tty);
-extern void tty_audit_push_task(struct task_struct *tsk, uid_t loginuid, u32 sessionid);
+extern void tty_audit_push_task(struct task_struct *tsk,
+					uid_t loginuid, u32 sessionid);
 #else
 static inline void tty_audit_add_data(struct tty_struct *tty,
 				      unsigned char *data, size_t size)
@@ -405,19 +409,20 @@ static inline void tty_audit_fork(struct signal_struct *sig)
 static inline void tty_audit_push(struct tty_struct *tty)
 {
 }
-static inline void tty_audit_push_task(struct task_struct *tsk, uid_t loginuid, u32 sessionid)
+static inline void tty_audit_push_task(struct task_struct *tsk,
+					uid_t loginuid, u32 sessionid)
 {
 }
 #endif
 
 /* tty_ioctl.c */
-extern int n_tty_ioctl(struct tty_struct * tty, struct file * file,
+extern int n_tty_ioctl(struct tty_struct *tty, struct file *file,
 		       unsigned int cmd, unsigned long arg);
 
 /* serial.c */
 
 extern void serial_console_init(void);
- 
+
 /* pcxx.c */
 
 extern int pcxe_open(struct tty_struct *tty, struct file *filp);
@@ -428,7 +433,7 @@ extern void console_print(const char *);
 
 /* vt.c */
 
-extern int vt_ioctl(struct tty_struct *tty, struct file * file,
+extern int vt_ioctl(struct tty_struct *tty, struct file *file,
 		    unsigned int cmd, unsigned long arg);
 
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3


From 4982d6b37a5ccebe6c2af79970c7a15c1939243a Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:55:11 +0100
Subject: esp: use tty_port

Switch esp to use the new tty_port structures

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/esp.c       | 272 +++++++++++++++++++++++------------------------
 include/linux/hayesesp.h |   9 +-
 2 files changed, 138 insertions(+), 143 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/esp.c b/drivers/char/esp.c
index 84840ba13ff0..2eaf09f93e3d 100644
--- a/drivers/char/esp.c
+++ b/drivers/char/esp.c
@@ -128,9 +128,9 @@ static struct tty_driver *esp_driver;
 
 #if defined(MODULE) && defined(SERIAL_DEBUG_MCOUNT)
 #define DBG_CNT(s) printk(KERN_DEBUG "(%s): [%x] refc=%d, serc=%d, ttyc=%d -> %s\n", \
-				tty->name, info->flags, \
+				tty->name, info->port.flags, \
 				serial_driver.refcount, \
-				info->count, tty->count, s)
+				info->port.count, tty->count, s)
 #else
 #define DBG_CNT(s)
 #endif
@@ -172,13 +172,13 @@ static inline int serial_paranoia_check(struct esp_struct *info,
 
 static inline unsigned int serial_in(struct esp_struct *info, int offset)
 {
-	return inb(info->port + offset);
+	return inb(info->io_port + offset);
 }
 
 static inline void serial_out(struct esp_struct *info, int offset,
 			      unsigned char value)
 {
-	outb(value, info->port+offset);
+	outb(value, info->io_port+offset);
 }
 
 /*
@@ -273,7 +273,7 @@ static inline void release_pio_buffer(struct esp_pio_buffer *buf)
 
 static inline void receive_chars_pio(struct esp_struct *info, int num_bytes)
 {
-	struct tty_struct *tty = info->tty;
+	struct tty_struct *tty = info->port.tty;
 	int i;
 	struct esp_pio_buffer *pio_buf;
 	struct esp_pio_buffer *err_buf;
@@ -295,7 +295,7 @@ static inline void receive_chars_pio(struct esp_struct *info, int num_bytes)
 
 	for (i = 0; i < num_bytes - 1; i += 2) {
 		*((unsigned short *)(pio_buf->data + i)) =
-			inw(info->port + UART_ESI_RX);
+			inw(info->io_port + UART_ESI_RX);
 		err_buf->data[i] = serial_in(info, UART_ESI_RWS);
 		err_buf->data[i + 1] = (err_buf->data[i] >> 3) & status_mask;
 		err_buf->data[i] &= status_mask;
@@ -308,7 +308,7 @@ static inline void receive_chars_pio(struct esp_struct *info, int num_bytes)
 	}
 
 	/* make sure everything is still ok since interrupts were enabled */
-	tty = info->tty;
+	tty = info->port.tty;
 
 	if (!tty) {
 		release_pio_buffer(pio_buf);
@@ -325,7 +325,7 @@ static inline void receive_chars_pio(struct esp_struct *info, int num_bytes)
 
 			if (err_buf->data[i] & 0x04) {
 				flag = TTY_BREAK;
-				if (info->flags & ASYNC_SAK)
+				if (info->port.flags & ASYNC_SAK)
 					do_SAK(tty);
 			} else if (err_buf->data[i] & 0x02)
 				flag = TTY_FRAME;
@@ -370,7 +370,7 @@ static void receive_chars_dma(struct esp_struct *info, int num_bytes)
 static inline void receive_chars_dma_done(struct esp_struct *info,
 					    int status)
 {
-	struct tty_struct *tty = info->tty;
+	struct tty_struct *tty = info->port.tty;
 	int num_bytes;
 	unsigned long flags;
 
@@ -396,7 +396,7 @@ static inline void receive_chars_dma_done(struct esp_struct *info,
 			if (status & 0x10) {
 				statflag = TTY_BREAK;
 				(info->icount.brk)++;
-				if (info->flags & ASYNC_SAK)
+				if (info->port.flags & ASYNC_SAK)
 					do_SAK(tty);
 			} else if (status & 0x08) {
 				statflag = TTY_FRAME;
@@ -451,7 +451,7 @@ static inline void transmit_chars_pio(struct esp_struct *info,
 
 		for (i = 0; i < space_avail - 1; i += 2) {
 			outw(*((unsigned short *)(pio_buf->data + i)),
-			     info->port + UART_ESI_TX);
+			     info->io_port + UART_ESI_TX);
 		}
 
 		if (space_avail & 0x0001)
@@ -470,8 +470,8 @@ static inline void transmit_chars_pio(struct esp_struct *info,
 	}
 
 	if (info->xmit_cnt < WAKEUP_CHARS) {
-		if (info->tty)
-			tty_wakeup(info->tty);
+		if (info->port.tty)
+			tty_wakeup(info->port.tty);
 
 #ifdef SERIAL_DEBUG_INTR
 		printk("THRE...");
@@ -507,8 +507,8 @@ static inline void transmit_chars_dma(struct esp_struct *info, int num_bytes)
 	info->xmit_tail = (info->xmit_tail + dma_bytes) & (ESP_XMIT_SIZE - 1);
 
 	if (info->xmit_cnt < WAKEUP_CHARS) {
-		if (info->tty)
-			tty_wakeup(info->tty);
+		if (info->port.tty)
+			tty_wakeup(info->port.tty);
 
 #ifdef SERIAL_DEBUG_INTR
 		printk("THRE...");
@@ -575,18 +575,18 @@ static void check_modem_status(struct esp_struct *info)
 		wake_up_interruptible(&info->delta_msr_wait);
 	}
 
-	if ((info->flags & ASYNC_CHECK_CD) && (status & UART_MSR_DDCD)) {
+	if ((info->port.flags & ASYNC_CHECK_CD) && (status & UART_MSR_DDCD)) {
 #if (defined(SERIAL_DEBUG_OPEN) || defined(SERIAL_DEBUG_INTR))
 		printk("ttys%d CD now %s...", info->line,
 		       (status & UART_MSR_DCD) ? "on" : "off");
 #endif
 		if (status & UART_MSR_DCD)
-			wake_up_interruptible(&info->open_wait);
+			wake_up_interruptible(&info->port.open_wait);
 		else {
 #ifdef SERIAL_DEBUG_OPEN
 			printk("scheduling hangup...");
 #endif
-			tty_hangup(info->tty);
+			tty_hangup(info->port.tty);
 		}
 	}
 }
@@ -609,7 +609,7 @@ static irqreturn_t rs_interrupt_single(int irq, void *dev_id)
 
 	spin_lock(&info->lock);
 
-	if (!info->tty) {
+	if (!info->port.tty) {
 		spin_unlock(&info->lock);
 		return IRQ_NONE;
 	}
@@ -647,7 +647,7 @@ static irqreturn_t rs_interrupt_single(int irq, void *dev_id)
 		num_bytes = serial_in(info, UART_ESI_STAT1) << 8;
 		num_bytes |= serial_in(info, UART_ESI_STAT2);
 
-		num_bytes = tty_buffer_request_room(info->tty, num_bytes);
+		num_bytes = tty_buffer_request_room(info->port.tty, num_bytes);
 
 		if (num_bytes) {
 			if (dma_bytes ||
@@ -661,7 +661,7 @@ static irqreturn_t rs_interrupt_single(int irq, void *dev_id)
 
 	if (!(info->stat_flags & (ESP_STAT_DMA_RX | ESP_STAT_DMA_TX)) &&
 	    (scratch & 0x02) && (info->IER & UART_IER_THRI)) {
-		if ((info->xmit_cnt <= 0) || info->tty->stopped) {
+		if ((info->xmit_cnt <= 0) || info->port.tty->stopped) {
 			info->IER &= ~UART_IER_THRI;
 			serial_out(info, UART_ESI_CMD1, ESI_SET_SRV_MASK);
 			serial_out(info, UART_ESI_CMD2, info->IER);
@@ -782,7 +782,7 @@ static int startup(struct esp_struct *info)
 
 	spin_lock_irqsave(&info->lock, flags);
 
-	if (info->flags & ASYNC_INITIALIZED)
+	if (info->port.flags & ASYNC_INITIALIZED)
 		goto out;
 
 	if (!info->xmit_buf) {
@@ -806,7 +806,7 @@ static int startup(struct esp_struct *info)
 	num_chars |= serial_in(info, UART_ESI_STAT2);
 
 	while (num_chars > 1) {
-		inw(info->port + UART_ESI_RX);
+		inw(info->io_port + UART_ESI_RX);
 		num_chars -= 2;
 	}
 
@@ -834,9 +834,9 @@ static int startup(struct esp_struct *info)
 
 	if (retval) {
 		if (capable(CAP_SYS_ADMIN)) {
-			if (info->tty)
+			if (info->port.tty)
 				set_bit(TTY_IO_ERROR,
-					&info->tty->flags);
+					&info->port.tty->flags);
 			retval = 0;
 		}
 		goto out_unlocked;
@@ -874,30 +874,30 @@ static int startup(struct esp_struct *info)
 	serial_out(info, UART_ESI_CMD1, ESI_SET_SRV_MASK);
 	serial_out(info, UART_ESI_CMD2, info->IER);
 
-	if (info->tty)
-		clear_bit(TTY_IO_ERROR, &info->tty->flags);
+	if (info->port.tty)
+		clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
 	info->xmit_cnt = info->xmit_head = info->xmit_tail = 0;
 	spin_unlock_irqrestore(&info->lock, flags);
 
 	/*
 	 * Set up the tty->alt_speed kludge
 	 */
-	if (info->tty) {
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
-			info->tty->alt_speed = 57600;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
-			info->tty->alt_speed = 115200;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
-			info->tty->alt_speed = 230400;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
-			info->tty->alt_speed = 460800;
+	if (info->port.tty) {
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+			info->port.tty->alt_speed = 57600;
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+			info->port.tty->alt_speed = 115200;
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
+			info->port.tty->alt_speed = 230400;
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
+			info->port.tty->alt_speed = 460800;
 	}
 
 	/*
 	 * set the speed of the serial port
 	 */
 	change_speed(info);
-	info->flags |= ASYNC_INITIALIZED;
+	info->port.flags |= ASYNC_INITIALIZED;
 	return 0;
 
 out:
@@ -914,7 +914,7 @@ static void shutdown(struct esp_struct *info)
 {
 	unsigned long	flags, f;
 
-	if (!(info->flags & ASYNC_INITIALIZED))
+	if (!(info->port.flags & ASYNC_INITIALIZED))
 		return;
 
 #ifdef SERIAL_DEBUG_OPEN
@@ -951,7 +951,7 @@ static void shutdown(struct esp_struct *info)
 
 		while (current_port) {
 			if ((current_port != info) &&
-			    (current_port->flags & ASYNC_INITIALIZED))
+			    (current_port->port.flags & ASYNC_INITIALIZED))
 				break;
 
 			current_port = current_port->next_port;
@@ -974,7 +974,7 @@ static void shutdown(struct esp_struct *info)
 	serial_out(info, UART_ESI_CMD1, ESI_SET_SRV_MASK);
 	serial_out(info, UART_ESI_CMD2, 0x00);
 
-	if (!info->tty || (info->tty->termios->c_cflag & HUPCL))
+	if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL))
 		info->MCR &= ~(UART_MCR_DTR|UART_MCR_RTS);
 
 	info->MCR &= ~UART_MCR_OUT2;
@@ -982,10 +982,10 @@ static void shutdown(struct esp_struct *info)
 	serial_out(info, UART_ESI_CMD2, UART_MCR);
 	serial_out(info, UART_ESI_CMD2, info->MCR);
 
-	if (info->tty)
-		set_bit(TTY_IO_ERROR, &info->tty->flags);
+	if (info->port.tty)
+		set_bit(TTY_IO_ERROR, &info->port.tty->flags);
 
-	info->flags &= ~ASYNC_INITIALIZED;
+	info->port.flags &= ~ASYNC_INITIALIZED;
 	spin_unlock_irqrestore(&info->lock, flags);
 }
 
@@ -1002,10 +1002,10 @@ static void change_speed(struct esp_struct *info)
 	unsigned char flow1 = 0, flow2 = 0;
 	unsigned long flags;
 
-	if (!info->tty || !info->tty->termios)
+	if (!info->port.tty || !info->port.tty->termios)
 		return;
-	cflag = info->tty->termios->c_cflag;
-	port = info->port;
+	cflag = info->port.tty->termios->c_cflag;
+	port = info->io_port;
 
 	/* byte size and parity */
 	switch (cflag & CSIZE) {
@@ -1029,9 +1029,9 @@ static void change_speed(struct esp_struct *info)
 	if (cflag & CMSPAR)
 		cval |= UART_LCR_SPAR;
 #endif
-	baud = tty_get_baud_rate(info->tty);
+	baud = tty_get_baud_rate(info->port.tty);
 	if (baud == 38400 &&
-		((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST))
+		((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST))
 		quot = info->custom_divisor;
 	else {
 		if (baud == 134) /* Special case since 134 is really 134.5 */
@@ -1046,49 +1046,49 @@ static void change_speed(struct esp_struct *info)
 	if (baud) {
 		/* Actual rate */
 		baud = BASE_BAUD/quot;
-		tty_encode_baud_rate(info->tty, baud, baud);
+		tty_encode_baud_rate(info->port.tty, baud, baud);
 	}
 	info->timeout = ((1024 * HZ * bits * quot) / BASE_BAUD) + (HZ / 50);
 
 	/* CTS flow control flag and modem status interrupts */
 	/* info->IER &= ~UART_IER_MSI; */
 	if (cflag & CRTSCTS) {
-		info->flags |= ASYNC_CTS_FLOW;
+		info->port.flags |= ASYNC_CTS_FLOW;
 		/* info->IER |= UART_IER_MSI; */
 		flow1 = 0x04;
 		flow2 = 0x10;
 	} else
-		info->flags &= ~ASYNC_CTS_FLOW;
+		info->port.flags &= ~ASYNC_CTS_FLOW;
 	if (cflag & CLOCAL)
-		info->flags &= ~ASYNC_CHECK_CD;
+		info->port.flags &= ~ASYNC_CHECK_CD;
 	else
-		info->flags |= ASYNC_CHECK_CD;
+		info->port.flags |= ASYNC_CHECK_CD;
 
 	/*
 	 * Set up parity check flag
 	 */
 	info->read_status_mask = UART_LSR_OE | UART_LSR_THRE | UART_LSR_DR;
-	if (I_INPCK(info->tty))
+	if (I_INPCK(info->port.tty))
 		info->read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (I_BRKINT(info->tty) || I_PARMRK(info->tty))
+	if (I_BRKINT(info->port.tty) || I_PARMRK(info->port.tty))
 		info->read_status_mask |= UART_LSR_BI;
 
 	info->ignore_status_mask = 0;
 #if 0
 	/* This should be safe, but for some broken bits of hardware... */
-	if (I_IGNPAR(info->tty)) {
+	if (I_IGNPAR(info->port.tty)) {
 		info->ignore_status_mask |= UART_LSR_PE | UART_LSR_FE;
 		info->read_status_mask |= UART_LSR_PE | UART_LSR_FE;
 	}
 #endif
-	if (I_IGNBRK(info->tty)) {
+	if (I_IGNBRK(info->port.tty)) {
 		info->ignore_status_mask |= UART_LSR_BI;
 		info->read_status_mask |= UART_LSR_BI;
 		/*
 		 * If we're ignore parity and break indicators, ignore
 		 * overruns too.  (For real raw support).
 		 */
-		if (I_IGNPAR(info->tty)) {
+		if (I_IGNPAR(info->port.tty)) {
 			info->ignore_status_mask |= UART_LSR_OE | \
 				UART_LSR_PE | UART_LSR_FE;
 			info->read_status_mask |= UART_LSR_OE | \
@@ -1096,7 +1096,7 @@ static void change_speed(struct esp_struct *info)
 		}
 	}
 
-	if (I_IXOFF(info->tty))
+	if (I_IXOFF(info->port.tty))
 		flow1 |= 0x81;
 
 	spin_lock_irqsave(&info->lock, flags);
@@ -1116,10 +1116,10 @@ static void change_speed(struct esp_struct *info)
 	serial_out(info, UART_ESI_CMD2, flow2);
 
 	/* set flow control characters (XON/XOFF only) */
-	if (I_IXOFF(info->tty)) {
+	if (I_IXOFF(info->port.tty)) {
 		serial_out(info, UART_ESI_CMD1, ESI_SET_FLOW_CHARS);
-		serial_out(info, UART_ESI_CMD2, START_CHAR(info->tty));
-		serial_out(info, UART_ESI_CMD2, STOP_CHAR(info->tty));
+		serial_out(info, UART_ESI_CMD2, START_CHAR(info->port.tty));
+		serial_out(info, UART_ESI_CMD2, STOP_CHAR(info->port.tty));
 		serial_out(info, UART_ESI_CMD2, 0x10);
 		serial_out(info, UART_ESI_CMD2, 0x21);
 		switch (cflag & CSIZE) {
@@ -1355,9 +1355,9 @@ static int get_serial_info(struct esp_struct *info,
 	memset(&tmp, 0, sizeof(tmp));
 	tmp.type = PORT_16550A;
 	tmp.line = info->line;
-	tmp.port = info->port;
+	tmp.port = info->io_port;
 	tmp.irq = info->irq;
-	tmp.flags = info->flags;
+	tmp.flags = info->port.flags;
 	tmp.xmit_fifo_size = 1024;
 	tmp.baud_base = BASE_BAUD;
 	tmp.close_delay = info->close_delay;
@@ -1407,7 +1407,7 @@ static int set_serial_info(struct esp_struct *info,
 
 	if ((new_serial.type != PORT_16550A) ||
 	    (new_serial.hub6) ||
-	    (info->port != new_serial.port) ||
+	    (info->io_port != new_serial.port) ||
 	    (new_serial.baud_base != BASE_BAUD) ||
 	    (new_serial.irq > 15) ||
 	    (new_serial.irq < 2) ||
@@ -1425,9 +1425,9 @@ static int set_serial_info(struct esp_struct *info,
 		if (change_irq ||
 		    (new_serial.close_delay != info->close_delay) ||
 		    ((new_serial.flags & ~ASYNC_USR_MASK) !=
-		     (info->flags & ~ASYNC_USR_MASK)))
+		     (info->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
-		info->flags = ((info->flags & ~ASYNC_USR_MASK) |
+		info->port.flags = ((info->port.flags & ~ASYNC_USR_MASK) |
 			       (new_serial.flags & ASYNC_USR_MASK));
 		info->custom_divisor = new_serial.custom_divisor;
 	} else {
@@ -1441,9 +1441,9 @@ static int set_serial_info(struct esp_struct *info,
 				if ((current_async->line >= info->line) &&
 				    (current_async->line < (info->line + 8))) {
 					if (current_async == info) {
-						if (current_async->count > 1)
+						if (current_async->port.count > 1)
 							return -EBUSY;
-					} else if (current_async->count)
+					} else if (current_async->port.count)
 						return -EBUSY;
 				}
 
@@ -1456,7 +1456,7 @@ static int set_serial_info(struct esp_struct *info,
 		 * At this point, we start making changes.....
 		 */
 
-		info->flags = ((info->flags & ~ASYNC_FLAGS) |
+		info->port.flags = ((info->port.flags & ~ASYNC_FLAGS) |
 			       (new_serial.flags & ASYNC_FLAGS));
 		info->custom_divisor = new_serial.custom_divisor;
 		info->close_delay = new_serial.close_delay * HZ/100;
@@ -1487,18 +1487,18 @@ static int set_serial_info(struct esp_struct *info,
 		}
 	}
 
-	if (info->flags & ASYNC_INITIALIZED) {
-		if (((old_info.flags & ASYNC_SPD_MASK) !=
-		     (info->flags & ASYNC_SPD_MASK)) ||
+	if (info->port.flags & ASYNC_INITIALIZED) {
+		if (((old_info.port.flags & ASYNC_SPD_MASK) !=
+		     (info->port.flags & ASYNC_SPD_MASK)) ||
 		    (old_info.custom_divisor != info->custom_divisor)) {
-			if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
-				info->tty->alt_speed = 57600;
-			if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
-				info->tty->alt_speed = 115200;
-			if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
-				info->tty->alt_speed = 230400;
-			if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
-				info->tty->alt_speed = 460800;
+			if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+				info->port.tty->alt_speed = 57600;
+			if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+				info->port.tty->alt_speed = 115200;
+			if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
+				info->port.tty->alt_speed = 230400;
+			if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
+				info->port.tty->alt_speed = 460800;
 			change_speed(info);
 		}
 	} else
@@ -1554,9 +1554,9 @@ static int set_esp_config(struct esp_struct *info,
 
 			while (current_async) {
 				if (current_async == info) {
-					if (current_async->count > 1)
+					if (current_async->port.count > 1)
 						return -EBUSY;
-				} else if (current_async->count)
+				} else if (current_async->port.count)
 					return -EBUSY;
 
 				current_async = current_async->next_port;
@@ -1578,7 +1578,7 @@ static int set_esp_config(struct esp_struct *info,
 			spin_unlock_irqrestore(&info->lock, flags);
 		} else {
 			/* DMA mode to PIO mode only */
-			if (info->count > 1)
+			if (info->port.count > 1)
 				return -EBUSY;
 
 			shutdown(info);
@@ -1634,7 +1634,7 @@ static int set_esp_config(struct esp_struct *info,
 		spin_unlock_irqrestore(&info->lock, flags);
 	}
 
-	if (!(info->flags & ASYNC_INITIALIZED))
+	if (!(info->port.flags & ASYNC_INITIALIZED))
 		retval = startup(info);
 
 	return retval;
@@ -1917,9 +1917,9 @@ static void rs_close(struct tty_struct *tty, struct file *filp)
 
 #ifdef SERIAL_DEBUG_OPEN
 	printk(KERN_DEBUG "rs_close ttys%d, count = %d\n",
-						info->line, info->count);
+						info->line, info->port.count);
 #endif
-	if (tty->count == 1 && info->count != 1) {
+	if (tty->count == 1 && info->port.count != 1) {
 		/*
 		 * Uh, oh.  tty->count is 1, which means that the tty
 		 * structure will be freed.  Info->count should always
@@ -1927,19 +1927,19 @@ static void rs_close(struct tty_struct *tty, struct file *filp)
 		 * one, we've got real problems, since it means the
 		 * serial port won't be shutdown.
 		 */
-		printk(KERN_DEBUG "rs_close: bad serial port count; tty->count is 1, info->count is %d\n", info->count);
-		info->count = 1;
+		printk(KERN_DEBUG "rs_close: bad serial port count; tty->count is 1, info->port.count is %d\n", info->port.count);
+		info->port.count = 1;
 	}
-	if (--info->count < 0) {
+	if (--info->port.count < 0) {
 		printk(KERN_ERR "rs_close: bad serial port count for ttys%d: %d\n",
-		       info->line, info->count);
-		info->count = 0;
+		       info->line, info->port.count);
+		info->port.count = 0;
 	}
-	if (info->count) {
+	if (info->port.count) {
 		DBG_CNT("before DEC-2");
 		goto out;
 	}
-	info->flags |= ASYNC_CLOSING;
+	info->port.flags |= ASYNC_CLOSING;
 
 	spin_unlock_irqrestore(&info->lock, flags);
 	/*
@@ -1958,7 +1958,7 @@ static void rs_close(struct tty_struct *tty, struct file *filp)
 	/* info->IER &= ~UART_IER_RLSI; */
 	info->IER &= ~UART_IER_RDI;
 	info->read_status_mask &= ~UART_LSR_DR;
-	if (info->flags & ASYNC_INITIALIZED) {
+	if (info->port.flags & ASYNC_INITIALIZED) {
 
 		spin_lock_irqsave(&info->lock, flags);
 		serial_out(info, UART_ESI_CMD1, ESI_SET_SRV_MASK);
@@ -1981,15 +1981,15 @@ static void rs_close(struct tty_struct *tty, struct file *filp)
 	rs_flush_buffer(tty);
 	tty_ldisc_flush(tty);
 	tty->closing = 0;
-	info->tty = NULL;
+	info->port.tty = NULL;
 
-	if (info->blocked_open) {
+	if (info->port.blocked_open) {
 		if (info->close_delay)
 			msleep_interruptible(jiffies_to_msecs(info->close_delay));
-		wake_up_interruptible(&info->open_wait);
+		wake_up_interruptible(&info->port.open_wait);
 	}
-	info->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&info->close_wait);
+	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
+	wake_up_interruptible(&info->port.close_wait);
 	return;
 
 out:
@@ -2047,10 +2047,10 @@ static void esp_hangup(struct tty_struct *tty)
 
 	rs_flush_buffer(tty);
 	shutdown(info);
-	info->count = 0;
-	info->flags &= ~ASYNC_NORMAL_ACTIVE;
-	info->tty = NULL;
-	wake_up_interruptible(&info->open_wait);
+	info->port.count = 0;
+	info->port.flags &= ~ASYNC_NORMAL_ACTIVE;
+	info->port.tty = NULL;
+	wake_up_interruptible(&info->port.open_wait);
 }
 
 /*
@@ -2071,11 +2071,11 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	 * until it's done, and then try again.
 	 */
 	if (tty_hung_up_p(filp) ||
-	    (info->flags & ASYNC_CLOSING)) {
-		if (info->flags & ASYNC_CLOSING)
-			interruptible_sleep_on(&info->close_wait);
+	    (info->port.flags & ASYNC_CLOSING)) {
+		if (info->port.flags & ASYNC_CLOSING)
+			interruptible_sleep_on(&info->port.close_wait);
 #ifdef SERIAL_DO_RESTART
-		if (info->flags & ASYNC_HUP_NOTIFY)
+		if (info->port.flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -2090,7 +2090,7 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		info->flags |= ASYNC_NORMAL_ACTIVE;
+		info->port.flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -2100,20 +2100,20 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->count is dropped by one, so that
+	 * this loop, info->port.count is dropped by one, so that
 	 * rs_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&info->open_wait, &wait);
+	add_wait_queue(&info->port.open_wait, &wait);
 #ifdef SERIAL_DEBUG_OPEN
 	printk(KERN_DEBUG "block_til_ready before block: ttys%d, count = %d\n",
-	       info->line, info->count);
+	       info->line, info->port.count);
 #endif
 	spin_lock_irqsave(&info->lock, flags);
 	if (!tty_hung_up_p(filp))
-		info->count--;
-	info->blocked_open++;
+		info->port.count--;
+	info->port.blocked_open++;
 	while (1) {
 		if ((tty->termios->c_cflag & CBAUD)) {
 			unsigned int scratch;
@@ -2128,9 +2128,9 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		}
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(info->flags & ASYNC_INITIALIZED)) {
+		    !(info->port.flags & ASYNC_INITIALIZED)) {
 #ifdef SERIAL_DO_RESTART
-			if (info->flags & ASYNC_HUP_NOTIFY)
+			if (info->port.flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
@@ -2144,7 +2144,7 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		if (serial_in(info, UART_ESI_STAT2) & UART_MSR_DCD)
 			do_clocal = 1;
 
-		if (!(info->flags & ASYNC_CLOSING) &&
+		if (!(info->port.flags & ASYNC_CLOSING) &&
 		    (do_clocal))
 			break;
 		if (signal_pending(current)) {
@@ -2153,25 +2153,25 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		}
 #ifdef SERIAL_DEBUG_OPEN
 		printk(KERN_DEBUG "block_til_ready blocking: ttys%d, count = %d\n",
-		       info->line, info->count);
+		       info->line, info->port.count);
 #endif
 		spin_unlock_irqrestore(&info->lock, flags);
 		schedule();
 		spin_lock_irqsave(&info->lock, flags);
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->open_wait, &wait);
+	remove_wait_queue(&info->port.open_wait, &wait);
 	if (!tty_hung_up_p(filp))
-		info->count++;
-	info->blocked_open--;
+		info->port.count++;
+	info->port.blocked_open--;
 	spin_unlock_irqrestore(&info->lock, flags);
 #ifdef SERIAL_DEBUG_OPEN
 	printk(KERN_DEBUG "block_til_ready after blocking: ttys%d, count = %d\n",
-	       info->line, info->count);
+	       info->line, info->port.count);
 #endif
 	if (retval)
 		return retval;
-	info->flags |= ASYNC_NORMAL_ACTIVE;
+	info->port.flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }
 
@@ -2204,12 +2204,12 @@ static int esp_open(struct tty_struct *tty, struct file *filp)
 	}
 
 #ifdef SERIAL_DEBUG_OPEN
-	printk(KERN_DEBUG "esp_open %s, count = %d\n", tty->name, info->count);
+	printk(KERN_DEBUG "esp_open %s, count = %d\n", tty->name, info->port.count);
 #endif
 	spin_lock_irqsave(&info->lock, flags);
-	info->count++;
+	info->port.count++;
 	tty->driver_data = info;
-	info->tty = tty;
+	info->port.tty = tty;
 
 	spin_unlock_irqrestore(&info->lock, flags);
 
@@ -2263,7 +2263,7 @@ static int autoconfig(struct esp_struct *info)
 	int port_detected = 0;
 	unsigned long flags;
 
-	if (!request_region(info->port, REGION_SIZE, "esp serial"))
+	if (!request_region(info->io_port, REGION_SIZE, "esp serial"))
 		return -EIO;
 
 	spin_lock_irqsave(&info->lock, flags);
@@ -2300,7 +2300,7 @@ static int autoconfig(struct esp_struct *info)
 		}
 	}
 	if (!port_detected)
-		release_region(info->port, REGION_SIZE);
+		release_region(info->io_port, REGION_SIZE);
 
 	spin_unlock_irqrestore(&info->lock, flags);
 	return (port_detected);
@@ -2414,7 +2414,7 @@ static int __init espserial_init(void)
 	offset = 0;
 
 	do {
-		info->port = esp[i] + offset;
+		info->io_port = esp[i] + offset;
 		info->irq = irq[i];
 		info->line = (i * 8) + (offset / 8);
 
@@ -2425,9 +2425,9 @@ static int __init espserial_init(void)
 		}
 
 		info->custom_divisor = (divisor[i] >> (offset / 2)) & 0xf;
-		info->flags = STD_COM_FLAGS;
+		info->port.flags = STD_COM_FLAGS;
 		if (info->custom_divisor)
-			info->flags |= ASYNC_SPD_CUST;
+			info->port.flags |= ASYNC_SPD_CUST;
 		info->magic = ESP_MAGIC;
 		info->close_delay = 5*HZ/10;
 		info->closing_wait = 30*HZ;
@@ -2436,13 +2436,13 @@ static int __init espserial_init(void)
 		info->config.flow_off = flow_off;
 		info->config.pio_threshold = pio_threshold;
 		info->next_port = ports;
-		init_waitqueue_head(&info->open_wait);
-		init_waitqueue_head(&info->close_wait);
+		init_waitqueue_head(&info->port.open_wait);
+		init_waitqueue_head(&info->port.close_wait);
 		init_waitqueue_head(&info->delta_msr_wait);
 		init_waitqueue_head(&info->break_wait);
 		ports = info;
 		printk(KERN_INFO "ttyP%d at 0x%04x (irq = %d) is an ESP ",
-			info->line, info->port, info->irq);
+			info->line, info->io_port, info->irq);
 
 		if (info->line % 8) {
 			printk("secondary port\n");
@@ -2498,8 +2498,8 @@ static void __exit espserial_exit(void)
 	put_tty_driver(esp_driver);
 
 	while (ports) {
-		if (ports->port)
-			release_region(ports->port, REGION_SIZE);
+		if (ports->io_port)
+			release_region(ports->io_port, REGION_SIZE);
 		temp_async = ports->next_port;
 		kfree(ports);
 		ports = temp_async;
diff --git a/include/linux/hayesesp.h b/include/linux/hayesesp.h
index 2177ee5b2fe2..940aeb51d53f 100644
--- a/include/linux/hayesesp.h
+++ b/include/linux/hayesesp.h
@@ -76,11 +76,10 @@ struct hayes_esp_config {
 
 struct esp_struct {
 	int			magic;
+	struct tty_port		port;
 	spinlock_t		lock;
-	int			port;
+	int			io_port;
 	int			irq;
-	int			flags; 		/* defined in tty.h */
-	struct tty_struct 	*tty;
 	int			read_status_mask;
 	int			ignore_status_mask;
 	int			timeout;
@@ -93,14 +92,10 @@ struct esp_struct {
 	int			MCR; 	/* Modem control register */
 	unsigned long		last_active;
 	int			line;
-	int			count;	    /* # of fd on device */
-	int			blocked_open; /* # of blocked opens */
 	unsigned char 		*xmit_buf;
 	int			xmit_head;
 	int			xmit_tail;
 	int			xmit_cnt;
-	wait_queue_head_t	open_wait;
-	wait_queue_head_t	close_wait;
 	wait_queue_head_t	delta_msr_wait;
 	wait_queue_head_t	break_wait;
 	struct async_icount	icount;	/* kernel counters for the 4 input interrupts */
-- 
cgit v1.2.3


From b5391e29f428d11755ca2c91074c6db6f5c69d7c Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:55:20 +0100
Subject: gs: use tty_port

Switch drivers using the old "generic serial" driver to use the tty_port
structures

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/generic_serial.c  | 158 ++++++++++++++++++++---------------------
 drivers/char/rio/rio_linux.c   |  20 +++---
 drivers/char/rio/riocmd.c      |  10 +--
 drivers/char/rio/riointr.c     |  10 +--
 drivers/char/rio/rioparam.c    |   2 +-
 drivers/char/rio/riotty.c      |  16 ++---
 drivers/char/sx.c              | 115 +++++++++++++++---------------
 include/linux/generic_serial.h |   8 +--
 8 files changed, 166 insertions(+), 173 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/generic_serial.c b/drivers/char/generic_serial.c
index 252f73e48596..19d3afb0e50c 100644
--- a/drivers/char/generic_serial.c
+++ b/drivers/char/generic_serial.c
@@ -60,7 +60,7 @@ int gs_put_char(struct tty_struct * tty, unsigned char ch)
 
 	if (!port) return 0;
 
-	if (! (port->flags & ASYNC_INITIALIZED)) return 0;
+	if (! (port->port.flags & ASYNC_INITIALIZED)) return 0;
 
 	/* Take a lock on the serial tranmit buffer! */
 	mutex_lock(& port->port_write_mutex);
@@ -103,7 +103,7 @@ int gs_write(struct tty_struct * tty,
 
 	if (!port) return 0;
 
-	if (! (port->flags & ASYNC_INITIALIZED))
+	if (! (port->port.flags & ASYNC_INITIALIZED))
 		return 0;
 
 	/* get exclusive "write" access to this port (problem 3) */
@@ -141,13 +141,13 @@ int gs_write(struct tty_struct * tty,
 	mutex_unlock(& port->port_write_mutex);
 
 	gs_dprintk (GS_DEBUG_WRITE, "write: interrupts are %s\n", 
-	            (port->flags & GS_TX_INTEN)?"enabled": "disabled"); 
+	            (port->port.flags & GS_TX_INTEN)?"enabled": "disabled");
 
 	if (port->xmit_cnt && 
 	    !tty->stopped && 
 	    !tty->hw_stopped &&
-	    !(port->flags & GS_TX_INTEN)) {
-		port->flags |= GS_TX_INTEN;
+	    !(port->port.flags & GS_TX_INTEN)) {
+		port->port.flags |= GS_TX_INTEN;
 		port->rd->enable_tx_interrupts (port);
 	}
 	func_exit ();
@@ -208,7 +208,7 @@ static int gs_wait_tx_flushed (void * ptr, unsigned long timeout)
 	gs_dprintk (GS_DEBUG_FLUSH, "port=%p.\n", port);
 	if (port) {
 		gs_dprintk (GS_DEBUG_FLUSH, "xmit_cnt=%x, xmit_buf=%p, tty=%p.\n", 
-		port->xmit_cnt, port->xmit_buf, port->tty);
+		port->xmit_cnt, port->xmit_buf, port->port.tty);
 	}
 
 	if (!port || port->xmit_cnt < 0 || !port->xmit_buf) {
@@ -217,7 +217,7 @@ static int gs_wait_tx_flushed (void * ptr, unsigned long timeout)
 		return -EINVAL;  /* This is an error which we don't know how to handle. */
 	}
 
-	rcib = gs_real_chars_in_buffer(port->tty);
+	rcib = gs_real_chars_in_buffer(port->port.tty);
 
 	if(rcib <= 0) {
 		gs_dprintk (GS_DEBUG_FLUSH, "nothing to wait for.\n");
@@ -236,7 +236,7 @@ static int gs_wait_tx_flushed (void * ptr, unsigned long timeout)
 
 	/* the expression is actually jiffies < end_jiffies, but that won't
 	   work around the wraparound. Tricky eh? */
-	while ((charsleft = gs_real_chars_in_buffer (port->tty)) &&
+	while ((charsleft = gs_real_chars_in_buffer (port->port.tty)) &&
 	        time_after (end_jiffies, jiffies)) {
 		/* Units check: 
 		   chars * (bits/char) * (jiffies /sec) / (bits/sec) = jiffies!
@@ -309,7 +309,7 @@ void gs_flush_chars(struct tty_struct * tty)
 	}
 
 	/* Beats me -- REW */
-	port->flags |= GS_TX_INTEN;
+	port->port.flags |= GS_TX_INTEN;
 	port->rd->enable_tx_interrupts (port);
 	func_exit ();
 }
@@ -329,8 +329,8 @@ void gs_stop(struct tty_struct * tty)
 
 	if (port->xmit_cnt && 
 	    port->xmit_buf && 
-	    (port->flags & GS_TX_INTEN) ) {
-		port->flags &= ~GS_TX_INTEN;
+	    (port->port.flags & GS_TX_INTEN) ) {
+		port->port.flags &= ~GS_TX_INTEN;
 		port->rd->disable_tx_interrupts (port);
 	}
 	func_exit ();
@@ -349,8 +349,8 @@ void gs_start(struct tty_struct * tty)
 
 	if (port->xmit_cnt && 
 	    port->xmit_buf && 
-	    !(port->flags & GS_TX_INTEN) ) {
-		port->flags |= GS_TX_INTEN;
+	    !(port->port.flags & GS_TX_INTEN) ) {
+		port->port.flags |= GS_TX_INTEN;
 		port->rd->enable_tx_interrupts (port);
 	}
 	func_exit ();
@@ -365,7 +365,7 @@ static void gs_shutdown_port (struct gs_port *port)
 	
 	if (!port) return;
 	
-	if (!(port->flags & ASYNC_INITIALIZED))
+	if (!(port->port.flags & ASYNC_INITIALIZED))
 		return;
 
 	spin_lock_irqsave(&port->driver_lock, flags);
@@ -375,12 +375,12 @@ static void gs_shutdown_port (struct gs_port *port)
 		port->xmit_buf = NULL;
 	}
 
-	if (port->tty)
-		set_bit(TTY_IO_ERROR, &port->tty->flags);
+	if (port->port.tty)
+		set_bit(TTY_IO_ERROR, &port->port.tty->flags);
 
 	port->rd->shutdown_port (port);
 
-	port->flags &= ~ASYNC_INITIALIZED;
+	port->port.flags &= ~ASYNC_INITIALIZED;
 	spin_unlock_irqrestore(&port->driver_lock, flags);
 
 	func_exit();
@@ -396,16 +396,16 @@ void gs_hangup(struct tty_struct *tty)
 	if (!tty) return;
 
 	port = tty->driver_data;
-	tty = port->tty;
+	tty = port->port.tty;
 	if (!tty) 
 		return;
 
 	gs_shutdown_port (port);
-	port->flags &= ~(ASYNC_NORMAL_ACTIVE|GS_ACTIVE);
-	port->tty = NULL;
-	port->count = 0;
+	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|GS_ACTIVE);
+	port->port.tty = NULL;
+	port->port.count = 0;
 
-	wake_up_interruptible(&port->open_wait);
+	wake_up_interruptible(&port->port.open_wait);
 	func_exit ();
 }
 
@@ -424,7 +424,7 @@ int gs_block_til_ready(void *port_, struct file * filp)
 
 	if (!port) return 0;
 
-	tty = port->tty;
+	tty = port->port.tty;
 
 	if (!tty) return 0;
 
@@ -433,9 +433,9 @@ int gs_block_til_ready(void *port_, struct file * filp)
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
-	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&port->close_wait);
-		if (port->flags & ASYNC_HUP_NOTIFY)
+	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&port->port.close_wait);
+		if (port->port.flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -449,7 +449,7 @@ int gs_block_til_ready(void *port_, struct file * filp)
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		port->flags |= ASYNC_NORMAL_ACTIVE;
+		port->port.flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -461,34 +461,34 @@ int gs_block_til_ready(void *port_, struct file * filp)
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, port->count is dropped by one, so that
+	 * this loop, port->port.count is dropped by one, so that
 	 * rs_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
 
-	add_wait_queue(&port->open_wait, &wait);
+	add_wait_queue(&port->port.open_wait, &wait);
 
 	gs_dprintk (GS_DEBUG_BTR, "after add waitq.\n"); 
 	spin_lock_irqsave(&port->driver_lock, flags);
 	if (!tty_hung_up_p(filp)) {
-		port->count--;
+		port->port.count--;
 	}
 	spin_unlock_irqrestore(&port->driver_lock, flags);
-	port->blocked_open++;
+	port->port.blocked_open++;
 	while (1) {
 		CD = port->rd->get_CD (port);
 		gs_dprintk (GS_DEBUG_BTR, "CD is now %d.\n", CD);
 		set_current_state (TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(port->flags & ASYNC_INITIALIZED)) {
-			if (port->flags & ASYNC_HUP_NOTIFY)
+		    !(port->port.flags & ASYNC_INITIALIZED)) {
+			if (port->port.flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(port->flags & ASYNC_CLOSING) &&
+		if (!(port->port.flags & ASYNC_CLOSING) &&
 		    (do_clocal || CD))
 			break;
 		gs_dprintk (GS_DEBUG_BTR, "signal_pending is now: %d (%lx)\n", 
@@ -500,17 +500,17 @@ int gs_block_til_ready(void *port_, struct file * filp)
 		schedule();
 	}
 	gs_dprintk (GS_DEBUG_BTR, "Got out of the loop. (%d)\n",
-		    port->blocked_open);
+		    port->port.blocked_open);
 	set_current_state (TASK_RUNNING);
-	remove_wait_queue(&port->open_wait, &wait);
+	remove_wait_queue(&port->port.open_wait, &wait);
 	if (!tty_hung_up_p(filp)) {
-		port->count++;
+		port->port.count++;
 	}
-	port->blocked_open--;
+	port->port.blocked_open--;
 	if (retval)
 		return retval;
 
-	port->flags |= ASYNC_NORMAL_ACTIVE;
+	port->port.flags |= ASYNC_NORMAL_ACTIVE;
 	func_exit ();
 	return 0;
 }			 
@@ -529,10 +529,10 @@ void gs_close(struct tty_struct * tty, struct file * filp)
 
 	if (!port) return;
 
-	if (!port->tty) {
+	if (!port->port.tty) {
 		/* This seems to happen when this is called from vhangup. */
-		gs_dprintk (GS_DEBUG_CLOSE, "gs: Odd: port->tty is NULL\n");
-		port->tty = tty;
+		gs_dprintk (GS_DEBUG_CLOSE, "gs: Odd: port->port.tty is NULL\n");
+		port->port.tty = tty;
 	}
 
 	spin_lock_irqsave(&port->driver_lock, flags);
@@ -545,23 +545,23 @@ void gs_close(struct tty_struct * tty, struct file * filp)
 		return;
 	}
 
-	if ((tty->count == 1) && (port->count != 1)) {
+	if ((tty->count == 1) && (port->port.count != 1)) {
 		printk(KERN_ERR "gs: gs_close port %p: bad port count;"
-		       " tty->count is 1, port count is %d\n", port, port->count);
-		port->count = 1;
+		       " tty->count is 1, port count is %d\n", port, port->port.count);
+		port->port.count = 1;
 	}
-	if (--port->count < 0) {
-		printk(KERN_ERR "gs: gs_close port %p: bad port count: %d\n", port, port->count);
-		port->count = 0;
+	if (--port->port.count < 0) {
+		printk(KERN_ERR "gs: gs_close port %p: bad port count: %d\n", port, port->port.count);
+		port->port.count = 0;
 	}
 
-	if (port->count) {
-		gs_dprintk(GS_DEBUG_CLOSE, "gs_close port %p: count: %d\n", port, port->count);
+	if (port->port.count) {
+		gs_dprintk(GS_DEBUG_CLOSE, "gs_close port %p: count: %d\n", port, port->port.count);
 		spin_unlock_irqrestore(&port->driver_lock, flags);
 		func_exit ();
 		return;
 	}
-	port->flags |= ASYNC_CLOSING;
+	port->port.flags |= ASYNC_CLOSING;
 
 	/*
 	 * Now we wait for the transmit buffer to clear; and we notify 
@@ -585,7 +585,7 @@ void gs_close(struct tty_struct * tty, struct file * filp)
 	if (port->closing_wait != ASYNC_CLOSING_WAIT_NONE)
 		gs_wait_tx_flushed (port, port->closing_wait);
 
-	port->flags &= ~GS_ACTIVE;
+	port->port.flags &= ~GS_ACTIVE;
 
 	gs_flush_buffer(tty);
 
@@ -595,18 +595,18 @@ void gs_close(struct tty_struct * tty, struct file * filp)
 	port->event = 0;
 	port->rd->close (port);
 	port->rd->shutdown_port (port);
-	port->tty = NULL;
+	port->port.tty = NULL;
 
-	if (port->blocked_open) {
+	if (port->port.blocked_open) {
 		if (port->close_delay) {
 			spin_unlock_irqrestore(&port->driver_lock, flags);
 			msleep_interruptible(jiffies_to_msecs(port->close_delay));
 			spin_lock_irqsave(&port->driver_lock, flags);
 		}
-		wake_up_interruptible(&port->open_wait);
+		wake_up_interruptible(&port->port.open_wait);
 	}
-	port->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING | ASYNC_INITIALIZED);
-	wake_up_interruptible(&port->close_wait);
+	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING | ASYNC_INITIALIZED);
+	wake_up_interruptible(&port->port.close_wait);
 
 	func_exit ();
 }
@@ -626,10 +626,10 @@ void gs_set_termios (struct tty_struct * tty,
 	port = tty->driver_data;
 
 	if (!port) return;
-	if (!port->tty) {
+	if (!port->port.tty) {
 		/* This seems to happen when this is called after gs_close. */
-		gs_dprintk (GS_DEBUG_TERMIOS, "gs: Odd: port->tty is NULL\n");
-		port->tty = tty;
+		gs_dprintk (GS_DEBUG_TERMIOS, "gs: Odd: port->port.tty is NULL\n");
+		port->port.tty = tty;
 	}
 
 
@@ -651,15 +651,15 @@ void gs_set_termios (struct tty_struct * tty,
 	baudrate = tty_get_baud_rate(tty);
 
 	if ((tiosp->c_cflag & CBAUD) == B38400) {
-		if (     (port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+		if (     (port->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
 			baudrate = 57600;
-		else if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+		else if ((port->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
 			baudrate = 115200;
-		else if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
+		else if ((port->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
 			baudrate = 230400;
-		else if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
+		else if ((port->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
 			baudrate = 460800;
-		else if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
+		else if ((port->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
 			baudrate = (port->baud_base / port->custom_divisor);
 	}
 
@@ -715,7 +715,7 @@ int gs_init_port(struct gs_port *port)
 
 	func_enter ();
 
-	if (port->flags & ASYNC_INITIALIZED) {
+	if (port->port.flags & ASYNC_INITIALIZED) {
 		func_exit ();
 		return 0;
 	}
@@ -737,15 +737,15 @@ int gs_init_port(struct gs_port *port)
 	}
 
 	spin_lock_irqsave (&port->driver_lock, flags);
-	if (port->tty) 
-		clear_bit(TTY_IO_ERROR, &port->tty->flags);
+	if (port->port.tty)
+		clear_bit(TTY_IO_ERROR, &port->port.tty->flags);
 	mutex_init(&port->port_write_mutex);
 	port->xmit_cnt = port->xmit_head = port->xmit_tail = 0;
 	spin_unlock_irqrestore(&port->driver_lock, flags);
-	gs_set_termios(port->tty, NULL);
+	gs_set_termios(port->port.tty, NULL);
 	spin_lock_irqsave (&port->driver_lock, flags);
-	port->flags |= ASYNC_INITIALIZED;
-	port->flags &= ~GS_TX_INTEN;
+	port->port.flags |= ASYNC_INITIALIZED;
+	port->port.flags &= ~GS_TX_INTEN;
 
 	spin_unlock_irqrestore(&port->driver_lock, flags);
 	func_exit ();
@@ -764,11 +764,11 @@ int gs_setserial(struct gs_port *port, struct serial_struct __user *sp)
 		if ((sio.baud_base != port->baud_base) ||
 		    (sio.close_delay != port->close_delay) ||
 		    ((sio.flags & ~ASYNC_USR_MASK) !=
-		     (port->flags & ~ASYNC_USR_MASK)))
+		     (port->port.flags & ~ASYNC_USR_MASK)))
 			return(-EPERM);
 	} 
 
-	port->flags = (port->flags & ~ASYNC_USR_MASK) |
+	port->port.flags = (port->port.flags & ~ASYNC_USR_MASK) |
 		(sio.flags & ASYNC_USR_MASK);
   
 	port->baud_base = sio.baud_base;
@@ -776,7 +776,7 @@ int gs_setserial(struct gs_port *port, struct serial_struct __user *sp)
 	port->closing_wait = sio.closing_wait;
 	port->custom_divisor = sio.custom_divisor;
 
-	gs_set_termios (port->tty, NULL);
+	gs_set_termios (port->port.tty, NULL);
 
 	return 0;
 }
@@ -793,7 +793,7 @@ int gs_getserial(struct gs_port *port, struct serial_struct __user *sp)
 	struct serial_struct    sio;
 
 	memset(&sio, 0, sizeof(struct serial_struct));
-	sio.flags = port->flags;
+	sio.flags = port->port.flags;
 	sio.baud_base = port->baud_base;
 	sio.close_delay = port->close_delay;
 	sio.closing_wait = port->closing_wait;
@@ -821,10 +821,10 @@ void gs_got_break(struct gs_port *port)
 {
 	func_enter ();
 
-	tty_insert_flip_char(port->tty, 0, TTY_BREAK);
-	tty_schedule_flip(port->tty);
-	if (port->flags & ASYNC_SAK) {
-		do_SAK (port->tty);
+	tty_insert_flip_char(port->port.tty, 0, TTY_BREAK);
+	tty_schedule_flip(port->port.tty);
+	if (port->port.flags & ASYNC_SAK) {
+		do_SAK (port->port.tty);
 	}
 
 	func_exit ();
diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c
index e49e6e6372f3..0cdfee152916 100644
--- a/drivers/char/rio/rio_linux.c
+++ b/drivers/char/rio/rio_linux.c
@@ -431,7 +431,7 @@ static void rio_disable_tx_interrupts(void *ptr)
 {
 	func_enter();
 
-	/*  port->gs.flags &= ~GS_TX_INTEN; */
+	/*  port->gs.port.flags &= ~GS_TX_INTEN; */
 
 	func_exit();
 }
@@ -455,7 +455,7 @@ static void rio_enable_tx_interrupts(void *ptr)
 	 * In general we cannot count on "tx empty" interrupts, although
 	 * the interrupt routine seems to be able to tell the difference.
 	 */
-	PortP->gs.flags &= ~GS_TX_INTEN;
+	PortP->gs.port.flags &= ~GS_TX_INTEN;
 
 	func_exit();
 }
@@ -510,7 +510,7 @@ static void rio_shutdown_port(void *ptr)
 	func_enter();
 
 	PortP = (struct Port *) ptr;
-	PortP->gs.tty = NULL;
+	PortP->gs.port.tty = NULL;
 	func_exit();
 }
 
@@ -529,7 +529,7 @@ static void rio_hungup(void *ptr)
 	func_enter();
 
 	PortP = (struct Port *) ptr;
-	PortP->gs.tty = NULL;
+	PortP->gs.port.tty = NULL;
 
 	func_exit();
 }
@@ -549,12 +549,12 @@ static void rio_close(void *ptr)
 
 	riotclose(ptr);
 
-	if (PortP->gs.count) {
-		printk(KERN_ERR "WARNING port count:%d\n", PortP->gs.count);
-		PortP->gs.count = 0;
+	if (PortP->gs.port.count) {
+		printk(KERN_ERR "WARNING port count:%d\n", PortP->gs.port.count);
+		PortP->gs.port.count = 0;
 	}
 
-	PortP->gs.tty = NULL;
+	PortP->gs.port.tty = NULL;
 	func_exit();
 }
 
@@ -849,8 +849,8 @@ static int rio_init_datastructures(void)
 		/*
 		 * Initializing wait queue
 		 */
-		init_waitqueue_head(&port->gs.open_wait);
-		init_waitqueue_head(&port->gs.close_wait);
+		init_waitqueue_head(&port->gs.port.open_wait);
+		init_waitqueue_head(&port->gs.port.close_wait);
 	}
 #else
 	/* We could postpone initializing them to when they are configured. */
diff --git a/drivers/char/rio/riocmd.c b/drivers/char/rio/riocmd.c
index 391f0b4da7ea..01f2654d5a2e 100644
--- a/drivers/char/rio/riocmd.c
+++ b/drivers/char/rio/riocmd.c
@@ -484,12 +484,12 @@ static int RIOCommandRup(struct rio_info *p, uint Rup, struct Host *HostP, struc
 				 ** If the device is a modem, then check the modem
 				 ** carrier.
 				 */
-				if (PortP->gs.tty == NULL)
+				if (PortP->gs.port.tty == NULL)
 					break;
-				if (PortP->gs.tty->termios == NULL)
+				if (PortP->gs.port.tty->termios == NULL)
 					break;
 
-				if (!(PortP->gs.tty->termios->c_cflag & CLOCAL) && ((PortP->State & (RIO_MOPEN | RIO_WOPEN)))) {
+				if (!(PortP->gs.port.tty->termios->c_cflag & CLOCAL) && ((PortP->State & (RIO_MOPEN | RIO_WOPEN)))) {
 
 					rio_dprintk(RIO_DEBUG_CMD, "Is there a Carrier?\n");
 					/*
@@ -506,7 +506,7 @@ static int RIOCommandRup(struct rio_info *p, uint Rup, struct Host *HostP, struc
 							 ** wakeup anyone in WOPEN
 							 */
 							if (PortP->State & (PORT_ISOPEN | RIO_WOPEN))
-								wake_up_interruptible(&PortP->gs.open_wait);
+								wake_up_interruptible(&PortP->gs.port.open_wait);
 						}
 					} else {
 						/*
@@ -514,7 +514,7 @@ static int RIOCommandRup(struct rio_info *p, uint Rup, struct Host *HostP, struc
 						 */
 						if (PortP->State & RIO_CARR_ON) {
 							if (PortP->State & (PORT_ISOPEN | RIO_WOPEN | RIO_MOPEN))
-								tty_hangup(PortP->gs.tty);
+								tty_hangup(PortP->gs.port.tty);
 							PortP->State &= ~RIO_CARR_ON;
 							rio_dprintk(RIO_DEBUG_CMD, "Carrirer just went down\n");
 						}
diff --git a/drivers/char/rio/riointr.c b/drivers/char/rio/riointr.c
index 11c7987821c4..71f87600907c 100644
--- a/drivers/char/rio/riointr.c
+++ b/drivers/char/rio/riointr.c
@@ -102,7 +102,7 @@ void RIOTxEnable(char *en)
 
 	PortP = (struct Port *) en;
 	p = (struct rio_info *) PortP->p;
-	tty = PortP->gs.tty;
+	tty = PortP->gs.port.tty;
 
 
 	rio_dprintk(RIO_DEBUG_INTR, "tx port %d: %d chars queued.\n", PortP->PortNum, PortP->gs.xmit_cnt);
@@ -158,7 +158,7 @@ void RIOTxEnable(char *en)
 	rio_spin_unlock_irqrestore(&PortP->portSem, flags);
 
 	if (PortP->gs.xmit_cnt <= (PortP->gs.wakeup_chars + 2 * PKT_MAX_DATA_LEN))
-		tty_wakeup(PortP->gs.tty);
+		tty_wakeup(PortP->gs.port.tty);
 
 }
 
@@ -241,7 +241,7 @@ void RIOServiceHost(struct rio_info *p, struct Host *HostP)
 			 ** find corresponding tty structure. The process of mapping
 			 ** the ports puts these here.
 			 */
-			ttyP = PortP->gs.tty;
+			ttyP = PortP->gs.port.tty;
 
 			/*
 			 ** Lock the port before we begin working on it.
@@ -335,7 +335,7 @@ void RIOServiceHost(struct rio_info *p, struct Host *HostP)
 			 ** find corresponding tty structure. The process of mapping
 			 ** the ports puts these here.
 			 */
-			ttyP = PortP->gs.tty;
+			ttyP = PortP->gs.port.tty;
 			/* If ttyP is NULL, the port is getting closed. Forget about it. */
 			if (!ttyP) {
 				rio_dprintk(RIO_DEBUG_INTR, "no tty, so skipping.\n");
@@ -542,7 +542,7 @@ static void RIOReceive(struct rio_info *p, struct Port *PortP)
 
 	intCount++;
 
-	TtyP = PortP->gs.tty;
+	TtyP = PortP->gs.port.tty;
 	if (!TtyP) {
 		rio_dprintk(RIO_DEBUG_INTR, "RIOReceive: tty is null. \n");
 		return;
diff --git a/drivers/char/rio/rioparam.c b/drivers/char/rio/rioparam.c
index 447ca34a6a72..d687c17be152 100644
--- a/drivers/char/rio/rioparam.c
+++ b/drivers/char/rio/rioparam.c
@@ -160,7 +160,7 @@ int RIOParam(struct Port *PortP, int cmd, int Modem, int SleepFlag)
 
 	func_enter();
 
-	TtyP = PortP->gs.tty;
+	TtyP = PortP->gs.port.tty;
 
 	rio_dprintk(RIO_DEBUG_PARAM, "RIOParam: Port:%d cmd:%d Modem:%d SleepFlag:%d Mapped: %d, tty=%p\n", PortP->PortNum, cmd, Modem, SleepFlag, PortP->Mapped, TtyP);
 
diff --git a/drivers/char/rio/riotty.c b/drivers/char/rio/riotty.c
index 95a88a4138e8..2fb49e89b324 100644
--- a/drivers/char/rio/riotty.c
+++ b/drivers/char/rio/riotty.c
@@ -140,14 +140,14 @@ int riotopen(struct tty_struct *tty, struct file *filp)
 
 	tty->driver_data = PortP;
 
-	PortP->gs.tty = tty;
-	PortP->gs.count++;
+	PortP->gs.port.tty = tty;
+	PortP->gs.port.count++;
 
 	rio_dprintk(RIO_DEBUG_TTY, "%d bytes in tx buffer\n", PortP->gs.xmit_cnt);
 
 	retval = gs_init_port(&PortP->gs);
 	if (retval) {
-		PortP->gs.count--;
+		PortP->gs.port.count--;
 		return -ENXIO;
 	}
 	/*
@@ -293,7 +293,7 @@ int riotopen(struct tty_struct *tty, struct file *filp)
 	 ** insert test for carrier here. -- ???
 	 ** I already see that test here. What's the deal? -- REW
 	 */
-	if ((PortP->gs.tty->termios->c_cflag & CLOCAL) ||
+	if ((PortP->gs.port.tty->termios->c_cflag & CLOCAL) ||
 			(PortP->ModemState & RIOC_MSVR1_CD)) {
 		rio_dprintk(RIO_DEBUG_TTY, "open(%d) Modem carr on\n", SysPort);
 		/*
@@ -301,16 +301,16 @@ int riotopen(struct tty_struct *tty, struct file *filp)
 		   wakeup((caddr_t) &tp->tm.c_canq);
 		 */
 		PortP->State |= RIO_CARR_ON;
-		wake_up_interruptible(&PortP->gs.open_wait);
+		wake_up_interruptible(&PortP->gs.port.open_wait);
 	} else {	/* no carrier - wait for DCD */
 			/*
-		   while (!(PortP->gs.tty->termios->c_state & CARR_ON) &&
+		   while (!(PortP->gs.port.tty->termios->c_state & CARR_ON) &&
 		   !(filp->f_flags & O_NONBLOCK) && !p->RIOHalted )
 		 */
 		while (!(PortP->State & RIO_CARR_ON) && !(filp->f_flags & O_NONBLOCK) && !p->RIOHalted) {
 				rio_dprintk(RIO_DEBUG_TTY, "open(%d) sleeping for carr on\n", SysPort);
 			/*
-			   PortP->gs.tty->termios->c_state |= WOPEN;
+			   PortP->gs.port.tty->termios->c_state |= WOPEN;
 			 */
 			PortP->State |= RIO_WOPEN;
 			rio_spin_unlock_irqrestore(&PortP->portSem, flags);
@@ -380,7 +380,7 @@ int riotclose(void *ptr)
 	/* PortP = p->RIOPortp[SysPort]; */
 	rio_dprintk(RIO_DEBUG_TTY, "Port is at address %p\n", PortP);
 	/* tp = PortP->TtyP; *//* Get tty */
-	tty = PortP->gs.tty;
+	tty = PortP->gs.port.tty;
 	rio_dprintk(RIO_DEBUG_TTY, "TTY is at address %p\n", tty);
 
 	if (PortP->gs.closing_wait)
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index b1a7a8cb65ea..b1239ee48b78 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -1,4 +1,3 @@
-
 /* sx.c -- driver for the Specialix SX series cards. 
  *
  *  This driver will also support the older SI, and XIO cards.
@@ -930,7 +929,7 @@ static int sx_set_real_termios(void *ptr)
 
 	func_enter2();
 
-	if (!port->gs.tty)
+	if (!port->gs.port.tty)
 		return 0;
 
 	/* What is this doing here? -- REW
@@ -941,19 +940,19 @@ static int sx_set_real_termios(void *ptr)
 
 	sx_set_baud(port);
 
-#define CFLAG port->gs.tty->termios->c_cflag
+#define CFLAG port->gs.port.tty->termios->c_cflag
 	sx_write_channel_byte(port, hi_mr1,
-			(C_PARENB(port->gs.tty) ? MR1_WITH : MR1_NONE) |
-			(C_PARODD(port->gs.tty) ? MR1_ODD : MR1_EVEN) |
-			(C_CRTSCTS(port->gs.tty) ? MR1_RTS_RXFLOW : 0) |
+			(C_PARENB(port->gs.port.tty) ? MR1_WITH : MR1_NONE) |
+			(C_PARODD(port->gs.port.tty) ? MR1_ODD : MR1_EVEN) |
+			(C_CRTSCTS(port->gs.port.tty) ? MR1_RTS_RXFLOW : 0) |
 			(((CFLAG & CSIZE) == CS8) ? MR1_8_BITS : 0) |
 			(((CFLAG & CSIZE) == CS7) ? MR1_7_BITS : 0) |
 			(((CFLAG & CSIZE) == CS6) ? MR1_6_BITS : 0) |
 			(((CFLAG & CSIZE) == CS5) ? MR1_5_BITS : 0));
 
 	sx_write_channel_byte(port, hi_mr2,
-			(C_CRTSCTS(port->gs.tty) ? MR2_CTS_TXFLOW : 0) |
-			(C_CSTOPB(port->gs.tty) ? MR2_2_STOP :
+			(C_CRTSCTS(port->gs.port.tty) ? MR2_CTS_TXFLOW : 0) |
+			(C_CSTOPB(port->gs.port.tty) ? MR2_2_STOP :
 			MR2_1_STOP));
 
 	switch (CFLAG & CSIZE) {
@@ -976,44 +975,44 @@ static int sx_set_real_termios(void *ptr)
 	}
 
 	sx_write_channel_byte(port, hi_prtcl,
-			(I_IXON(port->gs.tty) ? SP_TXEN : 0) |
-			(I_IXOFF(port->gs.tty) ? SP_RXEN : 0) |
-			(I_IXANY(port->gs.tty) ? SP_TANY : 0) | SP_DCEN);
+			(I_IXON(port->gs.port.tty) ? SP_TXEN : 0) |
+			(I_IXOFF(port->gs.port.tty) ? SP_RXEN : 0) |
+			(I_IXANY(port->gs.port.tty) ? SP_TANY : 0) | SP_DCEN);
 
 	sx_write_channel_byte(port, hi_break,
-			(I_IGNBRK(port->gs.tty) ? BR_IGN : 0 |
-			I_BRKINT(port->gs.tty) ? BR_INT : 0));
+			(I_IGNBRK(port->gs.port.tty) ? BR_IGN : 0 |
+			I_BRKINT(port->gs.port.tty) ? BR_INT : 0));
 
-	sx_write_channel_byte(port, hi_txon, START_CHAR(port->gs.tty));
-	sx_write_channel_byte(port, hi_rxon, START_CHAR(port->gs.tty));
-	sx_write_channel_byte(port, hi_txoff, STOP_CHAR(port->gs.tty));
-	sx_write_channel_byte(port, hi_rxoff, STOP_CHAR(port->gs.tty));
+	sx_write_channel_byte(port, hi_txon, START_CHAR(port->gs.port.tty));
+	sx_write_channel_byte(port, hi_rxon, START_CHAR(port->gs.port.tty));
+	sx_write_channel_byte(port, hi_txoff, STOP_CHAR(port->gs.port.tty));
+	sx_write_channel_byte(port, hi_rxoff, STOP_CHAR(port->gs.port.tty));
 
 	sx_reconfigure_port(port);
 
 	/* Tell line discipline whether we will do input cooking */
-	if (I_OTHER(port->gs.tty)) {
-		clear_bit(TTY_HW_COOK_IN, &port->gs.tty->flags);
+	if (I_OTHER(port->gs.port.tty)) {
+		clear_bit(TTY_HW_COOK_IN, &port->gs.port.tty->flags);
 	} else {
-		set_bit(TTY_HW_COOK_IN, &port->gs.tty->flags);
+		set_bit(TTY_HW_COOK_IN, &port->gs.port.tty->flags);
 	}
 	sx_dprintk(SX_DEBUG_TERMIOS, "iflags: %x(%d) ",
-			(unsigned int)port->gs.tty->termios->c_iflag,
-			I_OTHER(port->gs.tty));
+			(unsigned int)port->gs.port.tty->termios->c_iflag,
+			I_OTHER(port->gs.port.tty));
 
 /* Tell line discipline whether we will do output cooking.
  * If OPOST is set and no other output flags are set then we can do output
  * processing.  Even if only *one* other flag in the O_OTHER group is set
  * we do cooking in software.
  */
-	if (O_OPOST(port->gs.tty) && !O_OTHER(port->gs.tty)) {
-		set_bit(TTY_HW_COOK_OUT, &port->gs.tty->flags);
+	if (O_OPOST(port->gs.port.tty) && !O_OTHER(port->gs.port.tty)) {
+		set_bit(TTY_HW_COOK_OUT, &port->gs.port.tty->flags);
 	} else {
-		clear_bit(TTY_HW_COOK_OUT, &port->gs.tty->flags);
+		clear_bit(TTY_HW_COOK_OUT, &port->gs.port.tty->flags);
 	}
 	sx_dprintk(SX_DEBUG_TERMIOS, "oflags: %x(%d)\n",
-			(unsigned int)port->gs.tty->termios->c_oflag,
-			O_OTHER(port->gs.tty));
+			(unsigned int)port->gs.port.tty->termios->c_oflag,
+			O_OTHER(port->gs.port.tty));
 	/* port->c_dcd = sx_get_CD (port); */
 	func_exit();
 	return 0;
@@ -1102,8 +1101,8 @@ static void sx_transmit_chars(struct sx_port *port)
 		sx_disable_tx_interrupts(port);
 	}
 
-	if ((port->gs.xmit_cnt <= port->gs.wakeup_chars) && port->gs.tty) {
-		tty_wakeup(port->gs.tty);
+	if ((port->gs.xmit_cnt <= port->gs.wakeup_chars) && port->gs.port.tty) {
+		tty_wakeup(port->gs.port.tty);
 		sx_dprintk(SX_DEBUG_TRANSMIT, "Waking up.... ldisc (%d)....\n",
 				port->gs.wakeup_chars);
 	}
@@ -1126,7 +1125,7 @@ static inline void sx_receive_chars(struct sx_port *port)
 	unsigned char *rp;
 
 	func_enter2();
-	tty = port->gs.tty;
+	tty = port->gs.port.tty;
 	while (1) {
 		rx_op = sx_read_channel_byte(port, hi_rxopos);
 		c = (sx_read_channel_byte(port, hi_rxipos) - rx_op) & 0xff;
@@ -1211,12 +1210,12 @@ static inline void sx_check_modem_signals(struct sx_port *port)
 				/* DCD went UP */
 				if ((sx_read_channel_byte(port, hi_hstat) !=
 						HS_IDLE_CLOSED) &&
-						!(port->gs.tty->termios->
+						!(port->gs.port.tty->termios->
 							c_cflag & CLOCAL)) {
 					/* Are we blocking in open? */
 					sx_dprintk(SX_DEBUG_MODEMSIGNALS, "DCD "
 						"active, unblocking open\n");
-					wake_up_interruptible(&port->gs.
+					wake_up_interruptible(&port->gs.port.
 							open_wait);
 				} else {
 					sx_dprintk(SX_DEBUG_MODEMSIGNALS, "DCD "
@@ -1224,10 +1223,10 @@ static inline void sx_check_modem_signals(struct sx_port *port)
 				}
 			} else {
 				/* DCD went down! */
-				if (!(port->gs.tty->termios->c_cflag & CLOCAL)){
+				if (!(port->gs.port.tty->termios->c_cflag & CLOCAL)){
 					sx_dprintk(SX_DEBUG_MODEMSIGNALS, "DCD "
 						"dropped. hanging up....\n");
-					tty_hangup(port->gs.tty);
+					tty_hangup(port->gs.port.tty);
 				} else {
 					sx_dprintk(SX_DEBUG_MODEMSIGNALS, "DCD "
 						"dropped. ignoring.\n");
@@ -1325,7 +1324,7 @@ static irqreturn_t sx_interrupt(int irq, void *ptr)
 
 	for (i = 0; i < board->nports; i++) {
 		port = &board->ports[i];
-		if (port->gs.flags & GS_ACTIVE) {
+		if (port->gs.port.flags & GS_ACTIVE) {
 			if (sx_read_channel_byte(port, hi_state)) {
 				sx_dprintk(SX_DEBUG_INTERRUPTS, "Port %d: "
 						"modem signal change?... \n",i);
@@ -1334,7 +1333,7 @@ static irqreturn_t sx_interrupt(int irq, void *ptr)
 			if (port->gs.xmit_cnt) {
 				sx_transmit_chars(port);
 			}
-			if (!(port->gs.flags & SX_RX_THROTTLE)) {
+			if (!(port->gs.port.flags & SX_RX_THROTTLE)) {
 				sx_receive_chars(port);
 			}
 		}
@@ -1373,7 +1372,7 @@ static void sx_disable_tx_interrupts(void *ptr)
 	struct sx_port *port = ptr;
 	func_enter2();
 
-	port->gs.flags &= ~GS_TX_INTEN;
+	port->gs.port.flags &= ~GS_TX_INTEN;
 
 	func_exit();
 }
@@ -1394,7 +1393,7 @@ static void sx_enable_tx_interrupts(void *ptr)
 
 	/* XXX Must be "HIGH_WATER" for SI card according to doc. */
 	if (data_in_buffer < LOW_WATER)
-		port->gs.flags &= ~GS_TX_INTEN;
+		port->gs.port.flags &= ~GS_TX_INTEN;
 
 	func_exit();
 }
@@ -1442,8 +1441,8 @@ static void sx_shutdown_port(void *ptr)
 
 	func_enter();
 
-	port->gs.flags &= ~GS_ACTIVE;
-	if (port->gs.tty && (port->gs.tty->termios->c_cflag & HUPCL)) {
+	port->gs.port.flags &= ~GS_ACTIVE;
+	if (port->gs.port.tty && (port->gs.port.tty->termios->c_cflag & HUPCL)) {
 		sx_setsignals(port, 0, 0);
 		sx_reconfigure_port(port);
 	}
@@ -1485,8 +1484,8 @@ static int sx_open(struct tty_struct *tty, struct file *filp)
 	spin_lock_irqsave(&port->gs.driver_lock, flags);
 
 	tty->driver_data = port;
-	port->gs.tty = tty;
-	port->gs.count++;
+	port->gs.port.tty = tty;
+	port->gs.port.count++;
 	spin_unlock_irqrestore(&port->gs.driver_lock, flags);
 
 	sx_dprintk(SX_DEBUG_OPEN, "starting port\n");
@@ -1497,12 +1496,12 @@ static int sx_open(struct tty_struct *tty, struct file *filp)
 	retval = gs_init_port(&port->gs);
 	sx_dprintk(SX_DEBUG_OPEN, "done gs_init\n");
 	if (retval) {
-		port->gs.count--;
+		port->gs.port.count--;
 		return retval;
 	}
 
-	port->gs.flags |= GS_ACTIVE;
-	if (port->gs.count <= 1)
+	port->gs.port.flags |= GS_ACTIVE;
+	if (port->gs.port.count <= 1)
 		sx_setsignals(port, 1, 1);
 
 #if 0
@@ -1513,12 +1512,12 @@ static int sx_open(struct tty_struct *tty, struct file *filp)
 		my_hd_io(port->board->base + port->ch_base, sizeof(*port));
 #endif
 
-	if (port->gs.count <= 1) {
+	if (port->gs.port.count <= 1) {
 		if (sx_send_command(port, HS_LOPEN, -1, HS_IDLE_OPEN) != 1) {
 			printk(KERN_ERR "sx: Card didn't respond to LOPEN "
 					"command.\n");
 			spin_lock_irqsave(&port->gs.driver_lock, flags);
-			port->gs.count--;
+			port->gs.port.count--;
 			spin_unlock_irqrestore(&port->gs.driver_lock, flags);
 			return -EIO;
 		}
@@ -1526,11 +1525,11 @@ static int sx_open(struct tty_struct *tty, struct file *filp)
 
 	retval = gs_block_til_ready(port, filp);
 	sx_dprintk(SX_DEBUG_OPEN, "Block til ready returned %d. Count=%d\n",
-			retval, port->gs.count);
+			retval, port->gs.port.count);
 
 	if (retval) {
 /*
- * Don't lower gs.count here because sx_close() will be called later
+ * Don't lower gs.port.count here because sx_close() will be called later
  */
 
 		return retval;
@@ -1571,14 +1570,14 @@ static void sx_close(void *ptr)
 	}
 
 	sx_dprintk(SX_DEBUG_CLOSE, "waited %d jiffies for close. count=%d\n",
-			5 * HZ - to - 1, port->gs.count);
+			5 * HZ - to - 1, port->gs.port.count);
 
-	if (port->gs.count) {
+	if (port->gs.port.count) {
 		sx_dprintk(SX_DEBUG_CLOSE, "WARNING port count:%d\n",
-				port->gs.count);
+				port->gs.port.count);
 		/*printk("%s SETTING port count to zero: %p count: %d\n",
-				__func__, port, port->gs.count);
-		port->gs.count = 0;*/
+				__func__, port, port->gs.port.count);
+		port->gs.port.count = 0;*/
 	}
 
 	func_exit();
@@ -1939,7 +1938,7 @@ static void sx_throttle(struct tty_struct *tty)
 	 * control then throttle the port.
 	 */
 	if ((tty->termios->c_cflag & CRTSCTS) || (I_IXOFF(tty))) {
-		port->gs.flags |= SX_RX_THROTTLE;
+		port->gs.port.flags |= SX_RX_THROTTLE;
 	}
 	func_exit();
 }
@@ -1953,7 +1952,7 @@ static void sx_unthrottle(struct tty_struct *tty)
 	 * this port in case we disabled flow control while the port
 	 * was throttled
 	 */
-	port->gs.flags &= ~SX_RX_THROTTLE;
+	port->gs.port.flags &= ~SX_RX_THROTTLE;
 	func_exit();
 	return;
 }
@@ -2408,9 +2407,7 @@ static int sx_init_portstructs(int nboards, int nports)
 			/*
 			 * Initializing wait queue
 			 */
-			init_waitqueue_head(&port->gs.open_wait);
-			init_waitqueue_head(&port->gs.close_wait);
-
+			tty_port_init(&port->gs.port);
 			port++;
 		}
 	}
diff --git a/include/linux/generic_serial.h b/include/linux/generic_serial.h
index 110833666e37..4cc913939817 100644
--- a/include/linux/generic_serial.h
+++ b/include/linux/generic_serial.h
@@ -14,6 +14,7 @@
 
 #ifdef __KERNEL__
 #include <linux/mutex.h>
+#include <linux/tty.h>
 
 struct real_driver {
   void                    (*disable_tx_interrupts) (void *);
@@ -33,17 +34,12 @@ struct real_driver {
 
 struct gs_port {
   int                     magic;
+  struct tty_port	  port;
   unsigned char           *xmit_buf; 
   int                     xmit_head;
   int                     xmit_tail;
   int                     xmit_cnt;
   struct mutex            port_write_mutex;
-  int                     flags;
-  wait_queue_head_t       open_wait;
-  wait_queue_head_t       close_wait;
-  int                     count;
-  int                     blocked_open;
-  struct tty_struct       *tty;
   unsigned long           event;
   unsigned short          closing_wait;
   int                     close_delay;
-- 
cgit v1.2.3


From b02f5ad6a3ff5a1ee2a7b8ec6eee338de553b060 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:55:53 +0100
Subject: istallion: use tty_port

Switch istallion to use the new tty_port structure

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/istallion.c  | 116 +++++++++++++++++++++++-----------------------
 include/linux/istallion.h |   6 +--
 2 files changed, 59 insertions(+), 63 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 7c8b62f162bf..6ef1c565705c 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -735,8 +735,8 @@ static void stli_cleanup_ports(struct stlibrd *brdp)
 	for (j = 0; j < STL_MAXPORTS; j++) {
 		portp = brdp->ports[j];
 		if (portp != NULL) {
-			if (portp->tty != NULL)
-				tty_hangup(portp->tty);
+			if (portp->port.tty != NULL)
+				tty_hangup(portp->port.tty);
 			kfree(portp);
 		}
 	}
@@ -811,9 +811,9 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
  *	The sleep here does not need interrupt protection since the wakeup
  *	for it is done with the same context.
  */
-	if (portp->flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->close_wait);
-		if (portp->flags & ASYNC_HUP_NOTIFY)
+	if (portp->port.flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&portp->port.close_wait);
+		if (portp->port.flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		return -ERESTARTSYS;
 	}
@@ -824,7 +824,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
  *	requires several commands to the board we will need to wait for any
  *	other open that is already initializing the port.
  */
-	portp->tty = tty;
+	portp->port.tty = tty;
 	tty->driver_data = portp;
 	portp->refcount++;
 
@@ -833,10 +833,10 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
 	if (signal_pending(current))
 		return -ERESTARTSYS;
 
-	if ((portp->flags & ASYNC_INITIALIZED) == 0) {
+	if ((portp->port.flags & ASYNC_INITIALIZED) == 0) {
 		set_bit(ST_INITIALIZING, &portp->state);
 		if ((rc = stli_initopen(brdp, portp)) >= 0) {
-			portp->flags |= ASYNC_INITIALIZED;
+			portp->port.flags |= ASYNC_INITIALIZED;
 			clear_bit(TTY_IO_ERROR, &tty->flags);
 		}
 		clear_bit(ST_INITIALIZING, &portp->state);
@@ -851,9 +851,9 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
  *	The sleep here does not need interrupt protection since the wakeup
  *	for it is done with the same context.
  */
-	if (portp->flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->close_wait);
-		if (portp->flags & ASYNC_HUP_NOTIFY)
+	if (portp->port.flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&portp->port.close_wait);
+		if (portp->port.flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		return -ERESTARTSYS;
 	}
@@ -867,7 +867,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
 		if ((rc = stli_waitcarrier(brdp, portp, filp)) != 0)
 			return rc;
 	}
-	portp->flags |= ASYNC_NORMAL_ACTIVE;
+	portp->port.flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }
 
@@ -895,7 +895,7 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 		return;
 	}
 
-	portp->flags |= ASYNC_CLOSING;
+	portp->port.flags |= ASYNC_CLOSING;
 
 /*
  *	May want to wait for data to drain before closing. The BUSY flag
@@ -911,7 +911,7 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 	if (portp->closing_wait != ASYNC_CLOSING_WAIT_NONE)
 		tty_wait_until_sent(tty, portp->closing_wait);
 
-	portp->flags &= ~ASYNC_INITIALIZED;
+	portp->port.flags &= ~ASYNC_INITIALIZED;
 	brdp = stli_brds[portp->brdnr];
 	stli_rawclose(brdp, portp, 0, 0);
 	if (tty->termios->c_cflag & HUPCL) {
@@ -931,16 +931,16 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 	stli_flushbuffer(tty);
 
 	tty->closing = 0;
-	portp->tty = NULL;
+	portp->port.tty = NULL;
 
 	if (portp->openwaitcnt) {
 		if (portp->close_delay)
 			msleep_interruptible(jiffies_to_msecs(portp->close_delay));
-		wake_up_interruptible(&portp->open_wait);
+		wake_up_interruptible(&portp->port.open_wait);
 	}
 
-	portp->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&portp->close_wait);
+	portp->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
+	wake_up_interruptible(&portp->port.close_wait);
 }
 
 /*****************************************************************************/
@@ -970,7 +970,7 @@ static int stli_initopen(struct stlibrd *brdp, struct stliport *portp)
 	    sizeof(asynotify_t), 0)) < 0)
 		return rc;
 
-	tty = portp->tty;
+	tty = portp->port.tty;
 	if (tty == NULL)
 		return -ENODEV;
 	stli_mkasyport(portp, &aport, tty->termios);
@@ -1169,7 +1169,7 @@ static int stli_setport(struct stliport *portp)
 
 	if (portp == NULL)
 		return -ENODEV;
-	if (portp->tty == NULL)
+	if (portp->port.tty == NULL)
 		return -ENODEV;
 	if (portp->brdnr >= stli_nrbrds)
 		return -ENODEV;
@@ -1177,7 +1177,7 @@ static int stli_setport(struct stliport *portp)
 	if (brdp == NULL)
 		return -ENODEV;
 
-	stli_mkasyport(portp, &aport, portp->tty->termios);
+	stli_mkasyport(portp, &aport, portp->port.tty->termios);
 	return(stli_cmdwait(brdp, portp, A_SETPORT, &aport, sizeof(asyport_t), 0));
 }
 
@@ -1196,7 +1196,7 @@ static int stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct
 	rc = 0;
 	doclocal = 0;
 
-	if (portp->tty->termios->c_cflag & CLOCAL)
+	if (portp->port.tty->termios->c_cflag & CLOCAL)
 		doclocal++;
 
 	spin_lock_irqsave(&stli_lock, flags);
@@ -1211,14 +1211,14 @@ static int stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct
 		    &portp->asig, sizeof(asysigs_t), 0)) < 0)
 			break;
 		if (tty_hung_up_p(filp) ||
-		    ((portp->flags & ASYNC_INITIALIZED) == 0)) {
-			if (portp->flags & ASYNC_HUP_NOTIFY)
+		    ((portp->port.flags & ASYNC_INITIALIZED) == 0)) {
+			if (portp->port.flags & ASYNC_HUP_NOTIFY)
 				rc = -EBUSY;
 			else
 				rc = -ERESTARTSYS;
 			break;
 		}
-		if (((portp->flags & ASYNC_CLOSING) == 0) &&
+		if (((portp->port.flags & ASYNC_CLOSING) == 0) &&
 		    (doclocal || (portp->sigs & TIOCM_CD))) {
 			break;
 		}
@@ -1226,7 +1226,7 @@ static int stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct
 			rc = -ERESTARTSYS;
 			break;
 		}
-		interruptible_sleep_on(&portp->open_wait);
+		interruptible_sleep_on(&portp->port.open_wait);
 	}
 
 	spin_lock_irqsave(&stli_lock, flags);
@@ -1548,7 +1548,7 @@ static int stli_getserial(struct stliport *portp, struct serial_struct __user *s
 	sio.type = PORT_UNKNOWN;
 	sio.line = portp->portnr;
 	sio.irq = 0;
-	sio.flags = portp->flags;
+	sio.flags = portp->port.flags;
 	sio.baud_base = portp->baud_base;
 	sio.close_delay = portp->close_delay;
 	sio.closing_wait = portp->closing_wait;
@@ -1583,11 +1583,11 @@ static int stli_setserial(struct stliport *portp, struct serial_struct __user *s
 		if ((sio.baud_base != portp->baud_base) ||
 		    (sio.close_delay != portp->close_delay) ||
 		    ((sio.flags & ~ASYNC_USR_MASK) !=
-		    (portp->flags & ~ASYNC_USR_MASK)))
+		    (portp->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
 	} 
 
-	portp->flags = (portp->flags & ~ASYNC_USR_MASK) |
+	portp->port.flags = (portp->port.flags & ~ASYNC_USR_MASK) |
 		(sio.flags & ASYNC_USR_MASK);
 	portp->baud_base = sio.baud_base;
 	portp->close_delay = sio.close_delay;
@@ -1751,7 +1751,7 @@ static void stli_settermios(struct tty_struct *tty, struct ktermios *old)
 	if ((old->c_cflag & CRTSCTS) && ((tiosp->c_cflag & CRTSCTS) == 0))
 		tty->hw_stopped = 0;
 	if (((old->c_cflag & CLOCAL) == 0) && (tiosp->c_cflag & CLOCAL))
-		wake_up_interruptible(&portp->open_wait);
+		wake_up_interruptible(&portp->port.open_wait);
 }
 
 /*****************************************************************************/
@@ -1834,7 +1834,7 @@ static void stli_hangup(struct tty_struct *tty)
 	if (brdp == NULL)
 		return;
 
-	portp->flags &= ~ASYNC_INITIALIZED;
+	portp->port.flags &= ~ASYNC_INITIALIZED;
 
 	if (!test_bit(ST_CLOSING, &portp->state))
 		stli_rawclose(brdp, portp, 0, 0);
@@ -1855,12 +1855,12 @@ static void stli_hangup(struct tty_struct *tty)
 	clear_bit(ST_TXBUSY, &portp->state);
 	clear_bit(ST_RXSTOP, &portp->state);
 	set_bit(TTY_IO_ERROR, &tty->flags);
-	portp->tty = NULL;
-	portp->flags &= ~ASYNC_NORMAL_ACTIVE;
+	portp->port.tty = NULL;
+	portp->port.flags &= ~ASYNC_NORMAL_ACTIVE;
 	portp->refcount = 0;
 	spin_unlock_irqrestore(&stli_lock, flags);
 
-	wake_up_interruptible(&portp->open_wait);
+	wake_up_interruptible(&portp->port.open_wait);
 }
 
 /*****************************************************************************/
@@ -2188,7 +2188,7 @@ static void stli_read(struct stlibrd *brdp, struct stliport *portp)
 
 	if (test_bit(ST_RXSTOP, &portp->state))
 		return;
-	tty = portp->tty;
+	tty = portp->port.tty;
 	if (tty == NULL)
 		return;
 
@@ -2362,7 +2362,7 @@ static int stli_hostcmd(struct stlibrd *brdp, struct stliport *portp)
 	if (ap->notify) {
 		nt = ap->changed;
 		ap->notify = 0;
-		tty = portp->tty;
+		tty = portp->port.tty;
 
 		if (nt.signal & SG_DCD) {
 			oldsigs = portp->sigs;
@@ -2370,10 +2370,10 @@ static int stli_hostcmd(struct stlibrd *brdp, struct stliport *portp)
 			clear_bit(ST_GETSIGS, &portp->state);
 			if ((portp->sigs & TIOCM_CD) &&
 			    ((oldsigs & TIOCM_CD) == 0))
-				wake_up_interruptible(&portp->open_wait);
+				wake_up_interruptible(&portp->port.open_wait);
 			if ((oldsigs & TIOCM_CD) &&
 			    ((portp->sigs & TIOCM_CD) == 0)) {
-				if (portp->flags & ASYNC_CHECK_CD) {
+				if (portp->port.flags & ASYNC_CHECK_CD) {
 					if (tty)
 						tty_hangup(tty);
 				}
@@ -2392,7 +2392,7 @@ static int stli_hostcmd(struct stlibrd *brdp, struct stliport *portp)
 		if ((nt.data & DT_RXBREAK) && (portp->rxmarkmsk & BRKINT)) {
 			if (tty != NULL) {
 				tty_insert_flip_char(tty, 0, TTY_BREAK);
-				if (portp->flags & ASYNC_SAK) {
+				if (portp->port.flags & ASYNC_SAK) {
 					do_SAK(tty);
 					EBRDENABLE(brdp);
 				}
@@ -2542,17 +2542,17 @@ static void stli_mkasyport(struct stliport *portp, asyport_t *pp, struct ktermio
 /*
  *	Start of by setting the baud, char size, parity and stop bit info.
  */
-	pp->baudout = tty_get_baud_rate(portp->tty);
+	pp->baudout = tty_get_baud_rate(portp->port.tty);
 	if ((tiosp->c_cflag & CBAUD) == B38400) {
-		if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+		if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
 			pp->baudout = 57600;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
 			pp->baudout = 115200;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
 			pp->baudout = 230400;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
 			pp->baudout = 460800;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
 			pp->baudout = (portp->baud_base / portp->custom_divisor);
 	}
 	if (pp->baudout > STL_MAXBAUD)
@@ -2625,9 +2625,9 @@ static void stli_mkasyport(struct stliport *portp, asyport_t *pp, struct ktermio
  *	Set up clocal processing as required.
  */
 	if (tiosp->c_cflag & CLOCAL)
-		portp->flags &= ~ASYNC_CHECK_CD;
+		portp->port.flags &= ~ASYNC_CHECK_CD;
 	else
-		portp->flags |= ASYNC_CHECK_CD;
+		portp->port.flags |= ASYNC_CHECK_CD;
 
 /*
  *	Transfer any persistent flags into the asyport structure.
@@ -2703,8 +2703,8 @@ static int stli_initports(struct stlibrd *brdp)
 		portp->baud_base = STL_BAUDBASE;
 		portp->close_delay = STL_CLOSEDELAY;
 		portp->closing_wait = 30 * HZ;
-		init_waitqueue_head(&portp->open_wait);
-		init_waitqueue_head(&portp->close_wait);
+		init_waitqueue_head(&portp->port.open_wait);
+		init_waitqueue_head(&portp->port.close_wait);
 		init_waitqueue_head(&portp->raw_wait);
 		panelport++;
 		if (panelport >= brdp->panels[panelnr]) {
@@ -4246,18 +4246,18 @@ static int stli_portcmdstats(struct stliport *portp)
 	stli_comstats.panel = portp->panelnr;
 	stli_comstats.port = portp->portnr;
 	stli_comstats.state = portp->state;
-	stli_comstats.flags = portp->flags;
+	stli_comstats.flags = portp->port.flag;
 
 	spin_lock_irqsave(&brd_lock, flags);
-	if (portp->tty != NULL) {
-		if (portp->tty->driver_data == portp) {
-			stli_comstats.ttystate = portp->tty->flags;
+	if (portp->port.tty != NULL) {
+		if (portp->port.tty->driver_data == portp) {
+			stli_comstats.ttystate = portp->port.tty->flags;
 			stli_comstats.rxbuffered = -1;
-			if (portp->tty->termios != NULL) {
-				stli_comstats.cflags = portp->tty->termios->c_cflag;
-				stli_comstats.iflags = portp->tty->termios->c_iflag;
-				stli_comstats.oflags = portp->tty->termios->c_oflag;
-				stli_comstats.lflags = portp->tty->termios->c_lflag;
+			if (portp->port.tty->termios != NULL) {
+				stli_comstats.cflags = portp->port.tty->termios->c_cflag;
+				stli_comstats.iflags = portp->port.tty->termios->c_iflag;
+				stli_comstats.oflags = portp->port.tty->termios->c_oflag;
+				stli_comstats.lflags = portp->port.tty->termios->c_lflag;
 			}
 		}
 	}
diff --git a/include/linux/istallion.h b/include/linux/istallion.h
index 5a84fe944b74..0d1840723249 100644
--- a/include/linux/istallion.h
+++ b/include/linux/istallion.h
@@ -51,25 +51,21 @@
  */
 struct stliport {
 	unsigned long		magic;
+	struct tty_port		port;
 	unsigned int		portnr;
 	unsigned int		panelnr;
 	unsigned int		brdnr;
 	unsigned long		state;
 	unsigned int		devnr;
-	int			flags;
 	int			baud_base;
 	int			custom_divisor;
 	int			close_delay;
 	int			closing_wait;
-	int			refcount;
 	int			openwaitcnt;
 	int			rc;
 	int			argsize;
 	void			*argp;
 	unsigned int		rxmarkmsk;
-	struct tty_struct	*tty;
-	wait_queue_head_t	open_wait;
-	wait_queue_head_t	close_wait;
 	wait_queue_head_t	raw_wait;
 	struct asysigs		asig;
 	unsigned long		addr;
-- 
cgit v1.2.3


From f8ae47641611fcdf175ab8bbe89054731b16971d Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:56:37 +0100
Subject: stallion: use tty_port

Switch the stallion driver to use the tty_port structure

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/stallion.c  | 160 +++++++++++++++++++++++------------------------
 include/linux/stallion.h |   6 +-
 2 files changed, 81 insertions(+), 85 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index d17be10c5d21..0243efb0be95 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -613,17 +613,17 @@ static void stl_cd_change(struct stlport *portp)
 {
 	unsigned int oldsigs = portp->sigs;
 
-	if (!portp->tty)
+	if (!portp->port.tty)
 		return;
 
 	portp->sigs = stl_getsignals(portp);
 
 	if ((portp->sigs & TIOCM_CD) && ((oldsigs & TIOCM_CD) == 0))
-		wake_up_interruptible(&portp->open_wait);
+		wake_up_interruptible(&portp->port.open_wait);
 
 	if ((oldsigs & TIOCM_CD) && ((portp->sigs & TIOCM_CD) == 0))
-		if (portp->flags & ASYNC_CHECK_CD)
-			tty_hangup(portp->tty);
+		if (portp->port.flags & ASYNC_CHECK_CD)
+			tty_hangup(portp->port.tty);
 }
 
 /*
@@ -734,11 +734,11 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
  *	On the first open of the device setup the port hardware, and
  *	initialize the per port data structure.
  */
-	portp->tty = tty;
+	portp->port.tty = tty;
 	tty->driver_data = portp;
-	portp->refcount++;
+	portp->port.count++;
 
-	if ((portp->flags & ASYNC_INITIALIZED) == 0) {
+	if ((portp->port.flags & ASYNC_INITIALIZED) == 0) {
 		if (!portp->tx.buf) {
 			portp->tx.buf = kmalloc(STL_TXBUFSIZE, GFP_KERNEL);
 			if (!portp->tx.buf)
@@ -752,7 +752,7 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
 		stl_enablerxtx(portp, 1, 1);
 		stl_startrxtx(portp, 1, 0);
 		clear_bit(TTY_IO_ERROR, &tty->flags);
-		portp->flags |= ASYNC_INITIALIZED;
+		portp->port.flags |= ASYNC_INITIALIZED;
 	}
 
 /*
@@ -761,9 +761,9 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
  *	The sleep here does not need interrupt protection since the wakeup
  *	for it is done with the same context.
  */
-	if (portp->flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->close_wait);
-		if (portp->flags & ASYNC_HUP_NOTIFY)
+	if (portp->port.flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&portp->port.close_wait);
+		if (portp->port.flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		return -ERESTARTSYS;
 	}
@@ -777,7 +777,7 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
 		if ((rc = stl_waitcarrier(portp, filp)) != 0)
 			return rc;
 
-	portp->flags |= ASYNC_NORMAL_ACTIVE;
+	portp->port.flags |= ASYNC_NORMAL_ACTIVE;
 
 	return 0;
 }
@@ -801,25 +801,25 @@ static int stl_waitcarrier(struct stlport *portp, struct file *filp)
 
 	spin_lock_irqsave(&stallion_lock, flags);
 
-	if (portp->tty->termios->c_cflag & CLOCAL)
+	if (portp->port.tty->termios->c_cflag & CLOCAL)
 		doclocal++;
 
 	portp->openwaitcnt++;
 	if (! tty_hung_up_p(filp))
-		portp->refcount--;
+		portp->port.count--;
 
 	for (;;) {
 		/* Takes brd_lock internally */
 		stl_setsignals(portp, 1, 1);
 		if (tty_hung_up_p(filp) ||
-		    ((portp->flags & ASYNC_INITIALIZED) == 0)) {
-			if (portp->flags & ASYNC_HUP_NOTIFY)
+		    ((portp->port.flags & ASYNC_INITIALIZED) == 0)) {
+			if (portp->port.flags & ASYNC_HUP_NOTIFY)
 				rc = -EBUSY;
 			else
 				rc = -ERESTARTSYS;
 			break;
 		}
-		if (((portp->flags & ASYNC_CLOSING) == 0) &&
+		if (((portp->port.flags & ASYNC_CLOSING) == 0) &&
 		    (doclocal || (portp->sigs & TIOCM_CD)))
 			break;
 		if (signal_pending(current)) {
@@ -827,11 +827,11 @@ static int stl_waitcarrier(struct stlport *portp, struct file *filp)
 			break;
 		}
 		/* FIXME */
-		interruptible_sleep_on(&portp->open_wait);
+		interruptible_sleep_on(&portp->port.open_wait);
 	}
 
 	if (! tty_hung_up_p(filp))
-		portp->refcount++;
+		portp->port.count++;
 	portp->openwaitcnt--;
 	spin_unlock_irqrestore(&stallion_lock, flags);
 
@@ -904,15 +904,15 @@ static void stl_close(struct tty_struct *tty, struct file *filp)
 		spin_unlock_irqrestore(&stallion_lock, flags);
 		return;
 	}
-	if ((tty->count == 1) && (portp->refcount != 1))
-		portp->refcount = 1;
-	if (portp->refcount-- > 1) {
+	if ((tty->count == 1) && (portp->port.count != 1))
+		portp->port.count = 1;
+	if (portp->port.count-- > 1) {
 		spin_unlock_irqrestore(&stallion_lock, flags);
 		return;
 	}
 
-	portp->refcount = 0;
-	portp->flags |= ASYNC_CLOSING;
+	portp->port.count = 0;
+	portp->port.flags |= ASYNC_CLOSING;
 
 /*
  *	May want to wait for any data to drain before closing. The BUSY
@@ -930,7 +930,7 @@ static void stl_close(struct tty_struct *tty, struct file *filp)
 
 
 	spin_lock_irqsave(&stallion_lock, flags);
-	portp->flags &= ~ASYNC_INITIALIZED;
+	portp->port.flags &= ~ASYNC_INITIALIZED;
 	spin_unlock_irqrestore(&stallion_lock, flags);
 
 	stl_disableintrs(portp);
@@ -949,16 +949,16 @@ static void stl_close(struct tty_struct *tty, struct file *filp)
 	tty_ldisc_flush(tty);
 
 	tty->closing = 0;
-	portp->tty = NULL;
+	portp->port.tty = NULL;
 
 	if (portp->openwaitcnt) {
 		if (portp->close_delay)
 			msleep_interruptible(jiffies_to_msecs(portp->close_delay));
-		wake_up_interruptible(&portp->open_wait);
+		wake_up_interruptible(&portp->port.open_wait);
 	}
 
-	portp->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&portp->close_wait);
+	portp->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
+	wake_up_interruptible(&portp->port.close_wait);
 }
 
 /*****************************************************************************/
@@ -1153,7 +1153,7 @@ static int stl_getserial(struct stlport *portp, struct serial_struct __user *sp)
 	memset(&sio, 0, sizeof(struct serial_struct));
 	sio.line = portp->portnr;
 	sio.port = portp->ioaddr;
-	sio.flags = portp->flags;
+	sio.flags = portp->port.flags;
 	sio.baud_base = portp->baud_base;
 	sio.close_delay = portp->close_delay;
 	sio.closing_wait = portp->closing_wait;
@@ -1194,17 +1194,17 @@ static int stl_setserial(struct stlport *portp, struct serial_struct __user *sp)
 		if ((sio.baud_base != portp->baud_base) ||
 		    (sio.close_delay != portp->close_delay) ||
 		    ((sio.flags & ~ASYNC_USR_MASK) !=
-		    (portp->flags & ~ASYNC_USR_MASK)))
+		    (portp->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
 	} 
 
-	portp->flags = (portp->flags & ~ASYNC_USR_MASK) |
+	portp->port.flags = (portp->port.flags & ~ASYNC_USR_MASK) |
 		(sio.flags & ASYNC_USR_MASK);
 	portp->baud_base = sio.baud_base;
 	portp->close_delay = sio.close_delay;
 	portp->closing_wait = sio.closing_wait;
 	portp->custom_divisor = sio.custom_divisor;
-	stl_setport(portp, portp->tty->termios);
+	stl_setport(portp, portp->port.tty->termios);
 	return 0;
 }
 
@@ -1353,7 +1353,7 @@ static void stl_settermios(struct tty_struct *tty, struct ktermios *old)
 		stl_start(tty);
 	}
 	if (((old->c_cflag & CLOCAL) == 0) && (tiosp->c_cflag & CLOCAL))
-		wake_up_interruptible(&portp->open_wait);
+		wake_up_interruptible(&portp->port.open_wait);
 }
 
 /*****************************************************************************/
@@ -1438,7 +1438,7 @@ static void stl_hangup(struct tty_struct *tty)
 	if (portp == NULL)
 		return;
 
-	portp->flags &= ~ASYNC_INITIALIZED;
+	portp->port.flags &= ~ASYNC_INITIALIZED;
 	stl_disableintrs(portp);
 	if (tty->termios->c_cflag & HUPCL)
 		stl_setsignals(portp, 0, 0);
@@ -1452,10 +1452,10 @@ static void stl_hangup(struct tty_struct *tty)
 		portp->tx.head = NULL;
 		portp->tx.tail = NULL;
 	}
-	portp->tty = NULL;
-	portp->flags &= ~ASYNC_NORMAL_ACTIVE;
-	portp->refcount = 0;
-	wake_up_interruptible(&portp->open_wait);
+	portp->port.tty = NULL;
+	portp->port.flags &= ~ASYNC_NORMAL_ACTIVE;
+	portp->port.count = 0;
+	wake_up_interruptible(&portp->port.open_wait);
 }
 
 /*****************************************************************************/
@@ -1814,8 +1814,8 @@ static int __devinit stl_initports(struct stlbrd *brdp, struct stlpanel *panelp)
 		portp->baud_base = STL_BAUDBASE;
 		portp->close_delay = STL_CLOSEDELAY;
 		portp->closing_wait = 30 * HZ;
-		init_waitqueue_head(&portp->open_wait);
-		init_waitqueue_head(&portp->close_wait);
+		init_waitqueue_head(&portp->port.open_wait);
+		init_waitqueue_head(&portp->port.close_wait);
 		portp->stats.brd = portp->brdnr;
 		portp->stats.panel = portp->panelnr;
 		portp->stats.port = portp->portnr;
@@ -1840,8 +1840,8 @@ static void stl_cleanup_panels(struct stlbrd *brdp)
 			portp = panelp->ports[k];
 			if (portp == NULL)
 				continue;
-			if (portp->tty != NULL)
-				stl_hangup(portp->tty);
+			if (portp->port.tty != NULL)
+				stl_hangup(portp->port.tty);
 			kfree(portp->tx.buf);
 			kfree(portp);
 		}
@@ -2513,7 +2513,7 @@ static int stl_getportstats(struct stlport *portp, comstats_t __user *cp)
 	}
 
 	portp->stats.state = portp->istate;
-	portp->stats.flags = portp->flags;
+	portp->stats.flags = portp->port.flags;
 	portp->stats.hwid = portp->hwid;
 
 	portp->stats.ttystate = 0;
@@ -2524,16 +2524,16 @@ static int stl_getportstats(struct stlport *portp, comstats_t __user *cp)
 	portp->stats.rxbuffered = 0;
 
 	spin_lock_irqsave(&stallion_lock, flags);
-	if (portp->tty != NULL)
-		if (portp->tty->driver_data == portp) {
-			portp->stats.ttystate = portp->tty->flags;
+	if (portp->port.tty != NULL)
+		if (portp->port.tty->driver_data == portp) {
+			portp->stats.ttystate = portp->port.tty->flags;
 			/* No longer available as a statistic */
-			portp->stats.rxbuffered = 1; /*portp->tty->flip.count; */
-			if (portp->tty->termios != NULL) {
-				portp->stats.cflags = portp->tty->termios->c_cflag;
-				portp->stats.iflags = portp->tty->termios->c_iflag;
-				portp->stats.oflags = portp->tty->termios->c_oflag;
-				portp->stats.lflags = portp->tty->termios->c_lflag;
+			portp->stats.rxbuffered = 1; /*portp->port.tty->flip.count; */
+			if (portp->port.tty->termios != NULL) {
+				portp->stats.cflags = portp->port.tty->termios->c_cflag;
+				portp->stats.iflags = portp->port.tty->termios->c_iflag;
+				portp->stats.oflags = portp->port.tty->termios->c_oflag;
+				portp->stats.lflags = portp->port.tty->termios->c_lflag;
 			}
 		}
 	spin_unlock_irqrestore(&stallion_lock, flags);
@@ -2939,15 +2939,15 @@ static void stl_cd1400setport(struct stlport *portp, struct ktermios *tiosp)
 	}
 	baudrate = stl_baudrates[baudrate];
 	if ((tiosp->c_cflag & CBAUD) == B38400) {
-		if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+		if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
 			baudrate = 57600;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
 			baudrate = 115200;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
 			baudrate = 230400;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
 			baudrate = 460800;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
 			baudrate = (portp->baud_base / portp->custom_divisor);
 	}
 	if (baudrate > STL_CD1400MAXBAUD)
@@ -2969,9 +2969,9 @@ static void stl_cd1400setport(struct stlport *portp, struct ktermios *tiosp)
 		mcor1 |= MCOR1_DCD;
 		mcor2 |= MCOR2_DCD;
 		sreron |= SRER_MODEM;
-		portp->flags |= ASYNC_CHECK_CD;
+		portp->port.flags |= ASYNC_CHECK_CD;
 	} else
-		portp->flags &= ~ASYNC_CHECK_CD;
+		portp->port.flags &= ~ASYNC_CHECK_CD;
 
 /*
  *	Setup cd1400 enhanced modes if we can. In particular we want to
@@ -3242,7 +3242,7 @@ static void stl_cd1400flowctrl(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->tty;
+	tty = portp->port.tty;
 	if (tty == NULL)
 		return;
 
@@ -3304,7 +3304,7 @@ static void stl_cd1400sendflow(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->tty;
+	tty = portp->port.tty;
 	if (tty == NULL)
 		return;
 
@@ -3503,8 +3503,8 @@ static void stl_cd1400txisr(struct stlpanel *panelp, int ioaddr)
 	if ((len == 0) || ((len < STL_TXBUFLOW) &&
 	    (test_bit(ASYI_TXLOW, &portp->istate) == 0))) {
 		set_bit(ASYI_TXLOW, &portp->istate);
-		if (portp->tty)
-			tty_wakeup(portp->tty);
+		if (portp->port.tty)
+			tty_wakeup(portp->port.tty);
 	}
 
 	if (len == 0) {
@@ -3568,7 +3568,7 @@ static void stl_cd1400rxisr(struct stlpanel *panelp, int ioaddr)
 		return;
 	}
 	portp = panelp->ports[(ioack >> 3)];
-	tty = portp->tty;
+	tty = portp->port.tty;
 
 	if ((ioack & ACK_TYPMASK) == ACK_TYPRXGOOD) {
 		outb((RDCR + portp->uartaddr), ioaddr);
@@ -3613,7 +3613,7 @@ static void stl_cd1400rxisr(struct stlpanel *panelp, int ioaddr)
 			if (portp->rxmarkmsk & status) {
 				if (status & ST_BREAK) {
 					status = TTY_BREAK;
-					if (portp->flags & ASYNC_SAK) {
+					if (portp->port.flags & ASYNC_SAK) {
 						do_SAK(tty);
 						BRDENABLE(portp->brdnr, portp->pagenr);
 					}
@@ -3899,15 +3899,15 @@ static void stl_sc26198setport(struct stlport *portp, struct ktermios *tiosp)
 	}
 	baudrate = stl_baudrates[baudrate];
 	if ((tiosp->c_cflag & CBAUD) == B38400) {
-		if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+		if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
 			baudrate = 57600;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
 			baudrate = 115200;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
 			baudrate = 230400;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
 			baudrate = 460800;
-		else if ((portp->flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
+		else if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST)
 			baudrate = (portp->baud_base / portp->custom_divisor);
 	}
 	if (baudrate > STL_SC26198MAXBAUD)
@@ -3922,11 +3922,11 @@ static void stl_sc26198setport(struct stlport *portp, struct ktermios *tiosp)
  *	Check what form of modem signaling is required and set it up.
  */
 	if (tiosp->c_cflag & CLOCAL) {
-		portp->flags &= ~ASYNC_CHECK_CD;
+		portp->port.flags &= ~ASYNC_CHECK_CD;
 	} else {
 		iopr |= IOPR_DCDCOS;
 		imron |= IR_IOPORT;
-		portp->flags |= ASYNC_CHECK_CD;
+		portp->port.flags |= ASYNC_CHECK_CD;
 	}
 
 /*
@@ -4174,7 +4174,7 @@ static void stl_sc26198flowctrl(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->tty;
+	tty = portp->port.tty;
 	if (tty == NULL)
 		return;
 
@@ -4243,7 +4243,7 @@ static void stl_sc26198sendflow(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->tty;
+	tty = portp->port.tty;
 	if (tty == NULL)
 		return;
 
@@ -4421,8 +4421,8 @@ static void stl_sc26198txisr(struct stlport *portp)
 	if ((len == 0) || ((len < STL_TXBUFLOW) &&
 	    (test_bit(ASYI_TXLOW, &portp->istate) == 0))) {
 		set_bit(ASYI_TXLOW, &portp->istate);
-		if (portp->tty)
-			tty_wakeup(portp->tty);
+		if (portp->port.tty)
+			tty_wakeup(portp->port.tty);
 	}
 
 	if (len == 0) {
@@ -4475,7 +4475,7 @@ static void stl_sc26198rxisr(struct stlport *portp, unsigned int iack)
 
 	pr_debug("stl_sc26198rxisr(portp=%p,iack=%x)\n", portp, iack);
 
-	tty = portp->tty;
+	tty = portp->port.tty;
 	ioaddr = portp->ioaddr;
 	outb(GIBCR, (ioaddr + XP_ADDR));
 	len = inb(ioaddr + XP_DATA) + 1;
@@ -4527,7 +4527,7 @@ static void stl_sc26198rxbadch(struct stlport *portp, unsigned char status, char
 	struct tty_struct	*tty;
 	unsigned int		ioaddr;
 
-	tty = portp->tty;
+	tty = portp->port.tty;
 	ioaddr = portp->ioaddr;
 
 	if (status & SR_RXPARITY)
@@ -4544,7 +4544,7 @@ static void stl_sc26198rxbadch(struct stlport *portp, unsigned char status, char
 		if (portp->rxmarkmsk & status) {
 			if (status & SR_RXBREAK) {
 				status = TTY_BREAK;
-				if (portp->flags & ASYNC_SAK) {
+				if (portp->port.flags & ASYNC_SAK) {
 					do_SAK(tty);
 					BRDENABLE(portp->brdnr, portp->pagenr);
 				}
diff --git a/include/linux/stallion.h b/include/linux/stallion.h
index 0424d75a5aaa..336af33c6ea4 100644
--- a/include/linux/stallion.h
+++ b/include/linux/stallion.h
@@ -69,6 +69,7 @@ struct stlrq {
  */
 struct stlport {
 	unsigned long		magic;
+	struct tty_port		port;
 	unsigned int		portnr;
 	unsigned int		panelnr;
 	unsigned int		brdnr;
@@ -76,12 +77,10 @@ struct stlport {
 	int			uartaddr;
 	unsigned int		pagenr;
 	unsigned long		istate;
-	int			flags;
 	int			baud_base;
 	int			custom_divisor;
 	int			close_delay;
 	int			closing_wait;
-	int			refcount;
 	int			openwaitcnt;
 	int			brklen;
 	unsigned int		sigs;
@@ -92,9 +91,6 @@ struct stlport {
 	unsigned long		clk;
 	unsigned long		hwid;
 	void			*uartp;
-	struct tty_struct	*tty;
-	wait_queue_head_t	open_wait;
-	wait_queue_head_t	close_wait;
 	comstats_t		stats;
 	struct stlrq		tx;
 };
-- 
cgit v1.2.3


From 77451e53e0a509a98eda272567869cfe96431ba9 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:57:02 +0100
Subject: cyclades: use tty_port

Switch cyclades to use the new tty_port structure

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/cyclades.c  | 315 +++++++++++++++++++++++------------------------
 include/linux/cyclades.h |   7 +-
 2 files changed, 158 insertions(+), 164 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c
index a957dbcc5a46..0144b19d1036 100644
--- a/drivers/char/cyclades.c
+++ b/drivers/char/cyclades.c
@@ -762,7 +762,7 @@ static int cy_next_channel;	/* next minor available */
 /*
  * This is used to look up the divisor speeds and the timeouts
  * We're normally limited to 15 distinct baud rates.  The extra
- * are accessed via settings in info->flags.
+ * are accessed via settings in info->port.flags.
  *      0,     1,     2,     3,     4,     5,     6,     7,     8,     9,
  *     10,    11,    12,    13,    14,    15,    16,    17,    18,    19,
  *                                               HI            VHI
@@ -1003,7 +1003,7 @@ static void cyy_chip_rx(struct cyclades_card *cinfo, int chip,
 	cy_writeb(base_addr + (CyCAR << index), save_xir);
 
 	/* if there is nowhere to put the data, discard it */
-	if (info->tty == NULL) {
+	if (info->port.tty == NULL) {
 		if ((readb(base_addr + (CyRIVR << index)) & CyIVRMask) ==
 				CyIVRRxEx) {	/* exception */
 			data = readb(base_addr + (CyRDSR << index));
@@ -1015,7 +1015,7 @@ static void cyy_chip_rx(struct cyclades_card *cinfo, int chip,
 		goto end;
 	}
 	/* there is an open port for this data */
-	tty = info->tty;
+	tty = info->port.tty;
 	if ((readb(base_addr + (CyRIVR << index)) & CyIVRMask) ==
 			CyIVRRxEx) {	/* exception */
 		data = readb(base_addr + (CyRDSR << index));
@@ -1041,7 +1041,7 @@ static void cyy_chip_rx(struct cyclades_card *cinfo, int chip,
 						readb(base_addr + (CyRDSR <<
 							index)), TTY_BREAK);
 					info->icount.rx++;
-					if (info->flags & ASYNC_SAK)
+					if (info->port.flags & ASYNC_SAK)
 						do_SAK(tty);
 				} else if (data & CyFRAME) {
 					tty_insert_flip_char(tty,
@@ -1145,7 +1145,7 @@ static void cyy_chip_tx(struct cyclades_card *cinfo, unsigned int chip,
 		goto end;
 	}
 	info = &cinfo->ports[channel + chip * 4];
-	if (info->tty == NULL) {
+	if (info->port.tty == NULL) {
 		cy_writeb(base_addr + (CySRER << index),
 			  readb(base_addr + (CySRER << index)) & ~CyTxRdy);
 		goto end;
@@ -1190,13 +1190,13 @@ static void cyy_chip_tx(struct cyclades_card *cinfo, unsigned int chip,
 			}
 			goto done;
 		}
-		if (info->xmit_buf == NULL) {
+		if (info->port.xmit_buf == NULL) {
 			cy_writeb(base_addr + (CySRER << index),
 				readb(base_addr + (CySRER << index)) &
 					~CyTxRdy);
 			goto done;
 		}
-		if (info->tty->stopped || info->tty->hw_stopped) {
+		if (info->port.tty->stopped || info->port.tty->hw_stopped) {
 			cy_writeb(base_addr + (CySRER << index),
 				readb(base_addr + (CySRER << index)) &
 					~CyTxRdy);
@@ -1211,7 +1211,7 @@ static void cyy_chip_tx(struct cyclades_card *cinfo, unsigned int chip,
 		 * character. This is necessary because there may not be room
 		 * for the two chars needed to send a NULL.)
 		 */
-		outch = info->xmit_buf[info->xmit_tail];
+		outch = info->port.xmit_buf[info->xmit_tail];
 		if (outch) {
 			info->xmit_cnt--;
 			info->xmit_tail = (info->xmit_tail + 1) &
@@ -1232,7 +1232,7 @@ static void cyy_chip_tx(struct cyclades_card *cinfo, unsigned int chip,
 	}
 
 done:
-	tty_wakeup(info->tty);
+	tty_wakeup(info->port.tty);
 end:
 	/* end of service */
 	cy_writeb(base_addr + (CyTIR << index), save_xir & 0x3f);
@@ -1256,7 +1256,7 @@ static void cyy_chip_modem(struct cyclades_card *cinfo, int chip,
 	mdm_change = readb(base_addr + (CyMISR << index));
 	mdm_status = readb(base_addr + (CyMSVR1 << index));
 
-	if (!info->tty)
+	if (!info->port.tty)
 		goto end;
 
 	if (mdm_change & CyANY_DELTA) {
@@ -1273,29 +1273,29 @@ static void cyy_chip_modem(struct cyclades_card *cinfo, int chip,
 		wake_up_interruptible(&info->delta_msr_wait);
 	}
 
-	if ((mdm_change & CyDCD) && (info->flags & ASYNC_CHECK_CD)) {
+	if ((mdm_change & CyDCD) && (info->port.flags & ASYNC_CHECK_CD)) {
 		if (!(mdm_status & CyDCD)) {
-			tty_hangup(info->tty);
-			info->flags &= ~ASYNC_NORMAL_ACTIVE;
+			tty_hangup(info->port.tty);
+			info->port.flags &= ~ASYNC_NORMAL_ACTIVE;
 		}
-		wake_up_interruptible(&info->open_wait);
+		wake_up_interruptible(&info->port.open_wait);
 	}
-	if ((mdm_change & CyCTS) && (info->flags & ASYNC_CTS_FLOW)) {
-		if (info->tty->hw_stopped) {
+	if ((mdm_change & CyCTS) && (info->port.flags & ASYNC_CTS_FLOW)) {
+		if (info->port.tty->hw_stopped) {
 			if (mdm_status & CyCTS) {
 				/* cy_start isn't used
 				   because... !!! */
-				info->tty->hw_stopped = 0;
+				info->port.tty->hw_stopped = 0;
 				cy_writeb(base_addr + (CySRER << index),
 					readb(base_addr + (CySRER << index)) |
 						CyTxRdy);
-				tty_wakeup(info->tty);
+				tty_wakeup(info->port.tty);
 			}
 		} else {
 			if (!(mdm_status & CyCTS)) {
 				/* cy_stop isn't used
 				   because ... !!! */
-				info->tty->hw_stopped = 1;
+				info->port.tty->hw_stopped = 1;
 				cy_writeb(base_addr + (CySRER << index),
 					readb(base_addr + (CySRER << index)) &
 						~CyTxRdy);
@@ -1449,7 +1449,7 @@ static void cyz_handle_rx(struct cyclades_port *info,
 		struct BUF_CTRL __iomem *buf_ctrl)
 {
 	struct cyclades_card *cinfo = info->card;
-	struct tty_struct *tty = info->tty;
+	struct tty_struct *tty = info->port.tty;
 	unsigned int char_count;
 	int len;
 #ifdef BLOCKMOVE
@@ -1542,7 +1542,7 @@ static void cyz_handle_tx(struct cyclades_port *info,
 		struct BUF_CTRL __iomem *buf_ctrl)
 {
 	struct cyclades_card *cinfo = info->card;
-	struct tty_struct *tty = info->tty;
+	struct tty_struct *tty = info->port.tty;
 	u8 data;
 	unsigned int char_count;
 #ifdef BLOCKMOVE
@@ -1585,7 +1585,7 @@ static void cyz_handle_tx(struct cyclades_port *info,
 
 			memcpy_toio((char *)(cinfo->base_addr + tx_bufaddr +
 					tx_put),
-					&info->xmit_buf[info->xmit_tail],
+					&info->port.xmit_buf[info->xmit_tail],
 					small_count);
 
 			tx_put = (tx_put + small_count) & (tx_bufsize - 1);
@@ -1597,7 +1597,7 @@ static void cyz_handle_tx(struct cyclades_port *info,
 		}
 #else
 		while (info->xmit_cnt && char_count) {
-			data = info->xmit_buf[info->xmit_tail];
+			data = info->port.xmit_buf[info->xmit_tail];
 			info->xmit_cnt--;
 			info->xmit_tail = (info->xmit_tail + 1) &
 					(SERIAL_XMIT_SIZE - 1);
@@ -1642,7 +1642,7 @@ static void cyz_handle_cmd(struct cyclades_card *cinfo)
 		special_count = 0;
 		delta_count = 0;
 		info = &cinfo->ports[channel];
-		tty = info->tty;
+		tty = info->port.tty;
 		if (tty == NULL)
 			continue;
 
@@ -1668,15 +1668,15 @@ static void cyz_handle_cmd(struct cyclades_card *cinfo)
 		case C_CM_MDCD:
 			info->icount.dcd++;
 			delta_count++;
-			if (info->flags & ASYNC_CHECK_CD) {
+			if (info->port.flags & ASYNC_CHECK_CD) {
 				if ((fw_ver > 241 ? ((u_long) param) :
 						readl(&ch_ctrl->rs_status)) &
 						C_RS_DCD) {
-					wake_up_interruptible(&info->open_wait);
+					wake_up_interruptible(&info->port.open_wait);
 				} else {
-					tty_hangup(info->tty);
-					wake_up_interruptible(&info->open_wait);
-					info->flags &= ~ASYNC_NORMAL_ACTIVE;
+					tty_hangup(info->port.tty);
+					wake_up_interruptible(&info->port.open_wait);
+					info->port.flags &= ~ASYNC_NORMAL_ACTIVE;
 				}
 			}
 			break;
@@ -1814,7 +1814,7 @@ static void cyz_poll(unsigned long arg)
 
 		for (port = 0; port < cinfo->nports; port++) {
 			info = &cinfo->ports[port];
-			tty = info->tty;
+			tty = info->port.tty;
 			buf_ctrl = &(zfw_ctrl->buf_ctrl[port]);
 
 			if (!info->throttle)
@@ -1853,22 +1853,22 @@ static int startup(struct cyclades_port *info)
 
 	spin_lock_irqsave(&card->card_lock, flags);
 
-	if (info->flags & ASYNC_INITIALIZED) {
+	if (info->port.flags & ASYNC_INITIALIZED) {
 		free_page(page);
 		goto errout;
 	}
 
 	if (!info->type) {
-		if (info->tty)
-			set_bit(TTY_IO_ERROR, &info->tty->flags);
+		if (info->port.tty)
+			set_bit(TTY_IO_ERROR, &info->port.tty->flags);
 		free_page(page);
 		goto errout;
 	}
 
-	if (info->xmit_buf)
+	if (info->port.xmit_buf)
 		free_page(page);
 	else
-		info->xmit_buf = (unsigned char *)page;
+		info->port.xmit_buf = (unsigned char *)page;
 
 	spin_unlock_irqrestore(&card->card_lock, flags);
 
@@ -1909,10 +1909,10 @@ static int startup(struct cyclades_port *info)
 
 		cy_writeb(base_addr + (CySRER << index),
 			readb(base_addr + (CySRER << index)) | CyRxData);
-		info->flags |= ASYNC_INITIALIZED;
+		info->port.flags |= ASYNC_INITIALIZED;
 
-		if (info->tty)
-			clear_bit(TTY_IO_ERROR, &info->tty->flags);
+		if (info->port.tty)
+			clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
 		info->xmit_cnt = info->xmit_head = info->xmit_tail = 0;
 		info->breakon = info->breakoff = 0;
 		memset((char *)&info->idle_stats, 0, sizeof(info->idle_stats));
@@ -1994,9 +1994,9 @@ static int startup(struct cyclades_port *info)
 
 		/* enable send, recv, modem !!! */
 
-		info->flags |= ASYNC_INITIALIZED;
-		if (info->tty)
-			clear_bit(TTY_IO_ERROR, &info->tty->flags);
+		info->port.flags |= ASYNC_INITIALIZED;
+		if (info->port.tty)
+			clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
 		info->xmit_cnt = info->xmit_head = info->xmit_tail = 0;
 		info->breakon = info->breakoff = 0;
 		memset((char *)&info->idle_stats, 0, sizeof(info->idle_stats));
@@ -2065,7 +2065,7 @@ static void shutdown(struct cyclades_port *info)
 	void __iomem *base_addr;
 	int chip, channel, index;
 
-	if (!(info->flags & ASYNC_INITIALIZED))
+	if (!(info->port.flags & ASYNC_INITIALIZED))
 		return;
 
 	card = info->card;
@@ -2087,14 +2087,14 @@ static void shutdown(struct cyclades_port *info)
 		/* Clear delta_msr_wait queue to avoid mem leaks. */
 		wake_up_interruptible(&info->delta_msr_wait);
 
-		if (info->xmit_buf) {
+		if (info->port.xmit_buf) {
 			unsigned char *temp;
-			temp = info->xmit_buf;
-			info->xmit_buf = NULL;
+			temp = info->port.xmit_buf;
+			info->port.xmit_buf = NULL;
 			free_page((unsigned long)temp);
 		}
 		cy_writeb(base_addr + (CyCAR << index), (u_char) channel);
-		if (!info->tty || (info->tty->termios->c_cflag & HUPCL)) {
+		if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL)) {
 			cy_writeb(base_addr + (CyMSVR1 << index), ~CyRTS);
 			cy_writeb(base_addr + (CyMSVR2 << index), ~CyDTR);
 #ifdef CY_DEBUG_DTR
@@ -2108,9 +2108,9 @@ static void shutdown(struct cyclades_port *info)
 		/* it may be appropriate to clear _XMIT at
 		   some later date (after testing)!!! */
 
-		if (info->tty)
-			set_bit(TTY_IO_ERROR, &info->tty->flags);
-		info->flags &= ~ASYNC_INITIALIZED;
+		if (info->port.tty)
+			set_bit(TTY_IO_ERROR, &info->port.tty->flags);
+		info->port.flags &= ~ASYNC_INITIALIZED;
 		spin_unlock_irqrestore(&card->card_lock, flags);
 	} else {
 		struct FIRM_ID __iomem *firm_id;
@@ -2136,14 +2136,14 @@ static void shutdown(struct cyclades_port *info)
 
 		spin_lock_irqsave(&card->card_lock, flags);
 
-		if (info->xmit_buf) {
+		if (info->port.xmit_buf) {
 			unsigned char *temp;
-			temp = info->xmit_buf;
-			info->xmit_buf = NULL;
+			temp = info->port.xmit_buf;
+			info->port.xmit_buf = NULL;
 			free_page((unsigned long)temp);
 		}
 
-		if (!info->tty || (info->tty->termios->c_cflag & HUPCL)) {
+		if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL)) {
 			cy_writel(&ch_ctrl[channel].rs_control,
 				(__u32)(readl(&ch_ctrl[channel].rs_control) &
 					~(C_RS_RTS | C_RS_DTR)));
@@ -2158,9 +2158,9 @@ static void shutdown(struct cyclades_port *info)
 #endif
 		}
 
-		if (info->tty)
-			set_bit(TTY_IO_ERROR, &info->tty->flags);
-		info->flags &= ~ASYNC_INITIALIZED;
+		if (info->port.tty)
+			set_bit(TTY_IO_ERROR, &info->port.tty->flags);
+		info->port.flags &= ~ASYNC_INITIALIZED;
 
 		spin_unlock_irqrestore(&card->card_lock, flags);
 	}
@@ -2194,10 +2194,10 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
-	if (tty_hung_up_p(filp) || (info->flags & ASYNC_CLOSING)) {
-		wait_event_interruptible(info->close_wait,
-				!(info->flags & ASYNC_CLOSING));
-		return (info->flags & ASYNC_HUP_NOTIFY) ? -EAGAIN: -ERESTARTSYS;
+	if (tty_hung_up_p(filp) || (info->port.flags & ASYNC_CLOSING)) {
+		wait_event_interruptible(info->port.close_wait,
+				!(info->port.flags & ASYNC_CLOSING));
+		return (info->port.flags & ASYNC_HUP_NOTIFY) ? -EAGAIN: -ERESTARTSYS;
 	}
 
 	/*
@@ -2206,32 +2206,32 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 					(tty->flags & (1 << TTY_IO_ERROR))) {
-		info->flags |= ASYNC_NORMAL_ACTIVE;
+		info->port.flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->count is dropped by one, so that
+	 * this loop, info->port.count is dropped by one, so that
 	 * cy_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&info->open_wait, &wait);
+	add_wait_queue(&info->port.open_wait, &wait);
 #ifdef CY_DEBUG_OPEN
 	printk(KERN_DEBUG "cyc block_til_ready before block: ttyC%d, "
-		"count = %d\n", info->line, info->count);
+		"count = %d\n", info->line, info->port.count);
 #endif
 	spin_lock_irqsave(&cinfo->card_lock, flags);
 	if (!tty_hung_up_p(filp))
-		info->count--;
+		info->port.count--;
 	spin_unlock_irqrestore(&cinfo->card_lock, flags);
 #ifdef CY_DEBUG_COUNT
 	printk(KERN_DEBUG "cyc block_til_ready: (%d): decrementing count to "
-		"%d\n", current->pid, info->count);
+		"%d\n", current->pid, info->port.count);
 #endif
-	info->blocked_open++;
+	info->port.blocked_open++;
 
 	if (!IS_CYC_Z(*cinfo)) {
 		chip = channel >> 2;
@@ -2260,8 +2260,8 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 
 			set_current_state(TASK_INTERRUPTIBLE);
 			if (tty_hung_up_p(filp) ||
-					!(info->flags & ASYNC_INITIALIZED)) {
-				retval = ((info->flags & ASYNC_HUP_NOTIFY) ?
+					!(info->port.flags & ASYNC_INITIALIZED)) {
+				retval = ((info->port.flags & ASYNC_HUP_NOTIFY) ?
 					  -EAGAIN : -ERESTARTSYS);
 				break;
 			}
@@ -2269,7 +2269,7 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 			spin_lock_irqsave(&cinfo->card_lock, flags);
 			cy_writeb(base_addr + (CyCAR << index),
 				  (u_char) channel);
-			if (!(info->flags & ASYNC_CLOSING) && (C_CLOCAL(tty) ||
+			if (!(info->port.flags & ASYNC_CLOSING) && (C_CLOCAL(tty) ||
 					(readb(base_addr +
 						(CyMSVR1 << index)) & CyDCD))) {
 				spin_unlock_irqrestore(&cinfo->card_lock, flags);
@@ -2284,7 +2284,7 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 #ifdef CY_DEBUG_OPEN
 			printk(KERN_DEBUG "cyc block_til_ready blocking: "
 				"ttyC%d, count = %d\n",
-				info->line, info->count);
+				info->line, info->port.count);
 #endif
 			schedule();
 		}
@@ -2298,7 +2298,7 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 		firm_id = base_addr + ID_ADDRESS;
 		if (!ISZLOADED(*cinfo)) {
 			__set_current_state(TASK_RUNNING);
-			remove_wait_queue(&info->open_wait, &wait);
+			remove_wait_queue(&info->port.open_wait, &wait);
 			return -EINVAL;
 		}
 
@@ -2327,12 +2327,12 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 
 			set_current_state(TASK_INTERRUPTIBLE);
 			if (tty_hung_up_p(filp) ||
-					!(info->flags & ASYNC_INITIALIZED)) {
-				retval = ((info->flags & ASYNC_HUP_NOTIFY) ?
+					!(info->port.flags & ASYNC_INITIALIZED)) {
+				retval = ((info->port.flags & ASYNC_HUP_NOTIFY) ?
 					  -EAGAIN : -ERESTARTSYS);
 				break;
 			}
-			if (!(info->flags & ASYNC_CLOSING) && (C_CLOCAL(tty) ||
+			if (!(info->port.flags & ASYNC_CLOSING) && (C_CLOCAL(tty) ||
 					(readl(&ch_ctrl[channel].rs_status) &
 						C_RS_DCD))) {
 				break;
@@ -2344,28 +2344,28 @@ block_til_ready(struct tty_struct *tty, struct file *filp,
 #ifdef CY_DEBUG_OPEN
 			printk(KERN_DEBUG "cyc block_til_ready blocking: "
 				"ttyC%d, count = %d\n",
-				info->line, info->count);
+				info->line, info->port.count);
 #endif
 			schedule();
 		}
 	}
 	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->open_wait, &wait);
+	remove_wait_queue(&info->port.open_wait, &wait);
 	if (!tty_hung_up_p(filp)) {
-		info->count++;
+		info->port.count++;
 #ifdef CY_DEBUG_COUNT
 		printk(KERN_DEBUG "cyc:block_til_ready (%d): incrementing "
-			"count to %d\n", current->pid, info->count);
+			"count to %d\n", current->pid, info->port.count);
 #endif
 	}
-	info->blocked_open--;
+	info->port.blocked_open--;
 #ifdef CY_DEBUG_OPEN
 	printk(KERN_DEBUG "cyc:block_til_ready after blocking: ttyC%d, "
-		"count = %d\n", info->line, info->count);
+		"count = %d\n", info->line, info->port.count);
 #endif
 	if (retval)
 		return retval;
-	info->flags |= ASYNC_NORMAL_ACTIVE;
+	info->port.flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }				/* block_til_ready */
 
@@ -2456,27 +2456,27 @@ static int cy_open(struct tty_struct *tty, struct file *filp)
 	printk(KERN_DEBUG "cyc:cy_open ttyC%d\n", info->line);
 #endif
 	tty->driver_data = info;
-	info->tty = tty;
+	info->port.tty = tty;
 	if (serial_paranoia_check(info, tty->name, "cy_open"))
 		return -ENODEV;
 
 #ifdef CY_DEBUG_OPEN
 	printk(KERN_DEBUG "cyc:cy_open ttyC%d, count = %d\n", info->line,
-			info->count);
+			info->port.count);
 #endif
-	info->count++;
+	info->port.count++;
 #ifdef CY_DEBUG_COUNT
 	printk(KERN_DEBUG "cyc:cy_open (%d): incrementing count to %d\n",
-		current->pid, info->count);
+		current->pid, info->port.count);
 #endif
 
 	/*
 	 * If the port is the middle of closing, bail out now
 	 */
-	if (tty_hung_up_p(filp) || (info->flags & ASYNC_CLOSING)) {
-		wait_event_interruptible(info->close_wait,
-				!(info->flags & ASYNC_CLOSING));
-		return (info->flags & ASYNC_HUP_NOTIFY) ? -EAGAIN: -ERESTARTSYS;
+	if (tty_hung_up_p(filp) || (info->port.flags & ASYNC_CLOSING)) {
+		wait_event_interruptible(info->port.close_wait,
+				!(info->port.flags & ASYNC_CLOSING));
+		return (info->port.flags & ASYNC_HUP_NOTIFY) ? -EAGAIN: -ERESTARTSYS;
 	}
 
 	/*
@@ -2641,9 +2641,9 @@ static void cy_close(struct tty_struct *tty, struct file *filp)
 	}
 #ifdef CY_DEBUG_OPEN
 	printk(KERN_DEBUG "cyc:cy_close ttyC%d, count = %d\n", info->line,
-		info->count);
+		info->port.count);
 #endif
-	if ((tty->count == 1) && (info->count != 1)) {
+	if ((tty->count == 1) && (info->port.count != 1)) {
 		/*
 		 * Uh, oh.  tty->count is 1, which means that the tty
 		 * structure will be freed.  Info->count should always
@@ -2652,24 +2652,24 @@ static void cy_close(struct tty_struct *tty, struct file *filp)
 		 * serial port won't be shutdown.
 		 */
 		printk(KERN_ERR "cyc:cy_close: bad serial port count; "
-			"tty->count is 1, info->count is %d\n", info->count);
-		info->count = 1;
+			"tty->count is 1, info->port.count is %d\n", info->port.count);
+		info->port.count = 1;
 	}
 #ifdef CY_DEBUG_COUNT
 	printk(KERN_DEBUG  "cyc:cy_close at (%d): decrementing count to %d\n",
-		current->pid, info->count - 1);
+		current->pid, info->port.count - 1);
 #endif
-	if (--info->count < 0) {
+	if (--info->port.count < 0) {
 #ifdef CY_DEBUG_COUNT
 		printk(KERN_DEBUG "cyc:cyc_close setting count to 0\n");
 #endif
-		info->count = 0;
+		info->port.count = 0;
 	}
-	if (info->count) {
+	if (info->port.count) {
 		spin_unlock_irqrestore(&card->card_lock, flags);
 		return;
 	}
-	info->flags |= ASYNC_CLOSING;
+	info->port.flags |= ASYNC_CLOSING;
 
 	/*
 	 * Now we wait for the transmit buffer to clear; and we notify
@@ -2692,7 +2692,7 @@ static void cy_close(struct tty_struct *tty, struct file *filp)
 		cy_writeb(base_addr + (CyCAR << index), (u_char) channel);
 		cy_writeb(base_addr + (CySRER << index),
 			  readb(base_addr + (CySRER << index)) & ~CyRxData);
-		if (info->flags & ASYNC_INITIALIZED) {
+		if (info->port.flags & ASYNC_INITIALIZED) {
 			/* Waiting for on-board buffers to be empty before
 			   closing the port */
 			spin_unlock_irqrestore(&card->card_lock, flags);
@@ -2731,18 +2731,18 @@ static void cy_close(struct tty_struct *tty, struct file *filp)
 	spin_lock_irqsave(&card->card_lock, flags);
 
 	tty->closing = 0;
-	info->tty = NULL;
-	if (info->blocked_open) {
+	info->port.tty = NULL;
+	if (info->port.blocked_open) {
 		spin_unlock_irqrestore(&card->card_lock, flags);
 		if (info->close_delay) {
 			msleep_interruptible(jiffies_to_msecs
 						(info->close_delay));
 		}
-		wake_up_interruptible(&info->open_wait);
+		wake_up_interruptible(&info->port.open_wait);
 		spin_lock_irqsave(&card->card_lock, flags);
 	}
-	info->flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
-	wake_up_interruptible(&info->close_wait);
+	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
+	wake_up_interruptible(&info->port.close_wait);
 
 #ifdef CY_DEBUG_OTHER
 	printk(KERN_DEBUG "cyc:cy_close done\n");
@@ -2777,7 +2777,7 @@ static int cy_write(struct tty_struct *tty, const unsigned char *buf, int count)
 	if (serial_paranoia_check(info, tty->name, "cy_write"))
 		return 0;
 
-	if (!info->xmit_buf)
+	if (!info->port.xmit_buf)
 		return 0;
 
 	spin_lock_irqsave(&info->card->card_lock, flags);
@@ -2788,7 +2788,7 @@ static int cy_write(struct tty_struct *tty, const unsigned char *buf, int count)
 		if (c <= 0)
 			break;
 
-		memcpy(info->xmit_buf + info->xmit_head, buf, c);
+		memcpy(info->port.xmit_buf + info->xmit_head, buf, c);
 		info->xmit_head = (info->xmit_head + c) &
 			(SERIAL_XMIT_SIZE - 1);
 		info->xmit_cnt += c;
@@ -2826,7 +2826,7 @@ static int cy_put_char(struct tty_struct *tty, unsigned char ch)
 	if (serial_paranoia_check(info, tty->name, "cy_put_char"))
 		return 0;
 
-	if (!info->xmit_buf)
+	if (!info->port.xmit_buf)
 		return 0;
 
 	spin_lock_irqsave(&info->card->card_lock, flags);
@@ -2835,7 +2835,7 @@ static int cy_put_char(struct tty_struct *tty, unsigned char ch)
 		return 0;
 	}
 
-	info->xmit_buf[info->xmit_head++] = ch;
+	info->port.xmit_buf[info->xmit_head++] = ch;
 	info->xmit_head &= SERIAL_XMIT_SIZE - 1;
 	info->xmit_cnt++;
 	info->idle_stats.xmit_bytes++;
@@ -2860,7 +2860,7 @@ static void cy_flush_chars(struct tty_struct *tty)
 		return;
 
 	if (info->xmit_cnt <= 0 || tty->stopped || tty->hw_stopped ||
-			!info->xmit_buf)
+			!info->port.xmit_buf)
 		return;
 
 	start_xmit(info);
@@ -2988,27 +2988,27 @@ static void set_line_char(struct cyclades_port *info)
 	int baud, baud_rate = 0;
 	int i;
 
-	if (!info->tty || !info->tty->termios)
+	if (!info->port.tty || !info->port.tty->termios)
 		return;
 
 	if (info->line == -1)
 		return;
 
-	cflag = info->tty->termios->c_cflag;
-	iflag = info->tty->termios->c_iflag;
+	cflag = info->port.tty->termios->c_cflag;
+	iflag = info->port.tty->termios->c_iflag;
 
 	/*
 	 * Set up the tty->alt_speed kludge
 	 */
-	if (info->tty) {
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
-			info->tty->alt_speed = 57600;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
-			info->tty->alt_speed = 115200;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
-			info->tty->alt_speed = 230400;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
-			info->tty->alt_speed = 460800;
+	if (info->port.tty) {
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+			info->port.tty->alt_speed = 57600;
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+			info->port.tty->alt_speed = 115200;
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
+			info->port.tty->alt_speed = 230400;
+		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
+			info->port.tty->alt_speed = 460800;
 	}
 
 	card = info->card;
@@ -3020,8 +3020,8 @@ static void set_line_char(struct cyclades_port *info)
 		index = card->bus_index;
 
 		/* baud rate */
-		baud = tty_get_baud_rate(info->tty);
-		if (baud == 38400 && (info->flags & ASYNC_SPD_MASK) ==
+		baud = tty_get_baud_rate(info->port.tty);
+		if (baud == 38400 && (info->port.flags & ASYNC_SPD_MASK) ==
 				ASYNC_SPD_CUST) {
 			if (info->custom_divisor)
 				baud_rate = info->baud / info->custom_divisor;
@@ -3038,7 +3038,7 @@ static void set_line_char(struct cyclades_port *info)
 		if (i == 20)
 			i = 19;	/* CD1400_MAX_SPEED */
 
-		if (baud == 38400 && (info->flags & ASYNC_SPD_MASK) ==
+		if (baud == 38400 && (info->port.flags & ASYNC_SPD_MASK) ==
 				ASYNC_SPD_CUST) {
 			cyy_baud_calc(info, baud_rate);
 		} else {
@@ -3059,7 +3059,7 @@ static void set_line_char(struct cyclades_port *info)
 			/* get it right for 134.5 baud */
 			info->timeout = (info->xmit_fifo_size * HZ * 30 / 269) +
 					2;
-		} else if (baud == 38400 && (info->flags & ASYNC_SPD_MASK) ==
+		} else if (baud == 38400 && (info->port.flags & ASYNC_SPD_MASK) ==
 				ASYNC_SPD_CUST) {
 			info->timeout = (info->xmit_fifo_size * HZ * 15 /
 					baud_rate) + 2;
@@ -3108,16 +3108,16 @@ static void set_line_char(struct cyclades_port *info)
 
 		/* CTS flow control flag */
 		if (cflag & CRTSCTS) {
-			info->flags |= ASYNC_CTS_FLOW;
+			info->port.flags |= ASYNC_CTS_FLOW;
 			info->cor2 |= CyCtsAE;
 		} else {
-			info->flags &= ~ASYNC_CTS_FLOW;
+			info->port.flags &= ~ASYNC_CTS_FLOW;
 			info->cor2 &= ~CyCtsAE;
 		}
 		if (cflag & CLOCAL)
-			info->flags &= ~ASYNC_CHECK_CD;
+			info->port.flags &= ~ASYNC_CHECK_CD;
 		else
-			info->flags |= ASYNC_CHECK_CD;
+			info->port.flags |= ASYNC_CHECK_CD;
 
 	 /***********************************************
 	    The hardware option, CyRtsAO, presents RTS when
@@ -3146,8 +3146,8 @@ static void set_line_char(struct cyclades_port *info)
 		/* set line characteristics  according configuration */
 
 		cy_writeb(base_addr + (CySCHR1 << index),
-			  START_CHAR(info->tty));
-		cy_writeb(base_addr + (CySCHR2 << index), STOP_CHAR(info->tty));
+			  START_CHAR(info->port.tty));
+		cy_writeb(base_addr + (CySCHR2 << index), STOP_CHAR(info->port.tty));
 		cy_writeb(base_addr + (CyCOR1 << index), info->cor1);
 		cy_writeb(base_addr + (CyCOR2 << index), info->cor2);
 		cy_writeb(base_addr + (CyCOR3 << index), info->cor3);
@@ -3163,7 +3163,7 @@ static void set_line_char(struct cyclades_port *info)
 			(info->default_timeout ? info->default_timeout : 0x02));
 		/* 10ms rx timeout */
 
-		if (C_CLOCAL(info->tty)) {
+		if (C_CLOCAL(info->port.tty)) {
 			/* without modem intr */
 			cy_writeb(base_addr + (CySRER << index),
 				readb(base_addr + (CySRER << index)) | CyMdmCh);
@@ -3226,8 +3226,8 @@ static void set_line_char(struct cyclades_port *info)
 #endif
 		}
 
-		if (info->tty)
-			clear_bit(TTY_IO_ERROR, &info->tty->flags);
+		if (info->port.tty)
+			clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
 		spin_unlock_irqrestore(&card->card_lock, flags);
 
 	} else {
@@ -3250,8 +3250,8 @@ static void set_line_char(struct cyclades_port *info)
 		buf_ctrl = &zfw_ctrl->buf_ctrl[channel];
 
 		/* baud rate */
-		baud = tty_get_baud_rate(info->tty);
-		if (baud == 38400 && (info->flags & ASYNC_SPD_MASK) ==
+		baud = tty_get_baud_rate(info->port.tty);
+		if (baud == 38400 && (info->port.flags & ASYNC_SPD_MASK) ==
 				ASYNC_SPD_CUST) {
 			if (info->custom_divisor)
 				baud_rate = info->baud / info->custom_divisor;
@@ -3266,7 +3266,7 @@ static void set_line_char(struct cyclades_port *info)
 			/* get it right for 134.5 baud */
 			info->timeout = (info->xmit_fifo_size * HZ * 30 / 269) +
 					2;
-		} else if (baud == 38400 && (info->flags & ASYNC_SPD_MASK) ==
+		} else if (baud == 38400 && (info->port.flags & ASYNC_SPD_MASK) ==
 				ASYNC_SPD_CUST) {
 			info->timeout = (info->xmit_fifo_size * HZ * 15 /
 					baud_rate) + 2;
@@ -3318,7 +3318,7 @@ static void set_line_char(struct cyclades_port *info)
 		}
 		/* As the HW flow control is done in firmware, the driver
 		   doesn't need to care about it */
-		info->flags &= ~ASYNC_CTS_FLOW;
+		info->port.flags &= ~ASYNC_CTS_FLOW;
 
 		/* XON/XOFF/XANY flow control flags */
 		sw_flow = 0;
@@ -3337,9 +3337,9 @@ static void set_line_char(struct cyclades_port *info)
 
 		/* CD sensitivity */
 		if (cflag & CLOCAL)
-			info->flags &= ~ASYNC_CHECK_CD;
+			info->port.flags &= ~ASYNC_CHECK_CD;
 		else
-			info->flags |= ASYNC_CHECK_CD;
+			info->port.flags |= ASYNC_CHECK_CD;
 
 		if (baud == 0) {	/* baud rate is zero, turn off line */
 			cy_writel(&ch_ctrl->rs_control,
@@ -3361,8 +3361,8 @@ static void set_line_char(struct cyclades_port *info)
 				"was %x\n", info->line, retval);
 		}
 
-		if (info->tty)
-			clear_bit(TTY_IO_ERROR, &info->tty->flags);
+		if (info->port.tty)
+			clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
 	}
 }				/* set_line_char */
 
@@ -3381,7 +3381,7 @@ get_serial_info(struct cyclades_port *info,
 	tmp.port = (info->card - cy_card) * 0x100 + info->line -
 		cinfo->first_line;
 	tmp.irq = cinfo->irq;
-	tmp.flags = info->flags;
+	tmp.flags = info->port.flags;
 	tmp.close_delay = info->close_delay;
 	tmp.closing_wait = info->closing_wait;
 	tmp.baud_base = info->baud;
@@ -3406,9 +3406,9 @@ set_serial_info(struct cyclades_port *info,
 				new_serial.baud_base != info->baud ||
 				(new_serial.flags & ASYNC_FLAGS &
 					~ASYNC_USR_MASK) !=
-				(info->flags & ASYNC_FLAGS & ~ASYNC_USR_MASK))
+				(info->port.flags & ASYNC_FLAGS & ~ASYNC_USR_MASK))
 			return -EPERM;
-		info->flags = (info->flags & ~ASYNC_USR_MASK) |
+		info->port.flags = (info->port.flags & ~ASYNC_USR_MASK) |
 				(new_serial.flags & ASYNC_USR_MASK);
 		info->baud = new_serial.baud_base;
 		info->custom_divisor = new_serial.custom_divisor;
@@ -3422,13 +3422,13 @@ set_serial_info(struct cyclades_port *info,
 
 	info->baud = new_serial.baud_base;
 	info->custom_divisor = new_serial.custom_divisor;
-	info->flags = (info->flags & ~ASYNC_FLAGS) |
+	info->port.flags = (info->port.flags & ~ASYNC_FLAGS) |
 			(new_serial.flags & ASYNC_FLAGS);
 	info->close_delay = new_serial.close_delay * HZ / 100;
 	info->closing_wait = new_serial.closing_wait * HZ / 100;
 
 check_and_exit:
-	if (info->flags & ASYNC_INITIALIZED) {
+	if (info->port.flags & ASYNC_INITIALIZED) {
 		set_line_char(info);
 		return 0;
 	} else {
@@ -4097,7 +4097,7 @@ static void cy_set_termios(struct tty_struct *tty, struct ktermios *old_termios)
 	 */
 	if (!(old_termios->c_cflag & CLOCAL) &&
 	    (tty->termios->c_cflag & CLOCAL))
-		wake_up_interruptible(&info->open_wait);
+		wake_up_interruptible(&info->port.open_wait);
 #endif
 }				/* cy_set_termios */
 
@@ -4326,14 +4326,14 @@ static void cy_hangup(struct tty_struct *tty)
 
 	cy_flush_buffer(tty);
 	shutdown(info);
-	info->count = 0;
+	info->port.count = 0;
 #ifdef CY_DEBUG_COUNT
 	printk(KERN_DEBUG "cyc:cy_hangup (%d): setting count to 0\n",
 		current->pid);
 #endif
-	info->tty = NULL;
-	info->flags &= ~ASYNC_NORMAL_ACTIVE;
-	wake_up_interruptible(&info->open_wait);
+	info->port.tty = NULL;
+	info->port.flags &= ~ASYNC_NORMAL_ACTIVE;
+	wake_up_interruptible(&info->port.open_wait);
 }				/* cy_hangup */
 
 /*
@@ -4379,12 +4379,11 @@ static int __devinit cy_init_card(struct cyclades_card *cinfo)
 		info->magic = CYCLADES_MAGIC;
 		info->card = cinfo;
 		info->line = port;
-		info->flags = STD_COM_FLAGS;
 		info->closing_wait = CLOSING_WAIT_DELAY;
 		info->close_delay = 5 * HZ / 10;
 
-		init_waitqueue_head(&info->open_wait);
-		init_waitqueue_head(&info->close_wait);
+		tty_port_init(&info->port);
+		info->port.flags = STD_COM_FLAGS;
 		init_completion(&info->shutdown_wait);
 		init_waitqueue_head(&info->delta_msr_wait);
 
@@ -5237,7 +5236,7 @@ cyclades_get_proc_info(char *buf, char **start, off_t offset, int length,
 		for (j = 0; j < cy_card[i].nports; j++) {
 			info = &cy_card[i].ports[j];
 
-			if (info->count)
+			if (info->port.count)
 				size = sprintf(buf + len, "%3d %8lu %10lu %8lu "
 					"%10lu %8lu %9lu %6ld\n", info->line,
 					(cur_jifs - info->idle_stats.in_use) /
@@ -5247,7 +5246,7 @@ cyclades_get_proc_info(char *buf, char **start, off_t offset, int length,
 					(cur_jifs - info->idle_stats.recv_idle)/
 					HZ, info->idle_stats.overruns,
 					/* FIXME: double check locking */
-					(long)info->tty->ldisc.ops->num);
+					(long)info->port.tty->ldisc.ops->num);
 			else
 				size = sprintf(buf + len, "%3d %8lu %10lu %8lu "
 					"%10lu %8lu %9lu %6ld\n",
diff --git a/include/linux/cyclades.h b/include/linux/cyclades.h
index 504cb2c3fa9a..a982b74a6ee6 100644
--- a/include/linux/cyclades.h
+++ b/include/linux/cyclades.h
@@ -550,11 +550,11 @@ struct cyclades_icount {
 
 struct cyclades_port {
 	int                     magic;
+	struct tty_port		port;
 	struct cyclades_card	*card;
 	int			line;
 	int			flags; 		/* defined in tty.h */
 	int                     type;		/* UART type */
-	struct tty_struct 	*tty;
 	int			read_status_mask;
 	int			ignore_status_mask;
 	int			timeout;
@@ -569,11 +569,8 @@ struct cyclades_port {
 	u8			x_char; /* to be pushed out ASAP */
 	int			close_delay;
 	unsigned short		closing_wait;
-	int			count;	/* # of fd on device */
 	int                     breakon;
 	int                     breakoff;
-	int			blocked_open; /* # of blocked opens */
-	unsigned char 		*xmit_buf;
 	int			xmit_head;
 	int			xmit_tail;
 	int			xmit_cnt;
@@ -583,8 +580,6 @@ struct cyclades_port {
 	struct cyclades_monitor	mon;
 	struct cyclades_idle_stats	idle_stats;
 	struct cyclades_icount	icount;
-	wait_queue_head_t       open_wait;
-	wait_queue_head_t       close_wait;
 	struct completion       shutdown_wait;
 	wait_queue_head_t       delta_msr_wait;
 	int throttle;
-- 
cgit v1.2.3


From 44b7d1b37f786c61d0e382b6f72f605f73de284b Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Wed, 16 Jul 2008 21:57:18 +0100
Subject: tty: add more tty_port fields

Move more bits into the tty_port structure

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/cyclades.c      |  28 ++++----
 drivers/char/isicom.c        |  28 ++++----
 drivers/char/moxa.c          |   8 +--
 drivers/char/mxser.c         |  29 ++++----
 drivers/char/riscom8.c       |  24 ++++---
 drivers/char/riscom8.h       |   2 -
 drivers/char/rocket.c        |  23 +++----
 drivers/char/rocket.h        |   4 +-
 drivers/char/rocket_int.h    |   2 -
 drivers/char/specialix.c     | 153 +++++++++++++++++++++----------------------
 drivers/char/specialix_io8.h |   8 +--
 drivers/char/sx.c            |   2 +-
 drivers/char/synclink.c      |  16 ++---
 drivers/char/synclink_gt.c   |  16 ++---
 drivers/char/synclinkmp.c    |  14 ++--
 drivers/char/tty_io.c        |   2 +
 include/linux/cyclades.h     |   6 +-
 include/linux/tty.h          |   2 +
 18 files changed, 169 insertions(+), 198 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c
index 0144b19d1036..e991dc85f2fb 100644
--- a/drivers/char/cyclades.c
+++ b/drivers/char/cyclades.c
@@ -2677,8 +2677,8 @@ static void cy_close(struct tty_struct *tty, struct file *filp)
 	 */
 	tty->closing = 1;
 	spin_unlock_irqrestore(&card->card_lock, flags);
-	if (info->closing_wait != CY_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, info->closing_wait);
+	if (info->port.closing_wait != CY_CLOSING_WAIT_NONE)
+		tty_wait_until_sent(tty, info->port.closing_wait);
 
 	spin_lock_irqsave(&card->card_lock, flags);
 
@@ -2734,9 +2734,9 @@ static void cy_close(struct tty_struct *tty, struct file *filp)
 	info->port.tty = NULL;
 	if (info->port.blocked_open) {
 		spin_unlock_irqrestore(&card->card_lock, flags);
-		if (info->close_delay) {
+		if (info->port.close_delay) {
 			msleep_interruptible(jiffies_to_msecs
-						(info->close_delay));
+						(info->port.close_delay));
 		}
 		wake_up_interruptible(&info->port.open_wait);
 		spin_lock_irqsave(&card->card_lock, flags);
@@ -3382,8 +3382,8 @@ get_serial_info(struct cyclades_port *info,
 		cinfo->first_line;
 	tmp.irq = cinfo->irq;
 	tmp.flags = info->port.flags;
-	tmp.close_delay = info->close_delay;
-	tmp.closing_wait = info->closing_wait;
+	tmp.close_delay = info->port.close_delay;
+	tmp.closing_wait = info->port.closing_wait;
 	tmp.baud_base = info->baud;
 	tmp.custom_divisor = info->custom_divisor;
 	tmp.hub6 = 0;		/*!!! */
@@ -3402,7 +3402,7 @@ set_serial_info(struct cyclades_port *info,
 	old_info = *info;
 
 	if (!capable(CAP_SYS_ADMIN)) {
-		if (new_serial.close_delay != info->close_delay ||
+		if (new_serial.close_delay != info->port.close_delay ||
 				new_serial.baud_base != info->baud ||
 				(new_serial.flags & ASYNC_FLAGS &
 					~ASYNC_USR_MASK) !=
@@ -3424,8 +3424,8 @@ set_serial_info(struct cyclades_port *info,
 	info->custom_divisor = new_serial.custom_divisor;
 	info->port.flags = (info->port.flags & ~ASYNC_FLAGS) |
 			(new_serial.flags & ASYNC_FLAGS);
-	info->close_delay = new_serial.close_delay * HZ / 100;
-	info->closing_wait = new_serial.closing_wait * HZ / 100;
+	info->port.close_delay = new_serial.close_delay * HZ / 100;
+	info->port.closing_wait = new_serial.closing_wait * HZ / 100;
 
 check_and_exit:
 	if (info->port.flags & ASYNC_INITIALIZED) {
@@ -3971,11 +3971,11 @@ cy_ioctl(struct tty_struct *tty, struct file *file,
 		break;
 #endif				/* CONFIG_CYZ_INTR */
 	case CYSETWAIT:
-		info->closing_wait = (unsigned short)arg * HZ / 100;
+		info->port.closing_wait = (unsigned short)arg * HZ / 100;
 		ret_val = 0;
 		break;
 	case CYGETWAIT:
-		ret_val = info->closing_wait / (HZ / 100);
+		ret_val = info->port.closing_wait / (HZ / 100);
 		break;
 	case TIOCGSERIAL:
 		ret_val = get_serial_info(info, argp);
@@ -4376,13 +4376,13 @@ static int __devinit cy_init_card(struct cyclades_card *cinfo)
 	for (port = cinfo->first_line; port < cinfo->first_line + nports;
 			port++) {
 		info = &cinfo->ports[port - cinfo->first_line];
+		tty_port_init(&info->port);
 		info->magic = CYCLADES_MAGIC;
 		info->card = cinfo;
 		info->line = port;
-		info->closing_wait = CLOSING_WAIT_DELAY;
-		info->close_delay = 5 * HZ / 10;
 
-		tty_port_init(&info->port);
+		info->port.closing_wait = CLOSING_WAIT_DELAY;
+		info->port.close_delay = 5 * HZ / 10;
 		info->port.flags = STD_COM_FLAGS;
 		init_completion(&info->shutdown_wait);
 		init_waitqueue_head(&info->delta_msr_wait);
diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index 5a53c15b0dc2..d4281df10c22 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -199,10 +199,8 @@ struct	isi_board {
 struct	isi_port {
 	unsigned short		magic;
 	struct tty_port		port;
-	int			close_delay;
 	u16			channel;
 	u16			status;
-	u16			closing_wait;
 	struct isi_board	*card;
 	unsigned char		*xmit_buf;
 	int			xmit_head;
@@ -1051,8 +1049,8 @@ static void isicom_close(struct tty_struct *tty, struct file *filp)
 	tty->closing = 1;
 	spin_unlock_irqrestore(&card->card_lock, flags);
 
-	if (port->closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, port->closing_wait);
+	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
+		tty_wait_until_sent(tty, port->port.closing_wait);
 	/* indicate to the card that no more data can be received
 	   on this port */
 	spin_lock_irqsave(&card->card_lock, flags);
@@ -1071,10 +1069,10 @@ static void isicom_close(struct tty_struct *tty, struct file *filp)
 
 	if (port->port.blocked_open) {
 		spin_unlock_irqrestore(&card->card_lock, flags);
-		if (port->close_delay) {
+		if (port->port.close_delay) {
 			pr_dbg("scheduling until time out.\n");
 			msleep_interruptible(
-				jiffies_to_msecs(port->close_delay));
+				jiffies_to_msecs(port->port.close_delay));
 		}
 		spin_lock_irqsave(&card->card_lock, flags);
 		wake_up_interruptible(&port->port.open_wait);
@@ -1256,8 +1254,8 @@ static int isicom_set_serial_info(struct isi_port *port,
 		(newinfo.flags & ASYNC_SPD_MASK));
 
 	if (!capable(CAP_SYS_ADMIN)) {
-		if ((newinfo.close_delay != port->close_delay) ||
-				(newinfo.closing_wait != port->closing_wait) ||
+		if ((newinfo.close_delay != port->port.close_delay) ||
+				(newinfo.closing_wait != port->port.closing_wait) ||
 				((newinfo.flags & ~ASYNC_USR_MASK) !=
 				(port->port.flags & ~ASYNC_USR_MASK))) {
 			unlock_kernel();
@@ -1266,8 +1264,8 @@ static int isicom_set_serial_info(struct isi_port *port,
 		port->port.flags = ((port->port.flags & ~ASYNC_USR_MASK) |
 				(newinfo.flags & ASYNC_USR_MASK));
 	} else {
-		port->close_delay = newinfo.close_delay;
-		port->closing_wait = newinfo.closing_wait;
+		port->port.close_delay = newinfo.close_delay;
+		port->port.closing_wait = newinfo.closing_wait;
 		port->port.flags = ((port->port.flags & ~ASYNC_FLAGS) |
 				(newinfo.flags & ASYNC_FLAGS));
 	}
@@ -1294,8 +1292,8 @@ static int isicom_get_serial_info(struct isi_port *port,
 	out_info.irq = port->card->irq;
 	out_info.flags = port->port.flags;
 /*	out_info.baud_base = ? */
-	out_info.close_delay = port->close_delay;
-	out_info.closing_wait = port->closing_wait;
+	out_info.close_delay = port->port.close_delay;
+	out_info.closing_wait = port->port.closing_wait;
 	unlock_kernel();
 	if (copy_to_user(info, &out_info, sizeof(out_info)))
 		return -EFAULT;
@@ -1804,13 +1802,13 @@ static int __init isicom_init(void)
 		isi_card[idx].ports = port;
 		spin_lock_init(&isi_card[idx].card_lock);
 		for (channel = 0; channel < 16; channel++, port++) {
+			tty_port_init(&port->port);
 			port->magic = ISICOM_MAGIC;
 			port->card = &isi_card[idx];
 			port->channel = channel;
-			port->close_delay = 50 * HZ/100;
-			port->closing_wait = 3000 * HZ/100;
+			port->port.close_delay = 50 * HZ/100;
+			port->port.closing_wait = 3000 * HZ/100;
 			port->status = 0;
-			tty_port_init(&port->port);
 			/*  . . .  */
 		}
 		isi_card[idx].base = 0;
diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c
index dd10f143d96f..2bba250ffc8e 100644
--- a/drivers/char/moxa.c
+++ b/drivers/char/moxa.c
@@ -135,7 +135,6 @@ struct moxa_port {
 	void __iomem *tableAddr;
 
 	int type;
-	int close_delay;
 	int cflag;
 	unsigned long statusflags;
 
@@ -822,10 +821,9 @@ static int moxa_init_board(struct moxa_board_conf *brd, struct device *dev)
 	}
 
 	for (i = 0, p = brd->ports; i < MAX_PORTS_PER_BOARD; i++, p++) {
+		tty_port_init(&p->port);
 		p->type = PORT_16550A;
-		p->close_delay = 5 * HZ / 10;
 		p->cflag = B9600 | CS8 | CREAD | CLOCAL | HUPCL;
-		tty_port_init(&p->port);
 	}
 
 	switch (brd->boardType) {
@@ -2124,7 +2122,7 @@ static int moxa_get_serial_info(struct moxa_port *info,
 		.line = info->port.tty->index,
 		.flags = info->port.flags,
 		.baud_base = 921600,
-		.close_delay = info->close_delay
+		.close_delay = info->port.close_delay
 	};
 	return copy_to_user(retinfo, &tmp, sizeof(*retinfo)) ? -EFAULT : 0;
 }
@@ -2148,7 +2146,7 @@ static int moxa_set_serial_info(struct moxa_port *info,
 		     (info->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
 	} else
-		info->close_delay = new_serial.close_delay * HZ / 100;
+		info->port.close_delay = new_serial.close_delay * HZ / 100;
 
 	new_serial.flags = (new_serial.flags & ~ASYNC_FLAGS);
 	new_serial.flags |= (info->port.flags & ASYNC_FLAGS);
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index e83ccee03161..6307e301bd26 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -243,10 +243,7 @@ struct mxser_port {
 	unsigned char ldisc_stop_rx;
 
 	int custom_divisor;
-	int close_delay;
-	unsigned short closing_wait;
 	unsigned char err_shadow;
-	unsigned long event;
 
 	struct async_icount icount; /* kernel counters for 4 input interrupts */
 	int timeout;
@@ -1199,8 +1196,8 @@ static void mxser_close(struct tty_struct *tty, struct file *filp)
 	 * the line discipline to only process XON/XOFF characters.
 	 */
 	tty->closing = 1;
-	if (info->closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, info->closing_wait);
+	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
+		tty_wait_until_sent(tty, info->port.closing_wait);
 	/*
 	 * At this point we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts, and tell the
@@ -1231,11 +1228,10 @@ static void mxser_close(struct tty_struct *tty, struct file *filp)
 	tty_ldisc_flush(tty);
 
 	tty->closing = 0;
-	info->event = 0;
 	info->port.tty = NULL;
 	if (info->port.blocked_open) {
-		if (info->close_delay)
-			schedule_timeout_interruptible(info->close_delay);
+		if (info->port.close_delay)
+			schedule_timeout_interruptible(info->port.close_delay);
 		wake_up_interruptible(&info->port.open_wait);
 	}
 
@@ -1370,8 +1366,8 @@ static int mxser_get_serial_info(struct mxser_port *info,
 		.irq = info->board->irq,
 		.flags = info->port.flags,
 		.baud_base = info->baud_base,
-		.close_delay = info->close_delay,
-		.closing_wait = info->closing_wait,
+		.close_delay = info->port.close_delay,
+		.closing_wait = info->port.closing_wait,
 		.custom_divisor = info->custom_divisor,
 		.hub6 = 0
 	};
@@ -1402,7 +1398,7 @@ static int mxser_set_serial_info(struct mxser_port *info,
 
 	if (!capable(CAP_SYS_ADMIN)) {
 		if ((new_serial.baud_base != info->baud_base) ||
-				(new_serial.close_delay != info->close_delay) ||
+				(new_serial.close_delay != info->port.close_delay) ||
 				((new_serial.flags & ~ASYNC_USR_MASK) != (info->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
 		info->port.flags = ((info->port.flags & ~ASYNC_USR_MASK) |
@@ -1414,8 +1410,8 @@ static int mxser_set_serial_info(struct mxser_port *info,
 		 */
 		info->port.flags = ((info->port.flags & ~ASYNC_FLAGS) |
 				(new_serial.flags & ASYNC_FLAGS));
-		info->close_delay = new_serial.close_delay * HZ / 100;
-		info->closing_wait = new_serial.closing_wait * HZ / 100;
+		info->port.close_delay = new_serial.close_delay * HZ / 100;
+		info->port.closing_wait = new_serial.closing_wait * HZ / 100;
 		info->port.tty->low_latency =
 				(info->port.flags & ASYNC_LOW_LATENCY) ? 1 : 0;
 		info->port.tty->low_latency = 0;
@@ -2214,7 +2210,6 @@ static void mxser_hangup(struct tty_struct *tty)
 
 	mxser_flush_buffer(tty);
 	mxser_shutdown(info);
-	info->event = 0;
 	info->port.count = 0;
 	info->port.flags &= ~ASYNC_NORMAL_ACTIVE;
 	info->port.tty = NULL;
@@ -2545,6 +2540,7 @@ static int __devinit mxser_initbrd(struct mxser_board *brd,
 
 	for (i = 0; i < brd->info->nports; i++) {
 		info = &brd->ports[i];
+		tty_port_init(&info->port);
 		info->board = brd;
 		info->stop_rx = 0;
 		info->ldisc_stop_rx = 0;
@@ -2559,10 +2555,9 @@ static int __devinit mxser_initbrd(struct mxser_board *brd,
 		process_txrx_fifo(info);
 
 		info->custom_divisor = info->baud_base * 16;
-		info->close_delay = 5 * HZ / 10;
-		info->closing_wait = 30 * HZ;
+		info->port.close_delay = 5 * HZ / 10;
+		info->port.closing_wait = 30 * HZ;
 		info->normal_termios = mxvar_sdriver->init_termios;
-		tty_port_init(&info->port);
 		init_waitqueue_head(&info->delta_msr_wait);
 		memset(&info->mon_data, 0, sizeof(struct mxser_mon));
 		info->err_shadow = 0;
diff --git a/drivers/char/riscom8.c b/drivers/char/riscom8.c
index 3ca8957ba327..724b2b20f4b2 100644
--- a/drivers/char/riscom8.c
+++ b/drivers/char/riscom8.c
@@ -1032,8 +1032,8 @@ static void rc_close(struct tty_struct *tty, struct file *filp)
 	 * the line discipline to only process XON/XOFF characters.
 	 */
 	tty->closing = 1;
-	if (port->closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, port->closing_wait);
+	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
+		tty_wait_until_sent(tty, port->port.closing_wait);
 	/*
 	 * At this point we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts, and tell the
@@ -1065,8 +1065,8 @@ static void rc_close(struct tty_struct *tty, struct file *filp)
 	tty->closing = 0;
 	port->port.tty = NULL;
 	if (port->port.blocked_open) {
-		if (port->close_delay)
-			msleep_interruptible(jiffies_to_msecs(port->close_delay));
+		if (port->port.close_delay)
+			msleep_interruptible(jiffies_to_msecs(port->port.close_delay));
 		wake_up_interruptible(&port->port.open_wait);
 	}
 	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
@@ -1295,8 +1295,8 @@ static int rc_set_serial_info(struct riscom_port *port,
 			(tmp.flags & ASYNC_SPD_MASK));
 
 	if (!capable(CAP_SYS_ADMIN)) {
-		if ((tmp.close_delay != port->close_delay) ||
-		    (tmp.closing_wait != port->closing_wait) ||
+		if ((tmp.close_delay != port->port.close_delay) ||
+		    (tmp.closing_wait != port->port.closing_wait) ||
 		    ((tmp.flags & ~ASYNC_USR_MASK) !=
 		     (port->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
@@ -1305,8 +1305,8 @@ static int rc_set_serial_info(struct riscom_port *port,
 	} else  {
 		port->port.flags = ((port->port.flags & ~ASYNC_FLAGS) |
 			       (tmp.flags & ASYNC_FLAGS));
-		port->close_delay = tmp.close_delay;
-		port->closing_wait = tmp.closing_wait;
+		port->port.close_delay = tmp.close_delay;
+		port->port.closing_wait = tmp.closing_wait;
 	}
 	if (change_speed)  {
 		unsigned long flags;
@@ -1331,8 +1331,8 @@ static int rc_get_serial_info(struct riscom_port *port,
 	tmp.irq  = bp->irq;
 	tmp.flags = port->port.flags;
 	tmp.baud_base = (RC_OSCFREQ + CD180_TPC/2) / CD180_TPC;
-	tmp.close_delay = port->close_delay * HZ/100;
-	tmp.closing_wait = port->closing_wait * HZ/100;
+	tmp.close_delay = port->port.close_delay * HZ/100;
+	tmp.closing_wait = port->port.closing_wait * HZ/100;
 	tmp.xmit_fifo_size = CD180_NFIFO;
 	return copy_to_user(retinfo, &tmp, sizeof(tmp)) ? -EFAULT : 0;
 }
@@ -1549,10 +1549,8 @@ static int __init rc_init_drivers(void)
 	}
 	memset(rc_port, 0, sizeof(rc_port));
 	for (i = 0; i < RC_NPORT * RC_NBOARD; i++)  {
-		rc_port[i].magic = RISCOM8_MAGIC;
-		rc_port[i].close_delay = 50 * HZ / 100;
-		rc_port[i].closing_wait = 3000 * HZ / 100;
 		tty_port_init(&rc_port[i].port);
+		rc_port[i].magic = RISCOM8_MAGIC;
 	}
 	return 0;
 }
diff --git a/drivers/char/riscom8.h b/drivers/char/riscom8.h
index 29ddbab78801..c9876b3f9714 100644
--- a/drivers/char/riscom8.h
+++ b/drivers/char/riscom8.h
@@ -69,14 +69,12 @@ struct riscom_port {
 	struct			tty_port port;
 	int			baud_base;
 	int			timeout;
-	int			close_delay;
 	int			custom_divisor;
 	int			xmit_head;
 	int			xmit_tail;
 	int			xmit_cnt;
 	short			wakeup_chars;
 	short			break_length;
-	unsigned short		closing_wait;
 	unsigned char		mark_mask;
 	unsigned char		IER;
 	unsigned char		MSVR;
diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c
index bc35b900f9c6..e670eae2f510 100644
--- a/drivers/char/rocket.c
+++ b/drivers/char/rocket.c
@@ -72,6 +72,7 @@
 #include <linux/tty.h>
 #include <linux/tty_driver.h>
 #include <linux/tty_flip.h>
+#include <linux/serial.h>
 #include <linux/string.h>
 #include <linux/fcntl.h>
 #include <linux/ptrace.h>
@@ -81,7 +82,7 @@
 #include <linux/completion.h>
 #include <linux/wait.h>
 #include <linux/pci.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <asm/atomic.h>
 #include <asm/unaligned.h>
 #include <linux/bitops.h>
@@ -648,8 +649,8 @@ static void init_r_port(int board, int aiop, int chan, struct pci_dev *pci_dev)
 	info->board = board;
 	info->aiop = aiop;
 	info->chan = chan;
-	info->closing_wait = 3000;
-	info->close_delay = 50;
+	info->port.closing_wait = 3000;
+	info->port.close_delay = 50;
 	init_waitqueue_head(&info->port.open_wait);
 	init_completion(&info->close_wait);
 	info->flags &= ~ROCKET_MODE_MASK;
@@ -1137,8 +1138,8 @@ static void rp_close(struct tty_struct *tty, struct file *filp)
 	/*
 	 * Wait for the transmit buffer to clear
 	 */
-	if (info->closing_wait != ROCKET_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, info->closing_wait);
+	if (info->port.closing_wait != ROCKET_CLOSING_WAIT_NONE)
+		tty_wait_until_sent(tty, info->port.closing_wait);
 	/*
 	 * Before we drop DTR, make sure the UART transmitter
 	 * has completely drained; this is especially
@@ -1168,8 +1169,8 @@ static void rp_close(struct tty_struct *tty, struct file *filp)
 	clear_bit((info->aiop * 8) + info->chan, (void *) &xmit_flags[info->board]);
 
 	if (info->port.blocked_open) {
-		if (info->close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->close_delay));
+		if (info->port.close_delay) {
+			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
 		}
 		wake_up_interruptible(&info->port.open_wait);
 	} else {
@@ -1327,8 +1328,8 @@ static int get_config(struct r_port *info, struct rocket_config __user *retinfo)
 	memset(&tmp, 0, sizeof (tmp));
 	tmp.line = info->line;
 	tmp.flags = info->flags;
-	tmp.close_delay = info->close_delay;
-	tmp.closing_wait = info->closing_wait;
+	tmp.close_delay = info->port.close_delay;
+	tmp.closing_wait = info->port.closing_wait;
 	tmp.port = rcktpt_io_addr[(info->line >> 5) & 3];
 
 	if (copy_to_user(retinfo, &tmp, sizeof (*retinfo)))
@@ -1353,8 +1354,8 @@ static int set_config(struct r_port *info, struct rocket_config __user *new_info
 	}
 
 	info->flags = ((info->flags & ~ROCKET_FLAGS) | (new_serial.flags & ROCKET_FLAGS));
-	info->close_delay = new_serial.close_delay;
-	info->closing_wait = new_serial.closing_wait;
+	info->port.close_delay = new_serial.close_delay;
+	info->port.closing_wait = new_serial.closing_wait;
 
 	if ((info->flags & ROCKET_SPD_MASK) == ROCKET_SPD_HI)
 		info->port.tty->alt_speed = 57600;
diff --git a/drivers/char/rocket.h b/drivers/char/rocket.h
index ae6b04f90c03..a8b09195ebba 100644
--- a/drivers/char/rocket.h
+++ b/drivers/char/rocket.h
@@ -64,8 +64,8 @@ struct rocket_version {
 /*
  * For closing_wait and closing_wait2
  */
-#define ROCKET_CLOSING_WAIT_NONE	65535
-#define ROCKET_CLOSING_WAIT_INF		0
+#define ROCKET_CLOSING_WAIT_NONE	ASYNC_CLOSING_WAIT_NONE
+#define ROCKET_CLOSING_WAIT_INF		ASYNC_CLOSING_WAIT_INF
 
 /*
  * Rocketport ioctls -- "RP"
diff --git a/drivers/char/rocket_int.h b/drivers/char/rocket_int.h
index 3affc48f6a57..21f3ff53ba32 100644
--- a/drivers/char/rocket_int.h
+++ b/drivers/char/rocket_int.h
@@ -1133,8 +1133,6 @@ struct r_port {
 	unsigned int chan:3;
 	CONTROLLER_t *ctlp;
 	CHANNEL_t channel;
-	int closing_wait;
-	int close_delay;
 	int intmask;
 	int xmit_fifo_room;	/* room in xmit fifo */
 	unsigned char *xmit_buf;
diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c
index 2ee4d9893757..037dc47e4cb1 100644
--- a/drivers/char/specialix.c
+++ b/drivers/char/specialix.c
@@ -608,9 +608,9 @@ static inline struct specialix_port * sx_get_port(struct specialix_board * bp,
 	dprintk (SX_DEBUG_CHAN, "channel: %d\n", channel);
 	if (channel < CD186x_NCH) {
 		port = &sx_port[board_No(bp) * SX_NPORT + channel];
-		dprintk (SX_DEBUG_CHAN, "port: %d %p flags: 0x%x\n",board_No(bp) * SX_NPORT + channel,  port, port->flags & ASYNC_INITIALIZED);
+		dprintk (SX_DEBUG_CHAN, "port: %d %p flags: 0x%lx\n",board_No(bp) * SX_NPORT + channel,  port, port->port.flags & ASYNC_INITIALIZED);
 
-		if (port->flags & ASYNC_INITIALIZED) {
+		if (port->port.flags & ASYNC_INITIALIZED) {
 			dprintk (SX_DEBUG_CHAN, "port: %d %p\n", channel, port);
 			func_exit();
 			return port;
@@ -637,7 +637,7 @@ static inline void sx_receive_exc(struct specialix_board * bp)
 		func_exit();
 		return;
 	}
-	tty = port->tty;
+	tty = port->port.tty;
 
 	status = sx_in(bp, CD186x_RCSR);
 
@@ -673,7 +673,7 @@ static inline void sx_receive_exc(struct specialix_board * bp)
 		dprintk(SX_DEBUG_RX, "sx%d: port %d: Handling break...\n",
 		       board_No(bp), port_No(port));
 		flag = TTY_BREAK;
-		if (port->flags & ASYNC_SAK)
+		if (port->port.flags & ASYNC_SAK)
 			do_SAK(tty);
 
 	} else if (status & RCSR_PE)
@@ -707,7 +707,7 @@ static inline void sx_receive(struct specialix_board * bp)
 		func_exit();
 		return;
 	}
-	tty = port->tty;
+	tty = port->port.tty;
 
 	count = sx_in(bp, CD186x_RDCR);
 	dprintk (SX_DEBUG_RX, "port: %p: count: %d\n", port, count);
@@ -734,7 +734,7 @@ static inline void sx_transmit(struct specialix_board * bp)
 		return;
 	}
 	dprintk (SX_DEBUG_TX, "port: %p\n", port);
-	tty = port->tty;
+	tty = port->port.tty;
 
 	if (port->IER & IER_TXEMPTY) {
 		/* FIFO drained */
@@ -811,7 +811,7 @@ static inline void sx_check_modem(struct specialix_board * bp)
 	if (!(port = sx_get_port(bp, "Modem")))
 		return;
 
-	tty = port->tty;
+	tty = port->port.tty;
 
 	mcr = sx_in(bp, CD186x_MCR);
 	printk ("mcr = %02x.\n", mcr);
@@ -821,7 +821,7 @@ static inline void sx_check_modem(struct specialix_board * bp)
 		msvr_cd = sx_in(bp, CD186x_MSVR) & MSVR_CD;
 		if (msvr_cd) {
 			dprintk (SX_DEBUG_SIGNALS, "Waking up guys in open.\n");
-			wake_up_interruptible(&port->open_wait);
+			wake_up_interruptible(&port->port.open_wait);
 		} else {
 			dprintk (SX_DEBUG_SIGNALS, "Sending HUP.\n");
 			tty_hangup(tty);
@@ -1030,7 +1030,7 @@ static void sx_change_speed(struct specialix_board *bp, struct specialix_port *p
 
 	func_enter();
 
-	if (!(tty = port->tty) || !tty->termios) {
+	if (!(tty = port->port.tty) || !tty->termios) {
 		func_exit();
 		return;
 	}
@@ -1052,9 +1052,9 @@ static void sx_change_speed(struct specialix_board *bp, struct specialix_port *p
 	baud = tty_get_baud_rate(tty);
 
 	if (baud == 38400) {
-		if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
+		if ((port->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
 			baud = 57600;
-		if ((port->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
+		if ((port->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
 			baud = 115200;
 	}
 
@@ -1244,7 +1244,7 @@ static int sx_setup_port(struct specialix_board *bp, struct specialix_port *port
 
 	func_enter();
 
-	if (port->flags & ASYNC_INITIALIZED) {
+	if (port->port.flags & ASYNC_INITIALIZED) {
 		func_exit();
 		return 0;
 	}
@@ -1268,12 +1268,12 @@ static int sx_setup_port(struct specialix_board *bp, struct specialix_port *port
 
 	spin_lock_irqsave(&port->lock, flags);
 
-	if (port->tty)
-		clear_bit(TTY_IO_ERROR, &port->tty->flags);
+	if (port->port.tty)
+		clear_bit(TTY_IO_ERROR, &port->port.tty->flags);
 
 	port->xmit_cnt = port->xmit_head = port->xmit_tail = 0;
 	sx_change_speed(bp, port);
-	port->flags |= ASYNC_INITIALIZED;
+	port->port.flags |= ASYNC_INITIALIZED;
 
 	spin_unlock_irqrestore(&port->lock, flags);
 
@@ -1292,7 +1292,7 @@ static void sx_shutdown_port(struct specialix_board *bp, struct specialix_port *
 
 	func_enter();
 
-	if (!(port->flags & ASYNC_INITIALIZED)) {
+	if (!(port->port.flags & ASYNC_INITIALIZED)) {
 		func_exit();
 		return;
 	}
@@ -1315,7 +1315,7 @@ static void sx_shutdown_port(struct specialix_board *bp, struct specialix_port *
 	spin_lock_irqsave(&bp->lock, flags);
 	sx_out(bp, CD186x_CAR, port_No(port));
 
-	if (!(tty = port->tty) || C_HUPCL(tty)) {
+	if (!(tty = port->port.tty) || C_HUPCL(tty)) {
 		/* Drop DTR */
 		sx_out(bp, CD186x_MSVDTR, 0);
 	}
@@ -1330,7 +1330,7 @@ static void sx_shutdown_port(struct specialix_board *bp, struct specialix_port *
 	spin_unlock_irqrestore(&bp->lock, flags);
 	if (tty)
 		set_bit(TTY_IO_ERROR, &tty->flags);
-	port->flags &= ~ASYNC_INITIALIZED;
+	port->port.flags &= ~ASYNC_INITIALIZED;
 
 	if (!bp->count)
 		sx_shutdown_board(bp);
@@ -1354,9 +1354,9 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
-	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&port->close_wait);
-		if (port->flags & ASYNC_HUP_NOTIFY) {
+	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&port->port.close_wait);
+		if (port->port.flags & ASYNC_HUP_NOTIFY) {
 			func_exit();
 			return -EAGAIN;
 		} else {
@@ -1371,7 +1371,7 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		port->flags |= ASYNC_NORMAL_ACTIVE;
+		port->port.flags |= ASYNC_NORMAL_ACTIVE;
 		func_exit();
 		return 0;
 	}
@@ -1387,13 +1387,13 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&port->open_wait, &wait);
+	add_wait_queue(&port->port.open_wait, &wait);
 	spin_lock_irqsave(&port->lock, flags);
 	if (!tty_hung_up_p(filp)) {
-		port->count--;
+		port->port.count--;
 	}
 	spin_unlock_irqrestore(&port->lock, flags);
-	port->blocked_open++;
+	port->port.blocked_open++;
 	while (1) {
 		spin_lock_irqsave(&bp->lock, flags);
 		sx_out(bp, CD186x_CAR, port_No(port));
@@ -1410,14 +1410,14 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 		spin_unlock_irqrestore(&bp->lock, flags);
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(port->flags & ASYNC_INITIALIZED)) {
-			if (port->flags & ASYNC_HUP_NOTIFY)
+		    !(port->port.flags & ASYNC_INITIALIZED)) {
+			if (port->port.flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(port->flags & ASYNC_CLOSING) &&
+		if (!(port->port.flags & ASYNC_CLOSING) &&
 		    (do_clocal || CD))
 			break;
 		if (signal_pending(current)) {
@@ -1428,19 +1428,19 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 	}
 
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->open_wait, &wait);
+	remove_wait_queue(&port->port.open_wait, &wait);
 	spin_lock_irqsave(&port->lock, flags);
 	if (!tty_hung_up_p(filp)) {
-		port->count++;
+		port->port.count++;
 	}
-	port->blocked_open--;
+	port->port.blocked_open--;
 	spin_unlock_irqrestore(&port->lock, flags);
 	if (retval) {
 		func_exit();
 		return retval;
 	}
 
-	port->flags |= ASYNC_NORMAL_ACTIVE;
+	port->port.flags |= ASYNC_NORMAL_ACTIVE;
 	func_exit();
 	return 0;
 }
@@ -1484,10 +1484,10 @@ static int sx_open(struct tty_struct * tty, struct file * filp)
 	}
 
 	spin_lock_irqsave(&bp->lock, flags);
-	port->count++;
+	port->port.count++;
 	bp->count++;
 	tty->driver_data = port;
-	port->tty = tty;
+	port->port.tty = tty;
 	spin_unlock_irqrestore(&bp->lock, flags);
 
 	if ((error = sx_setup_port(bp, port))) {
@@ -1547,15 +1547,15 @@ static void sx_close(struct tty_struct * tty, struct file * filp)
 	}
 
 	bp = port_Board(port);
-	if ((tty->count == 1) && (port->count != 1)) {
+	if ((tty->count == 1) && (port->port.count != 1)) {
 		printk(KERN_ERR "sx%d: sx_close: bad port count;"
 		       " tty->count is 1, port count is %d\n",
-		       board_No(bp), port->count);
-		port->count = 1;
+		       board_No(bp), port->port.count);
+		port->port.count = 1;
 	}
 
-	if (port->count > 1) {
-		port->count--;
+	if (port->port.count > 1) {
+		port->port.count--;
 		bp->count--;
 
 		spin_unlock_irqrestore(&port->lock, flags);
@@ -1563,7 +1563,7 @@ static void sx_close(struct tty_struct * tty, struct file * filp)
 		func_exit();
 		return;
 	}
-	port->flags |= ASYNC_CLOSING;
+	port->port.flags |= ASYNC_CLOSING;
 	/*
 	 * Now we wait for the transmit buffer to clear; and we notify
 	 * the line discipline to only process XON/XOFF characters.
@@ -1571,8 +1571,8 @@ static void sx_close(struct tty_struct * tty, struct file * filp)
 	tty->closing = 1;
 	spin_unlock_irqrestore(&port->lock, flags);
 	dprintk (SX_DEBUG_OPEN, "Closing\n");
-	if (port->closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		tty_wait_until_sent(tty, port->closing_wait);
+	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
+		tty_wait_until_sent(tty, port->port.closing_wait);
 	}
 	/*
 	 * At this point we stop accepting input.  To do this, we
@@ -1582,7 +1582,7 @@ static void sx_close(struct tty_struct * tty, struct file * filp)
 	 */
 	dprintk (SX_DEBUG_OPEN, "Closed\n");
 	port->IER &= ~IER_RXD;
-	if (port->flags & ASYNC_INITIALIZED) {
+	if (port->port.flags & ASYNC_INITIALIZED) {
 		port->IER &= ~IER_TXRDY;
 		port->IER |= IER_TXEMPTY;
 		spin_lock_irqsave(&bp->lock, flags);
@@ -1611,10 +1611,10 @@ static void sx_close(struct tty_struct * tty, struct file * filp)
 		       board_No(bp), bp->count, tty->index);
 		bp->count = 0;
 	}
-	if (--port->count < 0) {
+	if (--port->port.count < 0) {
 		printk(KERN_ERR "sx%d: sx_close: bad port count for tty%d: %d\n",
-		       board_No(bp), port_No(port), port->count);
-		port->count = 0;
+		       board_No(bp), port_No(port), port->port.count);
+		port->port.count = 0;
 	}
 
 	sx_shutdown_port(bp, port);
@@ -1622,16 +1622,16 @@ static void sx_close(struct tty_struct * tty, struct file * filp)
 	tty_ldisc_flush(tty);
 	spin_lock_irqsave(&port->lock, flags);
 	tty->closing = 0;
-	port->tty = NULL;
+	port->port.tty = NULL;
 	spin_unlock_irqrestore(&port->lock, flags);
-	if (port->blocked_open) {
-		if (port->close_delay) {
-			msleep_interruptible(jiffies_to_msecs(port->close_delay));
+	if (port->port.blocked_open) {
+		if (port->port.close_delay) {
+			msleep_interruptible(jiffies_to_msecs(port->port.close_delay));
 		}
-		wake_up_interruptible(&port->open_wait);
+		wake_up_interruptible(&port->port.open_wait);
 	}
-	port->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&port->close_wait);
+	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
+	wake_up_interruptible(&port->port.close_wait);
 
 	func_exit();
 }
@@ -1815,7 +1815,7 @@ static int sx_tiocmget(struct tty_struct *tty, struct file *file)
 	dprintk (SX_DEBUG_INIT, "Got msvr[%d] = %02x, car = %d.\n",
 		port_No(port), status, sx_in (bp, CD186x_CAR));
 	dprintk (SX_DEBUG_INIT, "sx_port = %p, port = %p\n", sx_port, port);
-	if (SX_CRTSCTS(port->tty)) {
+	if (SX_CRTSCTS(port->port.tty)) {
 		result  = /*   (status & MSVR_RTS) ? */ TIOCM_DTR /* : 0) */
 		          |   ((status & MSVR_DTR) ? TIOCM_RTS : 0)
 		          |   ((status & MSVR_CD)  ? TIOCM_CAR : 0)
@@ -1857,7 +1857,7 @@ static int sx_tiocmset(struct tty_struct *tty, struct file *file,
    /*   if (set & TIOCM_DTR)
 		port->MSVR |= MSVR_DTR; */
 
-	if (SX_CRTSCTS(port->tty)) {
+	if (SX_CRTSCTS(port->port.tty)) {
 		if (set & TIOCM_RTS)
 			port->MSVR |= MSVR_DTR;
 	} else {
@@ -1869,7 +1869,7 @@ static int sx_tiocmset(struct tty_struct *tty, struct file *file,
 		port->MSVR &= ~MSVR_RTS; */
   /*    if (clear & TIOCM_DTR)
 		port->MSVR &= ~MSVR_DTR; */
-	if (SX_CRTSCTS(port->tty)) {
+	if (SX_CRTSCTS(port->port.tty)) {
 		if (clear & TIOCM_RTS)
 			port->MSVR &= ~MSVR_DTR;
 	} else {
@@ -1929,27 +1929,27 @@ static inline int sx_set_serial_info(struct specialix_port * port,
 
 	lock_kernel();
 
-	change_speed = ((port->flags & ASYNC_SPD_MASK) !=
+	change_speed = ((port->port.flags & ASYNC_SPD_MASK) !=
 			(tmp.flags & ASYNC_SPD_MASK));
 	change_speed |= (tmp.custom_divisor != port->custom_divisor);
 
 	if (!capable(CAP_SYS_ADMIN)) {
-		if ((tmp.close_delay != port->close_delay) ||
-		    (tmp.closing_wait != port->closing_wait) ||
+		if ((tmp.close_delay != port->port.close_delay) ||
+		    (tmp.closing_wait != port->port.closing_wait) ||
 		    ((tmp.flags & ~ASYNC_USR_MASK) !=
-		     (port->flags & ~ASYNC_USR_MASK))) {
+		     (port->port.flags & ~ASYNC_USR_MASK))) {
 			func_exit();
 			unlock_kernel();
 			return -EPERM;
 		}
-		port->flags = ((port->flags & ~ASYNC_USR_MASK) |
+		port->port.flags = ((port->port.flags & ~ASYNC_USR_MASK) |
 		                  (tmp.flags & ASYNC_USR_MASK));
 		port->custom_divisor = tmp.custom_divisor;
 	} else {
-		port->flags = ((port->flags & ~ASYNC_FLAGS) |
+		port->port.flags = ((port->port.flags & ~ASYNC_FLAGS) |
 		                  (tmp.flags & ASYNC_FLAGS));
-		port->close_delay = tmp.close_delay;
-		port->closing_wait = tmp.closing_wait;
+		port->port.close_delay = tmp.close_delay;
+		port->port.closing_wait = tmp.closing_wait;
 		port->custom_divisor = tmp.custom_divisor;
 	}
 	if (change_speed) {
@@ -1975,10 +1975,10 @@ static inline int sx_get_serial_info(struct specialix_port * port,
 	tmp.line = port - sx_port;
 	tmp.port = bp->base;
 	tmp.irq  = bp->irq;
-	tmp.flags = port->flags;
+	tmp.flags = port->port.flags;
 	tmp.baud_base = (SX_OSCFREQ + CD186x_TPC/2) / CD186x_TPC;
-	tmp.close_delay = port->close_delay * HZ/100;
-	tmp.closing_wait = port->closing_wait * HZ/100;
+	tmp.close_delay = port->port.close_delay * HZ/100;
+	tmp.closing_wait = port->port.closing_wait * HZ/100;
 	tmp.custom_divisor =  port->custom_divisor;
 	tmp.xmit_fifo_size = CD186x_NFIFO;
 	unlock_kernel();
@@ -2199,17 +2199,17 @@ static void sx_hangup(struct tty_struct * tty)
 
 	sx_shutdown_port(bp, port);
 	spin_lock_irqsave(&port->lock, flags);
-	bp->count -= port->count;
+	bp->count -= port->port.count;
 	if (bp->count < 0) {
 		printk(KERN_ERR "sx%d: sx_hangup: bad board count: %d port: %d\n",
 			board_No(bp), bp->count, tty->index);
 		bp->count = 0;
 	}
-	port->count = 0;
-	port->flags &= ~ASYNC_NORMAL_ACTIVE;
-	port->tty = NULL;
+	port->port.count = 0;
+	port->port.flags &= ~ASYNC_NORMAL_ACTIVE;
+	port->port.tty = NULL;
 	spin_unlock_irqrestore(&port->lock, flags);
-	wake_up_interruptible(&port->open_wait);
+	wake_up_interruptible(&port->port.open_wait);
 
 	func_exit();
 }
@@ -2224,10 +2224,6 @@ static void sx_set_termios(struct tty_struct * tty, struct ktermios * old_termio
 	if (sx_paranoia_check(port, tty->name, "sx_set_termios"))
 		return;
 
-	if (tty->termios->c_cflag == old_termios->c_cflag &&
-	    tty->termios->c_iflag == old_termios->c_iflag)
-		return;
-
 	bp = port_Board(port);
 	spin_lock_irqsave(&port->lock, flags);
 	sx_change_speed(port_Board(port), port);
@@ -2297,10 +2293,7 @@ static int sx_init_drivers(void)
 	memset(sx_port, 0, sizeof(sx_port));
 	for (i = 0; i < SX_NPORT * SX_NBOARD; i++) {
 		sx_port[i].magic = SPECIALIX_MAGIC;
-		sx_port[i].close_delay = 50 * HZ/100;
-		sx_port[i].closing_wait = 3000 * HZ/100;
-		init_waitqueue_head(&sx_port[i].open_wait);
-		init_waitqueue_head(&sx_port[i].close_wait);
+		tty_port_init(&sx_port[i].port);
 		spin_lock_init(&sx_port[i].lock);
 	}
 
diff --git a/drivers/char/specialix_io8.h b/drivers/char/specialix_io8.h
index 3f2f85bdf516..c63005274d9b 100644
--- a/drivers/char/specialix_io8.h
+++ b/drivers/char/specialix_io8.h
@@ -107,23 +107,17 @@ struct specialix_board {
 
 struct specialix_port {
 	int			magic;
+	struct tty_port		port;
 	int			baud_base;
 	int			flags;
-	struct tty_struct 	* tty;
-	int			count;
-	int			blocked_open;
 	int			timeout;
-	int			close_delay;
 	unsigned char 		* xmit_buf;
 	int			custom_divisor;
 	int			xmit_head;
 	int			xmit_tail;
 	int			xmit_cnt;
-	wait_queue_head_t	open_wait;
-	wait_queue_head_t	close_wait;
 	short			wakeup_chars;
 	short			break_length;
-	unsigned short		closing_wait;
 	unsigned char		mark_mask;
 	unsigned char		IER;
 	unsigned char		MSVR;
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index b1239ee48b78..d5cffcd6a572 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -2395,6 +2395,7 @@ static int sx_init_portstructs(int nboards, int nports)
 		board->ports = port;
 		for (j = 0; j < boards[i].nports; j++) {
 			sx_dprintk(SX_DEBUG_INIT, "initing port %d\n", j);
+			tty_port_init(&port->gs.port);
 			port->gs.magic = SX_MAGIC;
 			port->gs.close_delay = HZ / 2;
 			port->gs.closing_wait = 30 * HZ;
@@ -2407,7 +2408,6 @@ static int sx_init_portstructs(int nboards, int nports)
 			/*
 			 * Initializing wait queue
 			 */
-			tty_port_init(&port->gs.port);
 			port++;
 		}
 	}
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index 734098f7dfa2..9f14753fada1 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -183,8 +183,6 @@ struct mgsl_struct {
 	struct tty_port		port;
 	int			line;
 	int                     hw_version;
-	unsigned short		close_delay;
-	unsigned short		closing_wait;	/* time to wait before closing */
 	
 	struct mgsl_icount	icount;
 	
@@ -3142,11 +3140,11 @@ static void mgsl_close(struct tty_struct *tty, struct file * filp)
 	
 	/* wait for transmit data to clear all layers */
 	
-	if (info->closing_wait != ASYNC_CLOSING_WAIT_NONE) {
+	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
 		if (debug_level >= DEBUG_LEVEL_INFO)
 			printk("%s(%d):mgsl_close(%s) calling tty_wait_until_sent\n",
 				 __FILE__,__LINE__, info->device_name );
-		tty_wait_until_sent(tty, info->closing_wait);
+		tty_wait_until_sent(tty, info->port.closing_wait);
 	}
 		
  	if (info->port.flags & ASYNC_INITIALIZED)
@@ -3162,8 +3160,8 @@ static void mgsl_close(struct tty_struct *tty, struct file * filp)
 	info->port.tty = NULL;
 	
 	if (info->port.blocked_open) {
-		if (info->close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->close_delay));
+		if (info->port.close_delay) {
+			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
 		}
 		wake_up_interruptible(&info->port.open_wait);
 	}
@@ -4326,12 +4324,12 @@ static struct mgsl_struct* mgsl_allocate_device(void)
 	if (!info) {
 		printk("Error can't allocate device instance data\n");
 	} else {
+		tty_port_init(&info->port);
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, mgsl_bh_handler);
 		info->max_frame_size = 4096;
-		info->close_delay = 5*HZ/10;
-		info->closing_wait = 30*HZ;
-		tty_port_init(&info->port);
+		info->port.close_delay = 5*HZ/10;
+		info->port.closing_wait = 30*HZ;
 		init_waitqueue_head(&info->status_event_wait_q);
 		init_waitqueue_head(&info->event_wait_q);
 		spin_lock_init(&info->irq_spinlock);
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index fc71d9819165..07aa42a7f397 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -261,8 +261,6 @@ struct slgt_info {
 	struct slgt_info *port_array[SLGT_MAX_PORTS];
 
 	int			line;		/* tty line instance number */
-	unsigned short		close_delay;
-	unsigned short		closing_wait;	/* time to wait before closing */
 
 	struct mgsl_icount	icount;
 
@@ -758,9 +756,9 @@ static void close(struct tty_struct *tty, struct file *filp)
 
 	/* wait for transmit data to clear all layers */
 
-	if (info->closing_wait != ASYNC_CLOSING_WAIT_NONE) {
+	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
 		DBGINFO(("%s call tty_wait_until_sent\n", info->device_name));
-		tty_wait_until_sent(tty, info->closing_wait);
+		tty_wait_until_sent(tty, info->port.closing_wait);
 	}
 
  	if (info->port.flags & ASYNC_INITIALIZED)
@@ -774,8 +772,8 @@ static void close(struct tty_struct *tty, struct file *filp)
 	info->port.tty = NULL;
 
 	if (info->port.blocked_open) {
-		if (info->close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->close_delay));
+		if (info->port.close_delay) {
+			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
 		}
 		wake_up_interruptible(&info->port.open_wait);
 	}
@@ -3448,13 +3446,13 @@ static struct slgt_info *alloc_dev(int adapter_num, int port_num, struct pci_dev
 		DBGERR(("%s device alloc failed adapter=%d port=%d\n",
 			driver_name, adapter_num, port_num));
 	} else {
+		tty_port_init(&info->port);
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, bh_handler);
 		info->max_frame_size = 4096;
 		info->raw_rx_size = DMABUFSIZE;
-		info->close_delay = 5*HZ/10;
-		info->closing_wait = 30*HZ;
-		tty_port_init(&info->port);
+		info->port.close_delay = 5*HZ/10;
+		info->port.closing_wait = 30*HZ;
 		init_waitqueue_head(&info->status_event_wait_q);
 		init_waitqueue_head(&info->event_wait_q);
 		spin_lock_init(&info->netlock);
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index 5b5b292d046b..c4bc09018368 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -846,11 +846,11 @@ static void close(struct tty_struct *tty, struct file *filp)
 
 	/* wait for transmit data to clear all layers */
 
-	if (info->closing_wait != ASYNC_CLOSING_WAIT_NONE) {
+	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
 		if (debug_level >= DEBUG_LEVEL_INFO)
 			printk("%s(%d):%s close() calling tty_wait_until_sent\n",
 				 __FILE__,__LINE__, info->device_name );
-		tty_wait_until_sent(tty, info->closing_wait);
+		tty_wait_until_sent(tty, info->port.closing_wait);
 	}
 
  	if (info->port.flags & ASYNC_INITIALIZED)
@@ -866,8 +866,8 @@ static void close(struct tty_struct *tty, struct file *filp)
 	info->port.tty = NULL;
 
 	if (info->port.blocked_open) {
-		if (info->close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->close_delay));
+		if (info->port.close_delay) {
+			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
 		}
 		wake_up_interruptible(&info->port.open_wait);
 	}
@@ -3802,12 +3802,12 @@ static SLMP_INFO *alloc_dev(int adapter_num, int port_num, struct pci_dev *pdev)
 		printk("%s(%d) Error can't allocate device instance data for adapter %d, port %d\n",
 			__FILE__,__LINE__, adapter_num, port_num);
 	} else {
+		tty_port_init(&info->port);
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, bh_handler);
 		info->max_frame_size = 4096;
-		info->close_delay = 5*HZ/10;
-		info->closing_wait = 30*HZ;
-		tty_port_init(&info->port);
+		info->port.close_delay = 5*HZ/10;
+		info->port.closing_wait = 30*HZ;
 		init_waitqueue_head(&info->status_event_wait_q);
 		init_waitqueue_head(&info->event_wait_q);
 		spin_lock_init(&info->netlock);
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index a8cc416a23c0..82f6a8c86332 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -2094,6 +2094,8 @@ void tty_port_init(struct tty_port *port)
 	init_waitqueue_head(&port->open_wait);
 	init_waitqueue_head(&port->close_wait);
 	mutex_init(&port->mutex);
+	port->close_delay = (50 * HZ) / 100;
+	port->closing_wait = (3000 * HZ) / 100;
 }
 EXPORT_SYMBOL(tty_port_init);
 
diff --git a/include/linux/cyclades.h b/include/linux/cyclades.h
index a982b74a6ee6..2d3d1e04ba92 100644
--- a/include/linux/cyclades.h
+++ b/include/linux/cyclades.h
@@ -567,8 +567,6 @@ struct cyclades_port {
 	int			chip_rev;
 	int			custom_divisor;
 	u8			x_char; /* to be pushed out ASAP */
-	int			close_delay;
-	unsigned short		closing_wait;
 	int                     breakon;
 	int                     breakoff;
 	int			xmit_head;
@@ -586,8 +584,8 @@ struct cyclades_port {
 };
 
 #define	CLOSING_WAIT_DELAY	30*HZ
-#define CY_CLOSING_WAIT_NONE	65535
-#define CY_CLOSING_WAIT_INF	0
+#define CY_CLOSING_WAIT_NONE	ASYNC_CLOSING_WAIT_NONE
+#define CY_CLOSING_WAIT_INF	ASYNC_CLOSING_WAIT_INF
 
 
 #define CyMAX_CHIPS_PER_CARD	8
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 46008e86b191..4e5833073aa6 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -188,6 +188,8 @@ struct tty_port {
 	unsigned long		flags;		/* TTY flags ASY_*/
 	struct mutex		mutex;		/* Locking */
 	unsigned char		*xmit_buf;	/* Optional buffer */
+	int			close_delay;	/* Close port delay */
+	int			closing_wait;	/* Delay for output */
 };
 
 /*
-- 
cgit v1.2.3


From f233ea5c9e0d8b95e4283bf6a3436b88f6fd3586 Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Mon, 21 Jul 2008 17:05:22 +1000
Subject: md: Make mddev->array_size sector-based.

This patch renames the array_size field of struct mddev_s to array_sectors
and converts all instances to use units of 512 byte sectors instead of 1k
blocks.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/faulty.c       |  2 +-
 drivers/md/linear.c       |  6 +++---
 drivers/md/md.c           | 12 +++++++-----
 drivers/md/multipath.c    |  2 +-
 drivers/md/raid0.c        |  8 ++++----
 drivers/md/raid1.c        | 11 ++++++-----
 drivers/md/raid10.c       |  2 +-
 drivers/md/raid5.c        | 16 +++++++++-------
 include/linux/raid/md_k.h |  2 +-
 9 files changed, 33 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/faulty.c b/drivers/md/faulty.c
index d107ddceefcd..268547dbfbd3 100644
--- a/drivers/md/faulty.c
+++ b/drivers/md/faulty.c
@@ -297,7 +297,7 @@ static int run(mddev_t *mddev)
 	rdev_for_each(rdev, tmp, mddev)
 		conf->rdev = rdev;
 
-	mddev->array_size = mddev->size;
+	mddev->array_sectors = mddev->size * 2;
 	mddev->private = conf;
 
 	reconfig(mddev, mddev->layout, -1);
diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index ec921f58fbb8..57644a780f16 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -256,7 +256,7 @@ static int linear_run (mddev_t *mddev)
 	if (!conf)
 		return 1;
 	mddev->private = conf;
-	mddev->array_size = conf->array_size;
+	mddev->array_sectors = conf->array_size * 2;
 
 	blk_queue_merge_bvec(mddev->queue, linear_mergeable_bvec);
 	mddev->queue->unplug_fn = linear_unplug;
@@ -290,8 +290,8 @@ static int linear_add(mddev_t *mddev, mdk_rdev_t *rdev)
 	newconf->prev = mddev_to_conf(mddev);
 	mddev->private = newconf;
 	mddev->raid_disks++;
-	mddev->array_size = newconf->array_size;
-	set_capacity(mddev->gendisk, mddev->array_size << 1);
+	mddev->array_sectors = newconf->array_size * 2;
+	set_capacity(mddev->gendisk, mddev->array_sectors);
 	return 0;
 }
 
diff --git a/drivers/md/md.c b/drivers/md/md.c
index df13a17a9627..4bfbc1982cda 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3704,7 +3704,7 @@ static int do_md_run(mddev_t * mddev)
 	if (mddev->flags)
 		md_update_sb(mddev, 0);
 
-	set_capacity(disk, mddev->array_size<<1);
+	set_capacity(disk, mddev->array_sectors);
 
 	/* If we call blk_queue_make_request here, it will
 	 * re-initialise max_sectors etc which may have been
@@ -3905,7 +3905,7 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 
 		export_array(mddev);
 
-		mddev->array_size = 0;
+		mddev->array_sectors = 0;
 		mddev->size = 0;
 		mddev->raid_disks = 0;
 		mddev->recovery_cp = 0;
@@ -4644,7 +4644,8 @@ static int update_size(mddev_t *mddev, sector_t num_sectors)
 		bdev = bdget_disk(mddev->gendisk, 0);
 		if (bdev) {
 			mutex_lock(&bdev->bd_inode->i_mutex);
-			i_size_write(bdev->bd_inode, (loff_t)mddev->array_size << 10);
+			i_size_write(bdev->bd_inode,
+				     (loff_t)mddev->array_sectors << 9);
 			mutex_unlock(&bdev->bd_inode->i_mutex);
 			bdput(bdev);
 		}
@@ -5391,10 +5392,11 @@ static int md_seq_show(struct seq_file *seq, void *v)
 		if (!list_empty(&mddev->disks)) {
 			if (mddev->pers)
 				seq_printf(seq, "\n      %llu blocks",
-					(unsigned long long)mddev->array_size);
+					   (unsigned long long)
+					   mddev->array_sectors / 2);
 			else
 				seq_printf(seq, "\n      %llu blocks",
-					(unsigned long long)size);
+					   (unsigned long long)size);
 		}
 		if (mddev->persistent) {
 			if (mddev->major_version != 0 ||
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index 541cbe3414bd..c4779ccba1c3 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -504,7 +504,7 @@ static int multipath_run (mddev_t *mddev)
 	/*
 	 * Ok, everything is just fine now
 	 */
-	mddev->array_size = mddev->size;
+	mddev->array_sectors = mddev->size * 2;
 
 	mddev->queue->unplug_fn = multipath_unplug;
 	mddev->queue->backing_dev_info.congested_fn = multipath_congested;
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 914c04ddec7c..2f30ebd8b7ab 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -293,16 +293,16 @@ static int raid0_run (mddev_t *mddev)
 		goto out_free_conf;
 
 	/* calculate array device size */
-	mddev->array_size = 0;
+	mddev->array_sectors = 0;
 	rdev_for_each(rdev, tmp, mddev)
-		mddev->array_size += rdev->size;
+		mddev->array_sectors += rdev->size * 2;
 
 	printk("raid0 : md_size is %llu blocks.\n", 
-		(unsigned long long)mddev->array_size);
+		(unsigned long long)mddev->array_sectors / 2);
 	printk("raid0 : conf->hash_spacing is %llu blocks.\n",
 		(unsigned long long)conf->hash_spacing);
 	{
-		sector_t s = mddev->array_size;
+		sector_t s = mddev->array_sectors / 2;
 		sector_t space = conf->hash_spacing;
 		int round;
 		conf->preshift = 0;
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 491dc2d4ad5f..03a5ab705c20 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -2043,7 +2043,7 @@ static int run(mddev_t *mddev)
 	/*
 	 * Ok, everything is just fine now
 	 */
-	mddev->array_size = mddev->size;
+	mddev->array_sectors = mddev->size * 2;
 
 	mddev->queue->unplug_fn = raid1_unplug;
 	mddev->queue->backing_dev_info.congested_fn = raid1_congested;
@@ -2105,14 +2105,15 @@ static int raid1_resize(mddev_t *mddev, sector_t sectors)
 	 * any io in the removed space completes, but it hardly seems
 	 * worth it.
 	 */
-	mddev->array_size = sectors>>1;
-	set_capacity(mddev->gendisk, mddev->array_size << 1);
+	mddev->array_sectors = sectors;
+	set_capacity(mddev->gendisk, mddev->array_sectors);
 	mddev->changed = 1;
-	if (mddev->array_size > mddev->size && mddev->recovery_cp == MaxSector) {
+	if (mddev->array_sectors / 2 > mddev->size &&
+	    mddev->recovery_cp == MaxSector) {
 		mddev->recovery_cp = mddev->size << 1;
 		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	}
-	mddev->size = mddev->array_size;
+	mddev->size = mddev->array_sectors / 2;
 	mddev->resync_max_sectors = sectors;
 	return 0;
 }
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index df08a9fa3a1f..2acea4025243 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -2164,7 +2164,7 @@ static int run(mddev_t *mddev)
 	/*
 	 * Ok, everything is just fine now
 	 */
-	mddev->array_size = size << (conf->chunk_shift-1);
+	mddev->array_sectors = size << conf->chunk_shift;
 	mddev->resync_max_sectors = size << conf->chunk_shift;
 
 	mddev->queue->unplug_fn = raid10_unplug;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 8f4c70a53210..42a480ba767b 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3540,7 +3540,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
 			    j == raid6_next_disk(sh->pd_idx, sh->disks))
 				continue;
 			s = compute_blocknr(sh, j);
-			if (s < (mddev->array_size<<1)) {
+			if (s < mddev->array_sectors) {
 				skipped = 1;
 				continue;
 			}
@@ -4189,7 +4189,7 @@ static int run(mddev_t *mddev)
 	mddev->queue->backing_dev_info.congested_data = mddev;
 	mddev->queue->backing_dev_info.congested_fn = raid5_congested;
 
-	mddev->array_size =  mddev->size * (conf->previous_raid_disks -
+	mddev->array_sectors = 2 * mddev->size * (conf->previous_raid_disks -
 					    conf->max_degraded);
 
 	blk_queue_merge_bvec(mddev->queue, raid5_mergeable_bvec);
@@ -4413,8 +4413,9 @@ static int raid5_resize(mddev_t *mddev, sector_t sectors)
 	raid5_conf_t *conf = mddev_to_conf(mddev);
 
 	sectors &= ~((sector_t)mddev->chunk_size/512 - 1);
-	mddev->array_size = (sectors * (mddev->raid_disks-conf->max_degraded))>>1;
-	set_capacity(mddev->gendisk, mddev->array_size << 1);
+	mddev->array_sectors = sectors * (mddev->raid_disks
+					  - conf->max_degraded);
+	set_capacity(mddev->gendisk, mddev->array_sectors);
 	mddev->changed = 1;
 	if (sectors/2  > mddev->size && mddev->recovery_cp == MaxSector) {
 		mddev->recovery_cp = mddev->size << 1;
@@ -4547,15 +4548,16 @@ static void end_reshape(raid5_conf_t *conf)
 	struct block_device *bdev;
 
 	if (!test_bit(MD_RECOVERY_INTR, &conf->mddev->recovery)) {
-		conf->mddev->array_size = conf->mddev->size *
+		conf->mddev->array_sectors = 2 * conf->mddev->size *
 			(conf->raid_disks - conf->max_degraded);
-		set_capacity(conf->mddev->gendisk, conf->mddev->array_size << 1);
+		set_capacity(conf->mddev->gendisk, conf->mddev->array_sectors);
 		conf->mddev->changed = 1;
 
 		bdev = bdget_disk(conf->mddev->gendisk, 0);
 		if (bdev) {
 			mutex_lock(&bdev->bd_inode->i_mutex);
-			i_size_write(bdev->bd_inode, (loff_t)conf->mddev->array_size << 10);
+			i_size_write(bdev->bd_inode,
+				     (loff_t)conf->mddev->array_sectors << 9);
 			mutex_unlock(&bdev->bd_inode->i_mutex);
 			bdput(bdev);
 		}
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index e37aaa41efc6..6f72b47ae41c 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -150,7 +150,7 @@ struct mddev_s
 	int				raid_disks;
 	int				max_disks;
 	sector_t			size; /* used size of component devices */
-	sector_t			array_size; /* exported array size */
+	sector_t			array_sectors; /* exported array size */
 	__u64				events;
 
 	char				uuid[16];
-- 
cgit v1.2.3


From d6e2215052810678bc9782fd980b52706fc71f50 Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Mon, 21 Jul 2008 17:05:25 +1000
Subject: md: linear: Make array_size sector-based and rename it to
 array_sectors.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/linear.c         | 16 ++++++++--------
 include/linux/raid/linear.h |  2 +-
 2 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 57644a780f16..1cafaa959443 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -120,7 +120,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 		return NULL;
 
 	cnt = 0;
-	conf->array_size = 0;
+	conf->array_sectors = 0;
 
 	rdev_for_each(rdev, tmp, mddev) {
 		int j = rdev->raid_disk;
@@ -144,7 +144,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 			blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);
 
 		disk->size = rdev->size;
-		conf->array_size += rdev->size;
+		conf->array_sectors += rdev->size * 2;
 
 		cnt++;
 	}
@@ -153,7 +153,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 		goto out;
 	}
 
-	min_spacing = conf->array_size;
+	min_spacing = conf->array_sectors / 2;
 	sector_div(min_spacing, PAGE_SIZE/sizeof(struct dev_info *));
 
 	/* min_spacing is the minimum spacing that will fit the hash
@@ -162,7 +162,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 	 * that is larger than min_spacing as use the size of that as
 	 * the actual spacing
 	 */
-	conf->hash_spacing = conf->array_size;
+	conf->hash_spacing = conf->array_sectors / 2;
 	for (i=0; i < cnt-1 ; i++) {
 		sector_t sz = 0;
 		int j;
@@ -192,7 +192,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 		unsigned round;
 		unsigned long base;
 
-		sz = conf->array_size >> conf->preshift;
+		sz = conf->array_sectors >> (conf->preshift + 1);
 		sz += 1; /* force round-up */
 		base = conf->hash_spacing >> conf->preshift;
 		round = sector_div(sz, base);
@@ -219,7 +219,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 	curr_offset = 0;
 	i = 0;
 	for (curr_offset = 0;
-	     curr_offset < conf->array_size;
+	     curr_offset < conf->array_sectors / 2;
 	     curr_offset += conf->hash_spacing) {
 
 		while (i < raid_disks-1 &&
@@ -256,7 +256,7 @@ static int linear_run (mddev_t *mddev)
 	if (!conf)
 		return 1;
 	mddev->private = conf;
-	mddev->array_sectors = conf->array_size * 2;
+	mddev->array_sectors = conf->array_sectors;
 
 	blk_queue_merge_bvec(mddev->queue, linear_mergeable_bvec);
 	mddev->queue->unplug_fn = linear_unplug;
@@ -290,7 +290,7 @@ static int linear_add(mddev_t *mddev, mdk_rdev_t *rdev)
 	newconf->prev = mddev_to_conf(mddev);
 	mddev->private = newconf;
 	mddev->raid_disks++;
-	mddev->array_sectors = newconf->array_size * 2;
+	mddev->array_sectors = newconf->array_sectors;
 	set_capacity(mddev->gendisk, mddev->array_sectors);
 	return 0;
 }
diff --git a/include/linux/raid/linear.h b/include/linux/raid/linear.h
index ba15469daf11..7e375111d007 100644
--- a/include/linux/raid/linear.h
+++ b/include/linux/raid/linear.h
@@ -16,7 +16,7 @@ struct linear_private_data
 	struct linear_private_data *prev;	/* earlier version */
 	dev_info_t		**hash_table;
 	sector_t		hash_spacing;
-	sector_t		array_size;
+	sector_t		array_sectors;
 	int			preshift; /* shift before dividing by hash_spacing */
 	dev_info_t		disks[0];
 };
-- 
cgit v1.2.3


From f2ea68cf42aafdd93393b6b8b20fc3c2b5f4390c Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Mon, 21 Jul 2008 17:05:25 +1000
Subject: md: only count actual openers as access which prevent a 'stop'

Open isn't the only thing that increments ->active.  e.g. reading
/proc/mdstat will increment it briefly.  So to avoid false positives
in testing for concurrent access, introduce a new counter that counts
just the number of times the md device it open.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c           | 9 ++++++---
 include/linux/raid/md_k.h | 3 ++-
 2 files changed, 8 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 4bfbc1982cda..450f30b6edf9 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -273,6 +273,7 @@ static mddev_t * mddev_find(dev_t unit)
 	INIT_LIST_HEAD(&new->all_mddevs);
 	init_timer(&new->safemode_timer);
 	atomic_set(&new->active, 1);
+	atomic_set(&new->openers, 0);
 	spin_lock_init(&new->write_lock);
 	init_waitqueue_head(&new->sb_wait);
 	init_waitqueue_head(&new->recovery_wait);
@@ -2695,14 +2696,14 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len)
 		break;
 	case clear:
 		/* stopping an active array */
-		if (atomic_read(&mddev->active) > 1)
+		if (atomic_read(&mddev->openers) > 0)
 			return -EBUSY;
 		err = do_md_stop(mddev, 0, 0);
 		break;
 	case inactive:
 		/* stopping an active array */
 		if (mddev->pers) {
-			if (atomic_read(&mddev->active) > 1)
+			if (atomic_read(&mddev->openers) > 0)
 				return -EBUSY;
 			err = do_md_stop(mddev, 2, 0);
 		} else
@@ -3816,7 +3817,7 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 	int err = 0;
 	struct gendisk *disk = mddev->gendisk;
 
-	if (atomic_read(&mddev->active) > 1 + is_open) {
+	if (atomic_read(&mddev->openers) > is_open) {
 		printk("md: %s still in use.\n",mdname(mddev));
 		return -EBUSY;
 	}
@@ -5014,6 +5015,7 @@ static int md_open(struct inode *inode, struct file *file)
 
 	err = 0;
 	mddev_get(mddev);
+	atomic_inc(&mddev->openers);
 	mddev_unlock(mddev);
 
 	check_disk_change(inode->i_bdev);
@@ -5026,6 +5028,7 @@ static int md_release(struct inode *inode, struct file * file)
  	mddev_t *mddev = inode->i_bdev->bd_disk->private_data;
 
 	BUG_ON(!mddev);
+	atomic_dec(&mddev->openers);
 	mddev_put(mddev);
 
 	return 0;
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 6f72b47ae41c..4bef4791d80d 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -215,7 +215,8 @@ struct mddev_s
 
 	int				in_sync;	/* know to not need resync */
 	struct mutex			reconfig_mutex;
-	atomic_t			active;
+	atomic_t			active;		/* general refcount */
+	atomic_t			openers;	/* number of active opens */
 
 	int				changed;	/* true if we might need to reread partition info */
 	int				degraded;	/* whether md should consider
-- 
cgit v1.2.3


From 4b80991c6cb9efa607bc4fd6f3ecdf5511c31bb0 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Mon, 21 Jul 2008 17:05:25 +1000
Subject: md: Protect access to mddev->disks list using RCU

All modifications and most access to the mddev->disks list are made
under the reconfig_mutex lock.  However there are three places where
the list is walked without any locking.  If a reconfig happens at this
time, havoc (and oops) can ensue.

So use RCU to protect these accesses:
  - wrap them in rcu_read_{,un}lock()
  - use list_for_each_entry_rcu
  - add to the list with list_add_rcu
  - delete from the list with list_del_rcu
  - delay the 'free' with call_rcu rather than schedule_work

Note that export_rdev did a list_del_init on this list.  In almost all
cases the entry was not in the list anymore so it was a no-op and so
safe.  It is no longer safe as after list_del_rcu we may not touch
the list_head.
An audit shows that export_rdev is called:
  - after unbind_rdev_from_array, in which case the delete has
     already been done,
  - after bind_rdev_to_array fails, in which case the delete isn't needed.
  - before the device has been put on a list at all (e.g. in
      add_new_disk where reading the superblock fails).
  - and in autorun devices after a failure when the device is on a
      different list.

So remove the list_del_init call from export_rdev, and add it back
immediately before the called to export_rdev for that last case.

Note also that ->same_set is sometimes used for lists other than
mddev->list (e.g. candidates).  In these cases rcu is not needed.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/bitmap.c       | 15 ++++++++++-----
 drivers/md/md.c           | 30 ++++++++++++++++++------------
 include/linux/raid/md_k.h |  3 +++
 3 files changed, 31 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index eba83e25b678..621a272a2c74 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -241,10 +241,10 @@ static struct page *read_sb_page(mddev_t *mddev, long offset, unsigned long inde
 static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
 {
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	mddev_t *mddev = bitmap->mddev;
 
-	rdev_for_each(rdev, tmp, mddev)
+	rcu_read_lock();
+	rdev_for_each_rcu(rdev, mddev)
 		if (test_bit(In_sync, &rdev->flags)
 		    && !test_bit(Faulty, &rdev->flags)) {
 			int size = PAGE_SIZE;
@@ -260,11 +260,11 @@ static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
 				    + (long)(page->index * (PAGE_SIZE/512))
 				    + size/512 > 0)
 					/* bitmap runs in to metadata */
-					return -EINVAL;
+					goto bad_alignment;
 				if (rdev->data_offset + mddev->size*2
 				    > rdev->sb_start + bitmap->offset)
 					/* data runs in to bitmap */
-					return -EINVAL;
+					goto bad_alignment;
 			} else if (rdev->sb_start < rdev->data_offset) {
 				/* METADATA BITMAP DATA */
 				if (rdev->sb_start
@@ -272,7 +272,7 @@ static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
 				    + page->index*(PAGE_SIZE/512) + size/512
 				    > rdev->data_offset)
 					/* bitmap runs in to data */
-					return -EINVAL;
+					goto bad_alignment;
 			} else {
 				/* DATA METADATA BITMAP - no problems */
 			}
@@ -282,10 +282,15 @@ static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
 				       size,
 				       page);
 		}
+	rcu_read_unlock();
 
 	if (wait)
 		md_super_wait(mddev);
 	return 0;
+
+ bad_alignment:
+	rcu_read_unlock();
+	return -EINVAL;
 }
 
 static void bitmap_file_kick(struct bitmap *bitmap);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 450f30b6edf9..c2ff77ccec50 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1395,15 +1395,17 @@ static struct super_type super_types[] = {
 
 static int match_mddev_units(mddev_t *mddev1, mddev_t *mddev2)
 {
-	struct list_head *tmp, *tmp2;
 	mdk_rdev_t *rdev, *rdev2;
 
-	rdev_for_each(rdev, tmp, mddev1)
-		rdev_for_each(rdev2, tmp2, mddev2)
+	rcu_read_lock();
+	rdev_for_each_rcu(rdev, mddev1)
+		rdev_for_each_rcu(rdev2, mddev2)
 			if (rdev->bdev->bd_contains ==
-			    rdev2->bdev->bd_contains)
+			    rdev2->bdev->bd_contains) {
+				rcu_read_unlock();
 				return 1;
-
+			}
+	rcu_read_unlock();
 	return 0;
 }
 
@@ -1470,7 +1472,7 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 		kobject_del(&rdev->kobj);
 		goto fail;
 	}
-	list_add(&rdev->same_set, &mddev->disks);
+	list_add_rcu(&rdev->same_set, &mddev->disks);
 	bd_claim_by_disk(rdev->bdev, rdev->bdev->bd_holder, mddev->gendisk);
 	return 0;
 
@@ -1495,14 +1497,16 @@ static void unbind_rdev_from_array(mdk_rdev_t * rdev)
 		return;
 	}
 	bd_release_from_disk(rdev->bdev, rdev->mddev->gendisk);
-	list_del_init(&rdev->same_set);
+	list_del_rcu(&rdev->same_set);
 	printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b));
 	rdev->mddev = NULL;
 	sysfs_remove_link(&rdev->kobj, "block");
 
 	/* We need to delay this, otherwise we can deadlock when
-	 * writing to 'remove' to "dev/state"
+	 * writing to 'remove' to "dev/state".  We also need
+	 * to delay it due to rcu usage.
 	 */
+	synchronize_rcu();
 	INIT_WORK(&rdev->del_work, md_delayed_delete);
 	kobject_get(&rdev->kobj);
 	schedule_work(&rdev->del_work);
@@ -1558,7 +1562,6 @@ static void export_rdev(mdk_rdev_t * rdev)
 	if (rdev->mddev)
 		MD_BUG();
 	free_disk_sb(rdev);
-	list_del_init(&rdev->same_set);
 #ifndef MODULE
 	if (test_bit(AutoDetected, &rdev->flags))
 		md_autodetect_dev(rdev->bdev->bd_dev);
@@ -4062,8 +4065,10 @@ static void autorun_devices(int part)
 		/* on success, candidates will be empty, on error
 		 * it won't...
 		 */
-		rdev_for_each_list(rdev, tmp, candidates)
+		rdev_for_each_list(rdev, tmp, candidates) {
+			list_del_init(&rdev->same_set);
 			export_rdev(rdev);
+		}
 		mddev_put(mddev);
 	}
 	printk(KERN_INFO "md: ... autorun DONE.\n");
@@ -5529,12 +5534,12 @@ int unregister_md_personality(struct mdk_personality *p)
 static int is_mddev_idle(mddev_t *mddev)
 {
 	mdk_rdev_t * rdev;
-	struct list_head *tmp;
 	int idle;
 	long curr_events;
 
 	idle = 1;
-	rdev_for_each(rdev, tmp, mddev) {
+	rcu_read_lock();
+	rdev_for_each_rcu(rdev, mddev) {
 		struct gendisk *disk = rdev->bdev->bd_contains->bd_disk;
 		curr_events = disk_stat_read(disk, sectors[0]) + 
 				disk_stat_read(disk, sectors[1]) - 
@@ -5566,6 +5571,7 @@ static int is_mddev_idle(mddev_t *mddev)
 			idle = 0;
 		}
 	}
+	rcu_read_unlock();
 	return idle;
 }
 
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 4bef4791d80d..9f2549ac0e2d 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -339,6 +339,9 @@ static inline char * mdname (mddev_t * mddev)
 #define rdev_for_each(rdev, tmp, mddev)				\
 	rdev_for_each_list(rdev, tmp, (mddev)->disks)
 
+#define rdev_for_each_rcu(rdev, mddev)				\
+	list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)
+
 typedef struct mdk_thread_s {
 	void			(*run) (mddev_t *mddev);
 	mddev_t			*mddev;
-- 
cgit v1.2.3


From f6fccb1213ba3d661baeb2a5eee0a9701dc03e1b Mon Sep 17 00:00:00 2001
From: Milan Broz <mbroz@redhat.com>
Date: Mon, 21 Jul 2008 12:00:37 +0100
Subject: dm: introduce merge_bvec_fn

Introduce a bvec merge function for device mapper devices
for dynamic size restrictions.

This code ensures the requested biovec lies within a single
target and then calls a target-specific function to check
against any constraints imposed by underlying devices.

Signed-off-by: Milan Broz <mbroz@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c               | 44 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/device-mapper.h |  6 ++++++
 include/linux/dm-ioctl.h      |  4 ++--
 3 files changed, 52 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index efe969074928..bca448e11878 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -829,6 +829,49 @@ static int __split_bio(struct mapped_device *md, struct bio *bio)
  * CRUD END
  *---------------------------------------------------------------*/
 
+static int dm_merge_bvec(struct request_queue *q,
+			 struct bvec_merge_data *bvm,
+			 struct bio_vec *biovec)
+{
+	struct mapped_device *md = q->queuedata;
+	struct dm_table *map = dm_get_table(md);
+	struct dm_target *ti;
+	sector_t max_sectors;
+	int max_size;
+
+	if (unlikely(!map))
+		return 0;
+
+	ti = dm_table_find_target(map, bvm->bi_sector);
+
+	/*
+	 * Find maximum amount of I/O that won't need splitting
+	 */
+	max_sectors = min(max_io_len(md, bvm->bi_sector, ti),
+			  (sector_t) BIO_MAX_SECTORS);
+	max_size = (max_sectors << SECTOR_SHIFT) - bvm->bi_size;
+	if (max_size < 0)
+		max_size = 0;
+
+	/*
+	 * merge_bvec_fn() returns number of bytes
+	 * it can accept at this offset
+	 * max is precomputed maximal io size
+	 */
+	if (max_size && ti->type->merge)
+		max_size = ti->type->merge(ti, bvm, biovec, max_size);
+
+	/*
+	 * Always allow an entire first page
+	 */
+	if (max_size <= biovec->bv_len && !(bvm->bi_size >> SECTOR_SHIFT))
+		max_size = biovec->bv_len;
+
+	dm_table_put(map);
+
+	return max_size;
+}
+
 /*
  * The request function that just remaps the bio built up by
  * dm_merge_bvec.
@@ -1032,6 +1075,7 @@ static struct mapped_device *alloc_dev(int minor)
 	blk_queue_make_request(md->queue, dm_request);
 	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
 	md->queue->unplug_fn = dm_unplug_all;
+	blk_queue_merge_bvec(md->queue, dm_merge_bvec);
 
 	md->io_pool = mempool_create_slab_pool(MIN_IOS, _io_cache);
 	if (!md->io_pool)
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 0d8d419d191a..a90222e3297d 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -9,11 +9,13 @@
 #define _LINUX_DEVICE_MAPPER_H
 
 #include <linux/bio.h>
+#include <linux/blkdev.h>
 
 struct dm_target;
 struct dm_table;
 struct dm_dev;
 struct mapped_device;
+struct bio_vec;
 
 typedef enum { STATUSTYPE_INFO, STATUSTYPE_TABLE } status_type_t;
 
@@ -72,6 +74,9 @@ typedef int (*dm_ioctl_fn) (struct dm_target *ti, struct inode *inode,
 			    struct file *filp, unsigned int cmd,
 			    unsigned long arg);
 
+typedef int (*dm_merge_fn) (struct dm_target *ti, struct bvec_merge_data *bvm,
+			    struct bio_vec *biovec, int max_size);
+
 void dm_error(const char *message);
 
 /*
@@ -107,6 +112,7 @@ struct target_type {
 	dm_status_fn status;
 	dm_message_fn message;
 	dm_ioctl_fn ioctl;
+	dm_merge_fn merge;
 };
 
 struct io_restrictions {
diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h
index b03c41bbfa14..28c2940eb30d 100644
--- a/include/linux/dm-ioctl.h
+++ b/include/linux/dm-ioctl.h
@@ -256,9 +256,9 @@ enum {
 #define DM_DEV_SET_GEOMETRY	_IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl)
 
 #define DM_VERSION_MAJOR	4
-#define DM_VERSION_MINOR	13
+#define DM_VERSION_MINOR	14
 #define DM_VERSION_PATCHLEVEL	0
-#define DM_VERSION_EXTRA	"-ioctl (2007-10-18)"
+#define DM_VERSION_EXTRA	"-ioctl (2008-04-23)"
 
 /* Status bits */
 #define DM_READONLY_FLAG	(1 << 0) /* In/Out */
-- 
cgit v1.2.3


From 584015727a3b88b46602b20077b46cd04f8b4ab3 Mon Sep 17 00:00:00 2001
From: Krzysztof Piotr Oledzki <ole@ans.pl>
Date: Mon, 21 Jul 2008 10:01:34 -0700
Subject: netfilter: accounting rework: ct_extend + 64bit counters (v4)

Initially netfilter has had 64bit counters for conntrack-based accounting, but
it was changed in 2.6.14 to save memory. Unfortunately in-kernel 64bit counters are
still required, for example for "connbytes" extension. However, 64bit counters
waste a lot of memory and it was not possible to enable/disable it runtime.

This patch:
 - reimplements accounting with respect to the extension infrastructure,
 - makes one global version of seq_print_acct() instead of two seq_print_counters(),
 - makes it possible to enable it at boot time (for CONFIG_SYSCTL/CONFIG_SYSFS=n),
 - makes it possible to enable/disable it at runtime by sysctl or sysfs,
 - extends counters from 32bit to 64bit,
 - renames ip_conntrack_counter -> nf_conn_counter,
 - enables accounting code unconditionally (no longer depends on CONFIG_NF_CT_ACCT),
 - set initial accounting enable state based on CONFIG_NF_CT_ACCT
 - removes buggy IPCT_COUNTER_FILLING event handling.

If accounting is enabled newly created connections get additional acct extend.
Old connections are not changed as it is not possible to add a ct_extend area
to confirmed conntrack. Accounting is performed for all connections with
acct extend regardless of a current state of "net.netfilter.nf_conntrack_acct".

Signed-off-by: Krzysztof Piotr Oledzki <ole@ans.pl>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/feature-removal-schedule.txt         |  10 ++
 Documentation/kernel-parameters.txt                |   7 ++
 include/linux/netfilter/nf_conntrack_common.h      |   8 +-
 include/linux/netfilter/nfnetlink_conntrack.h      |   8 +-
 include/net/netfilter/nf_conntrack.h               |   6 --
 include/net/netfilter/nf_conntrack_acct.h          |  51 ++++++++++
 include/net/netfilter/nf_conntrack_extend.h        |   2 +
 .../netfilter/nf_conntrack_l3proto_ipv4_compat.c   |  18 +---
 net/netfilter/Kconfig                              |   9 ++
 net/netfilter/Makefile                             |   2 +-
 net/netfilter/nf_conntrack_acct.c                  | 104 +++++++++++++++++++++
 net/netfilter/nf_conntrack_core.c                  |  39 +++++---
 net/netfilter/nf_conntrack_netlink.c               |  44 +++++----
 net/netfilter/nf_conntrack_standalone.c            |  18 +---
 net/netfilter/xt_connbytes.c                       |   8 +-
 15 files changed, 248 insertions(+), 86 deletions(-)
 create mode 100644 include/net/netfilter/nf_conntrack_acct.h
 create mode 100644 net/netfilter/nf_conntrack_acct.c

(limited to 'include/linux')

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 86334b6f8238..9f73587219e8 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -336,3 +336,13 @@ When:	After the only user (hal) has seen a release with the patches
 Why:	Over 1K .text/.data size reduction, data is available in other
 	ways (ioctls)
 Who:	Johannes Berg <johannes@sipsolutions.net>
+
+---------------------------
+
+What: CONFIG_NF_CT_ACCT
+When: 2.6.29
+Why:  Accounting can now be enabled/disabled without kernel recompilation.
+      Currently used only to set a default value for a feature that is also
+      controlled by a kernel/module/sysfs/sysctl parameter.
+Who:  Krzysztof Piotr Oledzki <ole@ans.pl>
+
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 09ad7450647b..e4ef27584408 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1279,6 +1279,13 @@ and is between 256 and 4096 characters. It is defined in the file
 			This usage is only documented in each driver source
 			file if at all.
 
+	nf_conntrack.acct=
+			[NETFILTER] Enable connection tracking flow accounting
+			0 to disable accounting
+			1 to enable accounting
+			Default value depends on CONFIG_NF_CT_ACCT that is
+			going to be removed in 2.6.29.
+
 	nfsaddrs=	[NFS]
 			See Documentation/filesystems/nfsroot.txt.
 
diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h
index bad1eb760f61..885cbe282260 100644
--- a/include/linux/netfilter/nf_conntrack_common.h
+++ b/include/linux/netfilter/nf_conntrack_common.h
@@ -122,7 +122,7 @@ enum ip_conntrack_events
 	IPCT_NATINFO_BIT = 10,
 	IPCT_NATINFO = (1 << IPCT_NATINFO_BIT),
 
-	/* Counter highest bit has been set */
+	/* Counter highest bit has been set, unused */
 	IPCT_COUNTER_FILLING_BIT = 11,
 	IPCT_COUNTER_FILLING = (1 << IPCT_COUNTER_FILLING_BIT),
 
@@ -145,12 +145,6 @@ enum ip_conntrack_expect_events {
 };
 
 #ifdef __KERNEL__
-struct ip_conntrack_counter
-{
-	u_int32_t packets;
-	u_int32_t bytes;
-};
-
 struct ip_conntrack_stat
 {
 	unsigned int searched;
diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h
index 759bc043dc65..c19595c89304 100644
--- a/include/linux/netfilter/nfnetlink_conntrack.h
+++ b/include/linux/netfilter/nfnetlink_conntrack.h
@@ -115,10 +115,10 @@ enum ctattr_protoinfo_sctp {
 
 enum ctattr_counters {
 	CTA_COUNTERS_UNSPEC,
-	CTA_COUNTERS_PACKETS,		/* old 64bit counters */
-	CTA_COUNTERS_BYTES,		/* old 64bit counters */
-	CTA_COUNTERS32_PACKETS,
-	CTA_COUNTERS32_BYTES,
+	CTA_COUNTERS_PACKETS,		/* 64bit counters */
+	CTA_COUNTERS_BYTES,		/* 64bit counters */
+	CTA_COUNTERS32_PACKETS,		/* old 32bit counters, unused */
+	CTA_COUNTERS32_BYTES,		/* old 32bit counters, unused */
 	__CTA_COUNTERS_MAX
 };
 #define CTA_COUNTERS_MAX (__CTA_COUNTERS_MAX - 1)
diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index 8f5b75734dd0..0741ad592da0 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -88,7 +88,6 @@ struct nf_conn_help {
 	u8 expecting[NF_CT_MAX_EXPECT_CLASSES];
 };
 
-
 #include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
 #include <net/netfilter/ipv6/nf_conntrack_ipv6.h>
 
@@ -111,11 +110,6 @@ struct nf_conn
 	/* Timer function; drops refcnt when it goes off. */
 	struct timer_list timeout;
 
-#ifdef CONFIG_NF_CT_ACCT
-	/* Accounting Information (same cache line as other written members) */
-	struct ip_conntrack_counter counters[IP_CT_DIR_MAX];
-#endif
-
 #if defined(CONFIG_NF_CONNTRACK_MARK)
 	u_int32_t mark;
 #endif
diff --git a/include/net/netfilter/nf_conntrack_acct.h b/include/net/netfilter/nf_conntrack_acct.h
new file mode 100644
index 000000000000..5d5ae55d54c4
--- /dev/null
+++ b/include/net/netfilter/nf_conntrack_acct.h
@@ -0,0 +1,51 @@
+/*
+ * (C) 2008 Krzysztof Piotr Oledzki <ole@ans.pl>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef _NF_CONNTRACK_ACCT_H
+#define _NF_CONNTRACK_ACCT_H
+#include <linux/netfilter/nf_conntrack_common.h>
+#include <linux/netfilter/nf_conntrack_tuple_common.h>
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_extend.h>
+
+struct nf_conn_counter {
+	u_int64_t packets;
+	u_int64_t bytes;
+};
+
+extern int nf_ct_acct;
+
+static inline
+struct nf_conn_counter *nf_conn_acct_find(const struct nf_conn *ct)
+{
+	return nf_ct_ext_find(ct, NF_CT_EXT_ACCT);
+}
+
+static inline
+struct nf_conn_counter *nf_ct_acct_ext_add(struct nf_conn *ct, gfp_t gfp)
+{
+	struct nf_conn_counter *acct;
+
+	if (!nf_ct_acct)
+		return NULL;
+
+	acct = nf_ct_ext_add(ct, NF_CT_EXT_ACCT, gfp);
+	if (!acct)
+		pr_debug("failed to add accounting extension area");
+
+
+	return acct;
+};
+
+extern unsigned int
+seq_print_acct(struct seq_file *s, const struct nf_conn *ct, int dir);
+
+extern int nf_conntrack_acct_init(void);
+extern void nf_conntrack_acct_fini(void);
+
+#endif /* _NF_CONNTRACK_ACCT_H */
diff --git a/include/net/netfilter/nf_conntrack_extend.h b/include/net/netfilter/nf_conntrack_extend.h
index f80c0ed6d870..da8ee52613a5 100644
--- a/include/net/netfilter/nf_conntrack_extend.h
+++ b/include/net/netfilter/nf_conntrack_extend.h
@@ -7,11 +7,13 @@ enum nf_ct_ext_id
 {
 	NF_CT_EXT_HELPER,
 	NF_CT_EXT_NAT,
+	NF_CT_EXT_ACCT,
 	NF_CT_EXT_NUM,
 };
 
 #define NF_CT_EXT_HELPER_TYPE struct nf_conn_help
 #define NF_CT_EXT_NAT_TYPE struct nf_conn_nat
+#define NF_CT_EXT_ACCT_TYPE struct nf_conn_counter
 
 /* Extensions: optional stuff which isn't permanently in struct. */
 struct nf_ct_ext {
diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
index 40a46d482490..3a020720e40b 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c
@@ -18,19 +18,7 @@
 #include <net/netfilter/nf_conntrack_l3proto.h>
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_expect.h>
-
-#ifdef CONFIG_NF_CT_ACCT
-static unsigned int
-seq_print_counters(struct seq_file *s,
-		   const struct ip_conntrack_counter *counter)
-{
-	return seq_printf(s, "packets=%llu bytes=%llu ",
-			  (unsigned long long)counter->packets,
-			  (unsigned long long)counter->bytes);
-}
-#else
-#define seq_print_counters(x, y)	0
-#endif
+#include <net/netfilter/nf_conntrack_acct.h>
 
 struct ct_iter_state {
 	unsigned int bucket;
@@ -127,7 +115,7 @@ static int ct_seq_show(struct seq_file *s, void *v)
 			l3proto, l4proto))
 		return -ENOSPC;
 
-	if (seq_print_counters(s, &ct->counters[IP_CT_DIR_ORIGINAL]))
+	if (seq_print_acct(s, ct, IP_CT_DIR_ORIGINAL))
 		return -ENOSPC;
 
 	if (!(test_bit(IPS_SEEN_REPLY_BIT, &ct->status)))
@@ -138,7 +126,7 @@ static int ct_seq_show(struct seq_file *s, void *v)
 			l3proto, l4proto))
 		return -ENOSPC;
 
-	if (seq_print_counters(s, &ct->counters[IP_CT_DIR_REPLY]))
+	if (seq_print_acct(s, ct, IP_CT_DIR_REPLY))
 		return -ENOSPC;
 
 	if (test_bit(IPS_ASSURED_BIT, &ct->status))
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 316c7af1d2b1..ee898e74808d 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -49,6 +49,15 @@ config NF_CT_ACCT
 	  Those counters can be used for flow-based accounting or the
 	  `connbytes' match.
 
+	  Please note that currently this option only sets a default state.
+	  You may change it at boot time with nf_conntrack.acct=0/1 kernel
+	  paramater or by loading the nf_conntrack module with acct=0/1.
+
+	  You may also disable/enable it on a running system with:
+	   sysctl net.netfilter.nf_conntrack_acct=0/1
+
+	  This option will be removed in 2.6.29.
+
 	  If unsure, say `N'.
 
 config NF_CONNTRACK_MARK
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 5c4b183f6422..3bd2cc556aea 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -1,6 +1,6 @@
 netfilter-objs := core.o nf_log.o nf_queue.o nf_sockopt.o
 
-nf_conntrack-y	:= nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_proto.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o nf_conntrack_extend.o
+nf_conntrack-y	:= nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_proto.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o nf_conntrack_extend.o nf_conntrack_acct.o
 nf_conntrack-$(CONFIG_NF_CONNTRACK_EVENTS) += nf_conntrack_ecache.o
 
 obj-$(CONFIG_NETFILTER) = netfilter.o
diff --git a/net/netfilter/nf_conntrack_acct.c b/net/netfilter/nf_conntrack_acct.c
new file mode 100644
index 000000000000..59bd8b903a19
--- /dev/null
+++ b/net/netfilter/nf_conntrack_acct.c
@@ -0,0 +1,104 @@
+/* Accouting handling for netfilter. */
+
+/*
+ * (C) 2008 Krzysztof Piotr Oledzki <ole@ans.pl>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/netfilter.h>
+#include <linux/kernel.h>
+#include <linux/moduleparam.h>
+
+#include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_extend.h>
+#include <net/netfilter/nf_conntrack_acct.h>
+
+#ifdef CONFIG_NF_CT_ACCT
+#define NF_CT_ACCT_DEFAULT 1
+#else
+#define NF_CT_ACCT_DEFAULT 0
+#endif
+
+int nf_ct_acct __read_mostly = NF_CT_ACCT_DEFAULT;
+EXPORT_SYMBOL_GPL(nf_ct_acct);
+
+module_param_named(acct, nf_ct_acct, bool, 0644);
+MODULE_PARM_DESC(acct, "Enable connection tracking flow accounting.");
+
+#ifdef CONFIG_SYSCTL
+static struct ctl_table_header *acct_sysctl_header;
+static struct ctl_table acct_sysctl_table[] = {
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nf_conntrack_acct",
+		.data		= &nf_ct_acct,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{}
+};
+#endif /* CONFIG_SYSCTL */
+
+unsigned int
+seq_print_acct(struct seq_file *s, const struct nf_conn *ct, int dir)
+{
+	struct nf_conn_counter *acct;
+
+	acct = nf_conn_acct_find(ct);
+	if (!acct)
+		return 0;
+
+	return seq_printf(s, "packets=%llu bytes=%llu ",
+			  (unsigned long long)acct[dir].packets,
+			  (unsigned long long)acct[dir].bytes);
+};
+EXPORT_SYMBOL_GPL(seq_print_acct);
+
+static struct nf_ct_ext_type acct_extend __read_mostly = {
+	.len	= sizeof(struct nf_conn_counter[IP_CT_DIR_MAX]),
+	.align	= __alignof__(struct nf_conn_counter[IP_CT_DIR_MAX]),
+	.id	= NF_CT_EXT_ACCT,
+};
+
+int nf_conntrack_acct_init(void)
+{
+	int ret;
+
+#ifdef CONFIG_NF_CT_ACCT
+	printk(KERN_WARNING "CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Plase use\n");
+	printk(KERN_WARNING "nf_conntrack.acct=1 kernel paramater, acct=1 nf_conntrack module option or\n");
+	printk(KERN_WARNING "sysctl net.netfilter.nf_conntrack_acct=1 to enable it.\n");
+#endif
+
+	ret = nf_ct_extend_register(&acct_extend);
+	if (ret < 0) {
+		printk(KERN_ERR "nf_conntrack_acct: Unable to register extension\n");
+		return ret;
+	}
+
+#ifdef CONFIG_SYSCTL
+	acct_sysctl_header = register_sysctl_paths(nf_net_netfilter_sysctl_path,
+				acct_sysctl_table);
+
+	if (!acct_sysctl_header) {
+		nf_ct_extend_unregister(&acct_extend);
+
+		printk(KERN_ERR "nf_conntrack_acct: can't register to sysctl.\n");
+		return -ENOMEM;
+	}
+#endif
+
+	return 0;
+}
+
+void nf_conntrack_acct_fini(void)
+{
+#ifdef CONFIG_SYSCTL
+	unregister_sysctl_table(acct_sysctl_header);
+#endif
+	nf_ct_extend_unregister(&acct_extend);
+}
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 28d03e64200b..c519d090bdb9 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -37,6 +37,7 @@
 #include <net/netfilter/nf_conntrack_helper.h>
 #include <net/netfilter/nf_conntrack_core.h>
 #include <net/netfilter/nf_conntrack_extend.h>
+#include <net/netfilter/nf_conntrack_acct.h>
 
 #define NF_CONNTRACK_VERSION	"0.5.0"
 
@@ -555,6 +556,8 @@ init_conntrack(const struct nf_conntrack_tuple *tuple,
 		return NULL;
 	}
 
+	nf_ct_acct_ext_add(ct, GFP_ATOMIC);
+
 	spin_lock_bh(&nf_conntrack_lock);
 	exp = nf_ct_find_expectation(tuple);
 	if (exp) {
@@ -828,17 +831,16 @@ void __nf_ct_refresh_acct(struct nf_conn *ct,
 	}
 
 acct:
-#ifdef CONFIG_NF_CT_ACCT
 	if (do_acct) {
-		ct->counters[CTINFO2DIR(ctinfo)].packets++;
-		ct->counters[CTINFO2DIR(ctinfo)].bytes +=
-			skb->len - skb_network_offset(skb);
+		struct nf_conn_counter *acct;
 
-		if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000)
-		    || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000))
-			event |= IPCT_COUNTER_FILLING;
+		acct = nf_conn_acct_find(ct);
+		if (acct) {
+			acct[CTINFO2DIR(ctinfo)].packets++;
+			acct[CTINFO2DIR(ctinfo)].bytes +=
+				skb->len - skb_network_offset(skb);
+		}
 	}
-#endif
 
 	spin_unlock_bh(&nf_conntrack_lock);
 
@@ -853,15 +855,19 @@ bool __nf_ct_kill_acct(struct nf_conn *ct,
 		       const struct sk_buff *skb,
 		       int do_acct)
 {
-#ifdef CONFIG_NF_CT_ACCT
 	if (do_acct) {
+		struct nf_conn_counter *acct;
+
 		spin_lock_bh(&nf_conntrack_lock);
-		ct->counters[CTINFO2DIR(ctinfo)].packets++;
-		ct->counters[CTINFO2DIR(ctinfo)].bytes +=
-			skb->len - skb_network_offset(skb);
+		acct = nf_conn_acct_find(ct);
+		if (acct) {
+			acct[CTINFO2DIR(ctinfo)].packets++;
+			acct[CTINFO2DIR(ctinfo)].bytes +=
+				skb->len - skb_network_offset(skb);
+		}
 		spin_unlock_bh(&nf_conntrack_lock);
 	}
-#endif
+
 	if (del_timer(&ct->timeout)) {
 		ct->timeout.function((unsigned long)ct);
 		return true;
@@ -1029,6 +1035,7 @@ void nf_conntrack_cleanup(void)
 	nf_conntrack_proto_fini();
 	nf_conntrack_helper_fini();
 	nf_conntrack_expect_fini();
+	nf_conntrack_acct_fini();
 }
 
 struct hlist_head *nf_ct_alloc_hashtable(unsigned int *sizep, int *vmalloced)
@@ -1168,6 +1175,10 @@ int __init nf_conntrack_init(void)
 	if (ret < 0)
 		goto out_fini_expect;
 
+	ret = nf_conntrack_acct_init();
+	if (ret < 0)
+		goto out_fini_helper;
+
 	/* For use by REJECT target */
 	rcu_assign_pointer(ip_ct_attach, nf_conntrack_attach);
 	rcu_assign_pointer(nf_ct_destroy, destroy_conntrack);
@@ -1180,6 +1191,8 @@ int __init nf_conntrack_init(void)
 
 	return ret;
 
+out_fini_helper:
+	nf_conntrack_helper_fini();
 out_fini_expect:
 	nf_conntrack_expect_fini();
 out_fini_proto:
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 95a7967731f9..105a616c5c78 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -37,6 +37,7 @@
 #include <net/netfilter/nf_conntrack_l3proto.h>
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_tuple.h>
+#include <net/netfilter/nf_conntrack_acct.h>
 #ifdef CONFIG_NF_NAT_NEEDED
 #include <net/netfilter/nf_nat_core.h>
 #include <net/netfilter/nf_nat_protocol.h>
@@ -206,22 +207,26 @@ nla_put_failure:
 	return -1;
 }
 
-#ifdef CONFIG_NF_CT_ACCT
 static int
 ctnetlink_dump_counters(struct sk_buff *skb, const struct nf_conn *ct,
 			enum ip_conntrack_dir dir)
 {
 	enum ctattr_type type = dir ? CTA_COUNTERS_REPLY: CTA_COUNTERS_ORIG;
 	struct nlattr *nest_count;
+	const struct nf_conn_counter *acct;
+
+	acct = nf_conn_acct_find(ct);
+	if (!acct)
+		return 0;
 
 	nest_count = nla_nest_start(skb, type | NLA_F_NESTED);
 	if (!nest_count)
 		goto nla_put_failure;
 
-	NLA_PUT_BE32(skb, CTA_COUNTERS32_PACKETS,
-		     htonl(ct->counters[dir].packets));
-	NLA_PUT_BE32(skb, CTA_COUNTERS32_BYTES,
-		     htonl(ct->counters[dir].bytes));
+	NLA_PUT_BE64(skb, CTA_COUNTERS_PACKETS,
+		     cpu_to_be64(acct[dir].packets));
+	NLA_PUT_BE64(skb, CTA_COUNTERS_BYTES,
+		     cpu_to_be64(acct[dir].bytes));
 
 	nla_nest_end(skb, nest_count);
 
@@ -230,9 +235,6 @@ ctnetlink_dump_counters(struct sk_buff *skb, const struct nf_conn *ct,
 nla_put_failure:
 	return -1;
 }
-#else
-#define ctnetlink_dump_counters(a, b, c) (0)
-#endif
 
 #ifdef CONFIG_NF_CONNTRACK_MARK
 static inline int
@@ -501,11 +503,6 @@ static int ctnetlink_conntrack_event(struct notifier_block *this,
 			goto nla_put_failure;
 #endif
 
-		if (events & IPCT_COUNTER_FILLING &&
-		    (ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 ||
-		     ctnetlink_dump_counters(skb, ct, IP_CT_DIR_REPLY) < 0))
-			goto nla_put_failure;
-
 		if (events & IPCT_RELATED &&
 		    ctnetlink_dump_master(skb, ct) < 0)
 			goto nla_put_failure;
@@ -576,11 +573,15 @@ restart:
 				cb->args[1] = (unsigned long)ct;
 				goto out;
 			}
-#ifdef CONFIG_NF_CT_ACCT
+
 			if (NFNL_MSG_TYPE(cb->nlh->nlmsg_type) ==
-						IPCTNL_MSG_CT_GET_CTRZERO)
-				memset(&ct->counters, 0, sizeof(ct->counters));
-#endif
+						IPCTNL_MSG_CT_GET_CTRZERO) {
+				struct nf_conn_counter *acct;
+
+				acct = nf_conn_acct_find(ct);
+				if (acct)
+					memset(acct, 0, sizeof(struct nf_conn_counter[IP_CT_DIR_MAX]));
+			}
 		}
 		if (cb->args[1]) {
 			cb->args[1] = 0;
@@ -832,14 +833,9 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb,
 	u_int8_t u3 = nfmsg->nfgen_family;
 	int err = 0;
 
-	if (nlh->nlmsg_flags & NLM_F_DUMP) {
-#ifndef CONFIG_NF_CT_ACCT
-		if (NFNL_MSG_TYPE(nlh->nlmsg_type) == IPCTNL_MSG_CT_GET_CTRZERO)
-			return -ENOTSUPP;
-#endif
+	if (nlh->nlmsg_flags & NLM_F_DUMP)
 		return netlink_dump_start(ctnl, skb, nlh, ctnetlink_dump_table,
 					  ctnetlink_done);
-	}
 
 	if (cda[CTA_TUPLE_ORIG])
 		err = ctnetlink_parse_tuple(cda, &tuple, CTA_TUPLE_ORIG, u3);
@@ -1152,6 +1148,8 @@ ctnetlink_create_conntrack(struct nlattr *cda[],
 			goto err;
 	}
 
+	nf_ct_acct_ext_add(ct, GFP_KERNEL);
+
 #if defined(CONFIG_NF_CONNTRACK_MARK)
 	if (cda[CTA_MARK])
 		ct->mark = ntohl(nla_get_be32(cda[CTA_MARK]));
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index 46ea542d0df9..869ef9349d0f 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -25,6 +25,7 @@
 #include <net/netfilter/nf_conntrack_l4proto.h>
 #include <net/netfilter/nf_conntrack_expect.h>
 #include <net/netfilter/nf_conntrack_helper.h>
+#include <net/netfilter/nf_conntrack_acct.h>
 
 MODULE_LICENSE("GPL");
 
@@ -38,19 +39,6 @@ print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple,
 }
 EXPORT_SYMBOL_GPL(print_tuple);
 
-#ifdef CONFIG_NF_CT_ACCT
-static unsigned int
-seq_print_counters(struct seq_file *s,
-		   const struct ip_conntrack_counter *counter)
-{
-	return seq_printf(s, "packets=%llu bytes=%llu ",
-			  (unsigned long long)counter->packets,
-			  (unsigned long long)counter->bytes);
-}
-#else
-#define seq_print_counters(x, y)	0
-#endif
-
 struct ct_iter_state {
 	unsigned int bucket;
 };
@@ -146,7 +134,7 @@ static int ct_seq_show(struct seq_file *s, void *v)
 			l3proto, l4proto))
 		return -ENOSPC;
 
-	if (seq_print_counters(s, &ct->counters[IP_CT_DIR_ORIGINAL]))
+	if (seq_print_acct(s, ct, IP_CT_DIR_ORIGINAL))
 		return -ENOSPC;
 
 	if (!(test_bit(IPS_SEEN_REPLY_BIT, &ct->status)))
@@ -157,7 +145,7 @@ static int ct_seq_show(struct seq_file *s, void *v)
 			l3proto, l4proto))
 		return -ENOSPC;
 
-	if (seq_print_counters(s, &ct->counters[IP_CT_DIR_REPLY]))
+	if (seq_print_acct(s, ct, IP_CT_DIR_REPLY))
 		return -ENOSPC;
 
 	if (test_bit(IPS_ASSURED_BIT, &ct->status))
diff --git a/net/netfilter/xt_connbytes.c b/net/netfilter/xt_connbytes.c
index d7e8983cd37f..3e39c4fe1931 100644
--- a/net/netfilter/xt_connbytes.c
+++ b/net/netfilter/xt_connbytes.c
@@ -8,6 +8,7 @@
 #include <linux/netfilter/x_tables.h>
 #include <linux/netfilter/xt_connbytes.h>
 #include <net/netfilter/nf_conntrack.h>
+#include <net/netfilter/nf_conntrack_acct.h>
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
@@ -27,12 +28,15 @@ connbytes_mt(const struct sk_buff *skb, const struct net_device *in,
 	u_int64_t what = 0;	/* initialize to make gcc happy */
 	u_int64_t bytes = 0;
 	u_int64_t pkts = 0;
-	const struct ip_conntrack_counter *counters;
+	const struct nf_conn_counter *counters;
 
 	ct = nf_ct_get(skb, &ctinfo);
 	if (!ct)
 		return false;
-	counters = ct->counters;
+
+	counters = nf_conn_acct_find(ct);
+	if (!counters)
+		return false;
 
 	switch (sinfo->what) {
 	case XT_CONNBYTES_PKTS:
-- 
cgit v1.2.3


From 72961ecf84d67d6359a1b30f9b2a8427f13e1e71 Mon Sep 17 00:00:00 2001
From: Eric Leblond <eric@inl.fr>
Date: Mon, 21 Jul 2008 10:02:35 -0700
Subject: netfilter: nfnetlink_log: send complete hardware header

This patch adds some fields to NFLOG to be able to send the complete
hardware header with all necessary informations.
It sends to userspace:
 * the type of hardware link
 * the lenght of hardware header
 * the hardware header

Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/nfnetlink_log.h | 3 +++
 net/netfilter/nfnetlink_log.c           | 8 ++++++++
 2 files changed, 11 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/nfnetlink_log.h b/include/linux/netfilter/nfnetlink_log.h
index a85721332924..f661731f3cb1 100644
--- a/include/linux/netfilter/nfnetlink_log.h
+++ b/include/linux/netfilter/nfnetlink_log.h
@@ -48,6 +48,9 @@ enum nfulnl_attr_type {
 	NFULA_SEQ,			/* instance-local sequence number */
 	NFULA_SEQ_GLOBAL,		/* global sequence number */
 	NFULA_GID,			/* group id of socket */
+	NFULA_HWTYPE,			/* hardware type */
+	NFULA_HWHEADER,			/* hardware header */
+	NFULA_HWLEN,			/* hardware header length */
 
 	__NFULA_MAX
 };
diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c
index b8173af8c24a..9a35b57ab76d 100644
--- a/net/netfilter/nfnetlink_log.c
+++ b/net/netfilter/nfnetlink_log.c
@@ -453,6 +453,14 @@ __build_packet_message(struct nfulnl_instance *inst,
 		}
 	}
 
+	if (indev && skb_mac_header_was_set(skb)) {
+		NLA_PUT_BE16(inst->skb, NFULA_HWTYPE, htons(skb->dev->type));
+		NLA_PUT_BE16(inst->skb, NFULA_HWLEN,
+			     htons(skb->dev->hard_header_len));
+		NLA_PUT(inst->skb, NFULA_HWHEADER, skb->dev->hard_header_len,
+			skb_mac_header(skb));
+	}
+
 	if (skb->tstamp.tv64) {
 		struct nfulnl_msg_packet_timestamp ts;
 		struct timeval tv = ktime_to_timeval(skb->tstamp);
-- 
cgit v1.2.3


From 6579e57b31d79d31d9b806e41ba48774e73257dc Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Mon, 21 Jul 2008 13:31:48 -0700
Subject: net: Print the module name as part of the watchdog message

As suggested by Dave:

This patch adds a function to get the driver name from a struct net_device,
and consequently uses this in the watchdog timeout handler to print as
part of the message.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  2 ++
 net/core/dev.c            | 20 ++++++++++++++++++++
 net/sched/sch_generic.c   |  6 +++---
 3 files changed, 25 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 812bcd8b4363..f5ea445f89f0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1645,6 +1645,8 @@ extern void dev_seq_stop(struct seq_file *seq, void *v);
 extern int netdev_class_create_file(struct class_attribute *class_attr);
 extern void netdev_class_remove_file(struct class_attribute *class_attr);
 
+extern char *netdev_drivername(struct net_device *dev, char *buffer, int len);
+
 extern void linkwatch_run_queue(void);
 
 extern int netdev_compute_features(unsigned long all, unsigned long one);
diff --git a/net/core/dev.c b/net/core/dev.c
index 1698b3998981..ad5598d2bb37 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4686,6 +4686,26 @@ err_name:
 	return -ENOMEM;
 }
 
+char *netdev_drivername(struct net_device *dev, char *buffer, int len)
+{
+	struct device_driver *driver;
+	struct device *parent;
+
+	if (len <= 0 || !buffer)
+		return buffer;
+	buffer[0] = 0;
+
+	parent = dev->dev.parent;
+
+	if (!parent)
+		return buffer;
+
+	driver = parent->driver;
+	if (driver && driver->name)
+		strlcpy(buffer, driver->name, len);
+	return buffer;
+}
+
 static void __net_exit netdev_exit(struct net *net)
 {
 	kfree(net->dev_name_head);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index cb625b4d6da5..4ac7e3a8c253 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -212,9 +212,9 @@ static void dev_watchdog(unsigned long arg)
 			if (some_queue_stopped &&
 			    time_after(jiffies, (dev->trans_start +
 						 dev->watchdog_timeo))) {
-				printk(KERN_INFO "NETDEV WATCHDOG: %s: "
-				       "transmit timed out\n",
-				       dev->name);
+				char drivername[64];
+				printk(KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit timed out\n",
+				       dev->name, netdev_drivername(dev, drivername, 64));
 				dev->tx_timeout(dev);
 				WARN_ON_ONCE(1);
 			}
-- 
cgit v1.2.3


From ebb36a978131810c98e7198b1187090c697cf99f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 21 Jul 2008 13:41:16 -0700
Subject: ipv6: __KERNEL__ ifdef struct ipv6_devconf

Based upon a report by Olaf Hering.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ipv6.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 391ad0843a46..641e026eee8f 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -123,6 +123,7 @@ struct ipv6hdr {
 	struct	in6_addr	daddr;
 };
 
+#ifdef __KERNEL__
 /*
  * This structure contains configuration options per IPv6 link.
  */
@@ -167,6 +168,7 @@ struct ipv6_devconf {
 	__s32		accept_dad;
 	void		*sysctl;
 };
+#endif
 
 /* index values for the variables in ipv6_devconf */
 enum {
-- 
cgit v1.2.3


From 1b26da1510c02a2dac33c0ea48904256dcec4617 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 2 Jul 2008 12:46:22 -0700
Subject: USB: handle pci_name() being const

This changes usb_create_hcd() to be able to handle the fact that
pci_name() has changed to a constant string.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c | 2 +-
 drivers/usb/core/hcd.h | 2 +-
 include/linux/usb.h    | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 42a436478b78..07f7bedf4dca 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1764,7 +1764,7 @@ EXPORT_SYMBOL_GPL (usb_hc_died);
  * If memory is unavailable, returns NULL.
  */
 struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
-		struct device *dev, char *bus_name)
+		struct device *dev, const char *bus_name)
 {
 	struct usb_hcd *hcd;
 
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index b9de1569b39e..50b8bb2d1212 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -235,7 +235,7 @@ extern void usb_hcd_disable_endpoint(struct usb_device *udev,
 extern int usb_hcd_get_frame_number(struct usb_device *udev);
 
 extern struct usb_hcd *usb_create_hcd(const struct hc_driver *driver,
-		struct device *dev, char *bus_name);
+		struct device *dev, const char *bus_name);
 extern struct usb_hcd *usb_get_hcd(struct usb_hcd *hcd);
 extern void usb_put_hcd(struct usb_hcd *hcd);
 extern int usb_add_hcd(struct usb_hcd *hcd,
diff --git a/include/linux/usb.h b/include/linux/usb.h
index c08689ea9b4b..cee7fbb2b605 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -293,7 +293,7 @@ struct usb_devmap {
 struct usb_bus {
 	struct device *controller;	/* host/master side hardware */
 	int busnum;			/* Bus number (in order of reg) */
-	char *bus_name;			/* stable id (PCI slot_name etc) */
+	const char *bus_name;		/* stable id (PCI slot_name etc) */
 	u8 uses_dma;			/* Does the host controller use DMA? */
 	u8 otg_port;			/* 0, or number of OTG/HNP port */
 	unsigned is_b_host:1;		/* true during some HNP roleswitches */
-- 
cgit v1.2.3


From 9da82bd4649334817ef0e752a69eb99051645dad Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Thu, 8 May 2008 11:54:37 -0400
Subject: USB: implement "soft" unbinding

This patch (as1091) changes the way usbcore handles interface
unbinding.  If the interface's driver supports "soft" unbinding (a new
flag in the driver structure) then in-flight URBs are not cancelled
and endpoints are not disabled.  Instead the driver is allowed to
continue communicating with the device (although of course it should
stop before its disconnect routine returns).

The purpose of this change is to allow drivers to do a clean shutdown
when they get unbound from a device that is still plugged in.  Killing
all the URBs and disabling the endpoints before calling the driver's
disconnect method doesn't give the driver any control over what
happens, and it can leave devices in indeterminate states.  For
example, when usb-storage unbinds it doesn't want to stop while in the
middle of transmitting a SCSI command.

The soft_unbind flag is added because in the past, a number of drivers
have experienced problems related to ongoing I/O after their disconnect
routine returned.  Hence "soft" unbinding is made available only to
drivers that claim to support it.

The patch also replaces "interface_to_usbdev(intf)" with "udev" in a
couple of places, a minor simplification.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 11 ++++++-----
 include/linux/usb.h       |  3 +++
 2 files changed, 9 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 0a0e8cea0afc..8da1a56659be 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -257,15 +257,16 @@ static int usb_unbind_interface(struct device *dev)
 	udev = interface_to_usbdev(intf);
 	error = usb_autoresume_device(udev);
 
-	/* release all urbs for this interface */
-	usb_disable_interface(interface_to_usbdev(intf), intf);
+	/* Terminate all URBs for this interface unless the driver
+	 * supports "soft" unbinding.
+	 */
+	if (!driver->soft_unbind)
+		usb_disable_interface(udev, intf);
 
 	driver->disconnect(intf);
 
 	/* reset other interface state */
-	usb_set_interface(interface_to_usbdev(intf),
-			intf->altsetting[0].desc.bInterfaceNumber,
-			0);
+	usb_set_interface(udev, intf->altsetting[0].desc.bInterfaceNumber, 0);
 	usb_set_intfdata(intf, NULL);
 
 	intf->condition = USB_INTERFACE_UNBOUND;
diff --git a/include/linux/usb.h b/include/linux/usb.h
index cee7fbb2b605..8429d08bd2fd 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -972,6 +972,8 @@ struct usbdrv_wrap {
  *	added to this driver by preventing the sysfs file from being created.
  * @supports_autosuspend: if set to 0, the USB core will not allow autosuspend
  *	for interfaces bound to this driver.
+ * @soft_unbind: if set to 1, the USB core will not kill URBs and disable
+ *	endpoints before calling the driver's disconnect method.
  *
  * USB interface drivers must provide a name, probe() and disconnect()
  * methods, and an id_table.  Other driver fields are optional.
@@ -1012,6 +1014,7 @@ struct usb_driver {
 	struct usbdrv_wrap drvwrap;
 	unsigned int no_dynamic_id:1;
 	unsigned int supports_autosuspend:1;
+	unsigned int soft_unbind:1;
 };
 #define	to_usb_driver(d) container_of(d, struct usb_driver, drvwrap.driver)
 
-- 
cgit v1.2.3


From ea05af61a874ffbc158d9cf06df8a9396f299f38 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 20 May 2008 01:00:46 +0300
Subject: USB: remove CVS keywords

This patch removes CVS keywords that weren't updated for a long time
from comments.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/devices.c           | 2 --
 drivers/usb/core/devio.c             | 2 --
 drivers/usb/gadget/rndis.c           | 2 --
 drivers/usb/gadget/rndis.h           | 2 --
 drivers/usb/misc/emi62.c             | 2 --
 drivers/usb/serial/digi_acceleport.c | 2 --
 drivers/usb/storage/datafab.c        | 2 --
 drivers/usb/storage/debug.c          | 2 --
 drivers/usb/storage/debug.h          | 2 --
 drivers/usb/storage/dpcm.c           | 2 --
 drivers/usb/storage/dpcm.h           | 2 --
 drivers/usb/storage/freecom.c        | 2 --
 drivers/usb/storage/freecom.h        | 2 --
 drivers/usb/storage/initializers.c   | 2 --
 drivers/usb/storage/initializers.h   | 2 --
 drivers/usb/storage/isd200.c         | 2 --
 drivers/usb/storage/jumpshot.c       | 2 --
 drivers/usb/storage/protocol.c       | 2 --
 drivers/usb/storage/protocol.h       | 2 --
 drivers/usb/storage/scsiglue.c       | 2 --
 drivers/usb/storage/scsiglue.h       | 2 --
 drivers/usb/storage/sddr09.c         | 1 -
 drivers/usb/storage/sddr09.h         | 2 --
 drivers/usb/storage/sddr55.c         | 2 --
 drivers/usb/storage/sddr55.h         | 2 --
 drivers/usb/storage/shuttle_usbat.c  | 2 --
 drivers/usb/storage/shuttle_usbat.h  | 2 --
 drivers/usb/storage/transport.c      | 2 --
 drivers/usb/storage/transport.h      | 2 --
 drivers/usb/storage/unusual_devs.h   | 2 --
 drivers/usb/storage/usb.c            | 2 --
 drivers/usb/storage/usb.h            | 2 --
 include/linux/usbdevice_fs.h         | 2 --
 33 files changed, 65 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/devices.c b/drivers/usb/core/devices.c
index 83d9dc379d96..a681d9b92bda 100644
--- a/drivers/usb/core/devices.c
+++ b/drivers/usb/core/devices.c
@@ -46,8 +46,6 @@
  * 2000-07-05: Ashley Montanaro <ashley@compsoc.man.ac.uk>
  *   Converted file reading routine to dump to buffer once
  *   per device, not per bus
- *
- * $Id: devices.c,v 1.5 2000/01/11 13:58:21 tom Exp $
  */
 
 #include <linux/fs.h>
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 9218cca21043..7bee9c18b3bc 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -19,8 +19,6 @@
  *      along with this program; if not, write to the Free Software
  *      Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  *
- *  $Id: devio.c,v 1.7 2000/02/01 17:28:48 fliegl Exp $
- *
  *  This file implements the usbfs/x/y files, where
  *  x is the bus number and y the device number.
  *
diff --git a/drivers/usb/gadget/rndis.c b/drivers/usb/gadget/rndis.c
index d0677f5d3cd5..20b15fcc9d49 100644
--- a/drivers/usb/gadget/rndis.c
+++ b/drivers/usb/gadget/rndis.c
@@ -1,8 +1,6 @@
 /*
  * RNDIS MSG parser
  *
- * Version:     $Id: rndis.c,v 1.19 2004/03/25 21:33:46 robert Exp $
- *
  * Authors:	Benedikt Spranger, Pengutronix
  *		Robert Schwebel, Pengutronix
  *
diff --git a/drivers/usb/gadget/rndis.h b/drivers/usb/gadget/rndis.h
index 397b149f3ca7..b917b4f34ea8 100644
--- a/drivers/usb/gadget/rndis.h
+++ b/drivers/usb/gadget/rndis.h
@@ -1,8 +1,6 @@
 /*
  * RNDIS	Definitions for Remote NDIS
  *
- * Version:	$Id: rndis.h,v 1.15 2004/03/25 21:33:46 robert Exp $
- *
  * Authors:	Benedikt Spranger, Pengutronix
  *		Robert Schwebel, Pengutronix
  *
diff --git a/drivers/usb/misc/emi62.c b/drivers/usb/misc/emi62.c
index 20886c21e739..5d859ded5bbf 100644
--- a/drivers/usb/misc/emi62.c
+++ b/drivers/usb/misc/emi62.c
@@ -6,8 +6,6 @@
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License, as published by
  * the Free Software Foundation, version 2.
- * 
- * $Id: emi62.c,v 1.15 2002/04/23 06:13:59 tapio Exp $
  */
 #include <linux/kernel.h>
 #include <linux/errno.h>
diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c
index 04a56f300ea6..d96302d0d96e 100644
--- a/drivers/usb/serial/digi_acceleport.c
+++ b/drivers/usb/serial/digi_acceleport.c
@@ -229,8 +229,6 @@
 *    in case a wake up is lost.
 *  - Following Documentation/DocBook/kernel-locking.pdf no spin locks
 *    are held when calling copy_to/from_user or printk.
-*    
-*  $Id: digi_acceleport.c,v 1.80.1.2 2000/11/02 05:45:08 root Exp $
 */
 
 #include <linux/kernel.h>
diff --git a/drivers/usb/storage/datafab.c b/drivers/usb/storage/datafab.c
index 579e9f52053a..17f1ae232919 100644
--- a/drivers/usb/storage/datafab.c
+++ b/drivers/usb/storage/datafab.c
@@ -1,6 +1,4 @@
 /* Driver for Datafab USB Compact Flash reader
- *
- * $Id: datafab.c,v 1.7 2002/02/25 00:40:13 mdharm Exp $
  *
  * datafab driver v0.1:
  *
diff --git a/drivers/usb/storage/debug.c b/drivers/usb/storage/debug.c
index 01e430654a13..a2b5526c9fa0 100644
--- a/drivers/usb/storage/debug.c
+++ b/drivers/usb/storage/debug.c
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * Debugging Functions Source Code File
  *
- * $Id: debug.c,v 1.9 2002/04/22 03:39:43 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 1999-2002 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/drivers/usb/storage/debug.h b/drivers/usb/storage/debug.h
index 77e244a8c376..dbb985d52423 100644
--- a/drivers/usb/storage/debug.h
+++ b/drivers/usb/storage/debug.h
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * Debugging Functions Header File
  *
- * $Id: debug.h,v 1.6 2001/01/12 23:51:04 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 1999-2002 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/drivers/usb/storage/dpcm.c b/drivers/usb/storage/dpcm.c
index 9a410b5a6e5b..939923471af4 100644
--- a/drivers/usb/storage/dpcm.c
+++ b/drivers/usb/storage/dpcm.c
@@ -1,6 +1,4 @@
 /* Driver for Microtech DPCM-USB CompactFlash/SmartMedia reader
- *
- * $Id: dpcm.c,v 1.4 2001/06/11 02:54:25 mdharm Exp $
  *
  * DPCM driver v0.1:
  *
diff --git a/drivers/usb/storage/dpcm.h b/drivers/usb/storage/dpcm.h
index 81b464cfcc1e..e7b7b0f120d7 100644
--- a/drivers/usb/storage/dpcm.h
+++ b/drivers/usb/storage/dpcm.h
@@ -1,6 +1,4 @@
 /* Driver for Microtech DPCM-USB CompactFlash/SmartMedia reader
- *
- * $Id: dpcm.h,v 1.2 2000/08/25 00:13:51 mdharm Exp $
  *
  * DPCM driver v0.1:
  *
diff --git a/drivers/usb/storage/freecom.c b/drivers/usb/storage/freecom.c
index f5a4e8d6a3b1..7a4d45677227 100644
--- a/drivers/usb/storage/freecom.c
+++ b/drivers/usb/storage/freecom.c
@@ -1,6 +1,4 @@
 /* Driver for Freecom USB/IDE adaptor
- *
- * $Id: freecom.c,v 1.22 2002/04/22 03:39:43 mdharm Exp $
  *
  * Freecom v0.1:
  *
diff --git a/drivers/usb/storage/freecom.h b/drivers/usb/storage/freecom.h
index 1b012d62d0a8..20d0fe6ba0c8 100644
--- a/drivers/usb/storage/freecom.h
+++ b/drivers/usb/storage/freecom.h
@@ -1,6 +1,4 @@
 /* Driver for Freecom USB/IDE adaptor
- *
- * $Id: freecom.h,v 1.4 2000/08/29 14:49:15 dlbrown Exp $
  *
  * Freecom v0.1:
  *
diff --git a/drivers/usb/storage/initializers.c b/drivers/usb/storage/initializers.c
index 187dd1e01093..4995bb595aef 100644
--- a/drivers/usb/storage/initializers.c
+++ b/drivers/usb/storage/initializers.c
@@ -1,6 +1,4 @@
 /* Special Initializers for certain USB Mass Storage devices
- *
- * $Id: initializers.c,v 1.2 2000/09/06 22:35:57 mdharm Exp $
  *
  * Current development and maintenance by:
  *   (c) 1999, 2000 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
diff --git a/drivers/usb/storage/initializers.h b/drivers/usb/storage/initializers.h
index ad3ffd4236c2..529327fbb06b 100644
--- a/drivers/usb/storage/initializers.h
+++ b/drivers/usb/storage/initializers.h
@@ -1,6 +1,4 @@
 /* Header file for Special Initializers for certain USB Mass Storage devices
- *
- * $Id: initializers.h,v 1.1 2000/08/29 23:07:02 mdharm Exp $
  *
  * Current development and maintenance by:
  *   (c) 1999, 2000 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
diff --git a/drivers/usb/storage/isd200.c b/drivers/usb/storage/isd200.c
index a153335f3648..383abf2516a5 100644
--- a/drivers/usb/storage/isd200.c
+++ b/drivers/usb/storage/isd200.c
@@ -1,6 +1,4 @@
 /* Transport & Protocol Driver for In-System Design, Inc. ISD200 ASIC
- *
- * $Id: isd200.c,v 1.16 2002/04/22 03:39:43 mdharm Exp $
  *
  * Current development and maintenance:
  *   (C) 2001-2002 Björn Stenberg (bjorn@haxx.se)
diff --git a/drivers/usb/storage/jumpshot.c b/drivers/usb/storage/jumpshot.c
index 61097cbb1585..df67f13c9e73 100644
--- a/drivers/usb/storage/jumpshot.c
+++ b/drivers/usb/storage/jumpshot.c
@@ -1,6 +1,4 @@
 /* Driver for Lexar "Jumpshot" Compact Flash reader
- *
- * $Id: jumpshot.c,v 1.7 2002/02/25 00:40:13 mdharm Exp $
  *
  * jumpshot driver v0.1:
  *
diff --git a/drivers/usb/storage/protocol.c b/drivers/usb/storage/protocol.c
index b9b8ede61fb3..3b3357e20ea7 100644
--- a/drivers/usb/storage/protocol.c
+++ b/drivers/usb/storage/protocol.c
@@ -1,6 +1,4 @@
 /* Driver for USB Mass Storage compliant devices
- *
- * $Id: protocol.c,v 1.14 2002/04/22 03:39:43 mdharm Exp $
  *
  * Current development and maintenance by:
  *   (c) 1999-2002 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
diff --git a/drivers/usb/storage/protocol.h b/drivers/usb/storage/protocol.h
index 8737a36891ca..487056ffb516 100644
--- a/drivers/usb/storage/protocol.h
+++ b/drivers/usb/storage/protocol.h
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * Protocol Functions Header File
  *
- * $Id: protocol.h,v 1.4 2001/02/13 07:10:03 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 1999, 2000 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index 043b60b2ad17..b4c9e0f18a82 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * SCSI layer glue code
  *
- * $Id: scsiglue.c,v 1.26 2002/04/22 03:39:43 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 1999-2002 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/drivers/usb/storage/scsiglue.h b/drivers/usb/storage/scsiglue.h
index 737e4fa6045f..ffa1cca93d2c 100644
--- a/drivers/usb/storage/scsiglue.h
+++ b/drivers/usb/storage/scsiglue.h
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * SCSI Connecting Glue Header File
  *
- * $Id: scsiglue.h,v 1.4 2000/08/25 00:13:51 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 1999, 2000 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/drivers/usb/storage/sddr09.c b/drivers/usb/storage/sddr09.c
index 8972b17da843..c5a54b872c24 100644
--- a/drivers/usb/storage/sddr09.c
+++ b/drivers/usb/storage/sddr09.c
@@ -1,6 +1,5 @@
 /* Driver for SanDisk SDDR-09 SmartMedia reader
  *
- * $Id: sddr09.c,v 1.24 2002/04/22 03:39:43 mdharm Exp $
  *   (c) 2000, 2001 Robert Baruch (autophile@starband.net)
  *   (c) 2002 Andries Brouwer (aeb@cwi.nl)
  * Developed with the assistance of:
diff --git a/drivers/usb/storage/sddr09.h b/drivers/usb/storage/sddr09.h
index c03089a9ec38..e50033ad7b19 100644
--- a/drivers/usb/storage/sddr09.h
+++ b/drivers/usb/storage/sddr09.h
@@ -1,8 +1,6 @@
 /* Driver for SanDisk SDDR-09 SmartMedia reader
  * Header File
  *
- * $Id: sddr09.h,v 1.5 2000/08/25 00:13:51 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 2000 Robert Baruch (autophile@dol.net)
  *   (c) 2002 Andries Brouwer (aeb@cwi.nl)
diff --git a/drivers/usb/storage/sddr55.c b/drivers/usb/storage/sddr55.c
index 6d14327c921d..0d8df7577899 100644
--- a/drivers/usb/storage/sddr55.c
+++ b/drivers/usb/storage/sddr55.c
@@ -1,6 +1,4 @@
 /* Driver for SanDisk SDDR-55 SmartMedia reader
- *
- * $Id:$
  *
  * SDDR55 driver v0.1:
  *
diff --git a/drivers/usb/storage/sddr55.h b/drivers/usb/storage/sddr55.h
index d6bd32f6c9f3..a815a0470c84 100644
--- a/drivers/usb/storage/sddr55.h
+++ b/drivers/usb/storage/sddr55.h
@@ -1,8 +1,6 @@
 /* Driver for SanDisk SDDR-55 SmartMedia reader
  * Header File
  *
- * $Id:$
- *
  * Current development and maintenance by:
  *   (c) 2002 Simon Munton
  *
diff --git a/drivers/usb/storage/shuttle_usbat.c b/drivers/usb/storage/shuttle_usbat.c
index 570c1250f6f3..ae6d64810d2a 100644
--- a/drivers/usb/storage/shuttle_usbat.c
+++ b/drivers/usb/storage/shuttle_usbat.c
@@ -1,6 +1,4 @@
 /* Driver for SCM Microsystems (a.k.a. Shuttle) USB-ATAPI cable
- *
- * $Id: shuttle_usbat.c,v 1.17 2002/04/22 03:39:43 mdharm Exp $
  *
  * Current development and maintenance by:
  *   (c) 2000, 2001 Robert Baruch (autophile@starband.net)
diff --git a/drivers/usb/storage/shuttle_usbat.h b/drivers/usb/storage/shuttle_usbat.h
index 3ddf143a1dec..d8bfc43e9044 100644
--- a/drivers/usb/storage/shuttle_usbat.h
+++ b/drivers/usb/storage/shuttle_usbat.h
@@ -1,8 +1,6 @@
 /* Driver for SCM Microsystems USB-ATAPI cable
  * Header File
  *
- * $Id: shuttle_usbat.h,v 1.5 2000/09/17 14:44:52 groovyjava Exp $
- *
  * Current development and maintenance by:
  *   (c) 2000 Robert Baruch (autophile@dol.net)
  *   (c) 2004, 2005 Daniel Drake <dsd@gentoo.org>
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 94138df557b9..08d3a13fec2c 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -1,6 +1,4 @@
 /* Driver for USB Mass Storage compliant devices
- *
- * $Id: transport.c,v 1.47 2002/04/22 03:39:43 mdharm Exp $
  *
  * Current development and maintenance by:
  *   (c) 1999-2002 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
diff --git a/drivers/usb/storage/transport.h b/drivers/usb/storage/transport.h
index ada7c2f43f84..e70b88182f0e 100644
--- a/drivers/usb/storage/transport.h
+++ b/drivers/usb/storage/transport.h
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * Transport Functions Header File
  *
- * $Id: transport.h,v 1.18 2002/04/21 02:57:59 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 1999, 2000 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 3ba972c20f3c..7ae69f55aa96 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * Unusual Devices File
  *
- * $Id: unusual_devs.h,v 1.32 2002/02/25 02:41:24 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 2000-2002 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c
index 6bfd99dd57aa..bfea851be985 100644
--- a/drivers/usb/storage/usb.c
+++ b/drivers/usb/storage/usb.c
@@ -1,6 +1,4 @@
 /* Driver for USB Mass Storage compliant devices
- *
- * $Id: usb.c,v 1.75 2002/04/22 03:39:43 mdharm Exp $
  *
  * Current development and maintenance by:
  *   (c) 1999-2003 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
diff --git a/drivers/usb/storage/usb.h b/drivers/usb/storage/usb.h
index 47906dc620db..a4ad73bd832d 100644
--- a/drivers/usb/storage/usb.h
+++ b/drivers/usb/storage/usb.h
@@ -1,8 +1,6 @@
 /* Driver for USB Mass Storage compliant devices
  * Main Header File
  *
- * $Id: usb.h,v 1.21 2002/04/21 02:57:59 mdharm Exp $
- *
  * Current development and maintenance by:
  *   (c) 1999-2002 Matthew Dharm (mdharm-usb@one-eyed-alien.net)
  *
diff --git a/include/linux/usbdevice_fs.h b/include/linux/usbdevice_fs.h
index 3118ede2c67b..0044d9b4cb85 100644
--- a/include/linux/usbdevice_fs.h
+++ b/include/linux/usbdevice_fs.h
@@ -22,8 +22,6 @@
  *
  *  History:
  *   0.1  04.01.2000  Created
- *
- *  $Id: usbdevice_fs.h,v 1.1 2000/01/06 18:40:41 tom Exp $
  */
 
 /*****************************************************************************/
-- 
cgit v1.2.3


From a4c39c41bf3592684e36fa0dbbd4ab1a31f969b9 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Thu, 19 Jun 2008 17:52:25 -0700
Subject: usb gadget: descriptor copying support

Define three new descriptor manipulation utilities, for use when
setting up functions that may have multiple instances:

	usb_copy_descriptors() to copy a vector of descriptors
	usb_free_descriptors() to free the copy
	usb_find_endpoint() to find a copied version

These will be used as follows.  Functions will continue to have static
tables of descriptors they update, now used as __initdata templates.

When a function creates a new instance, it patches those tables with
relevant interface and string IDs, plus endpoint assignments.  Then it
copies those morphed descriptors, associates the copies with the new
function instance, and records the endpoint descriptors to use when
activating the endpoints.  When initialization is done, only the copies
remain in memory.  The copies are freed on driver removal.

This ensures that each instance has descriptors which hold the right
instance-specific data.  Two instances in the same configuration will
obviously never share the same interface IDs or use the same endpoints.
Instances in different configurations won't do so either, which means
this is slightly less memory-efficient in some cases.

This also includes a bugfix to the epautoconf code that shows up with
this usage model.  It must replace the previous endpoint number when
updating the template descriptors, not just mask in a few more bits.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/config.c     | 76 ++++++++++++++++++++++++++++++++++++++++-
 drivers/usb/gadget/epautoconf.c |  1 +
 include/linux/usb/gadget.h      | 19 +++++++++++
 3 files changed, 95 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/usb/gadget/config.c b/drivers/usb/gadget/config.c
index a4e54b2743f0..1ca1c326392a 100644
--- a/drivers/usb/gadget/config.c
+++ b/drivers/usb/gadget/config.c
@@ -96,7 +96,7 @@ int usb_gadget_config_buf(
 	/* config descriptor first */
 	if (length < USB_DT_CONFIG_SIZE || !desc)
 		return -EINVAL;
-	*cp = *config; 
+	*cp = *config;
 
 	/* then interface/endpoint/class/vendor/... */
 	len = usb_descriptor_fillbuf(USB_DT_CONFIG_SIZE + (u8*)buf,
@@ -115,3 +115,77 @@ int usb_gadget_config_buf(
 	return len;
 }
 
+/**
+ * usb_copy_descriptors - copy a vector of USB descriptors
+ * @src: null-terminated vector to copy
+ * Context: initialization code, which may sleep
+ *
+ * This makes a copy of a vector of USB descriptors.  Its primary use
+ * is to support usb_function objects which can have multiple copies,
+ * each needing different descriptors.  Functions may have static
+ * tables of descriptors, which are used as templates and customized
+ * with identifiers (for interfaces, strings, endpoints, and more)
+ * as needed by a given function instance.
+ */
+struct usb_descriptor_header **__init
+usb_copy_descriptors(struct usb_descriptor_header **src)
+{
+	struct usb_descriptor_header **tmp;
+	unsigned bytes;
+	unsigned n_desc;
+	void *mem;
+	struct usb_descriptor_header **ret;
+
+	/* count descriptors and their sizes; then add vector size */
+	for (bytes = 0, n_desc = 0, tmp = src; *tmp; tmp++, n_desc++)
+		bytes += (*tmp)->bLength;
+	bytes += (n_desc + 1) * sizeof(*tmp);
+
+	mem = kmalloc(bytes, GFP_KERNEL);
+	if (!mem)
+		return NULL;
+
+	/* fill in pointers starting at "tmp",
+	 * to descriptors copied starting at "mem";
+	 * and return "ret"
+	 */
+	tmp = mem;
+	ret = mem;
+	mem += (n_desc + 1) * sizeof(*tmp);
+	while (*src) {
+		memcpy(mem, *src, (*src)->bLength);
+		*tmp = mem;
+		tmp++;
+		mem += (*src)->bLength;
+		src++;
+	}
+	*tmp = NULL;
+
+	return ret;
+}
+
+/**
+ * usb_find_endpoint - find a copy of an endpoint descriptor
+ * @src: original vector of descriptors
+ * @copy: copy of @src
+ * @ep: endpoint descriptor found in @src
+ *
+ * This returns the copy of the @match descriptor made for @copy.  Its
+ * intended use is to help remembering the endpoint descriptor to use
+ * when enabling a given endpoint.
+ */
+struct usb_endpoint_descriptor *__init
+usb_find_endpoint(
+	struct usb_descriptor_header **src,
+	struct usb_descriptor_header **copy,
+	struct usb_endpoint_descriptor *match
+)
+{
+	while (*src) {
+		if (*src == (void *) match)
+			return (void *)*copy;
+		src++;
+		copy++;
+	}
+	return NULL;
+}
diff --git a/drivers/usb/gadget/epautoconf.c b/drivers/usb/gadget/epautoconf.c
index 8bdad221fa91..9462e30192d8 100644
--- a/drivers/usb/gadget/epautoconf.c
+++ b/drivers/usb/gadget/epautoconf.c
@@ -159,6 +159,7 @@ ep_matches (
 	/* MATCH!! */
 
 	/* report address */
+	desc->bEndpointAddress &= USB_DIR_IN;
 	if (isdigit (ep->name [2])) {
 		u8	num = simple_strtol (&ep->name [2], NULL, 10);
 		desc->bEndpointAddress |= num;
diff --git a/include/linux/usb/gadget.h b/include/linux/usb/gadget.h
index cf468fbdbf8e..0ebedaec075d 100644
--- a/include/linux/usb/gadget.h
+++ b/include/linux/usb/gadget.h
@@ -858,6 +858,25 @@ int usb_descriptor_fillbuf(void *, unsigned,
 int usb_gadget_config_buf(const struct usb_config_descriptor *config,
 	void *buf, unsigned buflen, const struct usb_descriptor_header **desc);
 
+/* copy a NULL-terminated vector of descriptors */
+struct usb_descriptor_header **usb_copy_descriptors(
+		struct usb_descriptor_header **);
+
+/* return copy of endpoint descriptor given original descriptor set */
+struct usb_endpoint_descriptor *usb_find_endpoint(
+	struct usb_descriptor_header **src,
+	struct usb_descriptor_header **copy,
+	struct usb_endpoint_descriptor *match);
+
+/**
+ * usb_free_descriptors - free descriptors returned by usb_copy_descriptors()
+ * @v: vector of descriptors
+ */
+static inline void usb_free_descriptors(struct usb_descriptor_header **v)
+{
+	kfree(v);
+}
+
 /*-------------------------------------------------------------------------*/
 
 /* utility wrapping a simple endpoint selection policy */
-- 
cgit v1.2.3


From 40982be52d8f64c3e10adce17e66ab755a4fa26b Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Thu, 19 Jun 2008 17:52:58 -0700
Subject: usb gadget: composite gadget core

Add <linux/usb/composite.h> interfaces for composite gadget drivers, and
basic implementation support behind it:

  - struct usb_function ... groups one or more interfaces into a function
    managed as one unit within a configuration, to which it's added by
    usb_add_function().

  - struct usb_configuration ... groups one or more such functions into
    a configuration managed as one unit by a driver, to which it's added
    by usb_add_config().  These operate at either high or full/low speeds
    and at a given bMaxPower.

  - struct usb_composite_driver ... groups one or more such configurations
    into a gadget driver, which may be registered or unregistered.

  - struct usb_composite_dev ... a usb_composite_driver manages this; it
    wraps the usb_gadget exposed by the controller driver.

This also includes some basic kerneldoc.

How to use it (the short version):  provide a usb_composite_driver with a
bind() that calls usb_add_config() for each of the needed configurations.
The configurations in turn have bind() calls, which will usb_add_function()
for each function required.  Each function's bind() allocates resources
needed to perform its tasks, like endpoints; sometimes configurations will
allocate resources too.

Separate patches will convert most gadget drivers to this infrastructure.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/DocBook/gadget.tmpl |   35 ++
 drivers/usb/gadget/composite.c    | 1040 +++++++++++++++++++++++++++++++++++++
 include/linux/usb/composite.h     |  338 ++++++++++++
 3 files changed, 1413 insertions(+)
 create mode 100644 drivers/usb/gadget/composite.c
 create mode 100644 include/linux/usb/composite.h

(limited to 'include/linux')

diff --git a/Documentation/DocBook/gadget.tmpl b/Documentation/DocBook/gadget.tmpl
index 5a8ffa761e09..478bfe16a19d 100644
--- a/Documentation/DocBook/gadget.tmpl
+++ b/Documentation/DocBook/gadget.tmpl
@@ -524,6 +524,41 @@ These utilities include endpoint autoconfiguration.
 <!-- !Edrivers/usb/gadget/epautoconf.c -->
 </sect1>
 
+<sect1 id="composite"><title>Composite Device Framework</title>
+
+<para>The core API is sufficient for writing drivers for composite
+USB devices (with more than one function in a given configuration),
+and also multi-configuration devices (also more than one function,
+but not necessarily sharing a given configuration).
+There is however an optional framework which makes it easier to
+reuse and combine functions.
+</para>
+
+<para>Devices using this framework provide a <emphasis>struct
+usb_composite_driver</emphasis>, which in turn provides one or
+more <emphasis>struct usb_configuration</emphasis> instances.
+Each such configuration includes at least one
+<emphasis>struct usb_function</emphasis>, which packages a user
+visible role such as "network link" or "mass storage device".
+Management functions may also exist, such as "Device Firmware
+Upgrade".
+</para>
+
+!Iinclude/linux/usb/composite.h
+!Edrivers/usb/gadget/composite.c
+
+</sect1>
+
+<sect1 id="functions"><title>Composite Device Functions</title>
+
+<para>At this writing, a few of the current gadget drivers have
+been converted to this framework.
+Near-term plans include converting all of them, except for "gadgetfs".
+</para>
+
+</sect1>
+
+
 </chapter>
 
 <chapter id="controllers"><title>Peripheral Controller Drivers</title>
diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c
new file mode 100644
index 000000000000..b3d302a6a081
--- /dev/null
+++ b/drivers/usb/gadget/composite.c
@@ -0,0 +1,1040 @@
+/*
+ * composite.c - infrastructure for Composite USB Gadgets
+ *
+ * Copyright (C) 2006-2008 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+/* #define VERBOSE_DEBUG */
+
+#include <linux/kallsyms.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/device.h>
+
+#include <linux/usb/composite.h>
+
+
+/*
+ * The code in this file is utility code, used to build a gadget driver
+ * from one or more "function" drivers, one or more "configuration"
+ * objects, and a "usb_composite_driver" by gluing them together along
+ * with the relevant device-wide data.
+ */
+
+/* big enough to hold our biggest descriptor */
+#define USB_BUFSIZ	512
+
+static struct usb_composite_driver *composite;
+
+/* Some systems will need runtime overrides for the  product identifers
+ * published in the device descriptor, either numbers or strings or both.
+ * String parameters are in UTF-8 (superset of ASCII's 7 bit characters).
+ */
+
+static ushort idVendor;
+module_param(idVendor, ushort, 0);
+MODULE_PARM_DESC(idVendor, "USB Vendor ID");
+
+static ushort idProduct;
+module_param(idProduct, ushort, 0);
+MODULE_PARM_DESC(idProduct, "USB Product ID");
+
+static ushort bcdDevice;
+module_param(bcdDevice, ushort, 0);
+MODULE_PARM_DESC(bcdDevice, "USB Device version (BCD)");
+
+static char *iManufacturer;
+module_param(iManufacturer, charp, 0);
+MODULE_PARM_DESC(iManufacturer, "USB Manufacturer string");
+
+static char *iProduct;
+module_param(iProduct, charp, 0);
+MODULE_PARM_DESC(iProduct, "USB Product string");
+
+static char *iSerialNumber;
+module_param(iSerialNumber, charp, 0);
+MODULE_PARM_DESC(iSerialNumber, "SerialNumber string");
+
+/*-------------------------------------------------------------------------*/
+
+/**
+ * usb_add_function() - add a function to a configuration
+ * @config: the configuration
+ * @function: the function being added
+ * Context: single threaded during gadget setup
+ *
+ * After initialization, each configuration must have one or more
+ * functions added to it.  Adding a function involves calling its @bind()
+ * method to allocate resources such as interface and string identifiers
+ * and endpoints.
+ *
+ * This function returns the value of the function's bind(), which is
+ * zero for success else a negative errno value.
+ */
+int __init usb_add_function(struct usb_configuration *config,
+		struct usb_function *function)
+{
+	int	value = -EINVAL;
+
+	DBG(config->cdev, "adding '%s'/%p to config '%s'/%p\n",
+			function->name, function,
+			config->label, config);
+
+	if (!function->set_alt || !function->disable)
+		goto done;
+
+	function->config = config;
+	list_add_tail(&function->list, &config->functions);
+
+	/* REVISIT *require* function->bind? */
+	if (function->bind) {
+		value = function->bind(config, function);
+		if (value < 0) {
+			list_del(&function->list);
+			function->config = NULL;
+		}
+	} else
+		value = 0;
+
+	/* We allow configurations that don't work at both speeds.
+	 * If we run into a lowspeed Linux system, treat it the same
+	 * as full speed ... it's the function drivers that will need
+	 * to avoid bulk and ISO transfers.
+	 */
+	if (!config->fullspeed && function->descriptors)
+		config->fullspeed = true;
+	if (!config->highspeed && function->hs_descriptors)
+		config->highspeed = true;
+
+done:
+	if (value)
+		DBG(config->cdev, "adding '%s'/%p --> %d\n",
+				function->name, function, value);
+	return value;
+}
+
+/**
+ * usb_interface_id() - allocate an unused interface ID
+ * @config: configuration associated with the interface
+ * @function: function handling the interface
+ * Context: single threaded during gadget setup
+ *
+ * usb_interface_id() is called from usb_function.bind() callbacks to
+ * allocate new interface IDs.  The function driver will then store that
+ * ID in interface, association, CDC union, and other descriptors.  It
+ * will also handle any control requests targetted at that interface,
+ * particularly changing its altsetting via set_alt().  There may
+ * also be class-specific or vendor-specific requests to handle.
+ *
+ * All interface identifier should be allocated using this routine, to
+ * ensure that for example different functions don't wrongly assign
+ * different meanings to the same identifier.  Note that since interface
+ * identifers are configuration-specific, functions used in more than
+ * one configuration (or more than once in a given configuration) need
+ * multiple versions of the relevant descriptors.
+ *
+ * Returns the interface ID which was allocated; or -ENODEV if no
+ * more interface IDs can be allocated.
+ */
+int __init usb_interface_id(struct usb_configuration *config,
+		struct usb_function *function)
+{
+	unsigned id = config->next_interface_id;
+
+	if (id < MAX_CONFIG_INTERFACES) {
+		config->interface[id] = function;
+		config->next_interface_id = id + 1;
+		return id;
+	}
+	return -ENODEV;
+}
+
+static int config_buf(struct usb_configuration *config,
+		enum usb_device_speed speed, void *buf, u8 type)
+{
+	struct usb_config_descriptor	*c = buf;
+	void				*next = buf + USB_DT_CONFIG_SIZE;
+	int				len = USB_BUFSIZ - USB_DT_CONFIG_SIZE;
+	struct usb_function		*f;
+	int				status;
+
+	/* write the config descriptor */
+	c = buf;
+	c->bLength = USB_DT_CONFIG_SIZE;
+	c->bDescriptorType = type;
+	/* wTotalLength is written later */
+	c->bNumInterfaces = config->next_interface_id;
+	c->bConfigurationValue = config->bConfigurationValue;
+	c->iConfiguration = config->iConfiguration;
+	c->bmAttributes = USB_CONFIG_ATT_ONE | config->bmAttributes;
+	c->bMaxPower = config->bMaxPower;
+
+	/* There may be e.g. OTG descriptors */
+	if (config->descriptors) {
+		status = usb_descriptor_fillbuf(next, len,
+				config->descriptors);
+		if (status < 0)
+			return status;
+		len -= status;
+		next += status;
+	}
+
+	/* add each function's descriptors */
+	list_for_each_entry(f, &config->functions, list) {
+		struct usb_descriptor_header **descriptors;
+
+		if (speed == USB_SPEED_HIGH)
+			descriptors = f->hs_descriptors;
+		else
+			descriptors = f->descriptors;
+		if (!descriptors)
+			continue;
+		status = usb_descriptor_fillbuf(next, len,
+			(const struct usb_descriptor_header **) descriptors);
+		if (status < 0)
+			return status;
+		len -= status;
+		next += status;
+	}
+
+	len = next - buf;
+	c->wTotalLength = cpu_to_le16(len);
+	return len;
+}
+
+static int config_desc(struct usb_composite_dev *cdev, unsigned w_value)
+{
+	struct usb_gadget		*gadget = cdev->gadget;
+	struct usb_configuration	*c;
+	u8				type = w_value >> 8;
+	enum usb_device_speed		speed = USB_SPEED_UNKNOWN;
+
+	if (gadget_is_dualspeed(gadget)) {
+		int			hs = 0;
+
+		if (gadget->speed == USB_SPEED_HIGH)
+			hs = 1;
+		if (type == USB_DT_OTHER_SPEED_CONFIG)
+			hs = !hs;
+		if (hs)
+			speed = USB_SPEED_HIGH;
+
+	}
+
+	/* This is a lookup by config *INDEX* */
+	w_value &= 0xff;
+	list_for_each_entry(c, &cdev->configs, list) {
+		/* ignore configs that won't work at this speed */
+		if (speed == USB_SPEED_HIGH) {
+			if (!c->highspeed)
+				continue;
+		} else {
+			if (!c->fullspeed)
+				continue;
+		}
+		if (w_value == 0)
+			return config_buf(c, speed, cdev->req->buf, type);
+		w_value--;
+	}
+	return -EINVAL;
+}
+
+static int count_configs(struct usb_composite_dev *cdev, unsigned type)
+{
+	struct usb_gadget		*gadget = cdev->gadget;
+	struct usb_configuration	*c;
+	unsigned			count = 0;
+	int				hs = 0;
+
+	if (gadget_is_dualspeed(gadget)) {
+		if (gadget->speed == USB_SPEED_HIGH)
+			hs = 1;
+		if (type == USB_DT_DEVICE_QUALIFIER)
+			hs = !hs;
+	}
+	list_for_each_entry(c, &cdev->configs, list) {
+		/* ignore configs that won't work at this speed */
+		if (hs) {
+			if (!c->highspeed)
+				continue;
+		} else {
+			if (!c->fullspeed)
+				continue;
+		}
+		count++;
+	}
+	return count;
+}
+
+static void device_qual(struct usb_composite_dev *cdev)
+{
+	struct usb_qualifier_descriptor	*qual = cdev->req->buf;
+
+	qual->bLength = sizeof(*qual);
+	qual->bDescriptorType = USB_DT_DEVICE_QUALIFIER;
+	/* POLICY: same bcdUSB and device type info at both speeds */
+	qual->bcdUSB = cdev->desc.bcdUSB;
+	qual->bDeviceClass = cdev->desc.bDeviceClass;
+	qual->bDeviceSubClass = cdev->desc.bDeviceSubClass;
+	qual->bDeviceProtocol = cdev->desc.bDeviceProtocol;
+	/* ASSUME same EP0 fifo size at both speeds */
+	qual->bMaxPacketSize0 = cdev->desc.bMaxPacketSize0;
+	qual->bNumConfigurations = count_configs(cdev, USB_DT_DEVICE_QUALIFIER);
+}
+
+/*-------------------------------------------------------------------------*/
+
+static void reset_config(struct usb_composite_dev *cdev)
+{
+	struct usb_function		*f;
+
+	DBG(cdev, "reset config\n");
+
+	list_for_each_entry(f, &cdev->config->functions, list) {
+		if (f->disable)
+			f->disable(f);
+	}
+	cdev->config = NULL;
+}
+
+static int set_config(struct usb_composite_dev *cdev,
+		const struct usb_ctrlrequest *ctrl, unsigned number)
+{
+	struct usb_gadget	*gadget = cdev->gadget;
+	struct usb_configuration *c = NULL;
+	int			result = -EINVAL;
+	unsigned		power = gadget_is_otg(gadget) ? 8 : 100;
+	int			tmp;
+
+	if (cdev->config)
+		reset_config(cdev);
+
+	if (number) {
+		list_for_each_entry(c, &cdev->configs, list) {
+			if (c->bConfigurationValue == number) {
+				result = 0;
+				break;
+			}
+		}
+		if (result < 0)
+			goto done;
+	} else
+		result = 0;
+
+	INFO(cdev, "%s speed config #%d: %s\n",
+		({ char *speed;
+		switch (gadget->speed) {
+		case USB_SPEED_LOW:	speed = "low"; break;
+		case USB_SPEED_FULL:	speed = "full"; break;
+		case USB_SPEED_HIGH:	speed = "high"; break;
+		default:		speed = "?"; break;
+		} ; speed; }), number, c ? c->label : "unconfigured");
+
+	if (!c)
+		goto done;
+
+	cdev->config = c;
+
+	/* Initialize all interfaces by setting them to altsetting zero. */
+	for (tmp = 0; tmp < MAX_CONFIG_INTERFACES; tmp++) {
+		struct usb_function	*f = c->interface[tmp];
+
+		if (!f)
+			break;
+
+		result = f->set_alt(f, tmp, 0);
+		if (result < 0) {
+			DBG(cdev, "interface %d (%s/%p) alt 0 --> %d\n",
+					tmp, f->name, f, result);
+
+			reset_config(cdev);
+			goto done;
+		}
+	}
+
+	/* when we return, be sure our power usage is valid */
+	power = 2 * c->bMaxPower;
+done:
+	usb_gadget_vbus_draw(gadget, power);
+	return result;
+}
+
+/**
+ * usb_add_config() - add a configuration to a device.
+ * @cdev: wraps the USB gadget
+ * @config: the configuration, with bConfigurationValue assigned
+ * Context: single threaded during gadget setup
+ *
+ * One of the main tasks of a composite driver's bind() routine is to
+ * add each of the configurations it supports, using this routine.
+ *
+ * This function returns the value of the configuration's bind(), which
+ * is zero for success else a negative errno value.  Binding configurations
+ * assigns global resources including string IDs, and per-configuration
+ * resources such as interface IDs and endpoints.
+ */
+int __init usb_add_config(struct usb_composite_dev *cdev,
+		struct usb_configuration *config)
+{
+	int				status = -EINVAL;
+	struct usb_configuration	*c;
+
+	DBG(cdev, "adding config #%u '%s'/%p\n",
+			config->bConfigurationValue,
+			config->label, config);
+
+	if (!config->bConfigurationValue || !config->bind)
+		goto done;
+
+	/* Prevent duplicate configuration identifiers */
+	list_for_each_entry(c, &cdev->configs, list) {
+		if (c->bConfigurationValue == config->bConfigurationValue) {
+			status = -EBUSY;
+			goto done;
+		}
+	}
+
+	config->cdev = cdev;
+	list_add_tail(&config->list, &cdev->configs);
+
+	INIT_LIST_HEAD(&config->functions);
+	config->next_interface_id = 0;
+
+	status = config->bind(config);
+	if (status < 0) {
+		list_del(&config->list);
+		config->cdev = NULL;
+	} else {
+		unsigned	i;
+
+		DBG(cdev, "cfg %d/%p speeds:%s%s\n",
+			config->bConfigurationValue, config,
+			config->highspeed ? " high" : "",
+			config->fullspeed
+				? (gadget_is_dualspeed(cdev->gadget)
+					? " full"
+					: " full/low")
+				: "");
+
+		for (i = 0; i < MAX_CONFIG_INTERFACES; i++) {
+			struct usb_function	*f = config->interface[i];
+
+			if (!f)
+				continue;
+			DBG(cdev, "  interface %d = %s/%p\n",
+				i, f->name, f);
+		}
+	}
+
+	/* set_alt(), or next config->bind(), sets up
+	 * ep->driver_data as needed.
+	 */
+	usb_ep_autoconfig_reset(cdev->gadget);
+
+done:
+	if (status)
+		DBG(cdev, "added config '%s'/%u --> %d\n", config->label,
+				config->bConfigurationValue, status);
+	return status;
+}
+
+/*-------------------------------------------------------------------------*/
+
+/* We support strings in multiple languages ... string descriptor zero
+ * says which languages are supported.  The typical case will be that
+ * only one language (probably English) is used, with I18N handled on
+ * the host side.
+ */
+
+static void collect_langs(struct usb_gadget_strings **sp, __le16 *buf)
+{
+	const struct usb_gadget_strings	*s;
+	u16				language;
+	__le16				*tmp;
+
+	while (*sp) {
+		s = *sp;
+		language = cpu_to_le16(s->language);
+		for (tmp = buf; *tmp && tmp < &buf[126]; tmp++) {
+			if (*tmp == language)
+				goto repeat;
+		}
+		*tmp++ = language;
+repeat:
+		sp++;
+	}
+}
+
+static int lookup_string(
+	struct usb_gadget_strings	**sp,
+	void				*buf,
+	u16				language,
+	int				id
+)
+{
+	struct usb_gadget_strings	*s;
+	int				value;
+
+	while (*sp) {
+		s = *sp++;
+		if (s->language != language)
+			continue;
+		value = usb_gadget_get_string(s, id, buf);
+		if (value > 0)
+			return value;
+	}
+	return -EINVAL;
+}
+
+static int get_string(struct usb_composite_dev *cdev,
+		void *buf, u16 language, int id)
+{
+	struct usb_configuration	*c;
+	struct usb_function		*f;
+	int				len;
+
+	/* Yes, not only is USB's I18N support probably more than most
+	 * folk will ever care about ... also, it's all supported here.
+	 * (Except for UTF8 support for Unicode's "Astral Planes".)
+	 */
+
+	/* 0 == report all available language codes */
+	if (id == 0) {
+		struct usb_string_descriptor	*s = buf;
+		struct usb_gadget_strings	**sp;
+
+		memset(s, 0, 256);
+		s->bDescriptorType = USB_DT_STRING;
+
+		sp = composite->strings;
+		if (sp)
+			collect_langs(sp, s->wData);
+
+		list_for_each_entry(c, &cdev->configs, list) {
+			sp = c->strings;
+			if (sp)
+				collect_langs(sp, s->wData);
+
+			list_for_each_entry(f, &c->functions, list) {
+				sp = f->strings;
+				if (sp)
+					collect_langs(sp, s->wData);
+			}
+		}
+
+		for (len = 0; s->wData[len] && len <= 126; len++)
+			continue;
+		if (!len)
+			return -EINVAL;
+
+		s->bLength = 2 * (len + 1);
+		return s->bLength;
+	}
+
+	/* Otherwise, look up and return a specified string.  String IDs
+	 * are device-scoped, so we look up each string table we're told
+	 * about.  These lookups are infrequent; simpler-is-better here.
+	 */
+	if (composite->strings) {
+		len = lookup_string(composite->strings, buf, language, id);
+		if (len > 0)
+			return len;
+	}
+	list_for_each_entry(c, &cdev->configs, list) {
+		if (c->strings) {
+			len = lookup_string(c->strings, buf, language, id);
+			if (len > 0)
+				return len;
+		}
+		list_for_each_entry(f, &c->functions, list) {
+			if (!f->strings)
+				continue;
+			len = lookup_string(f->strings, buf, language, id);
+			if (len > 0)
+				return len;
+		}
+	}
+	return -EINVAL;
+}
+
+/**
+ * usb_string_id() - allocate an unused string ID
+ * @cdev: the device whose string descriptor IDs are being allocated
+ * Context: single threaded during gadget setup
+ *
+ * @usb_string_id() is called from bind() callbacks to allocate
+ * string IDs.  Drivers for functions, configurations, or gadgets will
+ * then store that ID in the appropriate descriptors and string table.
+ *
+ * All string identifier should be allocated using this routine, to
+ * ensure that for example different functions don't wrongly assign
+ * different meanings to the same identifier.
+ */
+int __init usb_string_id(struct usb_composite_dev *cdev)
+{
+	if (cdev->next_string_id < 254) {
+		/* string id 0 is reserved */
+		cdev->next_string_id++;
+		return cdev->next_string_id;
+	}
+	return -ENODEV;
+}
+
+/*-------------------------------------------------------------------------*/
+
+static void composite_setup_complete(struct usb_ep *ep, struct usb_request *req)
+{
+	if (req->status || req->actual != req->length)
+		DBG((struct usb_composite_dev *) ep->driver_data,
+				"setup complete --> %d, %d/%d\n",
+				req->status, req->actual, req->length);
+}
+
+/*
+ * The setup() callback implements all the ep0 functionality that's
+ * not handled lower down, in hardware or the hardware driver(like
+ * device and endpoint feature flags, and their status).  It's all
+ * housekeeping for the gadget function we're implementing.  Most of
+ * the work is in config and function specific setup.
+ */
+static int
+composite_setup(struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl)
+{
+	struct usb_composite_dev	*cdev = get_gadget_data(gadget);
+	struct usb_request		*req = cdev->req;
+	int				value = -EOPNOTSUPP;
+	u16				w_index = le16_to_cpu(ctrl->wIndex);
+	u16				w_value = le16_to_cpu(ctrl->wValue);
+	u16				w_length = le16_to_cpu(ctrl->wLength);
+	struct usb_function		*f = NULL;
+
+	/* partial re-init of the response message; the function or the
+	 * gadget might need to intercept e.g. a control-OUT completion
+	 * when we delegate to it.
+	 */
+	req->zero = 0;
+	req->complete = composite_setup_complete;
+	req->length = USB_BUFSIZ;
+	gadget->ep0->driver_data = cdev;
+
+	switch (ctrl->bRequest) {
+
+	/* we handle all standard USB descriptors */
+	case USB_REQ_GET_DESCRIPTOR:
+		if (ctrl->bRequestType != USB_DIR_IN)
+			goto unknown;
+		switch (w_value >> 8) {
+
+		case USB_DT_DEVICE:
+			cdev->desc.bNumConfigurations =
+				count_configs(cdev, USB_DT_DEVICE);
+			value = min(w_length, (u16) sizeof cdev->desc);
+			memcpy(req->buf, &cdev->desc, value);
+			break;
+		case USB_DT_DEVICE_QUALIFIER:
+			if (!gadget_is_dualspeed(gadget))
+				break;
+			device_qual(cdev);
+			value = min_t(int, w_length,
+				sizeof(struct usb_qualifier_descriptor));
+			break;
+		case USB_DT_OTHER_SPEED_CONFIG:
+			if (!gadget_is_dualspeed(gadget))
+				break;
+			/* FALLTHROUGH */
+		case USB_DT_CONFIG:
+			value = config_desc(cdev, w_value);
+			if (value >= 0)
+				value = min(w_length, (u16) value);
+			break;
+		case USB_DT_STRING:
+			value = get_string(cdev, req->buf,
+					w_index, w_value & 0xff);
+			if (value >= 0)
+				value = min(w_length, (u16) value);
+			break;
+		}
+		break;
+
+	/* any number of configs can work */
+	case USB_REQ_SET_CONFIGURATION:
+		if (ctrl->bRequestType != 0)
+			goto unknown;
+		if (gadget_is_otg(gadget)) {
+			if (gadget->a_hnp_support)
+				DBG(cdev, "HNP available\n");
+			else if (gadget->a_alt_hnp_support)
+				DBG(cdev, "HNP on another port\n");
+			else
+				VDBG(cdev, "HNP inactive\n");
+		}
+		spin_lock(&cdev->lock);
+		value = set_config(cdev, ctrl, w_value);
+		spin_unlock(&cdev->lock);
+		break;
+	case USB_REQ_GET_CONFIGURATION:
+		if (ctrl->bRequestType != USB_DIR_IN)
+			goto unknown;
+		if (cdev->config)
+			*(u8 *)req->buf = cdev->config->bConfigurationValue;
+		else
+			*(u8 *)req->buf = 0;
+		value = min(w_length, (u16) 1);
+		break;
+
+	/* function drivers must handle get/set altsetting; if there's
+	 * no get() method, we know only altsetting zero works.
+	 */
+	case USB_REQ_SET_INTERFACE:
+		if (ctrl->bRequestType != USB_RECIP_INTERFACE)
+			goto unknown;
+		if (!cdev->config || w_index >= MAX_CONFIG_INTERFACES)
+			break;
+		f = cdev->config->interface[w_index];
+		if (!f)
+			break;
+		if (w_value && !f->get_alt)
+			break;
+		value = f->set_alt(f, w_index, w_value);
+		break;
+	case USB_REQ_GET_INTERFACE:
+		if (ctrl->bRequestType != (USB_DIR_IN|USB_RECIP_INTERFACE))
+			goto unknown;
+		if (!cdev->config || w_index >= MAX_CONFIG_INTERFACES)
+			break;
+		f = cdev->config->interface[w_index];
+		if (!f)
+			break;
+		/* lots of interfaces only need altsetting zero... */
+		value = f->get_alt ? f->get_alt(f, w_index) : 0;
+		if (value < 0)
+			break;
+		*((u8 *)req->buf) = value;
+		value = min(w_length, (u16) 1);
+		break;
+	default:
+unknown:
+		VDBG(cdev,
+			"non-core control req%02x.%02x v%04x i%04x l%d\n",
+			ctrl->bRequestType, ctrl->bRequest,
+			w_value, w_index, w_length);
+
+		/* functions always handle their interfaces ... punt other
+		 * recipients (endpoint, other, WUSB, ...) to the current
+		 * configuration code.
+		 *
+		 * REVISIT it could make sense to let the composite device
+		 * take such requests too, if that's ever needed:  to work
+		 * in config 0, etc.
+		 */
+		if ((ctrl->bRequestType & USB_RECIP_MASK)
+				== USB_RECIP_INTERFACE) {
+			f = cdev->config->interface[w_index];
+			if (f && f->setup)
+				value = f->setup(f, ctrl);
+			else
+				f = NULL;
+		}
+		if (value < 0 && !f) {
+			struct usb_configuration	*c;
+
+			c = cdev->config;
+			if (c && c->setup)
+				value = c->setup(c, ctrl);
+		}
+
+		goto done;
+	}
+
+	/* respond with data transfer before status phase? */
+	if (value >= 0) {
+		req->length = value;
+		req->zero = value < w_length;
+		value = usb_ep_queue(gadget->ep0, req, GFP_ATOMIC);
+		if (value < 0) {
+			DBG(cdev, "ep_queue --> %d\n", value);
+			req->status = 0;
+			composite_setup_complete(gadget->ep0, req);
+		}
+	}
+
+done:
+	/* device either stalls (value < 0) or reports success */
+	return value;
+}
+
+static void composite_disconnect(struct usb_gadget *gadget)
+{
+	struct usb_composite_dev	*cdev = get_gadget_data(gadget);
+	unsigned long			flags;
+
+	/* REVISIT:  should we have config and device level
+	 * disconnect callbacks?
+	 */
+	spin_lock_irqsave(&cdev->lock, flags);
+	if (cdev->config)
+		reset_config(cdev);
+	spin_unlock_irqrestore(&cdev->lock, flags);
+}
+
+/*-------------------------------------------------------------------------*/
+
+static void /* __init_or_exit */
+composite_unbind(struct usb_gadget *gadget)
+{
+	struct usb_composite_dev	*cdev = get_gadget_data(gadget);
+
+	/* composite_disconnect() must already have been called
+	 * by the underlying peripheral controller driver!
+	 * so there's no i/o concurrency that could affect the
+	 * state protected by cdev->lock.
+	 */
+	WARN_ON(cdev->config);
+
+	while (!list_empty(&cdev->configs)) {
+		struct usb_configuration	*c;
+
+		c = list_first_entry(&cdev->configs,
+				struct usb_configuration, list);
+		while (!list_empty(&c->functions)) {
+			struct usb_function		*f;
+
+			f = list_first_entry(&c->functions,
+					struct usb_function, list);
+			list_del(&f->list);
+			if (f->unbind) {
+				DBG(cdev, "unbind function '%s'/%p\n",
+						f->name, f);
+				f->unbind(c, f);
+				/* may free memory for "f" */
+			}
+		}
+		list_del(&c->list);
+		if (c->unbind) {
+			DBG(cdev, "unbind config '%s'/%p\n", c->label, c);
+			c->unbind(c);
+			/* may free memory for "c" */
+		}
+	}
+	if (composite->unbind)
+		composite->unbind(cdev);
+
+	if (cdev->req) {
+		kfree(cdev->req->buf);
+		usb_ep_free_request(gadget->ep0, cdev->req);
+	}
+	kfree(cdev);
+	set_gadget_data(gadget, NULL);
+	composite = NULL;
+}
+
+static void __init
+string_override_one(struct usb_gadget_strings *tab, u8 id, const char *s)
+{
+	struct usb_string		*str = tab->strings;
+
+	for (str = tab->strings; str->s; str++) {
+		if (str->id == id) {
+			str->s = s;
+			return;
+		}
+	}
+}
+
+static void __init
+string_override(struct usb_gadget_strings **tab, u8 id, const char *s)
+{
+	while (*tab) {
+		string_override_one(*tab, id, s);
+		tab++;
+	}
+}
+
+static int __init composite_bind(struct usb_gadget *gadget)
+{
+	struct usb_composite_dev	*cdev;
+	int				status = -ENOMEM;
+
+	cdev = kzalloc(sizeof *cdev, GFP_KERNEL);
+	if (!cdev)
+		return status;
+
+	spin_lock_init(&cdev->lock);
+	cdev->gadget = gadget;
+	set_gadget_data(gadget, cdev);
+	INIT_LIST_HEAD(&cdev->configs);
+
+	/* preallocate control response and buffer */
+	cdev->req = usb_ep_alloc_request(gadget->ep0, GFP_KERNEL);
+	if (!cdev->req)
+		goto fail;
+	cdev->req->buf = kmalloc(USB_BUFSIZ, GFP_KERNEL);
+	if (!cdev->req->buf)
+		goto fail;
+	cdev->req->complete = composite_setup_complete;
+	gadget->ep0->driver_data = cdev;
+
+	cdev->bufsiz = USB_BUFSIZ;
+	cdev->driver = composite;
+
+	usb_gadget_set_selfpowered(gadget);
+
+	/* interface and string IDs start at zero via kzalloc.
+	 * we force endpoints to start unassigned; few controller
+	 * drivers will zero ep->driver_data.
+	 */
+	usb_ep_autoconfig_reset(cdev->gadget);
+
+	/* composite gadget needs to assign strings for whole device (like
+	 * serial number), register function drivers, potentially update
+	 * power state and consumption, etc
+	 */
+	status = composite->bind(cdev);
+	if (status < 0)
+		goto fail;
+
+	cdev->desc = *composite->dev;
+	cdev->desc.bMaxPacketSize0 = gadget->ep0->maxpacket;
+
+	/* standardized runtime overrides for device ID data */
+	if (idVendor)
+		cdev->desc.idVendor = cpu_to_le16(idVendor);
+	if (idProduct)
+		cdev->desc.idProduct = cpu_to_le16(idProduct);
+	if (bcdDevice)
+		cdev->desc.bcdDevice = cpu_to_le16(bcdDevice);
+
+	/* strings can't be assigned before bind() allocates the
+	 * releavnt identifiers
+	 */
+	if (cdev->desc.iManufacturer && iManufacturer)
+		string_override(composite->strings,
+			cdev->desc.iManufacturer, iManufacturer);
+	if (cdev->desc.iProduct && iProduct)
+		string_override(composite->strings,
+			cdev->desc.iProduct, iProduct);
+	if (cdev->desc.iSerialNumber && iSerialNumber)
+		string_override(composite->strings,
+			cdev->desc.iSerialNumber, iSerialNumber);
+
+	INFO(cdev, "%s ready\n", composite->name);
+	return 0;
+
+fail:
+	composite_unbind(gadget);
+	return status;
+}
+
+/*-------------------------------------------------------------------------*/
+
+static void
+composite_suspend(struct usb_gadget *gadget)
+{
+	struct usb_composite_dev	*cdev = get_gadget_data(gadget);
+	struct usb_function		*f;
+
+	/* REVISIT:  should we have config and device level
+	 * suspend/resume callbacks?
+	 */
+	DBG(cdev, "suspend\n");
+	if (cdev->config) {
+		list_for_each_entry(f, &cdev->config->functions, list) {
+			if (f->suspend)
+				f->suspend(f);
+		}
+	}
+}
+
+static void
+composite_resume(struct usb_gadget *gadget)
+{
+	struct usb_composite_dev	*cdev = get_gadget_data(gadget);
+	struct usb_function		*f;
+
+	/* REVISIT:  should we have config and device level
+	 * suspend/resume callbacks?
+	 */
+	DBG(cdev, "resume\n");
+	if (cdev->config) {
+		list_for_each_entry(f, &cdev->config->functions, list) {
+			if (f->resume)
+				f->resume(f);
+		}
+	}
+}
+
+/*-------------------------------------------------------------------------*/
+
+static struct usb_gadget_driver composite_driver = {
+	.speed		= USB_SPEED_HIGH,
+
+	.bind		= composite_bind,
+	.unbind		= __exit_p(composite_unbind),
+
+	.setup		= composite_setup,
+	.disconnect	= composite_disconnect,
+
+	.suspend	= composite_suspend,
+	.resume		= composite_resume,
+
+	.driver	= {
+		.owner		= THIS_MODULE,
+	},
+};
+
+/**
+ * usb_composite_register() - register a composite driver
+ * @driver: the driver to register
+ * Context: single threaded during gadget setup
+ *
+ * This function is used to register drivers using the composite driver
+ * framework.  The return value is zero, or a negative errno value.
+ * Those values normally come from the driver's @bind method, which does
+ * all the work of setting up the driver to match the hardware.
+ *
+ * On successful return, the gadget is ready to respond to requests from
+ * the host, unless one of its components invokes usb_gadget_disconnect()
+ * while it was binding.  That would usually be done in order to wait for
+ * some userspace participation.
+ */
+int __init usb_composite_register(struct usb_composite_driver *driver)
+{
+	if (!driver || !driver->dev || !driver->bind || composite)
+		return -EINVAL;
+
+	if (!driver->name)
+		driver->name = "composite";
+	composite_driver.function =  (char *) driver->name;
+	composite_driver.driver.name = driver->name;
+	composite = driver;
+
+	return usb_gadget_register_driver(&composite_driver);
+}
+
+/**
+ * usb_composite_unregister() - unregister a composite driver
+ * @driver: the driver to unregister
+ *
+ * This function is used to unregister drivers using the composite
+ * driver framework.
+ */
+void __exit usb_composite_unregister(struct usb_composite_driver *driver)
+{
+	if (composite != driver)
+		return;
+	usb_gadget_unregister_driver(&composite_driver);
+}
diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h
new file mode 100644
index 000000000000..747c3a49cdc9
--- /dev/null
+++ b/include/linux/usb/composite.h
@@ -0,0 +1,338 @@
+/*
+ * composite.h -- framework for usb gadgets which are composite devices
+ *
+ * Copyright (C) 2006-2008 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#ifndef	__LINUX_USB_COMPOSITE_H
+#define	__LINUX_USB_COMPOSITE_H
+
+/*
+ * This framework is an optional layer on top of the USB Gadget interface,
+ * making it easier to build (a) Composite devices, supporting multiple
+ * functions within any single configuration, and (b) Multi-configuration
+ * devices, also supporting multiple functions but without necessarily
+ * having more than one function per configuration.
+ *
+ * Example:  a device with a single configuration supporting both network
+ * link and mass storage functions is a composite device.  Those functions
+ * might alternatively be packaged in individual configurations, but in
+ * the composite model the host can use both functions at the same time.
+ */
+
+#include <linux/usb/ch9.h>
+#include <linux/usb/gadget.h>
+
+
+struct usb_configuration;
+
+/**
+ * struct usb_function - describes one function of a configuration
+ * @name: For diagnostics, identifies the function.
+ * @strings: tables of strings, keyed by identifiers assigned during bind()
+ *	and by language IDs provided in control requests
+ * @descriptors: Table of full (or low) speed descriptors, using interface and
+ *	string identifiers assigned during @bind().  If this pointer is null,
+ *	the function will not be available at full speed (or at low speed).
+ * @hs_descriptors: Table of high speed descriptors, using interface and
+ *	string identifiers assigned during @bind().  If this pointer is null,
+ *	the function will not be available at high speed.
+ * @config: assigned when @usb_add_function() is called; this is the
+ *	configuration with which this function is associated.
+ * @bind: Before the gadget can register, all of its functions bind() to the
+ *	available resources including string and interface identifiers used
+ *	in interface or class descriptors; endpoints; I/O buffers; and so on.
+ * @unbind: Reverses @bind; called as a side effect of unregistering the
+ *	driver which added this function.
+ * @set_alt: (REQUIRED) Reconfigures altsettings; function drivers may
+ *	initialize usb_ep.driver data at this time (when it is used).
+ *	Note that setting an interface to its current altsetting resets
+ *	interface state, and that all interfaces have a disabled state.
+ * @get_alt: Returns the active altsetting.  If this is not provided,
+ *	then only altsetting zero is supported.
+ * @disable: (REQUIRED) Indicates the function should be disabled.  Reasons
+ *	include host resetting or reconfiguring the gadget, and disconnection.
+ * @setup: Used for interface-specific control requests.
+ * @suspend: Notifies functions when the host stops sending USB traffic.
+ * @resume: Notifies functions when the host restarts USB traffic.
+ *
+ * A single USB function uses one or more interfaces, and should in most
+ * cases support operation at both full and high speeds.  Each function is
+ * associated by @usb_add_function() with a one configuration; that function
+ * causes @bind() to be called so resources can be allocated as part of
+ * setting up a gadget driver.  Those resources include endpoints, which
+ * should be allocated using @usb_ep_autoconfig().
+ *
+ * To support dual speed operation, a function driver provides descriptors
+ * for both high and full speed operation.  Except in rare cases that don't
+ * involve bulk endpoints, each speed needs different endpoint descriptors.
+ *
+ * Function drivers choose their own strategies for managing instance data.
+ * The simplest strategy just declares it "static', which means the function
+ * can only be activated once.  If the function needs to be exposed in more
+ * than one configuration at a given speed, it needs to support multiple
+ * usb_function structures (one for each configuration).
+ *
+ * A more complex strategy might encapsulate a @usb_function structure inside
+ * a driver-specific instance structure to allows multiple activations.  An
+ * example of multiple activations might be a CDC ACM function that supports
+ * two or more distinct instances within the same configuration, providing
+ * several independent logical data links to a USB host.
+ */
+struct usb_function {
+	const char			*name;
+	struct usb_gadget_strings	**strings;
+	struct usb_descriptor_header	**descriptors;
+	struct usb_descriptor_header	**hs_descriptors;
+
+	struct usb_configuration	*config;
+
+	/* REVISIT:  bind() functions can be marked __init, which
+	 * makes trouble for section mismatch analysis.  See if
+	 * we can't restructure things to avoid mismatching.
+	 * Related:  unbind() may kfree() but bind() won't...
+	 */
+
+	/* configuration management:  bind/unbind */
+	int			(*bind)(struct usb_configuration *,
+					struct usb_function *);
+	void			(*unbind)(struct usb_configuration *,
+					struct usb_function *);
+
+	/* runtime state management */
+	int			(*set_alt)(struct usb_function *,
+					unsigned interface, unsigned alt);
+	int			(*get_alt)(struct usb_function *,
+					unsigned interface);
+	void			(*disable)(struct usb_function *);
+	int			(*setup)(struct usb_function *,
+					const struct usb_ctrlrequest *);
+	void			(*suspend)(struct usb_function *);
+	void			(*resume)(struct usb_function *);
+
+	/* internals */
+	struct list_head		list;
+};
+
+int usb_add_function(struct usb_configuration *, struct usb_function *);
+
+int usb_interface_id(struct usb_configuration *, struct usb_function *);
+
+/**
+ * ep_choose - select descriptor endpoint at current device speed
+ * @g: gadget, connected and running at some speed
+ * @hs: descriptor to use for high speed operation
+ * @fs: descriptor to use for full or low speed operation
+ */
+static inline struct usb_endpoint_descriptor *
+ep_choose(struct usb_gadget *g, struct usb_endpoint_descriptor *hs,
+		struct usb_endpoint_descriptor *fs)
+{
+	if (gadget_is_dualspeed(g) && g->speed == USB_SPEED_HIGH)
+		return hs;
+	return fs;
+}
+
+#define	MAX_CONFIG_INTERFACES		16	/* arbitrary; max 255 */
+
+/**
+ * struct usb_configuration - represents one gadget configuration
+ * @label: For diagnostics, describes the configuration.
+ * @strings: Tables of strings, keyed by identifiers assigned during @bind()
+ *	and by language IDs provided in control requests.
+ * @descriptors: Table of descriptors preceding all function descriptors.
+ *	Examples include OTG and vendor-specific descriptors.
+ * @bind: Called from @usb_add_config() to allocate resources unique to this
+ *	configuration and to call @usb_add_function() for each function used.
+ * @unbind: Reverses @bind; called as a side effect of unregistering the
+ *	driver which added this configuration.
+ * @setup: Used to delegate control requests that aren't handled by standard
+ *	device infrastructure or directed at a specific interface.
+ * @bConfigurationValue: Copied into configuration descriptor.
+ * @iConfiguration: Copied into configuration descriptor.
+ * @bmAttributes: Copied into configuration descriptor.
+ * @bMaxPower: Copied into configuration descriptor.
+ * @cdev: assigned by @usb_add_config() before calling @bind(); this is
+ *	the device associated with this configuration.
+ *
+ * Configurations are building blocks for gadget drivers structured around
+ * function drivers.  Simple USB gadgets require only one function and one
+ * configuration, and handle dual-speed hardware by always providing the same
+ * functionality.  Slightly more complex gadgets may have more than one
+ * single-function configuration at a given speed; or have configurations
+ * that only work at one speed.
+ *
+ * Composite devices are, by definition, ones with configurations which
+ * include more than one function.
+ *
+ * The lifecycle of a usb_configuration includes allocation, initialization
+ * of the fields described above, and calling @usb_add_config() to set up
+ * internal data and bind it to a specific device.  The configuration's
+ * @bind() method is then used to initialize all the functions and then
+ * call @usb_add_function() for them.
+ *
+ * Those functions would normally be independant of each other, but that's
+ * not mandatory.  CDC WMC devices are an example where functions often
+ * depend on other functions, with some functions subsidiary to others.
+ * Such interdependency may be managed in any way, so long as all of the
+ * descriptors complete by the time the composite driver returns from
+ * its bind() routine.
+ */
+struct usb_configuration {
+	const char			*label;
+	struct usb_gadget_strings	**strings;
+	const struct usb_descriptor_header **descriptors;
+
+	/* REVISIT:  bind() functions can be marked __init, which
+	 * makes trouble for section mismatch analysis.  See if
+	 * we can't restructure things to avoid mismatching...
+	 */
+
+	/* configuration management:  bind/unbind */
+	int			(*bind)(struct usb_configuration *);
+	void			(*unbind)(struct usb_configuration *);
+	int			(*setup)(struct usb_configuration *,
+					const struct usb_ctrlrequest *);
+
+	/* fields in the config descriptor */
+	u8			bConfigurationValue;
+	u8			iConfiguration;
+	u8			bmAttributes;
+	u8			bMaxPower;
+
+	struct usb_composite_dev	*cdev;
+
+	/* internals */
+	struct list_head	list;
+	struct list_head	functions;
+	u8			next_interface_id;
+	unsigned		highspeed:1;
+	unsigned		fullspeed:1;
+	struct usb_function	*interface[MAX_CONFIG_INTERFACES];
+};
+
+int usb_add_config(struct usb_composite_dev *,
+		struct usb_configuration *);
+
+/**
+ * struct usb_composite_driver - groups configurations into a gadget
+ * @name: For diagnostics, identifies the driver.
+ * @dev: Template descriptor for the device, including default device
+ *	identifiers.
+ * @strings: tables of strings, keyed by identifiers assigned during bind()
+ *	and language IDs provided in control requests
+ * @bind: (REQUIRED) Used to allocate resources that are shared across the
+ *	whole device, such as string IDs, and add its configurations using
+ *	@usb_add_config().  This may fail by returning a negative errno
+ *	value; it should return zero on successful initialization.
+ * @unbind: Reverses @bind(); called as a side effect of unregistering
+ *	this driver.
+ *
+ * Devices default to reporting self powered operation.  Devices which rely
+ * on bus powered operation should report this in their @bind() method.
+ *
+ * Before returning from @bind, various fields in the template descriptor
+ * may be overridden.  These include the idVendor/idProduct/bcdDevice values
+ * normally to bind the appropriate host side driver, and the three strings
+ * (iManufacturer, iProduct, iSerialNumber) normally used to provide user
+ * meaningful device identifiers.  (The strings will not be defined unless
+ * they are defined in @dev and @strings.)  The correct ep0 maxpacket size
+ * is also reported, as defined by the underlying controller driver.
+ */
+struct usb_composite_driver {
+	const char				*name;
+	const struct usb_device_descriptor	*dev;
+	struct usb_gadget_strings		**strings;
+
+	/* REVISIT:  bind() functions can be marked __init, which
+	 * makes trouble for section mismatch analysis.  See if
+	 * we can't restructure things to avoid mismatching...
+	 */
+
+	int			(*bind)(struct usb_composite_dev *);
+	int			(*unbind)(struct usb_composite_dev *);
+};
+
+extern int usb_composite_register(struct usb_composite_driver *);
+extern void usb_composite_unregister(struct usb_composite_driver *);
+
+
+/**
+ * struct usb_composite_device - represents one composite usb gadget
+ * @gadget: read-only, abstracts the gadget's usb peripheral controller
+ * @req: used for control responses; buffer is pre-allocated
+ * @bufsiz: size of buffer pre-allocated in @req
+ * @config: the currently active configuration
+ *
+ * One of these devices is allocated and initialized before the
+ * associated device driver's bind() is called.
+ *
+ * OPEN ISSUE:  it appears that some WUSB devices will need to be
+ * built by combining a normal (wired) gadget with a wireless one.
+ * This revision of the gadget framework should probably try to make
+ * sure doing that won't hurt too much.
+ *
+ * One notion for how to handle Wireless USB devices involves:
+ * (a) a second gadget here, discovery mechanism TBD, but likely
+ *     needing separate "register/unregister WUSB gadget" calls;
+ * (b) updates to usb_gadget to include flags "is it wireless",
+ *     "is it wired", plus (presumably in a wrapper structure)
+ *     bandgroup and PHY info;
+ * (c) presumably a wireless_ep wrapping a usb_ep, and reporting
+ *     wireless-specific parameters like maxburst and maxsequence;
+ * (d) configurations that are specific to wireless links;
+ * (e) function drivers that understand wireless configs and will
+ *     support wireless for (additional) function instances;
+ * (f) a function to support association setup (like CBAF), not
+ *     necessarily requiring a wireless adapter;
+ * (g) composite device setup that can create one or more wireless
+ *     configs, including appropriate association setup support;
+ * (h) more, TBD.
+ */
+struct usb_composite_dev {
+	struct usb_gadget		*gadget;
+	struct usb_request		*req;
+	unsigned			bufsiz;
+
+	struct usb_configuration	*config;
+
+	/* internals */
+	struct usb_device_descriptor	desc;
+	struct list_head		configs;
+	struct usb_composite_driver	*driver;
+	u8				next_string_id;
+
+	spinlock_t			lock;
+
+	/* REVISIT use and existence of lock ... */
+};
+
+extern int usb_string_id(struct usb_composite_dev *c);
+
+/* messaging utils */
+#define DBG(d, fmt, args...) \
+	dev_dbg(&(d)->gadget->dev , fmt , ## args)
+#define VDBG(d, fmt, args...) \
+	dev_vdbg(&(d)->gadget->dev , fmt , ## args)
+#define ERROR(d, fmt, args...) \
+	dev_err(&(d)->gadget->dev , fmt , ## args)
+#define WARN(d, fmt, args...) \
+	dev_warn(&(d)->gadget->dev , fmt , ## args)
+#define INFO(d, fmt, args...) \
+	dev_info(&(d)->gadget->dev , fmt , ## args)
+
+#endif	/* __LINUX_USB_COMPOSITE_H */
-- 
cgit v1.2.3


From e0d795e4f36c2d8949c1355b497fc5425dbb9437 Mon Sep 17 00:00:00 2001
From: Felipe Balbi <me@felipebalbi.com>
Date: Tue, 3 Jun 2008 14:47:52 +0300
Subject: usb: irda: cleanup on ir-usb module

General cleanup on ir-usb module. Introduced
a common header that could be used also on
usb gadget framework.

Lot's of cleanups and now using macros from the header
file.

Signed-off-by: Felipe Balbi <me@felipebalbi.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/ir-usb.c | 490 ++++++++++++++++++++++++--------------------
 include/linux/usb/irda.h    | 151 ++++++++++++++
 2 files changed, 421 insertions(+), 220 deletions(-)
 create mode 100644 include/linux/usb/irda.h

(limited to 'include/linux')

diff --git a/drivers/usb/serial/ir-usb.c b/drivers/usb/serial/ir-usb.c
index 004d57385a75..0063c11c8081 100644
--- a/drivers/usb/serial/ir-usb.c
+++ b/drivers/usb/serial/ir-usb.c
@@ -19,7 +19,12 @@
  * was written by Roman Weissgaerber <weissg@vienna.at>, Dag Brattli
  * <dag@brattli.net>, and Jean Tourrilhes <jt@hpl.hp.com>
  *
- * See Documentation/usb/usb-serial.txt for more information on using this driver
+ * See Documentation/usb/usb-serial.txt for more information on using this
+ * driver
+ *
+ * 2008_Jun_02  Felipe Balbi <me@felipebalbi.com>
+ *	Introduced common header to be used also in USB Gadget Framework.
+ *	Still needs some other style fixes.
  *
  * 2007_Jun_21  Alan Cox <alan@redhat.com>
  *	Minimal cleanups for some of the driver problens and tty layer abuse.
@@ -59,9 +64,10 @@
 #include <linux/tty_flip.h>
 #include <linux/module.h>
 #include <linux/spinlock.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <linux/usb.h>
 #include <linux/usb/serial.h>
+#include <linux/usb/irda.h>
 
 /*
  * Version Information
@@ -70,100 +76,75 @@
 #define DRIVER_AUTHOR "Greg Kroah-Hartman <greg@kroah.com>"
 #define DRIVER_DESC "USB IR Dongle driver"
 
-/* USB IrDA class spec information */
-#define USB_CLASS_IRDA		0x02
-#define USB_DT_IRDA		0x21
-#define IU_REQ_GET_CLASS_DESC	0x06
-#define SPEED_2400		0x01
-#define SPEED_9600		0x02
-#define SPEED_19200		0x03
-#define SPEED_38400		0x04
-#define SPEED_57600		0x05
-#define SPEED_115200		0x06
-#define SPEED_576000		0x07
-#define SPEED_1152000		0x08
-#define SPEED_4000000		0x09
-
-struct irda_class_desc {
-	u8	bLength;
-	u8	bDescriptorType;
-	u16	bcdSpecRevision;
-	u8	bmDataSize;
-	u8	bmWindowSize;
-	u8	bmMinTurnaroundTime;
-	u16	wBaudRate;
-	u8	bmAdditionalBOFs;
-	u8	bIrdaRateSniff;
-	u8	bMaxUnicastList;
-} __attribute__ ((packed));
-
 static int debug;
 
 /* if overridden by the user, then use their value for the size of the read and
  * write urbs */
 static int buffer_size;
+
 /* if overridden by the user, then use the specified number of XBOFs */
 static int xbof = -1;
 
-static int  ir_startup (struct usb_serial *serial);
-static int  ir_open (struct usb_serial_port *port, struct file *filep);
-static void ir_close (struct usb_serial_port *port, struct file *filep);
-static int  ir_write (struct usb_serial_port *port, const unsigned char *buf, int count);
-static void ir_write_bulk_callback (struct urb *urb);
-static void ir_read_bulk_callback (struct urb *urb);
-static void ir_set_termios (struct usb_serial_port *port, struct ktermios *old_termios);
+static int ir_startup(struct usb_serial *serial);
+static int ir_open(struct usb_serial_port *port, struct file *filep);
+static void ir_close(struct usb_serial_port *port, struct file *filep);
+static int ir_write(struct usb_serial_port *port,
+		const unsigned char *buf, int count);
+static void ir_write_bulk_callback(struct urb *urb);
+static void ir_read_bulk_callback(struct urb *urb);
+static void ir_set_termios(struct usb_serial_port *port,
+		struct ktermios *old_termios);
 
 /* Not that this lot means you can only have one per system */
-static u8 ir_baud = 0;
-static u8 ir_xbof = 0;
-static u8 ir_add_bof = 0;
+static u8 ir_baud;
+static u8 ir_xbof;
+static u8 ir_add_bof;
 
-static struct usb_device_id id_table [] = {
+static struct usb_device_id ir_id_table[] = {
 	{ USB_DEVICE(0x050f, 0x0180) },		/* KC Technology, KC-180 */
 	{ USB_DEVICE(0x08e9, 0x0100) },		/* XTNDAccess */
 	{ USB_DEVICE(0x09c4, 0x0011) },		/* ACTiSys ACT-IR2000U */
-	{ USB_INTERFACE_INFO (USB_CLASS_APP_SPEC, USB_CLASS_IRDA, 0) },
+	{ USB_INTERFACE_INFO(USB_CLASS_APP_SPEC, USB_SUBCLASS_IRDA, 0) },
 	{ }					/* Terminating entry */
 };
 
-MODULE_DEVICE_TABLE (usb, id_table);
+MODULE_DEVICE_TABLE(usb, ir_id_table);
 
 static struct usb_driver ir_driver = {
-	.name =		"ir-usb",
-	.probe =	usb_serial_probe,
-	.disconnect =	usb_serial_disconnect,
-	.id_table =	id_table,
-	.no_dynamic_id = 	1,
+	.name		= "ir-usb",
+	.probe		= usb_serial_probe,
+	.disconnect	= usb_serial_disconnect,
+	.id_table	= ir_id_table,
+	.no_dynamic_id	= 1,
 };
 
-
 static struct usb_serial_driver ir_device = {
-	.driver = {
-		.owner =	THIS_MODULE,
-		.name =		"ir-usb",
+	.driver	= {
+		.owner	= THIS_MODULE,
+		.name	= "ir-usb",
 	},
-	.description =		"IR Dongle",
-	.usb_driver = 		&ir_driver,
-	.id_table =		id_table,
-	.num_ports =		1,
-	.set_termios =		ir_set_termios,
-	.attach =		ir_startup,
-	.open =			ir_open,
-	.close =		ir_close,
-	.write =		ir_write,
-	.write_bulk_callback =	ir_write_bulk_callback,
-	.read_bulk_callback =	ir_read_bulk_callback,
+	.description		= "IR Dongle",
+	.usb_driver		= &ir_driver,
+	.id_table		= ir_id_table,
+	.num_ports		= 1,
+	.set_termios		= ir_set_termios,
+	.attach			= ir_startup,
+	.open			= ir_open,
+	.close			= ir_close,
+	.write			= ir_write,
+	.write_bulk_callback	= ir_write_bulk_callback,
+	.read_bulk_callback	= ir_read_bulk_callback,
 };
 
-static inline void irda_usb_dump_class_desc(struct irda_class_desc *desc)
+static inline void irda_usb_dump_class_desc(struct usb_irda_cs_descriptor *desc)
 {
 	dbg("bLength=%x", desc->bLength);
 	dbg("bDescriptorType=%x", desc->bDescriptorType);
-	dbg("bcdSpecRevision=%x", desc->bcdSpecRevision); 
+	dbg("bcdSpecRevision=%x", __le16_to_cpu(desc->bcdSpecRevision));
 	dbg("bmDataSize=%x", desc->bmDataSize);
 	dbg("bmWindowSize=%x", desc->bmWindowSize);
 	dbg("bmMinTurnaroundTime=%d", desc->bmMinTurnaroundTime);
-	dbg("wBaudRate=%x", desc->wBaudRate);
+	dbg("wBaudRate=%x", __le16_to_cpu(desc->wBaudRate));
 	dbg("bmAdditionalBOFs=%x", desc->bmAdditionalBOFs);
 	dbg("bIrdaRateSniff=%x", desc->bIrdaRateSniff);
 	dbg("bMaxUnicastList=%x", desc->bMaxUnicastList);
@@ -181,35 +162,37 @@ static inline void irda_usb_dump_class_desc(struct irda_class_desc *desc)
  *
  * Based on the same function in drivers/net/irda/irda-usb.c
  */
-static struct irda_class_desc *irda_usb_find_class_desc(struct usb_device *dev, unsigned int ifnum)
+static struct usb_irda_cs_descriptor *
+irda_usb_find_class_desc(struct usb_device *dev, unsigned int ifnum)
 {
-	struct irda_class_desc *desc;
+	struct usb_irda_cs_descriptor *desc;
 	int ret;
-		
-	desc = kzalloc(sizeof (struct irda_class_desc), GFP_KERNEL);
-	if (desc == NULL) 
+
+	desc = kzalloc(sizeof(*desc), GFP_KERNEL);
+	if (!desc)
 		return NULL;
-	
-	ret = usb_control_msg(dev, usb_rcvctrlpipe(dev,0),
-			IU_REQ_GET_CLASS_DESC,
+
+	ret = usb_control_msg(dev, usb_rcvctrlpipe(dev, 0),
+			USB_REQ_CS_IRDA_GET_CLASS_DESC,
 			USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE,
 			0, ifnum, desc, sizeof(*desc), 1000);
-	
+
 	dbg("%s -  ret=%d", __func__, ret);
 	if (ret < sizeof(*desc)) {
 		dbg("%s - class descriptor read %s (%d)",
 				__func__,
-				(ret<0) ? "failed" : "too short",
+				(ret < 0) ? "failed" : "too short",
 				ret);
 		goto error;
 	}
-	if (desc->bDescriptorType != USB_DT_IRDA) {
+	if (desc->bDescriptorType != USB_DT_CS_IRDA) {
 		dbg("%s - bad class descriptor type", __func__);
 		goto error;
 	}
-	
+
 	irda_usb_dump_class_desc(desc);
 	return desc;
+
 error:
 	kfree(desc);
 	return NULL;
@@ -219,64 +202,100 @@ error:
 static u8 ir_xbof_change(u8 xbof)
 {
 	u8 result;
+
 	/* reference irda-usb.c */
-	switch(xbof) {
-		case 48: result = 0x10; break;
-		case 28:
-		case 24: result = 0x20; break;
-		default:
-		case 12: result = 0x30; break;
-		case  5:
-		case  6: result = 0x40; break;
-		case  3: result = 0x50; break;
-		case  2: result = 0x60; break;
-		case  1: result = 0x70; break;
-		case  0: result = 0x80; break;
+	switch (xbof) {
+	case 48:
+		result = 0x10;
+		break;
+	case 28:
+	case 24:
+		result = 0x20;
+		break;
+	default:
+	case 12:
+		result = 0x30;
+		break;
+	case  5:
+	case  6:
+		result = 0x40;
+		break;
+	case  3:
+		result = 0x50;
+		break;
+	case  2:
+		result = 0x60;
+		break;
+	case  1:
+		result = 0x70;
+		break;
+	case  0:
+		result = 0x80;
+		break;
 	}
+
 	return(result);
 }
 
 
-static int ir_startup (struct usb_serial *serial)
+static int ir_startup(struct usb_serial *serial)
 {
-	struct irda_class_desc *irda_desc;
+	struct usb_irda_cs_descriptor *irda_desc;
 
-	irda_desc = irda_usb_find_class_desc (serial->dev, 0);
-	if (irda_desc == NULL) {
-		dev_err (&serial->dev->dev, "IRDA class descriptor not found, device not bound\n");
+	irda_desc = irda_usb_find_class_desc(serial->dev, 0);
+	if (!irda_desc) {
+		dev_err(&serial->dev->dev,
+			"IRDA class descriptor not found, device not bound\n");
 		return -ENODEV;
 	}
 
-	dbg ("%s - Baud rates supported:%s%s%s%s%s%s%s%s%s",
+	dbg("%s - Baud rates supported:%s%s%s%s%s%s%s%s%s",
 		__func__,
-		(irda_desc->wBaudRate & 0x0001) ? " 2400"    : "",
-		(irda_desc->wBaudRate & 0x0002) ? " 9600"    : "",
-		(irda_desc->wBaudRate & 0x0004) ? " 19200"   : "",
-		(irda_desc->wBaudRate & 0x0008) ? " 38400"   : "",
-		(irda_desc->wBaudRate & 0x0010) ? " 57600"   : "",
-		(irda_desc->wBaudRate & 0x0020) ? " 115200"  : "",
-		(irda_desc->wBaudRate & 0x0040) ? " 576000"  : "",
-		(irda_desc->wBaudRate & 0x0080) ? " 1152000" : "",
-		(irda_desc->wBaudRate & 0x0100) ? " 4000000" : "");
-
-	switch( irda_desc->bmAdditionalBOFs ) {
-		case 0x01: ir_add_bof = 48; break;
-		case 0x02: ir_add_bof = 24; break;
-		case 0x04: ir_add_bof = 12; break;
-		case 0x08: ir_add_bof =  6; break;
-		case 0x10: ir_add_bof =  3; break;
-		case 0x20: ir_add_bof =  2; break;
-		case 0x40: ir_add_bof =  1; break;
-		case 0x80: ir_add_bof =  0; break;
-		default:;
+		(irda_desc->wBaudRate & USB_IRDA_BR_2400) ? " 2400" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_9600) ? " 9600" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_19200) ? " 19200" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_38400) ? " 38400" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_57600) ? " 57600" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_115200) ? " 115200" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_576000) ? " 576000" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_1152000) ? " 1152000" : "",
+		(irda_desc->wBaudRate & USB_IRDA_BR_4000000) ? " 4000000" : "");
+
+	switch (irda_desc->bmAdditionalBOFs) {
+	case USB_IRDA_AB_48:
+		ir_add_bof = 48;
+		break;
+	case USB_IRDA_AB_24:
+		ir_add_bof = 24;
+		break;
+	case USB_IRDA_AB_12:
+		ir_add_bof = 12;
+		break;
+	case USB_IRDA_AB_6:
+		ir_add_bof = 6;
+		break;
+	case USB_IRDA_AB_3:
+		ir_add_bof = 3;
+		break;
+	case USB_IRDA_AB_2:
+		ir_add_bof = 2;
+		break;
+	case USB_IRDA_AB_1:
+		ir_add_bof = 1;
+		break;
+	case USB_IRDA_AB_0:
+		ir_add_bof = 0;
+		break;
+	default:
+		break;
 	}
 
-	kfree (irda_desc);
+	kfree(irda_desc);
 
-	return 0;		
+	return 0;
 }
 
-static int ir_open (struct usb_serial_port *port, struct file *filp)
+static int ir_open(struct usb_serial_port *port, struct file *filp)
 {
 	char *buffer;
 	int result = 0;
@@ -285,51 +304,55 @@ static int ir_open (struct usb_serial_port *port, struct file *filp)
 
 	if (buffer_size) {
 		/* override the default buffer sizes */
-		buffer = kmalloc (buffer_size, GFP_KERNEL);
+		buffer = kmalloc(buffer_size, GFP_KERNEL);
 		if (!buffer) {
-			dev_err (&port->dev, "%s - out of memory.\n", __func__);
+			dev_err(&port->dev, "%s - out of memory.\n", __func__);
 			return -ENOMEM;
 		}
-		kfree (port->read_urb->transfer_buffer);
+		kfree(port->read_urb->transfer_buffer);
 		port->read_urb->transfer_buffer = buffer;
 		port->read_urb->transfer_buffer_length = buffer_size;
 
-		buffer = kmalloc (buffer_size, GFP_KERNEL);
+		buffer = kmalloc(buffer_size, GFP_KERNEL);
 		if (!buffer) {
-			dev_err (&port->dev, "%s - out of memory.\n", __func__);
+			dev_err(&port->dev, "%s - out of memory.\n", __func__);
 			return -ENOMEM;
 		}
-		kfree (port->write_urb->transfer_buffer);
+		kfree(port->write_urb->transfer_buffer);
 		port->write_urb->transfer_buffer = buffer;
 		port->write_urb->transfer_buffer_length = buffer_size;
 		port->bulk_out_size = buffer_size;
 	}
 
 	/* Start reading from the device */
-	usb_fill_bulk_urb (
+	usb_fill_bulk_urb(
 		port->read_urb,
-		port->serial->dev, 
-		usb_rcvbulkpipe(port->serial->dev, port->bulk_in_endpointAddress),
+		port->serial->dev,
+		usb_rcvbulkpipe(port->serial->dev,
+			port->bulk_in_endpointAddress),
 		port->read_urb->transfer_buffer,
 		port->read_urb->transfer_buffer_length,
 		ir_read_bulk_callback,
 		port);
 	result = usb_submit_urb(port->read_urb, GFP_KERNEL);
 	if (result)
-		dev_err(&port->dev, "%s - failed submitting read urb, error %d\n", __func__, result);
+		dev_err(&port->dev,
+			"%s - failed submitting read urb, error %d\n",
+			__func__, result);
 
 	return result;
 }
 
-static void ir_close (struct usb_serial_port *port, struct file * filp)
+static void ir_close(struct usb_serial_port *port, struct file *filp)
 {
 	dbg("%s - port %d", __func__, port->number);
-			 
+
 	/* shutdown our bulk read */
 	usb_kill_urb(port->read_urb);
 }
 
-static int ir_write (struct usb_serial_port *port, const unsigned char *buf, int count)
+static int ir_write(struct usb_serial_port *port,
+		const unsigned char *buf, int count)
 {
 	unsigned char *transfer_buffer;
 	int result;
@@ -338,7 +361,7 @@ static int ir_write (struct usb_serial_port *port, const unsigned char *buf, int
 	dbg("%s - port = %d, count = %d", __func__, port->number, count);
 
 	if (!port->tty) {
-		dev_err (&port->dev, "%s - no tty???\n", __func__);
+		dev_err(&port->dev, "%s - no tty???\n", __func__);
 		return 0;
 	}
 
@@ -359,7 +382,7 @@ static int ir_write (struct usb_serial_port *port, const unsigned char *buf, int
 
 	/*
 	 * The first byte of the packet we send to the device contains an
-	 * inband header which indicates an additional number of BOFs and
+	 * inbound header which indicates an additional number of BOFs and
 	 * a baud rate change.
 	 *
 	 * See section 5.4.2.2 of the USB IrDA spec.
@@ -367,9 +390,9 @@ static int ir_write (struct usb_serial_port *port, const unsigned char *buf, int
 	*transfer_buffer = ir_xbof | ir_baud;
 	++transfer_buffer;
 
-	memcpy (transfer_buffer, buf, transfer_size);
+	memcpy(transfer_buffer, buf, transfer_size);
 
-	usb_fill_bulk_urb (
+	usb_fill_bulk_urb(
 		port->write_urb,
 		port->serial->dev,
 		usb_sndbulkpipe(port->serial->dev,
@@ -381,17 +404,19 @@ static int ir_write (struct usb_serial_port *port, const unsigned char *buf, int
 
 	port->write_urb->transfer_flags = URB_ZERO_PACKET;
 
-	result = usb_submit_urb (port->write_urb, GFP_ATOMIC);
+	result = usb_submit_urb(port->write_urb, GFP_ATOMIC);
 	if (result) {
 		port->write_urb_busy = 0;
-		dev_err(&port->dev, "%s - failed submitting write urb, error %d\n", __func__, result);
+		dev_err(&port->dev,
+			"%s - failed submitting write urb, error %d\n",
+			__func__, result);
 	} else
 		result = transfer_size;
 
 	return result;
 }
 
-static void ir_write_bulk_callback (struct urb *urb)
+static void ir_write_bulk_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	int status = urb->status;
@@ -405,7 +430,7 @@ static void ir_write_bulk_callback (struct urb *urb)
 		return;
 	}
 
-	usb_serial_debug_data (
+	usb_serial_debug_data(
 		debug,
 		&port->dev,
 		__func__,
@@ -415,7 +440,7 @@ static void ir_write_bulk_callback (struct urb *urb)
 	usb_serial_port_softint(port);
 }
 
-static void ir_read_bulk_callback (struct urb *urb)
+static void ir_read_bulk_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	struct tty_struct *tty;
@@ -431,68 +456,69 @@ static void ir_read_bulk_callback (struct urb *urb)
 	}
 
 	switch (status) {
-		case 0: /* Successful */
-
-			/*
-			 * The first byte of the packet we get from the device
-			 * contains a busy indicator and baud rate change.
-			 * See section 5.4.1.2 of the USB IrDA spec.
-			 */
-			if ((*data & 0x0f) > 0)
-				ir_baud = *data & 0x0f;
-
-			usb_serial_debug_data (
-				debug,
-				&port->dev,
-				__func__,
-				urb->actual_length,
-				data);
-
-			tty = port->tty;
-
-			if (tty_buffer_request_room(tty, urb->actual_length - 1)) {
-				tty_insert_flip_string(tty, data+1, urb->actual_length - 1);
-				tty_flip_buffer_push(tty);
-			}
-
-			/*
-			 * No break here.
-			 * We want to resubmit the urb so we can read
-			 * again.
-			 */
-
-		case -EPROTO: /* taking inspiration from pl2303.c */
-
-			/* Continue trying to always read */
-			usb_fill_bulk_urb (
-				port->read_urb,
-				port->serial->dev, 
-				usb_rcvbulkpipe(port->serial->dev,
-					port->bulk_in_endpointAddress),
-				port->read_urb->transfer_buffer,
-				port->read_urb->transfer_buffer_length,
-				ir_read_bulk_callback,
-				port);
-
-			result = usb_submit_urb(port->read_urb, GFP_ATOMIC);
-			if (result)
-				dev_err(&port->dev, "%s - failed resubmitting read urb, error %d\n",
-					__func__, result);
-
-			break ;
-
-		default:
-			dbg("%s - nonzero read bulk status received: %d",
-				__func__,
-				status);
-			break ;
+	case 0: /* Successful */
+
+		/*
+		 * The first byte of the packet we get from the device
+		 * contains a busy indicator and baud rate change.
+		 * See section 5.4.1.2 of the USB IrDA spec.
+		 */
+		if ((*data & 0x0f) > 0)
+			ir_baud = *data & 0x0f;
+
+		usb_serial_debug_data(
+			debug,
+			&port->dev,
+			__func__,
+			urb->actual_length,
+			data);
+
+		tty = port->tty;
+
+		if (tty_buffer_request_room(tty, urb->actual_length - 1)) {
+			tty_insert_flip_string(tty, data + 1,
+					urb->actual_length - 1);
+			tty_flip_buffer_push(tty);
+		}
 
+		/*
+		 * No break here.
+		 * We want to resubmit the urb so we can read
+		 * again.
+		 */
+
+	case -EPROTO: /* taking inspiration from pl2303.c */
+
+		/* Continue trying to always read */
+		usb_fill_bulk_urb(
+			port->read_urb,
+			port->serial->dev,
+			usb_rcvbulkpipe(port->serial->dev,
+				port->bulk_in_endpointAddress),
+			port->read_urb->transfer_buffer,
+			port->read_urb->transfer_buffer_length,
+			ir_read_bulk_callback,
+			port);
+
+		result = usb_submit_urb(port->read_urb, GFP_ATOMIC);
+		if (result)
+			dev_err(&port->dev,
+				"%s - failed resubmitting read urb, error %d\n",
+				__func__, result);
+		break;
+
+	default:
+		dbg("%s - nonzero read bulk status received: %d",
+			__func__,
+			status);
+		break;
 	}
 
 	return;
 }
 
-static void ir_set_termios (struct usb_serial_port *port, struct ktermios *old_termios)
+static void ir_set_termios(struct usb_serial_port *port,
+		struct ktermios *old_termios)
 {
 	unsigned char *transfer_buffer;
 	int result;
@@ -510,19 +536,36 @@ static void ir_set_termios (struct usb_serial_port *port, struct ktermios *old_t
 	 */
 
 	switch (baud) {
-		case 2400:	ir_baud = SPEED_2400; break;
-		case 9600:	ir_baud = SPEED_9600; break;
-		case 19200:	ir_baud = SPEED_19200; break;
-		case 38400:	ir_baud = SPEED_38400; break;
-		case 57600:	ir_baud = SPEED_57600; break;
-		case 115200:	ir_baud = SPEED_115200; break;
-		case 576000:	ir_baud = SPEED_576000; break;
-		case 1152000:	ir_baud = SPEED_1152000; break;
-		case 4000000:	ir_baud = SPEED_4000000; break;
-			break;
-		default:
-			ir_baud = SPEED_9600;
-			baud = 9600;
+	case 2400:
+		ir_baud = USB_IRDA_BR_2400;
+		break;
+	case 9600:
+		ir_baud = USB_IRDA_BR_9600;
+		break;
+	case 19200:
+		ir_baud = USB_IRDA_BR_19200;
+		break;
+	case 38400:
+		ir_baud = USB_IRDA_BR_38400;
+		break;
+	case 57600:
+		ir_baud = USB_IRDA_BR_57600;
+		break;
+	case 115200:
+		ir_baud = USB_IRDA_BR_115200;
+		break;
+	case 576000:
+		ir_baud = USB_IRDA_BR_576000;
+		break;
+	case 1152000:
+		ir_baud = USB_IRDA_BR_1152000;
+		break;
+	case 4000000:
+		ir_baud = USB_IRDA_BR_4000000;
+		break;
+	default:
+		ir_baud = USB_IRDA_BR_9600;
+		baud = 9600;
 	}
 
 	if (xbof == -1)
@@ -538,10 +581,11 @@ static void ir_set_termios (struct usb_serial_port *port, struct ktermios *old_t
 	transfer_buffer = port->write_urb->transfer_buffer;
 	*transfer_buffer = ir_xbof | ir_baud;
 
-	usb_fill_bulk_urb (
+	usb_fill_bulk_urb(
 		port->write_urb,
 		port->serial->dev,
-		usb_sndbulkpipe(port->serial->dev, port->bulk_out_endpointAddress),
+		usb_sndbulkpipe(port->serial->dev,
+			port->bulk_out_endpointAddress),
 		port->write_urb->transfer_buffer,
 		1,
 		ir_write_bulk_callback,
@@ -549,38 +593,44 @@ static void ir_set_termios (struct usb_serial_port *port, struct ktermios *old_t
 
 	port->write_urb->transfer_flags = URB_ZERO_PACKET;
 
-	result = usb_submit_urb (port->write_urb, GFP_KERNEL);
+	result = usb_submit_urb(port->write_urb, GFP_KERNEL);
 	if (result)
-		dev_err(&port->dev, "%s - failed submitting write urb, error %d\n", __func__, result);
+		dev_err(&port->dev,
+				"%s - failed submitting write urb, error %d\n",
+				__func__, result);
 
 	/* Only speed changes are supported */
 	tty_termios_copy_hw(port->tty->termios, old_termios);
 	tty_encode_baud_rate(port->tty, baud, baud);
 }
 
-
-static int __init ir_init (void)
+static int __init ir_init(void)
 {
 	int retval;
+
 	retval = usb_serial_register(&ir_device);
 	if (retval)
 		goto failed_usb_serial_register;
+
 	retval = usb_register(&ir_driver);
-	if (retval) 
+	if (retval)
 		goto failed_usb_register;
+
 	info(DRIVER_DESC " " DRIVER_VERSION);
+
 	return 0;
+
 failed_usb_register:
 	usb_serial_deregister(&ir_device);
+
 failed_usb_serial_register:
 	return retval;
 }
 
-
-static void __exit ir_exit (void)
+static void __exit ir_exit(void)
 {
-	usb_deregister (&ir_driver);
-	usb_serial_deregister (&ir_device);
+	usb_deregister(&ir_driver);
+	usb_serial_deregister(&ir_device);
 }
 
 
diff --git a/include/linux/usb/irda.h b/include/linux/usb/irda.h
new file mode 100644
index 000000000000..e345ceaf72d6
--- /dev/null
+++ b/include/linux/usb/irda.h
@@ -0,0 +1,151 @@
+/*
+ * USB IrDA Bridge Device Definition
+ */
+
+#ifndef __LINUX_USB_IRDA_H
+#define __LINUX_USB_IRDA_H
+
+/* This device should use Application-specific class */
+
+#define USB_SUBCLASS_IRDA			0x02
+
+/*-------------------------------------------------------------------------*/
+
+/* Class-Specific requests (bRequest field) */
+
+#define USB_REQ_CS_IRDA_RECEIVING		1
+#define USB_REQ_CS_IRDA_CHECK_MEDIA_BUSY	3
+#define USB_REQ_CS_IRDA_RATE_SNIFF		4
+#define USB_REQ_CS_IRDA_UNICAST_LIST		5
+#define USB_REQ_CS_IRDA_GET_CLASS_DESC		6
+
+/*-------------------------------------------------------------------------*/
+
+/* Class-Specific descriptor */
+
+#define USB_DT_CS_IRDA				0x21
+
+/*-------------------------------------------------------------------------*/
+
+/* Data sizes */
+
+#define USB_IRDA_DS_2048			(1 << 5)
+#define USB_IRDA_DS_1024			(1 << 4)
+#define USB_IRDA_DS_512				(1 << 3)
+#define USB_IRDA_DS_256				(1 << 2)
+#define USB_IRDA_DS_128				(1 << 1)
+#define USB_IRDA_DS_64				(1 << 0)
+
+/* Window sizes */
+
+#define USB_IRDA_WS_7				(1 << 6)
+#define USB_IRDA_WS_6				(1 << 5)
+#define USB_IRDA_WS_5				(1 << 4)
+#define USB_IRDA_WS_4				(1 << 3)
+#define USB_IRDA_WS_3				(1 << 2)
+#define USB_IRDA_WS_2				(1 << 1)
+#define USB_IRDA_WS_1				(1 << 0)
+
+/* Min turnaround times in usecs */
+
+#define USB_IRDA_MTT_0				(1 << 7)
+#define USB_IRDA_MTT_10				(1 << 6)
+#define USB_IRDA_MTT_50				(1 << 5)
+#define USB_IRDA_MTT_100			(1 << 4)
+#define USB_IRDA_MTT_500			(1 << 3)
+#define USB_IRDA_MTT_1000			(1 << 2)
+#define USB_IRDA_MTT_5000			(1 << 1)
+#define USB_IRDA_MTT_10000			(1 << 0)
+
+/* Baud rates */
+
+#define USB_IRDA_BR_4000000			(1 << 8)
+#define USB_IRDA_BR_1152000			(1 << 7)
+#define USB_IRDA_BR_576000			(1 << 6)
+#define USB_IRDA_BR_115200			(1 << 5)
+#define USB_IRDA_BR_57600			(1 << 4)
+#define USB_IRDA_BR_38400			(1 << 3)
+#define USB_IRDA_BR_19200			(1 << 2)
+#define USB_IRDA_BR_9600			(1 << 1)
+#define USB_IRDA_BR_2400			(1 << 0)
+
+/* Additional BOFs */
+
+#define USB_IRDA_AB_0				(1 << 7)
+#define USB_IRDA_AB_1				(1 << 6)
+#define USB_IRDA_AB_2				(1 << 5)
+#define USB_IRDA_AB_3				(1 << 4)
+#define USB_IRDA_AB_6				(1 << 3)
+#define USB_IRDA_AB_12				(1 << 2)
+#define USB_IRDA_AB_24				(1 << 1)
+#define USB_IRDA_AB_48				(1 << 0)
+
+/* IRDA Rate Sniff */
+
+#define USB_IRDA_RATE_SNIFF			1
+
+/*-------------------------------------------------------------------------*/
+
+struct usb_irda_cs_descriptor {
+	__u8	bLength;
+	__u8	bDescriptorType;
+
+	__le16	bcdSpecRevision;
+	__u8	bmDataSize;
+	__u8	bmWindowSize;
+	__u8	bmMinTurnaroundTime;
+	__le16	wBaudRate;
+	__u8	bmAdditionalBOFs;
+	__u8	bIrdaRateSniff;
+	__u8	bMaxUnicastList;
+} __attribute__ ((packed));
+
+/*-------------------------------------------------------------------------*/
+
+/* Data Format */
+
+#define USB_IRDA_STATUS_MEDIA_BUSY	(1 << 7)
+
+/* The following is a 4-bit value used for both
+ * inbound and outbound headers:
+ *
+ * 0 - speed ignored
+ * 1 - 2400 bps
+ * 2 - 9600 bps
+ * 3 - 19200 bps
+ * 4 - 38400 bps
+ * 5 - 57600 bps
+ * 6 - 115200 bps
+ * 7 - 576000 bps
+ * 8 - 1.152 Mbps
+ * 9 - 5 mbps
+ * 10..15 - Reserved
+ */
+#define USB_IRDA_STATUS_LINK_SPEED	0x0f
+
+/* The following is a 4-bit value used only for
+ * outbound header:
+ *
+ * 0 - No change (BOF ignored)
+ * 1 - 48 BOFs
+ * 2 - 24 BOFs
+ * 3 - 12 BOFs
+ * 4 - 6 BOFs
+ * 5 - 3 BOFs
+ * 6 - 2 BOFs
+ * 7 - 1 BOFs
+ * 8 - 0 BOFs
+ * 9..15 - Reserved
+ */
+#define USB_IRDA_EXTRA_BOFS		0xf0
+
+struct usb_irda_inbound_header {
+	__u8		bmStatus;
+};
+
+struct usb_irda_outbound_header {
+	__u8		bmChange;
+};
+
+#endif /* __LINUX_USB_IRDA_H */
+
-- 
cgit v1.2.3


From f579c2b46f74038e8f5a762c7f10c2385b33e3dc Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 2 Jun 2008 16:26:48 -0400
Subject: USB Gadget: documentation update

This patch (as1102) clarifies two points in the USB Gadget kerneldoc:

	Request completion callbacks are always made with interrupts
	disabled;

	Device controllers may not support STALLing the status stage
	of a control transfer after the data stage is over.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/gadget.h | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/usb/gadget.h b/include/linux/usb/gadget.h
index 0ebedaec075d..0460a746480c 100644
--- a/include/linux/usb/gadget.h
+++ b/include/linux/usb/gadget.h
@@ -33,7 +33,8 @@ struct usb_ep;
  * @short_not_ok: When reading data, makes short packets be
  *     treated as errors (queue stops advancing till cleanup).
  * @complete: Function called when request completes, so this request and
- *	its buffer may be re-used.
+ *	its buffer may be re-used.  The function will always be called with
+ *	interrupts disabled, and it must not sleep.
  *	Reads terminate with a short packet, or when the buffer fills,
  *	whichever comes first.  When writes terminate, some data bytes
  *	will usually still be in flight (often in a hardware fifo).
@@ -271,7 +272,10 @@ static inline void usb_ep_free_request(struct usb_ep *ep,
  * (Note that some USB device controllers disallow protocol stall responses
  * in some cases.)  When control responses are deferred (the response is
  * written after the setup callback returns), then usb_ep_set_halt() may be
- * used on ep0 to trigger protocol stalls.
+ * used on ep0 to trigger protocol stalls.  Depending on the controller,
+ * it may not be possible to trigger a status-stage protocol stall when the
+ * data stage is over, that is, from within the response's completion
+ * routine.
  *
  * For periodic endpoints, like interrupt or isochronous ones, the usb host
  * arranges to poll once per interval, and the gadget driver usually will
-- 
cgit v1.2.3


From 625f694936cbbdee98e6cc65f72724a7660e7946 Mon Sep 17 00:00:00 2001
From: Ming Lei <tom.leiming@gmail.com>
Date: Sun, 15 Jun 2008 09:42:02 +0800
Subject: USB: remove interface parameter of usb_reset_composite_device

From the current implementation of usb_reset_composite_device
function, the iface parameter is no longer useful. This function
doesn't do something special for the iface usb_interface,compared
with other interfaces in the usb_device. So remove the parameter
and fix the related caller.

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/hid/usbhid/hid-core.c   |  2 +-
 drivers/usb/core/devio.c        |  2 +-
 drivers/usb/core/hub.c          | 11 +++--------
 drivers/usb/storage/transport.c |  3 +--
 include/linux/usb.h             |  3 +--
 5 files changed, 7 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/hid/usbhid/hid-core.c b/drivers/hid/usbhid/hid-core.c
index 01427c51c7cc..69fa79b6b51c 100644
--- a/drivers/hid/usbhid/hid-core.c
+++ b/drivers/hid/usbhid/hid-core.c
@@ -122,7 +122,7 @@ static void hid_reset(struct work_struct *work)
 		dev_dbg(&usbhid->intf->dev, "resetting device\n");
 		rc = rc_lock = usb_lock_device_for_reset(hid_to_usb_dev(hid), usbhid->intf);
 		if (rc_lock >= 0) {
-			rc = usb_reset_composite_device(hid_to_usb_dev(hid), usbhid->intf);
+			rc = usb_reset_composite_device(hid_to_usb_dev(hid));
 			if (rc_lock)
 				usb_unlock_device(hid_to_usb_dev(hid));
 		}
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 7bee9c18b3bc..bc1cce5cf758 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -872,7 +872,7 @@ static int proc_connectinfo(struct dev_state *ps, void __user *arg)
 
 static int proc_resetdevice(struct dev_state *ps)
 {
-	return usb_reset_composite_device(ps->dev, NULL);
+	return usb_reset_composite_device(ps->dev);
 }
 
 static int proc_setintf(struct dev_state *ps, void __user *arg)
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 3251120b414e..207c33d369be 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2712,7 +2712,7 @@ static void hub_port_connect_change(struct usb_hub *hub, int port1,
 #endif
 
 		} else {
-			status = usb_reset_composite_device(udev, NULL);
+			status = usb_reset_composite_device(udev);
 		}
 		usb_unlock_device(udev);
 
@@ -2940,7 +2940,7 @@ static void hub_events(void)
 			dev_dbg (hub_dev, "resetting for error %d\n",
 				hub->error);
 
-			ret = usb_reset_composite_device(hdev, intf);
+			ret = usb_reset_composite_device(hdev);
 			if (ret) {
 				dev_dbg (hub_dev,
 					"error resetting hub: %d\n", ret);
@@ -3355,7 +3355,6 @@ EXPORT_SYMBOL_GPL(usb_reset_device);
 /**
  * usb_reset_composite_device - warn interface drivers and perform a USB port reset
  * @udev: device to reset (not in SUSPENDED or NOTATTACHED state)
- * @iface: interface bound to the driver making the request (optional)
  *
  * Warns all drivers bound to registered interfaces (using their pre_reset
  * method), performs the port reset, and then lets the drivers know that
@@ -3368,8 +3367,7 @@ EXPORT_SYMBOL_GPL(usb_reset_device);
  * For calls that might not occur during probe(), drivers should lock
  * the device using usb_lock_device_for_reset().
  */
-int usb_reset_composite_device(struct usb_device *udev,
-		struct usb_interface *iface)
+int usb_reset_composite_device(struct usb_device *udev)
 {
 	int ret;
 	int i;
@@ -3385,9 +3383,6 @@ int usb_reset_composite_device(struct usb_device *udev,
 	/* Prevent autosuspend during the reset */
 	usb_autoresume_device(udev);
 
-	if (iface && iface->condition != USB_INTERFACE_BINDING)
-		iface = NULL;
-
 	if (config) {
 		for (i = 0; i < config->desc.bNumInterfaces; ++i) {
 			struct usb_interface *cintf = config->interface[i];
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 08d3a13fec2c..670e4cbd1f06 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -1172,8 +1172,7 @@ int usb_stor_port_reset(struct us_data *us)
 			result = -EIO;
 			US_DEBUGP("No reset during disconnect\n");
 		} else {
-			result = usb_reset_composite_device(
-					us->pusb_dev, us->pusb_intf);
+			result = usb_reset_composite_device(us->pusb_dev);
 			US_DEBUGP("usb_reset_composite_device returns %d\n",
 					result);
 		}
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 8429d08bd2fd..c74cc64bddc8 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -497,8 +497,7 @@ extern int usb_lock_device_for_reset(struct usb_device *udev,
 
 /* USB port reset for device reinitialization */
 extern int usb_reset_device(struct usb_device *dev);
-extern int usb_reset_composite_device(struct usb_device *dev,
-		struct usb_interface *iface);
+extern int usb_reset_composite_device(struct usb_device *dev);
 
 extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id);
 
-- 
cgit v1.2.3


From 742120c63138651c898614001cb58cd607401eac Mon Sep 17 00:00:00 2001
From: Ming Lei <tom.leiming@gmail.com>
Date: Wed, 18 Jun 2008 22:00:29 +0800
Subject: USB: fix usb_reset_device and usb_reset_composite_device(take 3)

This patch renames the existing usb_reset_device in hub.c to
usb_reset_and_verify_device and renames the existing
usb_reset_composite_device to usb_reset_device. Also the new
usb_reset_and_verify_device does't need to be EXPORTED .

The idea of the patch is that external interface driver
should warn the other interfaces' driver of the same
device before and after reseting the usb device. One interface
driver shoud call _old_ usb_reset_composite_device instead of
_old_ usb_reset_device since it can't assume the device contains
only one interface. The _old_ usb_reset_composite_device
is safe for single interface device also. we rename the two
functions to make the change easily.

This patch is under guideline from Alan Stern.

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
---
 drivers/hid/usbhid/hid-core.c   |  2 +-
 drivers/usb/core/devio.c        |  2 +-
 drivers/usb/core/hub.c          | 33 +++++++++++++++++----------------
 drivers/usb/storage/transport.c |  4 ++--
 include/linux/usb.h             |  5 ++---
 5 files changed, 23 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/hid/usbhid/hid-core.c b/drivers/hid/usbhid/hid-core.c
index 69fa79b6b51c..27fe4d8912cb 100644
--- a/drivers/hid/usbhid/hid-core.c
+++ b/drivers/hid/usbhid/hid-core.c
@@ -122,7 +122,7 @@ static void hid_reset(struct work_struct *work)
 		dev_dbg(&usbhid->intf->dev, "resetting device\n");
 		rc = rc_lock = usb_lock_device_for_reset(hid_to_usb_dev(hid), usbhid->intf);
 		if (rc_lock >= 0) {
-			rc = usb_reset_composite_device(hid_to_usb_dev(hid));
+			rc = usb_reset_device(hid_to_usb_dev(hid));
 			if (rc_lock)
 				usb_unlock_device(hid_to_usb_dev(hid));
 		}
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index bc1cce5cf758..e09935acae80 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -872,7 +872,7 @@ static int proc_connectinfo(struct dev_state *ps, void __user *arg)
 
 static int proc_resetdevice(struct dev_state *ps)
 {
-	return usb_reset_composite_device(ps->dev);
+	return usb_reset_device(ps->dev);
 }
 
 static int proc_setintf(struct dev_state *ps, void __user *arg)
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 207c33d369be..bb3ecc4c08f2 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -135,6 +135,8 @@ EXPORT_SYMBOL_GPL(ehci_cf_port_reset_rwsem);
 #define HUB_DEBOUNCE_STABLE	 100
 
 
+static int usb_reset_and_verify_device(struct usb_device *udev);
+
 static inline char *portspeed(int portstatus)
 {
 	if (portstatus & (1 << USB_PORT_FEAT_HIGHSPEED))
@@ -1971,7 +1973,7 @@ static int finish_port_resume(struct usb_device *udev)
 	 * resumed.
 	 */
 	if (udev->reset_resume)
-		status = usb_reset_device(udev);
+		status = usb_reset_and_verify_device(udev);
 
  	/* 10.5.4.5 says be sure devices in the tree are still there.
  	 * For now let's assume the device didn't go crazy on resume,
@@ -2030,7 +2032,7 @@ static int finish_port_resume(struct usb_device *udev)
  * to it will be lost.  Using the USB_PERSIST facility, the device can be
  * made to appear as if it had not disconnected.
  *
- * This facility can be dangerous.  Although usb_reset_device() makes
+ * This facility can be dangerous.  Although usb_reset_and_verify_device() makes
  * every effort to insure that the same device is present after the
  * reset as before, it cannot provide a 100% guarantee.  Furthermore it's
  * quite possible for a device to remain unaltered but its media to be
@@ -2140,7 +2142,7 @@ int usb_port_resume(struct usb_device *udev)
 		hub_port_logical_disconnect(hub, port1);
 	} else if (udev->reset_resume) {
 		dev_dbg(&udev->dev, "reset-resume\n");
-		status = usb_reset_device(udev);
+		status = usb_reset_and_verify_device(udev);
 	}
 	return status;
 }
@@ -2321,7 +2323,7 @@ static int hub_set_address(struct usb_device *udev, int devnum)
  * Returns device in USB_STATE_ADDRESS, except on error.
  *
  * If this is called for an already-existing device (as part of
- * usb_reset_device), the caller must own the device lock.  For a
+ * usb_reset_and_verify_device), the caller must own the device lock.  For a
  * newly detected device that is not accessible through any global
  * pointers, it's not necessary to lock the device.
  */
@@ -2638,7 +2640,7 @@ hub_power_remaining (struct usb_hub *hub)
  * This routine is called when:
  * 	a port connection-change occurs;
  *	a port enable-change occurs (often caused by EMI);
- *	usb_reset_device() encounters changed descriptors (as from
+ *	usb_reset_and_verify_device() encounters changed descriptors (as from
  *		a firmware download)
  * caller already locked the hub
  */
@@ -2712,7 +2714,7 @@ static void hub_port_connect_change(struct usb_hub *hub, int port1,
 #endif
 
 		} else {
-			status = usb_reset_composite_device(udev);
+			status = usb_reset_device(udev);
 		}
 		usb_unlock_device(udev);
 
@@ -2940,7 +2942,7 @@ static void hub_events(void)
 			dev_dbg (hub_dev, "resetting for error %d\n",
 				hub->error);
 
-			ret = usb_reset_composite_device(hdev);
+			ret = usb_reset_device(hdev);
 			if (ret) {
 				dev_dbg (hub_dev,
 					"error resetting hub: %d\n", ret);
@@ -3233,12 +3235,12 @@ static int descriptors_changed(struct usb_device *udev,
 }
 
 /**
- * usb_reset_device - perform a USB port reset to reinitialize a device
+ * usb_reset_and_verify_device - perform a USB port reset to reinitialize a device
  * @udev: device to reset (not in SUSPENDED or NOTATTACHED state)
  *
  * WARNING - don't use this routine to reset a composite device
  * (one with multiple interfaces owned by separate drivers)!
- * Use usb_reset_composite_device() instead.
+ * Use usb_reset_device() instead.
  *
  * Do a port reset, reassign the device's address, and establish its
  * former operating configuration.  If the reset fails, or the device's
@@ -3262,7 +3264,7 @@ static int descriptors_changed(struct usb_device *udev,
  * holding the device lock because these tasks should always call
  * usb_autopm_resume_device(), thereby preventing any unwanted autoresume.
  */
-int usb_reset_device(struct usb_device *udev)
+static int usb_reset_and_verify_device(struct usb_device *udev)
 {
 	struct usb_device		*parent_hdev = udev->parent;
 	struct usb_hub			*parent_hub;
@@ -3350,24 +3352,23 @@ re_enumerate:
 	hub_port_logical_disconnect(parent_hub, port1);
 	return -ENODEV;
 }
-EXPORT_SYMBOL_GPL(usb_reset_device);
 
 /**
- * usb_reset_composite_device - warn interface drivers and perform a USB port reset
+ * usb_reset_device - warn interface drivers and perform a USB port reset
  * @udev: device to reset (not in SUSPENDED or NOTATTACHED state)
  *
  * Warns all drivers bound to registered interfaces (using their pre_reset
  * method), performs the port reset, and then lets the drivers know that
  * the reset is over (using their post_reset method).
  *
- * Return value is the same as for usb_reset_device().
+ * Return value is the same as for usb_reset_and_verify_device().
  *
  * The caller must own the device lock.  For example, it's safe to use
  * this from a driver probe() routine after downloading new firmware.
  * For calls that might not occur during probe(), drivers should lock
  * the device using usb_lock_device_for_reset().
  */
-int usb_reset_composite_device(struct usb_device *udev)
+int usb_reset_device(struct usb_device *udev)
 {
 	int ret;
 	int i;
@@ -3397,7 +3398,7 @@ int usb_reset_composite_device(struct usb_device *udev)
 		}
 	}
 
-	ret = usb_reset_device(udev);
+	ret = usb_reset_and_verify_device(udev);
 
 	if (config) {
 		for (i = config->desc.bNumInterfaces - 1; i >= 0; --i) {
@@ -3416,4 +3417,4 @@ int usb_reset_composite_device(struct usb_device *udev)
 	usb_autosuspend_device(udev);
 	return ret;
 }
-EXPORT_SYMBOL_GPL(usb_reset_composite_device);
+EXPORT_SYMBOL_GPL(usb_reset_device);
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 670e4cbd1f06..fcbbfdb7b2b0 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -1172,8 +1172,8 @@ int usb_stor_port_reset(struct us_data *us)
 			result = -EIO;
 			US_DEBUGP("No reset during disconnect\n");
 		} else {
-			result = usb_reset_composite_device(us->pusb_dev);
-			US_DEBUGP("usb_reset_composite_device returns %d\n",
+			result = usb_reset_device(us->pusb_dev);
+			US_DEBUGP("usb_reset_device returns %d\n",
 					result);
 		}
 		if (rc_lock)
diff --git a/include/linux/usb.h b/include/linux/usb.h
index c74cc64bddc8..3cc8db5254d1 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -497,7 +497,6 @@ extern int usb_lock_device_for_reset(struct usb_device *udev,
 
 /* USB port reset for device reinitialization */
 extern int usb_reset_device(struct usb_device *dev);
-extern int usb_reset_composite_device(struct usb_device *dev);
 
 extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id);
 
@@ -957,9 +956,9 @@ struct usbdrv_wrap {
  * @resume: Called when the device is being resumed by the system.
  * @reset_resume: Called when the suspended device has been reset instead
  *	of being resumed.
- * @pre_reset: Called by usb_reset_composite_device() when the device
+ * @pre_reset: Called by usb_reset_device() when the device
  *	is about to be reset.
- * @post_reset: Called by usb_reset_composite_device() after the device
+ * @post_reset: Called by usb_reset_device() after the device
  *	has been reset
  * @id_table: USB drivers use ID table to support hotplugging.
  *	Export this with MODULE_DEVICE_TABLE(usb,...).  This must be set
-- 
cgit v1.2.3


From 78d9a487ee961c356e1a934d9a92eca38ffb3a70 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 23 Jun 2008 16:00:40 -0400
Subject: USB: Force unbinding of drivers lacking reset_resume or other methods

This patch (as1024) takes care of a FIXME issue: Drivers that don't
have the necessary suspend, resume, reset_resume, pre_reset, or
post_reset methods will be unbound and their interface reprobed when
one of the unsupported events occurs.

This is made slightly more difficult by the fact that bind operations
won't work during a system sleep transition.  So instead the code has
to defer the operation until the transition ends.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 131 ++++++++++++++++++++++++++++++++++++++++------
 drivers/usb/core/hub.c    |  27 +++++++---
 drivers/usb/core/usb.h    |   2 +
 include/linux/usb.h       |   1 +
 4 files changed, 140 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 8da1a56659be..ddb54e14a5c5 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -201,6 +201,7 @@ static int usb_probe_interface(struct device *dev)
 
 	intf = to_usb_interface(dev);
 	udev = interface_to_usbdev(intf);
+	intf->needs_binding = 0;
 
 	if (udev->authorized == 0) {
 		dev_err(&intf->dev, "Device is not authorized for usage\n");
@@ -311,6 +312,7 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 
 	dev->driver = &driver->drvwrap.driver;
 	usb_set_intfdata(iface, priv);
+	iface->needs_binding = 0;
 
 	usb_pm_lock(udev);
 	iface->condition = USB_INTERFACE_BOUND;
@@ -772,6 +774,104 @@ void usb_deregister(struct usb_driver *driver)
 }
 EXPORT_SYMBOL_GPL(usb_deregister);
 
+
+/* Forced unbinding of a USB interface driver, either because
+ * it doesn't support pre_reset/post_reset/reset_resume or
+ * because it doesn't support suspend/resume.
+ *
+ * The caller must hold @intf's device's lock, but not its pm_mutex
+ * and not @intf->dev.sem.
+ */
+void usb_forced_unbind_intf(struct usb_interface *intf)
+{
+	struct usb_driver *driver = to_usb_driver(intf->dev.driver);
+
+	dev_dbg(&intf->dev, "forced unbind\n");
+	usb_driver_release_interface(driver, intf);
+
+	/* Mark the interface for later rebinding */
+	intf->needs_binding = 1;
+}
+
+/* Delayed forced unbinding of a USB interface driver and scan
+ * for rebinding.
+ *
+ * The caller must hold @intf's device's lock, but not its pm_mutex
+ * and not @intf->dev.sem.
+ *
+ * FIXME: The caller must block system sleep transitions.
+ */
+void usb_rebind_intf(struct usb_interface *intf)
+{
+	int rc;
+
+	/* Delayed unbind of an existing driver */
+	if (intf->dev.driver) {
+		struct usb_driver *driver =
+				to_usb_driver(intf->dev.driver);
+
+		dev_dbg(&intf->dev, "forced unbind\n");
+		usb_driver_release_interface(driver, intf);
+	}
+
+	/* Try to rebind the interface */
+	intf->needs_binding = 0;
+	rc = device_attach(&intf->dev);
+	if (rc < 0)
+		dev_warn(&intf->dev, "rebind failed: %d\n", rc);
+}
+
+#define DO_UNBIND	0
+#define DO_REBIND	1
+
+/* Unbind drivers for @udev's interfaces that don't support suspend/resume,
+ * or rebind interfaces that have been unbound, according to @action.
+ *
+ * The caller must hold @udev's device lock.
+ * FIXME: For rebinds, the caller must block system sleep transitions.
+ */
+static void do_unbind_rebind(struct usb_device *udev, int action)
+{
+	struct usb_host_config	*config;
+	int			i;
+	struct usb_interface	*intf;
+	struct usb_driver	*drv;
+
+	config = udev->actconfig;
+	if (config) {
+		for (i = 0; i < config->desc.bNumInterfaces; ++i) {
+			intf = config->interface[i];
+			switch (action) {
+			case DO_UNBIND:
+				if (intf->dev.driver) {
+					drv = to_usb_driver(intf->dev.driver);
+					if (!drv->suspend || !drv->resume)
+						usb_forced_unbind_intf(intf);
+				}
+				break;
+			case DO_REBIND:
+				if (intf->needs_binding) {
+
+	/* FIXME: The next line is needed because we are going to probe
+	 * the interface, but as far as the PM core is concerned the
+	 * interface is still suspended.  The problem wouldn't exist
+	 * if we could rebind the interface during the interface's own
+	 * resume() call, but at the time the usb_device isn't locked!
+	 *
+	 * The real solution will be to carry this out during the device's
+	 * complete() callback.  Until that is implemented, we have to
+	 * use this hack.
+	 */
+//					intf->dev.power.sleeping = 0;
+
+					usb_rebind_intf(intf);
+				}
+				break;
+			}
+		}
+	}
+}
+
 #ifdef CONFIG_PM
 
 /* Caller has locked udev's pm_mutex */
@@ -841,7 +941,7 @@ static int usb_suspend_interface(struct usb_interface *intf, pm_message_t msg)
 		goto done;
 	driver = to_usb_driver(intf->dev.driver);
 
-	if (driver->suspend && driver->resume) {
+	if (driver->suspend) {
 		status = driver->suspend(intf, msg);
 		if (status == 0)
 			mark_quiesced(intf);
@@ -849,12 +949,10 @@ static int usb_suspend_interface(struct usb_interface *intf, pm_message_t msg)
 			dev_err(&intf->dev, "%s error %d\n",
 					"suspend", status);
 	} else {
-		/*
-		 * FIXME else if there's no suspend method, disconnect...
-		 * Not possible if auto_pm is set...
-		 */
-		dev_warn(&intf->dev, "no suspend for driver %s?\n",
-				driver->name);
+		/* Later we will unbind the driver and reprobe */
+		intf->needs_binding = 1;
+		dev_warn(&intf->dev, "no %s for driver %s?\n",
+				"suspend", driver->name);
 		mark_quiesced(intf);
 	}
 
@@ -878,10 +976,12 @@ static int usb_resume_interface(struct usb_interface *intf, int reset_resume)
 		goto done;
 
 	/* Can't resume it if it doesn't have a driver. */
-	if (intf->condition == USB_INTERFACE_UNBOUND) {
-		status = -ENOTCONN;
+	if (intf->condition == USB_INTERFACE_UNBOUND)
+		goto done;
+
+	/* Don't resume if the interface is marked for rebinding */
+	if (intf->needs_binding)
 		goto done;
-	}
 	driver = to_usb_driver(intf->dev.driver);
 
 	if (reset_resume) {
@@ -891,7 +991,7 @@ static int usb_resume_interface(struct usb_interface *intf, int reset_resume)
 				dev_err(&intf->dev, "%s error %d\n",
 						"reset_resume", status);
 		} else {
-			/* status = -EOPNOTSUPP; */
+			intf->needs_binding = 1;
 			dev_warn(&intf->dev, "no %s for driver %s?\n",
 					"reset_resume", driver->name);
 		}
@@ -902,7 +1002,7 @@ static int usb_resume_interface(struct usb_interface *intf, int reset_resume)
 				dev_err(&intf->dev, "%s error %d\n",
 						"resume", status);
 		} else {
-			/* status = -EOPNOTSUPP; */
+			intf->needs_binding = 1;
 			dev_warn(&intf->dev, "no %s for driver %s?\n",
 					"resume", driver->name);
 		}
@@ -910,11 +1010,10 @@ static int usb_resume_interface(struct usb_interface *intf, int reset_resume)
 
 done:
 	dev_vdbg(&intf->dev, "%s: status %d\n", __func__, status);
-	if (status == 0)
+	if (status == 0 && intf->condition == USB_INTERFACE_BOUND)
 		mark_active(intf);
 
-	/* FIXME: Unbind the driver and reprobe if the resume failed
-	 * (not possible if auto_pm is set) */
+	/* Later we will unbind the driver and/or reprobe, if necessary */
 	return status;
 }
 
@@ -1470,6 +1569,7 @@ int usb_external_suspend_device(struct usb_device *udev, pm_message_t msg)
 {
 	int	status;
 
+	do_unbind_rebind(udev, DO_UNBIND);
 	usb_pm_lock(udev);
 	udev->auto_pm = 0;
 	status = usb_suspend_both(udev, msg);
@@ -1497,6 +1597,7 @@ int usb_external_resume_device(struct usb_device *udev)
 	status = usb_resume_both(udev);
 	udev->last_busy = jiffies;
 	usb_pm_unlock(udev);
+	do_unbind_rebind(udev, DO_REBIND);
 
 	/* Now that the device is awake, we can start trying to autosuspend
 	 * it again. */
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index bb3ecc4c08f2..f1efabbc1ca2 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -3367,6 +3367,11 @@ re_enumerate:
  * this from a driver probe() routine after downloading new firmware.
  * For calls that might not occur during probe(), drivers should lock
  * the device using usb_lock_device_for_reset().
+ *
+ * If an interface is currently being probed or disconnected, we assume
+ * its driver knows how to handle resets.  For all other interfaces,
+ * if the driver doesn't have pre_reset and post_reset methods then
+ * we attempt to unbind it and rebind afterward.
  */
 int usb_reset_device(struct usb_device *udev)
 {
@@ -3388,12 +3393,17 @@ int usb_reset_device(struct usb_device *udev)
 		for (i = 0; i < config->desc.bNumInterfaces; ++i) {
 			struct usb_interface *cintf = config->interface[i];
 			struct usb_driver *drv;
+			int unbind = 0;
 
 			if (cintf->dev.driver) {
 				drv = to_usb_driver(cintf->dev.driver);
-				if (drv->pre_reset)
-					(drv->pre_reset)(cintf);
-	/* FIXME: Unbind if pre_reset returns an error or isn't defined */
+				if (drv->pre_reset && drv->post_reset)
+					unbind = (drv->pre_reset)(cintf);
+				else if (cintf->condition ==
+						USB_INTERFACE_BOUND)
+					unbind = 1;
+				if (unbind)
+					usb_forced_unbind_intf(cintf);
 			}
 		}
 	}
@@ -3404,13 +3414,18 @@ int usb_reset_device(struct usb_device *udev)
 		for (i = config->desc.bNumInterfaces - 1; i >= 0; --i) {
 			struct usb_interface *cintf = config->interface[i];
 			struct usb_driver *drv;
+			int rebind = cintf->needs_binding;
 
-			if (cintf->dev.driver) {
+			if (!rebind && cintf->dev.driver) {
 				drv = to_usb_driver(cintf->dev.driver);
 				if (drv->post_reset)
-					(drv->post_reset)(cintf);
-	/* FIXME: Unbind if post_reset returns an error or isn't defined */
+					rebind = (drv->post_reset)(cintf);
+				else if (cintf->condition ==
+						USB_INTERFACE_BOUND)
+					rebind = 1;
 			}
+			if (rebind)
+				usb_rebind_intf(cintf);
 		}
 	}
 
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 1a8bc21c335e..d3eb0a29bca1 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -29,6 +29,8 @@ extern int usb_choose_configuration(struct usb_device *udev);
 extern void usb_kick_khubd(struct usb_device *dev);
 extern int usb_match_device(struct usb_device *dev,
 			    const struct usb_device_id *id);
+extern void usb_forced_unbind_intf(struct usb_interface *intf);
+extern void usb_rebind_intf(struct usb_interface *intf);
 
 extern int  usb_hub_init(void);
 extern void usb_hub_cleanup(void);
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 3cc8db5254d1..5811c5da69f9 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -160,6 +160,7 @@ struct usb_interface {
 	unsigned is_active:1;		/* the interface is not suspended */
 	unsigned sysfs_files_created:1;	/* the sysfs attributes exist */
 	unsigned needs_remote_wakeup:1;	/* driver requires remote wakeup */
+	unsigned needs_binding:1;	/* needs delayed unbind/rebind */
 
 	struct device dev;		/* interface specific device info */
 	struct device *usb_dev;
-- 
cgit v1.2.3


From 18ad7a61e1b700dfe177fabf1c350b4f5d4da8ac Mon Sep 17 00:00:00 2001
From: Wolfgang Grandegger <wg@grandegger.com>
Date: Fri, 30 May 2008 22:51:33 +1000
Subject: of_gpio: Should use new <linux/gpio.h> header

Since commit 7560fa60fcdcdb0da662f6a9fad9064b554ef46c (gpio: <linux/gpio.h>
and "no GPIO support here" stubs) drivers can use GPIOs if they're available,
but don't require them.

This patch actually enables this feature.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/linux/of_gpio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/of_gpio.h b/include/linux/of_gpio.h
index 2ee97e9877a7..67db101d0eb8 100644
--- a/include/linux/of_gpio.h
+++ b/include/linux/of_gpio.h
@@ -15,7 +15,7 @@
 #define __LINUX_OF_GPIO_H
 
 #include <linux/errno.h>
-#include <asm/gpio.h>
+#include <linux/gpio.h>
 
 #ifdef CONFIG_OF_GPIO
 
-- 
cgit v1.2.3


From 1ed6af73440c5ec920884bb800685a8cab4ce847 Mon Sep 17 00:00:00 2001
From: Mark Nelson <markn@au1.ibm.com>
Date: Fri, 18 Jul 2008 23:03:34 +1000
Subject: powerpc/cell: Add DMA_ATTR_WEAK_ORDERING dma attribute and use in
 Cell IOMMU code

Introduce a new dma attriblue DMA_ATTR_WEAK_ORDERING to use weak ordering
on DMA mappings in the Cell processor. Add the code to the Cell's IOMMU
implementation to use this code.

Dynamic mappings can be weakly or strongly ordered on an individual basis
but the fixed mapping has to be either completely strong or completely weak.
This is currently decided by a kernel boot option (pass iommu_fixed=weak
for a weakly ordered fixed linear mapping, strongly ordered is the default).

Signed-off-by: Mark Nelson <markn@au1.ibm.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 Documentation/DMA-attributes.txt    |   9 +++
 arch/powerpc/platforms/cell/iommu.c | 113 ++++++++++++++++++++++++++++++++++--
 include/linux/dma-attrs.h           |   1 +
 3 files changed, 118 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/DMA-attributes.txt b/Documentation/DMA-attributes.txt
index 6d772f84b477..b768cc0e402b 100644
--- a/Documentation/DMA-attributes.txt
+++ b/Documentation/DMA-attributes.txt
@@ -22,3 +22,12 @@ ready and available in memory.  The DMA of the "completion indication"
 could race with data DMA.  Mapping the memory used for completion
 indications with DMA_ATTR_WRITE_BARRIER would prevent the race.
 
+DMA_ATTR_WEAK_ORDERING
+----------------------
+
+DMA_ATTR_WEAK_ORDERING specifies that reads and writes to the mapping
+may be weakly ordered, that is that reads and writes may pass each other.
+
+Since it is optional for platforms to implement DMA_ATTR_WEAK_ORDERING,
+those that do not will simply ignore the attribute and exhibit default
+behavior.
diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c
index 3b7078453e7f..208005ca262c 100644
--- a/arch/powerpc/platforms/cell/iommu.c
+++ b/arch/powerpc/platforms/cell/iommu.c
@@ -199,6 +199,8 @@ static void tce_build_cell(struct iommu_table *tbl, long index, long npages,
 	base_pte = IOPTE_PP_W | IOPTE_PP_R | IOPTE_M | IOPTE_SO_RW |
 		(window->ioid & IOPTE_IOID_Mask);
 #endif
+	if (unlikely(dma_get_attr(DMA_ATTR_WEAK_ORDERING, attrs)))
+		base_pte &= ~IOPTE_SO_RW;
 
 	io_pte = (unsigned long *)tbl->it_base + (index - tbl->it_offset);
 
@@ -539,7 +541,9 @@ static struct cbe_iommu *cell_iommu_for_node(int nid)
 static unsigned long cell_dma_direct_offset;
 
 static unsigned long dma_iommu_fixed_base;
-struct dma_mapping_ops dma_iommu_fixed_ops;
+
+/* iommu_fixed_is_weak is set if booted with iommu_fixed=weak */
+static int iommu_fixed_is_weak;
 
 static struct iommu_table *cell_get_iommu_table(struct device *dev)
 {
@@ -563,6 +567,98 @@ static struct iommu_table *cell_get_iommu_table(struct device *dev)
 	return &window->table;
 }
 
+/* A coherent allocation implies strong ordering */
+
+static void *dma_fixed_alloc_coherent(struct device *dev, size_t size,
+				      dma_addr_t *dma_handle, gfp_t flag)
+{
+	if (iommu_fixed_is_weak)
+		return iommu_alloc_coherent(dev, cell_get_iommu_table(dev),
+					    size, dma_handle,
+					    device_to_mask(dev), flag,
+					    dev->archdata.numa_node);
+	else
+		return dma_direct_ops.alloc_coherent(dev, size, dma_handle,
+						     flag);
+}
+
+static void dma_fixed_free_coherent(struct device *dev, size_t size,
+				    void *vaddr, dma_addr_t dma_handle)
+{
+	if (iommu_fixed_is_weak)
+		iommu_free_coherent(cell_get_iommu_table(dev), size, vaddr,
+				    dma_handle);
+	else
+		dma_direct_ops.free_coherent(dev, size, vaddr, dma_handle);
+}
+
+static dma_addr_t dma_fixed_map_single(struct device *dev, void *ptr,
+				       size_t size,
+				       enum dma_data_direction direction,
+				       struct dma_attrs *attrs)
+{
+	if (iommu_fixed_is_weak == dma_get_attr(DMA_ATTR_WEAK_ORDERING, attrs))
+		return dma_direct_ops.map_single(dev, ptr, size, direction,
+						 attrs);
+	else
+		return iommu_map_single(dev, cell_get_iommu_table(dev), ptr,
+					size, device_to_mask(dev), direction,
+					attrs);
+}
+
+static void dma_fixed_unmap_single(struct device *dev, dma_addr_t dma_addr,
+				   size_t size,
+				   enum dma_data_direction direction,
+				   struct dma_attrs *attrs)
+{
+	if (iommu_fixed_is_weak == dma_get_attr(DMA_ATTR_WEAK_ORDERING, attrs))
+		dma_direct_ops.unmap_single(dev, dma_addr, size, direction,
+					    attrs);
+	else
+		iommu_unmap_single(cell_get_iommu_table(dev), dma_addr, size,
+				   direction, attrs);
+}
+
+static int dma_fixed_map_sg(struct device *dev, struct scatterlist *sg,
+			   int nents, enum dma_data_direction direction,
+			   struct dma_attrs *attrs)
+{
+	if (iommu_fixed_is_weak == dma_get_attr(DMA_ATTR_WEAK_ORDERING, attrs))
+		return dma_direct_ops.map_sg(dev, sg, nents, direction, attrs);
+	else
+		return iommu_map_sg(dev, cell_get_iommu_table(dev), sg, nents,
+				    device_to_mask(dev), direction, attrs);
+}
+
+static void dma_fixed_unmap_sg(struct device *dev, struct scatterlist *sg,
+			       int nents, enum dma_data_direction direction,
+			       struct dma_attrs *attrs)
+{
+	if (iommu_fixed_is_weak == dma_get_attr(DMA_ATTR_WEAK_ORDERING, attrs))
+		dma_direct_ops.unmap_sg(dev, sg, nents, direction, attrs);
+	else
+		iommu_unmap_sg(cell_get_iommu_table(dev), sg, nents, direction,
+			       attrs);
+}
+
+static int dma_fixed_dma_supported(struct device *dev, u64 mask)
+{
+	return mask == DMA_64BIT_MASK;
+}
+
+static int dma_set_mask_and_switch(struct device *dev, u64 dma_mask);
+
+struct dma_mapping_ops dma_iommu_fixed_ops = {
+	.alloc_coherent = dma_fixed_alloc_coherent,
+	.free_coherent  = dma_fixed_free_coherent,
+	.map_single     = dma_fixed_map_single,
+	.unmap_single   = dma_fixed_unmap_single,
+	.map_sg         = dma_fixed_map_sg,
+	.unmap_sg       = dma_fixed_unmap_sg,
+	.dma_supported  = dma_fixed_dma_supported,
+	.set_dma_mask   = dma_set_mask_and_switch,
+};
+
 static void cell_dma_dev_setup_fixed(struct device *dev);
 
 static void cell_dma_dev_setup(struct device *dev)
@@ -919,9 +1015,16 @@ static void cell_iommu_setup_fixed_ptab(struct cbe_iommu *iommu,
 
 	pr_debug("iommu: mapping 0x%lx pages from 0x%lx\n", fsize, fbase);
 
-	base_pte = IOPTE_PP_W | IOPTE_PP_R | IOPTE_M | IOPTE_SO_RW
+	base_pte = IOPTE_PP_W | IOPTE_PP_R | IOPTE_M
 		    | (cell_iommu_get_ioid(np) & IOPTE_IOID_Mask);
 
+	if (iommu_fixed_is_weak)
+		pr_info("IOMMU: Using weak ordering for fixed mapping\n");
+	else {
+		pr_info("IOMMU: Using strong ordering for fixed mapping\n");
+		base_pte |= IOPTE_SO_RW;
+	}
+
 	for (uaddr = 0; uaddr < fsize; uaddr += (1 << 24)) {
 		/* Don't touch the dynamic region */
 		ioaddr = uaddr + fbase;
@@ -1037,9 +1140,6 @@ static int __init cell_iommu_fixed_mapping_init(void)
 		cell_iommu_setup_window(iommu, np, dbase, dsize, 0);
 	}
 
-	dma_iommu_fixed_ops = dma_direct_ops;
-	dma_iommu_fixed_ops.set_dma_mask = dma_set_mask_and_switch;
-
 	dma_iommu_ops.set_dma_mask = dma_set_mask_and_switch;
 	set_pci_dma_ops(&dma_iommu_ops);
 
@@ -1053,6 +1153,9 @@ static int __init setup_iommu_fixed(char *str)
 	if (strcmp(str, "off") == 0)
 		iommu_fixed_disabled = 1;
 
+	else if (strcmp(str, "weak") == 0)
+		iommu_fixed_is_weak = 1;
+
 	return 1;
 }
 __setup("iommu_fixed=", setup_iommu_fixed);
diff --git a/include/linux/dma-attrs.h b/include/linux/dma-attrs.h
index 1677e2bfa00c..71ad34eca6e3 100644
--- a/include/linux/dma-attrs.h
+++ b/include/linux/dma-attrs.h
@@ -12,6 +12,7 @@
  */
 enum dma_attr {
 	DMA_ATTR_WRITE_BARRIER,
+	DMA_ATTR_WEAK_ORDERING,
 	DMA_ATTR_MAX,
 };
 
-- 
cgit v1.2.3


From e105b8bfc769b0545b6f0f395179d1e43cbee822 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Mon, 21 Apr 2008 10:51:07 -0700
Subject: sysfs: add /sys/dev/{char,block} to lookup sysfs path by major:minor

Why?:
There are occasions where userspace would like to access sysfs
attributes for a device but it may not know how sysfs has named the
device or the path.  For example what is the sysfs path for
/dev/disk/by-id/ata-ST3160827AS_5MT004CK?  With this change a call to
stat(2) returns the major:minor then userspace can see that
/sys/dev/block/8:32 links to /sys/block/sdc.

What are the alternatives?:
1/ Add an ioctl to return the path: Doable, but sysfs is meant to reduce
   the need to proliferate ioctl interfaces into the kernel, so this
   seems counter productive.

2/ Use udev to create these symlinks: Also doable, but it adds a
   udev dependency to utilities that might be running in a limited
   environment like an initramfs.

3/ Do a full-tree search of sysfs.

[kay.sievers@vrfy.org: fix duplicate registrations]
[kay.sievers@vrfy.org: cleanup suggestions]

Cc: Neil Brown <neilb@suse.de>
Cc: Tejun Heo <htejun@gmail.com>
Acked-by: Kay Sievers <kay.sievers@vrfy.org>
Reviewed-by: SL Baur <steve@xemacs.org>
Acked-by: Kay Sievers <kay.sievers@vrfy.org>
Acked-by: Mark Lord <lkml@rtr.ca>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/ABI/testing/sysfs-dev | 20 +++++++++
 Documentation/filesystems/sysfs.txt |  6 +++
 block/genhd.c                       |  5 ++-
 drivers/base/class.c                |  4 ++
 drivers/base/core.c                 | 83 ++++++++++++++++++++++++++++++++++++-
 drivers/usb/core/devio.c            |  5 +++
 include/linux/device.h              |  3 ++
 7 files changed, 124 insertions(+), 2 deletions(-)
 create mode 100644 Documentation/ABI/testing/sysfs-dev

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-dev b/Documentation/ABI/testing/sysfs-dev
new file mode 100644
index 000000000000..a9f2b8b0530f
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-dev
@@ -0,0 +1,20 @@
+What:		/sys/dev
+Date:		April 2008
+KernelVersion:	2.6.26
+Contact:	Dan Williams <dan.j.williams@intel.com>
+Description:	The /sys/dev tree provides a method to look up the sysfs
+		path for a device using the information returned from
+		stat(2).  There are two directories, 'block' and 'char',
+		beneath /sys/dev containing symbolic links with names of
+		the form "<major>:<minor>".  These links point to the
+		corresponding sysfs path for the given device.
+
+		Example:
+		$ readlink /sys/dev/block/8:32
+		../../block/sdc
+
+		Entries in /sys/dev/char and /sys/dev/block will be
+		dynamically created and destroyed as devices enter and
+		leave the system.
+
+Users:		mdadm <linux-raid@vger.kernel.org>
diff --git a/Documentation/filesystems/sysfs.txt b/Documentation/filesystems/sysfs.txt
index 7f27b8f840d0..9e9c348275a9 100644
--- a/Documentation/filesystems/sysfs.txt
+++ b/Documentation/filesystems/sysfs.txt
@@ -248,6 +248,7 @@ The top level sysfs directory looks like:
 block/
 bus/
 class/
+dev/
 devices/
 firmware/
 net/
@@ -274,6 +275,11 @@ fs/ contains a directory for some filesystems.  Currently each
 filesystem wanting to export attributes must create its own hierarchy
 below fs/ (see ./fuse.txt for an example).
 
+dev/ contains two directories char/ and block/. Inside these two
+directories there are symlinks named <major>:<minor>.  These symlinks
+point to the sysfs directory for the given device.  /sys/dev provides a
+quick way to lookup the sysfs interface for a device from the result of
+a stat(2) operation.
 
 More information can driver-model specific features can be found in
 Documentation/driver-model/. 
diff --git a/block/genhd.c b/block/genhd.c
index 9074f384b097..24e3fc9095fe 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -370,7 +370,10 @@ static struct kobject *base_probe(dev_t devt, int *part, void *data)
 
 static int __init genhd_device_init(void)
 {
-	int error = class_register(&block_class);
+	int error;
+
+	block_class.dev_kobj = sysfs_dev_block_kobj;
+	error = class_register(&block_class);
 	if (unlikely(error))
 		return error;
 	bdev_map = kobj_map_init(base_probe, &block_class_lock);
diff --git a/drivers/base/class.c b/drivers/base/class.c
index e085af0ff94f..71ce3ff6bdf5 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -148,6 +148,10 @@ int class_register(struct class *cls)
 	if (error)
 		return error;
 
+	/* set the default /sys/dev directory for devices of this class */
+	if (!cls->dev_kobj)
+		cls->dev_kobj = sysfs_dev_char_kobj;
+
 #if defined(CONFIG_SYSFS_DEPRECATED) && defined(CONFIG_BLOCK)
 	/* let the block class directory show up in the root of sysfs */
 	if (cls != &block_class)
diff --git a/drivers/base/core.c b/drivers/base/core.c
index ee0a51a3a41d..be9aba4dc2fb 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -27,6 +27,9 @@
 
 int (*platform_notify)(struct device *dev) = NULL;
 int (*platform_notify_remove)(struct device *dev) = NULL;
+static struct kobject *dev_kobj;
+struct kobject *sysfs_dev_char_kobj;
+struct kobject *sysfs_dev_block_kobj;
 
 #ifdef CONFIG_BLOCK
 static inline int device_is_not_partition(struct device *dev)
@@ -775,6 +778,54 @@ int dev_set_name(struct device *dev, const char *fmt, ...)
 }
 EXPORT_SYMBOL_GPL(dev_set_name);
 
+/**
+ * device_to_dev_kobj - select a /sys/dev/ directory for the device
+ * @dev: device
+ *
+ * By default we select char/ for new entries.  Setting class->dev_obj
+ * to NULL prevents an entry from being created.  class->dev_kobj must
+ * be set (or cleared) before any devices are registered to the class
+ * otherwise device_create_sys_dev_entry() and
+ * device_remove_sys_dev_entry() will disagree about the the presence
+ * of the link.
+ */
+static struct kobject *device_to_dev_kobj(struct device *dev)
+{
+	struct kobject *kobj;
+
+	if (dev->class)
+		kobj = dev->class->dev_kobj;
+	else
+		kobj = sysfs_dev_char_kobj;
+
+	return kobj;
+}
+
+static int device_create_sys_dev_entry(struct device *dev)
+{
+	struct kobject *kobj = device_to_dev_kobj(dev);
+	int error = 0;
+	char devt_str[15];
+
+	if (kobj) {
+		format_dev_t(devt_str, dev->devt);
+		error = sysfs_create_link(kobj, &dev->kobj, devt_str);
+	}
+
+	return error;
+}
+
+static void device_remove_sys_dev_entry(struct device *dev)
+{
+	struct kobject *kobj = device_to_dev_kobj(dev);
+	char devt_str[15];
+
+	if (kobj) {
+		format_dev_t(devt_str, dev->devt);
+		sysfs_remove_link(kobj, devt_str);
+	}
+}
+
 /**
  * device_add - add device to device hierarchy.
  * @dev: device.
@@ -829,6 +880,10 @@ int device_add(struct device *dev)
 		error = device_create_file(dev, &devt_attr);
 		if (error)
 			goto ueventattrError;
+
+		error = device_create_sys_dev_entry(dev);
+		if (error)
+			goto devtattrError;
 	}
 
 	error = device_add_class_symlinks(dev);
@@ -872,6 +927,9 @@ int device_add(struct device *dev)
  AttrsError:
 	device_remove_class_symlinks(dev);
  SymlinkError:
+	if (MAJOR(dev->devt))
+		device_remove_sys_dev_entry(dev);
+ devtattrError:
 	if (MAJOR(dev->devt))
 		device_remove_file(dev, &devt_attr);
  ueventattrError:
@@ -948,8 +1006,10 @@ void device_del(struct device *dev)
 	device_pm_remove(dev);
 	if (parent)
 		klist_del(&dev->knode_parent);
-	if (MAJOR(dev->devt))
+	if (MAJOR(dev->devt)) {
+		device_remove_sys_dev_entry(dev);
 		device_remove_file(dev, &devt_attr);
+	}
 	if (dev->class) {
 		device_remove_class_symlinks(dev);
 
@@ -1074,7 +1134,25 @@ int __init devices_init(void)
 	devices_kset = kset_create_and_add("devices", &device_uevent_ops, NULL);
 	if (!devices_kset)
 		return -ENOMEM;
+	dev_kobj = kobject_create_and_add("dev", NULL);
+	if (!dev_kobj)
+		goto dev_kobj_err;
+	sysfs_dev_block_kobj = kobject_create_and_add("block", dev_kobj);
+	if (!sysfs_dev_block_kobj)
+		goto block_kobj_err;
+	sysfs_dev_char_kobj = kobject_create_and_add("char", dev_kobj);
+	if (!sysfs_dev_char_kobj)
+		goto char_kobj_err;
+
 	return 0;
+
+ char_kobj_err:
+	kobject_put(sysfs_dev_block_kobj);
+ block_kobj_err:
+	kobject_put(dev_kobj);
+ dev_kobj_err:
+	kset_unregister(devices_kset);
+	return -ENOMEM;
 }
 
 EXPORT_SYMBOL_GPL(device_for_each_child);
@@ -1447,4 +1525,7 @@ void device_shutdown(void)
 			dev->driver->shutdown(dev);
 		}
 	}
+	kobject_put(sysfs_dev_char_kobj);
+	kobject_put(sysfs_dev_block_kobj);
+	kobject_put(dev_kobj);
 }
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 54a350ccd033..6fbc8f5ab80c 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -1792,6 +1792,11 @@ int __init usb_devio_init(void)
 		usb_classdev_class = NULL;
 		goto out;
 	}
+	/* devices of this class shadow the major:minor of their parent
+	 * device, so clear ->dev_kobj to prevent adding duplicate entries
+	 * to /sys/dev
+	 */
+	usb_classdev_class->dev_kobj = NULL;
 
 	usb_register_notify(&usbdev_nb);
 #endif
diff --git a/include/linux/device.h b/include/linux/device.h
index f71a78d123ae..e49aa74f248c 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -193,6 +193,7 @@ struct class {
 	struct semaphore	sem; /* locks children, devices, interfaces */
 	struct class_attribute		*class_attrs;
 	struct device_attribute		*dev_attrs;
+	struct kobject			*dev_kobj;
 
 	int (*dev_uevent)(struct device *dev, struct kobj_uevent_env *env);
 
@@ -205,6 +206,8 @@ struct class {
 	struct pm_ops *pm;
 };
 
+extern struct kobject *sysfs_dev_block_kobj;
+extern struct kobject *sysfs_dev_char_kobj;
 extern int __must_check class_register(struct class *class);
 extern void class_unregister(struct class *class);
 extern int class_for_each_device(struct class *class, void *data,
-- 
cgit v1.2.3


From ccea44fadca396b0f89aed5bdeb60e4abb212566 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 21 May 2008 12:52:33 -0700
Subject: driver core: remove device_create()

There are no more users of this, and it is racy.  Use
device_create_drvdata() or device_create_vargs() instead.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 34 ----------------------------------
 include/linux/device.h |  3 ---
 2 files changed, 37 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index be9aba4dc2fb..4dc0d272f6d9 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -1273,40 +1273,6 @@ struct device *device_create_drvdata(struct class *class,
 }
 EXPORT_SYMBOL_GPL(device_create_drvdata);
 
-/**
- * device_create - creates a device and registers it with sysfs
- * @class: pointer to the struct class that this device should be registered to
- * @parent: pointer to the parent struct device of this new device, if any
- * @devt: the dev_t for the char device to be added
- * @fmt: string for the device's name
- *
- * This function can be used by char device classes.  A struct device
- * will be created in sysfs, registered to the specified class.
- *
- * A "dev" file will be created, showing the dev_t for the device, if
- * the dev_t is not 0,0.
- * If a pointer to a parent struct device is passed in, the newly created
- * struct device will be a child of that device in sysfs.
- * The pointer to the struct device will be returned from the call.
- * Any further sysfs files that might be required can be created using this
- * pointer.
- *
- * Note: the struct class passed to this function must have previously
- * been created with a call to class_create().
- */
-struct device *device_create(struct class *class, struct device *parent,
-			     dev_t devt, const char *fmt, ...)
-{
-	va_list vargs;
-	struct device *dev;
-
-	va_start(vargs, fmt);
-	dev = device_create_vargs(class, parent, devt, NULL, fmt, vargs);
-	va_end(vargs);
-	return dev;
-}
-EXPORT_SYMBOL_GPL(device_create);
-
 static int __match_devt(struct device *dev, void *data)
 {
 	dev_t *devt = data;
diff --git a/include/linux/device.h b/include/linux/device.h
index e49aa74f248c..a3ef5a2d4fbb 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -470,9 +470,6 @@ extern struct device *device_create_vargs(struct class *cls,
 					  void *drvdata,
 					  const char *fmt,
 					  va_list vargs);
-extern struct device *device_create(struct class *cls, struct device *parent,
-				    dev_t devt, const char *fmt, ...)
-				    __attribute__((format(printf, 4, 5)));
 extern struct device *device_create_drvdata(struct class *cls,
 					    struct device *parent,
 					    dev_t devt,
-- 
cgit v1.2.3


From 4e10673944a5c386378ff9d692ae37e19993f9d5 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 21 Jul 2008 20:03:34 -0700
Subject: device create: convert device_create_drvdata to device_create

Now that device_create() has been audited, rename things back to the
original call to be sane.

Keep the device_create_drvdata macro around to make merges easier.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 11 ++++-------
 include/linux/device.h |  9 ++++-----
 2 files changed, 8 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 4dc0d272f6d9..9ae28aa709d5 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -1236,7 +1236,7 @@ error:
 EXPORT_SYMBOL_GPL(device_create_vargs);
 
 /**
- * device_create_drvdata - creates a device and registers it with sysfs
+ * device_create - creates a device and registers it with sysfs
  * @class: pointer to the struct class that this device should be registered to
  * @parent: pointer to the parent struct device of this new device, if any
  * @devt: the dev_t for the char device to be added
@@ -1257,11 +1257,8 @@ EXPORT_SYMBOL_GPL(device_create_vargs);
  * Note: the struct class passed to this function must have previously
  * been created with a call to class_create().
  */
-struct device *device_create_drvdata(struct class *class,
-				     struct device *parent,
-				     dev_t devt,
-				     void *drvdata,
-				     const char *fmt, ...)
+struct device *device_create(struct class *class, struct device *parent,
+			     dev_t devt, void *drvdata, const char *fmt, ...)
 {
 	va_list vargs;
 	struct device *dev;
@@ -1271,7 +1268,7 @@ struct device *device_create_drvdata(struct class *class,
 	va_end(vargs);
 	return dev;
 }
-EXPORT_SYMBOL_GPL(device_create_drvdata);
+EXPORT_SYMBOL_GPL(device_create);
 
 static int __match_devt(struct device *dev, void *data)
 {
diff --git a/include/linux/device.h b/include/linux/device.h
index a3ef5a2d4fbb..de178712e02c 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -470,12 +470,11 @@ extern struct device *device_create_vargs(struct class *cls,
 					  void *drvdata,
 					  const char *fmt,
 					  va_list vargs);
-extern struct device *device_create_drvdata(struct class *cls,
-					    struct device *parent,
-					    dev_t devt,
-					    void *drvdata,
-					    const char *fmt, ...)
+extern struct device *device_create(struct class *cls, struct device *parent,
+				    dev_t devt, void *drvdata,
+				    const char *fmt, ...)
 				    __attribute__((format(printf, 5, 6)));
+#define device_create_drvdata	device_create
 extern void device_destroy(struct class *cls, dev_t devt);
 
 /*
-- 
cgit v1.2.3


From 93562b537659fc0f63920fd4d9d24f54e434f4c4 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 22 May 2008 17:21:08 -0400
Subject: Driver Core: add ability for class_for_each_device to start in middle
 of list

This mirrors the functionality that driver_for_each_device has as well.

We add a start variable, and all callers of the function are fixed up at
the same time.

The block layer will be using this new functionality in a follow-on
patch.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c              | 21 +++++++++++++--------
 drivers/i2c/i2c-core.c            |  6 ++++--
 drivers/ieee1394/nodemgr.c        | 14 +++++++++-----
 drivers/power/apm_power.c         |  2 +-
 drivers/power/power_supply_core.c |  4 ++--
 include/linux/device.h            |  3 ++-
 6 files changed, 31 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index 71ce3ff6bdf5..2eb7048003a8 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -256,11 +256,14 @@ char *make_class_name(const char *name, struct kobject *kobj)
 /**
  * class_for_each_device - device iterator
  * @class: the class we're iterating
+ * @start: the device to start with in the list, if any.
  * @data: data for the callback
  * @fn: function to be called for each device
  *
  * Iterate over @class's list of devices, and call @fn for each,
- * passing it @data.
+ * passing it @data.  If @start is set, the list iteration will start
+ * there, otherwise if it is NULL, the iteration starts at the
+ * beginning of the list.
  *
  * We check the return of @fn each time. If it returns anything
  * other than 0, we break out and return that value.
@@ -269,8 +272,8 @@ char *make_class_name(const char *name, struct kobject *kobj)
  * re-acquired in @fn, otherwise it will self-deadlocking. For
  * example, calls to add or remove class members would be verboten.
  */
-int class_for_each_device(struct class *class, void *data,
-			   int (*fn)(struct device *, void *))
+int class_for_each_device(struct class *class, struct device *start,
+			  void *data, int (*fn)(struct device *, void *))
 {
 	struct device *dev;
 	int error = 0;
@@ -279,12 +282,14 @@ int class_for_each_device(struct class *class, void *data,
 		return -EINVAL;
 	down(&class->sem);
 	list_for_each_entry(dev, &class->devices, node) {
+		if (start) {
+			if (start == dev)
+				start = NULL;
+			continue;
+		}
 		dev = get_device(dev);
-		if (dev) {
-			error = fn(dev, data);
-			put_device(dev);
-		} else
-			error = -ENODEV;
+		error = fn(dev, data);
+		put_device(dev);
 		if (error)
 			break;
 	}
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 7608df83d6d1..7bf38c418086 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -722,7 +722,8 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver)
 
 	INIT_LIST_HEAD(&driver->clients);
 	/* Walk the adapters that are already present */
-	class_for_each_device(&i2c_adapter_class, driver, __attach_adapter);
+	class_for_each_device(&i2c_adapter_class, NULL, driver,
+			      __attach_adapter);
 
 	mutex_unlock(&core_lock);
 	return 0;
@@ -782,7 +783,8 @@ void i2c_del_driver(struct i2c_driver *driver)
 {
 	mutex_lock(&core_lock);
 
-	class_for_each_device(&i2c_adapter_class, driver, __detach_adapter);
+	class_for_each_device(&i2c_adapter_class, NULL, driver,
+			      __detach_adapter);
 
 	driver_unregister(&driver->driver);
 	pr_debug("i2c-core: driver [%s] unregistered\n", driver->driver.name);
diff --git a/drivers/ieee1394/nodemgr.c b/drivers/ieee1394/nodemgr.c
index 05710c7c1220..47c0d85e0f32 100644
--- a/drivers/ieee1394/nodemgr.c
+++ b/drivers/ieee1394/nodemgr.c
@@ -1453,7 +1453,8 @@ static void nodemgr_suspend_ne(struct node_entry *ne)
 	ne->in_limbo = 1;
 	WARN_ON(device_create_file(&ne->device, &dev_attr_ne_in_limbo));
 
-	class_for_each_device(&nodemgr_ud_class, ne, __nodemgr_driver_suspend);
+	class_for_each_device(&nodemgr_ud_class, NULL, ne,
+			      __nodemgr_driver_suspend);
 }
 
 
@@ -1462,7 +1463,8 @@ static void nodemgr_resume_ne(struct node_entry *ne)
 	ne->in_limbo = 0;
 	device_remove_file(&ne->device, &dev_attr_ne_in_limbo);
 
-	class_for_each_device(&nodemgr_ud_class, ne, __nodemgr_driver_resume);
+	class_for_each_device(&nodemgr_ud_class, NULL, ne,
+			      __nodemgr_driver_resume);
 	HPSB_DEBUG("Node resumed: ID:BUS[" NODE_BUS_FMT "]  GUID[%016Lx]",
 		   NODE_BUS_ARGS(ne->host, ne->nodeid), (unsigned long long)ne->guid);
 }
@@ -1498,7 +1500,8 @@ static int __nodemgr_update_pdrv(struct device *dev, void *data)
 
 static void nodemgr_update_pdrv(struct node_entry *ne)
 {
-	class_for_each_device(&nodemgr_ud_class, ne, __nodemgr_update_pdrv);
+	class_for_each_device(&nodemgr_ud_class, NULL, ne,
+			      __nodemgr_update_pdrv);
 }
 
 
@@ -1591,7 +1594,8 @@ static void nodemgr_node_probe(struct host_info *hi, int generation)
 	 * while probes are time-consuming. (Well, those probes need some
 	 * improvement...) */
 
-	class_for_each_device(&nodemgr_ne_class, &param, __nodemgr_node_probe);
+	class_for_each_device(&nodemgr_ne_class, NULL, &param,
+			      __nodemgr_node_probe);
 
 	/* If we had a bus reset while we were scanning the bus, it is
 	 * possible that we did not probe all nodes.  In that case, we
@@ -1826,7 +1830,7 @@ int nodemgr_for_each_host(void *data, int (*cb)(struct hpsb_host *, void *))
 
 	hip.cb = cb;
 	hip.data = data;
-	error = class_for_each_device(&hpsb_host_class, &hip,
+	error = class_for_each_device(&hpsb_host_class, NULL, &hip,
 				      __nodemgr_for_each_host);
 
 	return error;
diff --git a/drivers/power/apm_power.c b/drivers/power/apm_power.c
index a4892275659d..936bae560fa1 100644
--- a/drivers/power/apm_power.c
+++ b/drivers/power/apm_power.c
@@ -78,7 +78,7 @@ static void find_main_battery(void)
 	main_battery = NULL;
 	bp.main = main_battery;
 
-	error = class_for_each_device(power_supply_class, &bp,
+	error = class_for_each_device(power_supply_class, NULL, &bp,
 				      __find_main_battery);
 	if (error) {
 		main_battery = bp.main;
diff --git a/drivers/power/power_supply_core.c b/drivers/power/power_supply_core.c
index af1633eb3b70..cb1ccb472921 100644
--- a/drivers/power/power_supply_core.c
+++ b/drivers/power/power_supply_core.c
@@ -41,7 +41,7 @@ static void power_supply_changed_work(struct work_struct *work)
 
 	dev_dbg(psy->dev, "%s\n", __func__);
 
-	class_for_each_device(power_supply_class, psy,
+	class_for_each_device(power_supply_class, NULL, psy,
 			      __power_supply_changed_work);
 
 	power_supply_update_leds(psy);
@@ -79,7 +79,7 @@ int power_supply_am_i_supplied(struct power_supply *psy)
 {
 	int error;
 
-	error = class_for_each_device(power_supply_class, psy,
+	error = class_for_each_device(power_supply_class, NULL, psy,
 				      __power_supply_am_i_supplied);
 
 	dev_dbg(psy->dev, "%s %d\n", __func__, error);
diff --git a/include/linux/device.h b/include/linux/device.h
index de178712e02c..6d5b351b29c9 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -210,7 +210,8 @@ extern struct kobject *sysfs_dev_block_kobj;
 extern struct kobject *sysfs_dev_char_kobj;
 extern int __must_check class_register(struct class *class);
 extern void class_unregister(struct class *class);
-extern int class_for_each_device(struct class *class, void *data,
+extern int class_for_each_device(struct class *class, struct device *start,
+				 void *data,
 				 int (*fn)(struct device *dev, void *data));
 extern struct device *class_find_device(struct class *class, void *data,
 					int (*match)(struct device *, void *));
-- 
cgit v1.2.3


From 695794ae0c5bdd9bd06e35b118801e2e9be04f9e Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 22 May 2008 17:21:08 -0400
Subject: Driver Core: add ability for class_find_device to start in middle of
 list

This mirrors the functionality that driver_find_device has as well.

We add a start variable, and all callers of the function are fixed up at
the same time.

The block layer will be using this new functionality in a follow-on
patch.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c                | 22 +++++++++++++---------
 drivers/base/core.c                 |  2 +-
 drivers/ieee1394/nodemgr.c          |  9 ++++++---
 drivers/rtc/interface.c             |  2 +-
 drivers/scsi/hosts.c                |  3 ++-
 drivers/scsi/scsi_transport_iscsi.c |  4 ++--
 drivers/spi/spi.c                   |  2 +-
 include/linux/device.h              |  3 ++-
 8 files changed, 28 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index 2eb7048003a8..3918d0e432d4 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -302,6 +302,7 @@ EXPORT_SYMBOL_GPL(class_for_each_device);
 /**
  * class_find_device - device iterator for locating a particular device
  * @class: the class we're iterating
+ * @start: Device to begin with
  * @data: data for the match function
  * @match: function to check device
  *
@@ -319,8 +320,9 @@ EXPORT_SYMBOL_GPL(class_for_each_device);
  * re-acquired in @match, otherwise it will self-deadlocking. For
  * example, calls to add or remove class members would be verboten.
  */
-struct device *class_find_device(struct class *class, void *data,
-				   int (*match)(struct device *, void *))
+struct device *class_find_device(struct class *class, struct device *start,
+				 void *data,
+				 int (*match)(struct device *, void *))
 {
 	struct device *dev;
 	int found = 0;
@@ -330,15 +332,17 @@ struct device *class_find_device(struct class *class, void *data,
 
 	down(&class->sem);
 	list_for_each_entry(dev, &class->devices, node) {
+		if (start) {
+			if (start == dev)
+				start = NULL;
+			continue;
+		}
 		dev = get_device(dev);
-		if (dev) {
-			if (match(dev, data)) {
-				found = 1;
-				break;
-			} else
-				put_device(dev);
-		} else
+		if (match(dev, data)) {
+			found = 1;
 			break;
+		} else
+			put_device(dev);
 	}
 	up(&class->sem);
 
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 9ae28aa709d5..9f05de6f80b5 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -1289,7 +1289,7 @@ void device_destroy(struct class *class, dev_t devt)
 {
 	struct device *dev;
 
-	dev = class_find_device(class, &devt, __match_devt);
+	dev = class_find_device(class, NULL, &devt, __match_devt);
 	if (dev) {
 		put_device(dev);
 		device_unregister(dev);
diff --git a/drivers/ieee1394/nodemgr.c b/drivers/ieee1394/nodemgr.c
index 47c0d85e0f32..994a21e5a0aa 100644
--- a/drivers/ieee1394/nodemgr.c
+++ b/drivers/ieee1394/nodemgr.c
@@ -754,7 +754,8 @@ static void nodemgr_remove_uds(struct node_entry *ne)
 	 */
 	mutex_lock(&nodemgr_serialize_remove_uds);
 	for (;;) {
-		dev = class_find_device(&nodemgr_ud_class, ne, __match_ne);
+		dev = class_find_device(&nodemgr_ud_class, NULL, ne,
+					__match_ne);
 		if (!dev)
 			break;
 		ud = container_of(dev, struct unit_directory, unit_dev);
@@ -901,7 +902,8 @@ static struct node_entry *find_entry_by_guid(u64 guid)
 	struct device *dev;
 	struct node_entry *ne;
 
-	dev = class_find_device(&nodemgr_ne_class, &guid, __match_ne_guid);
+	dev = class_find_device(&nodemgr_ne_class, NULL, &guid,
+				__match_ne_guid);
 	if (!dev)
 		return NULL;
 	ne = container_of(dev, struct node_entry, node_dev);
@@ -940,7 +942,8 @@ static struct node_entry *find_entry_by_nodeid(struct hpsb_host *host,
 	param.host = host;
 	param.nodeid = nodeid;
 
-	dev = class_find_device(&nodemgr_ne_class, &param, __match_ne_nodeid);
+	dev = class_find_device(&nodemgr_ne_class, NULL, &param,
+				__match_ne_nodeid);
 	if (!dev)
 		return NULL;
 	ne = container_of(dev, struct node_entry, node_dev);
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index 58b7336640ff..d397fa5f3a91 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -345,7 +345,7 @@ struct rtc_device *rtc_class_open(char *name)
 	struct device *dev;
 	struct rtc_device *rtc = NULL;
 
-	dev = class_find_device(rtc_class, name, __rtc_match);
+	dev = class_find_device(rtc_class, NULL, name, __rtc_match);
 	if (dev)
 		rtc = to_rtc_device(dev);
 
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 35cd892dce04..78dad28b70d5 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -466,7 +466,8 @@ struct Scsi_Host *scsi_host_lookup(unsigned short hostnum)
 	struct device *cdev;
 	struct Scsi_Host *shost = ERR_PTR(-ENXIO);
 
-	cdev = class_find_device(&shost_class, &hostnum, __scsi_host_match);
+	cdev = class_find_device(&shost_class, NULL, &hostnum,
+				 __scsi_host_match);
 	if (cdev) {
 		shost = scsi_host_get(class_to_shost(cdev));
 		put_device(cdev);
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 3af7cbcc5c5d..06748f318cd5 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -170,7 +170,7 @@ iscsi_create_endpoint(int dd_size)
 	int err;
 
 	for (id = 1; id < ISCSI_MAX_EPID; id++) {
-		dev = class_find_device(&iscsi_endpoint_class, &id,
+		dev = class_find_device(&iscsi_endpoint_class, NULL, &id,
 					iscsi_match_epid);
 		if (!dev)
 			break;
@@ -222,7 +222,7 @@ struct iscsi_endpoint *iscsi_lookup_endpoint(u64 handle)
 	struct iscsi_endpoint *ep;
 	struct device *dev;
 
-	dev = class_find_device(&iscsi_endpoint_class, &handle,
+	dev = class_find_device(&iscsi_endpoint_class, NULL, &handle,
 				iscsi_match_epid);
 	if (!dev)
 		return NULL;
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 1ad12afc6ba0..1771b2456bfa 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -502,7 +502,7 @@ struct spi_master *spi_busnum_to_master(u16 bus_num)
 	struct device		*dev;
 	struct spi_master	*master = NULL;
 
-	dev = class_find_device(&spi_master_class, &bus_num,
+	dev = class_find_device(&spi_master_class, NULL, &bus_num,
 				__spi_master_match);
 	if (dev)
 		master = container_of(dev, struct spi_master, dev);
diff --git a/include/linux/device.h b/include/linux/device.h
index 6d5b351b29c9..c1f72984875f 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -213,7 +213,8 @@ extern void class_unregister(struct class *class);
 extern int class_for_each_device(struct class *class, struct device *start,
 				 void *data,
 				 int (*fn)(struct device *dev, void *data));
-extern struct device *class_find_device(struct class *class, void *data,
+extern struct device *class_find_device(struct class *class,
+					struct device *start, void *data,
 					int (*match)(struct device *, void *));
 
 struct class_attribute {
-- 
cgit v1.2.3


From 7c71448b8aa80123fc521563d5f7c63a099d97ab Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 22 Jan 2008 18:17:41 -0500
Subject: class: move driver core specific parts to a private structure

This moves the portions of struct class that are dynamic (kobject and
lock and lists) out of the main structure and into a dynamic, private,
structure.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    | 27 ++++++++++++++++
 drivers/base/class.c   | 87 ++++++++++++++++++++++++++++----------------------
 drivers/base/core.c    | 57 +++++++++++++++++----------------
 include/linux/device.h |  7 ++--
 4 files changed, 107 insertions(+), 71 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 2c9ae43e2219..0ec372a67762 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -36,6 +36,33 @@ struct driver_private {
 };
 #define to_driver(obj) container_of(obj, struct driver_private, kobj)
 
+
+/**
+ * struct class_private - structure to hold the private to the driver core portions of the class structure.
+ *
+ * @subsys - the struct kset that defines this class.  This is the main kobject
+ * @children - list of class_devices associated with this class
+ * @devices - list of devices associated with this class
+ * @interfaces - list of class_interfaces associated with this class
+ * @class_dirs -
+ * @sem - semaphore to protect the children, devices, and interfaces lists.
+ * @class - pointer back to the struct class that this structure is associated
+ * with.
+ *
+ * This structure is the one that is the actual kobject allowing struct
+ * class to be statically allocated safely.  Nothing outside of the driver
+ * core should ever touch these fields.
+ */
+struct class_private {
+	struct kset subsys;
+	struct list_head devices;
+	struct list_head interfaces;
+	struct kset class_dirs;
+	struct semaphore sem;
+	struct class *class;
+};
+#define to_class(obj) container_of(obj, struct class_private, subsys.kobj)
+
 /* initialisation functions */
 extern int devices_init(void);
 extern int buses_init(void);
diff --git a/drivers/base/class.c b/drivers/base/class.c
index 3918d0e432d4..06f09c929a91 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -21,17 +21,16 @@
 #include "base.h"
 
 #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr)
-#define to_class(obj) container_of(obj, struct class, subsys.kobj)
 
 static ssize_t class_attr_show(struct kobject *kobj, struct attribute *attr,
 			       char *buf)
 {
 	struct class_attribute *class_attr = to_class_attr(attr);
-	struct class *dc = to_class(kobj);
+	struct class_private *cp = to_class(kobj);
 	ssize_t ret = -EIO;
 
 	if (class_attr->show)
-		ret = class_attr->show(dc, buf);
+		ret = class_attr->show(cp->class, buf);
 	return ret;
 }
 
@@ -39,17 +38,18 @@ static ssize_t class_attr_store(struct kobject *kobj, struct attribute *attr,
 				const char *buf, size_t count)
 {
 	struct class_attribute *class_attr = to_class_attr(attr);
-	struct class *dc = to_class(kobj);
+	struct class_private *cp = to_class(kobj);
 	ssize_t ret = -EIO;
 
 	if (class_attr->store)
-		ret = class_attr->store(dc, buf, count);
+		ret = class_attr->store(cp->class, buf, count);
 	return ret;
 }
 
 static void class_release(struct kobject *kobj)
 {
-	struct class *class = to_class(kobj);
+	struct class_private *cp = to_class(kobj);
+	struct class *class = cp->class;
 
 	pr_debug("class '%s': release.\n", class->name);
 
@@ -78,7 +78,7 @@ int class_create_file(struct class *cls, const struct class_attribute *attr)
 {
 	int error;
 	if (cls)
-		error = sysfs_create_file(&cls->subsys.kobj, &attr->attr);
+		error = sysfs_create_file(&cls->p->subsys.kobj, &attr->attr);
 	else
 		error = -EINVAL;
 	return error;
@@ -87,21 +87,20 @@ int class_create_file(struct class *cls, const struct class_attribute *attr)
 void class_remove_file(struct class *cls, const struct class_attribute *attr)
 {
 	if (cls)
-		sysfs_remove_file(&cls->subsys.kobj, &attr->attr);
+		sysfs_remove_file(&cls->p->subsys.kobj, &attr->attr);
 }
 
 static struct class *class_get(struct class *cls)
 {
 	if (cls)
-		return container_of(kset_get(&cls->subsys),
-				    struct class, subsys);
-	return NULL;
+		kset_get(&cls->p->subsys);
+	return cls;
 }
 
 static void class_put(struct class *cls)
 {
 	if (cls)
-		kset_put(&cls->subsys);
+		kset_put(&cls->p->subsys);
 }
 
 static int add_class_attrs(struct class *cls)
@@ -136,17 +135,23 @@ static void remove_class_attrs(struct class *cls)
 
 int class_register(struct class *cls)
 {
+	struct class_private *cp;
 	int error;
 
 	pr_debug("device class '%s': registering\n", cls->name);
 
-	INIT_LIST_HEAD(&cls->devices);
-	INIT_LIST_HEAD(&cls->interfaces);
-	kset_init(&cls->class_dirs);
-	init_MUTEX(&cls->sem);
-	error = kobject_set_name(&cls->subsys.kobj, "%s", cls->name);
-	if (error)
+	cp = kzalloc(sizeof(*cp), GFP_KERNEL);
+	if (!cp)
+		return -ENOMEM;
+	INIT_LIST_HEAD(&cp->devices);
+	INIT_LIST_HEAD(&cp->interfaces);
+	kset_init(&cp->class_dirs);
+	init_MUTEX(&cp->sem);
+	error = kobject_set_name(&cp->subsys.kobj, "%s", cls->name);
+	if (error) {
+		kfree(cp);
 		return error;
+	}
 
 	/* set the default /sys/dev directory for devices of this class */
 	if (!cls->dev_kobj)
@@ -155,17 +160,21 @@ int class_register(struct class *cls)
 #if defined(CONFIG_SYSFS_DEPRECATED) && defined(CONFIG_BLOCK)
 	/* let the block class directory show up in the root of sysfs */
 	if (cls != &block_class)
-		cls->subsys.kobj.kset = class_kset;
+		cp->subsys.kobj.kset = class_kset;
 #else
-	cls->subsys.kobj.kset = class_kset;
+	cp->subsys.kobj.kset = class_kset;
 #endif
-	cls->subsys.kobj.ktype = &class_ktype;
+	cp->subsys.kobj.ktype = &class_ktype;
+	cp->class = cls;
+	cls->p = cp;
 
-	error = kset_register(&cls->subsys);
-	if (!error) {
-		error = add_class_attrs(class_get(cls));
-		class_put(cls);
+	error = kset_register(&cp->subsys);
+	if (error) {
+		kfree(cp);
+		return error;
 	}
+	error = add_class_attrs(class_get(cls));
+	class_put(cls);
 	return error;
 }
 
@@ -173,7 +182,7 @@ void class_unregister(struct class *cls)
 {
 	pr_debug("device class '%s': unregistering\n", cls->name);
 	remove_class_attrs(cls);
-	kset_unregister(&cls->subsys);
+	kset_unregister(&cls->p->subsys);
 }
 
 static void class_create_release(struct class *cls)
@@ -280,8 +289,8 @@ int class_for_each_device(struct class *class, struct device *start,
 
 	if (!class)
 		return -EINVAL;
-	down(&class->sem);
-	list_for_each_entry(dev, &class->devices, node) {
+	down(&class->p->sem);
+	list_for_each_entry(dev, &class->p->devices, node) {
 		if (start) {
 			if (start == dev)
 				start = NULL;
@@ -293,7 +302,7 @@ int class_for_each_device(struct class *class, struct device *start,
 		if (error)
 			break;
 	}
-	up(&class->sem);
+	up(&class->p->sem);
 
 	return error;
 }
@@ -330,8 +339,8 @@ struct device *class_find_device(struct class *class, struct device *start,
 	if (!class)
 		return NULL;
 
-	down(&class->sem);
-	list_for_each_entry(dev, &class->devices, node) {
+	down(&class->p->sem);
+	list_for_each_entry(dev, &class->p->devices, node) {
 		if (start) {
 			if (start == dev)
 				start = NULL;
@@ -344,7 +353,7 @@ struct device *class_find_device(struct class *class, struct device *start,
 		} else
 			put_device(dev);
 	}
-	up(&class->sem);
+	up(&class->p->sem);
 
 	return found ? dev : NULL;
 }
@@ -362,13 +371,13 @@ int class_interface_register(struct class_interface *class_intf)
 	if (!parent)
 		return -EINVAL;
 
-	down(&parent->sem);
-	list_add_tail(&class_intf->node, &parent->interfaces);
+	down(&parent->p->sem);
+	list_add_tail(&class_intf->node, &parent->p->interfaces);
 	if (class_intf->add_dev) {
-		list_for_each_entry(dev, &parent->devices, node)
+		list_for_each_entry(dev, &parent->p->devices, node)
 			class_intf->add_dev(dev, class_intf);
 	}
-	up(&parent->sem);
+	up(&parent->p->sem);
 
 	return 0;
 }
@@ -381,13 +390,13 @@ void class_interface_unregister(struct class_interface *class_intf)
 	if (!parent)
 		return;
 
-	down(&parent->sem);
+	down(&parent->p->sem);
 	list_del_init(&class_intf->node);
 	if (class_intf->remove_dev) {
-		list_for_each_entry(dev, &parent->devices, node)
+		list_for_each_entry(dev, &parent->p->devices, node)
 			class_intf->remove_dev(dev, class_intf);
 	}
-	up(&parent->sem);
+	up(&parent->p->sem);
 
 	class_put(parent);
 }
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 9f05de6f80b5..64c150b5a883 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -551,7 +551,7 @@ static struct kobject *get_device_parent(struct device *dev,
 {
 	/* class devices without a parent live in /sys/class/<classname>/ */
 	if (dev->class && (!parent || parent->class != dev->class))
-		return &dev->class->subsys.kobj;
+		return &dev->class->p->subsys.kobj;
 	/* all other devices keep their parent */
 	else if (parent)
 		return &parent->kobj;
@@ -597,13 +597,13 @@ static struct kobject *get_device_parent(struct device *dev,
 			parent_kobj = &parent->kobj;
 
 		/* find our class-directory at the parent and reference it */
-		spin_lock(&dev->class->class_dirs.list_lock);
-		list_for_each_entry(k, &dev->class->class_dirs.list, entry)
+		spin_lock(&dev->class->p->class_dirs.list_lock);
+		list_for_each_entry(k, &dev->class->p->class_dirs.list, entry)
 			if (k->parent == parent_kobj) {
 				kobj = kobject_get(k);
 				break;
 			}
-		spin_unlock(&dev->class->class_dirs.list_lock);
+		spin_unlock(&dev->class->p->class_dirs.list_lock);
 		if (kobj)
 			return kobj;
 
@@ -611,7 +611,7 @@ static struct kobject *get_device_parent(struct device *dev,
 		k = kobject_create();
 		if (!k)
 			return NULL;
-		k->kset = &dev->class->class_dirs;
+		k->kset = &dev->class->p->class_dirs;
 		retval = kobject_add(k, parent_kobj, "%s", dev->class->name);
 		if (retval < 0) {
 			kobject_put(k);
@@ -630,7 +630,7 @@ static void cleanup_glue_dir(struct device *dev, struct kobject *glue_dir)
 {
 	/* see if we live in a "glue" directory */
 	if (!glue_dir || !dev->class ||
-	    glue_dir->kset != &dev->class->class_dirs)
+	    glue_dir->kset != &dev->class->p->class_dirs)
 		return;
 
 	kobject_put(glue_dir);
@@ -657,17 +657,17 @@ static int device_add_class_symlinks(struct device *dev)
 	if (!dev->class)
 		return 0;
 
-	error = sysfs_create_link(&dev->kobj, &dev->class->subsys.kobj,
+	error = sysfs_create_link(&dev->kobj, &dev->class->p->subsys.kobj,
 				  "subsystem");
 	if (error)
 		goto out;
 
 #ifdef CONFIG_SYSFS_DEPRECATED
 	/* stacked class devices need a symlink in the class directory */
-	if (dev->kobj.parent != &dev->class->subsys.kobj &&
+	if (dev->kobj.parent != &dev->class->p->subsys.kobj &&
 	    device_is_not_partition(dev)) {
-		error = sysfs_create_link(&dev->class->subsys.kobj, &dev->kobj,
-					  dev->bus_id);
+		error = sysfs_create_link(&dev->class->p->subsys.kobj,
+					  &dev->kobj, dev->bus_id);
 		if (error)
 			goto out_subsys;
 	}
@@ -704,12 +704,12 @@ out_device:
 	if (dev->parent && device_is_not_partition(dev))
 		sysfs_remove_link(&dev->kobj, "device");
 out_busid:
-	if (dev->kobj.parent != &dev->class->subsys.kobj &&
+	if (dev->kobj.parent != &dev->class->p->subsys.kobj &&
 	    device_is_not_partition(dev))
-		sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id);
+		sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id);
 #else
 	/* link in the class directory pointing to the device */
-	error = sysfs_create_link(&dev->class->subsys.kobj, &dev->kobj,
+	error = sysfs_create_link(&dev->class->p->subsys.kobj, &dev->kobj,
 				  dev->bus_id);
 	if (error)
 		goto out_subsys;
@@ -723,7 +723,7 @@ out_busid:
 	return 0;
 
 out_busid:
-	sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id);
+	sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id);
 #endif
 
 out_subsys:
@@ -749,14 +749,14 @@ static void device_remove_class_symlinks(struct device *dev)
 		sysfs_remove_link(&dev->kobj, "device");
 	}
 
-	if (dev->kobj.parent != &dev->class->subsys.kobj &&
+	if (dev->kobj.parent != &dev->class->p->subsys.kobj &&
 	    device_is_not_partition(dev))
-		sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id);
+		sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id);
 #else
 	if (dev->parent && device_is_not_partition(dev))
 		sysfs_remove_link(&dev->kobj, "device");
 
-	sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id);
+	sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id);
 #endif
 
 	sysfs_remove_link(&dev->kobj, "subsystem");
@@ -904,15 +904,16 @@ int device_add(struct device *dev)
 		klist_add_tail(&dev->knode_parent, &parent->klist_children);
 
 	if (dev->class) {
-		down(&dev->class->sem);
+		down(&dev->class->p->sem);
 		/* tie the class to the device */
-		list_add_tail(&dev->node, &dev->class->devices);
+		list_add_tail(&dev->node, &dev->class->p->devices);
 
 		/* notify any interfaces that the device is here */
-		list_for_each_entry(class_intf, &dev->class->interfaces, node)
+		list_for_each_entry(class_intf, &dev->class->p->interfaces,
+				    node)
 			if (class_intf->add_dev)
 				class_intf->add_dev(dev, class_intf);
-		up(&dev->class->sem);
+		up(&dev->class->p->sem);
 	}
  Done:
 	put_device(dev);
@@ -1013,14 +1014,15 @@ void device_del(struct device *dev)
 	if (dev->class) {
 		device_remove_class_symlinks(dev);
 
-		down(&dev->class->sem);
+		down(&dev->class->p->sem);
 		/* notify any interfaces that the device is now gone */
-		list_for_each_entry(class_intf, &dev->class->interfaces, node)
+		list_for_each_entry(class_intf, &dev->class->p->interfaces,
+				    node)
 			if (class_intf->remove_dev)
 				class_intf->remove_dev(dev, class_intf);
 		/* remove the device from the class list */
 		list_del_init(&dev->node);
-		up(&dev->class->sem);
+		up(&dev->class->p->sem);
 	}
 	device_remove_file(dev, &uevent_attr);
 	device_remove_attrs(dev);
@@ -1348,11 +1350,12 @@ int device_rename(struct device *dev, char *new_name)
 	}
 #else
 	if (dev->class) {
-		error = sysfs_create_link(&dev->class->subsys.kobj, &dev->kobj,
-					  dev->bus_id);
+		error = sysfs_create_link(&dev->class->p->subsys.kobj,
+					  &dev->kobj, dev->bus_id);
 		if (error)
 			goto out;
-		sysfs_remove_link(&dev->class->subsys.kobj, old_device_name);
+		sysfs_remove_link(&dev->class->p->subsys.kobj,
+				  old_device_name);
 	}
 #endif
 
diff --git a/include/linux/device.h b/include/linux/device.h
index c1f72984875f..b0556082179b 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -35,6 +35,7 @@ struct device;
 struct device_driver;
 struct driver_private;
 struct class;
+struct class_private;
 struct bus_type;
 struct bus_type_private;
 
@@ -186,11 +187,6 @@ struct class {
 	const char		*name;
 	struct module		*owner;
 
-	struct kset		subsys;
-	struct list_head	devices;
-	struct list_head	interfaces;
-	struct kset		class_dirs;
-	struct semaphore	sem; /* locks children, devices, interfaces */
 	struct class_attribute		*class_attrs;
 	struct device_attribute		*dev_attrs;
 	struct kobject			*dev_kobj;
@@ -204,6 +200,7 @@ struct class {
 	int (*resume)(struct device *dev);
 
 	struct pm_ops *pm;
+	struct class_private *p;
 };
 
 extern struct kobject *sysfs_dev_block_kobj;
-- 
cgit v1.2.3


From d2a3b9146e4f40c2e872d7567c996ef95083d802 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Wed, 28 May 2008 09:28:39 -0700
Subject: class: add lockdep infrastructure

This adds the infrastructure to properly handle lockdep issues when the
internal class semaphore is changed to a mutex.

Matthew wrote the original patch, and Greg fixed it up to work properly
with the class_create() function.


From: Matthew Wilcox <matthew@wil.cx>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Dave Young <hidave.darkstar@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 11 ++++++-----
 include/linux/device.h | 25 +++++++++++++++++++++++--
 2 files changed, 29 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/class.c b/drivers/base/class.c
index d24d21114ccb..89000566690c 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -134,7 +134,7 @@ static void remove_class_attrs(struct class *cls)
 	}
 }
 
-int class_register(struct class *cls)
+int __class_register(struct class *cls, struct lock_class_key *key)
 {
 	struct class_private *cp;
 	int error;
@@ -178,6 +178,7 @@ int class_register(struct class *cls)
 	class_put(cls);
 	return error;
 }
+EXPORT_SYMBOL_GPL(__class_register);
 
 void class_unregister(struct class *cls)
 {
@@ -203,7 +204,8 @@ static void class_create_release(struct class *cls)
  * Note, the pointer created here is to be destroyed when finished by
  * making a call to class_destroy().
  */
-struct class *class_create(struct module *owner, const char *name)
+struct class *__class_create(struct module *owner, const char *name,
+			     struct lock_class_key *key)
 {
 	struct class *cls;
 	int retval;
@@ -218,7 +220,7 @@ struct class *class_create(struct module *owner, const char *name)
 	cls->owner = owner;
 	cls->class_release = class_create_release;
 
-	retval = class_register(cls);
+	retval = __class_register(cls, key);
 	if (retval)
 		goto error;
 
@@ -228,6 +230,7 @@ error:
 	kfree(cls);
 	return ERR_PTR(retval);
 }
+EXPORT_SYMBOL_GPL(__class_create);
 
 /**
  * class_destroy - destroys a struct class structure
@@ -412,9 +415,7 @@ int __init classes_init(void)
 
 EXPORT_SYMBOL_GPL(class_create_file);
 EXPORT_SYMBOL_GPL(class_remove_file);
-EXPORT_SYMBOL_GPL(class_register);
 EXPORT_SYMBOL_GPL(class_unregister);
-EXPORT_SYMBOL_GPL(class_create);
 EXPORT_SYMBOL_GPL(class_destroy);
 
 EXPORT_SYMBOL_GPL(class_interface_register);
diff --git a/include/linux/device.h b/include/linux/device.h
index b0556082179b..0e1d24c2ed41 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -16,6 +16,7 @@
 #include <linux/kobject.h>
 #include <linux/klist.h>
 #include <linux/list.h>
+#include <linux/lockdep.h>
 #include <linux/compiler.h>
 #include <linux/types.h>
 #include <linux/module.h>
@@ -205,8 +206,18 @@ struct class {
 
 extern struct kobject *sysfs_dev_block_kobj;
 extern struct kobject *sysfs_dev_char_kobj;
-extern int __must_check class_register(struct class *class);
+extern int __must_check __class_register(struct class *class,
+					 struct lock_class_key *key);
 extern void class_unregister(struct class *class);
+
+/* This is a #define to keep the compiler from merging different
+ * instances of the __key variable */
+#define class_register(class)			\
+({						\
+	static struct lock_class_key __key;	\
+	__class_register(class, &__key);	\
+})
+
 extern int class_for_each_device(struct class *class, struct device *start,
 				 void *data,
 				 int (*fn)(struct device *dev, void *data));
@@ -239,9 +250,19 @@ struct class_interface {
 extern int __must_check class_interface_register(struct class_interface *);
 extern void class_interface_unregister(struct class_interface *);
 
-extern struct class *class_create(struct module *owner, const char *name);
+extern struct class * __must_check __class_create(struct module *owner,
+						  const char *name,
+						  struct lock_class_key *key);
 extern void class_destroy(struct class *cls);
 
+/* This is a #define to keep the compiler from merging different
+ * instances of the __key variable */
+#define class_create(owner, name)		\
+({						\
+	static struct lock_class_key __key;	\
+	__class_create(owner, name, &__key);	\
+})
+
 /*
  * The type of device, "struct device" is embedded in. A class
  * or bus can contain devices of different types
-- 
cgit v1.2.3


From aab0de245150c09e61c30962feb16aacde508dc3 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 2 May 2008 06:02:41 +0200
Subject: driver core: remove KOBJ_NAME_LEN define

Kobjects do not have a limit in name size since a while, so stop
pretending that they do.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/plat-omap/mailbox.c        | 2 +-
 arch/sparc64/kernel/vio.c           | 2 +-
 drivers/message/fusion/mptbase.c    | 3 ++-
 drivers/message/fusion/mptbase.h    | 4 ++--
 drivers/message/fusion/mptfc.c      | 4 ++--
 drivers/pci/hotplug/acpiphp.h       | 4 ++--
 drivers/scsi/hosts.c                | 4 ++--
 drivers/scsi/scsi_transport_fc.c    | 9 +++++----
 drivers/scsi/scsi_transport_iscsi.c | 4 ++--
 fs/partitions/check.c               | 2 +-
 include/linux/device.h              | 3 +--
 include/linux/kobject.h             | 1 -
 include/linux/spi/spi.h             | 2 +-
 include/scsi/scsi_host.h            | 2 +-
 include/scsi/scsi_transport_fc.h    | 4 ++--
 include/scsi/scsi_transport_iscsi.h | 2 +-
 16 files changed, 26 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/plat-omap/mailbox.c b/arch/arm/plat-omap/mailbox.c
index 6f33f58bca45..848fdcafaa28 100644
--- a/arch/arm/plat-omap/mailbox.c
+++ b/arch/arm/plat-omap/mailbox.c
@@ -334,7 +334,7 @@ static int omap_mbox_init(struct omap_mbox *mbox)
 	}
 
 	mbox->dev.class = &omap_mbox_class;
-	strlcpy(mbox->dev.bus_id, mbox->name, KOBJ_NAME_LEN);
+	strlcpy(mbox->dev.bus_id, mbox->name, BUS_ID_SIZE);
 	dev_set_drvdata(&mbox->dev, mbox);
 
 	ret = device_register(&mbox->dev);
diff --git a/arch/sparc64/kernel/vio.c b/arch/sparc64/kernel/vio.c
index e78b3517940b..ecbb8b618b8c 100644
--- a/arch/sparc64/kernel/vio.c
+++ b/arch/sparc64/kernel/vio.c
@@ -224,7 +224,7 @@ static struct vio_dev *vio_create_one(struct mdesc_handle *hp, u64 mp,
 	if (!strcmp(type, "domain-services-port"))
 		bus_id_name = "ds";
 
-	if (strlen(bus_id_name) >= KOBJ_NAME_LEN - 4) {
+	if (strlen(bus_id_name) >= BUS_ID_SIZE - 4) {
 		printk(KERN_ERR "VIO: bus_id_name [%s] is too long.\n",
 		       bus_id_name);
 		return NULL;
diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c
index 75e599b85b64..34402c47027e 100644
--- a/drivers/message/fusion/mptbase.c
+++ b/drivers/message/fusion/mptbase.c
@@ -1670,7 +1670,8 @@ mpt_attach(struct pci_dev *pdev, const struct pci_device_id *id)
 	INIT_DELAYED_WORK(&ioc->fault_reset_work, mpt_fault_reset_work);
 	spin_lock_init(&ioc->fault_reset_work_lock);
 
-	snprintf(ioc->reset_work_q_name, KOBJ_NAME_LEN, "mpt_poll_%d", ioc->id);
+	snprintf(ioc->reset_work_q_name, sizeof(ioc->reset_work_q_name),
+		 "mpt_poll_%d", ioc->id);
 	ioc->reset_work_q =
 		create_singlethread_workqueue(ioc->reset_work_q_name);
 	if (!ioc->reset_work_q) {
diff --git a/drivers/message/fusion/mptbase.h b/drivers/message/fusion/mptbase.h
index 6adab648dbb9..dff048cfa101 100644
--- a/drivers/message/fusion/mptbase.h
+++ b/drivers/message/fusion/mptbase.h
@@ -707,12 +707,12 @@ typedef struct _MPT_ADAPTER
 	u8			 fc_link_speed[2];
 	spinlock_t		 fc_rescan_work_lock;
 	struct work_struct	 fc_rescan_work;
-	char			 fc_rescan_work_q_name[KOBJ_NAME_LEN];
+	char			 fc_rescan_work_q_name[20];
 	struct workqueue_struct *fc_rescan_work_q;
 	struct scsi_cmnd	**ScsiLookup;
 	spinlock_t		  scsi_lookup_lock;
 
-	char			 reset_work_q_name[KOBJ_NAME_LEN];
+	char			 reset_work_q_name[20];
 	struct workqueue_struct *reset_work_q;
 	struct delayed_work	 fault_reset_work;
 	spinlock_t		 fault_reset_work_lock;
diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c
index fc31ca6829d8..b36cae9ec6db 100644
--- a/drivers/message/fusion/mptfc.c
+++ b/drivers/message/fusion/mptfc.c
@@ -1326,8 +1326,8 @@ mptfc_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	/* initialize workqueue */
 
-	snprintf(ioc->fc_rescan_work_q_name, KOBJ_NAME_LEN, "mptfc_wq_%d",
-		sh->host_no);
+	snprintf(ioc->fc_rescan_work_q_name, sizeof(ioc->fc_rescan_work_q_name),
+		 "mptfc_wq_%d", sh->host_no);
 	ioc->fc_rescan_work_q =
 		create_singlethread_workqueue(ioc->fc_rescan_work_q_name);
 	if (!ioc->fc_rescan_work_q)
diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h
index eecf7cbf4139..5a58b075dd8d 100644
--- a/drivers/pci/hotplug/acpiphp.h
+++ b/drivers/pci/hotplug/acpiphp.h
@@ -36,7 +36,7 @@
 #define _ACPIPHP_H
 
 #include <linux/acpi.h>
-#include <linux/kobject.h>	/* for KOBJ_NAME_LEN */
+#include <linux/kobject.h>
 #include <linux/mutex.h>
 #include <linux/pci_hotplug.h>
 
@@ -51,7 +51,7 @@
 #define warn(format, arg...) printk(KERN_WARNING "%s: " format, MY_NAME , ## arg)
 
 /* name size which is used for entries in pcihpfs */
-#define SLOT_NAME_SIZE	KOBJ_NAME_LEN		/* {_SUN} */
+#define SLOT_NAME_SIZE	20		/* {_SUN} */
 
 struct acpiphp_bridge;
 struct acpiphp_slot;
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 78dad28b70d5..fed0b02ebc1d 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -232,8 +232,8 @@ int scsi_add_host(struct Scsi_Host *shost, struct device *dev)
 	}
 
 	if (shost->transportt->create_work_queue) {
-		snprintf(shost->work_q_name, KOBJ_NAME_LEN, "scsi_wq_%d",
-			shost->host_no);
+		snprintf(shost->work_q_name, sizeof(shost->work_q_name),
+			 "scsi_wq_%d", shost->host_no);
 		shost->work_q = create_singlethread_workqueue(
 					shost->work_q_name);
 		if (!shost->work_q) {
diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 5fd64e70029d..a272b9a2c869 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -417,15 +417,16 @@ static int fc_host_setup(struct transport_container *tc, struct device *dev,
 	fc_host->next_vport_number = 0;
 	fc_host->npiv_vports_inuse = 0;
 
-	snprintf(fc_host->work_q_name, KOBJ_NAME_LEN, "fc_wq_%d",
-		shost->host_no);
+	snprintf(fc_host->work_q_name, sizeof(fc_host->work_q_name),
+		 "fc_wq_%d", shost->host_no);
 	fc_host->work_q = create_singlethread_workqueue(
 					fc_host->work_q_name);
 	if (!fc_host->work_q)
 		return -ENOMEM;
 
-	snprintf(fc_host->devloss_work_q_name, KOBJ_NAME_LEN, "fc_dl_%d",
-		shost->host_no);
+	snprintf(fc_host->devloss_work_q_name,
+		 sizeof(fc_host->devloss_work_q_name),
+		 "fc_dl_%d", shost->host_no);
 	fc_host->devloss_work_q = create_singlethread_workqueue(
 					fc_host->devloss_work_q_name);
 	if (!fc_host->devloss_work_q) {
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 06748f318cd5..043c3921164f 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -247,8 +247,8 @@ static int iscsi_setup_host(struct transport_container *tc, struct device *dev,
 	atomic_set(&ihost->nr_scans, 0);
 	mutex_init(&ihost->mutex);
 
-	snprintf(ihost->scan_workq_name, KOBJ_NAME_LEN, "iscsi_scan_%d",
-		shost->host_no);
+	snprintf(ihost->scan_workq_name, sizeof(ihost->scan_workq_name),
+		 "iscsi_scan_%d", shost->host_no);
 	ihost->scan_workq = create_singlethread_workqueue(
 						ihost->scan_workq_name);
 	if (!ihost->scan_workq)
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 6149e4b58c88..efef715135d3 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -401,7 +401,7 @@ void register_disk(struct gendisk *disk)
 	disk->dev.parent = disk->driverfs_dev;
 	disk->dev.devt = MKDEV(disk->major, disk->first_minor);
 
-	strlcpy(disk->dev.bus_id, disk->disk_name, KOBJ_NAME_LEN);
+	strlcpy(disk->dev.bus_id, disk->disk_name, BUS_ID_SIZE);
 	/* ewww... some of these buggers have / in the name... */
 	s = strchr(disk->dev.bus_id, '/');
 	if (s)
diff --git a/include/linux/device.h b/include/linux/device.h
index 0e1d24c2ed41..fba1bb0d1758 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -29,8 +29,7 @@
 /* DEVICE_NAME_HALF is really less than half to accommodate slop */
 #define DEVICE_NAME_HALF	__stringify(20)
 #define DEVICE_ID_SIZE		32
-#define BUS_ID_SIZE		KOBJ_NAME_LEN
-
+#define BUS_ID_SIZE		20
 
 struct device;
 struct device_driver;
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 39e709f88aa0..d542faa6cb47 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -26,7 +26,6 @@
 #include <linux/wait.h>
 #include <asm/atomic.h>
 
-#define KOBJ_NAME_LEN			20
 #define UEVENT_HELPER_PATH_LEN		256
 #define UEVENT_NUM_ENVP			32	/* number of env pointers */
 #define UEVENT_BUFFER_SIZE		2048	/* buffer for the variables */
diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index 387e428f1cdf..b9a76c972084 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -733,7 +733,7 @@ struct spi_board_info {
 	 * controller_data goes to spi_device.controller_data,
 	 * irq is copied too
 	 */
-	char		modalias[KOBJ_NAME_LEN];
+	char		modalias[32];
 	const void	*platform_data;
 	void		*controller_data;
 	int		irq;
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index 1834fdfe82a7..a594bac4a77d 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -623,7 +623,7 @@ struct Scsi_Host {
 	/*
 	 * Optional work queue to be utilized by the transport
 	 */
-	char work_q_name[KOBJ_NAME_LEN];
+	char work_q_name[20];
 	struct workqueue_struct *work_q;
 
 	/*
diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h
index 06f72bab9df0..878373c32ef7 100644
--- a/include/scsi/scsi_transport_fc.h
+++ b/include/scsi/scsi_transport_fc.h
@@ -489,9 +489,9 @@ struct fc_host_attrs {
 	u16 npiv_vports_inuse;
 
 	/* work queues for rport state manipulation */
-	char work_q_name[KOBJ_NAME_LEN];
+	char work_q_name[20];
 	struct workqueue_struct *work_q;
-	char devloss_work_q_name[KOBJ_NAME_LEN];
+	char devloss_work_q_name[20];
 	struct workqueue_struct *devloss_work_q;
 };
 
diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h
index f5444e033cc9..8b6c91df4c7a 100644
--- a/include/scsi/scsi_transport_iscsi.h
+++ b/include/scsi/scsi_transport_iscsi.h
@@ -198,7 +198,7 @@ struct iscsi_cls_host {
 	atomic_t nr_scans;
 	struct mutex mutex;
 	struct workqueue_struct *scan_workq;
-	char scan_workq_name[KOBJ_NAME_LEN];
+	char scan_workq_name[20];
 };
 
 extern void iscsi_host_for_each_session(struct Scsi_Host *shost,
-- 
cgit v1.2.3


From ca52a49846f1451163c0dc14c40cd06ff808df3e Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 2 May 2008 06:02:41 +0200
Subject: driver core: remove DEVICE_NAME_SIZE define

There is no such thing as a "device name size" in the driver core, so
remove the define and fix up any users of this odd define in the rest of
the kernel.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/eisa/Makefile        | 2 +-
 drivers/eisa/eisa-bus.c      | 4 ++--
 drivers/mca/mca-bus.c        | 2 +-
 drivers/video/aty/aty128fb.c | 2 +-
 drivers/video/aty/radeonfb.h | 2 +-
 include/linux/device.h       | 3 ---
 include/linux/eisa.h         | 2 +-
 7 files changed, 7 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/eisa/Makefile b/drivers/eisa/Makefile
index 70abf93fe6b0..5369ce957c6d 100644
--- a/drivers/eisa/Makefile
+++ b/drivers/eisa/Makefile
@@ -9,7 +9,7 @@ obj-${CONFIG_EISA_VIRTUAL_ROOT} += virtual_root.o
 
 
 # Ugly hack to get DEVICE_NAME_SIZE value...
-DEVICE_NAME_SIZE =$(shell awk '$$1=="\#define" && $$2=="DEVICE_NAME_SIZE" {print $$3-1}' $(srctree)/include/linux/device.h)
+DEVICE_NAME_SIZE = 50
 
 $(obj)/eisa-bus.o: $(obj)/devlist.h
 
diff --git a/drivers/eisa/eisa-bus.c b/drivers/eisa/eisa-bus.c
index 65dcf0432653..c950bf8606d9 100644
--- a/drivers/eisa/eisa-bus.c
+++ b/drivers/eisa/eisa-bus.c
@@ -22,7 +22,7 @@
 
 struct eisa_device_info {
 	struct eisa_device_id id;
-	char name[DEVICE_NAME_SIZE];
+	char name[50];
 };
 
 #ifdef CONFIG_EISA_NAMES
@@ -63,7 +63,7 @@ static void __init eisa_name_device (struct eisa_device *edev)
 		if (!strcmp (edev->id.sig, eisa_table[i].id.sig)) {
 			strlcpy (edev->pretty_name,
 				 eisa_table[i].name,
-				 DEVICE_NAME_SIZE);
+				 sizeof(edev->pretty_name));
 			return;
 		}
 	}
diff --git a/drivers/mca/mca-bus.c b/drivers/mca/mca-bus.c
index 67b8e9453b19..ef2dbfe74714 100644
--- a/drivers/mca/mca-bus.c
+++ b/drivers/mca/mca-bus.c
@@ -40,7 +40,7 @@ static struct mca_bus *mca_root_busses[MAX_MCA_BUSSES];
 
 struct mca_device_info {
 	short pos_id;		/* the 2 byte pos id for this card */
-	char name[DEVICE_NAME_SIZE];
+	char name[50];
 };
 
 static int mca_bus_match (struct device *dev, struct device_driver *drv)
diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c
index 24ee96c4e9e9..07b6addbb3c1 100644
--- a/drivers/video/aty/aty128fb.c
+++ b/drivers/video/aty/aty128fb.c
@@ -1872,7 +1872,7 @@ static int __devinit aty128_init(struct pci_dev *pdev, const struct pci_device_i
 	struct fb_info *info = pci_get_drvdata(pdev);
 	struct aty128fb_par *par = info->par;
 	struct fb_var_screeninfo var;
-	char video_card[DEVICE_NAME_SIZE];
+	char video_card[50];
 	u8 chip_rev;
 	u32 dac;
 
diff --git a/drivers/video/aty/radeonfb.h b/drivers/video/aty/radeonfb.h
index c347e38cd0b0..ccbfffd12805 100644
--- a/drivers/video/aty/radeonfb.h
+++ b/drivers/video/aty/radeonfb.h
@@ -289,7 +289,7 @@ struct radeonfb_info {
 	struct radeon_regs 	state;
 	struct radeon_regs	init_state;
 
-	char			name[DEVICE_NAME_SIZE];
+	char			name[50];
 
 	unsigned long		mmio_base_phys;
 	unsigned long		fb_base_phys;
diff --git a/include/linux/device.h b/include/linux/device.h
index fba1bb0d1758..894fdb2e483a 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -25,9 +25,6 @@
 #include <asm/atomic.h>
 #include <asm/device.h>
 
-#define DEVICE_NAME_SIZE	50
-/* DEVICE_NAME_HALF is really less than half to accommodate slop */
-#define DEVICE_NAME_HALF	__stringify(20)
 #define DEVICE_ID_SIZE		32
 #define BUS_ID_SIZE		20
 
diff --git a/include/linux/eisa.h b/include/linux/eisa.h
index fe806b6f030d..e61c0be2a459 100644
--- a/include/linux/eisa.h
+++ b/include/linux/eisa.h
@@ -40,7 +40,7 @@ struct eisa_device {
 	u64                   dma_mask;
 	struct device         dev; /* generic device */
 #ifdef CONFIG_EISA_NAMES
-	char		      pretty_name[DEVICE_NAME_SIZE];
+	char		      pretty_name[50];
 #endif
 };
 
-- 
cgit v1.2.3


From b98cb4b7fe0e83238501b48489e46b3e0dce9aaf Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 2 May 2008 06:02:41 +0200
Subject: driver core: remove DEVICE_ID_SIZE define

There is no such thing as a "device id size" in the driver core, so
remove the define and fix up any users of this odd define in the rest of
the kernel.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/memstick/host/jmb38x_ms.c | 4 ++--
 include/linux/device.h            | 1 -
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/memstick/host/jmb38x_ms.c b/drivers/memstick/host/jmb38x_ms.c
index a054668eda16..4e3bfbcdf155 100644
--- a/drivers/memstick/host/jmb38x_ms.c
+++ b/drivers/memstick/host/jmb38x_ms.c
@@ -51,7 +51,7 @@ struct jmb38x_ms_host {
 	void __iomem            *addr;
 	spinlock_t              lock;
 	int                     id;
-	char                    host_id[DEVICE_ID_SIZE];
+	char                    host_id[32];
 	int                     irq;
 	unsigned int            block_pos;
 	unsigned long           timeout_jiffies;
@@ -781,7 +781,7 @@ static struct memstick_host *jmb38x_ms_alloc_host(struct jmb38x_ms *jm, int cnt)
 
 	spin_lock_init(&host->lock);
 	host->id = cnt;
-	snprintf(host->host_id, DEVICE_ID_SIZE, DRIVER_NAME ":slot%d",
+	snprintf(host->host_id, sizeof(host->host_id), DRIVER_NAME ":slot%d",
 		 host->id);
 	host->irq = jm->pdev->irq;
 	host->timeout_jiffies = msecs_to_jiffies(1000);
diff --git a/include/linux/device.h b/include/linux/device.h
index 894fdb2e483a..d24a47f80f9c 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -25,7 +25,6 @@
 #include <asm/atomic.h>
 #include <asm/device.h>
 
-#define DEVICE_ID_SIZE		32
 #define BUS_ID_SIZE		20
 
 struct device;
-- 
cgit v1.2.3


From 328a14e70e7f46997cb50d4258dd93d5377f98c6 Mon Sep 17 00:00:00 2001
From: "Hans J. Koch" <hjk@linutronix.de>
Date: Fri, 23 May 2008 13:50:14 +0200
Subject: UIO: Add write function to allow irq masking
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Sometimes it is necessary to enable/disable the interrupt of a UIO device
from the userspace part of the driver. With this patch, the UIO kernel driver
can implement an "irqcontrol()" function that does this. Userspace can write
an s32 value to /dev/uioX (usually 0 or 1 to turn the irq off or on). The
UIO core will then call the driver's irqcontrol function.

Signed-off-by: Hans J. Koch <hjk@linutronix.de>
Acked-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Acked-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/DocBook/uio-howto.tmpl | 40 +++++++++++++++++++++++++++++++++++-
 drivers/uio/uio.c                    | 26 +++++++++++++++++++++++
 include/linux/uio_driver.h           |  2 ++
 3 files changed, 67 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/Documentation/DocBook/uio-howto.tmpl b/Documentation/DocBook/uio-howto.tmpl
index fdd7f4f887b7..c4d187313963 100644
--- a/Documentation/DocBook/uio-howto.tmpl
+++ b/Documentation/DocBook/uio-howto.tmpl
@@ -29,6 +29,12 @@
 </abstract>
 
 <revhistory>
+	<revision>
+	<revnumber>0.5</revnumber>
+	<date>2008-05-22</date>
+	<authorinitials>hjk</authorinitials>
+	<revremark>Added description of write() function.</revremark>
+	</revision>
 	<revision>
 	<revnumber>0.4</revnumber>
 	<date>2007-11-26</date>
@@ -64,7 +70,7 @@
 <?dbhtml filename="copyright.html"?>
 <title>Copyright and License</title>
 <para>
-      Copyright (c) 2006 by Hans-Jürgen Koch.</para>
+      Copyright (c) 2006-2008 by Hans-Jürgen Koch.</para>
 <para>
 This documentation is Free Software licensed under the terms of the
 GPL version 2.
@@ -189,6 +195,30 @@ interested in translating it, please email me
 	represents the total interrupt count. You can use this number
 	to figure out if you missed some interrupts.
 	</para>
+	<para>
+	For some hardware that has more than one interrupt source internally,
+	but not separate IRQ mask and status registers, there might be
+	situations where userspace cannot determine what the interrupt source
+	was if the kernel handler disables them by writing to the chip's IRQ
+	register. In such a case, the kernel has to disable the IRQ completely
+	to leave the chip's register untouched. Now the userspace part can
+	determine the cause of the interrupt, but it cannot re-enable
+	interrupts. Another cornercase is chips where re-enabling interrupts
+	is a read-modify-write operation to a combined IRQ status/acknowledge
+	register. This would be racy if a new interrupt occurred
+	simultaneously.
+	</para>
+	<para>
+	To address these problems, UIO also implements a write() function. It
+	is normally not used and can be ignored for hardware that has only a
+	single interrupt source or has separate IRQ mask and status registers.
+	If you need it, however, a write to <filename>/dev/uioX</filename>
+	will call the <function>irqcontrol()</function> function implemented
+	by the driver. You have to write a 32-bit value that is usually either
+	0 or 1 to disable or enable interrupts. If a driver does not implement
+	<function>irqcontrol()</function>, <function>write()</function> will
+	return with <varname>-ENOSYS</varname>.
+	</para>
 
 	<para>
 	To handle interrupts properly, your custom kernel module can
@@ -362,6 +392,14 @@ device is actually used.
 <function>open()</function>, you will probably also want a custom
 <function>release()</function> function.
 </para></listitem>
+
+<listitem><para>
+<varname>int (*irqcontrol)(struct uio_info *info, s32 irq_on)
+</varname>: Optional. If you need to be able to enable or disable
+interrupts from userspace by writing to <filename>/dev/uioX</filename>,
+you can implement this function. The parameter <varname>irq_on</varname>
+will be 0 to disable interrupts and 1 to enable them.
+</para></listitem>
 </itemizedlist>
 
 <para>
diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index 5a7ca2e6094d..3a6934bf7131 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -427,6 +427,31 @@ static ssize_t uio_read(struct file *filep, char __user *buf,
 	return retval;
 }
 
+static ssize_t uio_write(struct file *filep, const char __user *buf,
+			size_t count, loff_t *ppos)
+{
+	struct uio_listener *listener = filep->private_data;
+	struct uio_device *idev = listener->dev;
+	ssize_t retval;
+	s32 irq_on;
+
+	if (idev->info->irq == UIO_IRQ_NONE)
+		return -EIO;
+
+	if (count != sizeof(s32))
+		return -EINVAL;
+
+	if (!idev->info->irqcontrol)
+		return -ENOSYS;
+
+	if (copy_from_user(&irq_on, buf, count))
+		return -EFAULT;
+
+	retval = idev->info->irqcontrol(idev->info, irq_on);
+
+	return retval ? retval : sizeof(s32);
+}
+
 static int uio_find_mem_index(struct vm_area_struct *vma)
 {
 	int mi;
@@ -546,6 +571,7 @@ static const struct file_operations uio_fops = {
 	.open		= uio_open,
 	.release	= uio_release,
 	.read		= uio_read,
+	.write		= uio_write,
 	.mmap		= uio_mmap,
 	.poll		= uio_poll,
 	.fasync		= uio_fasync,
diff --git a/include/linux/uio_driver.h b/include/linux/uio_driver.h
index 973386d439da..cf65e964102b 100644
--- a/include/linux/uio_driver.h
+++ b/include/linux/uio_driver.h
@@ -53,6 +53,7 @@ struct uio_device;
  * @mmap:		mmap operation for this uio device
  * @open:		open operation for this uio device
  * @release:		release operation for this uio device
+ * @irqcontrol:		disable/enable irqs when 0/1 is written to /dev/uioX
  */
 struct uio_info {
 	struct uio_device	*uio_dev;
@@ -66,6 +67,7 @@ struct uio_info {
 	int (*mmap)(struct uio_info *info, struct vm_area_struct *vma);
 	int (*open)(struct uio_info *info, struct inode *inode);
 	int (*release)(struct uio_info *info, struct inode *inode);
+	int (*irqcontrol)(struct uio_info *info, s32 irq_on);
 };
 
 extern int __must_check
-- 
cgit v1.2.3


From 6d8333c24d41637f0f847f6e17032189dae08c02 Mon Sep 17 00:00:00 2001
From: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Date: Tue, 10 Jun 2008 09:14:48 +0200
Subject: UIO: minor style and comment fixes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Signed-off-by: Hans J. Koch <hjk@linutronix.de>
---
 include/linux/uio_driver.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/uio_driver.h b/include/linux/uio_driver.h
index cf65e964102b..cdf338d94b7f 100644
--- a/include/linux/uio_driver.h
+++ b/include/linux/uio_driver.h
@@ -36,7 +36,7 @@ struct uio_mem {
 	struct uio_map		*map;
 };
 
-#define MAX_UIO_MAPS 	5
+#define MAX_UIO_MAPS	5
 
 struct uio_device;
 
@@ -82,11 +82,11 @@ static inline int __must_check
 extern void uio_unregister_device(struct uio_info *info);
 extern void uio_event_notify(struct uio_info *info);
 
-/* defines for uio_device->irq */
+/* defines for uio_info->irq */
 #define UIO_IRQ_CUSTOM	-1
 #define UIO_IRQ_NONE	-2
 
-/* defines for uio_device->memtype */
+/* defines for uio_mem->memtype */
 #define UIO_MEM_NONE	0
 #define UIO_MEM_PHYS	1
 #define UIO_MEM_LOGICAL	2
-- 
cgit v1.2.3


From a231934bdf086a4fefc0df06e669499125a9db6f Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Mon, 2 Jun 2008 11:07:25 +0100
Subject: kobject: reorder kobject to save space on 64 bit builds

reorder kobject to save space on 64 bit builds.
shrinks from 72 to 64 bytes & moves allocated kobject to a smaller
slab.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/kobject.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index d542faa6cb47..60f0d418ae32 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -58,12 +58,12 @@ enum kobject_action {
 
 struct kobject {
 	const char		*name;
-	struct kref		kref;
 	struct list_head	entry;
 	struct kobject		*parent;
 	struct kset		*kset;
 	struct kobj_type	*ktype;
 	struct sysfs_dirent	*sd;
+	struct kref		kref;
 	unsigned int state_initialized:1;
 	unsigned int state_in_sysfs:1;
 	unsigned int state_add_uevent_sent:1;
-- 
cgit v1.2.3


From 9505e6375640fc61d92d36c8e9f25a6a218f3f57 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Tue, 1 Jul 2008 15:14:51 +0200
Subject: debugfs: Implement debugfs_remove_recursive()

debugfs_remove_recursive() will remove a dentry and all its children.
Drivers can use this to zap their whole debugfs tree so that they don't
need to keep track of every single debugfs dentry they created.

It may fail to remove the whole tree in certain cases:

sh-3.2# rmmod atmel-mci < /sys/kernel/debug/mmc0/ios/clock
mmc0: card b368 removed
atmel_mci atmel_mci.0: Lost dma0chan1, falling back to PIO
sh-3.2# ls /sys/kernel/debug/mmc0/
ios

But I'm not sure if that case can be handled in any sane manner.

Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Cc: Pierre Ossman <drzeus-list@drzeus.cx>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/debugfs/inode.c      | 114 ++++++++++++++++++++++++++++++++++++++++--------
 include/linux/debugfs.h |   4 ++
 2 files changed, 100 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index e9602d85c11d..08e28c9bb416 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -309,6 +309,31 @@ struct dentry *debugfs_create_symlink(const char *name, struct dentry *parent,
 }
 EXPORT_SYMBOL_GPL(debugfs_create_symlink);
 
+static void __debugfs_remove(struct dentry *dentry, struct dentry *parent)
+{
+	int ret = 0;
+
+	if (debugfs_positive(dentry)) {
+		if (dentry->d_inode) {
+			dget(dentry);
+			switch (dentry->d_inode->i_mode & S_IFMT) {
+			case S_IFDIR:
+				ret = simple_rmdir(parent->d_inode, dentry);
+				break;
+			case S_IFLNK:
+				kfree(dentry->d_inode->i_private);
+				/* fall through */
+			default:
+				simple_unlink(parent->d_inode, dentry);
+				break;
+			}
+			if (!ret)
+				d_delete(dentry);
+			dput(dentry);
+		}
+	}
+}
+
 /**
  * debugfs_remove - removes a file or directory from the debugfs filesystem
  * @dentry: a pointer to a the dentry of the file or directory to be
@@ -325,7 +350,6 @@ EXPORT_SYMBOL_GPL(debugfs_create_symlink);
 void debugfs_remove(struct dentry *dentry)
 {
 	struct dentry *parent;
-	int ret = 0;
 	
 	if (!dentry)
 		return;
@@ -335,29 +359,83 @@ void debugfs_remove(struct dentry *dentry)
 		return;
 
 	mutex_lock(&parent->d_inode->i_mutex);
-	if (debugfs_positive(dentry)) {
-		if (dentry->d_inode) {
-			dget(dentry);
-			switch (dentry->d_inode->i_mode & S_IFMT) {
-			case S_IFDIR:
-				ret = simple_rmdir(parent->d_inode, dentry);
-				break;
-			case S_IFLNK:
-				kfree(dentry->d_inode->i_private);
-				/* fall through */
-			default:
-				simple_unlink(parent->d_inode, dentry);
+	__debugfs_remove(dentry, parent);
+	mutex_unlock(&parent->d_inode->i_mutex);
+	simple_release_fs(&debugfs_mount, &debugfs_mount_count);
+}
+EXPORT_SYMBOL_GPL(debugfs_remove);
+
+/**
+ * debugfs_remove_recursive - recursively removes a directory
+ * @dentry: a pointer to a the dentry of the directory to be removed.
+ *
+ * This function recursively removes a directory tree in debugfs that
+ * was previously created with a call to another debugfs function
+ * (like debugfs_create_file() or variants thereof.)
+ *
+ * This function is required to be called in order for the file to be
+ * removed, no automatic cleanup of files will happen when a module is
+ * removed, you are responsible here.
+ */
+void debugfs_remove_recursive(struct dentry *dentry)
+{
+	struct dentry *child;
+	struct dentry *parent;
+
+	if (!dentry)
+		return;
+
+	parent = dentry->d_parent;
+	if (!parent || !parent->d_inode)
+		return;
+
+	parent = dentry;
+	mutex_lock(&parent->d_inode->i_mutex);
+
+	while (1) {
+		/*
+		 * When all dentries under "parent" has been removed,
+		 * walk up the tree until we reach our starting point.
+		 */
+		if (list_empty(&parent->d_subdirs)) {
+			mutex_unlock(&parent->d_inode->i_mutex);
+			if (parent == dentry)
 				break;
-			}
-			if (!ret)
-				d_delete(dentry);
-			dput(dentry);
+			parent = parent->d_parent;
+			mutex_lock(&parent->d_inode->i_mutex);
+		}
+		child = list_entry(parent->d_subdirs.next, struct dentry,
+				d_u.d_child);
+
+		/*
+		 * If "child" isn't empty, walk down the tree and
+		 * remove all its descendants first.
+		 */
+		if (!list_empty(&child->d_subdirs)) {
+			mutex_unlock(&parent->d_inode->i_mutex);
+			parent = child;
+			mutex_lock(&parent->d_inode->i_mutex);
+			continue;
 		}
+		__debugfs_remove(child, parent);
+		if (parent->d_subdirs.next == &child->d_u.d_child) {
+			/*
+			 * Avoid infinite loop if we fail to remove
+			 * one dentry.
+			 */
+			mutex_unlock(&parent->d_inode->i_mutex);
+			break;
+		}
+		simple_release_fs(&debugfs_mount, &debugfs_mount_count);
 	}
+
+	parent = dentry->d_parent;
+	mutex_lock(&parent->d_inode->i_mutex);
+	__debugfs_remove(dentry, parent);
 	mutex_unlock(&parent->d_inode->i_mutex);
 	simple_release_fs(&debugfs_mount, &debugfs_mount_count);
 }
-EXPORT_SYMBOL_GPL(debugfs_remove);
+EXPORT_SYMBOL_GPL(debugfs_remove_recursive);
 
 /**
  * debugfs_rename - rename a file/directory in the debugfs filesystem
diff --git a/include/linux/debugfs.h b/include/linux/debugfs.h
index 32755cdf68db..e1a6c046cea3 100644
--- a/include/linux/debugfs.h
+++ b/include/linux/debugfs.h
@@ -44,6 +44,7 @@ struct dentry *debugfs_create_symlink(const char *name, struct dentry *parent,
 				      const char *dest);
 
 void debugfs_remove(struct dentry *dentry);
+void debugfs_remove_recursive(struct dentry *dentry);
 
 struct dentry *debugfs_rename(struct dentry *old_dir, struct dentry *old_dentry,
                 struct dentry *new_dir, const char *new_name);
@@ -101,6 +102,9 @@ static inline struct dentry *debugfs_create_symlink(const char *name,
 static inline void debugfs_remove(struct dentry *dentry)
 { }
 
+static inline void debugfs_remove_recursive(struct dentry *dentry)
+{ }
+
 static inline struct dentry *debugfs_rename(struct dentry *old_dir, struct dentry *old_dentry,
                 struct dentry *new_dir, char *new_name)
 {
-- 
cgit v1.2.3


From 36ce6dad6e3cb3f050ed41e0beac0070d2062b25 Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Tue, 10 Jun 2008 11:09:08 +0200
Subject: driver core: Suppress sysfs warnings for device_rename().

driver core: Suppress sysfs warnings for device_rename().

Renaming network devices to an already existing name is not
something we want sysfs to print a scary warning for, since the
callers can deal with this correctly. So let's introduce
sysfs_create_link_nowarn() which gets rid of the common warning.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c   |  9 +++++----
 fs/sysfs/dir.c        | 37 +++++++++++++++++++++++++++++++++++--
 fs/sysfs/symlink.c    | 41 +++++++++++++++++++++++++++++++++--------
 fs/sysfs/sysfs.h      |  1 +
 include/linux/sysfs.h | 10 ++++++++++
 5 files changed, 84 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index c05b1159023e..7d5c63c81a59 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -1345,8 +1345,9 @@ int device_rename(struct device *dev, char *new_name)
 	if (old_class_name) {
 		new_class_name = make_class_name(dev->class->name, &dev->kobj);
 		if (new_class_name) {
-			error = sysfs_create_link(&dev->parent->kobj,
-						  &dev->kobj, new_class_name);
+			error = sysfs_create_link_nowarn(&dev->parent->kobj,
+							 &dev->kobj,
+							 new_class_name);
 			if (error)
 				goto out;
 			sysfs_remove_link(&dev->parent->kobj, old_class_name);
@@ -1354,8 +1355,8 @@ int device_rename(struct device *dev, char *new_name)
 	}
 #else
 	if (dev->class) {
-		error = sysfs_create_link(&dev->class->p->class_subsys.kobj,
-					  &dev->kobj, dev->bus_id);
+		error = sysfs_create_link_nowarn(&dev->class->p->class_subsys.kobj,
+						 &dev->kobj, dev->bus_id);
 		if (error)
 			goto out;
 		sysfs_remove_link(&dev->class->p->class_subsys.kobj,
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 8c0e4b92574f..c1a7efb310bf 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -398,7 +398,7 @@ void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt,
 }
 
 /**
- *	sysfs_add_one - add sysfs_dirent to parent
+ *	__sysfs_add_one - add sysfs_dirent to parent without warning
  *	@acxt: addrm context to use
  *	@sd: sysfs_dirent to be added
  *
@@ -417,7 +417,7 @@ void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt,
  *	0 on success, -EEXIST if entry with the given name already
  *	exists.
  */
-int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
+int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
 {
 	if (sysfs_find_dirent(acxt->parent_sd, sd->s_name))
 		return -EEXIST;
@@ -434,6 +434,39 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
 	return 0;
 }
 
+/**
+ *	sysfs_add_one - add sysfs_dirent to parent
+ *	@acxt: addrm context to use
+ *	@sd: sysfs_dirent to be added
+ *
+ *	Get @acxt->parent_sd and set sd->s_parent to it and increment
+ *	nlink of parent inode if @sd is a directory and link into the
+ *	children list of the parent.
+ *
+ *	This function should be called between calls to
+ *	sysfs_addrm_start() and sysfs_addrm_finish() and should be
+ *	passed the same @acxt as passed to sysfs_addrm_start().
+ *
+ *	LOCKING:
+ *	Determined by sysfs_addrm_start().
+ *
+ *	RETURNS:
+ *	0 on success, -EEXIST if entry with the given name already
+ *	exists.
+ */
+int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
+{
+	int ret;
+
+	ret = __sysfs_add_one(acxt, sd);
+	if (ret == -EEXIST) {
+		printk(KERN_WARNING "sysfs: duplicate filename '%s' "
+		       "can not be created\n", sd->s_name);
+		WARN_ON(1);
+	}
+	return ret;
+}
+
 /**
  *	sysfs_remove_one - remove sysfs_dirent from parent
  *	@acxt: addrm context to use
diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c
index 817f5966edca..a3ba217fbe74 100644
--- a/fs/sysfs/symlink.c
+++ b/fs/sysfs/symlink.c
@@ -19,13 +19,8 @@
 
 #include "sysfs.h"
 
-/**
- *	sysfs_create_link - create symlink between two objects.
- *	@kobj:	object whose directory we're creating the link in.
- *	@target:	object we're pointing to.
- *	@name:		name of the symlink.
- */
-int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name)
+static int sysfs_do_create_link(struct kobject *kobj, struct kobject *target,
+				const char *name, int warn)
 {
 	struct sysfs_dirent *parent_sd = NULL;
 	struct sysfs_dirent *target_sd = NULL;
@@ -65,7 +60,10 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char
 	target_sd = NULL;	/* reference is now owned by the symlink */
 
 	sysfs_addrm_start(&acxt, parent_sd);
-	error = sysfs_add_one(&acxt, sd);
+	if (warn)
+		error = sysfs_add_one(&acxt, sd);
+	else
+		error = __sysfs_add_one(&acxt, sd);
 	sysfs_addrm_finish(&acxt);
 
 	if (error)
@@ -79,6 +77,33 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char
 	return error;
 }
 
+/**
+ *	sysfs_create_link - create symlink between two objects.
+ *	@kobj:	object whose directory we're creating the link in.
+ *	@target:	object we're pointing to.
+ *	@name:		name of the symlink.
+ */
+int sysfs_create_link(struct kobject *kobj, struct kobject *target,
+		      const char *name)
+{
+	return sysfs_do_create_link(kobj, target, name, 1);
+}
+
+/**
+ *	sysfs_create_link_nowarn - create symlink between two objects.
+ *	@kobj:	object whose directory we're creating the link in.
+ *	@target:	object we're pointing to.
+ *	@name:		name of the symlink.
+ *
+ *	This function does the same as sysf_create_link(), but it
+ *	doesn't warn if the link already exists.
+ */
+int sysfs_create_link_nowarn(struct kobject *kobj, struct kobject *target,
+			     const char *name)
+{
+	return sysfs_do_create_link(kobj, target, name, 0);
+}
+
 /**
  *	sysfs_remove_link - remove symlink in object's directory.
  *	@kobj:	object we're acting for.
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index ce4e15f8aaeb..a5db496f71c7 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -107,6 +107,7 @@ struct sysfs_dirent *sysfs_get_active_two(struct sysfs_dirent *sd);
 void sysfs_put_active_two(struct sysfs_dirent *sd);
 void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt,
 		       struct sysfs_dirent *parent_sd);
+int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd);
 int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd);
 void sysfs_remove_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd);
 void sysfs_addrm_finish(struct sysfs_addrm_cxt *acxt);
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 7858eac40aa7..37fa24152bd8 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -101,6 +101,9 @@ void sysfs_remove_bin_file(struct kobject *kobj, struct bin_attribute *attr);
 
 int __must_check sysfs_create_link(struct kobject *kobj, struct kobject *target,
 				   const char *name);
+int __must_check sysfs_create_link_nowarn(struct kobject *kobj,
+					  struct kobject *target,
+					  const char *name);
 void sysfs_remove_link(struct kobject *kobj, const char *name);
 
 int __must_check sysfs_create_group(struct kobject *kobj,
@@ -180,6 +183,13 @@ static inline int sysfs_create_link(struct kobject *kobj,
 	return 0;
 }
 
+static inline int sysfs_create_link_nowarn(struct kobject *kobj,
+					   struct kobject *target,
+					   const char *name)
+{
+	return 0;
+}
+
 static inline void sysfs_remove_link(struct kobject *kobj, const char *name)
 {
 }
-- 
cgit v1.2.3


From 4a0b2b4dbe1335b8b9886ba3dc85a145d5d938ed Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Tue, 1 Jul 2008 18:48:41 +0200
Subject: sysdev: Pass the attribute to the low level sysdev show/store
 function

This allow to dynamically generate attributes and share show/store
functions between attributes. Right now most attributes are generated
by special macros and lots of duplicated code. With the attribute
passed it's instead possible to attach some data to the attribute
and then use that in shared low level functions to do different things.

I need this for the dynamically generated bank attributes in the x86
machine check code, but it'll allow some further cleanups.

I converted all users in tree to the new show/store prototype. It's a single
huge patch to avoid unbisectable sections.

Runtime tested: x86-32, x86-64
Compiled only: ia64, powerpc
Not compile tested/only grep converted: sh, arm, avr32

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/kernel/time.c                    |  4 ++-
 arch/avr32/kernel/cpu.c                   | 38 +++++++++++++++++---------
 arch/ia64/kernel/err_inject.c             | 22 ++++++++++-----
 arch/powerpc/kernel/sysfs.c               | 15 ++++++++---
 arch/powerpc/platforms/cell/cbe_thermal.c | 45 ++++++++++++++++++++-----------
 arch/powerpc/platforms/cell/spu_base.c    |  3 ++-
 arch/s390/kernel/smp.c                    | 36 ++++++++++++++++---------
 arch/s390/kernel/time.c                   | 35 ++++++++++++++++--------
 arch/sh/drivers/dma/dma-sysfs.c           | 15 ++++++++---
 arch/sparc64/kernel/sysfs.c               | 16 +++++++----
 arch/x86/kernel/cpu/mcheck/mce_64.c       | 14 +++++++---
 arch/x86/kernel/cpu/mcheck/therm_throt.c  |  1 +
 arch/x86/kernel/microcode.c               | 10 ++++---
 drivers/base/cpu.c                        | 10 ++++---
 drivers/base/memory.c                     | 12 ++++++---
 drivers/base/node.c                       | 15 +++++++----
 drivers/base/sys.c                        |  4 +--
 drivers/base/topology.c                   | 17 ++++++++----
 drivers/cpuidle/sysfs.c                   | 10 ++++---
 drivers/xen/balloon.c                     |  1 +
 include/linux/sysdev.h                    |  5 ++--
 kernel/rtmutex-tester.c                   |  7 ++---
 kernel/sched.c                            |  8 ++++--
 kernel/time/clocksource.c                 |  8 ++++--
 24 files changed, 239 insertions(+), 112 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index cc5145b28e7f..368d171754cf 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -130,7 +130,9 @@ static const struct leds_evt_name evt_names[] = {
 	{ "red",   led_red_on,   led_red_off   },
 };
 
-static ssize_t leds_store(struct sys_device *dev, const char *buf, size_t size)
+static ssize_t leds_store(struct sys_device *dev,
+			struct sysdev_attribute *attr,
+			const char *buf, size_t size)
 {
 	int ret = -EINVAL, len = strcspn(buf, " ");
 
diff --git a/arch/avr32/kernel/cpu.c b/arch/avr32/kernel/cpu.c
index b8409caeb23d..e84faffbbeca 100644
--- a/arch/avr32/kernel/cpu.c
+++ b/arch/avr32/kernel/cpu.c
@@ -26,14 +26,16 @@ static DEFINE_PER_CPU(struct cpu, cpu_devices);
  * XXX: If/when a SMP-capable implementation of AVR32 will ever be
  * made, we must make sure that the code executes on the correct CPU.
  */
-static ssize_t show_pc0event(struct sys_device *dev, char *buf)
+static ssize_t show_pc0event(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	unsigned long pccr;
 
 	pccr = sysreg_read(PCCR);
 	return sprintf(buf, "0x%lx\n", (pccr >> 12) & 0x3f);
 }
-static ssize_t store_pc0event(struct sys_device *dev, const char *buf,
+static ssize_t store_pc0event(struct sys_device *dev,
+			struct sysdev_attribute *attr, const char *buf,
 			      size_t count)
 {
 	unsigned long val;
@@ -46,15 +48,17 @@ static ssize_t store_pc0event(struct sys_device *dev, const char *buf,
 	sysreg_write(PCCR, val);
 	return count;
 }
-static ssize_t show_pc0count(struct sys_device *dev, char *buf)
+static ssize_t show_pc0count(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	unsigned long pcnt0;
 
 	pcnt0 = sysreg_read(PCNT0);
 	return sprintf(buf, "%lu\n", pcnt0);
 }
-static ssize_t store_pc0count(struct sys_device *dev, const char *buf,
-			      size_t count)
+static ssize_t store_pc0count(struct sys_device *dev,
+				struct sysdev_attribute *attr,
+				const char *buf, size_t count)
 {
 	unsigned long val;
 	char *endp;
@@ -67,14 +71,16 @@ static ssize_t store_pc0count(struct sys_device *dev, const char *buf,
 	return count;
 }
 
-static ssize_t show_pc1event(struct sys_device *dev, char *buf)
+static ssize_t show_pc1event(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	unsigned long pccr;
 
 	pccr = sysreg_read(PCCR);
 	return sprintf(buf, "0x%lx\n", (pccr >> 18) & 0x3f);
 }
-static ssize_t store_pc1event(struct sys_device *dev, const char *buf,
+static ssize_t store_pc1event(struct sys_device *dev,
+			      struct sysdev_attribute *attr, const char *buf,
 			      size_t count)
 {
 	unsigned long val;
@@ -87,14 +93,16 @@ static ssize_t store_pc1event(struct sys_device *dev, const char *buf,
 	sysreg_write(PCCR, val);
 	return count;
 }
-static ssize_t show_pc1count(struct sys_device *dev, char *buf)
+static ssize_t show_pc1count(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	unsigned long pcnt1;
 
 	pcnt1 = sysreg_read(PCNT1);
 	return sprintf(buf, "%lu\n", pcnt1);
 }
-static ssize_t store_pc1count(struct sys_device *dev, const char *buf,
+static ssize_t store_pc1count(struct sys_device *dev,
+				struct sysdev_attribute *attr, const char *buf,
 			      size_t count)
 {
 	unsigned long val;
@@ -108,14 +116,16 @@ static ssize_t store_pc1count(struct sys_device *dev, const char *buf,
 	return count;
 }
 
-static ssize_t show_pccycles(struct sys_device *dev, char *buf)
+static ssize_t show_pccycles(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	unsigned long pccnt;
 
 	pccnt = sysreg_read(PCCNT);
 	return sprintf(buf, "%lu\n", pccnt);
 }
-static ssize_t store_pccycles(struct sys_device *dev, const char *buf,
+static ssize_t store_pccycles(struct sys_device *dev,
+				struct sysdev_attribute *attr, const char *buf,
 			      size_t count)
 {
 	unsigned long val;
@@ -129,14 +139,16 @@ static ssize_t store_pccycles(struct sys_device *dev, const char *buf,
 	return count;
 }
 
-static ssize_t show_pcenable(struct sys_device *dev, char *buf)
+static ssize_t show_pcenable(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	unsigned long pccr;
 
 	pccr = sysreg_read(PCCR);
 	return sprintf(buf, "%c\n", (pccr & 1)?'1':'0');
 }
-static ssize_t store_pcenable(struct sys_device *dev, const char *buf,
+static ssize_t store_pcenable(struct sys_device *dev,
+			      struct sysdev_attribute *attr, const char *buf,
 			      size_t count)
 {
 	unsigned long pccr, val;
diff --git a/arch/ia64/kernel/err_inject.c b/arch/ia64/kernel/err_inject.c
index b642648cc2ac..c539c689493b 100644
--- a/arch/ia64/kernel/err_inject.c
+++ b/arch/ia64/kernel/err_inject.c
@@ -55,7 +55,8 @@ static u64 resources[NR_CPUS];
 
 #define show(name) 							\
 static ssize_t 								\
-show_##name(struct sys_device *dev, char *buf)				\
+show_##name(struct sys_device *dev, struct sysdev_attribute *attr,	\
+		char *buf)						\
 {									\
 	u32 cpu=dev->id;						\
 	return sprintf(buf, "%lx\n", name[cpu]);			\
@@ -63,7 +64,8 @@ show_##name(struct sys_device *dev, char *buf)				\
 
 #define store(name)							\
 static ssize_t 								\
-store_##name(struct sys_device *dev, const char *buf, size_t size)	\
+store_##name(struct sys_device *dev, struct sysdev_attribute *attr,	\
+					const char *buf, size_t size)	\
 {									\
 	unsigned int cpu=dev->id;					\
 	name[cpu] = simple_strtoull(buf, NULL, 16);			\
@@ -76,7 +78,8 @@ show(call_start)
  * processor. The cpu number in driver is only used for storing data.
  */
 static ssize_t
-store_call_start(struct sys_device *dev, const char *buf, size_t size)
+store_call_start(struct sys_device *dev, struct sysdev_attribute *attr,
+		const char *buf, size_t size)
 {
 	unsigned int cpu=dev->id;
 	unsigned long call_start = simple_strtoull(buf, NULL, 16);
@@ -124,14 +127,16 @@ show(err_type_info)
 store(err_type_info)
 
 static ssize_t
-show_virtual_to_phys(struct sys_device *dev, char *buf)
+show_virtual_to_phys(struct sys_device *dev, struct sysdev_attribute *attr,
+			char *buf)
 {
 	unsigned int cpu=dev->id;
 	return sprintf(buf, "%lx\n", phys_addr[cpu]);
 }
 
 static ssize_t
-store_virtual_to_phys(struct sys_device *dev, const char *buf, size_t size)
+store_virtual_to_phys(struct sys_device *dev, struct sysdev_attribute *attr,
+			const char *buf, size_t size)
 {
 	unsigned int cpu=dev->id;
 	u64 virt_addr=simple_strtoull(buf, NULL, 16);
@@ -154,7 +159,8 @@ show(err_struct_info)
 store(err_struct_info)
 
 static ssize_t
-show_err_data_buffer(struct sys_device *dev, char *buf)
+show_err_data_buffer(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	unsigned int cpu=dev->id;
 
@@ -165,7 +171,9 @@ show_err_data_buffer(struct sys_device *dev, char *buf)
 }
 
 static ssize_t
-store_err_data_buffer(struct sys_device *dev, const char *buf, size_t size)
+store_err_data_buffer(struct sys_device *dev,
+			struct sysdev_attribute *attr,
+			const char *buf, size_t size)
 {
 	unsigned int cpu=dev->id;
 	int ret;
diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c
index c8127f832df0..aba0ba95f062 100644
--- a/arch/powerpc/kernel/sysfs.c
+++ b/arch/powerpc/kernel/sysfs.c
@@ -28,7 +28,9 @@ static DEFINE_PER_CPU(struct cpu, cpu_devices);
 /* Time in microseconds we delay before sleeping in the idle loop */
 DEFINE_PER_CPU(unsigned long, smt_snooze_delay) = { 100 };
 
-static ssize_t store_smt_snooze_delay(struct sys_device *dev, const char *buf,
+static ssize_t store_smt_snooze_delay(struct sys_device *dev,
+				      struct sysdev_attribute *attr,
+				      const char *buf,
 				      size_t count)
 {
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
@@ -44,7 +46,9 @@ static ssize_t store_smt_snooze_delay(struct sys_device *dev, const char *buf,
 	return count;
 }
 
-static ssize_t show_smt_snooze_delay(struct sys_device *dev, char *buf)
+static ssize_t show_smt_snooze_delay(struct sys_device *dev,
+				     struct sysdev_attribute *attr,
+				     char *buf)
 {
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
 
@@ -152,14 +156,17 @@ static unsigned long write_##NAME(unsigned long val) \
 	mtspr(ADDRESS, val); \
 	return 0; \
 } \
-static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+static ssize_t show_##NAME(struct sys_device *dev, \
+			struct sysdev_attribute *attr, \
+			char *buf) \
 { \
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev); \
 	unsigned long val = run_on_cpu(cpu->sysdev.id, read_##NAME, 0); \
 	return sprintf(buf, "%lx\n", val); \
 } \
 static ssize_t __used \
-	store_##NAME(struct sys_device *dev, const char *buf, size_t count) \
+	store_##NAME(struct sys_device *dev, struct sysdev_attribute *attr, \
+			const char *buf, size_t count) \
 { \
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev); \
 	unsigned long val; \
diff --git a/arch/powerpc/platforms/cell/cbe_thermal.c b/arch/powerpc/platforms/cell/cbe_thermal.c
index 4852bf312d83..4d4c8c169124 100644
--- a/arch/powerpc/platforms/cell/cbe_thermal.c
+++ b/arch/powerpc/platforms/cell/cbe_thermal.c
@@ -97,7 +97,8 @@ static u8 spu_read_register_value(struct sys_device *sysdev, union spe_reg __iom
 	return value.spe[spu->spe_id];
 }
 
-static ssize_t spu_show_temp(struct sys_device *sysdev, char *buf)
+static ssize_t spu_show_temp(struct sys_device *sysdev, struct sysdev_attribute *attr,
+			char *buf)
 {
 	u8 value;
 	struct cbe_pmd_regs __iomem *pmd_regs;
@@ -146,32 +147,38 @@ static ssize_t store_throttle(struct cbe_pmd_regs __iomem *pmd_regs, const char
 	return size;
 }
 
-static ssize_t spu_show_throttle_end(struct sys_device *sysdev, char *buf)
+static ssize_t spu_show_throttle_end(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return show_throttle(get_pmd_regs(sysdev), buf, 0);
 }
 
-static ssize_t spu_show_throttle_begin(struct sys_device *sysdev, char *buf)
+static ssize_t spu_show_throttle_begin(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return show_throttle(get_pmd_regs(sysdev), buf, 8);
 }
 
-static ssize_t spu_show_throttle_full_stop(struct sys_device *sysdev, char *buf)
+static ssize_t spu_show_throttle_full_stop(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return show_throttle(get_pmd_regs(sysdev), buf, 16);
 }
 
-static ssize_t spu_store_throttle_end(struct sys_device *sysdev, const char *buf, size_t size)
+static ssize_t spu_store_throttle_end(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, const char *buf, size_t size)
 {
 	return store_throttle(get_pmd_regs(sysdev), buf, size, 0);
 }
 
-static ssize_t spu_store_throttle_begin(struct sys_device *sysdev, const char *buf, size_t size)
+static ssize_t spu_store_throttle_begin(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, const char *buf, size_t size)
 {
 	return store_throttle(get_pmd_regs(sysdev), buf, size, 8);
 }
 
-static ssize_t spu_store_throttle_full_stop(struct sys_device *sysdev, const char *buf, size_t size)
+static ssize_t spu_store_throttle_full_stop(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, const char *buf, size_t size)
 {
 	return store_throttle(get_pmd_regs(sysdev), buf, size, 16);
 }
@@ -192,43 +199,51 @@ static ssize_t ppe_show_temp(struct sys_device *sysdev, char *buf, int pos)
 
 /* shows the temperature of the DTS on the PPE,
  * located near the linear thermal sensor */
-static ssize_t ppe_show_temp0(struct sys_device *sysdev, char *buf)
+static ssize_t ppe_show_temp0(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return ppe_show_temp(sysdev, buf, 32);
 }
 
 /* shows the temperature of the second DTS on the PPE */
-static ssize_t ppe_show_temp1(struct sys_device *sysdev, char *buf)
+static ssize_t ppe_show_temp1(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return ppe_show_temp(sysdev, buf, 0);
 }
 
-static ssize_t ppe_show_throttle_end(struct sys_device *sysdev, char *buf)
+static ssize_t ppe_show_throttle_end(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return show_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, 32);
 }
 
-static ssize_t ppe_show_throttle_begin(struct sys_device *sysdev, char *buf)
+static ssize_t ppe_show_throttle_begin(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return show_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, 40);
 }
 
-static ssize_t ppe_show_throttle_full_stop(struct sys_device *sysdev, char *buf)
+static ssize_t ppe_show_throttle_full_stop(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	return show_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, 48);
 }
 
-static ssize_t ppe_store_throttle_end(struct sys_device *sysdev, const char *buf, size_t size)
+static ssize_t ppe_store_throttle_end(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, const char *buf, size_t size)
 {
 	return store_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, size, 32);
 }
 
-static ssize_t ppe_store_throttle_begin(struct sys_device *sysdev, const char *buf, size_t size)
+static ssize_t ppe_store_throttle_begin(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, const char *buf, size_t size)
 {
 	return store_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, size, 40);
 }
 
-static ssize_t ppe_store_throttle_full_stop(struct sys_device *sysdev, const char *buf, size_t size)
+static ssize_t ppe_store_throttle_full_stop(struct sys_device *sysdev,
+			struct sysdev_attribute *attr, const char *buf, size_t size)
 {
 	return store_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, size, 48);
 }
diff --git a/arch/powerpc/platforms/cell/spu_base.c b/arch/powerpc/platforms/cell/spu_base.c
index 78f905bc6a42..a5bdb89a17c3 100644
--- a/arch/powerpc/platforms/cell/spu_base.c
+++ b/arch/powerpc/platforms/cell/spu_base.c
@@ -703,7 +703,8 @@ static unsigned long long spu_acct_time(struct spu *spu,
 }
 
 
-static ssize_t spu_stat_show(struct sys_device *sysdev, char *buf)
+static ssize_t spu_stat_show(struct sys_device *sysdev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct spu *spu = container_of(sysdev, struct spu, sysdev);
 
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index b6781030cfbd..b795b3e24afd 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -864,7 +864,8 @@ int setup_profiling_timer(unsigned int multiplier)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static ssize_t cpu_configure_show(struct sys_device *dev, char *buf)
+static ssize_t cpu_configure_show(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	ssize_t count;
 
@@ -874,8 +875,9 @@ static ssize_t cpu_configure_show(struct sys_device *dev, char *buf)
 	return count;
 }
 
-static ssize_t cpu_configure_store(struct sys_device *dev, const char *buf,
-				   size_t count)
+static ssize_t cpu_configure_store(struct sys_device *dev,
+				  struct sysdev_attribute *attr,
+				  const char *buf, size_t count)
 {
 	int cpu = dev->id;
 	int val, rc;
@@ -922,7 +924,8 @@ out:
 static SYSDEV_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store);
 #endif /* CONFIG_HOTPLUG_CPU */
 
-static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf)
+static ssize_t cpu_polarization_show(struct sys_device *dev,
+				     struct sysdev_attribute *attr, char *buf)
 {
 	int cpu = dev->id;
 	ssize_t count;
@@ -950,7 +953,8 @@ static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf)
 }
 static SYSDEV_ATTR(polarization, 0444, cpu_polarization_show, NULL);
 
-static ssize_t show_cpu_address(struct sys_device *dev, char *buf)
+static ssize_t show_cpu_address(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	return sprintf(buf, "%d\n", __cpu_logical_map[dev->id]);
 }
@@ -970,7 +974,8 @@ static struct attribute_group cpu_common_attr_group = {
 	.attrs = cpu_common_attrs,
 };
 
-static ssize_t show_capability(struct sys_device *dev, char *buf)
+static ssize_t show_capability(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	unsigned int capability;
 	int rc;
@@ -982,7 +987,8 @@ static ssize_t show_capability(struct sys_device *dev, char *buf)
 }
 static SYSDEV_ATTR(capability, 0444, show_capability, NULL);
 
-static ssize_t show_idle_count(struct sys_device *dev, char *buf)
+static ssize_t show_idle_count(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct s390_idle_data *idle;
 	unsigned long long idle_count;
@@ -995,7 +1001,8 @@ static ssize_t show_idle_count(struct sys_device *dev, char *buf)
 }
 static SYSDEV_ATTR(idle_count, 0444, show_idle_count, NULL);
 
-static ssize_t show_idle_time(struct sys_device *dev, char *buf)
+static ssize_t show_idle_time(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct s390_idle_data *idle;
 	unsigned long long new_time;
@@ -1112,7 +1119,9 @@ out:
 	return rc;
 }
 
-static ssize_t __ref rescan_store(struct sys_device *dev, const char *buf,
+static ssize_t __ref rescan_store(struct sys_device *dev,
+				  struct sysdev_attribute *attr,
+				  const char *buf,
 				  size_t count)
 {
 	int rc;
@@ -1123,7 +1132,9 @@ static ssize_t __ref rescan_store(struct sys_device *dev, const char *buf,
 static SYSDEV_ATTR(rescan, 0200, NULL, rescan_store);
 #endif /* CONFIG_HOTPLUG_CPU */
 
-static ssize_t dispatching_show(struct sys_device *dev, char *buf)
+static ssize_t dispatching_show(struct sys_device *dev,
+				struct sysdev_attribute *attr,
+				char *buf)
 {
 	ssize_t count;
 
@@ -1133,8 +1144,9 @@ static ssize_t dispatching_show(struct sys_device *dev, char *buf)
 	return count;
 }
 
-static ssize_t dispatching_store(struct sys_device *dev, const char *buf,
-				 size_t count)
+static ssize_t dispatching_store(struct sys_device *dev,
+				 struct sysdev_attribute *attr,
+				 const char *buf, size_t count)
 {
 	int val, rc;
 	char delim;
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index f2cede3947b2..ab70d9bd9261 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -1100,7 +1100,9 @@ static inline struct etr_aib *etr_aib_from_dev(struct sys_device *dev)
 		return etr_port1_online ? &etr_port1 : NULL;
 }
 
-static ssize_t etr_online_show(struct sys_device *dev, char *buf)
+static ssize_t etr_online_show(struct sys_device *dev,
+				struct sysdev_attribute *attr,
+				char *buf)
 {
 	unsigned int online;
 
@@ -1109,7 +1111,8 @@ static ssize_t etr_online_show(struct sys_device *dev, char *buf)
 }
 
 static ssize_t etr_online_store(struct sys_device *dev,
-			      const char *buf, size_t count)
+				struct sysdev_attribute *attr,
+				const char *buf, size_t count)
 {
 	unsigned int value;
 
@@ -1136,7 +1139,9 @@ static ssize_t etr_online_store(struct sys_device *dev,
 
 static SYSDEV_ATTR(online, 0600, etr_online_show, etr_online_store);
 
-static ssize_t etr_stepping_control_show(struct sys_device *dev, char *buf)
+static ssize_t etr_stepping_control_show(struct sys_device *dev,
+					struct sysdev_attribute *attr,
+					char *buf)
 {
 	return sprintf(buf, "%i\n", (dev == &etr_port0_dev) ?
 		       etr_eacr.e0 : etr_eacr.e1);
@@ -1144,7 +1149,8 @@ static ssize_t etr_stepping_control_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(stepping_control, 0400, etr_stepping_control_show, NULL);
 
-static ssize_t etr_mode_code_show(struct sys_device *dev, char *buf)
+static ssize_t etr_mode_code_show(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	if (!etr_port0_online && !etr_port1_online)
 		/* Status word is not uptodate if both ports are offline. */
@@ -1155,7 +1161,8 @@ static ssize_t etr_mode_code_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(state_code, 0400, etr_mode_code_show, NULL);
 
-static ssize_t etr_untuned_show(struct sys_device *dev, char *buf)
+static ssize_t etr_untuned_show(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct etr_aib *aib = etr_aib_from_dev(dev);
 
@@ -1166,7 +1173,8 @@ static ssize_t etr_untuned_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(untuned, 0400, etr_untuned_show, NULL);
 
-static ssize_t etr_network_id_show(struct sys_device *dev, char *buf)
+static ssize_t etr_network_id_show(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct etr_aib *aib = etr_aib_from_dev(dev);
 
@@ -1177,7 +1185,8 @@ static ssize_t etr_network_id_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(network, 0400, etr_network_id_show, NULL);
 
-static ssize_t etr_id_show(struct sys_device *dev, char *buf)
+static ssize_t etr_id_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct etr_aib *aib = etr_aib_from_dev(dev);
 
@@ -1188,7 +1197,8 @@ static ssize_t etr_id_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(id, 0400, etr_id_show, NULL);
 
-static ssize_t etr_port_number_show(struct sys_device *dev, char *buf)
+static ssize_t etr_port_number_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct etr_aib *aib = etr_aib_from_dev(dev);
 
@@ -1199,7 +1209,8 @@ static ssize_t etr_port_number_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(port, 0400, etr_port_number_show, NULL);
 
-static ssize_t etr_coupled_show(struct sys_device *dev, char *buf)
+static ssize_t etr_coupled_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct etr_aib *aib = etr_aib_from_dev(dev);
 
@@ -1210,7 +1221,8 @@ static ssize_t etr_coupled_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(coupled, 0400, etr_coupled_show, NULL);
 
-static ssize_t etr_local_time_show(struct sys_device *dev, char *buf)
+static ssize_t etr_local_time_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct etr_aib *aib = etr_aib_from_dev(dev);
 
@@ -1221,7 +1233,8 @@ static ssize_t etr_local_time_show(struct sys_device *dev, char *buf)
 
 static SYSDEV_ATTR(local_time, 0400, etr_local_time_show, NULL);
 
-static ssize_t etr_utc_offset_show(struct sys_device *dev, char *buf)
+static ssize_t etr_utc_offset_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct etr_aib *aib = etr_aib_from_dev(dev);
 
diff --git a/arch/sh/drivers/dma/dma-sysfs.c b/arch/sh/drivers/dma/dma-sysfs.c
index 51b57c0d1a3c..347ee11351ec 100644
--- a/arch/sh/drivers/dma/dma-sysfs.c
+++ b/arch/sh/drivers/dma/dma-sysfs.c
@@ -23,7 +23,8 @@ static struct sysdev_class dma_sysclass = {
 };
 EXPORT_SYMBOL(dma_sysclass);
 
-static ssize_t dma_show_devices(struct sys_device *dev, char *buf)
+static ssize_t dma_show_devices(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	ssize_t len = 0;
 	int i;
@@ -57,13 +58,15 @@ static int __init dma_sysclass_init(void)
 }
 postcore_initcall(dma_sysclass_init);
 
-static ssize_t dma_show_dev_id(struct sys_device *dev, char *buf)
+static ssize_t dma_show_dev_id(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct dma_channel *channel = to_dma_channel(dev);
 	return sprintf(buf, "%s\n", channel->dev_id);
 }
 
 static ssize_t dma_store_dev_id(struct sys_device *dev,
+				struct sysdev_attribute *attr,
 				const char *buf, size_t count)
 {
 	struct dma_channel *channel = to_dma_channel(dev);
@@ -74,6 +77,7 @@ static ssize_t dma_store_dev_id(struct sys_device *dev,
 static SYSDEV_ATTR(dev_id, S_IRUGO | S_IWUSR, dma_show_dev_id, dma_store_dev_id);
 
 static ssize_t dma_store_config(struct sys_device *dev,
+				struct sysdev_attribute *attr,
 				const char *buf, size_t count)
 {
 	struct dma_channel *channel = to_dma_channel(dev);
@@ -87,13 +91,15 @@ static ssize_t dma_store_config(struct sys_device *dev,
 
 static SYSDEV_ATTR(config, S_IWUSR, NULL, dma_store_config);
 
-static ssize_t dma_show_mode(struct sys_device *dev, char *buf)
+static ssize_t dma_show_mode(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct dma_channel *channel = to_dma_channel(dev);
 	return sprintf(buf, "0x%08x\n", channel->mode);
 }
 
 static ssize_t dma_store_mode(struct sys_device *dev,
+			      struct sysdev_attribute *attr,
 			      const char *buf, size_t count)
 {
 	struct dma_channel *channel = to_dma_channel(dev);
@@ -104,7 +110,8 @@ static ssize_t dma_store_mode(struct sys_device *dev,
 static SYSDEV_ATTR(mode, S_IRUGO | S_IWUSR, dma_show_mode, dma_store_mode);
 
 #define dma_ro_attr(field, fmt)						\
-static ssize_t dma_show_##field(struct sys_device *dev, char *buf)	\
+static ssize_t dma_show_##field(struct sys_device *dev, 		\
+				struct sysdev_attribute *attr, char *buf)\
 {									\
 	struct dma_channel *channel = to_dma_channel(dev);		\
 	return sprintf(buf, fmt, channel->field);			\
diff --git a/arch/sparc64/kernel/sysfs.c b/arch/sparc64/kernel/sysfs.c
index e885034a6b73..84e5ce146713 100644
--- a/arch/sparc64/kernel/sysfs.c
+++ b/arch/sparc64/kernel/sysfs.c
@@ -14,7 +14,8 @@
 static DEFINE_PER_CPU(struct hv_mmu_statistics, mmu_stats) __attribute__((aligned(64)));
 
 #define SHOW_MMUSTAT_ULONG(NAME) \
-static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+static ssize_t show_##NAME(struct sys_device *dev, \
+			struct sysdev_attribute *attr, char *buf) \
 { \
 	struct hv_mmu_statistics *p = &per_cpu(mmu_stats, dev->id); \
 	return sprintf(buf, "%lu\n", p->NAME); \
@@ -135,13 +136,16 @@ static unsigned long write_mmustat_enable(unsigned long val)
 	return sun4v_mmustat_conf(ra, &orig_ra);
 }
 
-static ssize_t show_mmustat_enable(struct sys_device *s, char *buf)
+static ssize_t show_mmustat_enable(struct sys_device *s,
+				struct sysdev_attribute *attr, char *buf)
 {
 	unsigned long val = run_on_cpu(s->id, read_mmustat_enable, 0);
 	return sprintf(buf, "%lx\n", val);
 }
 
-static ssize_t store_mmustat_enable(struct sys_device *s, const char *buf, size_t count)
+static ssize_t store_mmustat_enable(struct sys_device *s,
+			struct sysdev_attribute *attr, const char *buf,
+			size_t count)
 {
 	unsigned long val, err;
 	int ret = sscanf(buf, "%ld", &val);
@@ -179,14 +183,16 @@ static void unregister_mmu_stats(struct sys_device *s)
 #endif
 
 #define SHOW_CPUDATA_ULONG_NAME(NAME, MEMBER) \
-static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+static ssize_t show_##NAME(struct sys_device *dev, \
+		struct sysdev_attribute *attr, char *buf) \
 { \
 	cpuinfo_sparc *c = &cpu_data(dev->id); \
 	return sprintf(buf, "%lu\n", c->MEMBER); \
 }
 
 #define SHOW_CPUDATA_UINT_NAME(NAME, MEMBER) \
-static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+static ssize_t show_##NAME(struct sys_device *dev, \
+		struct sysdev_attribute *attr, char *buf) \
 { \
 	cpuinfo_sparc *c = &cpu_data(dev->id); \
 	return sprintf(buf, "%u\n", c->MEMBER); \
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index c4a7ec31394c..e6a4d5f67643 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -762,10 +762,14 @@ DEFINE_PER_CPU(struct sys_device, device_mce);
 
 /* Why are there no generic functions for this? */
 #define ACCESSOR(name, var, start) \
-	static ssize_t show_ ## name(struct sys_device *s, char *buf) {	\
+	static ssize_t show_ ## name(struct sys_device *s,		\
+				     struct sysdev_attribute *attr,	\
+				     char *buf) {			\
 		return sprintf(buf, "%lx\n", (unsigned long)var);	\
 	}								\
-	static ssize_t set_ ## name(struct sys_device *s,const char *buf,size_t siz) { \
+	static ssize_t set_ ## name(struct sys_device *s,		\
+				    struct sysdev_attribute *attr,	\
+				    const char *buf, size_t siz) {	\
 		char *end;						\
 		unsigned long new = simple_strtoul(buf, &end, 0);	\
 		if (end == buf) return -EINVAL;				\
@@ -786,14 +790,16 @@ ACCESSOR(bank3ctl,bank[3],mce_restart())
 ACCESSOR(bank4ctl,bank[4],mce_restart())
 ACCESSOR(bank5ctl,bank[5],mce_restart())
 
-static ssize_t show_trigger(struct sys_device *s, char *buf)
+static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr,
+				char *buf)
 {
 	strcpy(buf, trigger);
 	strcat(buf, "\n");
 	return strlen(trigger) + 1;
 }
 
-static ssize_t set_trigger(struct sys_device *s,const char *buf,size_t siz)
+static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr,
+				const char *buf,size_t siz)
 {
 	char *p;
 	int len;
diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c
index 1f4cc48c14c6..d5ae2243f0b9 100644
--- a/arch/x86/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c
@@ -35,6 +35,7 @@ atomic_t therm_throt_en = ATOMIC_INIT(0);
 
 #define define_therm_throt_sysdev_show_func(name)                            \
 static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev,        \
+					struct sysdev_attribute *attr,	     \
                                               char *buf)                     \
 {                                                                            \
 	unsigned int cpu = dev->id;                                          \
diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c
index 56b933119a04..fc4790638b69 100644
--- a/arch/x86/kernel/microcode.c
+++ b/arch/x86/kernel/microcode.c
@@ -644,7 +644,9 @@ static void microcode_fini_cpu(int cpu)
 	mutex_unlock(&microcode_mutex);
 }
 
-static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz)
+static ssize_t reload_store(struct sys_device *dev,
+			    struct sysdev_attribute *attr,
+			    const char *buf, size_t sz)
 {
 	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
 	char *end;
@@ -674,14 +676,16 @@ static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz)
 	return sz;
 }
 
-static ssize_t version_show(struct sys_device *dev, char *buf)
+static ssize_t version_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
 
 	return sprintf(buf, "0x%x\n", uci->rev);
 }
 
-static ssize_t pf_show(struct sys_device *dev, char *buf)
+static ssize_t pf_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
 
diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index e38dfed41d80..20537d507909 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -21,15 +21,16 @@ EXPORT_SYMBOL(cpu_sysdev_class);
 static DEFINE_PER_CPU(struct sys_device *, cpu_sys_devices);
 
 #ifdef CONFIG_HOTPLUG_CPU
-static ssize_t show_online(struct sys_device *dev, char *buf)
+static ssize_t show_online(struct sys_device *dev, struct sysdev_attribute *attr,
+			   char *buf)
 {
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
 
 	return sprintf(buf, "%u\n", !!cpu_online(cpu->sysdev.id));
 }
 
-static ssize_t __ref store_online(struct sys_device *dev, const char *buf,
-			    size_t count)
+static ssize_t __ref store_online(struct sys_device *dev, struct sysdev_attribute *attr,
+				 const char *buf, size_t count)
 {
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
 	ssize_t ret;
@@ -80,7 +81,8 @@ static inline void register_cpu_control(struct cpu *cpu)
 #ifdef CONFIG_KEXEC
 #include <linux/kexec.h>
 
-static ssize_t show_crash_notes(struct sys_device *dev, char *buf)
+static ssize_t show_crash_notes(struct sys_device *dev, struct sysdev_attribute *attr,
+				char *buf)
 {
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
 	ssize_t rc;
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 937e8258981d..4d4e0e7b6e92 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -92,7 +92,8 @@ unregister_memory(struct memory_block *memory, struct mem_section *section)
  * uses.
  */
 
-static ssize_t show_mem_phys_index(struct sys_device *dev, char *buf)
+static ssize_t show_mem_phys_index(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct memory_block *mem =
 		container_of(dev, struct memory_block, sysdev);
@@ -102,7 +103,8 @@ static ssize_t show_mem_phys_index(struct sys_device *dev, char *buf)
 /*
  * online, offline, going offline, etc.
  */
-static ssize_t show_mem_state(struct sys_device *dev, char *buf)
+static ssize_t show_mem_state(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	struct memory_block *mem =
 		container_of(dev, struct memory_block, sysdev);
@@ -217,7 +219,8 @@ out:
 }
 
 static ssize_t
-store_mem_state(struct sys_device *dev, const char *buf, size_t count)
+store_mem_state(struct sys_device *dev,
+		struct sysdev_attribute *attr, const char *buf, size_t count)
 {
 	struct memory_block *mem;
 	unsigned int phys_section_nr;
@@ -248,7 +251,8 @@ out:
  * s.t. if I offline all of these sections I can then
  * remove the physical device?
  */
-static ssize_t show_phys_device(struct sys_device *dev, char *buf)
+static ssize_t show_phys_device(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct memory_block *mem =
 		container_of(dev, struct memory_block, sysdev);
diff --git a/drivers/base/node.c b/drivers/base/node.c
index 0f867a083338..5116b78c6325 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -36,11 +36,13 @@ static ssize_t node_read_cpumap(struct sys_device *dev, int type, char *buf)
 	return len;
 }
 
-static inline ssize_t node_read_cpumask(struct sys_device *dev, char *buf)
+static inline ssize_t node_read_cpumask(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	return node_read_cpumap(dev, 0, buf);
 }
-static inline ssize_t node_read_cpulist(struct sys_device *dev, char *buf)
+static inline ssize_t node_read_cpulist(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	return node_read_cpumap(dev, 1, buf);
 }
@@ -49,7 +51,8 @@ static SYSDEV_ATTR(cpumap,  S_IRUGO, node_read_cpumask, NULL);
 static SYSDEV_ATTR(cpulist, S_IRUGO, node_read_cpulist, NULL);
 
 #define K(x) ((x) << (PAGE_SHIFT - 10))
-static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
+static ssize_t node_read_meminfo(struct sys_device * dev,
+			struct sysdev_attribute *attr, char * buf)
 {
 	int n;
 	int nid = dev->id;
@@ -112,7 +115,8 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 #undef K
 static SYSDEV_ATTR(meminfo, S_IRUGO, node_read_meminfo, NULL);
 
-static ssize_t node_read_numastat(struct sys_device * dev, char * buf)
+static ssize_t node_read_numastat(struct sys_device * dev,
+				struct sysdev_attribute *attr, char * buf)
 {
 	return sprintf(buf,
 		       "numa_hit %lu\n"
@@ -130,7 +134,8 @@ static ssize_t node_read_numastat(struct sys_device * dev, char * buf)
 }
 static SYSDEV_ATTR(numastat, S_IRUGO, node_read_numastat, NULL);
 
-static ssize_t node_read_distance(struct sys_device * dev, char * buf)
+static ssize_t node_read_distance(struct sys_device * dev,
+			struct sysdev_attribute *attr, char * buf)
 {
 	int nid = dev->id;
 	int len = 0;
diff --git a/drivers/base/sys.c b/drivers/base/sys.c
index 50690d9df248..dc7dace14e1c 100644
--- a/drivers/base/sys.c
+++ b/drivers/base/sys.c
@@ -36,7 +36,7 @@ sysdev_show(struct kobject * kobj, struct attribute * attr, char * buffer)
 	struct sysdev_attribute * sysdev_attr = to_sysdev_attr(attr);
 
 	if (sysdev_attr->show)
-		return sysdev_attr->show(sysdev, buffer);
+		return sysdev_attr->show(sysdev, sysdev_attr, buffer);
 	return -EIO;
 }
 
@@ -49,7 +49,7 @@ sysdev_store(struct kobject * kobj, struct attribute * attr,
 	struct sysdev_attribute * sysdev_attr = to_sysdev_attr(attr);
 
 	if (sysdev_attr->store)
-		return sysdev_attr->store(sysdev, buffer, count);
+		return sysdev_attr->store(sysdev, sysdev_attr, buffer, count);
 	return -EIO;
 }
 
diff --git a/drivers/base/topology.c b/drivers/base/topology.c
index 3f6d9b0a6abe..199cd97e32e6 100644
--- a/drivers/base/topology.c
+++ b/drivers/base/topology.c
@@ -34,7 +34,8 @@
 static SYSDEV_ATTR(_name, 0444, show_##_name, NULL)
 
 #define define_id_show_func(name)				\
-static ssize_t show_##name(struct sys_device *dev, char *buf)	\
+static ssize_t show_##name(struct sys_device *dev,		\
+		struct sysdev_attribute *attr, char *buf)	\
 {								\
 	unsigned int cpu = dev->id;				\
 	return sprintf(buf, "%d\n", topology_##name(cpu));	\
@@ -59,14 +60,17 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf)
 
 #ifdef arch_provides_topology_pointers
 #define define_siblings_show_map(name)					\
-static ssize_t show_##name(struct sys_device *dev, char *buf)	\
+static ssize_t show_##name(struct sys_device *dev,			\
+			   struct sysdev_attribute *attr, char *buf)	\
 {									\
 	unsigned int cpu = dev->id;					\
 	return show_cpumap(0, &(topology_##name(cpu)), buf);		\
 }
 
 #define define_siblings_show_list(name)					\
-static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \
+static ssize_t show_##name##_list(struct sys_device *dev,		\
+				  struct sysdev_attribute *attr,	\
+				  char *buf)				\
 {									\
 	unsigned int cpu = dev->id;					\
 	return show_cpumap(1, &(topology_##name(cpu)), buf);		\
@@ -74,7 +78,8 @@ static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \
 
 #else
 #define define_siblings_show_map(name)					\
-static ssize_t show_##name(struct sys_device *dev, char *buf)	\
+static ssize_t show_##name(struct sys_device *dev,			\
+			   struct sysdev_attribute *attr, char *buf)	\
 {									\
 	unsigned int cpu = dev->id;					\
 	cpumask_t mask = topology_##name(cpu);				\
@@ -82,7 +87,9 @@ static ssize_t show_##name(struct sys_device *dev, char *buf)	\
 }
 
 #define define_siblings_show_list(name)					\
-static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \
+static ssize_t show_##name##_list(struct sys_device *dev,		\
+				  struct sysdev_attribute *attr,	\
+				  char *buf)				\
 {									\
 	unsigned int cpu = dev->id;					\
 	cpumask_t mask = topology_##name(cpu);				\
diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c
index e949618b9be0..31a0e0b455b6 100644
--- a/drivers/cpuidle/sysfs.c
+++ b/drivers/cpuidle/sysfs.c
@@ -21,7 +21,8 @@ static int __init cpuidle_sysfs_setup(char *unused)
 }
 __setup("cpuidle_sysfs_switch", cpuidle_sysfs_setup);
 
-static ssize_t show_available_governors(struct sys_device *dev, char *buf)
+static ssize_t show_available_governors(struct sys_device *dev,
+		struct sysdev_attribute *attr, char *buf)
 {
 	ssize_t i = 0;
 	struct cpuidle_governor *tmp;
@@ -39,7 +40,8 @@ out:
 	return i;
 }
 
-static ssize_t show_current_driver(struct sys_device *dev, char *buf)
+static ssize_t show_current_driver(struct sys_device *dev,
+		struct sysdev_attribute *attr, char *buf)
 {
 	ssize_t ret;
 
@@ -53,7 +55,8 @@ static ssize_t show_current_driver(struct sys_device *dev, char *buf)
 	return ret;
 }
 
-static ssize_t show_current_governor(struct sys_device *dev, char *buf)
+static ssize_t show_current_governor(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
 {
 	ssize_t ret;
 
@@ -68,6 +71,7 @@ static ssize_t show_current_governor(struct sys_device *dev, char *buf)
 }
 
 static ssize_t store_current_governor(struct sys_device *dev,
+	struct sysdev_attribute *attr,
 	const char *buf, size_t count)
 {
 	char gov_name[CPUIDLE_NAME_LEN];
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index 591bc29b55f5..d4427cb86979 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -610,6 +610,7 @@ static ssize_t show_target_kb(struct sys_device *dev, char *buf)
 }
 
 static ssize_t store_target_kb(struct sys_device *dev,
+			       struct sysdev_attribute *attr,
 			       const char *buf,
 			       size_t count)
 {
diff --git a/include/linux/sysdev.h b/include/linux/sysdev.h
index f2767bc6b735..8dcf3162b21b 100644
--- a/include/linux/sysdev.h
+++ b/include/linux/sysdev.h
@@ -99,8 +99,9 @@ extern void sysdev_unregister(struct sys_device *);
 
 struct sysdev_attribute { 
 	struct attribute	attr;
-	ssize_t (*show)(struct sys_device *, char *);
-	ssize_t (*store)(struct sys_device *, const char *, size_t);
+	ssize_t (*show)(struct sys_device *, struct sysdev_attribute *, char *);
+	ssize_t (*store)(struct sys_device *, struct sysdev_attribute *,
+			 const char *, size_t);
 };
 
 
diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c
index 092e4c620af9..a56f629b057a 100644
--- a/kernel/rtmutex-tester.c
+++ b/kernel/rtmutex-tester.c
@@ -297,8 +297,8 @@ static int test_func(void *data)
  *
  * opcode:data
  */
-static ssize_t sysfs_test_command(struct sys_device *dev, const char *buf,
-				  size_t count)
+static ssize_t sysfs_test_command(struct sys_device *dev, struct sysdev_attribute *attr,
+				  const char *buf, size_t count)
 {
 	struct sched_param schedpar;
 	struct test_thread_data *td;
@@ -360,7 +360,8 @@ static ssize_t sysfs_test_command(struct sys_device *dev, const char *buf,
  * @dev:	thread to query
  * @buf:	char buffer to be filled with thread status info
  */
-static ssize_t sysfs_test_status(struct sys_device *dev, char *buf)
+static ssize_t sysfs_test_status(struct sys_device *dev, struct sysdev_attribute *attr,
+				 char *buf)
 {
 	struct test_thread_data *td;
 	struct task_struct *tsk;
diff --git a/kernel/sched.c b/kernel/sched.c
index 99e6d850ecab..b1104ea5d255 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7737,11 +7737,13 @@ static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
 }
 
 #ifdef CONFIG_SCHED_MC
-static ssize_t sched_mc_power_savings_show(struct sys_device *dev, char *page)
+static ssize_t sched_mc_power_savings_show(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *page)
 {
 	return sprintf(page, "%u\n", sched_mc_power_savings);
 }
 static ssize_t sched_mc_power_savings_store(struct sys_device *dev,
+					    struct sysdev_attribute *attr,
 					    const char *buf, size_t count)
 {
 	return sched_power_savings_store(buf, count, 0);
@@ -7751,11 +7753,13 @@ static SYSDEV_ATTR(sched_mc_power_savings, 0644, sched_mc_power_savings_show,
 #endif
 
 #ifdef CONFIG_SCHED_SMT
-static ssize_t sched_smt_power_savings_show(struct sys_device *dev, char *page)
+static ssize_t sched_smt_power_savings_show(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *page)
 {
 	return sprintf(page, "%u\n", sched_smt_power_savings);
 }
 static ssize_t sched_smt_power_savings_store(struct sys_device *dev,
+					     struct sysdev_attribute *attr,
 					     const char *buf, size_t count)
 {
 	return sched_power_savings_store(buf, count, 1);
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index dadde5361f32..b1c2da81b050 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -376,7 +376,8 @@ void clocksource_unregister(struct clocksource *cs)
  * Provides sysfs interface for listing current clocksource.
  */
 static ssize_t
-sysfs_show_current_clocksources(struct sys_device *dev, char *buf)
+sysfs_show_current_clocksources(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	ssize_t count = 0;
 
@@ -397,6 +398,7 @@ sysfs_show_current_clocksources(struct sys_device *dev, char *buf)
  * clocksource selction.
  */
 static ssize_t sysfs_override_clocksource(struct sys_device *dev,
+					  struct sysdev_attribute *attr,
 					  const char *buf, size_t count)
 {
 	struct clocksource *ovr = NULL;
@@ -449,7 +451,9 @@ static ssize_t sysfs_override_clocksource(struct sys_device *dev,
  * Provides sysfs interface for listing registered clocksources
  */
 static ssize_t
-sysfs_show_available_clocksources(struct sys_device *dev, char *buf)
+sysfs_show_available_clocksources(struct sys_device *dev,
+				  struct sysdev_attribute *attr,
+				  char *buf)
 {
 	struct clocksource *src;
 	ssize_t count = 0;
-- 
cgit v1.2.3


From 9800794ac11d4646384b3a310dfd1fe2eed577bf Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Tue, 1 Jul 2008 18:48:42 +0200
Subject: sysdev: Add utility functions for simple int/ulong variable sysdev
 attributes

This adds a new sysdev_ext_attribute that stores a pointer to the variable
it manages and some utility functions/macro to easily use them.

Previously all users wrote custom macros to generate show/store
functions for each variable, with this it is possible to avoid
that in many cases.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/sys.c     | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/sysdev.h | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 83 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/base/sys.c b/drivers/base/sys.c
index dc7dace14e1c..40fc14f03540 100644
--- a/drivers/base/sys.c
+++ b/drivers/base/sys.c
@@ -479,3 +479,52 @@ int __init system_bus_init(void)
 
 EXPORT_SYMBOL_GPL(sysdev_register);
 EXPORT_SYMBOL_GPL(sysdev_unregister);
+
+#define to_ext_attr(x) container_of(x, struct sysdev_ext_attribute, attr)
+
+ssize_t sysdev_store_ulong(struct sys_device *sysdev,
+			   struct sysdev_attribute *attr,
+			   const char *buf, size_t size)
+{
+	struct sysdev_ext_attribute *ea = to_ext_attr(attr);
+	char *end;
+	unsigned long new = simple_strtoul(buf, &end, 0);
+	if (end == buf)
+		return -EINVAL;
+	*(unsigned long *)(ea->var) = new;
+	return end - buf;
+}
+EXPORT_SYMBOL_GPL(sysdev_store_ulong);
+
+ssize_t sysdev_show_ulong(struct sys_device *sysdev,
+			  struct sysdev_attribute *attr,
+			  char *buf)
+{
+	struct sysdev_ext_attribute *ea = to_ext_attr(attr);
+	return snprintf(buf, PAGE_SIZE, "%lx\n", *(unsigned long *)(ea->var));
+}
+EXPORT_SYMBOL_GPL(sysdev_show_ulong);
+
+ssize_t sysdev_store_int(struct sys_device *sysdev,
+			   struct sysdev_attribute *attr,
+			   const char *buf, size_t size)
+{
+	struct sysdev_ext_attribute *ea = to_ext_attr(attr);
+	char *end;
+	long new = simple_strtol(buf, &end, 0);
+	if (end == buf || new > INT_MAX || new < INT_MIN)
+		return -EINVAL;
+	*(int *)(ea->var) = new;
+	return end - buf;
+}
+EXPORT_SYMBOL_GPL(sysdev_store_int);
+
+ssize_t sysdev_show_int(struct sys_device *sysdev,
+			  struct sysdev_attribute *attr,
+			  char *buf)
+{
+	struct sysdev_ext_attribute *ea = to_ext_attr(attr);
+	return snprintf(buf, PAGE_SIZE, "%d\n", *(int *)(ea->var));
+}
+EXPORT_SYMBOL_GPL(sysdev_show_int);
+
diff --git a/include/linux/sysdev.h b/include/linux/sysdev.h
index 8dcf3162b21b..f395bb3fa2f2 100644
--- a/include/linux/sysdev.h
+++ b/include/linux/sysdev.h
@@ -119,4 +119,38 @@ struct sysdev_attribute {
 extern int sysdev_create_file(struct sys_device *, struct sysdev_attribute *);
 extern void sysdev_remove_file(struct sys_device *, struct sysdev_attribute *);
 
+struct sysdev_ext_attribute {
+	struct sysdev_attribute attr;
+	void *var;
+};
+
+/*
+ * Support for simple variable sysdev attributes.
+ * The pointer to the variable is stored in a sysdev_ext_attribute
+ */
+
+/* Add more types as needed */
+
+extern ssize_t sysdev_show_ulong(struct sys_device *, struct sysdev_attribute *,
+				char *);
+extern ssize_t sysdev_store_ulong(struct sys_device *,
+			struct sysdev_attribute *, const char *, size_t);
+extern ssize_t sysdev_show_int(struct sys_device *, struct sysdev_attribute *,
+				char *);
+extern ssize_t sysdev_store_int(struct sys_device *,
+			struct sysdev_attribute *, const char *, size_t);
+
+#define _SYSDEV_ULONG_ATTR(_name, _mode, _var)				\
+	{ _SYSDEV_ATTR(_name, _mode, sysdev_show_ulong, sysdev_store_ulong), \
+	  &(_var) }
+#define SYSDEV_ULONG_ATTR(_name, _mode, _var)			\
+	struct sysdev_ext_attribute attr_##_name = 		\
+		_SYSDEV_ULONG_ATTR(_name, _mode, _var);
+#define _SYSDEV_INT_ATTR(_name, _mode, _var)				\
+	{ _SYSDEV_ATTR(_name, _mode, sysdev_show_int, sysdev_store_int), \
+	  &(_var) }
+#define SYSDEV_INT_ATTR(_name, _mode, _var)			\
+	struct sysdev_ext_attribute attr_##_name = 		\
+		_SYSDEV_INT_ATTR(_name, _mode, _var);
+
 #endif /* _SYSDEV_H_ */
-- 
cgit v1.2.3


From eadcf0d704a46979c29984fa05f1fc413c775bcb Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 2 Jul 2008 12:46:22 -0700
Subject: MTD: handle pci_name() being const

This changes the MTD core to handle pci_name() now returning a constant
string.

Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/mtd/devices/block2mtd.c | 8 +++++---
 include/linux/mtd/map.h         | 2 +-
 include/linux/mtd/mtd.h         | 2 +-
 3 files changed, 7 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 519d942e7940..7b72a1b36115 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -241,6 +241,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size)
 {
 	struct block_device *bdev;
 	struct block2mtd_dev *dev;
+	char *name;
 
 	if (!devname)
 		return NULL;
@@ -279,12 +280,13 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size)
 
 	/* Setup the MTD structure */
 	/* make the name contain the block device in */
-	dev->mtd.name = kmalloc(sizeof("block2mtd: ") + strlen(devname),
+	name = kmalloc(sizeof("block2mtd: ") + strlen(devname) + 1,
 			GFP_KERNEL);
-	if (!dev->mtd.name)
+	if (!name)
 		goto devinit_err;
 
-	sprintf(dev->mtd.name, "block2mtd: %s", devname);
+	sprintf(name, "block2mtd: %s", devname);
+	dev->mtd.name = name;
 
 	dev->mtd.size = dev->blkdev->bd_inode->i_size & PAGE_MASK;
 	dev->mtd.erasesize = erase_size;
diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h
index a9fae032ba81..9c1d95491f8b 100644
--- a/include/linux/mtd/map.h
+++ b/include/linux/mtd/map.h
@@ -189,7 +189,7 @@ typedef union {
 */
 
 struct map_info {
-	char *name;
+	const char *name;
 	unsigned long size;
 	resource_size_t phys;
 #define NO_XIP (-1UL)
diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h
index 245f9098e171..8b5d49133ec6 100644
--- a/include/linux/mtd/mtd.h
+++ b/include/linux/mtd/mtd.h
@@ -121,7 +121,7 @@ struct mtd_info {
 	u_int32_t oobavail;  // Available OOB bytes per block
 
 	// Kernel-only stuff starts here.
-	char *name;
+	const char *name;
 	int index;
 
 	/* ecc layout structure pointer - read only ! */
-- 
cgit v1.2.3


From af5406895a05720a879dc33e2f4878fa503e81b3 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Tue, 22 Jul 2008 19:24:26 -0500
Subject: module: reorder struct module to save space on 64 bit builds

reorder struct module to save space on 64 bit builds.
saves 1 cacheline_size  (128 on default x86_64 & 64 on AMD
Opteron/athlon) when CONFIG_MODULE_UNLOAD=y.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/module.h | 42 ++++++++++++++++++++++--------------------
 1 file changed, 22 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/module.h b/include/linux/module.h
index 3e03b1acbc94..63f0eb69e214 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -249,27 +249,28 @@ struct module
 
 	/* Exported symbols */
 	const struct kernel_symbol *syms;
-	unsigned int num_syms;
 	const unsigned long *crcs;
+	unsigned int num_syms;
 
 	/* GPL-only exported symbols. */
-	const struct kernel_symbol *gpl_syms;
 	unsigned int num_gpl_syms;
+	const struct kernel_symbol *gpl_syms;
 	const unsigned long *gpl_crcs;
 
 	/* unused exported symbols. */
 	const struct kernel_symbol *unused_syms;
-	unsigned int num_unused_syms;
 	const unsigned long *unused_crcs;
+	unsigned int num_unused_syms;
+
 	/* GPL-only, unused exported symbols. */
-	const struct kernel_symbol *unused_gpl_syms;
 	unsigned int num_unused_gpl_syms;
+	const struct kernel_symbol *unused_gpl_syms;
 	const unsigned long *unused_gpl_crcs;
 
 	/* symbols that will be GPL-only in the near future. */
 	const struct kernel_symbol *gpl_future_syms;
-	unsigned int num_gpl_future_syms;
 	const unsigned long *gpl_future_crcs;
+	unsigned int num_gpl_future_syms;
 
 	/* Exception table */
 	unsigned int num_exentries;
@@ -300,23 +301,9 @@ struct module
 
 #ifdef CONFIG_GENERIC_BUG
 	/* Support for BUG */
+	unsigned num_bugs;
 	struct list_head bug_list;
 	struct bug_entry *bug_table;
-	unsigned num_bugs;
-#endif
-
-#ifdef CONFIG_MODULE_UNLOAD
-	/* Reference counts */
-	struct module_ref ref[NR_CPUS];
-
-	/* What modules depend on me? */
-	struct list_head modules_which_use_me;
-
-	/* Who is waiting for us to be unloaded */
-	struct task_struct *waiter;
-
-	/* Destruction function. */
-	void (*exit)(void);
 #endif
 
 #ifdef CONFIG_KALLSYMS
@@ -342,6 +329,21 @@ struct module
 	struct marker *markers;
 	unsigned int num_markers;
 #endif
+
+#ifdef CONFIG_MODULE_UNLOAD
+	/* What modules depend on me? */
+	struct list_head modules_which_use_me;
+
+	/* Who is waiting for us to be unloaded */
+	struct task_struct *waiter;
+
+	/* Destruction function. */
+	void (*exit)(void);
+
+	/* Reference counts */
+	struct module_ref ref[NR_CPUS];
+#endif
+
 };
 #ifndef MODULE_ARCH_INIT
 #define MODULE_ARCH_INIT {}
-- 
cgit v1.2.3


From f7f5b67557eac1131ba6532522e3c50eced34238 Mon Sep 17 00:00:00 2001
From: Denys Vlasenko <vda.linux@googlemail.com>
Date: Tue, 22 Jul 2008 19:24:26 -0500
Subject: Shrink struct module: CONFIG_UNUSED_SYMBOLS ifdefs

module.c and module.h conatains code for finding
exported symbols which are declared with EXPORT_UNUSED_SYMBOL,
and this code is compiled in even if CONFIG_UNUSED_SYMBOLS is not set
and thus there can be no EXPORT_UNUSED_SYMBOLs in modules anyway
(because EXPORT_UNUSED_SYMBOL(x) are compiled out to nothing then).

This patch adds required #ifdefs.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/module.h |  2 ++
 init/Kconfig           |  4 ++--
 kernel/module.c        | 49 ++++++++++++++++++++++++++++++++++---------------
 3 files changed, 38 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/module.h b/include/linux/module.h
index 63f0eb69e214..a860a2c1f379 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -257,6 +257,7 @@ struct module
 	const struct kernel_symbol *gpl_syms;
 	const unsigned long *gpl_crcs;
 
+#ifdef CONFIG_UNUSED_SYMBOLS
 	/* unused exported symbols. */
 	const struct kernel_symbol *unused_syms;
 	const unsigned long *unused_crcs;
@@ -266,6 +267,7 @@ struct module
 	unsigned int num_unused_gpl_syms;
 	const struct kernel_symbol *unused_gpl_syms;
 	const unsigned long *unused_gpl_crcs;
+#endif
 
 	/* symbols that will be GPL-only in the near future. */
 	const struct kernel_symbol *gpl_future_syms;
diff --git a/init/Kconfig b/init/Kconfig
index 6199d1120900..c8578f9ee31d 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -856,8 +856,8 @@ config MODULE_UNLOAD
 	help
 	  Without this option you will not be able to unload any
 	  modules (note that some modules may not be unloadable
-	  anyway), which makes your kernel slightly smaller and
-	  simpler.  If unsure, say Y.
+	  anyway), which makes your kernel smaller, faster
+	  and simpler.  If unsure, say Y.
 
 config MODULE_FORCE_UNLOAD
 	bool "Forced module unloading"
diff --git a/kernel/module.c b/kernel/module.c
index c51c089c666e..ea9580521eb1 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -134,17 +134,19 @@ extern const struct kernel_symbol __start___ksymtab_gpl[];
 extern const struct kernel_symbol __stop___ksymtab_gpl[];
 extern const struct kernel_symbol __start___ksymtab_gpl_future[];
 extern const struct kernel_symbol __stop___ksymtab_gpl_future[];
-extern const struct kernel_symbol __start___ksymtab_unused[];
-extern const struct kernel_symbol __stop___ksymtab_unused[];
-extern const struct kernel_symbol __start___ksymtab_unused_gpl[];
-extern const struct kernel_symbol __stop___ksymtab_unused_gpl[];
 extern const struct kernel_symbol __start___ksymtab_gpl_future[];
 extern const struct kernel_symbol __stop___ksymtab_gpl_future[];
 extern const unsigned long __start___kcrctab[];
 extern const unsigned long __start___kcrctab_gpl[];
 extern const unsigned long __start___kcrctab_gpl_future[];
+#ifdef CONFIG_UNUSED_SYMBOLS
+extern const struct kernel_symbol __start___ksymtab_unused[];
+extern const struct kernel_symbol __stop___ksymtab_unused[];
+extern const struct kernel_symbol __start___ksymtab_unused_gpl[];
+extern const struct kernel_symbol __stop___ksymtab_unused_gpl[];
 extern const unsigned long __start___kcrctab_unused[];
 extern const unsigned long __start___kcrctab_unused_gpl[];
+#endif
 
 #ifndef CONFIG_MODVERSIONS
 #define symversion(base, idx) NULL
@@ -198,12 +200,14 @@ static bool each_symbol(bool (*fn)(const struct symsearch *arr,
 		{ __start___ksymtab_gpl_future, __stop___ksymtab_gpl_future,
 		  __start___kcrctab_gpl_future,
 		  WILL_BE_GPL_ONLY, false },
+#ifdef CONFIG_UNUSED_SYMBOLS
 		{ __start___ksymtab_unused, __stop___ksymtab_unused,
 		  __start___kcrctab_unused,
 		  NOT_GPL_ONLY, true },
 		{ __start___ksymtab_unused_gpl, __stop___ksymtab_unused_gpl,
 		  __start___kcrctab_unused_gpl,
 		  GPL_ONLY, true },
+#endif
 	};
 
 	if (each_symbol_in_section(arr, ARRAY_SIZE(arr), NULL, fn, data))
@@ -220,6 +224,7 @@ static bool each_symbol(bool (*fn)(const struct symsearch *arr,
 			  mod->gpl_future_syms + mod->num_gpl_future_syms,
 			  mod->gpl_future_crcs,
 			  WILL_BE_GPL_ONLY, false },
+#ifdef CONFIG_UNUSED_SYMBOLS
 			{ mod->unused_syms,
 			  mod->unused_syms + mod->num_unused_syms,
 			  mod->unused_crcs,
@@ -228,6 +233,7 @@ static bool each_symbol(bool (*fn)(const struct symsearch *arr,
 			  mod->unused_gpl_syms + mod->num_unused_gpl_syms,
 			  mod->unused_gpl_crcs,
 			  GPL_ONLY, true },
+#endif
 		};
 
 		if (each_symbol_in_section(arr, ARRAY_SIZE(arr), mod, fn, data))
@@ -270,6 +276,7 @@ static bool find_symbol_in_section(const struct symsearch *syms,
 		}
 	}
 
+#ifdef CONFIG_UNUSED_SYMBOLS
 	if (syms->unused && fsa->warn) {
 		printk(KERN_WARNING "Symbol %s is marked as UNUSED, "
 		       "however this module is using it.\n", fsa->name);
@@ -281,6 +288,7 @@ static bool find_symbol_in_section(const struct symsearch *syms,
 		       "mailinglist together with submitting your code for "
 		       "inclusion.\n");
 	}
+#endif
 
 	fsa->owner = owner;
 	fsa->crc = symversion(syms->crcs, symnum);
@@ -1476,8 +1484,10 @@ static int verify_export_symbols(struct module *mod)
 		{ mod->syms, mod->num_syms },
 		{ mod->gpl_syms, mod->num_gpl_syms },
 		{ mod->gpl_future_syms, mod->num_gpl_future_syms },
+#ifdef CONFIG_UNUSED_SYMBOLS
 		{ mod->unused_syms, mod->num_unused_syms },
 		{ mod->unused_gpl_syms, mod->num_unused_gpl_syms },
+#endif
 	};
 
 	for (i = 0; i < ARRAY_SIZE(arr); i++) {
@@ -1795,10 +1805,12 @@ static struct module *load_module(void __user *umod,
 	unsigned int gplfutureindex;
 	unsigned int gplfuturecrcindex;
 	unsigned int unwindex = 0;
+#ifdef CONFIG_UNUSED_SYMBOLS
 	unsigned int unusedindex;
 	unsigned int unusedcrcindex;
 	unsigned int unusedgplindex;
 	unsigned int unusedgplcrcindex;
+#endif
 	unsigned int markersindex;
 	unsigned int markersstringsindex;
 	struct module *mod;
@@ -1881,13 +1893,15 @@ static struct module *load_module(void __user *umod,
 	exportindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab");
 	gplindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_gpl");
 	gplfutureindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_gpl_future");
-	unusedindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_unused");
-	unusedgplindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_unused_gpl");
 	crcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab");
 	gplcrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_gpl");
 	gplfuturecrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_gpl_future");
+#ifdef CONFIG_UNUSED_SYMBOLS
+	unusedindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_unused");
+	unusedgplindex = find_sec(hdr, sechdrs, secstrings, "__ksymtab_unused_gpl");
 	unusedcrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_unused");
 	unusedgplcrcindex = find_sec(hdr, sechdrs, secstrings, "__kcrctab_unused_gpl");
+#endif
 	setupindex = find_sec(hdr, sechdrs, secstrings, "__param");
 	exindex = find_sec(hdr, sechdrs, secstrings, "__ex_table");
 	obsparmindex = find_sec(hdr, sechdrs, secstrings, "__obsparm");
@@ -2049,14 +2063,15 @@ static struct module *load_module(void __user *umod,
 		mod->gpl_crcs = (void *)sechdrs[gplcrcindex].sh_addr;
 	mod->num_gpl_future_syms = sechdrs[gplfutureindex].sh_size /
 					sizeof(*mod->gpl_future_syms);
-	mod->num_unused_syms = sechdrs[unusedindex].sh_size /
-					sizeof(*mod->unused_syms);
-	mod->num_unused_gpl_syms = sechdrs[unusedgplindex].sh_size /
-					sizeof(*mod->unused_gpl_syms);
 	mod->gpl_future_syms = (void *)sechdrs[gplfutureindex].sh_addr;
 	if (gplfuturecrcindex)
 		mod->gpl_future_crcs = (void *)sechdrs[gplfuturecrcindex].sh_addr;
 
+#ifdef CONFIG_UNUSED_SYMBOLS
+	mod->num_unused_syms = sechdrs[unusedindex].sh_size /
+					sizeof(*mod->unused_syms);
+	mod->num_unused_gpl_syms = sechdrs[unusedgplindex].sh_size /
+					sizeof(*mod->unused_gpl_syms);
 	mod->unused_syms = (void *)sechdrs[unusedindex].sh_addr;
 	if (unusedcrcindex)
 		mod->unused_crcs = (void *)sechdrs[unusedcrcindex].sh_addr;
@@ -2064,13 +2079,17 @@ static struct module *load_module(void __user *umod,
 	if (unusedgplcrcindex)
 		mod->unused_gpl_crcs
 			= (void *)sechdrs[unusedgplcrcindex].sh_addr;
+#endif
 
 #ifdef CONFIG_MODVERSIONS
-	if ((mod->num_syms && !crcindex) ||
-	    (mod->num_gpl_syms && !gplcrcindex) ||
-	    (mod->num_gpl_future_syms && !gplfuturecrcindex) ||
-	    (mod->num_unused_syms && !unusedcrcindex) ||
-	    (mod->num_unused_gpl_syms && !unusedgplcrcindex)) {
+	if ((mod->num_syms && !crcindex)
+	    || (mod->num_gpl_syms && !gplcrcindex)
+	    || (mod->num_gpl_future_syms && !gplfuturecrcindex)
+#ifdef CONFIG_UNUSED_SYMBOLS
+	    || (mod->num_unused_syms && !unusedcrcindex)
+	    || (mod->num_unused_gpl_syms && !unusedgplcrcindex)
+#endif
+		) {
 		printk(KERN_WARNING "%s: No versions for exported symbols.\n", mod->name);
 		err = try_to_force_load(mod, "nocrc");
 		if (err)
-- 
cgit v1.2.3


From 2f0f2a334bc38b61a9afca951185cd3844ee709d Mon Sep 17 00:00:00 2001
From: Denys Vlasenko <vda.linux@googlemail.com>
Date: Tue, 22 Jul 2008 19:24:27 -0500
Subject: module: turn longs into ints for module sizes

This shrinks module.o and each *.ko file.

And finally, structure members which hold length of module
code (four such members there) and count of symbols
are converted from longs to ints.

We cannot possibly have a module where 32 bits won't
be enough to hold such counts.

For one, module loading checks module size for sanity
before loading, so such insanely big module will fail
that test first.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/module.h | 6 +++---
 kernel/module.c        | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/module.h b/include/linux/module.h
index a860a2c1f379..fce15ebd0e1c 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -288,10 +288,10 @@ struct module
 	void *module_core;
 
 	/* Here are the sizes of the init and core sections */
-	unsigned long init_size, core_size;
+	unsigned int init_size, core_size;
 
 	/* The size of the executable code in each section.  */
-	unsigned long init_text_size, core_text_size;
+	unsigned int init_text_size, core_text_size;
 
 	/* The handle returned from unwind_add_table. */
 	void *unwind_info;
@@ -311,7 +311,7 @@ struct module
 #ifdef CONFIG_KALLSYMS
 	/* We keep the symbol and string tables for kallsyms. */
 	Elf_Sym *symtab;
-	unsigned long num_symtab;
+	unsigned int num_symtab;
 	char *strtab;
 
 	/* Section attributes */
diff --git a/kernel/module.c b/kernel/module.c
index ea9580521eb1..5c7eb0695b3c 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -1567,7 +1567,7 @@ static int simplify_symbols(Elf_Shdr *sechdrs,
 }
 
 /* Update size with this section: return offset. */
-static long get_offset(unsigned long *size, Elf_Shdr *sechdr)
+static long get_offset(unsigned int *size, Elf_Shdr *sechdr)
 {
 	long ret;
 
@@ -2562,7 +2562,7 @@ static int m_show(struct seq_file *m, void *p)
 	struct module *mod = list_entry(p, struct module, list);
 	char buf[8];
 
-	seq_printf(m, "%s %lu",
+	seq_printf(m, "%s %u",
 		   mod->name, mod->init_size + mod->core_size);
 	print_unload_info(m, mod);
 
-- 
cgit v1.2.3


From df648c9fbebb4de50e7a6e76cc253c7cb0421f9b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Tue, 8 Jul 2008 19:00:18 +0200
Subject: rework try_then_request_module to do less in non-modular kernels

This reworks try_then_request_module to only invoke the "lookup"
function "x" once when the kernel is not modular.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/kmod.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kmod.h b/include/linux/kmod.h
index 5dc13848891b..9bdb28d6660f 100644
--- a/include/linux/kmod.h
+++ b/include/linux/kmod.h
@@ -29,11 +29,12 @@
 /* modprobe exit status on success, -ve on error.  Return value
  * usually useless though. */
 extern int request_module(const char * name, ...) __attribute__ ((format (printf, 1, 2)));
+#define try_then_request_module(x, mod...) ((x) ?: (request_module(mod), (x)))
 #else
 static inline int request_module(const char * name, ...) { return -ENOSYS; }
+#define try_then_request_module(x, mod...) (x)
 #endif
 
-#define try_then_request_module(x, mod...) ((x) ?: (request_module(mod), (x)))
 
 struct key;
 struct file;
-- 
cgit v1.2.3


From a1ef5adb4cad43460ebba23c5a78cf4a55bb6a5b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Tue, 8 Jul 2008 19:00:17 +0200
Subject: remove CONFIG_KMOD from core kernel code

Always compile request_module when the kernel allows modules.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/kmod.h | 2 +-
 kernel/exec_domain.c | 2 +-
 kernel/kmod.c        | 2 +-
 kernel/sysctl.c      | 4 ++--
 4 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kmod.h b/include/linux/kmod.h
index 9bdb28d6660f..0509c4ce4857 100644
--- a/include/linux/kmod.h
+++ b/include/linux/kmod.h
@@ -25,7 +25,7 @@
 
 #define KMOD_PATH_LEN 256
 
-#ifdef CONFIG_KMOD
+#ifdef CONFIG_MODULES
 /* modprobe exit status on success, -ve on error.  Return value
  * usually useless though. */
 extern int request_module(const char * name, ...) __attribute__ ((format (printf, 1, 2)));
diff --git a/kernel/exec_domain.c b/kernel/exec_domain.c
index a9e6bad9f706..c1ef192aa655 100644
--- a/kernel/exec_domain.c
+++ b/kernel/exec_domain.c
@@ -65,7 +65,7 @@ lookup_exec_domain(u_long personality)
 				goto out;
 	}
 
-#ifdef CONFIG_KMOD
+#ifdef CONFIG_MODULES
 	read_unlock(&exec_domains_lock);
 	request_module("personality-%ld", pers);
 	read_lock(&exec_domains_lock);
diff --git a/kernel/kmod.c b/kernel/kmod.c
index 8df97d3dfda8..90d7af1c1655 100644
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -42,7 +42,7 @@ extern int max_threads;
 
 static struct workqueue_struct *khelper_wq;
 
-#ifdef CONFIG_KMOD
+#ifdef CONFIG_MODULES
 
 /*
 	modprobe_path is set via /proc/sys.
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 6b16e16428d8..b859e6b5a767 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -110,7 +110,7 @@ static int min_percpu_pagelist_fract = 8;
 
 static int ngroups_max = NGROUPS_MAX;
 
-#ifdef CONFIG_KMOD
+#ifdef CONFIG_MODULES
 extern char modprobe_path[];
 #endif
 #ifdef CONFIG_CHR_DEV_SG
@@ -475,7 +475,7 @@ static struct ctl_table kern_table[] = {
 		.proc_handler	= &ftrace_enable_sysctl,
 	},
 #endif
-#ifdef CONFIG_KMOD
+#ifdef CONFIG_MODULES
 	{
 		.ctl_name	= KERN_MODPROBE,
 		.procname	= "modprobe",
-- 
cgit v1.2.3


From 651910874633a75f4a726d44e449be0a56b4b2e0 Mon Sep 17 00:00:00 2001
From: John Reiser <jreiser@BitWagon.com>
Date: Mon, 21 Jul 2008 14:21:32 -0700
Subject: execve filename: document and export via auxiliary vector

The Linux kernel puts the filename argument of execve() into the new
address space.  Many developers are surprised to learn this.  Those who
know and could use it, object "But it's not documented."

Those who want to use it dislike the expression
  (char *)(1+ strlen(env[-1+ n_env]) + env[-1+ n_env])
because it requires locating the last original environment variable,
and assumes that the filename follows the characters.

This patch documents the insertion of the filename, and makes it easier
to find by adding a new tag AT_EXECFN in the ElfXX_auxv_t; see <elf.h>.

In many cases readlink("/proc/self/exe",) gives the same answer.  But if
all the original pages get unmapped, then the kernel erases the symlink
for /proc/self/exe.  This can happen when a program decompressor does a
good job of cleaning up after uncompressing directly to memory, so that
the address space of the target program looks the same as if compression
had never happened.  One example is http://upx.sourceforge.net .

One notable use of the underlying concept (what path containED the
executable) is glibc expanding $ORIGIN in DT_RUNPATH.  In practice for
the near term, it may be a good idea for user-mode code to use both
/proc/self/exe and AT_EXECFN as fall-back methods for each other.
/proc/self/exe can fail due to unmapping, AT_EXECFN can fail because it
won't be present on non-new systems.  The auxvec or {AT_EXECFN}.d_val
also can get overwritten, although in nearly all cases this would be the
result of a bug.

The runtime cost is one NEW_AUX_ENT using two words of stack space.  The
underlying value is maintained already as bprm->exec; setup_arg_pages()
in fs/exec.c slides it for stack_shift, etc.

Signed-off-by: John Reiser <jreiser@BitWagon.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Jakub Jelinek <jakub@redhat.com>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_elf.c        | 1 +
 include/linux/auxvec.h | 4 +++-
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index d48ff5f370f4..639d2d8b5710 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -204,6 +204,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	NEW_AUX_ENT(AT_GID, tsk->gid);
 	NEW_AUX_ENT(AT_EGID, tsk->egid);
  	NEW_AUX_ENT(AT_SECURE, security_bprm_secureexec(bprm));
+	NEW_AUX_ENT(AT_EXECFN, bprm->exec);
 	if (k_platform) {
 		NEW_AUX_ENT(AT_PLATFORM,
 			    (elf_addr_t)(unsigned long)u_platform);
diff --git a/include/linux/auxvec.h b/include/linux/auxvec.h
index ad895455ab72..0da17d14fd13 100644
--- a/include/linux/auxvec.h
+++ b/include/linux/auxvec.h
@@ -26,8 +26,10 @@
 
 #define AT_SECURE 23   /* secure mode boolean */
 
+#define AT_EXECFN  31	/* filename of program */
 #ifdef __KERNEL__
-#define AT_VECTOR_SIZE_BASE (14 + 2) /* NEW_AUX_ENT entries in auxiliary table */
+#define AT_VECTOR_SIZE_BASE 17 /* NEW_AUX_ENT entries in auxiliary table */
+  /* number of "#define AT_.*" above, minus {AT_NULL, AT_IGNORE, AT_NOTELF} */
 #endif
 
 #endif /* _LINUX_AUXVEC_H */
-- 
cgit v1.2.3


From 0c36ec31473593aa937ff04f3b3b630e81512734 Mon Sep 17 00:00:00 2001
From: Juergen Beisert <j.beisert@pengutronix.de>
Date: Mon, 21 Jul 2008 14:21:34 -0700
Subject: gpio: gpio driver for max7301 SPI GPIO expander

Maxim's MAX7301 is an SPI GPIO expander with 28 GPIOs.  Note: MAX7301's
interrupt feature is not supported yet.

[akpm@linux-foundation.org: coding-style fixes]
[g.liakhovetski@pengutronix.de: Fix inaccuracies in comments, check spi_setup()
return code, mask off high byte in max7301_read()]
Signed-off-by: Juergen Beisert <j.beisert@pengutronix.de>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@pengutronix.de>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/gpio/Kconfig        |   6 +
 drivers/gpio/Makefile       |   1 +
 drivers/gpio/max7301.c      | 339 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/spi/max7301.h |   9 ++
 4 files changed, 355 insertions(+)
 create mode 100644 drivers/gpio/max7301.c
 create mode 100644 include/linux/spi/max7301.h

(limited to 'include/linux')

diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index 008c38ba774f..9e0c4fbfc51a 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -69,6 +69,12 @@ config GPIO_PCF857X
 
 comment "SPI GPIO expanders:"
 
+config GPIO_MAX7301
+	tristate "Maxim MAX7301 GPIO expander"
+	depends on SPI_MASTER
+	help
+	  gpio driver for Maxim MAX7301 SPI GPIO expander.
+
 config GPIO_MCP23S08
 	tristate "Microchip MCP23S08 I/O expander"
 	depends on SPI_MASTER
diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile
index fdde9923cf33..16e796dc5410 100644
--- a/drivers/gpio/Makefile
+++ b/drivers/gpio/Makefile
@@ -4,6 +4,7 @@ ccflags-$(CONFIG_DEBUG_GPIO)	+= -DDEBUG
 
 obj-$(CONFIG_HAVE_GPIO_LIB)	+= gpiolib.o
 
+obj-$(CONFIG_GPIO_MAX7301)	+= max7301.o
 obj-$(CONFIG_GPIO_MCP23S08)	+= mcp23s08.o
 obj-$(CONFIG_GPIO_PCA953X)	+= pca953x.o
 obj-$(CONFIG_GPIO_PCF857X)	+= pcf857x.o
diff --git a/drivers/gpio/max7301.c b/drivers/gpio/max7301.c
new file mode 100644
index 000000000000..39c795ad8312
--- /dev/null
+++ b/drivers/gpio/max7301.c
@@ -0,0 +1,339 @@
+/**
+ * drivers/gpio/max7301.c
+ *
+ * Copyright (C) 2006 Juergen Beisert, Pengutronix
+ * Copyright (C) 2008 Guennadi Liakhovetski, Pengutronix
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * The Maxim's MAX7301 device is an SPI driven GPIO expander. There are
+ * 28 GPIOs. 8 of them can trigger an interrupt. See datasheet for more
+ * details
+ * Note:
+ * - DIN must be stable at the rising edge of clock.
+ * - when writing:
+ *   - always clock in 16 clocks at once
+ *   - at DIN: D15 first, D0 last
+ *   - D0..D7 = databyte, D8..D14 = commandbyte
+ *   - D15 = low -> write command
+ * - when reading
+ *   - always clock in 16 clocks at once
+ *   - at DIN: D15 first, D0 last
+ *   - D0..D7 = dummy, D8..D14 = register address
+ *   - D15 = high -> read command
+ *   - raise CS and assert it again
+ *   - always clock in 16 clocks at once
+ *   - at DOUT: D15 first, D0 last
+ *   - D0..D7 contains the data from the first cycle
+ *
+ * The driver exports a standard gpiochip interface
+ */
+
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/mutex.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/max7301.h>
+#include <linux/gpio.h>
+
+#define DRIVER_NAME "max7301"
+
+/*
+ * Pin configurations, see MAX7301 datasheet page 6
+ */
+#define PIN_CONFIG_MASK 0x03
+#define PIN_CONFIG_IN_PULLUP 0x03
+#define PIN_CONFIG_IN_WO_PULLUP 0x02
+#define PIN_CONFIG_OUT 0x01
+
+#define PIN_NUMBER 28
+
+
+/*
+ * Some registers must be read back to modify.
+ * To save time we cache them here in memory
+ */
+struct max7301 {
+	struct mutex	lock;
+	u8		port_config[8];	/* field 0 is unused */
+	u32		out_level;	/* cached output levels */
+	struct gpio_chip chip;
+	struct spi_device *spi;
+};
+
+/**
+ * max7301_write - Write a new register content
+ * @spi: The SPI device
+ * @reg: Register offset
+ * @val: Value to write
+ *
+ * A write to the MAX7301 means one message with one transfer
+ *
+ * Returns 0 if successful or a negative value on error
+ */
+static int max7301_write(struct spi_device *spi, unsigned int reg, unsigned int val)
+{
+	u16 word = ((reg & 0x7F) << 8) | (val & 0xFF);
+	return spi_write(spi, (const u8 *)&word, sizeof(word));
+}
+
+/**
+ * max7301_read - Read back register content
+ * @spi: The SPI device
+ * @reg: Register offset
+ *
+ * A read from the MAX7301 means two transfers; here, one message each
+ *
+ * Returns positive 8 bit value from device if successful or a
+ * negative value on error
+ */
+static int max7301_read(struct spi_device *spi, unsigned int reg)
+{
+	int ret;
+	u16 word;
+
+	word = 0x8000 | (reg << 8);
+	ret = spi_write(spi, (const u8 *)&word, sizeof(word));
+	if (ret)
+		return ret;
+	/*
+	 * This relies on the fact, that a transfer with NULL tx_buf shifts out
+	 * zero bytes (=NOOP for MAX7301)
+	 */
+	ret = spi_read(spi, (u8 *)&word, sizeof(word));
+	if (ret)
+		return ret;
+	return word & 0xff;
+}
+
+static int max7301_direction_input(struct gpio_chip *chip, unsigned offset)
+{
+	struct max7301 *ts = container_of(chip, struct max7301, chip);
+	u8 *config;
+	int ret;
+
+	/* First 4 pins are unused in the controller */
+	offset += 4;
+
+	config = &ts->port_config[offset >> 2];
+
+	mutex_lock(&ts->lock);
+
+	/* Standard GPIO API doesn't support pull-ups, has to be extended.
+	 * Hard-coding no pollup for now. */
+	*config = (*config & ~(3 << (offset & 3))) | (1 << (offset & 3));
+
+	ret = max7301_write(ts->spi, 0x08 + (offset >> 2), *config);
+
+	mutex_unlock(&ts->lock);
+
+	return ret;
+}
+
+static int __max7301_set(struct max7301 *ts, unsigned offset, int value)
+{
+	if (value) {
+		ts->out_level |= 1 << offset;
+		return max7301_write(ts->spi, 0x20 + offset, 0x01);
+	} else {
+		ts->out_level &= ~(1 << offset);
+		return max7301_write(ts->spi, 0x20 + offset, 0x00);
+	}
+}
+
+static int max7301_direction_output(struct gpio_chip *chip, unsigned offset,
+				    int value)
+{
+	struct max7301 *ts = container_of(chip, struct max7301, chip);
+	u8 *config;
+	int ret;
+
+	/* First 4 pins are unused in the controller */
+	offset += 4;
+
+	config = &ts->port_config[offset >> 2];
+
+	mutex_lock(&ts->lock);
+
+	*config = (*config & ~(3 << (offset & 3))) | (1 << (offset & 3));
+
+	ret = __max7301_set(ts, offset, value);
+
+	if (!ret)
+		ret = max7301_write(ts->spi, 0x08 + (offset >> 2), *config);
+
+	mutex_unlock(&ts->lock);
+
+	return ret;
+}
+
+static int max7301_get(struct gpio_chip *chip, unsigned offset)
+{
+	struct max7301 *ts = container_of(chip, struct max7301, chip);
+	int config, level = -EINVAL;
+
+	/* First 4 pins are unused in the controller */
+	offset += 4;
+
+	mutex_lock(&ts->lock);
+
+	config = (ts->port_config[offset >> 2] >> ((offset & 3) * 2)) & 3;
+
+	switch (config) {
+	case 1:
+		/* Output: return cached level */
+		level =  !!(ts->out_level & (1 << offset));
+		break;
+	case 2:
+	case 3:
+		/* Input: read out */
+		level = max7301_read(ts->spi, 0x20 + offset) & 0x01;
+	}
+	mutex_unlock(&ts->lock);
+
+	return level;
+}
+
+static void max7301_set(struct gpio_chip *chip, unsigned offset, int value)
+{
+	struct max7301 *ts = container_of(chip, struct max7301, chip);
+
+	/* First 4 pins are unused in the controller */
+	offset += 4;
+
+	mutex_lock(&ts->lock);
+
+	__max7301_set(ts, offset, value);
+
+	mutex_unlock(&ts->lock);
+}
+
+static int __devinit max7301_probe(struct spi_device *spi)
+{
+	struct max7301 *ts;
+	struct max7301_platform_data *pdata;
+	int i, ret;
+
+	pdata = spi->dev.platform_data;
+	if (!pdata || !pdata->base)
+		return -ENODEV;
+
+	/*
+	 * bits_per_word cannot be configured in platform data
+	 */
+	spi->bits_per_word = 16;
+
+	ret = spi_setup(spi);
+	if (ret < 0)
+		return ret;
+
+	ts = kzalloc(sizeof(struct max7301), GFP_KERNEL);
+	if (!ts)
+		return -ENOMEM;
+
+	mutex_init(&ts->lock);
+
+	dev_set_drvdata(&spi->dev, ts);
+
+	/* Power up the chip and disable IRQ output */
+	max7301_write(spi, 0x04, 0x01);
+
+	ts->spi = spi;
+
+	ts->chip.label = DRIVER_NAME,
+
+	ts->chip.direction_input = max7301_direction_input;
+	ts->chip.get = max7301_get;
+	ts->chip.direction_output = max7301_direction_output;
+	ts->chip.set = max7301_set;
+
+	ts->chip.base = pdata->base;
+	ts->chip.ngpio = PIN_NUMBER;
+	ts->chip.can_sleep = 1;
+	ts->chip.dev = &spi->dev;
+	ts->chip.owner = THIS_MODULE;
+
+	ret = gpiochip_add(&ts->chip);
+	if (ret)
+		goto exit_destroy;
+
+	/*
+	 * tristate all pins in hardware and cache the
+	 * register values for later use.
+	 */
+	for (i = 1; i < 8; i++) {
+		int j;
+		/* 0xAA means input with internal pullup disabled */
+		max7301_write(spi, 0x08 + i, 0xAA);
+		ts->port_config[i] = 0xAA;
+		for (j = 0; j < 4; j++) {
+			int idx = ts->chip.base + (i - 1) * 4 + j;
+			ret = gpio_direction_input(idx);
+			if (ret)
+				goto exit_remove;
+			gpio_free(idx);
+		}
+	}
+	return ret;
+
+exit_remove:
+	gpiochip_remove(&ts->chip);
+exit_destroy:
+	dev_set_drvdata(&spi->dev, NULL);
+	mutex_destroy(&ts->lock);
+	kfree(ts);
+	return ret;
+}
+
+static int max7301_remove(struct spi_device *spi)
+{
+	struct max7301 *ts;
+	int ret;
+
+	ts = dev_get_drvdata(&spi->dev);
+	if (ts == NULL)
+		return -ENODEV;
+
+	dev_set_drvdata(&spi->dev, NULL);
+
+	/* Power down the chip and disable IRQ output */
+	max7301_write(spi, 0x04, 0x00);
+
+	ret = gpiochip_remove(&ts->chip);
+	if (!ret) {
+		mutex_destroy(&ts->lock);
+		kfree(ts);
+	} else
+		dev_err(&spi->dev, "Failed to remove the GPIO controller: %d\n",
+			ret);
+
+	return ret;
+}
+
+static struct spi_driver max7301_driver = {
+	.driver = {
+		.name		= DRIVER_NAME,
+		.owner		= THIS_MODULE,
+	},
+	.probe		= max7301_probe,
+	.remove		= __devexit_p(max7301_remove),
+};
+
+static int __init max7301_init(void)
+{
+	return spi_register_driver(&max7301_driver);
+}
+
+static void __exit max7301_exit(void)
+{
+	spi_unregister_driver(&max7301_driver);
+}
+
+module_init(max7301_init);
+module_exit(max7301_exit);
+
+MODULE_AUTHOR("Juergen Beisert");
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("MAX7301 SPI based GPIO-Expander");
diff --git a/include/linux/spi/max7301.h b/include/linux/spi/max7301.h
new file mode 100644
index 000000000000..6dfd83f19b4b
--- /dev/null
+++ b/include/linux/spi/max7301.h
@@ -0,0 +1,9 @@
+#ifndef LINUX_SPI_MAX7301_H
+#define LINUX_SPI_MAX7301_H
+
+struct max7301_platform_data {
+	/* number assigned to the first GPIO */
+	unsigned	base;
+};
+
+#endif
-- 
cgit v1.2.3


From 95da310e66ee8090119596c70ca8432e57f9a97f Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Tue, 22 Jul 2008 11:09:07 +0100
Subject: usb_serial: API all change

USB serial likes to use port->tty back pointers for the real work it does and
to do so without any actual locking. Unfortunately when you consider hangup
events, hangup/parallel reopen or even worse hangup followed by parallel close
events the tty->port and port->tty pointers are not guaranteed to be the same
as port->tty is the active tty while tty->port is the port the tty may or
may not still be attached to.

So rework the entire API to pass the tty struct. For console cases we need
to pass both for now. This shows up multiple drivers that immediately crash
with USB console some of which have been fixed in the process.

Longer term we need a proper tty as console abstraction

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/usb/serial/aircable.c         |  19 +-
 drivers/usb/serial/airprime.c         | 355 ++++++++++++++++++++++++++++++++++
 drivers/usb/serial/ark3116.c          |  32 +--
 drivers/usb/serial/belkin_sa.c        |  69 +++----
 drivers/usb/serial/ch341.c            |  10 +-
 drivers/usb/serial/console.c          |  34 ++--
 drivers/usb/serial/cp2101.c           |  74 +++----
 drivers/usb/serial/cyberjack.c        |  47 +++--
 drivers/usb/serial/cypress_m8.c       | 117 ++++++-----
 drivers/usb/serial/digi_acceleport.c  | 127 ++++++------
 drivers/usb/serial/empeg.c            |  78 +++-----
 drivers/usb/serial/ftdi_sio.c         | 211 ++++++++++----------
 drivers/usb/serial/garmin_gps.c       |  58 ++----
 drivers/usb/serial/generic.c          |  39 ++--
 drivers/usb/serial/io_edgeport.c      | 175 ++++++-----------
 drivers/usb/serial/io_ti.c            | 109 ++++++-----
 drivers/usb/serial/ipaq.c             |  48 +++--
 drivers/usb/serial/ipw.c              |  15 +-
 drivers/usb/serial/ir-usb.c           |  81 ++++----
 drivers/usb/serial/iuu_phoenix.c      |  43 ++--
 drivers/usb/serial/keyspan.c          | 138 ++++++-------
 drivers/usb/serial/keyspan.h          |  39 ++--
 drivers/usb/serial/keyspan_pda.c      |  69 +++----
 drivers/usb/serial/kl5kusb105.c       |  74 ++++---
 drivers/usb/serial/kobil_sct.c        |  80 ++++----
 drivers/usb/serial/mct_u232.c         | 121 +++++-------
 drivers/usb/serial/mos7720.c          |  87 ++++-----
 drivers/usb/serial/mos7840.c          | 165 +++++++---------
 drivers/usb/serial/navman.c           |  10 +-
 drivers/usb/serial/omninet.c          |  26 +--
 drivers/usb/serial/option.c           | 105 ++++------
 drivers/usb/serial/oti6858.c          | 119 +++++-------
 drivers/usb/serial/pl2303.c           |  76 ++++----
 drivers/usb/serial/safe_serial.c      |  14 +-
 drivers/usb/serial/sierra.c           |  93 +++------
 drivers/usb/serial/spcp8x5.c          |  69 ++++---
 drivers/usb/serial/ti_usb_3410_5052.c | 203 ++++++++++---------
 drivers/usb/serial/usb-serial.c       |  82 ++++----
 drivers/usb/serial/usb_debug.c        |   5 +-
 drivers/usb/serial/visor.c            |  66 ++-----
 drivers/usb/serial/whiteheat.c        | 102 +++++-----
 include/linux/usb/serial.h            |  56 +++---
 42 files changed, 1820 insertions(+), 1720 deletions(-)
 create mode 100644 drivers/usb/serial/airprime.c

(limited to 'include/linux')

diff --git a/drivers/usb/serial/aircable.c b/drivers/usb/serial/aircable.c
index db6f97a93c02..79ea98c66fa8 100644
--- a/drivers/usb/serial/aircable.c
+++ b/drivers/usb/serial/aircable.c
@@ -272,7 +272,7 @@ static void aircable_read(struct work_struct *work)
 	 * 64 bytes, to ensure I do not get throttled.
 	 * Ask USB mailing list for better aproach.
 	 */
-	tty = port->tty;
+	tty = port->port.tty;
 
 	if (!tty) {
 		schedule_work(&priv->rx_work);
@@ -378,13 +378,14 @@ static void aircable_shutdown(struct usb_serial *serial)
 	}
 }
 
-static int aircable_write_room(struct usb_serial_port *port)
+static int aircable_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct aircable_private *priv = usb_get_serial_port_data(port);
 	return serial_buf_data_avail(priv->tx_buf);
 }
 
-static int aircable_write(struct usb_serial_port *port,
+static int aircable_write(struct tty_struct *tty, struct usb_serial_port *port,
 			  const unsigned char *source, int count)
 {
 	struct aircable_private *priv = usb_get_serial_port_data(port);
@@ -466,7 +467,7 @@ static void aircable_read_bulk_callback(struct urb *urb)
 
 	if (status) {
 		dbg("%s - urb status = %d", __func__, status);
-		if (!port->open_count) {
+		if (!port->port.count) {
 			dbg("%s - port is closed, exiting.", __func__);
 			return;
 		}
@@ -494,7 +495,7 @@ static void aircable_read_bulk_callback(struct urb *urb)
 	usb_serial_debug_data(debug, &port->dev, __func__,
 				urb->actual_length, urb->transfer_buffer);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		if (urb->actual_length <= 2) {
 			/* This is an incomplete package */
@@ -528,7 +529,7 @@ static void aircable_read_bulk_callback(struct urb *urb)
 	}
 
 	/* Schedule the next read _if_ we are still open */
-	if (port->open_count) {
+	if (port->port.count) {
 		usb_fill_bulk_urb(port->read_urb, port->serial->dev,
 				  usb_rcvbulkpipe(port->serial->dev,
 					  port->bulk_in_endpointAddress),
@@ -547,8 +548,9 @@ static void aircable_read_bulk_callback(struct urb *urb)
 }
 
 /* Based on ftdi_sio.c throttle */
-static void aircable_throttle(struct usb_serial_port *port)
+static void aircable_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct aircable_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -560,8 +562,9 @@ static void aircable_throttle(struct usb_serial_port *port)
 }
 
 /* Based on ftdi_sio.c unthrottle */
-static void aircable_unthrottle(struct usb_serial_port *port)
+static void aircable_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct aircable_private *priv = usb_get_serial_port_data(port);
 	int actually_throttled;
 	unsigned long flags;
diff --git a/drivers/usb/serial/airprime.c b/drivers/usb/serial/airprime.c
new file mode 100644
index 000000000000..b3f1d1e82468
--- /dev/null
+++ b/drivers/usb/serial/airprime.c
@@ -0,0 +1,355 @@
+/*
+ * AirPrime CDMA Wireless Serial USB driver
+ *
+ * Copyright (C) 2005-2006 Greg Kroah-Hartman <gregkh@suse.de>
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License version
+ *	2 as published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+#include <linux/usb/serial.h>
+
+static struct usb_device_id id_table [] = {
+	{ USB_DEVICE(0x0c88, 0x17da) }, /* Kyocera Wireless KPC650/Passport */
+	{ },
+};
+MODULE_DEVICE_TABLE(usb, id_table);
+
+#define URB_TRANSFER_BUFFER_SIZE	4096
+#define NUM_READ_URBS			4
+#define NUM_WRITE_URBS			4
+#define NUM_BULK_EPS			3
+#define MAX_BULK_EPS			6
+
+/* if overridden by the user, then use their value for the size of the
+ * read and write urbs, and the number of endpoints */
+static int buffer_size = URB_TRANSFER_BUFFER_SIZE;
+static int endpoints = NUM_BULK_EPS;
+static int debug;
+struct airprime_private {
+	spinlock_t lock;
+	int outstanding_urbs;
+	int throttled;
+	struct urb *read_urbp[NUM_READ_URBS];
+
+	/* Settings for the port */
+	int rts_state;	/* Handshaking pins (outputs) */
+	int dtr_state;
+	int cts_state;	/* Handshaking pins (inputs) */
+	int dsr_state;
+	int dcd_state;
+	int ri_state;
+};
+
+static int airprime_send_setup(struct usb_serial_port *port)
+{
+	struct usb_serial *serial = port->serial;
+	struct airprime_private *priv;
+
+	dbg("%s", __func__);
+
+	if (port->number != 0)
+		return 0;
+
+	priv = usb_get_serial_port_data(port);
+
+	if (port->port.tty) {
+		int val = 0;
+		if (priv->dtr_state)
+			val |= 0x01;
+		if (priv->rts_state)
+			val |= 0x02;
+
+		return usb_control_msg(serial->dev,
+					usb_rcvctrlpipe(serial->dev, 0),
+					0x22, 0x21, val, 0, NULL, 0,
+					USB_CTRL_SET_TIMEOUT);
+	}
+
+	return 0;
+}
+
+static void airprime_read_bulk_callback(struct urb *urb)
+{
+	struct usb_serial_port *port = urb->context;
+	unsigned char *data = urb->transfer_buffer;
+	struct tty_struct *tty;
+	int result;
+	int status = urb->status;
+
+	dbg("%s - port %d", __func__, port->number);
+
+	if (status) {
+		dbg("%s - nonzero read bulk status received: %d",
+		    __func__, status);
+		return;
+	}
+	usb_serial_debug_data(debug, &port->dev, __func__,
+						urb->actual_length, data);
+
+	tty = port->port.tty;
+	if (tty && urb->actual_length) {
+		tty_insert_flip_string(tty, data, urb->actual_length);
+		tty_flip_buffer_push(tty);
+	}
+
+	result = usb_submit_urb(urb, GFP_ATOMIC);
+	if (result)
+		dev_err(&port->dev,
+			"%s - failed resubmitting read urb, error %d\n",
+			__func__, result);
+	return;
+}
+
+static void airprime_write_bulk_callback(struct urb *urb)
+{
+	struct usb_serial_port *port = urb->context;
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	int status = urb->status;
+	unsigned long flags;
+
+	dbg("%s - port %d", __func__, port->number);
+
+	/* free up the transfer buffer, as usb_free_urb() does not do this */
+	kfree(urb->transfer_buffer);
+
+	if (status)
+		dbg("%s - nonzero write bulk status received: %d",
+		    __func__, status);
+	spin_lock_irqsave(&priv->lock, flags);
+	--priv->outstanding_urbs;
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	usb_serial_port_softint(port);
+}
+
+static int airprime_open(struct tty_struct *tty, struct usb_serial_port *port,
+							struct file *filp)
+{
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	struct usb_serial *serial = port->serial;
+	struct urb *urb;
+	char *buffer = NULL;
+	int i;
+	int result = 0;
+
+	dbg("%s - port %d", __func__, port->number);
+
+	/* initialize our private data structure if it isn't already created */
+	if (!priv) {
+		priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+		if (!priv) {
+			result = -ENOMEM;
+			goto out;
+		}
+		spin_lock_init(&priv->lock);
+		usb_set_serial_port_data(port, priv);
+	}
+
+	/* Set some sane defaults */
+	priv->rts_state = 1;
+	priv->dtr_state = 1;
+
+	for (i = 0; i < NUM_READ_URBS; ++i) {
+		buffer = kmalloc(buffer_size, GFP_KERNEL);
+		if (!buffer) {
+			dev_err(&port->dev, "%s - out of memory.\n",
+				__func__);
+			result = -ENOMEM;
+			goto errout;
+		}
+		urb = usb_alloc_urb(0, GFP_KERNEL);
+		if (!urb) {
+			kfree(buffer);
+			dev_err(&port->dev, "%s - no more urbs?\n",
+				__func__);
+			result = -ENOMEM;
+			goto errout;
+		}
+		usb_fill_bulk_urb(urb, serial->dev,
+				  usb_rcvbulkpipe(serial->dev,
+					  port->bulk_out_endpointAddress),
+				  buffer, buffer_size,
+				  airprime_read_bulk_callback, port);
+		result = usb_submit_urb(urb, GFP_KERNEL);
+		if (result) {
+			usb_free_urb(urb);
+			kfree(buffer);
+			dev_err(&port->dev,
+				"%s - failed submitting read urb %d for port %d, error %d\n",
+				__func__, i, port->number, result);
+			goto errout;
+		}
+		/* remember this urb so we can kill it when the
+		   port is closed */
+		priv->read_urbp[i] = urb;
+	}
+
+	airprime_send_setup(port);
+
+	goto out;
+
+ errout:
+	/* some error happened, cancel any submitted urbs and clean up
+	   anything that got allocated successfully */
+
+	while (i-- != 0) {
+		urb = priv->read_urbp[i];
+		buffer = urb->transfer_buffer;
+		usb_kill_urb(urb);
+		usb_free_urb(urb);
+		kfree(buffer);
+	}
+
+ out:
+	return result;
+}
+
+static void airprime_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
+{
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	int i;
+
+	dbg("%s - port %d", __func__, port->number);
+
+	priv->rts_state = 0;
+	priv->dtr_state = 0;
+
+	mutex_lock(&port->serial->disc_mutex);
+	if (!port->serial->disconnected)
+		airprime_send_setup(port);
+	mutex_unlock(&port->serial->disc_mutex);
+
+	for (i = 0; i < NUM_READ_URBS; ++i) {
+		usb_kill_urb(priv->read_urbp[i]);
+		kfree(priv->read_urbp[i]->transfer_buffer);
+		usb_free_urb(priv->read_urbp[i]);
+	}
+
+	/* free up private structure */
+	kfree(priv);
+	usb_set_serial_port_data(port, NULL);
+}
+
+static int airprime_write(struct tty_struct *tty, struct usb_serial_port *port,
+			  const unsigned char *buf, int count)
+{
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	struct usb_serial *serial = port->serial;
+	struct urb *urb;
+	unsigned char *buffer;
+	unsigned long flags;
+	int status;
+	dbg("%s - port %d", __func__, port->number);
+
+	spin_lock_irqsave(&priv->lock, flags);
+	if (priv->outstanding_urbs > NUM_WRITE_URBS) {
+		spin_unlock_irqrestore(&priv->lock, flags);
+		dbg("%s - write limit hit\n", __func__);
+		return 0;
+	}
+	spin_unlock_irqrestore(&priv->lock, flags);
+	buffer = kmalloc(count, GFP_ATOMIC);
+	if (!buffer) {
+		dev_err(&port->dev, "out of memory\n");
+		return -ENOMEM;
+	}
+	urb = usb_alloc_urb(0, GFP_ATOMIC);
+	if (!urb) {
+		dev_err(&port->dev, "no more free urbs\n");
+		kfree(buffer);
+		return -ENOMEM;
+	}
+	memcpy(buffer, buf, count);
+
+	usb_serial_debug_data(debug, &port->dev, __func__, count, buffer);
+
+	usb_fill_bulk_urb(urb, serial->dev,
+			  usb_sndbulkpipe(serial->dev,
+					  port->bulk_out_endpointAddress),
+			  buffer, count,
+			  airprime_write_bulk_callback, port);
+
+	/* send it down the pipe */
+	status = usb_submit_urb(urb, GFP_ATOMIC);
+	if (status) {
+		dev_err(&port->dev,
+			"%s - usb_submit_urb(write bulk) failed with status = %d\n",
+			__func__, status);
+		count = status;
+		kfree(buffer);
+	} else {
+		spin_lock_irqsave(&priv->lock, flags);
+		++priv->outstanding_urbs;
+		spin_unlock_irqrestore(&priv->lock, flags);
+	}
+	/* we are done with this urb, so let the host driver
+	 * really free it when it is finished with it */
+	usb_free_urb(urb);
+	return count;
+}
+
+static struct usb_driver airprime_driver = {
+	.name =		"airprime",
+	.probe =	usb_serial_probe,
+	.disconnect =	usb_serial_disconnect,
+	.id_table =	id_table,
+	.no_dynamic_id =	1,
+};
+
+static struct usb_serial_driver airprime_device = {
+	.driver = {
+		.owner =	THIS_MODULE,
+		.name =		"airprime",
+	},
+	.usb_driver =		&airprime_driver,
+	.id_table =		id_table,
+	.open =			airprime_open,
+	.close =		airprime_close,
+	.write =		airprime_write,
+};
+
+static int __init airprime_init(void)
+{
+	int retval;
+
+	airprime_device.num_ports = endpoints;
+	if (endpoints < 0 || endpoints >= MAX_BULK_EPS)
+		airprime_device.num_ports = NUM_BULK_EPS;
+
+	retval = usb_serial_register(&airprime_device);
+	if (retval)
+		return retval;
+	retval = usb_register(&airprime_driver);
+	if (retval)
+		usb_serial_deregister(&airprime_device);
+	return retval;
+}
+
+static void __exit airprime_exit(void)
+{
+	dbg("%s", __func__);
+
+	usb_deregister(&airprime_driver);
+	usb_serial_deregister(&airprime_device);
+}
+
+module_init(airprime_init);
+module_exit(airprime_exit);
+MODULE_LICENSE("GPL");
+
+module_param(debug, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug, "Debug enabled");
+module_param(buffer_size, int, 0);
+MODULE_PARM_DESC(buffer_size,
+		"Size of the transfer buffers in bytes (default 4096)");
+module_param(endpoints, int, 0);
+MODULE_PARM_DESC(endpoints, "Number of bulk EPs to configure (default 3)");
diff --git a/drivers/usb/serial/ark3116.c b/drivers/usb/serial/ark3116.c
index 77895c8f8f31..aec61880f36c 100644
--- a/drivers/usb/serial/ark3116.c
+++ b/drivers/usb/serial/ark3116.c
@@ -158,12 +158,13 @@ cleanup:
 	return -ENOMEM;
 }
 
-static void ark3116_set_termios(struct usb_serial_port *port,
+static void ark3116_set_termios(struct tty_struct *tty,
+				struct usb_serial_port *port,
 				struct ktermios *old_termios)
 {
 	struct usb_serial *serial = port->serial;
 	struct ark3116_private *priv = usb_get_serial_port_data(port);
-	struct ktermios *termios = port->tty->termios;
+	struct ktermios *termios = tty->termios;
 	unsigned int cflag = termios->c_cflag;
 	unsigned long flags;
 	int baud;
@@ -177,8 +178,8 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 
 	spin_lock_irqsave(&priv->lock, flags);
 	if (!priv->termios_initialized) {
-		*(port->tty->termios) = tty_std_termios;
-		port->tty->termios->c_cflag = B9600 | CS8
+		*termios = tty_std_termios;
+		termios->c_cflag = B9600 | CS8
 					      | CREAD | HUPCL | CLOCAL;
 		termios->c_ispeed = 9600;
 		termios->c_ospeed = 9600;
@@ -192,7 +193,7 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 	buf = kmalloc(1, GFP_KERNEL);
 	if (!buf) {
 		dbg("error kmalloc");
-		*port->tty->termios = *old_termios;
+		*termios = *old_termios;
 		return;
 	}
 
@@ -243,7 +244,7 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 	}
 
 	/* set baudrate */
-	baud = tty_get_baud_rate(port->tty);
+	baud = tty_get_baud_rate(tty);
 
 	switch (baud) {
 	case 75:
@@ -262,11 +263,11 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 	case 230400:
 	case 460800:
 		/* Report the resulting rate back to the caller */
-		tty_encode_baud_rate(port->tty, baud, baud);
+		tty_encode_baud_rate(tty, baud, baud);
 		break;
 	/* set 9600 as default (if given baudrate is invalid for example) */
 	default:
-		tty_encode_baud_rate(port->tty, 9600, 9600);
+		tty_encode_baud_rate(tty, 9600, 9600);
 	case 0:
 		baud = 9600;
 	}
@@ -317,7 +318,8 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 	return;
 }
 
-static int ark3116_open(struct usb_serial_port *port, struct file *filp)
+static int ark3116_open(struct tty_struct *tty, struct usb_serial_port *port,
+					struct file *filp)
 {
 	struct ktermios tmp_termios;
 	struct usb_serial *serial = port->serial;
@@ -332,7 +334,7 @@ static int ark3116_open(struct usb_serial_port *port, struct file *filp)
 		return -ENOMEM;
 	}
 
-	result = usb_serial_generic_open(port, filp);
+	result = usb_serial_generic_open(tty, port, filp);
 	if (result)
 		goto err_out;
 
@@ -362,8 +364,8 @@ static int ark3116_open(struct usb_serial_port *port, struct file *filp)
 	ARK3116_RCV(serial, 124, 0xFE, 0xC0, 0x0000, 0x0006, 0xFF, buf);
 
 	/* initialise termios */
-	if (port->tty)
-		ark3116_set_termios(port, &tmp_termios);
+	if (tty)
+		ark3116_set_termios(tty, port, &tmp_termios);
 
 err_out:
 	kfree(buf);
@@ -371,9 +373,10 @@ err_out:
 	return result;
 }
 
-static int ark3116_ioctl(struct usb_serial_port *port, struct file *file,
+static int ark3116_ioctl(struct tty_struct *tty, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct serial_struct serstruct;
 	void __user *user_arg = (void __user *)arg;
 
@@ -403,8 +406,9 @@ static int ark3116_ioctl(struct usb_serial_port *port, struct file *file,
 	return -ENOIOCTLCMD;
 }
 
-static int ark3116_tiocmget(struct usb_serial_port *port, struct file *file)
+static int ark3116_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	char *buf;
 	char temp;
diff --git a/drivers/usb/serial/belkin_sa.c b/drivers/usb/serial/belkin_sa.c
index 0a322fc53d6e..1a762692c739 100644
--- a/drivers/usb/serial/belkin_sa.c
+++ b/drivers/usb/serial/belkin_sa.c
@@ -89,14 +89,13 @@ static int debug;
 /* function prototypes for a Belkin USB Serial Adapter F5U103 */
 static int  belkin_sa_startup		(struct usb_serial *serial);
 static void belkin_sa_shutdown		(struct usb_serial *serial);
-static int  belkin_sa_open		(struct usb_serial_port *port, struct file *filp);
-static void belkin_sa_close		(struct usb_serial_port *port, struct file *filp);
+static int  belkin_sa_open		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void belkin_sa_close		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
 static void belkin_sa_read_int_callback (struct urb *urb);
-static void belkin_sa_set_termios	(struct usb_serial_port *port, struct ktermios * old);
-static int  belkin_sa_ioctl		(struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg);
-static void belkin_sa_break_ctl		(struct usb_serial_port *port, int break_state );
-static int  belkin_sa_tiocmget		(struct usb_serial_port *port, struct file *file);
-static int  belkin_sa_tiocmset		(struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear);
+static void belkin_sa_set_termios	(struct tty_struct *tty, struct usb_serial_port *port, struct ktermios * old);
+static void belkin_sa_break_ctl		(struct tty_struct *tty, int break_state );
+static int  belkin_sa_tiocmget		(struct tty_struct *tty, struct file *file);
+static int  belkin_sa_tiocmset		(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear);
 
 
 static struct usb_device_id id_table_combined [] = {
@@ -132,7 +131,6 @@ static struct usb_serial_driver belkin_device = {
 	.open =			belkin_sa_open,
 	.close =		belkin_sa_close,
 	.read_int_callback =	belkin_sa_read_int_callback,	/* How we get the status info */
-	.ioctl =		belkin_sa_ioctl,
 	.set_termios =		belkin_sa_set_termios,
 	.break_ctl =		belkin_sa_break_ctl,
 	.tiocmget =		belkin_sa_tiocmget,
@@ -190,7 +188,7 @@ static int belkin_sa_startup (struct usb_serial *serial)
 }
 
 
-static void belkin_sa_shutdown (struct usb_serial *serial)
+static void belkin_sa_shutdown(struct usb_serial *serial)
 {
 	struct belkin_sa_private *priv;
 	int i;
@@ -206,7 +204,7 @@ static void belkin_sa_shutdown (struct usb_serial *serial)
 }
 
 
-static int  belkin_sa_open (struct usb_serial_port *port, struct file *filp)
+static int  belkin_sa_open(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp)
 {
 	int retval = 0;
 
@@ -235,7 +233,8 @@ exit:
 } /* belkin_sa_open */
 
 
-static void belkin_sa_close (struct usb_serial_port *port, struct file *filp)
+static void belkin_sa_close (struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	dbg("%s port %d", __func__, port->number);
 
@@ -246,7 +245,7 @@ static void belkin_sa_close (struct usb_serial_port *port, struct file *filp)
 } /* belkin_sa_close */
 
 
-static void belkin_sa_read_int_callback (struct urb *urb)
+static void belkin_sa_read_int_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	struct belkin_sa_private *priv;
@@ -311,7 +310,7 @@ static void belkin_sa_read_int_callback (struct urb *urb)
 	 * to look in to this before committing any code.
 	 */
 	if (priv->last_lsr & BELKIN_SA_LSR_ERR) {
-		tty = port->tty;
+		tty = port->port.tty;
 		/* Overrun Error */
 		if (priv->last_lsr & BELKIN_SA_LSR_OE) {
 		}
@@ -334,7 +333,8 @@ exit:
 		     __func__, retval);
 }
 
-static void belkin_sa_set_termios (struct usb_serial_port *port, struct ktermios *old_termios)
+static void belkin_sa_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct usb_serial *serial = port->serial;
 	struct belkin_sa_private *priv = usb_get_serial_port_data(port);
@@ -347,7 +347,7 @@ static void belkin_sa_set_termios (struct usb_serial_port *port, struct ktermios
 	unsigned long control_state;
 	int bad_flow_control;
 	speed_t baud;
-	struct ktermios *termios = port->tty->termios;
+	struct ktermios *termios = tty->termios;
 	
 	iflag = termios->c_iflag;
 	cflag = termios->c_cflag;
@@ -377,7 +377,7 @@ static void belkin_sa_set_termios (struct usb_serial_port *port, struct ktermios
 		}
 	}
 
-	baud = tty_get_baud_rate(port->tty);
+	baud = tty_get_baud_rate(tty);
 	if (baud) {
 		urb_value = BELKIN_SA_BAUD(baud);
 		/* Clip to maximum speed */
@@ -387,7 +387,7 @@ static void belkin_sa_set_termios (struct usb_serial_port *port, struct ktermios
 		baud = BELKIN_SA_BAUD(urb_value);
 
 		/* Report the actual baud rate back to the caller */
-		tty_encode_baud_rate(port->tty, baud, baud);
+		tty_encode_baud_rate(tty, baud, baud);
 		if (BSA_USB_CMD(BELKIN_SA_SET_BAUDRATE_REQUEST, urb_value) < 0)
 			err("Set baudrate error");
 	} else {
@@ -463,8 +463,9 @@ static void belkin_sa_set_termios (struct usb_serial_port *port, struct ktermios
 } /* belkin_sa_set_termios */
 
 
-static void belkin_sa_break_ctl( struct usb_serial_port *port, int break_state )
+static void belkin_sa_break_ctl(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 
 	if (BSA_USB_CMD(BELKIN_SA_SET_BREAK_REQUEST, break_state ? 1 : 0) < 0)
@@ -472,8 +473,9 @@ static void belkin_sa_break_ctl( struct usb_serial_port *port, int break_state )
 }
 
 
-static int belkin_sa_tiocmget (struct usb_serial_port *port, struct file *file)
+static int belkin_sa_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct belkin_sa_private *priv = usb_get_serial_port_data(port);
 	unsigned long control_state;
 	unsigned long flags;
@@ -488,9 +490,10 @@ static int belkin_sa_tiocmget (struct usb_serial_port *port, struct file *file)
 }
 
 
-static int belkin_sa_tiocmset (struct usb_serial_port *port, struct file *file,
+static int belkin_sa_tiocmset(struct tty_struct *tty, struct file *file,
 			       unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	struct belkin_sa_private *priv = usb_get_serial_port_data(port);
 	unsigned long control_state;
@@ -540,29 +543,7 @@ exit:
 }
 
 
-static int belkin_sa_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg)
-{
-	switch (cmd) {
-	case TIOCMIWAIT:
-		/* wait for any of the 4 modem inputs (DCD,RI,DSR,CTS)*/
-		/* TODO */
-		return( 0 );
-
-	case TIOCGICOUNT:
-		/* return count of modemline transitions */
-		/* TODO */
-		return 0;
-
-	default:
-		dbg("belkin_sa_ioctl arg not supported - 0x%04x",cmd);
-		return(-ENOIOCTLCMD);
-		break;
-	}
-	return 0;
-} /* belkin_sa_ioctl */
-
-
-static int __init belkin_sa_init (void)
+static int __init belkin_sa_init(void)
 {
 	int retval;
 	retval = usb_serial_register(&belkin_device);
@@ -583,7 +564,7 @@ failed_usb_serial_register:
 static void __exit belkin_sa_exit (void)
 {
 	usb_deregister (&belkin_driver);
-	usb_serial_deregister (&belkin_device);
+	usb_serial_deregister(&belkin_device);
 }
 
 
diff --git a/drivers/usb/serial/ch341.c b/drivers/usb/serial/ch341.c
index 1f7c86bd8297..f61e3ca64305 100644
--- a/drivers/usb/serial/ch341.c
+++ b/drivers/usb/serial/ch341.c
@@ -232,7 +232,8 @@ error:	kfree(priv);
 }
 
 /* open this device, set default parameters */
-static int ch341_open(struct usb_serial_port *port, struct file *filp)
+static int ch341_open(struct tty_struct *tty, struct usb_serial_port *port,
+				struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	struct ch341_private *priv = usb_get_serial_port_data(serial->port[0]);
@@ -256,7 +257,7 @@ static int ch341_open(struct usb_serial_port *port, struct file *filp)
 	if (r)
 		goto out;
 
-	r = usb_serial_generic_open(port, filp);
+	r = usb_serial_generic_open(tty, port, filp);
 
 out:	return r;
 }
@@ -264,11 +265,10 @@ out:	return r;
 /* Old_termios contains the original termios settings and
  * tty->termios contains the new setting to be used.
  */
-static void ch341_set_termios(struct usb_serial_port *port,
-			      struct ktermios *old_termios)
+static void ch341_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct ch341_private *priv = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	unsigned baud_rate;
 
 	dbg("ch341_set_termios()");
diff --git a/drivers/usb/serial/console.c b/drivers/usb/serial/console.c
index 201184c3fb87..940f5de68980 100644
--- a/drivers/usb/serial/console.c
+++ b/drivers/usb/serial/console.c
@@ -145,12 +145,12 @@ static int usb_console_setup(struct console *co, char *options)
 	}
 
 	port = serial->port[0];
-	port->tty = NULL;
+	port->port.tty = NULL;
 
 	info->port = port;
 	 
-	++port->open_count;
-	if (port->open_count == 1) {
+	++port->port.count;
+	if (port->port.count == 1) {
 		if (serial->type->set_termios) {
 			/*
 			 * allocate a fake tty so the driver can initialize
@@ -171,15 +171,15 @@ static int usb_console_setup(struct console *co, char *options)
 			}
 			memset(&dummy, 0, sizeof(struct ktermios));
 			tty->termios = termios;
-			port->tty = tty;
+			port->port.tty = tty;
 		}
 
 		/* only call the device specific open if this 
 		 * is the first time the port is opened */
 		if (serial->type->open)
-			retval = serial->type->open(port, NULL);
+			retval = serial->type->open(NULL, port, NULL);
 		else
-			retval = usb_serial_generic_open(port, NULL);
+			retval = usb_serial_generic_open(NULL, port, NULL);
 
 		if (retval) {
 			err("could not open USB console port");
@@ -188,9 +188,9 @@ static int usb_console_setup(struct console *co, char *options)
 
 		if (serial->type->set_termios) {
 			termios->c_cflag = cflag;
-			serial->type->set_termios(port, &dummy);
+			serial->type->set_termios(NULL, port, &dummy);
 
-			port->tty = NULL;
+			port->port.tty = NULL;
 			kfree(termios);
 			kfree(tty);
 		}
@@ -203,11 +203,11 @@ out:
 	return retval;
 free_termios:
 	kfree(termios);
-	port->tty = NULL;
+	port->port.tty = NULL;
 free_tty:
 	kfree(tty);
 reset_open_count:
-	port->open_count = 0;
+	port->port.count = 0;
 goto out;
 }
 
@@ -227,7 +227,7 @@ static void usb_console_write(struct console *co, const char *buf, unsigned coun
 
 	dbg("%s - port %d, %d byte(s)", __func__, port->number, count);
 
-	if (!port->open_count) {
+	if (!port->port.count) {
 		dbg ("%s - port not opened", __func__);
 		return;
 	}
@@ -245,17 +245,17 @@ static void usb_console_write(struct console *co, const char *buf, unsigned coun
 		}
 		/* pass on to the driver specific version of this function if it is available */
 		if (serial->type->write)
-			retval = serial->type->write(port, buf, i);
+			retval = serial->type->write(NULL, port, buf, i);
 		else
-			retval = usb_serial_generic_write(port, buf, i);
+			retval = usb_serial_generic_write(NULL, port, buf, i);
 		dbg("%s - return value : %d", __func__, retval);
 		if (lf) {
 			/* append CR after LF */
 			unsigned char cr = 13;
 			if (serial->type->write)
-				retval = serial->type->write(port, &cr, 1);
+				retval = serial->type->write(NULL, port, &cr, 1);
 			else
-				retval = usb_serial_generic_write(port, &cr, 1);
+				retval = usb_serial_generic_write(NULL, port, &cr, 1);
 			dbg("%s - return value : %d", __func__, retval);
 		}
 		buf += i;
@@ -306,8 +306,8 @@ void usb_serial_console_exit (void)
 {
 	if (usbcons_info.port) {
 		unregister_console(&usbcons);
-		if (usbcons_info.port->open_count)
-			usbcons_info.port->open_count--;
+		if (usbcons_info.port->port.count)
+			usbcons_info.port->port.count--;
 		usbcons_info.port = NULL;
 	}
 }
diff --git a/drivers/usb/serial/cp2101.c b/drivers/usb/serial/cp2101.c
index 2bc5576c443a..46c33fc9f6ce 100644
--- a/drivers/usb/serial/cp2101.c
+++ b/drivers/usb/serial/cp2101.c
@@ -37,15 +37,18 @@
 /*
  * Function Prototypes
  */
-static int cp2101_open(struct usb_serial_port*, struct file*);
-static void cp2101_cleanup(struct usb_serial_port*);
-static void cp2101_close(struct usb_serial_port*, struct file*);
-static void cp2101_get_termios(struct usb_serial_port*);
-static void cp2101_set_termios(struct usb_serial_port*, struct ktermios*);
-static int cp2101_tiocmget (struct usb_serial_port *, struct file *);
-static int cp2101_tiocmset (struct usb_serial_port *, struct file *,
+static int cp2101_open(struct tty_struct *, struct usb_serial_port *,
+							struct file *);
+static void cp2101_cleanup(struct usb_serial_port *);
+static void cp2101_close(struct tty_struct *, struct usb_serial_port *,
+							struct file*);
+static void cp2101_get_termios(struct tty_struct *);
+static void cp2101_set_termios(struct tty_struct *, struct usb_serial_port *,
+							struct ktermios*);
+static int cp2101_tiocmget (struct tty_struct *, struct file *);
+static int cp2101_tiocmset (struct tty_struct *, struct file *,
 		unsigned int, unsigned int);
-static void cp2101_break_ctl(struct usb_serial_port*, int);
+static void cp2101_break_ctl(struct tty_struct *, int);
 static int cp2101_startup (struct usb_serial *);
 static void cp2101_shutdown(struct usb_serial*);
 
@@ -182,7 +185,7 @@ static struct usb_serial_driver cp2101_device = {
  * 'data' is a pointer to a pre-allocated array of integers large
  * enough to hold 'size' bytes (with 4 bytes to each integer)
  */
-static int cp2101_get_config(struct usb_serial_port* port, u8 request,
+static int cp2101_get_config(struct usb_serial_port *port, u8 request,
 		unsigned int *data, int size)
 {
 	struct usb_serial *serial = port->serial;
@@ -228,7 +231,7 @@ static int cp2101_get_config(struct usb_serial_port* port, u8 request,
  * Values less than 16 bits wide are sent directly
  * 'size' is specified in bytes.
  */
-static int cp2101_set_config(struct usb_serial_port* port, u8 request,
+static int cp2101_set_config(struct usb_serial_port *port, u8 request,
 		unsigned int *data, int size)
 {
 	struct usb_serial *serial = port->serial;
@@ -283,13 +286,14 @@ static int cp2101_set_config(struct usb_serial_port* port, u8 request,
  * Convenience function for calling cp2101_set_config on single data values
  * without requiring an integer pointer
  */
-static inline int cp2101_set_config_single(struct usb_serial_port* port,
+static inline int cp2101_set_config_single(struct usb_serial_port *port,
 		u8 request, unsigned int data)
 {
 	return cp2101_set_config(port, request, &data, 2);
 }
 
-static int cp2101_open (struct usb_serial_port *port, struct file *filp)
+static int cp2101_open (struct tty_struct *tty, struct usb_serial_port *port,
+				struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	int result;
@@ -318,10 +322,10 @@ static int cp2101_open (struct usb_serial_port *port, struct file *filp)
 	}
 
 	/* Configure the termios structure */
-	cp2101_get_termios(port);
+	cp2101_get_termios(tty);
 
 	/* Set the DTR and RTS pins low */
-	cp2101_tiocmset(port, NULL, TIOCM_DTR | TIOCM_RTS, 0);
+	cp2101_tiocmset(tty, NULL, TIOCM_DTR | TIOCM_RTS, 0);
 
 	return 0;
 }
@@ -341,7 +345,8 @@ static void cp2101_cleanup (struct usb_serial_port *port)
 	}
 }
 
-static void cp2101_close (struct usb_serial_port *port, struct file * filp)
+static void cp2101_close(struct tty_struct *tty, struct usb_serial_port *port,
+					struct file * filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 
@@ -362,19 +367,15 @@ static void cp2101_close (struct usb_serial_port *port, struct file * filp)
  * from the device, corrects any unsupported values, and configures the
  * termios structure to reflect the state of the device
  */
-static void cp2101_get_termios (struct usb_serial_port *port)
+static void cp2101_get_termios (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned int cflag, modem_ctl[4];
 	unsigned int baud;
 	unsigned int bits;
 
 	dbg("%s - port %d", __func__, port->number);
 
-	if (!port->tty || !port->tty->termios) {
-		dbg("%s - no tty structures", __func__);
-		return;
-	}
-
 	cp2101_get_config(port, CP2101_BAUDRATE, &baud, 2);
 	/* Convert to baudrate */
 	if (baud)
@@ -382,8 +383,8 @@ static void cp2101_get_termios (struct usb_serial_port *port)
 
 	dbg("%s - baud rate = %d", __func__, baud);
 
-	tty_encode_baud_rate(port->tty, baud, baud);
-	cflag = port->tty->termios->c_cflag;
+	tty_encode_baud_rate(tty, baud, baud);
+	cflag = tty->termios->c_cflag;
 
 	cp2101_get_config(port, CP2101_BITS, &bits, 2);
 	cflag &= ~CSIZE;
@@ -491,11 +492,11 @@ static void cp2101_get_termios (struct usb_serial_port *port)
 		cflag &= ~CRTSCTS;
 	}
 
-	port->tty->termios->c_cflag = cflag;
+	tty->termios->c_cflag = cflag;
 }
 
-static void cp2101_set_termios (struct usb_serial_port *port,
-		struct ktermios *old_termios)
+static void cp2101_set_termios (struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	unsigned int cflag, old_cflag;
 	unsigned int baud = 0, bits;
@@ -503,15 +504,13 @@ static void cp2101_set_termios (struct usb_serial_port *port,
 
 	dbg("%s - port %d", __func__, port->number);
 
-	if (!port->tty || !port->tty->termios) {
-		dbg("%s - no tty structures", __func__);
+	if (!tty)
 		return;
-	}
-	port->tty->termios->c_cflag &= ~CMSPAR;
 
-	cflag = port->tty->termios->c_cflag;
+	tty->termios->c_cflag &= ~CMSPAR;
+	cflag = tty->termios->c_cflag;
 	old_cflag = old_termios->c_cflag;
-	baud = tty_get_baud_rate(port->tty);
+	baud = tty_get_baud_rate(tty);
 
 	/* If the baud rate is to be updated*/
 	if (baud != tty_termios_baud_rate(old_termios)) {
@@ -554,7 +553,7 @@ static void cp2101_set_termios (struct usb_serial_port *port,
 		}
 	}
 	/* Report back the resulting baud rate */
-	tty_encode_baud_rate(port->tty, baud, baud);
+	tty_encode_baud_rate(tty, baud, baud);
 
 	/* If the number of data bits is to be updated */
 	if ((cflag & CSIZE) != (old_cflag & CSIZE)) {
@@ -651,9 +650,10 @@ static void cp2101_set_termios (struct usb_serial_port *port,
 
 }
 
-static int cp2101_tiocmset (struct usb_serial_port *port, struct file *file,
+static int cp2101_tiocmset (struct tty_struct *tty, struct file *file,
 		unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned int control = 0;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -681,8 +681,9 @@ static int cp2101_tiocmset (struct usb_serial_port *port, struct file *file,
 
 }
 
-static int cp2101_tiocmget (struct usb_serial_port *port, struct file *file)
+static int cp2101_tiocmget (struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned int control;
 	int result;
 
@@ -702,8 +703,9 @@ static int cp2101_tiocmget (struct usb_serial_port *port, struct file *file)
 	return result;
 }
 
-static void cp2101_break_ctl (struct usb_serial_port *port, int break_state)
+static void cp2101_break_ctl (struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned int state;
 
 	dbg("%s - port %d", __func__, port->number);
diff --git a/drivers/usb/serial/cyberjack.c b/drivers/usb/serial/cyberjack.c
index c164e2cf2752..546178ea6f2d 100644
--- a/drivers/usb/serial/cyberjack.c
+++ b/drivers/usb/serial/cyberjack.c
@@ -57,15 +57,18 @@ static int debug;
 #define CYBERJACK_PRODUCT_ID	0x0100
 
 /* Function prototypes */
-static int cyberjack_startup (struct usb_serial *serial);
-static void cyberjack_shutdown (struct usb_serial *serial);
-static int  cyberjack_open (struct usb_serial_port *port, struct file *filp);
-static void cyberjack_close (struct usb_serial_port *port, struct file *filp);
-static int cyberjack_write (struct usb_serial_port *port, const unsigned char *buf, int count);
-static int cyberjack_write_room( struct usb_serial_port *port );
-static void cyberjack_read_int_callback (struct urb *urb);
-static void cyberjack_read_bulk_callback (struct urb *urb);
-static void cyberjack_write_bulk_callback (struct urb *urb);
+static int cyberjack_startup(struct usb_serial *serial);
+static void cyberjack_shutdown(struct usb_serial *serial);
+static int  cyberjack_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+static void cyberjack_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+static int cyberjack_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count);
+static int cyberjack_write_room( struct tty_struct *tty);
+static void cyberjack_read_int_callback(struct urb *urb);
+static void cyberjack_read_bulk_callback(struct urb *urb);
+static void cyberjack_write_bulk_callback(struct urb *urb);
 
 static struct usb_device_id id_table [] = {
 	{ USB_DEVICE(CYBERJACK_VENDOR_ID, CYBERJACK_PRODUCT_ID) },
@@ -111,7 +114,7 @@ struct cyberjack_private {
 };
 
 /* do some startup allocations not currently performed by usb_serial_probe() */
-static int cyberjack_startup (struct usb_serial *serial)
+static int cyberjack_startup(struct usb_serial *serial)
 {
 	struct cyberjack_private *priv;
 	int i;
@@ -145,7 +148,7 @@ static int cyberjack_startup (struct usb_serial *serial)
 	return( 0 );
 }
 
-static void cyberjack_shutdown (struct usb_serial *serial)
+static void cyberjack_shutdown(struct usb_serial *serial)
 {
 	int i;
 	
@@ -159,7 +162,8 @@ static void cyberjack_shutdown (struct usb_serial *serial)
 	}
 }
 	
-static int  cyberjack_open (struct usb_serial_port *port, struct file *filp)
+static int  cyberjack_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct cyberjack_private *priv;
 	unsigned long flags;
@@ -174,7 +178,8 @@ static int  cyberjack_open (struct usb_serial_port *port, struct file *filp)
 	 * the data through, otherwise it is scheduled, and with high
 	 * data rates (like with OHCI) data can get lost.
 	 */
-	port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	priv = usb_get_serial_port_data(port);
 	spin_lock_irqsave(&priv->lock, flags);
@@ -186,7 +191,8 @@ static int  cyberjack_open (struct usb_serial_port *port, struct file *filp)
 	return result;
 }
 
-static void cyberjack_close (struct usb_serial_port *port, struct file *filp)
+static void cyberjack_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 
@@ -197,7 +203,8 @@ static void cyberjack_close (struct usb_serial_port *port, struct file *filp)
 	}
 }
 
-static int cyberjack_write (struct usb_serial_port *port, const unsigned char *buf, int count)
+static int cyberjack_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count)
 {
 	struct usb_serial *serial = port->serial;
 	struct cyberjack_private *priv = usb_get_serial_port_data(port);
@@ -292,13 +299,13 @@ static int cyberjack_write (struct usb_serial_port *port, const unsigned char *b
 	return (count);
 } 
 
-static int cyberjack_write_room( struct usb_serial_port *port )
+static int cyberjack_write_room(struct tty_struct *tty)
 {
 	/* FIXME: .... */
 	return CYBERJACK_LOCAL_BUF_SIZE;
 }
 
-static void cyberjack_read_int_callback( struct urb *urb )
+static void cyberjack_read_int_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	struct cyberjack_private *priv = usb_get_serial_port_data(port);
@@ -355,7 +362,7 @@ resubmit:
 	dbg("%s - usb_submit_urb(int urb)", __func__);
 }
 
-static void cyberjack_read_bulk_callback (struct urb *urb)
+static void cyberjack_read_bulk_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	struct cyberjack_private *priv = usb_get_serial_port_data(port);
@@ -374,7 +381,7 @@ static void cyberjack_read_bulk_callback (struct urb *urb)
 		return;
 	}
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (!tty) {
 		dbg("%s - ignoring since device not open\n", __func__);
 		return;
@@ -407,7 +414,7 @@ static void cyberjack_read_bulk_callback (struct urb *urb)
 	}
 }
 
-static void cyberjack_write_bulk_callback (struct urb *urb)
+static void cyberjack_write_bulk_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	struct cyberjack_private *priv = usb_get_serial_port_data(port);
diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c
index 0230d3c0888a..6999d3372d85 100644
--- a/drivers/usb/serial/cypress_m8.c
+++ b/drivers/usb/serial/cypress_m8.c
@@ -167,18 +167,18 @@ static int  cypress_earthmate_startup	(struct usb_serial *serial);
 static int  cypress_hidcom_startup	(struct usb_serial *serial);
 static int  cypress_ca42v2_startup	(struct usb_serial *serial);
 static void cypress_shutdown		(struct usb_serial *serial);
-static int  cypress_open		(struct usb_serial_port *port, struct file *filp);
-static void cypress_close		(struct usb_serial_port *port, struct file *filp);
-static int  cypress_write		(struct usb_serial_port *port, const unsigned char *buf, int count);
+static int  cypress_open		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void cypress_close		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static int  cypress_write		(struct tty_struct *tty, struct usb_serial_port *port, const unsigned char *buf, int count);
 static void cypress_send		(struct usb_serial_port *port);
-static int  cypress_write_room		(struct usb_serial_port *port);
-static int  cypress_ioctl		(struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg);
-static void cypress_set_termios		(struct usb_serial_port *port, struct ktermios * old);
-static int  cypress_tiocmget		(struct usb_serial_port *port, struct file *file);
-static int  cypress_tiocmset		(struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear);
-static int  cypress_chars_in_buffer	(struct usb_serial_port *port);
-static void cypress_throttle		(struct usb_serial_port *port);
-static void cypress_unthrottle		(struct usb_serial_port *port);
+static int  cypress_write_room		(struct tty_struct *tty);
+static int  cypress_ioctl		(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg);
+static void cypress_set_termios		(struct tty_struct *tty, struct usb_serial_port *port, struct ktermios * old);
+static int  cypress_tiocmget		(struct tty_struct *tty, struct file *file);
+static int  cypress_tiocmset		(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear);
+static int  cypress_chars_in_buffer	(struct tty_struct *tty);
+static void cypress_throttle		(struct tty_struct *tty);
+static void cypress_unthrottle		(struct tty_struct *tty);
 static void cypress_set_dead		(struct usb_serial_port *port);
 static void cypress_read_int_callback	(struct urb *urb);
 static void cypress_write_int_callback	(struct urb *urb);
@@ -322,8 +322,10 @@ static int analyze_baud_rate(struct usb_serial_port *port, speed_t new_rate)
 
 
 /* This function can either set or retrieve the current serial line settings */
-static int cypress_serial_control (struct usb_serial_port *port, speed_t baud_rate, int data_bits, int stop_bits,
-				   int parity_enable, int parity_type, int reset, int cypress_request_type)
+static int cypress_serial_control (struct tty_struct *tty,
+	struct usb_serial_port *port, speed_t baud_rate, int data_bits,
+	int stop_bits, int parity_enable, int parity_type, int reset,
+	int cypress_request_type)
 {
 	int new_baudrate = 0, retval = 0, tries = 0;
 	struct cypress_private *priv;
@@ -395,7 +397,7 @@ static int cypress_serial_control (struct usb_serial_port *port, speed_t baud_ra
 				spin_unlock_irqrestore(&priv->lock, flags);
 				/* If we asked for a speed change encode it */
 				if (baud_rate)
-					tty_encode_baud_rate(port->tty,
+					tty_encode_baud_rate(tty,
 							new_baudrate, new_baudrate);
 			}
 		break;
@@ -611,7 +613,8 @@ static void cypress_shutdown (struct usb_serial *serial)
 }
 
 
-static int cypress_open (struct usb_serial_port *port, struct file *filp)
+static int cypress_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	struct usb_serial *serial = port->serial;
@@ -636,14 +639,15 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp)
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	/* setting to zero could cause data loss */
-	port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	/* raise both lines and set termios */
 	spin_lock_irqsave(&priv->lock, flags);
 	priv->line_control = CONTROL_DTR | CONTROL_RTS;
 	priv->cmd_ctrl = 1;
 	spin_unlock_irqrestore(&priv->lock, flags);
-	result = cypress_write(port, NULL, 0);
+	result = cypress_write(tty, port, NULL, 0);
 
 	if (result) {
 		dev_err(&port->dev, "%s - failed setting the control lines - error %d\n", __func__, result);
@@ -651,7 +655,8 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp)
 	} else
 		dbg("%s - success setting the control lines", __func__);
 
-	cypress_set_termios(port, &priv->tmp_termios);
+	if (tty)
+		cypress_set_termios(tty, port, &priv->tmp_termios);
 
 	/* setup the port and start reading from the device */
 	if(!port->interrupt_in_urb){
@@ -674,7 +679,8 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp)
 } /* cypress_open */
 
 
-static void cypress_close(struct usb_serial_port *port, struct file * filp)
+static void cypress_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	unsigned int c_cflag;
@@ -688,7 +694,7 @@ static void cypress_close(struct usb_serial_port *port, struct file * filp)
 	spin_lock_irq(&priv->lock);
 	timeout = CYPRESS_CLOSING_WAIT;
 	init_waitqueue_entry(&wait, current);
-	add_wait_queue(&port->tty->write_wait, &wait);
+	add_wait_queue(&tty->write_wait, &wait);
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (cypress_buf_data_avail(priv->buf) == 0
@@ -701,7 +707,7 @@ static void cypress_close(struct usb_serial_port *port, struct file * filp)
 		spin_lock_irq(&priv->lock);
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->tty->write_wait, &wait);
+	remove_wait_queue(&tty->write_wait, &wait);
 	/* clear out any remaining data in the buffer */
 	cypress_buf_clear(priv->buf);
 	spin_unlock_irq(&priv->lock);
@@ -713,19 +719,21 @@ static void cypress_close(struct usb_serial_port *port, struct file * filp)
 		return;
 	}
 	/* wait for characters to drain from device */
-	bps = tty_get_baud_rate(port->tty);
-	if (bps > 1200)
-		timeout = max((HZ*2560)/bps,HZ/10);
-	else
-		timeout = 2*HZ;
-	schedule_timeout_interruptible(timeout);
+	if (tty) {
+		bps = tty_get_baud_rate(tty);
+		if (bps > 1200)
+			timeout = max((HZ*2560)/bps,HZ/10);
+		else
+			timeout = 2*HZ;
+		schedule_timeout_interruptible(timeout);
+	}
 
 	dbg("%s - stopping urbs", __func__);
 	usb_kill_urb (port->interrupt_in_urb);
 	usb_kill_urb (port->interrupt_out_urb);
 
-	if (port->tty) {
-		c_cflag = port->tty->termios->c_cflag;
+	if (tty) {
+		c_cflag = tty->termios->c_cflag;
 		if (c_cflag & HUPCL) {
 			/* drop dtr and rts */
 			priv = usb_get_serial_port_data(port);
@@ -733,7 +741,7 @@ static void cypress_close(struct usb_serial_port *port, struct file * filp)
 			priv->line_control = 0;
 			priv->cmd_ctrl = 1;
 			spin_unlock_irq(&priv->lock);
-			cypress_write(port, NULL, 0);
+			cypress_write(tty, port, NULL, 0);
 		}
 	}
 
@@ -744,7 +752,8 @@ static void cypress_close(struct usb_serial_port *port, struct file * filp)
 } /* cypress_close */
 
 
-static int cypress_write(struct usb_serial_port *port, const unsigned char *buf, int count)
+static int cypress_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -878,8 +887,9 @@ send:
 
 
 /* returns how much space is available in the soft buffer */
-static int cypress_write_room(struct usb_serial_port *port)
+static int cypress_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	int room = 0;
 	unsigned long flags;
@@ -895,8 +905,9 @@ static int cypress_write_room(struct usb_serial_port *port)
 }
 
 
-static int cypress_tiocmget (struct usb_serial_port *port, struct file *file)
+static int cypress_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	__u8 status, control;
 	unsigned int result = 0;
@@ -922,9 +933,10 @@ static int cypress_tiocmget (struct usb_serial_port *port, struct file *file)
 }
 
 
-static int cypress_tiocmset (struct usb_serial_port *port, struct file *file,
+static int cypress_tiocmset(struct tty_struct *tty, struct file *file,
 			       unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	
@@ -942,12 +954,14 @@ static int cypress_tiocmset (struct usb_serial_port *port, struct file *file,
 	priv->cmd_ctrl = 1;
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	return cypress_write(port, NULL, 0);
+	return cypress_write(tty, port, NULL, 0);
 }
 
 
-static int cypress_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg)
+static int cypress_ioctl(struct tty_struct *tty, struct file * file,
+					unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 
 	dbg("%s - port %d, cmd 0x%.4x", __func__, port->number, cmd);
@@ -983,22 +997,18 @@ static int cypress_ioctl (struct usb_serial_port *port, struct file * file, unsi
 				}
 			}
 			return 0;
-			break;
 		default:
 			break;
 	}
-
 	dbg("%s - arg not supported - it was 0x%04x - check include/asm/ioctls.h", __func__, cmd);
-
 	return -ENOIOCTLCMD;
 } /* cypress_ioctl */
 
 
-static void cypress_set_termios (struct usb_serial_port *port,
-		struct ktermios *old_termios)
+static void cypress_set_termios(struct tty_struct *tty,
+	struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct cypress_private *priv = usb_get_serial_port_data(port);
-	struct tty_struct *tty;
 	int data_bits, stop_bits, parity_type, parity_enable;
 	unsigned cflag, iflag;
 	unsigned long flags;
@@ -1007,8 +1017,6 @@ static void cypress_set_termios (struct usb_serial_port *port,
 
 	dbg("%s - port %d", __func__, port->number);
 
-	tty = port->tty;
-
 	spin_lock_irqsave(&priv->lock, flags);
 	if (!priv->termios_initialized) {
 		if (priv->chiptype == CT_EARTHMATE) {
@@ -1096,13 +1104,13 @@ static void cypress_set_termios (struct usb_serial_port *port,
 			"%d data_bits (+5)", __func__, stop_bits,
 			parity_enable, parity_type, data_bits);
 
-	cypress_serial_control(port, tty_get_baud_rate(tty), data_bits, stop_bits,
+	cypress_serial_control(tty, port, tty_get_baud_rate(tty), data_bits, stop_bits,
 			parity_enable, parity_type, 0, CYPRESS_SET_CONFIG);
 
 	/* we perform a CYPRESS_GET_CONFIG so that the current settings are
 	 * filled into the private structure this should confirm that all is
 	 * working if it returns what we just set */
-	cypress_serial_control(port, 0, 0, 0, 0, 0, 0, CYPRESS_GET_CONFIG);
+	cypress_serial_control(tty, port, 0, 0, 0, 0, 0, 0, CYPRESS_GET_CONFIG);
 
 	/* Here we can define custom tty settings for devices; the main tty
 	 * termios flag base comes from empeg.c */
@@ -1142,14 +1150,15 @@ static void cypress_set_termios (struct usb_serial_port *port,
 	/* if necessary, set lines */
 	if (linechange) {
 		priv->cmd_ctrl = 1;
-		cypress_write(port, NULL, 0);
+		cypress_write(tty, port, NULL, 0);
 	}
 } /* cypress_set_termios */
 
 
 /* returns amount of data still left in soft buffer */
-static int cypress_chars_in_buffer(struct usb_serial_port *port)
+static int cypress_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	int chars = 0;
 	unsigned long flags;
@@ -1165,8 +1174,9 @@ static int cypress_chars_in_buffer(struct usb_serial_port *port)
 }
 
 
-static void cypress_throttle (struct usb_serial_port *port)
+static void cypress_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -1178,8 +1188,9 @@ static void cypress_throttle (struct usb_serial_port *port)
 }
 
 
-static void cypress_unthrottle (struct usb_serial_port *port)
+static void cypress_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	int actually_throttled, result;
 	unsigned long flags;
@@ -1251,7 +1262,7 @@ static void cypress_read_int_callback(struct urb *urb)
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (!tty) {
 		dbg("%s - bad tty pointer - exiting", __func__);
 		return;
@@ -1327,7 +1338,7 @@ static void cypress_read_int_callback(struct urb *urb)
 					data[i]);
 			tty_insert_flip_char(tty, data[i], tty_flag);
 		}
-		tty_flip_buffer_push(port->tty);
+		tty_flip_buffer_push(port->port.tty);
 	}
 
 	spin_lock_irqsave(&priv->lock, flags);
@@ -1339,7 +1350,7 @@ continue_read:
 
 	/* Continue trying to always read... unless the port has closed. */
 
-	if (port->open_count > 0 && priv->comm_is_ok) {
+	if (port->port.count > 0 && priv->comm_is_ok) {
 		usb_fill_int_urb(port->interrupt_in_urb, port->serial->dev,
 				usb_rcvintpipe(port->serial->dev,
 					port->interrupt_in_endpointAddress),
diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c
index 28bc6fcf44f0..dc92dbfc962e 100644
--- a/drivers/usb/serial/digi_acceleport.c
+++ b/drivers/usb/serial/digi_acceleport.c
@@ -15,7 +15,7 @@
 *  Al Borchers (borchers@steinerpoint.com)
 * 
 * (12/03/2001) gkh
-*	switched to using port->open_count instead of private version.
+*	switched to using port->port.count instead of private version.
 *	Removed port->active
 *
 * (04/08/2001) gb
@@ -441,22 +441,23 @@ static int digi_set_modem_signals(struct usb_serial_port *port,
 	unsigned int modem_signals, int interruptible);
 static int digi_transmit_idle(struct usb_serial_port *port,
 	unsigned long timeout);
-static void digi_rx_throttle (struct usb_serial_port *port);
-static void digi_rx_unthrottle (struct usb_serial_port *port);
-static void digi_set_termios(struct usb_serial_port *port,
-	struct ktermios *old_termios);
-static void digi_break_ctl(struct usb_serial_port *port, int break_state);
-static int digi_ioctl(struct usb_serial_port *port, struct file *file,
-	unsigned int cmd, unsigned long arg);
-static int digi_tiocmget(struct usb_serial_port *port, struct file *file);
-static int digi_tiocmset(struct usb_serial_port *port, struct file *file,
+static void digi_rx_throttle (struct tty_struct *tty);
+static void digi_rx_unthrottle (struct tty_struct *tty);
+static void digi_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios);
+static void digi_break_ctl(struct tty_struct *tty, int break_state);
+static int digi_tiocmget(struct tty_struct *tty, struct file *file);
+static int digi_tiocmset(struct tty_struct *tty, struct file *file,
 	unsigned int set, unsigned int clear);
-static int digi_write(struct usb_serial_port *port, const unsigned char *buf, int count);
+static int digi_write(struct tty_struct *tty, struct usb_serial_port *port,
+	const unsigned char *buf, int count);
 static void digi_write_bulk_callback(struct urb *urb);
-static int digi_write_room(struct usb_serial_port *port);
-static int digi_chars_in_buffer(struct usb_serial_port *port);
-static int digi_open(struct usb_serial_port *port, struct file *filp);
-static void digi_close(struct usb_serial_port *port, struct file *filp);
+static int digi_write_room(struct tty_struct *tty);
+static int digi_chars_in_buffer(struct tty_struct *tty);
+static int digi_open(struct tty_struct *tty, struct usb_serial_port *port,
+	struct file *filp);
+static void digi_close(struct tty_struct *tty, struct usb_serial_port *port,
+	struct file *filp);
 static int digi_startup_device(struct usb_serial *serial);
 static int digi_startup(struct usb_serial *serial);
 static void digi_shutdown(struct usb_serial *serial);
@@ -516,7 +517,6 @@ static struct usb_serial_driver digi_acceleport_2_device = {
 	.chars_in_buffer =		digi_chars_in_buffer,
 	.throttle =			digi_rx_throttle,
 	.unthrottle =			digi_rx_unthrottle,
-	.ioctl =			digi_ioctl,
 	.set_termios =			digi_set_termios,
 	.break_ctl =			digi_break_ctl,
 	.tiocmget =			digi_tiocmget,
@@ -543,7 +543,6 @@ static struct usb_serial_driver digi_acceleport_4_device = {
 	.chars_in_buffer =		digi_chars_in_buffer,
 	.throttle =			digi_rx_throttle,
 	.unthrottle =			digi_rx_unthrottle,
-	.ioctl =			digi_ioctl,
 	.set_termios =			digi_set_termios,
 	.break_ctl =			digi_break_ctl,
 	.tiocmget =			digi_tiocmget,
@@ -604,7 +603,7 @@ static void digi_wakeup_write_lock(struct work_struct *work)
 
 static void digi_wakeup_write(struct usb_serial_port *port)
 {
-	tty_wakeup(port->tty);
+	tty_wakeup(port->port.tty);
 }
 
 
@@ -856,9 +855,10 @@ static int digi_transmit_idle(struct usb_serial_port *port,
 }
 
 
-static void digi_rx_throttle(struct usb_serial_port *port)
+static void digi_rx_throttle(struct tty_struct *tty)
 {
 	unsigned long flags;
+	struct usb_serial_port *port = tty->driver_data;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 
 
@@ -872,10 +872,11 @@ static void digi_rx_throttle(struct usb_serial_port *port)
 }
 
 
-static void digi_rx_unthrottle(struct usb_serial_port *port)
+static void digi_rx_unthrottle(struct tty_struct *tty)
 {
 	int ret = 0;
 	unsigned long flags;
+	struct usb_serial_port *port = tty->driver_data;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 
 	dbg("digi_rx_unthrottle: TOP: port=%d", priv->dp_port_num);
@@ -900,12 +901,10 @@ static void digi_rx_unthrottle(struct usb_serial_port *port)
 }
 
 
-static void digi_set_termios(struct usb_serial_port *port,
-					struct ktermios *old_termios)
+static void digi_set_termios(struct tty_struct *tty, 
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
-
 	struct digi_port *priv = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	unsigned int iflag = tty->termios->c_iflag;
 	unsigned int cflag = tty->termios->c_cflag;
 	unsigned int old_iflag = old_termios->c_iflag;
@@ -1088,8 +1087,9 @@ static void digi_set_termios(struct usb_serial_port *port,
 }
 
 
-static void digi_break_ctl(struct usb_serial_port *port, int break_state)
+static void digi_break_ctl(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned char buf[4];
 
 	buf[0] = DIGI_CMD_BREAK_CONTROL;
@@ -1100,8 +1100,9 @@ static void digi_break_ctl(struct usb_serial_port *port, int break_state)
 }
 
 
-static int digi_tiocmget(struct usb_serial_port *port, struct file *file)
+static int digi_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 	unsigned int val;
 	unsigned long flags;
@@ -1115,9 +1116,10 @@ static int digi_tiocmget(struct usb_serial_port *port, struct file *file)
 }
 
 
-static int digi_tiocmset(struct usb_serial_port *port, struct file *file,
+static int digi_tiocmset(struct tty_struct *tty, struct file *file,
 	unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 	unsigned int val;
 	unsigned long flags;
@@ -1131,27 +1133,8 @@ static int digi_tiocmset(struct usb_serial_port *port, struct file *file,
 }
 
 
-static int digi_ioctl(struct usb_serial_port *port, struct file *file,
-	unsigned int cmd, unsigned long arg)
-{
-	struct digi_port *priv = usb_get_serial_port_data(port);
-	dbg("digi_ioctl: TOP: port=%d, cmd=0x%x", priv->dp_port_num, cmd);
-
-	switch (cmd) {
-	case TIOCMIWAIT:
-		/* wait for any of the 4 modem inputs (DCD,RI,DSR,CTS)*/
-		/* TODO */
-		return 0;
-	case TIOCGICOUNT:
-		/* return count of modemline transitions */
-		/* TODO */
-		return 0;
-	}
-	return -ENOIOCTLCMD;
-
-}
-
-static int digi_write(struct usb_serial_port *port, const unsigned char *buf, int count)
+static int digi_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 
 	int ret,data_len,new_len;
@@ -1261,7 +1244,7 @@ static void digi_write_bulk_callback(struct urb *urb)
 	/* try to send any buffered data on this port, if it is open */
 	spin_lock(&priv->dp_port_lock);
 	priv->dp_write_urb_in_use = 0;
-	if (port->open_count && port->write_urb->status != -EINPROGRESS
+	if (port->port.count && port->write_urb->status != -EINPROGRESS
 	    && priv->dp_out_buf_len > 0) {
 		*((unsigned char *)(port->write_urb->transfer_buffer))
 			= (unsigned char)DIGI_CMD_SEND_DATA;
@@ -1288,11 +1271,11 @@ static void digi_write_bulk_callback(struct urb *urb)
 			__func__, ret, priv->dp_port_num);
 }
 
-static int digi_write_room(struct usb_serial_port *port)
+static int digi_write_room(struct tty_struct *tty)
 {
-
-	int room;
+	struct usb_serial_port *port = tty->driver_data;
 	struct digi_port *priv = usb_get_serial_port_data(port);
+	int room;
 	unsigned long flags = 0;
 
 	spin_lock_irqsave(&priv->dp_port_lock, flags);
@@ -1308,12 +1291,11 @@ static int digi_write_room(struct usb_serial_port *port)
 
 }
 
-static int digi_chars_in_buffer(struct usb_serial_port *port)
+static int digi_chars_in_buffer(struct tty_struct *tty)
 {
-
+	struct usb_serial_port *port = tty->driver_data;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 
-
 	if (port->write_urb->status == -EINPROGRESS
 	    || priv->dp_write_urb_in_use) {
 		dbg("digi_chars_in_buffer: port=%d, chars=%d",
@@ -1329,7 +1311,8 @@ static int digi_chars_in_buffer(struct usb_serial_port *port)
 }
 
 
-static int digi_open(struct usb_serial_port *port, struct file *filp)
+static int digi_open(struct tty_struct *tty, struct usb_serial_port *port,
+				struct file *filp)
 {
 	int ret;
 	unsigned char buf[32];
@@ -1338,7 +1321,7 @@ static int digi_open(struct usb_serial_port *port, struct file *filp)
 	unsigned long flags = 0;
 
 	dbg("digi_open: TOP: port=%d, open_count=%d",
-		priv->dp_port_num, port->open_count);
+		priv->dp_port_num, port->port.count);
 
 	/* be sure the device is started up */
 	if (digi_startup_device(port->serial) != 0)
@@ -1380,9 +1363,11 @@ static int digi_open(struct usb_serial_port *port, struct file *filp)
 		dbg("digi_open: write oob failed, ret=%d", ret);
 
 	/* set termios settings */
-	not_termios.c_cflag = ~port->tty->termios->c_cflag;
-	not_termios.c_iflag = ~port->tty->termios->c_iflag;
-	digi_set_termios(port, &not_termios);
+	if (tty) {
+		not_termios.c_cflag = ~tty->termios->c_cflag;
+		not_termios.c_iflag = ~tty->termios->c_iflag;
+		digi_set_termios(tty, port, &not_termios);
+	}
 
 	/* set DTR and RTS */
 	digi_set_modem_signals(port, TIOCM_DTR|TIOCM_RTS, 1);
@@ -1391,16 +1376,16 @@ static int digi_open(struct usb_serial_port *port, struct file *filp)
 }
 
 
-static void digi_close(struct usb_serial_port *port, struct file *filp)
+static void digi_close(struct tty_struct *tty, struct usb_serial_port *port,
+				struct file *filp)
 {
 	DEFINE_WAIT(wait);
 	int ret;
 	unsigned char buf[32];
-	struct tty_struct *tty = port->tty;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 
 	dbg("digi_close: TOP: port=%d, open_count=%d",
-		priv->dp_port_num, port->open_count);
+		priv->dp_port_num, port->port.count);
 
 	mutex_lock(&port->serial->disc_mutex);
 	/* if disconnected, just clear flags */
@@ -1663,7 +1648,7 @@ static int digi_read_inb_callback(struct urb *urb)
 {
 
 	struct usb_serial_port *port = urb->context;
-	struct tty_struct *tty = port->tty;
+	struct tty_struct *tty = port->port.tty;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 	int opcode = ((unsigned char *)urb->transfer_buffer)[0];
 	int len = ((unsigned char *)urb->transfer_buffer)[1];
@@ -1675,7 +1660,7 @@ static int digi_read_inb_callback(struct urb *urb)
 
 	/* do not process callbacks on closed ports */
 	/* but do continue the read chain */
-	if (port->open_count == 0)
+	if (port->port.count == 0)
 		return 0;
 
 	/* short/multiple packet check */
@@ -1785,17 +1770,17 @@ static int digi_read_oob_callback(struct urb *urb)
 			if (val & DIGI_READ_INPUT_SIGNALS_CTS) {
 				priv->dp_modem_signals |= TIOCM_CTS;
 				/* port must be open to use tty struct */
-				if (port->open_count
-					&& port->tty->termios->c_cflag & CRTSCTS) {
-					port->tty->hw_stopped = 0;
+				if (port->port.count
+					&& port->port.tty->termios->c_cflag & CRTSCTS) {
+					port->port.tty->hw_stopped = 0;
 					digi_wakeup_write(port);
 				}
 			} else {
 				priv->dp_modem_signals &= ~TIOCM_CTS;
 				/* port must be open to use tty struct */
-				if (port->open_count
-					&& port->tty->termios->c_cflag & CRTSCTS) {
-					port->tty->hw_stopped = 1;
+				if (port->port.count
+					&& port->port.tty->termios->c_cflag & CRTSCTS) {
+					port->port.tty->hw_stopped = 1;
 				}
 			}
 			if (val & DIGI_READ_INPUT_SIGNALS_DSR)
diff --git a/drivers/usb/serial/empeg.c b/drivers/usb/serial/empeg.c
index c5ec309a3cb1..47ebdf5fad8e 100644
--- a/drivers/usb/serial/empeg.c
+++ b/drivers/usb/serial/empeg.c
@@ -31,7 +31,7 @@
  *	Moved MOD_DEC_USE_COUNT to end of empeg_close().
  * 
  * (12/03/2000) gb
- *	Added port->tty->ldisc.set_termios(port->tty, NULL) to empeg_open()
+ *	Added port->port.tty->ldisc.set_termios(port->port.tty, NULL) to empeg_open()
  *	This notifies the tty driver that the termios have changed.
  * 
  * (11/13/2000) gb
@@ -77,22 +77,18 @@ static int debug;
 #define EMPEG_PRODUCT_ID		0x0001
 
 /* function prototypes for an empeg-car player */
-static int  empeg_open			(struct usb_serial_port *port, struct file *filp);
-static void empeg_close			(struct usb_serial_port *port, struct file *filp);
-static int  empeg_write			(struct usb_serial_port *port,
+static int  empeg_open			(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void empeg_close			(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static int  empeg_write			(struct tty_struct *tty, struct usb_serial_port *port,
 					const unsigned char *buf,
 					int count);
-static int  empeg_write_room		(struct usb_serial_port *port);
-static int  empeg_chars_in_buffer	(struct usb_serial_port *port);
-static void empeg_throttle		(struct usb_serial_port *port);
-static void empeg_unthrottle		(struct usb_serial_port *port);
+static int  empeg_write_room		(struct tty_struct *tty);
+static int  empeg_chars_in_buffer	(struct tty_struct *tty);
+static void empeg_throttle		(struct tty_struct *tty);
+static void empeg_unthrottle		(struct tty_struct *tty);
 static int  empeg_startup		(struct usb_serial *serial);
 static void empeg_shutdown		(struct usb_serial *serial);
-static int  empeg_ioctl			(struct usb_serial_port *port,
-					struct file * file,
-					unsigned int cmd,
-					unsigned long arg);
-static void empeg_set_termios		(struct usb_serial_port *port, struct ktermios *old_termios);
+static void empeg_set_termios		(struct tty_struct *tty, struct usb_serial_port *port, struct ktermios *old_termios);
 static void empeg_write_bulk_callback	(struct urb *urb);
 static void empeg_read_bulk_callback	(struct urb *urb);
 
@@ -125,7 +121,6 @@ static struct usb_serial_driver empeg_device = {
 	.unthrottle =		empeg_unthrottle,
 	.attach =		empeg_startup,
 	.shutdown =		empeg_shutdown,
-	.ioctl =		empeg_ioctl,
 	.set_termios =		empeg_set_termios,
 	.write =		empeg_write,
 	.write_room =		empeg_write_room,
@@ -145,7 +140,8 @@ static int		bytes_out;
 /******************************************************************************
  * Empeg specific driver functions
  ******************************************************************************/
-static int empeg_open (struct usb_serial_port *port, struct file *filp)
+static int empeg_open(struct tty_struct *tty, struct usb_serial_port *port,
+				struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	int result = 0;
@@ -153,7 +149,7 @@ static int empeg_open (struct usb_serial_port *port, struct file *filp)
 	dbg("%s - port %d", __func__, port->number);
 
 	/* Force default termio settings */
-	empeg_set_termios (port, NULL) ;
+	empeg_set_termios (tty, port, NULL) ;
 
 	bytes_in = 0;
 	bytes_out = 0;
@@ -178,7 +174,8 @@ static int empeg_open (struct usb_serial_port *port, struct file *filp)
 }
 
 
-static void empeg_close (struct usb_serial_port *port, struct file * filp)
+static void empeg_close(struct tty_struct *tty, struct usb_serial_port *port,
+				struct file * filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 
@@ -189,7 +186,7 @@ static void empeg_close (struct usb_serial_port *port, struct file * filp)
 }
 
 
-static int empeg_write (struct usb_serial_port *port, const unsigned char *buf, int count)
+static int empeg_write(struct tty_struct *tty, struct usb_serial_port *port, const unsigned char *buf, int count)
 {
 	struct usb_serial *serial = port->serial;
 	struct urb *urb;
@@ -203,7 +200,6 @@ static int empeg_write (struct usb_serial_port *port, const unsigned char *buf,
 	dbg("%s - port %d", __func__, port->number);
 
 	while (count > 0) {
-
 		/* try to find a free urb in our list of them */
 		urb = NULL;
 
@@ -262,15 +258,14 @@ static int empeg_write (struct usb_serial_port *port, const unsigned char *buf,
 		bytes_out += transfer_size;
 
 	}
-
 exit:
 	return bytes_sent;
-
 } 
 
 
-static int empeg_write_room (struct usb_serial_port *port)
+static int empeg_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned long flags;
 	int i;
 	int room = 0;
@@ -278,25 +273,22 @@ static int empeg_write_room (struct usb_serial_port *port)
 	dbg("%s - port %d", __func__, port->number);
 
 	spin_lock_irqsave (&write_urb_pool_lock, flags);
-
 	/* tally up the number of bytes available */
 	for (i = 0; i < NUM_URBS; ++i) {
 		if (write_urb_pool[i]->status != -EINPROGRESS) {
 			room += URB_TRANSFER_BUFFER_SIZE;
 		}
 	} 
-
 	spin_unlock_irqrestore (&write_urb_pool_lock, flags);
-
 	dbg("%s - returns %d", __func__, room);
-
-	return (room);
+	return room;
 
 }
 
 
-static int empeg_chars_in_buffer (struct usb_serial_port *port)
+static int empeg_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned long flags;
 	int i;
 	int chars = 0;
@@ -356,7 +348,7 @@ static void empeg_read_bulk_callback (struct urb *urb)
 
 	usb_serial_debug_data(debug, &port->dev, __func__, urb->actual_length, data);
 
-	tty = port->tty;
+	tty = port->port.tty;
 
 	if (urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
@@ -386,27 +378,24 @@ static void empeg_read_bulk_callback (struct urb *urb)
 }
 
 
-static void empeg_throttle (struct usb_serial_port *port)
+static void empeg_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s - port %d", __func__, port->number);
 	usb_kill_urb(port->read_urb);
 }
 
 
-static void empeg_unthrottle (struct usb_serial_port *port)
+static void empeg_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int result;
-
 	dbg("%s - port %d", __func__, port->number);
 
 	port->read_urb->dev = port->serial->dev;
-
 	result = usb_submit_urb(port->read_urb, GFP_ATOMIC);
-
 	if (result)
 		dev_err(&port->dev, "%s - failed submitting read urb, error %d\n", __func__, result);
-
-	return;
 }
 
 
@@ -436,17 +425,10 @@ static void empeg_shutdown (struct usb_serial *serial)
 }
 
 
-static int empeg_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg)
-{
-	dbg("%s - port %d, cmd 0x%.4x", __func__, port->number, cmd);
-
-	return -ENOIOCTLCMD;
-}
-
-
-static void empeg_set_termios (struct usb_serial_port *port, struct ktermios *old_termios)
+static void empeg_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
-	struct ktermios *termios = port->tty->termios;
+	struct ktermios *termios = tty->termios;
 	dbg("%s - port %d", __func__, port->number);
 
 	/*
@@ -491,8 +473,8 @@ static void empeg_set_termios (struct usb_serial_port *port, struct ktermios *ol
 	 * this is bad as it opens up the possibility of dropping bytes
 	 * on the floor.  We don't want to drop bytes on the floor. :)
 	 */
-	port->tty->low_latency = 1;
-	tty_encode_baud_rate(port->tty, 115200, 115200);
+	tty->low_latency = 1;
+	tty_encode_baud_rate(tty, 115200, 115200);
 }
 
 
diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index 0ff4a3971e45..abbb447e5375 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -682,21 +682,21 @@ static int  ftdi_sio_probe	(struct usb_serial *serial, const struct usb_device_i
 static void ftdi_shutdown		(struct usb_serial *serial);
 static int  ftdi_sio_port_probe	(struct usb_serial_port *port);
 static int  ftdi_sio_port_remove	(struct usb_serial_port *port);
-static int  ftdi_open			(struct usb_serial_port *port, struct file *filp);
-static void ftdi_close			(struct usb_serial_port *port, struct file *filp);
-static int  ftdi_write			(struct usb_serial_port *port, const unsigned char *buf, int count);
-static int  ftdi_write_room		(struct usb_serial_port *port);
-static int  ftdi_chars_in_buffer	(struct usb_serial_port *port);
+static int  ftdi_open			(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void ftdi_close			(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static int  ftdi_write			(struct tty_struct *tty, struct usb_serial_port *port, const unsigned char *buf, int count);
+static int  ftdi_write_room		(struct tty_struct *tty);
+static int  ftdi_chars_in_buffer	(struct tty_struct *tty);
 static void ftdi_write_bulk_callback	(struct urb *urb);
 static void ftdi_read_bulk_callback	(struct urb *urb);
 static void ftdi_process_read		(struct work_struct *work);
-static void ftdi_set_termios		(struct usb_serial_port *port, struct ktermios * old);
-static int  ftdi_tiocmget               (struct usb_serial_port *port, struct file *file);
-static int  ftdi_tiocmset		(struct usb_serial_port *port, struct file * file, unsigned int set, unsigned int clear);
-static int  ftdi_ioctl			(struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg);
-static void ftdi_break_ctl		(struct usb_serial_port *port, int break_state );
-static void ftdi_throttle		(struct usb_serial_port *port);
-static void ftdi_unthrottle		(struct usb_serial_port *port);
+static void ftdi_set_termios		(struct tty_struct *tty, struct usb_serial_port *port, struct ktermios * old);
+static int  ftdi_tiocmget               (struct tty_struct *tty, struct file *file);
+static int  ftdi_tiocmset		(struct tty_struct *tty, struct file * file, unsigned int set, unsigned int clear);
+static int  ftdi_ioctl			(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg);
+static void ftdi_break_ctl		(struct tty_struct *tty, int break_state );
+static void ftdi_throttle		(struct tty_struct *tty);
+static void ftdi_unthrottle		(struct tty_struct *tty);
 
 static unsigned short int ftdi_232am_baud_base_to_divisor (int baud, int base);
 static unsigned short int ftdi_232am_baud_to_divisor (int baud);
@@ -843,42 +843,7 @@ static int update_mctrl(struct usb_serial_port *port, unsigned int set, unsigned
 }
 
 
-static __u32 get_ftdi_divisor(struct usb_serial_port * port);
-
-
-static int change_speed(struct usb_serial_port *port)
-{
-	struct ftdi_private *priv = usb_get_serial_port_data(port);
-	char *buf;
-        __u16 urb_value;
-	__u16 urb_index;
-	__u32 urb_index_value;
-	int rv;
-
-	buf = kmalloc(1, GFP_NOIO);
-	if (!buf)
-		return -ENOMEM;
-
-	urb_index_value = get_ftdi_divisor(port);
-	urb_value = (__u16)urb_index_value;
-	urb_index = (__u16)(urb_index_value >> 16);
-	if (priv->interface) {	/* FT2232C */
-		urb_index = (__u16)((urb_index << 8) | priv->interface);
-	}
-
-	rv = usb_control_msg(port->serial->dev,
-			    usb_sndctrlpipe(port->serial->dev, 0),
-			    FTDI_SIO_SET_BAUDRATE_REQUEST,
-			    FTDI_SIO_SET_BAUDRATE_REQUEST_TYPE,
-			    urb_value, urb_index,
-			    buf, 0, WDR_SHORT_TIMEOUT);
-
-	kfree(buf);
-	return rv;
-}
-
-
-static __u32 get_ftdi_divisor(struct usb_serial_port * port)
+static __u32 get_ftdi_divisor(struct tty_struct *tty, struct usb_serial_port *port)
 { /* get_ftdi_divisor */
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	__u32 div_value = 0;
@@ -910,7 +875,7 @@ static __u32 get_ftdi_divisor(struct usb_serial_port * port)
 
 	/* 1. Get the baud rate from the tty settings, this observes alt_speed hack */
 
-	baud = tty_get_baud_rate(port->tty);
+	baud = tty_get_baud_rate(tty);
 	dbg("%s - tty_get_baud_rate reports speed %d", __func__, baud);
 
 	/* 2. Observe async-compatible custom_divisor hack, update baudrate if needed */
@@ -976,10 +941,42 @@ static __u32 get_ftdi_divisor(struct usb_serial_port * port)
 			ftdi_chip_name[priv->chip_type]);
 	}
 
-	tty_encode_baud_rate(port->tty, baud, baud);
+	tty_encode_baud_rate(tty, baud, baud);
 	return(div_value);
 }
 
+static int change_speed(struct tty_struct *tty, struct usb_serial_port *port)
+{
+	struct ftdi_private *priv = usb_get_serial_port_data(port);
+	char *buf;
+        __u16 urb_value;
+	__u16 urb_index;
+	__u32 urb_index_value;
+	int rv;
+
+	buf = kmalloc(1, GFP_NOIO);
+	if (!buf)
+		return -ENOMEM;
+
+	urb_index_value = get_ftdi_divisor(tty, port);
+	urb_value = (__u16)urb_index_value;
+	urb_index = (__u16)(urb_index_value >> 16);
+	if (priv->interface) {	/* FT2232C */
+		urb_index = (__u16)((urb_index << 8) | priv->interface);
+	}
+
+	rv = usb_control_msg(port->serial->dev,
+			    usb_sndctrlpipe(port->serial->dev, 0),
+			    FTDI_SIO_SET_BAUDRATE_REQUEST,
+			    FTDI_SIO_SET_BAUDRATE_REQUEST_TYPE,
+			    urb_value, urb_index,
+			    buf, 0, WDR_SHORT_TIMEOUT);
+
+	kfree(buf);
+	return rv;
+}
+
+
 
 static int get_serial_info(struct usb_serial_port * port, struct serial_struct __user * retinfo)
 {
@@ -998,7 +995,8 @@ static int get_serial_info(struct usb_serial_port * port, struct serial_struct _
 } /* get_serial_info */
 
 
-static int set_serial_info(struct usb_serial_port * port, struct serial_struct __user * newinfo)
+static int set_serial_info(struct tty_struct *tty,
+	struct usb_serial_port * port, struct serial_struct __user * newinfo)
 { /* set_serial_info */
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	struct serial_struct new_serial;
@@ -1030,30 +1028,29 @@ static int set_serial_info(struct usb_serial_port * port, struct serial_struct _
 	               (new_serial.flags & ASYNC_FLAGS));
 	priv->custom_divisor = new_serial.custom_divisor;
 
-	port->tty->low_latency = (priv->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
+	tty->low_latency = (priv->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
 
 check_and_exit:
 	if ((old_priv.flags & ASYNC_SPD_MASK) !=
 	     (priv->flags & ASYNC_SPD_MASK)) {
 		if ((priv->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
-			port->tty->alt_speed = 57600;
+			tty->alt_speed = 57600;
 		else if ((priv->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
-			port->tty->alt_speed = 115200;
+			tty->alt_speed = 115200;
 		else if ((priv->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
-			port->tty->alt_speed = 230400;
+			tty->alt_speed = 230400;
 		else if ((priv->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
-			port->tty->alt_speed = 460800;
+			tty->alt_speed = 460800;
 		else
-			port->tty->alt_speed = 0;
+			tty->alt_speed = 0;
 	}
 	if (((old_priv.flags & ASYNC_SPD_MASK) !=
 	     (priv->flags & ASYNC_SPD_MASK)) ||
 	    (((priv->flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST) &&
 	     (old_priv.custom_divisor != priv->custom_divisor))) {
-		change_speed(port);
+		change_speed(tty, port);
 	}
-
-	return (0);
+	return 0;
 
 } /* set_serial_info */
 
@@ -1415,7 +1412,8 @@ static int ftdi_sio_port_remove(struct usb_serial_port *port)
 	return 0;
 }
 
-static int  ftdi_open (struct usb_serial_port *port, struct file *filp)
+static int ftdi_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 { /* ftdi_open */
 	struct usb_device *dev = port->serial->dev;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
@@ -1433,8 +1431,8 @@ static int  ftdi_open (struct usb_serial_port *port, struct file *filp)
 	priv->rx_bytes = 0;
 	spin_unlock_irqrestore(&priv->rx_lock, flags);
 
-	if (port->tty)
-		port->tty->low_latency = (priv->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
+	if (tty)
+		tty->low_latency = (priv->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
 
 	/* No error checking for this (will get errors later anyway) */
 	/* See ftdi_sio.h for description of what is reset */
@@ -1448,8 +1446,8 @@ static int  ftdi_open (struct usb_serial_port *port, struct file *filp)
 	   This is same behaviour as serial.c/rs_open() - Kuba */
 
 	/* ftdi_set_termios  will send usb control messages */
-	if (port->tty)
-		ftdi_set_termios(port, port->tty->termios);
+	if (tty)
+		ftdi_set_termios(tty, port, tty->termios);
 
 	/* FIXME: Flow control might be enabled, so it should be checked -
 	   we have no control of defaults! */
@@ -1485,9 +1483,10 @@ static int  ftdi_open (struct usb_serial_port *port, struct file *filp)
  *
  */
 
-static void ftdi_close (struct usb_serial_port *port, struct file *filp)
+static void ftdi_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 { /* ftdi_close */
-	unsigned int c_cflag = port->tty->termios->c_cflag;
+	unsigned int c_cflag = tty->termios->c_cflag;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	char buf[1];
 
@@ -1527,7 +1526,7 @@ static void ftdi_close (struct usb_serial_port *port, struct file *filp)
  *
  * The new devices do not require this byte
  */
-static int ftdi_write (struct usb_serial_port *port,
+static int ftdi_write(struct tty_struct *tty, struct usb_serial_port *port,
 			   const unsigned char *buf, int count)
 { /* ftdi_write */
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
@@ -1686,8 +1685,9 @@ static void ftdi_write_bulk_callback (struct urb *urb)
 } /* ftdi_write_bulk_callback */
 
 
-static int ftdi_write_room( struct usb_serial_port *port )
+static int ftdi_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	int room;
 	unsigned long flags;
@@ -1707,11 +1707,11 @@ static int ftdi_write_room( struct usb_serial_port *port )
 	}
 	spin_unlock_irqrestore(&priv->tx_lock, flags);
 	return room;
-} /* ftdi_write_room */
-
+}
 
-static int ftdi_chars_in_buffer (struct usb_serial_port *port)
-{ /* ftdi_chars_in_buffer */
+static int ftdi_chars_in_buffer(struct tty_struct *tty)
+{
+	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	int buffered;
 	unsigned long flags;
@@ -1726,12 +1726,10 @@ static int ftdi_chars_in_buffer (struct usb_serial_port *port)
 		buffered = 0;
 	}
 	return buffered;
-} /* ftdi_chars_in_buffer */
-
-
+}
 
-static void ftdi_read_bulk_callback (struct urb *urb)
-{ /* ftdi_read_bulk_callback */
+static void ftdi_read_bulk_callback(struct urb *urb)
+{
 	struct usb_serial_port *port = urb->context;
 	struct tty_struct *tty;
 	struct ftdi_private *priv;
@@ -1747,10 +1745,10 @@ static void ftdi_read_bulk_callback (struct urb *urb)
 
 	dbg("%s - port %d", __func__, port->number);
 
-	if (port->open_count <= 0)
+	if (port->port.count <= 0)
 		return;
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (!tty) {
 		dbg("%s - bad tty pointer - exiting",__func__);
 		return;
@@ -1803,10 +1801,10 @@ static void ftdi_process_read (struct work_struct *work)
 
 	dbg("%s - port %d", __func__, port->number);
 
-	if (port->open_count <= 0)
+	if (port->port.count <= 0)
 		return;
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (!tty) {
 		dbg("%s - bad tty pointer - exiting",__func__);
 		return;
@@ -1954,7 +1952,7 @@ static void ftdi_process_read (struct work_struct *work)
 		}
 		spin_unlock_irqrestore(&priv->rx_lock, flags);
 		/* if the port is closed stop trying to read */
-		if (port->open_count > 0){
+		if (port->port.count > 0){
 			/* delay processing of remainder */
 			schedule_delayed_work(&priv->rx_work, 1);
 		} else {
@@ -1967,7 +1965,7 @@ static void ftdi_process_read (struct work_struct *work)
 	priv->rx_processed = 0;
 
 	/* if the port is closed stop trying to read */
-	if (port->open_count > 0){
+	if (port->port.count > 0){
 		/* Continue trying to always read  */
 		usb_fill_bulk_urb(port->read_urb, port->serial->dev,
 			      usb_rcvbulkpipe(port->serial->dev, port->bulk_in_endpointAddress),
@@ -1983,8 +1981,9 @@ static void ftdi_process_read (struct work_struct *work)
 } /* ftdi_process_read */
 
 
-static void ftdi_break_ctl( struct usb_serial_port *port, int break_state )
+static void ftdi_break_ctl(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	__u16 urb_value = 0;
 	char buf[1];
@@ -2018,11 +2017,12 @@ static void ftdi_break_ctl( struct usb_serial_port *port, int break_state )
  * WARNING: set_termios calls this with old_termios in kernel space
  */
 
-static void ftdi_set_termios (struct usb_serial_port *port, struct ktermios *old_termios)
+static void ftdi_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 { /* ftdi_termios */
 	struct usb_device *dev = port->serial->dev;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
-	struct ktermios *termios = port->tty->termios;
+	struct ktermios *termios = tty->termios;
 	unsigned int cflag = termios->c_cflag;
 	__u16 urb_value; /* will hold the new flags */
 	char buf[1]; /* Perhaps I should dynamically alloc this? */
@@ -2037,7 +2037,7 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct ktermios *old
 	/* Force baud rate if this device requires it, unless it is set to B0. */
 	if (priv->force_baud && ((termios->c_cflag & CBAUD) != B0)) {
 		dbg("%s: forcing baud rate for this device", __func__);
-		tty_encode_baud_rate(port->tty, priv->force_baud,
+		tty_encode_baud_rate(tty, priv->force_baud,
 					priv->force_baud);
 	}
 
@@ -2104,7 +2104,7 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct ktermios *old
 		clear_mctrl(port, TIOCM_DTR | TIOCM_RTS);
 	} else {
 		/* set the baudrate determined before */
-		if (change_speed(port)) {
+		if (change_speed(tty, port)) {
 			err("%s urb failed to set baudrate", __func__);
 		}
 		/* Ensure RTS and DTR are raised when baudrate changed from 0 */
@@ -2168,11 +2168,11 @@ static void ftdi_set_termios (struct usb_serial_port *port, struct ktermios *old
 
 	}
 	return;
-} /* ftdi_termios */
-
+}
 
-static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file)
+static int ftdi_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	unsigned char buf[2];
 	int ret;
@@ -2221,15 +2221,18 @@ static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file)
 		priv->last_dtr_rts;
 }
 
-static int ftdi_tiocmset(struct usb_serial_port *port, struct file * file, unsigned int set, unsigned int clear)
+static int ftdi_tiocmset(struct tty_struct *tty, struct file * file,
+			unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s TIOCMSET", __func__);
 	return update_mctrl(port, set, clear);
 }
 
 
-static int ftdi_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg)
+static int ftdi_ioctl(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 
 	dbg("%s cmd 0x%04x", __func__, cmd);
@@ -2241,7 +2244,7 @@ static int ftdi_ioctl (struct usb_serial_port *port, struct file * file, unsigne
 		return get_serial_info(port, (struct serial_struct __user *) arg);
 
 	case TIOCSSERIAL: /* sets serial port data */
-		return set_serial_info(port, (struct serial_struct __user *) arg);
+		return set_serial_info(tty, port, (struct serial_struct __user *) arg);
 
 	/*
 	 * Wait for any of the 4 modem inputs (DCD,RI,DSR,CTS) to change
@@ -2280,25 +2283,20 @@ static int ftdi_ioctl (struct usb_serial_port *port, struct file * file, unsigne
 				 */
 			}
 		}
-		return(0);
-		break;
+		return 0;
 	default:
 		break;
-
 	}
-
-
 	/* This is not necessarily an error - turns out the higher layers will do
 	 *  some ioctls itself (see comment above)
 	 */
 	dbg("%s arg not supported - it was 0x%04x - check /usr/include/asm/ioctls.h", __func__, cmd);
+	return -ENOIOCTLCMD;
+}
 
-	return(-ENOIOCTLCMD);
-} /* ftdi_ioctl */
-
-
-static void ftdi_throttle (struct usb_serial_port *port)
+static void ftdi_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -2310,8 +2308,9 @@ static void ftdi_throttle (struct usb_serial_port *port)
 }
 
 
-static void ftdi_unthrottle (struct usb_serial_port *port)
+static void ftdi_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 	int actually_throttled;
 	unsigned long flags;
diff --git a/drivers/usb/serial/garmin_gps.c b/drivers/usb/serial/garmin_gps.c
index 8ce5a56a48e3..06cfa43c6f02 100644
--- a/drivers/usb/serial/garmin_gps.c
+++ b/drivers/usb/serial/garmin_gps.c
@@ -275,7 +275,7 @@ static inline int isAbortTrfCmnd(const unsigned char *buf)
 static void send_to_tty(struct usb_serial_port *port,
 			char *data, unsigned int actual_length)
 {
-	struct tty_struct *tty = port->tty;
+	struct tty_struct *tty = port->port.tty;
 
 	if (tty && actual_length) {
 
@@ -970,7 +970,8 @@ static int garmin_init_session(struct usb_serial_port *port)
 
 
 
-static int garmin_open (struct usb_serial_port *port, struct file *filp)
+static int garmin_open (struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	unsigned long flags;
 	int status = 0;
@@ -983,8 +984,8 @@ static int garmin_open (struct usb_serial_port *port, struct file *filp)
 	 * through, otherwise it is scheduled, and with high data rates (like
 	 * with OHCI) data can get lost.
 	 */
-	if (port->tty)
-		port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	spin_lock_irqsave(&garmin_data_p->lock, flags);
 	garmin_data_p->mode  = initial_mode;
@@ -998,17 +999,16 @@ static int garmin_open (struct usb_serial_port *port, struct file *filp)
 	usb_kill_urb (port->write_urb);
 	usb_kill_urb (port->read_urb);
 
-	if (garmin_data_p->state == STATE_RESET) {
+	if (garmin_data_p->state == STATE_RESET)
 		status = garmin_init_session(port);
-	}
 
 	garmin_data_p->state = STATE_ACTIVE;
-
 	return status;
 }
 
 
-static void garmin_close (struct usb_serial_port *port, struct file * filp)
+static void garmin_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct usb_serial *serial = port->serial;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
@@ -1042,7 +1042,6 @@ static void garmin_close (struct usb_serial_port *port, struct file * filp)
 	mutex_unlock(&port->serial->disc_mutex);
 }
 
-
 static void garmin_write_bulk_callback (struct urb *urb)
 {
 	unsigned long flags;
@@ -1145,10 +1144,8 @@ static int garmin_write_bulk (struct usb_serial_port *port,
 	return count;
 }
 
-
-
-static int garmin_write (struct usb_serial_port *port,
-			 const unsigned char *buf, int count)
+static int garmin_write (struct tty_struct *tty, struct usb_serial_port *port,
+					 const unsigned char *buf, int count)
 {
 	int pktid, pktsiz, len;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
@@ -1158,7 +1155,6 @@ static int garmin_write (struct usb_serial_port *port,
 
 	/* check for our private packets */
 	if (count >= GARMIN_PKTHDR_LENGTH) {
-
 		len = PRIVPKTSIZ;
 		if (count < len)
 			len = count;
@@ -1226,8 +1222,9 @@ static int garmin_write (struct usb_serial_port *port,
 }
 
 
-static int garmin_write_room (struct usb_serial_port *port)
+static int garmin_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	/*
 	 * Report back the bytes currently available in the output buffer.
 	 */
@@ -1236,20 +1233,6 @@ static int garmin_write_room (struct usb_serial_port *port)
 }
 
 
-static int garmin_chars_in_buffer (struct usb_serial_port *port)
-{
-	/*
-	 * Report back the number of bytes currently in our input buffer.
-	 * Will this lock up the driver - the buffer contains an incomplete
-	 * package which will not be written to the device until it
-	 * has been completed ?
-	 */
-	//struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
-	//return garmin_data_p->insize;
-	return 0;
-}
-
-
 static void garmin_read_process(struct garmin_data * garmin_data_p,
 				 unsigned char *data, unsigned data_length)
 {
@@ -1468,10 +1451,11 @@ static int garmin_flush_queue(struct garmin_data * garmin_data_p)
 }
 
 
-static void garmin_throttle (struct usb_serial_port *port)
+static void garmin_throttle(struct tty_struct *tty)
 {
-	unsigned long flags;
+	struct usb_serial_port *port = tty->driver_data;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
+	unsigned long flags;
 
 	dbg("%s - port %d", __func__, port->number);
 	/* set flag, data received will be put into a queue
@@ -1482,10 +1466,11 @@ static void garmin_throttle (struct usb_serial_port *port)
 }
 
 
-static void garmin_unthrottle (struct usb_serial_port *port)
+static void garmin_unthrottle (struct tty_struct *tty)
 {
-	unsigned long flags;
+	struct usb_serial_port *port = tty->driver_data;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
+	unsigned long flags;
 	int status;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -1507,8 +1492,6 @@ static void garmin_unthrottle (struct usb_serial_port *port)
 	}
 }
 
-
-
 /*
  * The timer is currently only used to send queued packets to
  * the tty in cases where the protocol provides no own handshaking
@@ -1526,7 +1509,7 @@ static void timeout_handler(unsigned long data)
 
 
 
-static int garmin_attach (struct usb_serial *serial)
+static int garmin_attach(struct usb_serial *serial)
 {
 	int status = 0;
 	struct usb_serial_port *port = serial->port[0];
@@ -1556,7 +1539,7 @@ static int garmin_attach (struct usb_serial *serial)
 }
 
 
-static void garmin_shutdown (struct usb_serial *serial)
+static void garmin_shutdown(struct usb_serial *serial)
 {
 	struct usb_serial_port *port = serial->port[0];
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
@@ -1588,7 +1571,6 @@ static struct usb_serial_driver garmin_device = {
 	.shutdown            = garmin_shutdown,
 	.write               = garmin_write,
 	.write_room          = garmin_write_room,
-	.chars_in_buffer     = garmin_chars_in_buffer,
 	.write_bulk_callback = garmin_write_bulk_callback,
 	.read_bulk_callback  = garmin_read_bulk_callback,
 	.read_int_callback   = garmin_read_int_callback,
diff --git a/drivers/usb/serial/generic.c b/drivers/usb/serial/generic.c
index 537f12a027c2..5128018c2766 100644
--- a/drivers/usb/serial/generic.c
+++ b/drivers/usb/serial/generic.c
@@ -112,7 +112,8 @@ void usb_serial_generic_deregister (void)
 #endif
 }
 
-int usb_serial_generic_open (struct usb_serial_port *port, struct file *filp)
+int usb_serial_generic_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	int result = 0;
@@ -123,8 +124,8 @@ int usb_serial_generic_open (struct usb_serial_port *port, struct file *filp)
 	/* force low_latency on so that our tty_push actually forces the data through, 
 	   otherwise it is scheduled, and with high data rates (like with OHCI) data
 	   can get lost. */
-	if (port->tty)
-		port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	/* clear the throttle flags */
 	spin_lock_irqsave(&port->lock, flags);
@@ -152,7 +153,7 @@ int usb_serial_generic_open (struct usb_serial_port *port, struct file *filp)
 }
 EXPORT_SYMBOL_GPL(usb_serial_generic_open);
 
-static void generic_cleanup (struct usb_serial_port *port)
+static void generic_cleanup(struct usb_serial_port *port)
 {
 	struct usb_serial *serial = port->serial;
 
@@ -182,7 +183,7 @@ int usb_serial_generic_resume(struct usb_serial *serial)
 #endif
 	for (i = 0; i < serial->num_ports; i++) {
 		port = serial->port[i];
-		if (port->open_count && port->read_urb) {
+		if (port->port.count && port->read_urb) {
 			r = usb_submit_urb(port->read_urb, GFP_NOIO);
 			if (r < 0)
 				c++;
@@ -192,13 +193,15 @@ int usb_serial_generic_resume(struct usb_serial *serial)
 	return c ? -EIO : 0;
 }
 
-void usb_serial_generic_close (struct usb_serial_port *port, struct file * filp)
+void usb_serial_generic_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 	generic_cleanup (port);
 }
 
-int usb_serial_generic_write(struct usb_serial_port *port, const unsigned char *buf, int count)
+int usb_serial_generic_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count)
 {
 	struct usb_serial *serial = port->serial;
 	int result;
@@ -255,8 +258,9 @@ int usb_serial_generic_write(struct usb_serial_port *port, const unsigned char *
 	return 0;
 }
 
-int usb_serial_generic_write_room (struct usb_serial_port *port)
+int usb_serial_generic_write_room (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	int room = 0;
 
@@ -272,8 +276,9 @@ int usb_serial_generic_write_room (struct usb_serial_port *port)
 	return room;
 }
 
-int usb_serial_generic_chars_in_buffer (struct usb_serial_port *port)
+int usb_serial_generic_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	int chars = 0;
 
@@ -286,7 +291,7 @@ int usb_serial_generic_chars_in_buffer (struct usb_serial_port *port)
 	}
 
 	dbg("%s - returns %d", __func__, chars);
-	return (chars);
+	return chars;
 }
 
 
@@ -311,10 +316,10 @@ static void resubmit_read_urb(struct usb_serial_port *port, gfp_t mem_flags)
 }
 
 /* Push data to tty layer and resubmit the bulk read URB */
-static void flush_and_resubmit_read_urb (struct usb_serial_port *port)
+static void flush_and_resubmit_read_urb(struct usb_serial_port *port)
 {
 	struct urb *urb = port->read_urb;
-	struct tty_struct *tty = port->tty;
+	struct tty_struct *tty = port->port.tty;
 	int room;
 
 	/* Push data to tty */
@@ -329,7 +334,7 @@ static void flush_and_resubmit_read_urb (struct usb_serial_port *port)
 	resubmit_read_urb(port, GFP_ATOMIC);
 }
 
-void usb_serial_generic_read_bulk_callback (struct urb *urb)
+void usb_serial_generic_read_bulk_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	unsigned char *data = urb->transfer_buffer;
@@ -357,7 +362,7 @@ void usb_serial_generic_read_bulk_callback (struct urb *urb)
 }
 EXPORT_SYMBOL_GPL(usb_serial_generic_read_bulk_callback);
 
-void usb_serial_generic_write_bulk_callback (struct urb *urb)
+void usb_serial_generic_write_bulk_callback(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
 	int status = urb->status;
@@ -374,8 +379,9 @@ void usb_serial_generic_write_bulk_callback (struct urb *urb)
 }
 EXPORT_SYMBOL_GPL(usb_serial_generic_write_bulk_callback);
 
-void usb_serial_generic_throttle (struct usb_serial_port *port)
+void usb_serial_generic_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned long flags;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -387,8 +393,9 @@ void usb_serial_generic_throttle (struct usb_serial_port *port)
 	spin_unlock_irqrestore(&port->lock, flags);
 }
 
-void usb_serial_generic_unthrottle (struct usb_serial_port *port)
+void usb_serial_generic_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int was_throttled;
 	unsigned long flags;
 
diff --git a/drivers/usb/serial/io_edgeport.c b/drivers/usb/serial/io_edgeport.c
index 2fd449bcfa35..29346d79df7b 100644
--- a/drivers/usb/serial/io_edgeport.c
+++ b/drivers/usb/serial/io_edgeport.c
@@ -206,18 +206,18 @@ static void edge_bulk_out_data_callback	(struct urb *urb);
 static void edge_bulk_out_cmd_callback	(struct urb *urb);
 
 /* function prototypes for the usbserial callbacks */
-static int  edge_open			(struct usb_serial_port *port, struct file *filp);
-static void edge_close			(struct usb_serial_port *port, struct file *filp);
-static int  edge_write			(struct usb_serial_port *port, const unsigned char *buf, int count);
-static int  edge_write_room		(struct usb_serial_port *port);
-static int  edge_chars_in_buffer	(struct usb_serial_port *port);
-static void edge_throttle		(struct usb_serial_port *port);
-static void edge_unthrottle		(struct usb_serial_port *port);
-static void edge_set_termios		(struct usb_serial_port *port, struct ktermios *old_termios);
-static int  edge_ioctl			(struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg);
-static void edge_break			(struct usb_serial_port *port, int break_state);
-static int  edge_tiocmget		(struct usb_serial_port *port, struct file *file);
-static int  edge_tiocmset		(struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear);
+static int  edge_open			(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void edge_close			(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static int  edge_write			(struct tty_struct *tty, struct usb_serial_port *port, const unsigned char *buf, int count);
+static int  edge_write_room		(struct tty_struct *tty);
+static int  edge_chars_in_buffer	(struct tty_struct *tty);
+static void edge_throttle		(struct tty_struct *tty);
+static void edge_unthrottle		(struct tty_struct *tty);
+static void edge_set_termios		(struct tty_struct *tty, struct usb_serial_port *port, struct ktermios *old_termios);
+static int  edge_ioctl			(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg);
+static void edge_break			(struct tty_struct *tty, int break_state);
+static int  edge_tiocmget		(struct tty_struct *tty, struct file *file);
+static int  edge_tiocmset		(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear);
 static int  edge_startup		(struct usb_serial *serial);
 static void edge_shutdown		(struct usb_serial *serial);
 
@@ -233,7 +233,8 @@ static void handle_new_lsr		(struct edgeport_port *edge_port, __u8 lsrData, __u8
 static int  send_iosp_ext_cmd		(struct edgeport_port *edge_port, __u8 command, __u8 param);
 static int  calc_baud_rate_divisor	(int baud_rate, int *divisor);
 static int  send_cmd_write_baud_rate	(struct edgeport_port *edge_port, int baudRate);
-static void change_port_settings	(struct edgeport_port *edge_port, struct ktermios *old_termios);
+static void change_port_settings	(struct tty_struct *tty, struct edgeport_port *edge_port,
+					 struct ktermios *old_termios);
 static int  send_cmd_write_uart_register	(struct edgeport_port *edge_port, __u8 regNum, __u8 regValue);
 static int  write_cmd_usb		(struct edgeport_port *edge_port, unsigned char *buffer, int writeLength);
 static void send_more_port_data		(struct edgeport_serial *edge_serial, struct edgeport_port *edge_port);
@@ -639,8 +640,8 @@ static void edge_interrupt_callback (struct urb *urb)
 					dbg("%s - txcredits for port%d = %d", __func__, portNumber, edge_port->txCredits);
 
 					/* tell the tty driver that something has changed */
-					if (edge_port->port->tty)
-						tty_wakeup(edge_port->port->tty);
+					if (edge_port->port->port.tty)
+						tty_wakeup(edge_port->port->port.tty);
 
 					// Since we have more credit, check if more data can be sent
 					send_more_port_data(edge_serial, edge_port);
@@ -737,7 +738,7 @@ static void edge_bulk_out_data_callback (struct urb *urb)
 		    __func__, status);
 	}
 
-	tty = edge_port->port->tty;
+	tty = edge_port->port->port.tty;
 
 	if (tty && edge_port->open) {
 		/* let the tty driver wakeup if it has a special write_wakeup function */
@@ -781,7 +782,7 @@ static void edge_bulk_out_cmd_callback (struct urb *urb)
 	}
 
 	/* Get pointer to tty */
-	tty = edge_port->port->tty;
+	tty = edge_port->port->port.tty;
 
 	/* tell the tty driver that something has changed */
 	if (tty && edge_port->open)
@@ -803,7 +804,8 @@ static void edge_bulk_out_cmd_callback (struct urb *urb)
  *	If successful, we return 0
  *	Otherwise we return a negative error number.
  *****************************************************************************/
-static int edge_open (struct usb_serial_port *port, struct file * filp)
+static int edge_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	struct usb_serial *serial;
@@ -815,16 +817,15 @@ static int edge_open (struct usb_serial_port *port, struct file * filp)
 	if (edge_port == NULL)
 		return -ENODEV;
 
-	if (port->tty)
-		port->tty->low_latency = low_latency;
+	if (tty)
+		tty->low_latency = low_latency;
 
 	/* see if we've set up our endpoint info yet (can't set it up in edge_startup
 	   as the structures were not set up at that time.) */
 	serial = port->serial;
 	edge_serial = usb_get_serial_data(serial);
-	if (edge_serial == NULL) {
+	if (edge_serial == NULL)
 		return -ENODEV;
-	}
 	if (edge_serial->interrupt_in_buffer == NULL) {
 		struct usb_serial_port *port0 = serial->port[0];
 		
@@ -908,7 +909,7 @@ static int edge_open (struct usb_serial_port *port, struct file * filp)
 
 	if (!edge_port->txfifo.fifo) {
 		dbg("%s - no memory", __func__);
-		edge_close (port, filp);
+		edge_close (tty, port, filp);
 		return -ENOMEM;
 	}
 
@@ -918,7 +919,7 @@ static int edge_open (struct usb_serial_port *port, struct file * filp)
 
 	if (!edge_port->write_urb) {
 		dbg("%s - no memory", __func__);
-		edge_close (port, filp);
+		edge_close (tty, port, filp);
 		return -ENOMEM;
 	}
 
@@ -1038,7 +1039,8 @@ static void block_until_tx_empty (struct edgeport_port *edge_port)
  * edge_close
  *	this function is called by the tty driver when a port is closed
  *****************************************************************************/
-static void edge_close (struct usb_serial_port *port, struct file * filp)
+static void edge_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct edgeport_serial *edge_serial;
 	struct edgeport_port *edge_port;
@@ -1106,7 +1108,8 @@ static void edge_close (struct usb_serial_port *port, struct file * filp)
  *	If successful, we return the number of bytes written, otherwise we return
  *	a negative error number.
  *****************************************************************************/
-static int edge_write (struct usb_serial_port *port, const unsigned char *data, int count)
+static int edge_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *data, int count)
 {
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	struct TxFifo *fifo;
@@ -1308,8 +1311,9 @@ exit_send:
  *	(the txCredits), 
  *	Otherwise we return a negative error number.
  *****************************************************************************/
-static int edge_write_room (struct usb_serial_port *port)
+static int edge_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	int room;
 	unsigned long flags;
@@ -1347,8 +1351,9 @@ static int edge_write_room (struct usb_serial_port *port)
  *	system, 
  *	Otherwise we return a negative error number.
  *****************************************************************************/
-static int edge_chars_in_buffer (struct usb_serial_port *port)
+static int edge_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	int num_chars;
 	unsigned long flags;
@@ -1381,10 +1386,10 @@ static int edge_chars_in_buffer (struct usb_serial_port *port)
  *	this function is called by the tty driver when it wants to stop the data
  *	being read from the port.
  *****************************************************************************/
-static void edge_throttle (struct usb_serial_port *port)
+static void edge_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
-	struct tty_struct *tty;
 	int status;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -1397,16 +1402,10 @@ static void edge_throttle (struct usb_serial_port *port)
 		return;
 	}
 
-	tty = port->tty;
-	if (!tty) {
-		dbg ("%s - no tty available", __func__);
-		return;
-	}
-
 	/* if we are implementing XON/XOFF, send the stop character */
 	if (I_IXOFF(tty)) {
 		unsigned char stop_char = STOP_CHAR(tty);
-		status = edge_write (port, &stop_char, 1);
+		status = edge_write (tty, port, &stop_char, 1);
 		if (status <= 0) {
 			return;
 		}
@@ -1430,10 +1429,10 @@ static void edge_throttle (struct usb_serial_port *port)
  *	this function is called by the tty driver when it wants to resume the data
  *	being read from the port (called after SerialThrottle is called)
  *****************************************************************************/
-static void edge_unthrottle (struct usb_serial_port *port)
+static void edge_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
-	struct tty_struct *tty;
 	int status;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -1446,31 +1445,18 @@ static void edge_unthrottle (struct usb_serial_port *port)
 		return;
 	}
 
-	tty = port->tty;
-	if (!tty) {
-		dbg ("%s - no tty available", __func__);
-		return;
-	}
-
 	/* if we are implementing XON/XOFF, send the start character */
 	if (I_IXOFF(tty)) {
 		unsigned char start_char = START_CHAR(tty);
-		status = edge_write (port, &start_char, 1);
-		if (status <= 0) {
+		status = edge_write(tty, port, &start_char, 1);
+		if (status <= 0)
 			return;
-		}
 	}
-
 	/* if we are implementing RTS/CTS, toggle that line */
 	if (tty->termios->c_cflag & CRTSCTS) {
 		edge_port->shadowMCR |= MCR_RTS;
-		status = send_cmd_write_uart_register(edge_port, MCR, edge_port->shadowMCR);
-		if (status != 0) {
-			return;
-		}
+		send_cmd_write_uart_register(edge_port, MCR, edge_port->shadowMCR);
 	}
-
-	return;
 }
 
 
@@ -1478,11 +1464,10 @@ static void edge_unthrottle (struct usb_serial_port *port)
  * SerialSetTermios
  *	this function is called by the tty driver when it wants to change the termios structure
  *****************************************************************************/
-static void edge_set_termios (struct usb_serial_port *port, struct ktermios *old_termios)
+static void edge_set_termios(struct tty_struct *tty,
+	struct usb_serial_port *port, struct ktermios *old_termios)
 {
-	/* FIXME: This function appears unused ?? */
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	unsigned int cflag;
 
 	cflag = tty->termios->c_cflag;
@@ -1502,9 +1487,7 @@ static void edge_set_termios (struct usb_serial_port *port, struct ktermios *old
 	}
 
 	/* change the port settings to the new ones specified */
-	change_port_settings (edge_port, old_termios);
-
-	return;
+	change_port_settings(tty, edge_port, old_termios);
 }
 
 
@@ -1536,25 +1519,9 @@ static int get_lsr_info(struct edgeport_port *edge_port, unsigned int __user *va
 	return 0;
 }
 
-static int get_number_bytes_avail(struct edgeport_port *edge_port, unsigned int __user *value)
-{
-	unsigned int result = 0;
-	struct tty_struct *tty = edge_port->port->tty;
-
-	if (!tty)
-		return -ENOIOCTLCMD;
-
-	result = tty->read_cnt;
-
-	dbg("%s(%d) = %d", __func__,  edge_port->port->number, result);
-	if (copy_to_user(value, &result, sizeof(int)))
-		return -EFAULT;
-	//return 0;
-	return -ENOIOCTLCMD;
-}
-
-static int edge_tiocmset (struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear)
+static int edge_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	unsigned int mcr;
 
@@ -1582,8 +1549,9 @@ static int edge_tiocmset (struct usb_serial_port *port, struct file *file, unsig
 	return 0;
 }
 
-static int edge_tiocmget(struct usb_serial_port *port, struct file *file)
+static int edge_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	unsigned int result = 0;
 	unsigned int msr;
@@ -1624,9 +1592,6 @@ static int get_serial_info(struct edgeport_port *edge_port, struct serial_struct
 	tmp.baud_base		= 9600;
 	tmp.close_delay		= 5*HZ;
 	tmp.closing_wait	= 30*HZ;
-//	tmp.custom_divisor	= state->custom_divisor;
-//	tmp.hub6		= state->hub6;
-//	tmp.io_type		= state->io_type;
 
 	if (copy_to_user(retinfo, &tmp, sizeof(*retinfo)))
 		return -EFAULT;
@@ -1639,8 +1604,10 @@ static int get_serial_info(struct edgeport_port *edge_port, struct serial_struct
  * SerialIoctl
  *	this function handles any ioctl calls to the driver
  *****************************************************************************/
-static int edge_ioctl (struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg)
+static int edge_ioctl(struct tty_struct *tty, struct file *file,
+					unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	DEFINE_WAIT(wait);
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	struct async_icount cnow;
@@ -1650,25 +1617,14 @@ static int edge_ioctl (struct usb_serial_port *port, struct file *file, unsigned
 	dbg("%s - port %d, cmd = 0x%x", __func__, port->number, cmd);
 
 	switch (cmd) {
-		// return number of bytes available
-		case TIOCINQ:
-			dbg("%s (%d) TIOCINQ", __func__,  port->number);
-			return get_number_bytes_avail(edge_port, (unsigned int __user *) arg);
-			break;
-
 		case TIOCSERGETLSR:
 			dbg("%s (%d) TIOCSERGETLSR", __func__,  port->number);
 			return get_lsr_info(edge_port, (unsigned int __user *) arg);
-			return 0;
 
 		case TIOCGSERIAL:
 			dbg("%s (%d) TIOCGSERIAL", __func__,  port->number);
 			return get_serial_info(edge_port, (struct serial_struct __user *) arg);
 
-		case TIOCSSERIAL:
-			dbg("%s (%d) TIOCSSERIAL", __func__,  port->number);
-			break;
-
 		case TIOCMIWAIT:
 			dbg("%s (%d) TIOCMIWAIT", __func__,  port->number);
 			cprev = edge_port->icount;
@@ -1723,8 +1679,9 @@ static int edge_ioctl (struct usb_serial_port *port, struct file *file, unsigned
  * SerialBreak
  *	this function sends a break to the port
  *****************************************************************************/
-static void edge_break (struct usb_serial_port *port, int break_state)
+static void edge_break (struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	struct edgeport_serial *edge_serial = usb_get_serial_data(port->serial);
 	int status;
@@ -1866,7 +1823,7 @@ static void process_rcvd_data (struct edgeport_serial *edge_serial, unsigned cha
 					port = edge_serial->serial->port[edge_serial->rxPort];
 					edge_port = usb_get_serial_port_data(port);
 					if (edge_port->open) {
-						tty = edge_port->port->tty;
+						tty = edge_port->port->port.tty;
 						if (tty) {
 							dbg("%s - Sending %d bytes to TTY for port %d", __func__, rxLen, edge_serial->rxPort);
 							edge_tty_recv(&edge_serial->serial->dev->dev, tty, buffer, rxLen);
@@ -1941,8 +1898,9 @@ static void process_rcvd_status (struct edgeport_serial *edge_serial, __u8 byte2
 		handle_new_msr (edge_port, byte2);
 
 		/* send the current line settings to the port so we are in sync with any further termios calls */
-		if (edge_port->port->tty)
-			change_port_settings (edge_port, edge_port->port->tty->termios);
+		/* FIXME: locking on tty */
+		if (edge_port->port->port.tty)
+			change_port_settings(edge_port->port->port.tty, edge_port, edge_port->port->port.tty->termios);
 
 		/* we have completed the open */
 		edge_port->openPending = false;
@@ -2078,8 +2036,8 @@ static void handle_new_lsr(struct edgeport_port *edge_port, __u8 lsrData, __u8 l
 	}
 
 	/* Place LSR data byte into Rx buffer */
-	if (lsrData && edge_port->port->tty)
-		edge_tty_recv(&edge_port->port->dev, edge_port->port->tty, &data, 1);
+	if (lsrData && edge_port->port->port.tty)
+		edge_tty_recv(&edge_port->port->dev, edge_port->port->port.tty, &data, 1);
 
 	/* update input line counters */
 	icount = &edge_port->icount;
@@ -2473,13 +2431,11 @@ static int send_cmd_write_uart_register (struct edgeport_port *edge_port, __u8 r
  *	This routine is called to set the UART on the device to match the specified
  *	new settings.
  *****************************************************************************/
-#ifndef CMSPAR
-#define CMSPAR 0
-#endif
-static void change_port_settings (struct edgeport_port *edge_port, struct ktermios *old_termios)
+
+static void change_port_settings(struct tty_struct *tty,
+	struct edgeport_port *edge_port, struct ktermios *old_termios)
 {
 	struct edgeport_serial *edge_serial = usb_get_serial_data(edge_port->port->serial);
-	struct tty_struct *tty;
 	int baud;
 	unsigned cflag;
 	__u8 mask = 0xff;
@@ -2498,13 +2454,6 @@ static void change_port_settings (struct edgeport_port *edge_port, struct ktermi
 		return;
 	}
 
-	tty = edge_port->port->tty;
-	if ((!tty) ||
-	    (!tty->termios)) {
-		dbg("%s - no tty structures", __func__);
-		return;
-	}
-
 	cflag = tty->termios->c_cflag;
 
 	switch (cflag & CSIZE) {
diff --git a/drivers/usb/serial/io_ti.c b/drivers/usb/serial/io_ti.c
index a58822a14a87..7cf383a2a556 100644
--- a/drivers/usb/serial/io_ti.c
+++ b/drivers/usb/serial/io_ti.c
@@ -243,9 +243,9 @@ static void edge_tty_recv(struct device *dev, struct tty_struct *tty,
 static void stop_read(struct edgeport_port *edge_port);
 static int restart_read(struct edgeport_port *edge_port);
 
-static void edge_set_termios(struct usb_serial_port *port,
-			     struct ktermios *old_termios);
-static void edge_send(struct usb_serial_port *port);
+static void edge_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios);
+static void edge_send(struct tty_struct *tty);
 
 /* sysfs attributes */
 static int edge_create_sysfs_attrs(struct usb_serial_port *port);
@@ -572,7 +572,7 @@ static void chase_port(struct edgeport_port *port, unsigned long timeout,
 								int flush)
 {
 	int baud_rate;
-	struct tty_struct *tty = port->port->tty;
+	struct tty_struct *tty = port->port->port.tty;
 	wait_queue_t wait;
 	unsigned long flags;
 
@@ -1554,7 +1554,7 @@ static void handle_new_msr(struct edgeport_port *edge_port, __u8 msr)
 	/* Save the new modem status */
 	edge_port->shadow_msr = msr & 0xf0;
 
-	tty = edge_port->port->tty;
+	tty = edge_port->port->port.tty;
 	/* handle CTS flow control */
 	if (tty && C_CRTSCTS(tty)) {
 		if (msr & EDGEPORT_MSR_CTS) {
@@ -1587,9 +1587,8 @@ static void handle_new_lsr(struct edgeport_port *edge_port, int lsr_data,
 		new_lsr &= (__u8)(LSR_OVER_ERR | LSR_BREAK);
 
 	/* Place LSR data byte into Rx buffer */
-	if (lsr_data && edge_port->port->tty)
-		edge_tty_recv(&edge_port->port->dev, edge_port->port->tty,
-								&data, 1);
+	if (lsr_data && edge_port->port->port.tty)
+		edge_tty_recv(&edge_port->port->dev, edge_port->port->port.tty, &data, 1);
 
 	/* update input line counters */
 	icount = &edge_port->icount;
@@ -1750,7 +1749,7 @@ static void edge_bulk_in_callback(struct urb *urb)
 		++data;
 	}
 
-	tty = edge_port->port->tty;
+	tty = edge_port->port->port.tty;
 	if (tty && urb->actual_length) {
 		usb_serial_debug_data(debug, &edge_port->port->dev,
 					__func__, urb->actual_length, data);
@@ -1819,10 +1818,11 @@ static void edge_bulk_out_callback(struct urb *urb)
 	}
 
 	/* send any buffered data */
-	edge_send(port);
+	edge_send(port->port.tty);
 }
 
-static int edge_open(struct usb_serial_port *port, struct file *filp)
+static int edge_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	struct edgeport_serial *edge_serial;
@@ -1838,7 +1838,8 @@ static int edge_open(struct usb_serial_port *port, struct file *filp)
 	if (edge_port == NULL)
 		return -ENODEV;
 
-	port->tty->low_latency = low_latency;
+	if (tty)
+		tty->low_latency = low_latency;
 
 	port_number = port->number - port->serial->minor;
 	switch (port_number) {
@@ -1874,7 +1875,8 @@ static int edge_open(struct usb_serial_port *port, struct file *filp)
 	}
 
 	/* set up the port settings */
-	edge_set_termios(port, port->tty->termios);
+	if (tty)
+		edge_set_termios(tty, port, port->port.tty->termios);
 
 	/* open up the port */
 
@@ -2000,7 +2002,8 @@ release_es_lock:
 	return status;
 }
 
-static void edge_close(struct usb_serial_port *port, struct file *filp)
+static void edge_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct edgeport_serial *edge_serial;
 	struct edgeport_port *edge_port;
@@ -2048,8 +2051,8 @@ static void edge_close(struct usb_serial_port *port, struct file *filp)
 	dbg("%s - exited", __func__);
 }
 
-static int edge_write(struct usb_serial_port *port, const unsigned char *data,
-								int count)
+static int edge_write(struct tty_struct *tty, struct usb_serial_port *port,
+				const unsigned char *data, int count)
 {
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -2070,16 +2073,16 @@ static int edge_write(struct usb_serial_port *port, const unsigned char *data,
 	count = edge_buf_put(edge_port->ep_out_buf, data, count);
 	spin_unlock_irqrestore(&edge_port->ep_lock, flags);
 
-	edge_send(port);
+	edge_send(tty);
 
 	return count;
 }
 
-static void edge_send(struct usb_serial_port *port)
+static void edge_send(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int count, result;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	unsigned long flags;
 
 
@@ -2133,8 +2136,9 @@ static void edge_send(struct usb_serial_port *port)
 		tty_wakeup(tty);
 }
 
-static int edge_write_room(struct usb_serial_port *port)
+static int edge_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	int room = 0;
 	unsigned long flags;
@@ -2154,8 +2158,9 @@ static int edge_write_room(struct usb_serial_port *port)
 	return room;
 }
 
-static int edge_chars_in_buffer(struct usb_serial_port *port)
+static int edge_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	int chars = 0;
 	unsigned long flags;
@@ -2175,10 +2180,10 @@ static int edge_chars_in_buffer(struct usb_serial_port *port)
 	return chars;
 }
 
-static void edge_throttle(struct usb_serial_port *port)
+static void edge_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	int status;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -2189,11 +2194,10 @@ static void edge_throttle(struct usb_serial_port *port)
 	/* if we are implementing XON/XOFF, send the stop character */
 	if (I_IXOFF(tty)) {
 		unsigned char stop_char = STOP_CHAR(tty);
-		status = edge_write(port, &stop_char, 1);
-		if (status <= 0)
-			dev_err(&port->dev,
-				"%s - failed to write stop character, %d\n",
-							__func__, status);
+		status = edge_write(tty, port, &stop_char, 1);
+		if (status <= 0) {
+			dev_err(&port->dev, "%s - failed to write stop character, %d\n", __func__, status);
+		}
 	}
 
 	/* if we are implementing RTS/CTS, stop reads */
@@ -2203,10 +2207,10 @@ static void edge_throttle(struct usb_serial_port *port)
 
 }
 
-static void edge_unthrottle(struct usb_serial_port *port)
+static void edge_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	int status;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -2217,11 +2221,10 @@ static void edge_unthrottle(struct usb_serial_port *port)
 	/* if we are implementing XON/XOFF, send the start character */
 	if (I_IXOFF(tty)) {
 		unsigned char start_char = START_CHAR(tty);
-		status = edge_write(port, &start_char, 1);
-		if (status <= 0)
-			dev_err(&port->dev,
-				"%s - failed to write start character, %d\n",
-							__func__, status);
+		status = edge_write(tty, port, &start_char, 1);
+		if (status <= 0) {
+			dev_err(&port->dev, "%s - failed to write start character, %d\n", __func__, status);
+		}
 	}
 	/* if we are implementing RTS/CTS, restart reads */
 	/* are the Edgeport will assert the RTS line */
@@ -2271,11 +2274,10 @@ static int restart_read(struct edgeport_port *edge_port)
 	return status;
 }
 
-static void change_port_settings(struct edgeport_port *edge_port,
-						struct ktermios *old_termios)
+static void change_port_settings(struct tty_struct *tty,
+		struct edgeport_port *edge_port, struct ktermios *old_termios)
 {
 	struct ump_uart_config *config;
-	struct tty_struct *tty;
 	int baud;
 	unsigned cflag;
 	int status;
@@ -2284,9 +2286,7 @@ static void change_port_settings(struct edgeport_port *edge_port,
 
 	dbg("%s - port %d", __func__, edge_port->port->number);
 
-	tty = edge_port->port->tty;
-
-	config = kmalloc(sizeof(*config), GFP_KERNEL);
+	config = kmalloc (sizeof (*config), GFP_KERNEL);
 	if (!config) {
 		*tty->termios = *old_termios;
 		dev_err(&edge_port->port->dev, "%s - out of memory\n",
@@ -2419,11 +2419,13 @@ static void change_port_settings(struct edgeport_port *edge_port,
 	return;
 }
 
-static void edge_set_termios(struct usb_serial_port *port,
-					struct ktermios *old_termios)
+static void edge_set_termios(struct tty_struct *tty, 
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
+	unsigned int cflag;
+
+	cflag = tty->termios->c_cflag;
 
 	dbg("%s - clfag %08x iflag %08x", __func__,
 	    tty->termios->c_cflag, tty->termios->c_iflag);
@@ -2434,12 +2436,14 @@ static void edge_set_termios(struct usb_serial_port *port,
 	if (edge_port == NULL)
 		return;
 	/* change the port settings to the new ones specified */
-	change_port_settings(edge_port, old_termios);
+	change_port_settings(tty, edge_port, old_termios);
+	return;
 }
 
-static int edge_tiocmset(struct usb_serial_port *port, struct file *file,
+static int edge_tiocmset(struct tty_struct *tty, struct file *file,
 					unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	unsigned int mcr;
 	unsigned long flags;
@@ -2469,8 +2473,9 @@ static int edge_tiocmset(struct usb_serial_port *port, struct file *file,
 	return 0;
 }
 
-static int edge_tiocmget(struct usb_serial_port *port, struct file *file)
+static int edge_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	unsigned int result = 0;
 	unsigned int msr;
@@ -2522,9 +2527,10 @@ static int get_serial_info(struct edgeport_port *edge_port,
 	return 0;
 }
 
-static int edge_ioctl(struct usb_serial_port *port, struct file *file,
+static int edge_ioctl(struct tty_struct *tty, struct file *file,
 					unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	struct async_icount cnow;
 	struct async_icount cprev;
@@ -2569,18 +2575,19 @@ static int edge_ioctl(struct usb_serial_port *port, struct file *file,
 	return -ENOIOCTLCMD;
 }
 
-static void edge_break(struct usb_serial_port *port, int on)
+static void edge_break(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	int status;
 	int bv = 0;	/* Off */
 
-	dbg("%s - state = %d", __func__, on);
+	dbg("%s - state = %d", __func__, break_state);
 
 	/* chase the port close */
 	chase_port(edge_port, 0, 0);
 
-	if (on == -1)
+	if (break_state == -1)
 		bv = 1;	/* On */
 	status = ti_do_config(edge_port, UMPC_SET_CLR_BREAK, bv);
 	if (status)
diff --git a/drivers/usb/serial/ipaq.c b/drivers/usb/serial/ipaq.c
index 80d9ec5570d6..a7784642d6a1 100644
--- a/drivers/usb/serial/ipaq.c
+++ b/drivers/usb/serial/ipaq.c
@@ -74,19 +74,21 @@ static int connect_retries = KP_RETRIES;
 static int initial_wait;
 
 /* Function prototypes for an ipaq */
-static int  ipaq_open (struct usb_serial_port *port, struct file *filp);
-static void ipaq_close (struct usb_serial_port *port, struct file *filp);
-static int  ipaq_startup (struct usb_serial *serial);
-static void ipaq_shutdown (struct usb_serial *serial);
-static int ipaq_write(struct usb_serial_port *port, const unsigned char *buf,
-		       int count);
+static int  ipaq_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+static void ipaq_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+static int  ipaq_startup(struct usb_serial *serial);
+static void ipaq_shutdown(struct usb_serial *serial);
+static int ipaq_write(struct tty_struct *tty, struct usb_serial_port *port,
+			const unsigned char *buf, int count);
 static int ipaq_write_bulk(struct usb_serial_port *port, const unsigned char *buf,
-			   int count);
+			int count);
 static void ipaq_write_gather(struct usb_serial_port *port);
 static void ipaq_read_bulk_callback (struct urb *urb);
 static void ipaq_write_bulk_callback(struct urb *urb);
-static int ipaq_write_room(struct usb_serial_port *port);
-static int ipaq_chars_in_buffer(struct usb_serial_port *port);
+static int ipaq_write_room(struct tty_struct *tty);
+static int ipaq_chars_in_buffer(struct tty_struct *tty);
 static void ipaq_destroy_lists(struct usb_serial_port *port);
 
 
@@ -591,7 +593,8 @@ static spinlock_t	write_list_lock;
 static int		bytes_in;
 static int		bytes_out;
 
-static int ipaq_open(struct usb_serial_port *port, struct file *filp)
+static int ipaq_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial	*serial = port->serial;
 	struct ipaq_private	*priv;
@@ -637,10 +640,12 @@ static int ipaq_open(struct usb_serial_port *port, struct file *filp)
 	 * discipline instead of queueing.
 	 */
 
-	port->tty->low_latency = 1;
-	port->tty->raw = 1;
-	port->tty->real_raw = 1;
-
+	if (tty) {
+		tty->low_latency = 1;
+		/* FIXME: These two are bogus */
+		tty->raw = 1;
+		tty->real_raw = 1;
+	}
 	/*
 	 * Lose the small buffers usbserial provides. Make larger ones.
 	 */
@@ -714,7 +719,8 @@ error:
 }
 
 
-static void ipaq_close(struct usb_serial_port *port, struct file *filp)
+static void ipaq_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct ipaq_private	*priv = usb_get_serial_port_data(port);
 
@@ -751,7 +757,7 @@ static void ipaq_read_bulk_callback(struct urb *urb)
 
 	usb_serial_debug_data(debug, &port->dev, __func__, urb->actual_length, data);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
@@ -770,8 +776,8 @@ static void ipaq_read_bulk_callback(struct urb *urb)
 	return;
 }
 
-static int ipaq_write(struct usb_serial_port *port, const unsigned char *buf,
-		       int count)
+static int ipaq_write(struct tty_struct *tty, struct usb_serial_port *port,
+			const unsigned char *buf, int count)
 {
 	const unsigned char	*current_position = buf;
 	int			bytes_sent = 0;
@@ -905,16 +911,18 @@ static void ipaq_write_bulk_callback(struct urb *urb)
 	usb_serial_port_softint(port);
 }
 
-static int ipaq_write_room(struct usb_serial_port *port)
+static int ipaq_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ipaq_private	*priv = usb_get_serial_port_data(port);
 
 	dbg("%s - freelen %d", __func__, priv->free_len);
 	return priv->free_len;
 }
 
-static int ipaq_chars_in_buffer(struct usb_serial_port *port)
+static int ipaq_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ipaq_private	*priv = usb_get_serial_port_data(port);
 
 	dbg("%s - queuelen %d", __func__, priv->queue_len);
diff --git a/drivers/usb/serial/ipw.c b/drivers/usb/serial/ipw.c
index bc85ca5c1c37..a89ebfe9e915 100644
--- a/drivers/usb/serial/ipw.c
+++ b/drivers/usb/serial/ipw.c
@@ -179,7 +179,7 @@ static void ipw_read_bulk_callback(struct urb *urb)
 
 	usb_serial_debug_data(debug, &port->dev, __func__, urb->actual_length, data);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
@@ -199,7 +199,8 @@ static void ipw_read_bulk_callback(struct urb *urb)
 	return;
 }
 
-static int ipw_open(struct usb_serial_port *port, struct file *filp)
+static int ipw_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_device *dev = port->serial->dev;
 	u8 buf_flow_static[16] = IPW_BYTES_FLOWINIT;
@@ -212,8 +213,8 @@ static int ipw_open(struct usb_serial_port *port, struct file *filp)
 	if (!buf_flow_init)
 		return -ENOMEM;
 
-	if (port->tty)
-		port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	/* --1: Tell the modem to initialize (we think) From sniffs this is always the
 	 * first thing that gets sent to the modem during opening of the device */
@@ -301,7 +302,8 @@ static int ipw_open(struct usb_serial_port *port, struct file *filp)
 	return 0;
 }
 
-static void ipw_close(struct usb_serial_port *port, struct file * filp)
+static void ipw_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct usb_device *dev = port->serial->dev;
 	int result;
@@ -384,7 +386,8 @@ static void ipw_write_bulk_callback(struct urb *urb)
 	usb_serial_port_softint(port);
 }
 
-static int ipw_write(struct usb_serial_port *port, const unsigned char *buf, int count)
+static int ipw_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 	struct usb_device *dev = port->serial->dev;
 	int ret;
diff --git a/drivers/usb/serial/ir-usb.c b/drivers/usb/serial/ir-usb.c
index 0063c11c8081..e59155c6607d 100644
--- a/drivers/usb/serial/ir-usb.c
+++ b/drivers/usb/serial/ir-usb.c
@@ -85,15 +85,17 @@ static int buffer_size;
 /* if overridden by the user, then use the specified number of XBOFs */
 static int xbof = -1;
 
-static int ir_startup(struct usb_serial *serial);
-static int ir_open(struct usb_serial_port *port, struct file *filep);
-static void ir_close(struct usb_serial_port *port, struct file *filep);
-static int ir_write(struct usb_serial_port *port,
-		const unsigned char *buf, int count);
-static void ir_write_bulk_callback(struct urb *urb);
-static void ir_read_bulk_callback(struct urb *urb);
-static void ir_set_termios(struct usb_serial_port *port,
-		struct ktermios *old_termios);
+static int  ir_startup (struct usb_serial *serial);
+static int  ir_open(struct tty_struct *tty, struct usb_serial_port *port,
+					struct file *filep);
+static void ir_close(struct tty_struct *tty, struct usb_serial_port *port,
+					struct file *filep);
+static int  ir_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count);
+static void ir_write_bulk_callback (struct urb *urb);
+static void ir_read_bulk_callback (struct urb *urb);
+static void ir_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios);
 
 /* Not that this lot means you can only have one per system */
 static u8 ir_baud;
@@ -295,7 +297,8 @@ static int ir_startup(struct usb_serial *serial)
 	return 0;
 }
 
-static int ir_open(struct usb_serial_port *port, struct file *filp)
+static int ir_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	char *buffer;
 	int result = 0;
@@ -343,7 +346,8 @@ static int ir_open(struct usb_serial_port *port, struct file *filp)
 	return result;
 }
 
-static void ir_close(struct usb_serial_port *port, struct file *filp)
+static void ir_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 
@@ -351,8 +355,8 @@ static void ir_close(struct usb_serial_port *port, struct file *filp)
 	usb_kill_urb(port->read_urb);
 }
 
-static int ir_write(struct usb_serial_port *port,
-		const unsigned char *buf, int count)
+static int ir_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 	unsigned char *transfer_buffer;
 	int result;
@@ -360,11 +364,6 @@ static int ir_write(struct usb_serial_port *port,
 
 	dbg("%s - port = %d, count = %d", __func__, port->number, count);
 
-	if (!port->tty) {
-		dev_err(&port->dev, "%s - no tty???\n", __func__);
-		return 0;
-	}
-
 	if (count == 0)
 		return 0;
 
@@ -450,14 +449,13 @@ static void ir_read_bulk_callback(struct urb *urb)
 
 	dbg("%s - port %d", __func__, port->number);
 
-	if (!port->open_count) {
+	if (!port->port.count) {
 		dbg("%s - port closed.", __func__);
 		return;
 	}
 
 	switch (status) {
 	case 0: /* Successful */
-
 		/*
 		 * The first byte of the packet we get from the device
 		 * contains a busy indicator and baud rate change.
@@ -465,19 +463,11 @@ static void ir_read_bulk_callback(struct urb *urb)
 		 */
 		if ((*data & 0x0f) > 0)
 			ir_baud = *data & 0x0f;
-
-		usb_serial_debug_data(
-			debug,
-			&port->dev,
-			__func__,
-			urb->actual_length,
-			data);
-
-		tty = port->tty;
-
+		usb_serial_debug_data(debug, &port->dev, __func__,
+						urb->actual_length, data);
+ 		tty = port->port.tty;
 		if (tty_buffer_request_room(tty, urb->actual_length - 1)) {
-			tty_insert_flip_string(tty, data + 1,
-					urb->actual_length - 1);
+			tty_insert_flip_string(tty, data+1, urb->actual_length - 1);
 			tty_flip_buffer_push(tty);
 		}
 
@@ -488,11 +478,10 @@ static void ir_read_bulk_callback(struct urb *urb)
 		 */
 
 	case -EPROTO: /* taking inspiration from pl2303.c */
-
-		/* Continue trying to always read */
+			/* Continue trying to always read */
 		usb_fill_bulk_urb(
 			port->read_urb,
-			port->serial->dev,
+			port->serial->dev, 
 			usb_rcvbulkpipe(port->serial->dev,
 				port->bulk_in_endpointAddress),
 			port->read_urb->transfer_buffer,
@@ -502,23 +491,19 @@ static void ir_read_bulk_callback(struct urb *urb)
 
 		result = usb_submit_urb(port->read_urb, GFP_ATOMIC);
 		if (result)
-			dev_err(&port->dev,
-				"%s - failed resubmitting read urb, error %d\n",
+			dev_err(&port->dev, "%s - failed resubmitting read urb, error %d\n",
 				__func__, result);
-		break;
-
+			break ;
 	default:
 		dbg("%s - nonzero read bulk status received: %d",
-			__func__,
-			status);
-		break;
+			__func__, status);
+		break ;
 	}
-
 	return;
 }
 
-static void ir_set_termios(struct usb_serial_port *port,
-		struct ktermios *old_termios)
+static void ir_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	unsigned char *transfer_buffer;
 	int result;
@@ -527,7 +512,7 @@ static void ir_set_termios(struct usb_serial_port *port,
 
 	dbg("%s - port %d", __func__, port->number);
 
-	baud = tty_get_baud_rate(port->tty);
+	baud = tty_get_baud_rate(tty);
 
 	/*
 	 * FIXME, we should compare the baud request against the
@@ -600,8 +585,8 @@ static void ir_set_termios(struct usb_serial_port *port,
 				__func__, result);
 
 	/* Only speed changes are supported */
-	tty_termios_copy_hw(port->tty->termios, old_termios);
-	tty_encode_baud_rate(port->tty, baud, baud);
+	tty_termios_copy_hw(tty->termios, old_termios);
+	tty_encode_baud_rate(tty, baud, baud);
 }
 
 static int __init ir_init(void)
diff --git a/drivers/usb/serial/iuu_phoenix.c b/drivers/usb/serial/iuu_phoenix.c
index a01e987c7d32..d65414888349 100644
--- a/drivers/usb/serial/iuu_phoenix.c
+++ b/drivers/usb/serial/iuu_phoenix.c
@@ -144,9 +144,10 @@ static void iuu_shutdown(struct usb_serial *serial)
 	}
 }
 
-static int iuu_tiocmset(struct usb_serial_port *port, struct file *file,
+static int iuu_tiocmset(struct tty_struct *tty, struct file *file,
 			unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct iuu_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -171,8 +172,9 @@ static int iuu_tiocmset(struct usb_serial_port *port, struct file *file,
  * When no card , the reader respond with TIOCM_CD
  * This is known as CD autodetect mechanism
  */
-static int iuu_tiocmget(struct usb_serial_port *port, struct file *file)
+static int iuu_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct iuu_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	int rc;
@@ -630,7 +632,7 @@ static void read_buf_callback(struct urb *urb)
 	}
 
 	dbg("%s - %i chars to write", __func__, urb->actual_length);
-	tty = port->tty;
+	tty = port->port.tty;
 	if (data == NULL)
 		dbg("%s - data is NULL !!!", __func__);
 	if (tty && urb->actual_length && data) {
@@ -752,11 +754,10 @@ static void iuu_uart_read_callback(struct urb *urb)
 	/* if nothing to write call again rxcmd */
 	dbg("%s - rxcmd recall", __func__);
 	iuu_led_activity_off(urb);
-	return;
 }
 
-static int iuu_uart_write(struct usb_serial_port *port, const u8 *buf,
-			  int count)
+static int iuu_uart_write(struct tty_struct *tty, struct usb_serial_port *port,
+			  const u8 *buf, int count)
 {
 	struct iuu_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -948,7 +949,8 @@ static int set_control_lines(struct usb_device *dev, u8 value)
 	return 0;
 }
 
-static void iuu_close(struct usb_serial_port *port, struct file *filp)
+static void iuu_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	/* iuu_led (port,255,0,0,0); */
 	struct usb_serial *serial;
@@ -964,8 +966,8 @@ static void iuu_close(struct usb_serial_port *port, struct file *filp)
 
 	iuu_uart_off(port);
 	if (serial->dev) {
-		if (port->tty) {
-			c_cflag = port->tty->termios->c_cflag;
+		if (tty) {
+			c_cflag = tty->termios->c_cflag;
 			if (c_cflag & HUPCL) {
 				/* drop DTR and RTS */
 				priv = usb_get_serial_port_data(port);
@@ -989,7 +991,8 @@ static void iuu_close(struct usb_serial_port *port, struct file *filp)
 	}
 }
 
-static int iuu_open(struct usb_serial_port *port, struct file *filp)
+static int iuu_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	u8 *buf;
@@ -1036,15 +1039,17 @@ static int iuu_open(struct usb_serial_port *port, struct file *filp)
 
 	/* set the termios structure */
 	spin_lock_irqsave(&priv->lock, flags);
-	if (!priv->termios_initialized) {
-		*(port->tty->termios) = tty_std_termios;
-		port->tty->termios->c_cflag = CLOCAL | CREAD | CS8 | B9600
-						| TIOCM_CTS | CSTOPB | PARENB;
-		port->tty->termios->c_lflag = 0;
-		port->tty->termios->c_oflag = 0;
-		port->tty->termios->c_iflag = 0;
+	if (tty && !priv->termios_initialized) {
+		*(tty->termios) = tty_std_termios;
+		tty->termios->c_cflag = CLOCAL | CREAD | CS8 | B9600
+					| TIOCM_CTS | CSTOPB | PARENB;
+		tty->termios->c_ispeed = 9600;
+		tty->termios->c_ospeed = 9600;
+		tty->termios->c_lflag = 0;
+		tty->termios->c_oflag = 0;
+		tty->termios->c_iflag = 0;
 		priv->termios_initialized = 1;
-		port->tty->low_latency = 1;
+		tty->low_latency = 1;
 		priv->poll = 0;
 	 }
 	spin_unlock_irqrestore(&priv->lock, flags);
@@ -1148,7 +1153,7 @@ static int iuu_open(struct usb_serial_port *port, struct file *filp)
 	if (result) {
 		dev_err(&port->dev, "%s - failed submitting read urb,"
 			" error %d\n", __func__, result);
-		iuu_close(port, NULL);
+		iuu_close(tty, port, NULL);
 		return -EPROTO;
 	} else {
 		dbg("%s - rxcmd OK", __func__);
diff --git a/drivers/usb/serial/keyspan.c b/drivers/usb/serial/keyspan.c
index 11e439b90eac..a371c41bb3ab 100644
--- a/drivers/usb/serial/keyspan.c
+++ b/drivers/usb/serial/keyspan.c
@@ -244,20 +244,9 @@ static void __exit keyspan_exit (void)
 module_init(keyspan_init);
 module_exit(keyspan_exit);
 
-static void keyspan_rx_throttle (struct usb_serial_port *port)
-{
-	dbg("%s - port %d", __func__, port->number);
-}
-
-
-static void keyspan_rx_unthrottle (struct usb_serial_port *port)
-{
-	dbg("%s - port %d", __func__, port->number);
-}
-
-
-static void keyspan_break_ctl (struct usb_serial_port *port, int break_state)
+static void keyspan_break_ctl(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct keyspan_port_private 	*p_priv;
 
  	dbg("%s", __func__);
@@ -273,14 +262,13 @@ static void keyspan_break_ctl (struct usb_serial_port *port, int break_state)
 }
 
 
-static void keyspan_set_termios (struct usb_serial_port *port, 
-				     struct ktermios *old_termios)
+static void keyspan_set_termios (struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	int				baud_rate, device_port;
 	struct keyspan_port_private 	*p_priv;
 	const struct keyspan_device_details	*d_details;
 	unsigned int 			cflag;
-	struct tty_struct		*tty = port->tty;
 
 	dbg("%s", __func__);
 
@@ -312,12 +300,11 @@ static void keyspan_set_termios (struct usb_serial_port *port,
 	keyspan_send_setup(port, 0);
 }
 
-static int keyspan_tiocmget(struct usb_serial_port *port, struct file *file)
+static int keyspan_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
+	struct keyspan_port_private *p_priv = usb_get_serial_port_data(port);
 	unsigned int			value;
-	struct keyspan_port_private 	*p_priv;
-
-	p_priv = usb_get_serial_port_data(port);
 	
 	value = ((p_priv->rts_state) ? TIOCM_RTS : 0) |
 		((p_priv->dtr_state) ? TIOCM_DTR : 0) |
@@ -329,18 +316,16 @@ static int keyspan_tiocmget(struct usb_serial_port *port, struct file *file)
 	return value;
 }
 
-static int keyspan_tiocmset(struct usb_serial_port *port, struct file *file,
+static int keyspan_tiocmset(struct tty_struct *tty, struct file *file,
 			    unsigned int set, unsigned int clear)
 {
-	struct keyspan_port_private 	*p_priv;
-
-	p_priv = usb_get_serial_port_data(port);
+	struct usb_serial_port *port = tty->driver_data;
+	struct keyspan_port_private *p_priv = usb_get_serial_port_data(port);
 	
 	if (set & TIOCM_RTS)
 		p_priv->rts_state = 1;
 	if (set & TIOCM_DTR)
 		p_priv->dtr_state = 1;
-
 	if (clear & TIOCM_RTS)
 		p_priv->rts_state = 0;
 	if (clear & TIOCM_DTR)
@@ -349,16 +334,10 @@ static int keyspan_tiocmset(struct usb_serial_port *port, struct file *file,
 	return 0;
 }
 
-static int keyspan_ioctl(struct usb_serial_port *port, struct file *file,
-			     unsigned int cmd, unsigned long arg)
-{
-	return -ENOIOCTLCMD;
-}
-
-	/* Write function is similar for the four protocols used
-	   with only a minor change for usa90 (usa19hs) required */
-static int keyspan_write(struct usb_serial_port *port, 
-			 const unsigned char *buf, int count)
+/* Write function is similar for the four protocols used
+   with only a minor change for usa90 (usa19hs) required */
+static int keyspan_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count)
 {
 	struct keyspan_port_private 	*p_priv;
 	const struct keyspan_device_details	*d_details;
@@ -448,7 +427,7 @@ static void	usa26_indat_callback(struct urb *urb)
 	}
 
 	port =  urb->context;
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		/* 0x80 bit is error flag */
 		if ((data[0] & 0x80) == 0) {
@@ -479,7 +458,7 @@ static void	usa26_indat_callback(struct urb *urb)
 				
 		/* Resubmit urb so we continue receiving */
 	urb->dev = port->serial->dev;
-	if (port->open_count)
+	if (port->port.count)
 		if ((err = usb_submit_urb(urb, GFP_ATOMIC)) != 0) {
 			dbg("%s - resubmit read urb failed. (%d)", __func__, err);
 		}
@@ -496,7 +475,7 @@ static void	usa2x_outdat_callback(struct urb *urb)
 	p_priv = usb_get_serial_port_data(port);
 	dbg ("%s - urb %d", __func__, urb == p_priv->out_urbs[1]);
 
-	if (port->open_count)
+	if (port->port.count)
 		usb_serial_port_softint(port);
 }
 
@@ -567,10 +546,10 @@ static void	usa26_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->gpia_dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->tty && !C_CLOCAL(port->tty)
+	if (port->port.tty && !C_CLOCAL(port->port.tty)
 	    && old_dcd_state != p_priv->dcd_state) {
 		if (old_dcd_state)
-			tty_hangup(port->tty);
+			tty_hangup(port->port.tty);
 		/*  else */
 		/*	wake_up_interruptible(&p_priv->open_wait); */
 	}
@@ -619,7 +598,7 @@ static void usa28_indat_callback(struct urb *urb)
 		p_priv = usb_get_serial_port_data(port);
 		data = urb->transfer_buffer;
 
-		tty = port->tty;
+		tty = port->port.tty;
 		if (urb->actual_length) {
 			for (i = 0; i < urb->actual_length ; ++i) {
 				tty_insert_flip_char(tty, data[i], 0);
@@ -629,7 +608,7 @@ static void usa28_indat_callback(struct urb *urb)
 
 		/* Resubmit urb so we continue receiving */
 		urb->dev = port->serial->dev;
-		if (port->open_count)
+		if (port->port.count)
 			if ((err = usb_submit_urb(urb, GFP_ATOMIC)) != 0) {
 				dbg("%s - resubmit read urb failed. (%d)", __func__, err);
 			}
@@ -704,10 +683,10 @@ static void	usa28_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->tty && !C_CLOCAL(port->tty)
+	if (port->port.tty && !C_CLOCAL(port->port.tty)
 	    && old_dcd_state != p_priv->dcd_state) {
 		if (old_dcd_state)
-			tty_hangup(port->tty);
+			tty_hangup(port->port.tty);
 		/*  else */
 		/*	wake_up_interruptible(&p_priv->open_wait); */
 	}
@@ -797,10 +776,10 @@ static void	usa49_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->tty && !C_CLOCAL(port->tty)
+	if (port->port.tty && !C_CLOCAL(port->port.tty)
 	    && old_dcd_state != p_priv->dcd_state) {
 		if (old_dcd_state)
-			tty_hangup(port->tty);
+			tty_hangup(port->port.tty);
 		/*  else */
 		/*	wake_up_interruptible(&p_priv->open_wait); */
 	}
@@ -839,7 +818,7 @@ static void	usa49_indat_callback(struct urb *urb)
 	}
 
 	port =  urb->context;
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		/* 0x80 bit is error flag */
 		if ((data[0] & 0x80) == 0) {
@@ -866,7 +845,7 @@ static void	usa49_indat_callback(struct urb *urb)
 				
 		/* Resubmit urb so we continue receiving */
 	urb->dev = port->serial->dev;
-	if (port->open_count)
+	if (port->port.count)
 		if ((err = usb_submit_urb(urb, GFP_ATOMIC)) != 0) {
 			dbg("%s - resubmit read urb failed. (%d)", __func__, err);
 		}
@@ -904,7 +883,7 @@ static void usa49wg_indat_callback(struct urb *urb)
 				return;
 			}
 			port = serial->port[data[i++]];
-			tty = port->tty;
+			tty = port->port.tty;
 			len = data[i++];
 
 			/* 0x80 bit is error flag */
@@ -912,7 +891,7 @@ static void usa49wg_indat_callback(struct urb *urb)
 				/* no error on any byte */
 				i++;
 				for (x = 1; x < len ; ++x)
-					if (port->open_count)
+					if (port->port.count)
 						tty_insert_flip_char(tty,
 								data[i++], 0);
 					else
@@ -930,13 +909,13 @@ static void usa49wg_indat_callback(struct urb *urb)
 					if (stat & RXERROR_PARITY)
 						flag |= TTY_PARITY;
 					/* XXX should handle break (0x10) */
-					if (port->open_count)
+					if (port->port.count)
 						tty_insert_flip_char(tty,
 							data[i+1], flag);
 					i += 2;
 				}
 			}
-			if (port->open_count)
+			if (port->port.count)
 				tty_flip_buffer_push(tty);
 		}
 	}
@@ -978,7 +957,7 @@ static void usa90_indat_callback(struct urb *urb)
 	port =  urb->context;
 	p_priv = usb_get_serial_port_data(port);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (urb->actual_length) {
 	
 		/* if current mode is DMA, looks like usa28 format
@@ -1021,7 +1000,7 @@ static void usa90_indat_callback(struct urb *urb)
 				
 	/* Resubmit urb so we continue receiving */
 	urb->dev = port->serial->dev;
-	if (port->open_count)
+	if (port->port.count)
 		if ((err = usb_submit_urb(urb, GFP_ATOMIC)) != 0) {
 			dbg("%s - resubmit read urb failed. (%d)", __func__, err);
 		}
@@ -1064,10 +1043,10 @@ static void	usa90_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->tty && !C_CLOCAL(port->tty)
+	if (port->port.tty && !C_CLOCAL(port->port.tty)
 	    && old_dcd_state != p_priv->dcd_state) {
 		if (old_dcd_state)
-			tty_hangup(port->tty);
+			tty_hangup(port->port.tty);
 		/*  else */
 		/*	wake_up_interruptible(&p_priv->open_wait); */
 	}
@@ -1139,10 +1118,10 @@ static void	usa67_instat_callback(struct urb *urb)
 	p_priv->cts_state = ((msg->hskia_cts) ? 1 : 0);
 	p_priv->dcd_state = ((msg->gpia_dcd) ? 1 : 0);
 
-	if (port->tty && !C_CLOCAL(port->tty)
+	if (port->port.tty && !C_CLOCAL(port->port.tty)
 	    && old_dcd_state != p_priv->dcd_state) {
 		if (old_dcd_state)
-			tty_hangup(port->tty);
+			tty_hangup(port->port.tty);
 		/*  else */
 		/*	wake_up_interruptible(&p_priv->open_wait); */
 	}
@@ -1177,8 +1156,9 @@ static void usa67_glocont_callback(struct urb *urb)
 	}
 }
 
-static int keyspan_write_room (struct usb_serial_port *port)
+static int keyspan_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct keyspan_port_private	*p_priv;
 	const struct keyspan_device_details	*d_details;
 	int				flip;
@@ -1210,13 +1190,8 @@ static int keyspan_write_room (struct usb_serial_port *port)
 }
 
 
-static int keyspan_chars_in_buffer (struct usb_serial_port *port)
-{
-	return 0;
-}
-
-
-static int keyspan_open (struct usb_serial_port *port, struct file *filp)
+static int keyspan_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct keyspan_port_private 	*p_priv;
 	struct keyspan_serial_private 	*s_priv;
@@ -1225,7 +1200,7 @@ static int keyspan_open (struct usb_serial_port *port, struct file *filp)
 	int				i, err;
 	int				baud_rate, device_port;
 	struct urb			*urb;
-	unsigned int			cflag;
+	unsigned int			cflag = 0;
 
 	s_priv = usb_get_serial_data(serial);
 	p_priv = usb_get_serial_port_data(port);
@@ -1271,19 +1246,19 @@ static int keyspan_open (struct usb_serial_port *port, struct file *filp)
 	/* get the terminal config for the setup message now so we don't
 	 * need to send 2 of them */
 
-	cflag = port->tty->termios->c_cflag;
 	device_port = port->number - port->serial->minor;
-
-	/* Baud rate calculation takes baud rate as an integer
-	   so other rates can be generated if desired. */
-	baud_rate = tty_get_baud_rate(port->tty);
-	/* If no match or invalid, leave as default */
-	if (baud_rate >= 0
-	    && d_details->calculate_baud_rate(baud_rate, d_details->baudclk,
-				NULL, NULL, NULL, device_port) == KEYSPAN_BAUD_RATE_OK) {
-		p_priv->baud = baud_rate;
+	if (tty) {
+		cflag = tty->termios->c_cflag;
+		/* Baud rate calculation takes baud rate as an integer
+		   so other rates can be generated if desired. */
+		baud_rate = tty_get_baud_rate(tty);
+		/* If no match or invalid, leave as default */
+		if (baud_rate >= 0
+		    && d_details->calculate_baud_rate(baud_rate, d_details->baudclk,
+					NULL, NULL, NULL, device_port) == KEYSPAN_BAUD_RATE_OK) {
+			p_priv->baud = baud_rate;
+		}
 	}
-
 	/* set CTS/RTS handshake etc. */
 	p_priv->cflag = cflag;
 	p_priv->flow_control = (cflag & CRTSCTS)? flow_cts: flow_none;
@@ -1301,7 +1276,8 @@ static inline void stop_urb(struct urb *urb)
 		usb_kill_urb(urb);
 }
 
-static void keyspan_close(struct usb_serial_port *port, struct file *filp)
+static void keyspan_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	int			i;
 	struct usb_serial	*serial = port->serial;
@@ -1338,7 +1314,7 @@ static void keyspan_close(struct usb_serial_port *port, struct file *filp)
 			stop_urb(p_priv->out_urbs[i]);
 		}
 	}
-	port->tty = NULL;
+	port->port.tty = NULL;
 }
 
 	/* download the firmware to a pre-renumeration device */
@@ -2427,7 +2403,7 @@ static int keyspan_usa90_send_setup(struct usb_serial *serial,
 	}
 	/* Sending intermediate configs */
 	else {
-		if (port->open_count)
+		if (port->port.count)
 			msg.portEnabled = 1;
 		msg.txBreak = (p_priv->break_on);
 	}
diff --git a/drivers/usb/serial/keyspan.h b/drivers/usb/serial/keyspan.h
index 8bf72639b148..38b4582e0734 100644
--- a/drivers/usb/serial/keyspan.h
+++ b/drivers/usb/serial/keyspan.h
@@ -35,17 +35,18 @@
 
 
 /* Function prototypes for Keyspan serial converter */
-static int  keyspan_open		(struct usb_serial_port *port,
+static int  keyspan_open		(struct tty_struct *tty,
+					 struct usb_serial_port *port,
 					 struct file *filp);
-static void keyspan_close		(struct usb_serial_port *port,
+static void keyspan_close		(struct tty_struct *tty,
+					 struct usb_serial_port *port,
 					 struct file *filp);
 static int  keyspan_startup		(struct usb_serial *serial);
 static void keyspan_shutdown		(struct usb_serial *serial);
-static void keyspan_rx_throttle		(struct usb_serial_port *port);
-static void keyspan_rx_unthrottle	(struct usb_serial_port *port);
-static int  keyspan_write_room		(struct usb_serial_port *port);
+static int  keyspan_write_room		(struct tty_struct *tty);
 
-static int  keyspan_write		(struct usb_serial_port *port,
+static int  keyspan_write		(struct tty_struct *tty,
+					 struct usb_serial_port *port,
 					 const unsigned char *buf,
 					 int count);
 
@@ -53,18 +54,14 @@ static void keyspan_send_setup		(struct usb_serial_port *port,
 					 int reset_port);
 
 
-static int  keyspan_chars_in_buffer 	(struct usb_serial_port *port);
-static int  keyspan_ioctl		(struct usb_serial_port *port,
-					 struct file *file,
-					 unsigned int cmd,
-					 unsigned long arg);
-static void keyspan_set_termios		(struct usb_serial_port *port,
+static void keyspan_set_termios		(struct tty_struct *tty,
+					 struct usb_serial_port *port,
 					 struct ktermios *old);
-static void keyspan_break_ctl		(struct usb_serial_port *port,
+static void keyspan_break_ctl		(struct tty_struct *tty,
 					 int break_state);
-static int  keyspan_tiocmget		(struct usb_serial_port *port,
+static int  keyspan_tiocmget		(struct tty_struct *tty,
 					 struct file *file);
-static int  keyspan_tiocmset		(struct usb_serial_port *port,
+static int  keyspan_tiocmset		(struct tty_struct *tty,
 					 struct file *file, unsigned int set,
 					 unsigned int clear);
 static int  keyspan_fake_startup	(struct usb_serial *serial);
@@ -567,10 +564,6 @@ static struct usb_serial_driver keyspan_1port_device = {
 	.close			= keyspan_close,
 	.write			= keyspan_write,
 	.write_room		= keyspan_write_room,
-	.chars_in_buffer	= keyspan_chars_in_buffer,
-	.throttle		= keyspan_rx_throttle,
-	.unthrottle		= keyspan_rx_unthrottle,
-	.ioctl			= keyspan_ioctl,
 	.set_termios		= keyspan_set_termios,
 	.break_ctl		= keyspan_break_ctl,
 	.tiocmget		= keyspan_tiocmget,
@@ -591,10 +584,6 @@ static struct usb_serial_driver keyspan_2port_device = {
 	.close			= keyspan_close,
 	.write			= keyspan_write,
 	.write_room		= keyspan_write_room,
-	.chars_in_buffer	= keyspan_chars_in_buffer,
-	.throttle		= keyspan_rx_throttle,
-	.unthrottle		= keyspan_rx_unthrottle,
-	.ioctl			= keyspan_ioctl,
 	.set_termios		= keyspan_set_termios,
 	.break_ctl		= keyspan_break_ctl,
 	.tiocmget		= keyspan_tiocmget,
@@ -615,10 +604,6 @@ static struct usb_serial_driver keyspan_4port_device = {
 	.close			= keyspan_close,
 	.write			= keyspan_write,
 	.write_room		= keyspan_write_room,
-	.chars_in_buffer	= keyspan_chars_in_buffer,
-	.throttle		= keyspan_rx_throttle,
-	.unthrottle		= keyspan_rx_unthrottle,
-	.ioctl			= keyspan_ioctl,
 	.set_termios		= keyspan_set_termios,
 	.break_ctl		= keyspan_break_ctl,
 	.tiocmget		= keyspan_tiocmget,
diff --git a/drivers/usb/serial/keyspan_pda.c b/drivers/usb/serial/keyspan_pda.c
index 60b3e22bd633..24a08ac2e4ee 100644
--- a/drivers/usb/serial/keyspan_pda.c
+++ b/drivers/usb/serial/keyspan_pda.c
@@ -171,7 +171,7 @@ static void keyspan_pda_wakeup_write(struct work_struct *work)
 		container_of(work, struct keyspan_pda_private, wakeup_work);
 	struct usb_serial_port *port = priv->port;
 
-	tty_wakeup(port->tty);
+	tty_wakeup(port->port.tty);
 }
 
 static void keyspan_pda_request_unthrottle(struct work_struct *work)
@@ -203,7 +203,7 @@ static void keyspan_pda_request_unthrottle(struct work_struct *work)
 static void keyspan_pda_rx_interrupt (struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
-       	struct tty_struct *tty = port->tty;
+       	struct tty_struct *tty = port->port.tty;
 	unsigned char *data = urb->transfer_buffer;
 	int i;
 	int retval;
@@ -266,7 +266,7 @@ exit:
 }
 
 
-static void keyspan_pda_rx_throttle (struct usb_serial_port *port)
+static void keyspan_pda_rx_throttle(struct tty_struct *tty)
 {
 	/* stop receiving characters. We just turn off the URB request, and
 	   let chars pile up in the device. If we're doing hardware
@@ -274,14 +274,15 @@ static void keyspan_pda_rx_throttle (struct usb_serial_port *port)
 	   fills up. If we're doing XON/XOFF, this would be a good time to
 	   send an XOFF, although it might make sense to foist that off
 	   upon the device too. */
-
+	struct usb_serial_port *port = tty->driver_data;
 	dbg("keyspan_pda_rx_throttle port %d", port->number);
 	usb_kill_urb(port->interrupt_in_urb);
 }
 
 
-static void keyspan_pda_rx_unthrottle (struct usb_serial_port *port)
+static void keyspan_pda_rx_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	/* just restart the receive interrupt URB */
 	dbg("keyspan_pda_rx_unthrottle port %d", port->number);
 	port->interrupt_in_urb->dev = port->serial->dev;
@@ -330,8 +331,9 @@ static speed_t keyspan_pda_setbaud (struct usb_serial *serial, speed_t baud)
 }
 
 
-static void keyspan_pda_break_ctl (struct usb_serial_port *port, int break_state)
+static void keyspan_pda_break_ctl(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	int value;
 	int result;
@@ -354,8 +356,8 @@ static void keyspan_pda_break_ctl (struct usb_serial_port *port, int break_state
 }
 
 
-static void keyspan_pda_set_termios (struct usb_serial_port *port, 
-				     struct ktermios *old_termios)
+static void keyspan_pda_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct usb_serial *serial = port->serial;
 	speed_t speed;
@@ -380,7 +382,7 @@ static void keyspan_pda_set_termios (struct usb_serial_port *port,
 
 	   For now, just do baud. */
 
-	speed = tty_get_baud_rate(port->tty);
+	speed = tty_get_baud_rate(tty);
 	speed = keyspan_pda_setbaud(serial, speed);
 
 	if (speed == 0) {
@@ -390,8 +392,8 @@ static void keyspan_pda_set_termios (struct usb_serial_port *port,
 	}
 	/* Only speed can change so copy the old h/w parameters
 	   then encode the new speed */
-	tty_termios_copy_hw(port->tty->termios, old_termios);
-	tty_encode_baud_rate(port->tty, speed, speed);
+	tty_termios_copy_hw(tty->termios, old_termios);
+	tty_encode_baud_rate(tty, speed, speed);
 }
 
 
@@ -425,8 +427,9 @@ static int keyspan_pda_set_modem_info(struct usb_serial *serial,
 	return rc;
 }
 
-static int keyspan_pda_tiocmget(struct usb_serial_port *port, struct file *file)
+static int keyspan_pda_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	int rc;
 	unsigned char status;
@@ -445,9 +448,10 @@ static int keyspan_pda_tiocmget(struct usb_serial_port *port, struct file *file)
 	return value;
 }
 
-static int keyspan_pda_tiocmset(struct usb_serial_port *port, struct file *file,
+static int keyspan_pda_tiocmset(struct tty_struct *tty, struct file *file,
 				unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	int rc;
 	unsigned char status;
@@ -469,23 +473,8 @@ static int keyspan_pda_tiocmset(struct usb_serial_port *port, struct file *file,
 	return rc;
 }
 
-static int keyspan_pda_ioctl(struct usb_serial_port *port, struct file *file,
-			     unsigned int cmd, unsigned long arg)
-{
-	switch (cmd) {
-	case TIOCMIWAIT:
-		/* wait for any of the 4 modem inputs (DCD,RI,DSR,CTS)*/
-		/* TODO */
-	case TIOCGICOUNT:
-		/* return count of modemline transitions */
-		return 0; /* TODO */
-	}
-	
-	return -ENOIOCTLCMD;
-}
-
-static int keyspan_pda_write(struct usb_serial_port *port, 
-			     const unsigned char *buf, int count)
+static int keyspan_pda_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count)
 {
 	struct usb_serial *serial = port->serial;
 	int request_unthrottle = 0;
@@ -607,22 +596,21 @@ static void keyspan_pda_write_bulk_callback (struct urb *urb)
 }
 
 
-static int keyspan_pda_write_room (struct usb_serial_port *port)
+static int keyspan_pda_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct keyspan_pda_private *priv;
-
 	priv = usb_get_serial_port_data(port);
-
 	/* used by n_tty.c for processing of tabs and such. Giving it our
 	   conservative guess is probably good enough, but needs testing by
 	   running a console through the device. */
-
 	return (priv->tx_room);
 }
 
 
-static int keyspan_pda_chars_in_buffer (struct usb_serial_port *port)
+static int keyspan_pda_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct keyspan_pda_private *priv;
 	unsigned long flags;
 	int ret = 0;
@@ -640,7 +628,8 @@ static int keyspan_pda_chars_in_buffer (struct usb_serial_port *port)
 }
 
 
-static int keyspan_pda_open (struct usb_serial_port *port, struct file *filp)
+static int keyspan_pda_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	unsigned char room;
@@ -672,7 +661,7 @@ static int keyspan_pda_open (struct usb_serial_port *port, struct file *filp)
 
 	/* the normal serial device seems to always turn on DTR and RTS here,
 	   so do the same */
-	if (port->tty->termios->c_cflag & CBAUD)
+	if (tty && (tty->termios->c_cflag & CBAUD))
 		keyspan_pda_set_modem_info(serial, (1<<7) | (1<<2) );
 	else
 		keyspan_pda_set_modem_info(serial, 0);
@@ -690,13 +679,14 @@ error:
 }
 
 
-static void keyspan_pda_close(struct usb_serial_port *port, struct file *filp)
+static void keyspan_pda_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 
 	if (serial->dev) {
 		/* the normal serial device seems to always shut off DTR and RTS now */
-		if (port->tty->termios->c_cflag & HUPCL)
+		if (tty->termios->c_cflag & HUPCL)
 			keyspan_pda_set_modem_info(serial, 0);
 
 		/* shutdown our bulk reads and writes */
@@ -832,7 +822,6 @@ static struct usb_serial_driver keyspan_pda_device = {
 	.chars_in_buffer =	keyspan_pda_chars_in_buffer,
 	.throttle =		keyspan_pda_rx_throttle,
 	.unthrottle =		keyspan_pda_rx_unthrottle,
-	.ioctl =		keyspan_pda_ioctl,
 	.set_termios =		keyspan_pda_set_termios,
 	.break_ctl =		keyspan_pda_break_ctl,
 	.tiocmget =		keyspan_pda_tiocmget,
diff --git a/drivers/usb/serial/kl5kusb105.c b/drivers/usb/serial/kl5kusb105.c
index 79787eda9524..4a38ec8f5fe3 100644
--- a/drivers/usb/serial/kl5kusb105.c
+++ b/drivers/usb/serial/kl5kusb105.c
@@ -74,29 +74,33 @@ static int debug;
  */
 static int  klsi_105_startup	         (struct usb_serial *serial);
 static void klsi_105_shutdown	         (struct usb_serial *serial);
-static int  klsi_105_open	         (struct usb_serial_port *port,
+static int  klsi_105_open	         (struct tty_struct *tty,
+					  struct usb_serial_port *port,
 					  struct file *filp);
-static void klsi_105_close	         (struct usb_serial_port *port,
+static void klsi_105_close	         (struct tty_struct *tty,
+					  struct usb_serial_port *port,
 					  struct file *filp);
-static int  klsi_105_write	         (struct usb_serial_port *port,
+static int  klsi_105_write	         (struct tty_struct *tty,
+					  struct usb_serial_port *port,
 					  const unsigned char *buf,
 					  int count);
 static void klsi_105_write_bulk_callback (struct urb *urb);
-static int  klsi_105_chars_in_buffer     (struct usb_serial_port *port);
-static int  klsi_105_write_room          (struct usb_serial_port *port);
+static int  klsi_105_chars_in_buffer     (struct tty_struct *tty);
+static int  klsi_105_write_room          (struct tty_struct *tty);
 
 static void klsi_105_read_bulk_callback  (struct urb *urb);
-static void klsi_105_set_termios         (struct usb_serial_port *port,
+static void klsi_105_set_termios         (struct tty_struct *tty,
+					  struct usb_serial_port *port,
 					  struct ktermios *old);
-static void klsi_105_throttle		 (struct usb_serial_port *port);
-static void klsi_105_unthrottle		 (struct usb_serial_port *port);
+static void klsi_105_throttle		 (struct tty_struct *tty);
+static void klsi_105_unthrottle		 (struct tty_struct *tty);
 /*
-static void klsi_105_break_ctl	         (struct usb_serial_port *port,
+static void klsi_105_break_ctl	         (struct tty_struct *tty,
 					  int break_state );
  */
-static int  klsi_105_tiocmget	         (struct usb_serial_port *port,
+static int  klsi_105_tiocmget	         (struct tty_struct *tty,
 					  struct file *file);
-static int  klsi_105_tiocmset	         (struct usb_serial_port *port,
+static int  klsi_105_tiocmset	         (struct tty_struct *tty,
 					  struct file *file, unsigned int set,
 					  unsigned int clear);
 
@@ -361,7 +365,8 @@ static void klsi_105_shutdown (struct usb_serial *serial)
 	}
 } /* klsi_105_shutdown */
 
-static int  klsi_105_open (struct usb_serial_port *port, struct file *filp)
+static int  klsi_105_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct klsi_105_private *priv = usb_get_serial_port_data(port);
 	int retval = 0;
@@ -375,7 +380,7 @@ static int  klsi_105_open (struct usb_serial_port *port, struct file *filp)
 
 	/* force low_latency on so that our tty_push actually forces
 	 * the data through
-	 * port->tty->low_latency = 1; */
+	 * tty->low_latency = 1; */
 
 	/* Do a defined restart:
 	 * Set up sane default baud rate and send the 'READ_ON'
@@ -393,12 +398,12 @@ static int  klsi_105_open (struct usb_serial_port *port, struct file *filp)
 	
 	/* set up termios structure */
 	spin_lock_irqsave (&priv->lock, flags);
-	priv->termios.c_iflag = port->tty->termios->c_iflag;
-	priv->termios.c_oflag = port->tty->termios->c_oflag;
-	priv->termios.c_cflag = port->tty->termios->c_cflag;
-	priv->termios.c_lflag = port->tty->termios->c_lflag;
+	priv->termios.c_iflag = tty->termios->c_iflag;
+	priv->termios.c_oflag = tty->termios->c_oflag;
+	priv->termios.c_cflag = tty->termios->c_cflag;
+	priv->termios.c_lflag = tty->termios->c_lflag;
 	for (i=0; i<NCCS; i++)
-		priv->termios.c_cc[i] = port->tty->termios->c_cc[i];
+		priv->termios.c_cc[i] = tty->termios->c_cc[i];
 	priv->cfg.pktlen   = cfg.pktlen;
 	priv->cfg.baudrate = cfg.baudrate;
 	priv->cfg.databits = cfg.databits;
@@ -452,7 +457,8 @@ exit:
 } /* klsi_105_open */
 
 
-static void klsi_105_close (struct usb_serial_port *port, struct file *filp)
+static void klsi_105_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct klsi_105_private *priv = usb_get_serial_port_data(port);
 	int rc;
@@ -493,8 +499,8 @@ static void klsi_105_close (struct usb_serial_port *port, struct file *filp)
 #define KLSI_105_DATA_OFFSET	2   /* in the bulk urb data block */
 
 
-static int klsi_105_write (struct usb_serial_port *port,
-			   const unsigned char *buf, int count)
+static int klsi_105_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count)
 {
 	struct klsi_105_private *priv = usb_get_serial_port_data(port);
 	int result, size;
@@ -584,8 +590,9 @@ static void klsi_105_write_bulk_callback ( struct urb *urb)
 
 
 /* return number of characters currently in the writing process */
-static int klsi_105_chars_in_buffer (struct usb_serial_port *port)
+static int klsi_105_chars_in_buffer (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int chars = 0;
 	int i;
 	unsigned long flags;
@@ -605,8 +612,9 @@ static int klsi_105_chars_in_buffer (struct usb_serial_port *port)
 	return (chars);
 }
 
-static int klsi_105_write_room (struct usb_serial_port *port)
+static int klsi_105_write_room (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned long flags;
 	int i;
 	int room = 0;
@@ -660,7 +668,7 @@ static void klsi_105_read_bulk_callback (struct urb *urb)
 	} else {
 		int bytes_sent = ((__u8 *) data)[0] +
 				 ((unsigned int) ((__u8 *) data)[1] << 8);
-		tty = port->tty;
+		tty = port->port.tty;
 		/* we should immediately resubmit the URB, before attempting
 		 * to pass the data on to the tty layer. But that needs locking
 		 * against re-entry an then mixed-up data because of
@@ -699,11 +707,11 @@ static void klsi_105_read_bulk_callback (struct urb *urb)
 } /* klsi_105_read_bulk_callback */
 
 
-static void klsi_105_set_termios (struct usb_serial_port *port,
+static void klsi_105_set_termios (struct tty_struct *tty,
+				  struct usb_serial_port *port,
 				  struct ktermios *old_termios)
 {
 	struct klsi_105_private *priv = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	unsigned int iflag = tty->termios->c_iflag;
 	unsigned int old_iflag = old_termios->c_iflag;
 	unsigned int cflag = tty->termios->c_cflag;
@@ -863,8 +871,9 @@ static void klsi_105_set_termios (struct usb_serial_port *port,
 
 
 #if 0
-static void mct_u232_break_ctl( struct usb_serial_port *port, int break_state )
+static void mct_u232_break_ctl( struct tty_struct *tty, int break_state )
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	struct mct_u232_private *priv = (struct mct_u232_private *)port->private;
 	unsigned char lcr = priv->last_lcr;
@@ -878,8 +887,9 @@ static void mct_u232_break_ctl( struct usb_serial_port *port, int break_state )
 } /* mct_u232_break_ctl */
 #endif
 
-static int klsi_105_tiocmget (struct usb_serial_port *port, struct file *file)
+static int klsi_105_tiocmget (struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct klsi_105_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	int rc;
@@ -900,7 +910,7 @@ static int klsi_105_tiocmget (struct usb_serial_port *port, struct file *file)
 	return (int)line_state;
 }
 
-static int klsi_105_tiocmset (struct usb_serial_port *port, struct file *file,
+static int klsi_105_tiocmset (struct tty_struct *tty, struct file *file,
 			      unsigned int set, unsigned int clear)
 {
 	int retval = -EINVAL;
@@ -929,14 +939,16 @@ static int klsi_105_tiocmset (struct usb_serial_port *port, struct file *file,
 	return retval;
 }
 
-static void klsi_105_throttle (struct usb_serial_port *port)
+static void klsi_105_throttle (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s - port %d", __func__, port->number);
 	usb_kill_urb(port->read_urb);
 }
 
-static void klsi_105_unthrottle (struct usb_serial_port *port)
+static void klsi_105_unthrottle (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int result;
 
 	dbg("%s - port %d", __func__, port->number);
diff --git a/drivers/usb/serial/kobil_sct.c b/drivers/usb/serial/kobil_sct.c
index 693f00da7c03..40c67f0096b1 100644
--- a/drivers/usb/serial/kobil_sct.c
+++ b/drivers/usb/serial/kobil_sct.c
@@ -70,19 +70,22 @@ static int debug;
 /* Function prototypes */
 static int  kobil_startup (struct usb_serial *serial);
 static void kobil_shutdown (struct usb_serial *serial);
-static int  kobil_open (struct usb_serial_port *port, struct file *filp);
-static void kobil_close (struct usb_serial_port *port, struct file *filp);
-static int  kobil_write (struct usb_serial_port *port, 
+static int  kobil_open (struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+static void kobil_close (struct tty_struct *tty, struct usb_serial_port *port,
+			struct file *filp);
+static int  kobil_write (struct tty_struct *tty, struct usb_serial_port *port, 
 			 const unsigned char *buf, int count);
-static int  kobil_write_room(struct usb_serial_port *port);
-static int  kobil_ioctl(struct usb_serial_port *port, struct file *file,
+static int  kobil_write_room(struct tty_struct *tty);
+static int  kobil_ioctl(struct tty_struct *tty, struct file *file,
 			unsigned int cmd, unsigned long arg);
-static int  kobil_tiocmget(struct usb_serial_port *port, struct file *file);
-static int  kobil_tiocmset(struct usb_serial_port *port, struct file *file,
+static int  kobil_tiocmget(struct tty_struct *tty, struct file *file);
+static int  kobil_tiocmset(struct tty_struct *tty, struct file *file,
 			   unsigned int set, unsigned int clear);
 static void kobil_read_int_callback( struct urb *urb );
 static void kobil_write_callback( struct urb *purb );
-static void kobil_set_termios(struct usb_serial_port *port, struct ktermios *old);
+static void kobil_set_termios(struct tty_struct *tty, 
+			struct usb_serial_port *port, struct ktermios *old);
 
 
 static struct usb_device_id id_table [] = {
@@ -201,8 +204,8 @@ static void kobil_shutdown (struct usb_serial *serial)
 	dbg("%s - port %d", __func__, serial->port[0]->number);
 
 	for (i=0; i < serial->num_ports; ++i) {
-		while (serial->port[i]->open_count > 0) {
-			kobil_close (serial->port[i], NULL);
+		while (serial->port[i]->port.count > 0) {
+			kobil_close (NULL, serial->port[i], NULL);
 		}
 		kfree(usb_get_serial_port_data(serial->port[i]));
 		usb_set_serial_port_data(serial->port[i], NULL);
@@ -210,7 +213,8 @@ static void kobil_shutdown (struct usb_serial *serial)
 }
 
 
-static int kobil_open (struct usb_serial_port *port, struct file *filp)
+static int kobil_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	int result = 0;
 	struct kobil_private *priv;
@@ -229,14 +233,15 @@ static int kobil_open (struct usb_serial_port *port, struct file *filp)
 	 * the data through, otherwise it is scheduled, and with high
 	 * data rates (like with OHCI) data can get lost.
 	 */
-	port->tty->low_latency = 1;
-
-	// without this, every push_tty_char is echoed :-(  
-	port->tty->termios->c_lflag = 0;
-	port->tty->termios->c_lflag &= ~(ISIG | ICANON | ECHO | IEXTEN | XCASE);
-	port->tty->termios->c_iflag = IGNBRK | IGNPAR | IXOFF;
-	port->tty->termios->c_oflag &= ~ONLCR; // do NOT translate CR to CR-NL (0x0A -> 0x0A 0x0D)
-	
+	if (tty) {
+		tty->low_latency = 1;
+
+		/* Default to echo off and other sane device settings */
+		tty->termios->c_lflag = 0;
+		tty->termios->c_lflag &= ~(ISIG | ICANON | ECHO | IEXTEN | XCASE);
+		tty->termios->c_iflag = IGNBRK | IGNPAR | IXOFF;
+		tty->termios->c_oflag &= ~ONLCR; // do NOT translate CR to CR-NL (0x0A -> 0x0A 0x0D)
+	}	
 	// allocate memory for transfer buffer
 	transfer_buffer = kzalloc(transfer_buffer_length, GFP_KERNEL);
 	if (! transfer_buffer) {
@@ -330,7 +335,8 @@ static int kobil_open (struct usb_serial_port *port, struct file *filp)
 }
 
 
-static void kobil_close (struct usb_serial_port *port, struct file *filp)
+static void kobil_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 
@@ -360,7 +366,7 @@ static void kobil_read_int_callback(struct urb *urb)
 		return;
 	}
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (urb->actual_length) {
 
 		// BEGIN DEBUG
@@ -395,7 +401,7 @@ static void kobil_write_callback( struct urb *purb )
 }
 
 
-static int kobil_write (struct usb_serial_port *port, 
+static int kobil_write (struct tty_struct *tty, struct usb_serial_port *port, 
 			const unsigned char *buf, int count)
 {
 	int length = 0;
@@ -417,12 +423,9 @@ static int kobil_write (struct usb_serial_port *port,
 
 	// Copy data to buffer
 	memcpy (priv->buf + priv->filled, buf, count);
-
 	usb_serial_debug_data(debug, &port->dev, __func__, count, priv->buf + priv->filled);
-
 	priv->filled = priv->filled + count;
 
-
 	// only send complete block. TWIN, KAAN SIM and adapter K use the same protocol.
 	if ( ((priv->device_type != KOBIL_ADAPTER_B_PRODUCT_ID) && (priv->filled > 2) && (priv->filled >= (priv->buf[1] + 3))) || 
 	     ((priv->device_type == KOBIL_ADAPTER_B_PRODUCT_ID) && (priv->filled > 3) && (priv->filled >= (priv->buf[2] + 4))) ) {
@@ -478,15 +481,17 @@ static int kobil_write (struct usb_serial_port *port,
 }
 
 
-static int kobil_write_room (struct usb_serial_port *port)
+static int kobil_write_room (struct tty_struct *tty)
 {
 	//dbg("%s - port %d", __func__, port->number);
+	/* FIXME */
 	return 8;
 }
 
 
-static int kobil_tiocmget(struct usb_serial_port *port, struct file *file)
+static int kobil_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct kobil_private * priv;
 	int result;
 	unsigned char *transfer_buffer;
@@ -524,9 +529,10 @@ static int kobil_tiocmget(struct usb_serial_port *port, struct file *file)
 	return result;
 }
 
-static int  kobil_tiocmset(struct usb_serial_port *port, struct file *file,
+static int kobil_tiocmset(struct tty_struct *tty, struct file *file,
 			   unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct kobil_private * priv;
 	int result;
 	int dtr = 0;
@@ -590,12 +596,13 @@ static int  kobil_tiocmset(struct usb_serial_port *port, struct file *file,
 	return (result < 0) ? result : 0;
 }
 
-static void kobil_set_termios(struct usb_serial_port *port, struct ktermios *old)
+static void kobil_set_termios(struct tty_struct *tty,
+			struct usb_serial_port *port, struct ktermios *old)
 {
 	struct kobil_private * priv;
 	int result;
 	unsigned short urb_val = 0;
-	int c_cflag = port->tty->termios->c_cflag;
+	int c_cflag = tty->termios->c_cflag;
 	speed_t speed;
 	void * settings;
 
@@ -604,7 +611,7 @@ static void kobil_set_termios(struct usb_serial_port *port, struct ktermios *old
 		// This device doesn't support ioctl calls
 		return;
 
-	switch (speed = tty_get_baud_rate(port->tty)) {
+	switch (speed = tty_get_baud_rate(tty)) {
 		case 1200:
 			urb_val = SUSBCR_SBR_1200;
 			break;
@@ -634,8 +641,8 @@ static void kobil_set_termios(struct usb_serial_port *port, struct ktermios *old
 		urb_val |= SUSBCR_SPASB_NoParity;
 		strcat(settings, "No Parity");
 	}
-	port->tty->termios->c_cflag &= ~CMSPAR;
-	tty_encode_baud_rate(port->tty, speed, speed);
+	tty->termios->c_cflag &= ~CMSPAR;
+	tty_encode_baud_rate(tty, speed, speed);
 
 	result = usb_control_msg( port->serial->dev,
 				  usb_rcvctrlpipe(port->serial->dev, 0 ),
@@ -650,8 +657,9 @@ static void kobil_set_termios(struct usb_serial_port *port, struct ktermios *old
 	kfree(settings);
 }
 
-static int kobil_ioctl(struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg)
+static int kobil_ioctl(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct kobil_private * priv = usb_get_serial_port_data(port);
 	unsigned char *transfer_buffer;
 	int transfer_buffer_length = 8;
@@ -662,7 +670,7 @@ static int kobil_ioctl(struct usb_serial_port *port, struct file * file, unsigne
 		return 0;
 
 	switch (cmd) {
-	case TCFLSH:   // 0x540B
+	case TCFLSH:
 		transfer_buffer = kmalloc(transfer_buffer_length, GFP_KERNEL);
 		if (! transfer_buffer)
 		 	return -ENOBUFS;
@@ -680,7 +688,7 @@ static int kobil_ioctl(struct usb_serial_port *port, struct file * file, unsigne
 		
 		dbg("%s - port %d Send reset_all_queues (FLUSH) URB returns: %i", __func__, port->number, result);
 		kfree(transfer_buffer);
-		return (result < 0) ? -EFAULT : 0;
+		return (result < 0) ? -EIO: 0;
 	default:
 		return -ENOIOCTLCMD;
 	}
diff --git a/drivers/usb/serial/mct_u232.c b/drivers/usb/serial/mct_u232.c
index 5fc2cef30e39..7bce4302a5f9 100644
--- a/drivers/usb/serial/mct_u232.c
+++ b/drivers/usb/serial/mct_u232.c
@@ -92,26 +92,25 @@ static int debug;
  */
 static int  mct_u232_startup	         (struct usb_serial *serial);
 static void mct_u232_shutdown	         (struct usb_serial *serial);
-static int  mct_u232_open	         (struct usb_serial_port *port,
+static int  mct_u232_open	         (struct tty_struct *tty,
+					  struct usb_serial_port *port,
 					  struct file *filp);
-static void mct_u232_close	         (struct usb_serial_port *port,
+static void mct_u232_close	         (struct tty_struct *tty,
+					  struct usb_serial_port *port,
 					  struct file *filp);
 static void mct_u232_read_int_callback   (struct urb *urb);
-static void mct_u232_set_termios         (struct usb_serial_port *port,
+static void mct_u232_set_termios         (struct tty_struct *tty,
+					  struct usb_serial_port *port,
 					  struct ktermios * old);
-static int  mct_u232_ioctl	         (struct usb_serial_port *port,
-					  struct file * file,
-					  unsigned int cmd,
-					  unsigned long arg);
-static void mct_u232_break_ctl	         (struct usb_serial_port *port,
+static void mct_u232_break_ctl	         (struct tty_struct *tty,
 					  int break_state );
-static int  mct_u232_tiocmget		 (struct usb_serial_port *port,
+static int  mct_u232_tiocmget		 (struct tty_struct *tty,
 					  struct file *file);
-static int  mct_u232_tiocmset		 (struct usb_serial_port *port,
+static int  mct_u232_tiocmset		 (struct tty_struct *tty,
 					  struct file *file, unsigned int set,
 					  unsigned int clear);
-static void mct_u232_throttle		 (struct usb_serial_port *port);
-static void mct_u232_unthrottle		 (struct usb_serial_port *port);
+static void mct_u232_throttle		 (struct tty_struct *tty);
+static void mct_u232_unthrottle		 (struct tty_struct *tty);
 
 
 /*
@@ -149,7 +148,6 @@ static struct usb_serial_driver mct_u232_device = {
 	.throttle =	     mct_u232_throttle,
 	.unthrottle =	     mct_u232_unthrottle,
 	.read_int_callback = mct_u232_read_int_callback,
-	.ioctl =	     mct_u232_ioctl,
 	.set_termios =	     mct_u232_set_termios,
 	.break_ctl =	     mct_u232_break_ctl,
 	.tiocmget =	     mct_u232_tiocmget,
@@ -224,8 +222,8 @@ static int mct_u232_calculate_baud_rate(struct usb_serial *serial, speed_t value
 	}
 }
 
-static int mct_u232_set_baud_rate(struct usb_serial *serial, struct usb_serial_port *port,
-				  speed_t value)
+static int mct_u232_set_baud_rate(struct tty_struct *tty,
+	struct usb_serial *serial, struct usb_serial_port *port, speed_t value)
 {
 	__le32 divisor;
         int rc;
@@ -243,7 +241,7 @@ static int mct_u232_set_baud_rate(struct usb_serial *serial, struct usb_serial_p
 	if (rc < 0)	/*FIXME: What value speed results */
 		err("Set BAUD RATE %d failed (error = %d)", value, rc);
 	else
-		tty_encode_baud_rate(port->tty, speed, speed);
+		tty_encode_baud_rate(tty, speed, speed);
 	dbg("set_baud_rate: value: 0x%x, divisor: 0x%x", value, divisor);
 
 	/* Mimic the MCT-supplied Windows driver (version 1.21P.0104), which
@@ -272,7 +270,7 @@ static int mct_u232_set_baud_rate(struct usb_serial *serial, struct usb_serial_p
 		err("Sending USB device request code %d failed (error = %d)", 
 		    MCT_U232_SET_UNKNOWN1_REQUEST, rc);
 
-	if (port && C_CRTSCTS(port->tty)) {
+	if (port && C_CRTSCTS(tty)) {
 	   cts_enable_byte = 1;
 	}
 
@@ -411,7 +409,8 @@ static void mct_u232_shutdown (struct usb_serial *serial)
 	}
 } /* mct_u232_shutdown */
 
-static int  mct_u232_open (struct usb_serial_port *port, struct file *filp)
+static int  mct_u232_open (struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
@@ -437,7 +436,7 @@ static int  mct_u232_open (struct usb_serial_port *port, struct file *filp)
 	 * either.
 	 */
 	spin_lock_irqsave(&priv->lock, flags);
-	if (port->tty->termios->c_cflag & CBAUD)
+	if (tty && (tty->termios->c_cflag & CBAUD))
 		priv->control_state = TIOCM_DTR | TIOCM_RTS;
 	else
 		priv->control_state = 0;
@@ -481,15 +480,16 @@ error:
 } /* mct_u232_open */
 
 
-static void mct_u232_close (struct usb_serial_port *port, struct file *filp)
+static void mct_u232_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	unsigned int c_cflag;
 	unsigned int control_state;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
 	dbg("%s port %d", __func__, port->number);
 
-   	if (port->tty) {
-		c_cflag = port->tty->termios->c_cflag;
+   	if (tty) {
+		c_cflag = tty->termios->c_cflag;
 		mutex_lock(&port->serial->disc_mutex);
 		if (c_cflag & HUPCL && !port->serial->disconnected) {
 			/* drop DTR and RTS */
@@ -553,7 +553,7 @@ static void mct_u232_read_int_callback (struct urb *urb)
 	 */
 	if (urb->transfer_buffer_length > 2) {
 		int i;
-		tty = port->tty;
+		tty = port->port.tty;
 		if (urb->actual_length) {
 			for (i = 0; i < urb->actual_length ; ++i) {
 				tty_insert_flip_char(tty, data[i], 0);
@@ -583,7 +583,7 @@ static void mct_u232_read_int_callback (struct urb *urb)
 	 * to look in to this before committing any code.
 	 */
 	if (priv->last_lsr & MCT_U232_LSR_ERR) {
-		tty = port->tty;
+		tty = port->port.tty;
 		/* Overrun Error */
 		if (priv->last_lsr & MCT_U232_LSR_OE) {
 		}
@@ -606,12 +606,13 @@ exit:
 		     __func__, retval);
 } /* mct_u232_read_int_callback */
 
-static void mct_u232_set_termios (struct usb_serial_port *port,
+static void mct_u232_set_termios (struct tty_struct *tty,
+				  struct usb_serial_port *port,
 				  struct ktermios *old_termios)
 {
 	struct usb_serial *serial = port->serial;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
-	struct ktermios *termios = port->tty->termios;
+	struct ktermios *termios = tty->termios;
 	unsigned int cflag = termios->c_cflag;
 	unsigned int old_cflag = old_termios->c_cflag;
 	unsigned long flags;
@@ -638,7 +639,7 @@ static void mct_u232_set_termios (struct usb_serial_port *port,
 		mct_u232_set_modem_ctrl(serial, control_state);
 	}
 
-	mct_u232_set_baud_rate(serial, port, tty_get_baud_rate(port->tty));
+	mct_u232_set_baud_rate(tty, serial, port, tty_get_baud_rate(tty));
 
 	if ((cflag & CBAUD) == B0 ) {
 		dbg("%s: baud is B0", __func__);
@@ -689,8 +690,9 @@ static void mct_u232_set_termios (struct usb_serial_port *port,
 	spin_unlock_irqrestore(&priv->lock, flags);
 } /* mct_u232_set_termios */
 
-static void mct_u232_break_ctl( struct usb_serial_port *port, int break_state )
+static void mct_u232_break_ctl(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
 	unsigned char lcr;
@@ -709,8 +711,9 @@ static void mct_u232_break_ctl( struct usb_serial_port *port, int break_state )
 } /* mct_u232_break_ctl */
 
 
-static int mct_u232_tiocmget (struct usb_serial_port *port, struct file *file)
+static int mct_u232_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
 	unsigned int control_state;
 	unsigned long flags;
@@ -724,9 +727,10 @@ static int mct_u232_tiocmget (struct usb_serial_port *port, struct file *file)
 	return control_state;
 }
 
-static int mct_u232_tiocmset (struct usb_serial_port *port, struct file *file,
+static int mct_u232_tiocmset(struct tty_struct *tty, struct file *file,
 			      unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
 	unsigned int control_state;
@@ -751,73 +755,46 @@ static int mct_u232_tiocmset (struct usb_serial_port *port, struct file *file,
 	return mct_u232_set_modem_ctrl(serial, control_state);
 }
 
-static int mct_u232_ioctl (struct usb_serial_port *port, struct file * file,
-			   unsigned int cmd, unsigned long arg)
-{
-	dbg("%scmd=0x%x", __func__, cmd);
-
-	/* Based on code from acm.c and others */
-	switch (cmd) {
-	case TIOCMIWAIT:
-		/* wait for any of the 4 modem inputs (DCD,RI,DSR,CTS)*/
-		/* TODO */
-		return( 0 );
-
-	case TIOCGICOUNT:
-		/* return count of modemline transitions */
-		/* TODO */
-		return 0;
-
-	default:
-		dbg("%s: arg not supported - 0x%04x", __func__,cmd);
-		return(-ENOIOCTLCMD);
-		break;
-	}
-	return 0;
-} /* mct_u232_ioctl */
-
-static void mct_u232_throttle (struct usb_serial_port *port)
+static void mct_u232_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	unsigned int control_state;
-	struct tty_struct *tty;
 
-	tty = port->tty;
 	dbg("%s - port %d", __func__, port->number);
 
 	spin_lock_irqsave(&priv->lock, flags);
 	priv->rx_flags |= THROTTLED;
 	if (C_CRTSCTS(tty)) {
-	  priv->control_state &= ~TIOCM_RTS;
-	  control_state = priv->control_state;
-	  spin_unlock_irqrestore(&priv->lock, flags);
-	  (void) mct_u232_set_modem_ctrl(port->serial, control_state);
+		priv->control_state &= ~TIOCM_RTS;
+		control_state = priv->control_state;
+		spin_unlock_irqrestore(&priv->lock, flags);
+		(void) mct_u232_set_modem_ctrl(port->serial, control_state);
 	} else {
-	  spin_unlock_irqrestore(&priv->lock, flags);
+		spin_unlock_irqrestore(&priv->lock, flags);
 	}
 }
 
 
-static void mct_u232_unthrottle (struct usb_serial_port *port)
+static void mct_u232_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct mct_u232_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	unsigned int control_state;
-	struct tty_struct *tty;
 
 	dbg("%s - port %d", __func__, port->number);
 
-	tty = port->tty;
 	spin_lock_irqsave(&priv->lock, flags);
 	if ((priv->rx_flags & THROTTLED) && C_CRTSCTS(tty)) {
-	  priv->rx_flags &= ~THROTTLED;
-	  priv->control_state |= TIOCM_RTS;
-	  control_state = priv->control_state;
-	  spin_unlock_irqrestore(&priv->lock, flags);
-	  (void) mct_u232_set_modem_ctrl(port->serial, control_state);
+		priv->rx_flags &= ~THROTTLED;
+		priv->control_state |= TIOCM_RTS;
+		control_state = priv->control_state;
+		spin_unlock_irqrestore(&priv->lock, flags);
+		(void) mct_u232_set_modem_ctrl(port->serial, control_state);
 	} else {
-	  spin_unlock_irqrestore(&priv->lock, flags);
+		spin_unlock_irqrestore(&priv->lock, flags);
 	}
 }
 
diff --git a/drivers/usb/serial/mos7720.c b/drivers/usb/serial/mos7720.c
index 50f1fe263338..d47f0814ce2d 100644
--- a/drivers/usb/serial/mos7720.c
+++ b/drivers/usb/serial/mos7720.c
@@ -218,7 +218,7 @@ static void mos7720_bulk_in_callback(struct urb *urb)
 
 	data = urb->transfer_buffer;
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
@@ -264,7 +264,7 @@ static void mos7720_bulk_out_data_callback(struct urb *urb)
 
 	dbg("Entering .........");
 
-	tty = mos7720_port->port->tty;
+	tty = mos7720_port->port->port.tty;
 
 	if (tty && mos7720_port->open)
 		tty_wakeup(tty);
@@ -320,7 +320,8 @@ static int send_mos_cmd(struct usb_serial *serial, __u8 request, __u16 value,
 	return status;
 }
 
-static int mos7720_open(struct usb_serial_port *port, struct file * filp)
+static int mos7720_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct usb_serial *serial;
 	struct usb_serial_port *port0;
@@ -443,14 +444,12 @@ static int mos7720_open(struct usb_serial_port *port, struct file * filp)
 	data = 0x0c;
 	send_mos_cmd(serial, MOS_WRITE, port_number, 0x01, &data);
 
-//Matrix
-
 	/* force low_latency on so that our tty_push actually forces *
 	 * the data through,otherwise it is scheduled, and with      *
 	 * high data rates (like with OHCI) data can get lost.       */
 
-	if (port->tty)
-		port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	/* see if we've set up our endpoint info yet   *
 	 * (can't set it up in mos7720_startup as the  *
@@ -515,8 +514,9 @@ static int mos7720_open(struct usb_serial_port *port, struct file * filp)
  *	system,
  *	Otherwise we return a negative error number.
  */
-static int mos7720_chars_in_buffer(struct usb_serial_port *port)
+static int mos7720_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int i;
 	int chars = 0;
 	struct moschip_port *mos7720_port;
@@ -537,7 +537,8 @@ static int mos7720_chars_in_buffer(struct usb_serial_port *port)
 	return chars;
 }
 
-static void mos7720_close(struct usb_serial_port *port, struct file *filp)
+static void mos7720_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial;
 	struct moschip_port *mos7720_port;
@@ -588,8 +589,9 @@ static void mos7720_close(struct usb_serial_port *port, struct file *filp)
 	dbg("Leaving %s", __func__);
 }
 
-static void mos7720_break(struct usb_serial_port *port, int break_state)
+static void mos7720_break(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
         unsigned char data;
 	struct usb_serial *serial;
 	struct moschip_port *mos7720_port;
@@ -621,8 +623,9 @@ static void mos7720_break(struct usb_serial_port *port, int break_state)
  *	If successful, we return the amount of room that we have for this port
  *	Otherwise we return a negative error number.
  */
-static int mos7720_write_room(struct usb_serial_port *port)
+static int mos7720_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct moschip_port *mos7720_port;
 	int room = 0;
 	int i;
@@ -645,8 +648,8 @@ static int mos7720_write_room(struct usb_serial_port *port)
 	return room;
 }
 
-static int mos7720_write(struct usb_serial_port *port,
-			 const unsigned char *data, int count)
+static int mos7720_write(struct tty_struct *tty, struct usb_serial_port *port,
+				 const unsigned char *data, int count)
 {
 	int status;
 	int i;
@@ -719,10 +722,10 @@ exit:
 	return bytes_sent;
 }
 
-static void mos7720_throttle(struct usb_serial_port *port)
+static void mos7720_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct moschip_port *mos7720_port;
-	struct tty_struct *tty;
 	int status;
 
 	dbg("%s- port %d\n", __func__, port->number);
@@ -739,16 +742,10 @@ static void mos7720_throttle(struct usb_serial_port *port)
 
 	dbg("%s: Entering ..........", __func__);
 
-	tty = port->tty;
-	if (!tty) {
-		dbg("%s - no tty available", __func__);
-		return;
-	}
-
 	/* if we are implementing XON/XOFF, send the stop character */
 	if (I_IXOFF(tty)) {
 		unsigned char stop_char = STOP_CHAR(tty);
-		status = mos7720_write(port, &stop_char, 1);
+		status = mos7720_write(tty, port, &stop_char, 1);
 		if (status <= 0)
 			return;
 	}
@@ -764,11 +761,11 @@ static void mos7720_throttle(struct usb_serial_port *port)
 	}
 }
 
-static void mos7720_unthrottle(struct usb_serial_port *port)
+static void mos7720_unthrottle(struct tty_struct *tty)
 {
-	struct tty_struct *tty;
-	int status;
+	struct usb_serial_port *port = tty->driver_data;
 	struct moschip_port *mos7720_port = usb_get_serial_port_data(port);
+	int status;
 
 	if (mos7720_port == NULL)
 		return;
@@ -780,16 +777,10 @@ static void mos7720_unthrottle(struct usb_serial_port *port)
 
 	dbg("%s: Entering ..........", __func__);
 
-	tty = port->tty;
-	if (!tty) {
-		dbg("%s - no tty available", __func__);
-		return;
-	}
-
 	/* if we are implementing XON/XOFF, send the start character */
 	if (I_IXOFF(tty)) {
 		unsigned char start_char = START_CHAR(tty);
-		status = mos7720_write(port, &start_char, 1);
+		status = mos7720_write(tty, port, &start_char, 1);
 		if (status <= 0)
 			return;
 	}
@@ -1011,12 +1002,12 @@ static int send_cmd_write_baud_rate(struct moschip_port *mos7720_port,
  *	This routine is called to set the UART on the device to match
  *      the specified new settings.
  */
-static void change_port_settings(struct moschip_port *mos7720_port,
+static void change_port_settings(struct tty_struct *tty,
+				 struct moschip_port *mos7720_port,
 				 struct ktermios *old_termios)
 {
 	struct usb_serial_port *port;
 	struct usb_serial *serial;
-	struct tty_struct *tty;
 	int baud;
 	unsigned cflag;
 	unsigned iflag;
@@ -1042,8 +1033,6 @@ static void change_port_settings(struct moschip_port *mos7720_port,
 		return;
 	}
 
-	tty = mos7720_port->port->tty;
-
 	dbg("%s: Entering ..........", __func__);
 
 	lData = UART_LCR_WLEN8;
@@ -1198,14 +1187,13 @@ static void change_port_settings(struct moschip_port *mos7720_port,
  *	this function is called by the tty driver when it wants to change the
  *	termios structure.
  */
-static void mos7720_set_termios(struct usb_serial_port *port,
-				struct ktermios *old_termios)
+static void mos7720_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	int status;
 	unsigned int cflag;
 	struct usb_serial *serial;
 	struct moschip_port *mos7720_port;
-	struct tty_struct *tty;
 
 	serial = port->serial;
 
@@ -1214,9 +1202,6 @@ static void mos7720_set_termios(struct usb_serial_port *port,
 	if (mos7720_port == NULL)
 		return;
 
-	tty = port->tty;
-
-
 	if (!mos7720_port->open) {
 		dbg("%s - port not opened", __func__);
 		return;
@@ -1237,7 +1222,7 @@ static void mos7720_set_termios(struct usb_serial_port *port,
 	dbg("%s - port %d", __func__, port->number);
 
 	/* change the port settings to the new ones specified */
-	change_port_settings(mos7720_port, old_termios);
+	change_port_settings(tty, mos7720_port, old_termios);
 
 	if(!port->read_urb) {
 		dbg("%s","URB KILLED !!!!!\n");
@@ -1264,13 +1249,13 @@ static void mos7720_set_termios(struct usb_serial_port *port,
  * 	    transmit holding register is empty.  This functionality
  * 	    allows an RS485 driver to be written in user space.
  */
-static int get_lsr_info(struct moschip_port *mos7720_port,
+static int get_lsr_info(struct tty_struct *tty, struct moschip_port *mos7720_port,
 			unsigned int __user *value)
 {
 	int count;
 	unsigned int result = 0;
 
-	count = mos7720_chars_in_buffer(mos7720_port->port);
+	count = mos7720_chars_in_buffer(tty);
 	if (count == 0) {
 		dbg("%s -- Empty", __func__);
 		result = TIOCSER_TEMT;
@@ -1290,7 +1275,7 @@ static int get_number_bytes_avail(struct moschip_port *mos7720_port,
 				  unsigned int __user *value)
 {
 	unsigned int result = 0;
-	struct tty_struct *tty = mos7720_port->port->tty;
+	struct tty_struct *tty = mos7720_port->port->port.tty;
 
 	if (!tty)
 		return -ENOIOCTLCMD;
@@ -1407,9 +1392,10 @@ static int get_serial_info(struct moschip_port *mos7720_port,
 	return 0;
 }
 
-static int mos7720_ioctl(struct usb_serial_port *port, struct file *file,
+static int mos7720_ioctl(struct tty_struct *tty, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct moschip_port *mos7720_port;
 	struct async_icount cnow;
 	struct async_icount cprev;
@@ -1431,9 +1417,10 @@ static int mos7720_ioctl(struct usb_serial_port *port, struct file *file,
 
 	case TIOCSERGETLSR:
 		dbg("%s (%d) TIOCSERGETLSR", __func__,  port->number);
-		return get_lsr_info(mos7720_port, (unsigned int __user *)arg);
+		return get_lsr_info(tty, mos7720_port, (unsigned int __user *)arg);
 		return 0;
 
+	/* FIXME: These should be using the mode methods */
 	case TIOCMBIS:
 	case TIOCMBIC:
 	case TIOCMSET:
@@ -1452,10 +1439,6 @@ static int mos7720_ioctl(struct usb_serial_port *port, struct file *file,
 		return get_serial_info(mos7720_port,
 				       (struct serial_struct __user *)arg);
 
-	case TIOCSSERIAL:
-		dbg("%s (%d) TIOCSSERIAL", __func__,  port->number);
-		break;
-
 	case TIOCMIWAIT:
 		dbg("%s (%d) TIOCMIWAIT", __func__,  port->number);
 		cprev = mos7720_port->icount;
diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c
index 78f2f6db494d..2b1fded6619d 100644
--- a/drivers/usb/serial/mos7840.c
+++ b/drivers/usb/serial/mos7840.c
@@ -710,7 +710,7 @@ static void mos7840_bulk_in_callback(struct urb *urb)
 	dbg("%s", "Entering ........... \n");
 
 	if (urb->actual_length) {
-		tty = mos7840_port->port->tty;
+		tty = mos7840_port->port->port.tty;
 		if (tty) {
 			tty_buffer_request_room(tty, urb->actual_length);
 			tty_insert_flip_string(tty, data, urb->actual_length);
@@ -774,7 +774,7 @@ static void mos7840_bulk_out_data_callback(struct urb *urb)
 
 	dbg("%s \n", "Entering .........");
 
-	tty = mos7840_port->port->tty;
+	tty = mos7840_port->port->port.tty;
 
 	if (tty && mos7840_port->open)
 		tty_wakeup(tty);
@@ -804,7 +804,8 @@ static int mos7840_serial_probe(struct usb_serial *serial,
  *	Otherwise we return a negative error number.
  *****************************************************************************/
 
-static int mos7840_open(struct usb_serial_port *port, struct file *filp)
+static int mos7840_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	int response;
 	int j;
@@ -1008,8 +1009,8 @@ static int mos7840_open(struct usb_serial_port *port, struct file *filp)
 	 * the data through,otherwise it is scheduled, and with      *
 	 * high data rates (like with OHCI) data can get lost.       */
 
-	if (port->tty)
-		port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 /* Check to see if we've set up our endpoint info yet    *
      * (can't set it up in mos7840_startup as the structures *
      * were not set up at that time.)                        */
@@ -1104,11 +1105,12 @@ static int mos7840_open(struct usb_serial_port *port, struct file *filp)
  *	been written, but hasn't made it out the port yet)
  *	If successful, we return the number of bytes left to be written in the
  *	system,
- *	Otherwise we return a negative error number.
+ *	Otherwise we return zero.
  *****************************************************************************/
 
-static int mos7840_chars_in_buffer(struct usb_serial_port *port)
+static int mos7840_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int i;
 	int chars = 0;
 	unsigned long flags;
@@ -1118,21 +1120,19 @@ static int mos7840_chars_in_buffer(struct usb_serial_port *port)
 
 	if (mos7840_port_paranoia_check(port, __func__)) {
 		dbg("%s", "Invalid port \n");
-		return -1;
+		return 0;
 	}
 
 	mos7840_port = mos7840_get_port_private(port);
 	if (mos7840_port == NULL) {
 		dbg("%s \n", "mos7840_break:leaving ...........");
-		return -1;
+		return 0;
 	}
 
 	spin_lock_irqsave(&mos7840_port->pool_lock,flags);
-	for (i = 0; i < NUM_URBS; ++i) {
-		if (mos7840_port->busy[i]) {
+	for (i = 0; i < NUM_URBS; ++i)
+		if (mos7840_port->busy[i])
 			chars += URB_TRANSFER_BUFFER_SIZE;
-		}
-	}
 	spin_unlock_irqrestore(&mos7840_port->pool_lock,flags);
 	dbg("%s - returns %d", __func__, chars);
 	return chars;
@@ -1149,7 +1149,8 @@ static int mos7840_chars_in_buffer(struct usb_serial_port *port)
  *		3. A timeout of 3 seconds without activity has expired
  *
  ************************************************************************/
-static void mos7840_block_until_tx_empty(struct moschip_port *mos7840_port)
+static void mos7840_block_until_tx_empty(struct tty_struct *tty,
+				struct moschip_port *mos7840_port)
 {
 	int timeout = HZ / 10;
 	int wait = 30;
@@ -1157,7 +1158,7 @@ static void mos7840_block_until_tx_empty(struct moschip_port *mos7840_port)
 
 	while (1) {
 
-		count = mos7840_chars_in_buffer(mos7840_port->port);
+		count = mos7840_chars_in_buffer(tty);
 
 		/* Check for Buffer status */
 		if (count <= 0) {
@@ -1185,7 +1186,8 @@ static void mos7840_block_until_tx_empty(struct moschip_port *mos7840_port)
  *	this function is called by the tty driver when a port is closed
  *****************************************************************************/
 
-static void mos7840_close(struct usb_serial_port *port, struct file *filp)
+static void mos7840_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial;
 	struct moschip_port *mos7840_port;
@@ -1226,20 +1228,17 @@ static void mos7840_close(struct usb_serial_port *port, struct file *filp)
 		}
 	}
 
-	if (serial->dev) {
+	if (serial->dev)
 		/* flush and block until tx is empty */
-		mos7840_block_until_tx_empty(mos7840_port);
-	}
+		mos7840_block_until_tx_empty(tty, mos7840_port);
 
 	/* While closing port, shutdown all bulk read, write  *
 	 * and interrupt read if they exists                  */
 	if (serial->dev) {
-
 		if (mos7840_port->write_urb) {
 			dbg("%s", "Shutdown bulk write\n");
 			usb_kill_urb(mos7840_port->write_urb);
 		}
-
 		if (mos7840_port->read_urb) {
 			dbg("%s", "Shutdown bulk read\n");
 			usb_kill_urb(mos7840_port->read_urb);
@@ -1247,11 +1246,10 @@ static void mos7840_close(struct usb_serial_port *port, struct file *filp)
 		if ((&mos7840_port->control_urb)) {
 			dbg("%s", "Shutdown control read\n");
 			//      usb_kill_urb (mos7840_port->control_urb);
-
 		}
 	}
-//              if(mos7840_port->ctrl_buf != NULL)
-//                      kfree(mos7840_port->ctrl_buf);
+//      if(mos7840_port->ctrl_buf != NULL)
+//              kfree(mos7840_port->ctrl_buf);
 	port0->open_ports--;
 	dbg("mos7840_num_open_ports in close%d:in port%d\n",
 	    port0->open_ports, port->number);
@@ -1293,15 +1291,15 @@ static void mos7840_close(struct usb_serial_port *port, struct file *filp)
  *
  ************************************************************************/
 
-static void mos7840_block_until_chase_response(struct moschip_port
-					       *mos7840_port)
+static void mos7840_block_until_chase_response(struct tty_struct *tty,
+					struct moschip_port *mos7840_port)
 {
 	int timeout = 1 * HZ;
 	int wait = 10;
 	int count;
 
 	while (1) {
-		count = mos7840_chars_in_buffer(mos7840_port->port);
+		count = mos7840_chars_in_buffer(tty);
 
 		/* Check for Buffer status */
 		if (count <= 0) {
@@ -1328,8 +1326,9 @@ static void mos7840_block_until_chase_response(struct moschip_port
  * mos7840_break
  *	this function sends a break to the port
  *****************************************************************************/
-static void mos7840_break(struct usb_serial_port *port, int break_state)
+static void mos7840_break(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned char data;
 	struct usb_serial *serial;
 	struct moschip_port *mos7840_port;
@@ -1354,17 +1353,14 @@ static void mos7840_break(struct usb_serial_port *port, int break_state)
 		return;
 	}
 
-	if (serial->dev) {
-
+	if (serial->dev)
 		/* flush and block until tx is empty */
-		mos7840_block_until_chase_response(mos7840_port);
-	}
+		mos7840_block_until_chase_response(tty, mos7840_port);
 
-	if (break_state == -1) {
+	if (break_state == -1)
 		data = mos7840_port->shadowLCR | LCR_SET_BREAK;
-	} else {
+	else
 		data = mos7840_port->shadowLCR & ~LCR_SET_BREAK;
-	}
 
 	mos7840_port->shadowLCR = data;
 	dbg("mcs7840_break mos7840_port->shadowLCR is %x\n",
@@ -1383,8 +1379,9 @@ static void mos7840_break(struct usb_serial_port *port, int break_state)
  *	Otherwise we return a negative error number.
  *****************************************************************************/
 
-static int mos7840_write_room(struct usb_serial_port *port)
+static int mos7840_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int i;
 	int room = 0;
 	unsigned long flags;
@@ -1426,7 +1423,7 @@ static int mos7840_write_room(struct usb_serial_port *port)
  *      return a negative error number.
  *****************************************************************************/
 
-static int mos7840_write(struct usb_serial_port *port,
+static int mos7840_write(struct tty_struct *tty, struct usb_serial_port *port,
 			 const unsigned char *data, int count)
 {
 	int status;
@@ -1555,8 +1552,7 @@ static int mos7840_write(struct usb_serial_port *port,
 	mos7840_port->icount.tx += transfer_size;
 	smp_wmb();
 	dbg("mos7840_port->icount.tx is %d:\n", mos7840_port->icount.tx);
-      exit:
-
+exit:
 	return bytes_sent;
 
 }
@@ -1567,10 +1563,10 @@ static int mos7840_write(struct usb_serial_port *port,
  *	being read from the port.
  *****************************************************************************/
 
-static void mos7840_throttle(struct usb_serial_port *port)
+static void mos7840_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct moschip_port *mos7840_port;
-	struct tty_struct *tty;
 	int status;
 
 	if (mos7840_port_paranoia_check(port, __func__)) {
@@ -1592,32 +1588,21 @@ static void mos7840_throttle(struct usb_serial_port *port)
 
 	dbg("%s", "Entering .......... \n");
 
-	tty = port->tty;
-	if (!tty) {
-		dbg("%s - no tty available", __func__);
-		return;
-	}
-
 	/* if we are implementing XON/XOFF, send the stop character */
 	if (I_IXOFF(tty)) {
 		unsigned char stop_char = STOP_CHAR(tty);
-		status = mos7840_write(port, &stop_char, 1);
-		if (status <= 0) {
+		status = mos7840_write(tty, port, &stop_char, 1);
+		if (status <= 0)
 			return;
-		}
 	}
-
 	/* if we are implementing RTS/CTS, toggle that line */
 	if (tty->termios->c_cflag & CRTSCTS) {
 		mos7840_port->shadowMCR &= ~MCR_RTS;
 		status = 0;
-		status =
-		    mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER,
+		status = mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER,
 					 mos7840_port->shadowMCR);
-
-		if (status < 0) {
+		if (status < 0)
 			return;
-		}
 	}
 
 	return;
@@ -1628,9 +1613,9 @@ static void mos7840_throttle(struct usb_serial_port *port)
  *	this function is called by the tty driver when it wants to resume the data
  *	being read from the port (called after SerialThrottle is called)
  *****************************************************************************/
-static void mos7840_unthrottle(struct usb_serial_port *port)
+static void mos7840_unthrottle(struct tty_struct *tty)
 {
-	struct tty_struct *tty;
+	struct usb_serial_port *port = tty->driver_data;
 	int status;
 	struct moschip_port *mos7840_port = mos7840_get_port_private(port);
 
@@ -1649,38 +1634,28 @@ static void mos7840_unthrottle(struct usb_serial_port *port)
 
 	dbg("%s", "Entering .......... \n");
 
-	tty = port->tty;
-	if (!tty) {
-		dbg("%s - no tty available", __func__);
-		return;
-	}
-
 	/* if we are implementing XON/XOFF, send the start character */
 	if (I_IXOFF(tty)) {
 		unsigned char start_char = START_CHAR(tty);
-		status = mos7840_write(port, &start_char, 1);
-		if (status <= 0) {
+		status = mos7840_write(tty, port, &start_char, 1);
+		if (status <= 0)
 			return;
-		}
 	}
 
 	/* if we are implementing RTS/CTS, toggle that line */
 	if (tty->termios->c_cflag & CRTSCTS) {
 		mos7840_port->shadowMCR |= MCR_RTS;
 		status = 0;
-		status =
-		    mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER,
+		status = mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER,
 					 mos7840_port->shadowMCR);
-		if (status < 0) {
+		if (status < 0)
 			return;
-		}
 	}
-
-	return;
 }
 
-static int mos7840_tiocmget(struct usb_serial_port *port, struct file *file)
+static int mos7840_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct moschip_port *mos7840_port;
 	unsigned int result;
 	__u16 msr;
@@ -1708,9 +1683,10 @@ static int mos7840_tiocmget(struct usb_serial_port *port, struct file *file)
 	return result;
 }
 
-static int mos7840_tiocmset(struct usb_serial_port *port, struct file *file,
+static int mos7840_tiocmset(struct tty_struct *tty, struct file *file,
 			    unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct moschip_port *mos7840_port;
 	unsigned int mcr;
 	int status;
@@ -1949,10 +1925,9 @@ static int mos7840_send_cmd_write_baud_rate(struct moschip_port *mos7840_port,
  *      the specified new settings.
  *****************************************************************************/
 
-static void mos7840_change_port_settings(struct moschip_port *mos7840_port,
-					 struct ktermios *old_termios)
+static void mos7840_change_port_settings(struct tty_struct *tty,
+	struct moschip_port *mos7840_port, struct ktermios *old_termios)
 {
-	struct tty_struct *tty;
 	int baud;
 	unsigned cflag;
 	unsigned iflag;
@@ -1988,8 +1963,6 @@ static void mos7840_change_port_settings(struct moschip_port *mos7840_port,
 		return;
 	}
 
-	tty = mos7840_port->port->tty;
-
 	dbg("%s", "Entering .......... \n");
 
 	lData = LCR_BITS_8;
@@ -2131,14 +2104,14 @@ static void mos7840_change_port_settings(struct moschip_port *mos7840_port,
  *	the termios structure
  *****************************************************************************/
 
-static void mos7840_set_termios(struct usb_serial_port *port,
+static void mos7840_set_termios(struct tty_struct *tty,
+				struct usb_serial_port *port,
 				struct ktermios *old_termios)
 {
 	int status;
 	unsigned int cflag;
 	struct usb_serial *serial;
 	struct moschip_port *mos7840_port;
-	struct tty_struct *tty;
 	dbg("mos7840_set_termios: START\n");
 	if (mos7840_port_paranoia_check(port, __func__)) {
 		dbg("%s", "Invalid port \n");
@@ -2157,8 +2130,6 @@ static void mos7840_set_termios(struct usb_serial_port *port,
 	if (mos7840_port == NULL)
 		return;
 
-	tty = port->tty;
-
 	if (!mos7840_port->open) {
 		dbg("%s - port not opened", __func__);
 		return;
@@ -2176,7 +2147,7 @@ static void mos7840_set_termios(struct usb_serial_port *port,
 
 	/* change the port settings to the new ones specified */
 
-	mos7840_change_port_settings(mos7840_port, old_termios);
+	mos7840_change_port_settings(tty, mos7840_port, old_termios);
 
 	if (!mos7840_port->read_urb) {
 		dbg("%s", "URB KILLED !!!!!\n");
@@ -2205,13 +2176,13 @@ static void mos7840_set_termios(struct usb_serial_port *port,
  * 	    allows an RS485 driver to be written in user space.
  *****************************************************************************/
 
-static int mos7840_get_lsr_info(struct moschip_port *mos7840_port,
+static int mos7840_get_lsr_info(struct tty_struct *tty,
 				unsigned int __user *value)
 {
 	int count;
 	unsigned int result = 0;
 
-	count = mos7840_chars_in_buffer(mos7840_port->port);
+	count = mos7840_chars_in_buffer(tty);
 	if (count == 0) {
 		dbg("%s -- Empty", __func__);
 		result = TIOCSER_TEMT;
@@ -2227,6 +2198,8 @@ static int mos7840_get_lsr_info(struct moschip_port *mos7840_port,
  *      function to set modem info
  *****************************************************************************/
 
+/* FIXME: Should be using the model control hooks */
+
 static int mos7840_set_modem_info(struct moschip_port *mos7840_port,
 				  unsigned int cmd, unsigned int __user *value)
 {
@@ -2304,9 +2277,8 @@ static int mos7840_get_modem_info(struct moschip_port *mos7840_port,
 	__u16 msr;
 	unsigned int mcr = mos7840_port->shadowMCR;
 	int status = 0;
-	status =
-	    mos7840_get_uart_reg(mos7840_port->port, MODEM_STATUS_REGISTER,
-				 &msr);
+	status = mos7840_get_uart_reg(mos7840_port->port,
+						MODEM_STATUS_REGISTER, &msr);
 	result = ((mcr & MCR_DTR) ? TIOCM_DTR : 0)	/* 0x002 */
 	    |((mcr & MCR_RTS) ? TIOCM_RTS : 0)	/* 0x004 */
 	    |((msr & MOS7840_MSR_CTS) ? TIOCM_CTS : 0)	/* 0x020 */
@@ -2359,12 +2331,12 @@ static int mos7840_get_serial_info(struct moschip_port *mos7840_port,
  *	this function handles any ioctl calls to the driver
  *****************************************************************************/
 
-static int mos7840_ioctl(struct usb_serial_port *port, struct file *file,
+static int mos7840_ioctl(struct tty_struct *tty, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	void __user *argp = (void __user *)arg;
 	struct moschip_port *mos7840_port;
-	struct tty_struct *tty;
 
 	struct async_icount cnow;
 	struct async_icount cprev;
@@ -2381,8 +2353,6 @@ static int mos7840_ioctl(struct usb_serial_port *port, struct file *file,
 	if (mos7840_port == NULL)
 		return -1;
 
-	tty = mos7840_port->port->tty;
-
 	dbg("%s - port %d, cmd = 0x%x", __func__, port->number, cmd);
 
 	switch (cmd) {
@@ -2390,9 +2360,10 @@ static int mos7840_ioctl(struct usb_serial_port *port, struct file *file,
 
 	case TIOCSERGETLSR:
 		dbg("%s (%d) TIOCSERGETLSR", __func__, port->number);
-		return mos7840_get_lsr_info(mos7840_port, argp);
+		return mos7840_get_lsr_info(tty, argp);
 		return 0;
 
+	/* FIXME: use the modem hooks and remove this */
 	case TIOCMBIS:
 	case TIOCMBIC:
 	case TIOCMSET:
@@ -2463,13 +2434,9 @@ static int mos7840_ioctl(struct usb_serial_port *port, struct file *file,
 		if (copy_to_user(argp, &icount, sizeof(icount)))
 			return -EFAULT;
 		return 0;
-
-	case TIOCEXBAUD:
-		return 0;
 	default:
 		break;
 	}
-
 	return -ENOIOCTLCMD;
 }
 
diff --git a/drivers/usb/serial/navman.c b/drivers/usb/serial/navman.c
index 43c8894353bf..d6736531a0fa 100644
--- a/drivers/usb/serial/navman.c
+++ b/drivers/usb/serial/navman.c
@@ -64,7 +64,7 @@ static void navman_read_int_callback(struct urb *urb)
 	usb_serial_debug_data(debug, &port->dev, __func__,
 			      urb->actual_length, data);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
@@ -79,7 +79,8 @@ exit:
 			__func__, result);
 }
 
-static int navman_open(struct usb_serial_port *port, struct file *filp)
+static int navman_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	int result = 0;
 
@@ -96,14 +97,15 @@ static int navman_open(struct usb_serial_port *port, struct file *filp)
 	return result;
 }
 
-static void navman_close(struct usb_serial_port *port, struct file *filp)
+static void navman_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 
 	usb_kill_urb(port->interrupt_in_urb);
 }
 
-static int navman_write(struct usb_serial_port *port,
+static int navman_write(struct tty_struct *tty, struct usb_serial_port *port,
 			const unsigned char *buf, int count)
 {
 	dbg("%s - port %d", __func__, port->number);
diff --git a/drivers/usb/serial/omninet.c b/drivers/usb/serial/omninet.c
index 7b7422f49478..5a2d045562f0 100644
--- a/drivers/usb/serial/omninet.c
+++ b/drivers/usb/serial/omninet.c
@@ -61,12 +61,12 @@ static int debug;
 #define BT_IGNITIONPRO_ID	0x2000  /* This one seems to be a re-branded ZyXEL device */
 
 /* function prototypes */
-static int  omninet_open		(struct usb_serial_port *port, struct file *filp);
-static void omninet_close		(struct usb_serial_port *port, struct file *filp);
+static int  omninet_open		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void omninet_close		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
 static void omninet_read_bulk_callback	(struct urb *urb);
 static void omninet_write_bulk_callback	(struct urb *urb);
-static int  omninet_write		(struct usb_serial_port *port, const unsigned char *buf, int count);
-static int  omninet_write_room		(struct usb_serial_port *port);
+static int  omninet_write		(struct tty_struct *tty, struct usb_serial_port *port, const unsigned char *buf, int count);
+static int  omninet_write_room		(struct tty_struct *tty);
 static void omninet_shutdown		(struct usb_serial *serial);
 static int omninet_attach		(struct usb_serial *serial);
 
@@ -157,7 +157,8 @@ static int omninet_attach (struct usb_serial *serial)
 	return 0;
 }
 
-static int omninet_open (struct usb_serial_port *port, struct file *filp)
+static int omninet_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial	*serial = port->serial;
 	struct usb_serial_port	*wport;
@@ -166,7 +167,7 @@ static int omninet_open (struct usb_serial_port *port, struct file *filp)
 	dbg("%s - port %d", __func__, port->number);
 
 	wport = serial->port[1];
-	wport->tty = port->tty;
+	wport->port.tty = tty;		/* FIXME */
 
 	/* Start reading from the device */
 	usb_fill_bulk_urb(port->read_urb, serial->dev, 
@@ -181,7 +182,8 @@ static int omninet_open (struct usb_serial_port *port, struct file *filp)
 	return result;
 }
 
-static void omninet_close (struct usb_serial_port *port, struct file * filp)
+static void omninet_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	dbg("%s - port %d", __func__, port->number);
 	usb_kill_urb(port->read_urb);
@@ -221,9 +223,9 @@ static void omninet_read_bulk_callback (struct urb *urb)
 
 	if (urb->actual_length && header->oh_len) {
 		for (i = 0; i < header->oh_len; i++) {
-			 tty_insert_flip_char(port->tty, data[OMNINET_DATAOFFSET + i], 0);
+			 tty_insert_flip_char(port->port.tty, data[OMNINET_DATAOFFSET + i], 0);
 	  	}
-	  	tty_flip_buffer_push(port->tty);
+	  	tty_flip_buffer_push(port->port.tty);
 	}
 
 	/* Continue trying to always read  */
@@ -238,7 +240,8 @@ static void omninet_read_bulk_callback (struct urb *urb)
 	return;
 }
 
-static int omninet_write (struct usb_serial_port *port, const unsigned char *buf, int count)
+static int omninet_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 	struct usb_serial 	*serial	= port->serial;
 	struct usb_serial_port 	*wport	= serial->port[1];
@@ -290,8 +293,9 @@ static int omninet_write (struct usb_serial_port *port, const unsigned char *buf
 }
 
 
-static int omninet_write_room (struct usb_serial_port *port)
+static int omninet_write_room (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial 	*serial = port->serial;
 	struct usb_serial_port 	*wport 	= serial->port[1];
 
diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c
index 1e936a1cbe0b..4350990abf14 100644
--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -43,29 +43,23 @@
 #include <linux/usb/serial.h>
 
 /* Function prototypes */
-static int  option_open(struct usb_serial_port *port, struct file *filp);
-static void option_close(struct usb_serial_port *port, struct file *filp);
+static int  option_open(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void option_close(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
 static int  option_startup(struct usb_serial *serial);
 static void option_shutdown(struct usb_serial *serial);
-static void option_rx_throttle(struct usb_serial_port *port);
-static void option_rx_unthrottle(struct usb_serial_port *port);
-static int  option_write_room(struct usb_serial_port *port);
+static int  option_write_room(struct tty_struct *tty);
 
 static void option_instat_callback(struct urb *urb);
 
-static int option_write(struct usb_serial_port *port,
+static int option_write(struct tty_struct *tty, struct usb_serial_port *port,
 			const unsigned char *buf, int count);
-
-static int  option_chars_in_buffer(struct usb_serial_port *port);
-static int  option_ioctl(struct usb_serial_port *port, struct file *file,
-			unsigned int cmd, unsigned long arg);
-static void option_set_termios(struct usb_serial_port *port,
-				struct ktermios *old);
-static void option_break_ctl(struct usb_serial_port *port, int break_state);
-static int  option_tiocmget(struct usb_serial_port *port, struct file *file);
-static int  option_tiocmset(struct usb_serial_port *port, struct file *file,
+static int  option_chars_in_buffer(struct tty_struct *tty);
+static void option_set_termios(struct tty_struct *tty,
+			struct usb_serial_port *port, struct ktermios *old);
+static int  option_tiocmget(struct tty_struct *tty, struct file *file);
+static int  option_tiocmset(struct tty_struct *tty, struct file *file,
 				unsigned int set, unsigned int clear);
-static int  option_send_setup(struct usb_serial_port *port);
+static int  option_send_setup(struct tty_struct *tty, struct usb_serial_port *port);
 
 /* Vendor and product IDs */
 #define OPTION_VENDOR_ID			0x0AF0
@@ -342,11 +336,7 @@ static struct usb_serial_driver option_1port_device = {
 	.write             = option_write,
 	.write_room        = option_write_room,
 	.chars_in_buffer   = option_chars_in_buffer,
-	.throttle          = option_rx_throttle,
-	.unthrottle        = option_rx_unthrottle,
-	.ioctl             = option_ioctl,
 	.set_termios       = option_set_termios,
-	.break_ctl         = option_break_ctl,
 	.tiocmget          = option_tiocmget,
 	.tiocmset          = option_tiocmset,
 	.attach            = option_startup,
@@ -417,33 +407,18 @@ static void __exit option_exit(void)
 module_init(option_init);
 module_exit(option_exit);
 
-static void option_rx_throttle(struct usb_serial_port *port)
-{
-	dbg("%s", __func__);
-}
-
-static void option_rx_unthrottle(struct usb_serial_port *port)
-{
-	dbg("%s", __func__);
-}
-
-static void option_break_ctl(struct usb_serial_port *port, int break_state)
-{
-	/* Unfortunately, I don't know how to send a break */
-	dbg("%s", __func__);
-}
-
-static void option_set_termios(struct usb_serial_port *port,
-			struct ktermios *old_termios)
+static void option_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	dbg("%s", __func__);
 	/* Doesn't support option setting */
-	tty_termios_copy_hw(port->tty->termios, old_termios);
-	option_send_setup(port);
+	tty_termios_copy_hw(tty->termios, old_termios);
+	option_send_setup(tty, port);
 }
 
-static int option_tiocmget(struct usb_serial_port *port, struct file *file)
+static int option_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned int value;
 	struct option_port_private *portdata;
 
@@ -459,9 +434,10 @@ static int option_tiocmget(struct usb_serial_port *port, struct file *file)
 	return value;
 }
 
-static int option_tiocmset(struct usb_serial_port *port, struct file *file,
+static int option_tiocmset(struct tty_struct *tty, struct file *file,
 			unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct option_port_private *portdata;
 
 	portdata = usb_get_serial_port_data(port);
@@ -476,17 +452,11 @@ static int option_tiocmset(struct usb_serial_port *port, struct file *file,
 		portdata->rts_state = 0;
 	if (clear & TIOCM_DTR)
 		portdata->dtr_state = 0;
-	return option_send_setup(port);
-}
-
-static int option_ioctl(struct usb_serial_port *port, struct file *file,
-			unsigned int cmd, unsigned long arg)
-{
-	return -ENOIOCTLCMD;
+	return option_send_setup(tty, port);
 }
 
 /* Write */
-static int option_write(struct usb_serial_port *port,
+static int option_write(struct tty_struct *tty, struct usb_serial_port *port,
 			const unsigned char *buf, int count)
 {
 	struct option_port_private *portdata;
@@ -562,7 +532,7 @@ static void option_indat_callback(struct urb *urb)
 		dbg("%s: nonzero status: %d on endpoint %02x.",
 		    __func__, status, endpoint);
 	} else {
-		tty = port->tty;
+		tty = port->port.tty;
 		if (urb->actual_length) {
 			tty_buffer_request_room(tty, urb->actual_length);
 			tty_insert_flip_string(tty, data, urb->actual_length);
@@ -572,7 +542,7 @@ static void option_indat_callback(struct urb *urb)
 		}
 
 		/* Resubmit urb so we continue receiving */
-		if (port->open_count && status != -ESHUTDOWN) {
+		if (port->port.count && status != -ESHUTDOWN) {
 			err = usb_submit_urb(urb, GFP_ATOMIC);
 			if (err)
 				printk(KERN_ERR "%s: resubmit read urb failed. "
@@ -638,9 +608,9 @@ static void option_instat_callback(struct urb *urb)
 			portdata->dsr_state = ((signals & 0x02) ? 1 : 0);
 			portdata->ri_state = ((signals & 0x08) ? 1 : 0);
 
-			if (port->tty && !C_CLOCAL(port->tty) &&
+			if (port->port.tty && !C_CLOCAL(port->port.tty) &&
 					old_dcd_state && !portdata->dcd_state)
-				tty_hangup(port->tty);
+				tty_hangup(port->port.tty);
 		} else {
 			dbg("%s: type %x req %x", __func__,
 				req_pkt->bRequestType,req_pkt->bRequest);
@@ -658,8 +628,9 @@ static void option_instat_callback(struct urb *urb)
 	}
 }
 
-static int option_write_room(struct usb_serial_port *port)
+static int option_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct option_port_private *portdata;
 	int i;
 	int data_len = 0;
@@ -678,8 +649,9 @@ static int option_write_room(struct usb_serial_port *port)
 	return data_len;
 }
 
-static int option_chars_in_buffer(struct usb_serial_port *port)
+static int option_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct option_port_private *portdata;
 	int i;
 	int data_len = 0;
@@ -698,7 +670,8 @@ static int option_chars_in_buffer(struct usb_serial_port *port)
 	return data_len;
 }
 
-static int option_open(struct usb_serial_port *port, struct file *filp)
+static int option_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct option_port_private *portdata;
 	struct usb_serial *serial = port->serial;
@@ -748,14 +721,16 @@ static int option_open(struct usb_serial_port *port, struct file *filp)
 				usb_pipeout(urb->pipe), 0); */
 	}
 
-	port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
-	option_send_setup(port);
+	option_send_setup(tty, port);
 
 	return (0);
 }
 
-static void option_close(struct usb_serial_port *port, struct file *filp)
+static void option_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	int i;
 	struct usb_serial *serial = port->serial;
@@ -770,7 +745,7 @@ static void option_close(struct usb_serial_port *port, struct file *filp)
 	if (serial->dev) {
 		mutex_lock(&serial->disc_mutex);
 		if (!serial->disconnected)
-			option_send_setup(port);
+			option_send_setup(tty, port);
 		mutex_unlock(&serial->disc_mutex);
 
 		/* Stop reading/writing urbs */
@@ -779,7 +754,7 @@ static void option_close(struct usb_serial_port *port, struct file *filp)
 		for (i = 0; i < N_OUT_URB; i++)
 			usb_kill_urb(portdata->out_urbs[i]);
 	}
-	port->tty = NULL;
+	port->port.tty = NULL;	/* FIXME */
 }
 
 /* Helper functions used by option_setup_urbs */
@@ -841,7 +816,8 @@ static void option_setup_urbs(struct usb_serial *serial)
  * This is exactly the same as SET_CONTROL_LINE_STATE from the PSTN
  * CDC.
 */
-static int option_send_setup(struct usb_serial_port *port)
+static int option_send_setup(struct tty_struct *tty,
+						struct usb_serial_port *port)
 {
 	struct usb_serial *serial = port->serial;
 	struct option_port_private *portdata;
@@ -850,7 +826,7 @@ static int option_send_setup(struct usb_serial_port *port)
 
 	portdata = usb_get_serial_port_data(port);
 
-	if (port->tty) {
+	if (tty) {
 		int val = 0;
 		if (portdata->dtr_state)
 			val |= 0x01;
@@ -861,7 +837,6 @@ static int option_send_setup(struct usb_serial_port *port)
 				usb_rcvctrlpipe(serial->dev, 0),
 				0x22,0x21,val,ifNum,NULL,0,USB_CTRL_SET_TIMEOUT);
 	}
-
 	return 0;
 }
 
diff --git a/drivers/usb/serial/oti6858.c b/drivers/usb/serial/oti6858.c
index a9625c180dc3..069d276a5276 100644
--- a/drivers/usb/serial/oti6858.c
+++ b/drivers/usb/serial/oti6858.c
@@ -140,22 +140,23 @@ struct oti6858_control_pkt {
 	  && ((a)->frame_fmt == (priv)->pending_setup.frame_fmt) )
 
 /* function prototypes */
-static int oti6858_open(struct usb_serial_port *port, struct file *filp);
-static void oti6858_close(struct usb_serial_port *port, struct file *filp);
-static void oti6858_set_termios(struct usb_serial_port *port,
-				struct ktermios *old);
-static int oti6858_ioctl(struct usb_serial_port *port, struct file *file,
+static int oti6858_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+static void oti6858_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+static void oti6858_set_termios(struct tty_struct *tty,
+			struct usb_serial_port *port, struct ktermios *old);
+static int oti6858_ioctl(struct tty_struct *tty, struct file *file,
 			unsigned int cmd, unsigned long arg);
 static void oti6858_read_int_callback(struct urb *urb);
 static void oti6858_read_bulk_callback(struct urb *urb);
 static void oti6858_write_bulk_callback(struct urb *urb);
-static int oti6858_write(struct usb_serial_port *port,
+static int oti6858_write(struct tty_struct *tty, struct usb_serial_port *port,
 			const unsigned char *buf, int count);
-static int oti6858_write_room(struct usb_serial_port *port);
-static void oti6858_break_ctl(struct usb_serial_port *port, int break_state);
-static int oti6858_chars_in_buffer(struct usb_serial_port *port);
-static int oti6858_tiocmget(struct usb_serial_port *port, struct file *file);
-static int oti6858_tiocmset(struct usb_serial_port *port, struct file *file,
+static int oti6858_write_room(struct tty_struct *tty);
+static int oti6858_chars_in_buffer(struct tty_struct *tty);
+static int oti6858_tiocmget(struct tty_struct *tty, struct file *file);
+static int oti6858_tiocmset(struct tty_struct *tty, struct file *file,
 				unsigned int set, unsigned int clear);
 static int oti6858_startup(struct usb_serial *serial);
 static void oti6858_shutdown(struct usb_serial *serial);
@@ -184,7 +185,6 @@ static struct usb_serial_driver oti6858_device = {
 	.close =		oti6858_close,
 	.write =		oti6858_write,
 	.ioctl =		oti6858_ioctl,
-	.break_ctl =		oti6858_break_ctl,
 	.set_termios =		oti6858_set_termios,
 	.tiocmget =		oti6858_tiocmget,
 	.tiocmset =		oti6858_tiocmset,
@@ -395,7 +395,7 @@ static int oti6858_startup(struct usb_serial *serial)
 	return -ENOMEM;
 }
 
-static int oti6858_write(struct usb_serial_port *port,
+static int oti6858_write(struct tty_struct *tty, struct usb_serial_port *port,
 			const unsigned char *buf, int count)
 {
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
@@ -413,8 +413,9 @@ static int oti6858_write(struct usb_serial_port *port,
 	return count;
 }
 
-static int oti6858_write_room(struct usb_serial_port *port)
+static int oti6858_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
 	int room = 0;
 	unsigned long flags;
@@ -428,8 +429,9 @@ static int oti6858_write_room(struct usb_serial_port *port)
 	return room;
 }
 
-static int oti6858_chars_in_buffer(struct usb_serial_port *port)
+static int oti6858_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
 	int chars = 0;
 	unsigned long flags;
@@ -443,8 +445,8 @@ static int oti6858_chars_in_buffer(struct usb_serial_port *port)
 	return chars;
 }
 
-static void oti6858_set_termios(struct usb_serial_port *port,
-				struct ktermios *old_termios)
+static void oti6858_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -455,22 +457,22 @@ static void oti6858_set_termios(struct usb_serial_port *port,
 
 	dbg("%s(port = %d)", __func__, port->number);
 
-	if (!port->tty || !port->tty->termios) {
+	if (!tty) {
 		dbg("%s(): no tty structures", __func__);
 		return;
 	}
 
 	spin_lock_irqsave(&priv->lock, flags);
 	if (!priv->flags.termios_initialized) {
-		*(port->tty->termios) = tty_std_termios;
-		port->tty->termios->c_cflag = B38400 | CS8 | CREAD | HUPCL | CLOCAL;
+		*(tty->termios) = tty_std_termios;
+		tty->termios->c_cflag = B38400 | CS8 | CREAD | HUPCL | CLOCAL;
+		tty->termios->c_ispeed = 38400;
+		tty->termios->c_ospeed = 38400;
 		priv->flags.termios_initialized = 1;
-		port->tty->termios->c_ispeed = 38400;
-		port->tty->termios->c_ospeed = 38400;
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	cflag = port->tty->termios->c_cflag;
+	cflag = tty->termios->c_cflag;
 
 	spin_lock_irqsave(&priv->lock, flags);
 	divisor = priv->pending_setup.divisor;
@@ -500,7 +502,7 @@ static void oti6858_set_termios(struct usb_serial_port *port,
 	 * guarantee that any other baud rate will work (especially
 	 * the higher ones)
 	 */
-	br = tty_get_baud_rate(port->tty);
+	br = tty_get_baud_rate(tty);
 	if (br == 0) {
 		divisor = 0;
 	} else {
@@ -511,7 +513,7 @@ static void oti6858_set_termios(struct usb_serial_port *port,
 		new_divisor = (96000000 + 8 * br) / (16 * br);
 		real_br = 96000000 / (16 * new_divisor);
 		divisor = cpu_to_le16(new_divisor);
-		tty_encode_baud_rate(port->tty, real_br, real_br);
+		tty_encode_baud_rate(tty, real_br, real_br);
 	}
 
 	frame_fmt &= ~FMT_STOP_BITS_MASK;
@@ -564,7 +566,8 @@ static void oti6858_set_termios(struct usb_serial_port *port,
 	spin_unlock_irqrestore(&priv->lock, flags);
 }
 
-static int oti6858_open(struct usb_serial_port *port, struct file *filp)
+static int oti6858_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
 	struct ktermios tmp_termios;
@@ -578,7 +581,7 @@ static int oti6858_open(struct usb_serial_port *port, struct file *filp)
 	usb_clear_halt(serial->dev, port->write_urb->pipe);
 	usb_clear_halt(serial->dev, port->read_urb->pipe);
 
-	if (port->open_count != 1)
+	if (port->port.count != 1)
 		return 0;
 
 	if ((buf = kmalloc(OTI6858_CTRL_PKT_SIZE, GFP_KERNEL)) == NULL) {
@@ -617,18 +620,19 @@ static int oti6858_open(struct usb_serial_port *port, struct file *filp)
 	if (result != 0) {
 		dev_err(&port->dev, "%s(): usb_submit_urb() failed"
 			       " with error %d\n", __func__, result);
-		oti6858_close(port, NULL);
+		oti6858_close(tty, port, NULL);
 		return -EPROTO;
 	}
 
 	/* setup termios */
-	if (port->tty)
-		oti6858_set_termios(port, &tmp_termios);
+	if (tty)
+		oti6858_set_termios(tty, port, &tmp_termios);
 
 	return 0;
 }
 
-static void oti6858_close(struct usb_serial_port *port, struct file *filp)
+static void oti6858_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -641,7 +645,7 @@ static void oti6858_close(struct usb_serial_port *port, struct file *filp)
 	spin_lock_irqsave(&priv->lock, flags);
 	timeout = 30 * HZ;	/* PL2303_CLOSING_WAIT */
 	init_waitqueue_entry(&wait, current);
-	add_wait_queue(&port->tty->write_wait, &wait);
+	add_wait_queue(&tty->write_wait, &wait);
 	dbg("%s(): entering wait loop", __func__);
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
@@ -654,7 +658,7 @@ static void oti6858_close(struct usb_serial_port *port, struct file *filp)
 		spin_lock_irqsave(&priv->lock, flags);
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->tty->write_wait, &wait);
+	remove_wait_queue(&tty->write_wait, &wait);
 	dbg("%s(): after wait loop", __func__);
 
 	/* clear out any remaining data in the buffer */
@@ -669,7 +673,7 @@ static void oti6858_close(struct usb_serial_port *port, struct file *filp)
 	/* data is in the buffer to compute a delay */
 	/* that is not unnecessarily long) */
 	/* FIXME
-	bps = tty_get_baud_rate(port->tty);
+	bps = tty_get_baud_rate(tty);
 	if (bps > 1200)
 		timeout = max((HZ*2560)/bps,HZ/10);
 	else
@@ -690,7 +694,7 @@ static void oti6858_close(struct usb_serial_port *port, struct file *filp)
 	usb_kill_urb(port->interrupt_in_urb);
 
 	/*
-	if (port->tty && (port->tty->termios->c_cflag) & HUPCL) {
+	if (tty && (tty->termios->c_cflag) & HUPCL) {
 		// drop DTR and RTS
 		spin_lock_irqsave(&priv->lock, flags);
 		priv->pending_setup.control &= ~CONTROL_MASK;
@@ -699,9 +703,10 @@ static void oti6858_close(struct usb_serial_port *port, struct file *filp)
 	*/
 }
 
-static int oti6858_tiocmset(struct usb_serial_port *port, struct file *file,
+static int oti6858_tiocmset(struct tty_struct *tty, struct file *file,
 				unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	u8 control;
@@ -732,8 +737,9 @@ static int oti6858_tiocmset(struct usb_serial_port *port, struct file *file,
 	return 0;
 }
 
-static int oti6858_tiocmget(struct usb_serial_port *port, struct file *file)
+static int oti6858_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct oti6858_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	unsigned pin_state;
@@ -802,26 +808,15 @@ static int wait_modem_info(struct usb_serial_port *port, unsigned int arg)
 	return 0;
 }
 
-static int oti6858_ioctl(struct usb_serial_port *port, struct file *file,
+static int oti6858_ioctl(struct tty_struct *tty, struct file *file,
 			unsigned int cmd, unsigned long arg)
 {
-	void __user *user_arg = (void __user *) arg;
-	unsigned int x;
+	struct usb_serial_port *port = tty->driver_data;
 
 	dbg("%s(port = %d, cmd = 0x%04x, arg = 0x%08lx)",
 				__func__, port->number, cmd, arg);
 
 	switch (cmd) {
-		case TIOCMBIS:
-			if (copy_from_user(&x, user_arg, sizeof(x)))
-				return -EFAULT;
-			return oti6858_tiocmset(port, NULL, x, 0);
-
-		case TIOCMBIC:
-			if (copy_from_user(&x, user_arg, sizeof(x)))
-				return -EFAULT;
-			return oti6858_tiocmset(port, NULL, 0, x);
-
 		case TIOCMIWAIT:
 			dbg("%s(): TIOCMIWAIT", __func__);
 			return wait_modem_info(port, arg);
@@ -834,24 +829,6 @@ static int oti6858_ioctl(struct usb_serial_port *port, struct file *file,
 	return -ENOIOCTLCMD;
 }
 
-static void oti6858_break_ctl(struct usb_serial_port *port, int break_state)
-{
-	int state;
-
-	dbg("%s(port = %d)", __func__, port->number);
-
-	state = (break_state == 0) ? 0 : 1;
-	dbg("%s(): turning break %s", __func__, state ? "on" : "off");
-
-	/* FIXME */
-/*
-	result = usb_control_msg (serial->dev, usb_sndctrlpipe (serial->dev, 0),
-				  BREAK_REQUEST, BREAK_REQUEST_TYPE, state,
-				  0, NULL, 0, 100);
-	if (result != 0)
-		dbg("%s(): error sending break", __func__);
- */
-}
 
 static void oti6858_shutdown(struct usb_serial *serial)
 {
@@ -1002,7 +979,7 @@ static void oti6858_read_bulk_callback(struct urb *urb)
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	if (status != 0) {
-		if (!port->open_count) {
+		if (!port->port.count) {
 			dbg("%s(): port is closed, exiting", __func__);
 			return;
 		}
@@ -1020,14 +997,14 @@ static void oti6858_read_bulk_callback(struct urb *urb)
 		return;
 	}
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty != NULL && urb->actual_length > 0) {
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
 
 	// schedule the interrupt urb if we are still open */
-	if (port->open_count != 0) {
+	if (port->port.count != 0) {
 		port->interrupt_in_urb->dev = port->serial->dev;
 		result = usb_submit_urb(port->interrupt_in_urb, GFP_ATOMIC);
 		if (result != 0) {
diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 2a0dd1b50dc4..a0016725d314 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -458,8 +458,8 @@ static void pl2303_send(struct usb_serial_port *port)
 	usb_serial_port_softint(port);
 }
 
-static int pl2303_write(struct usb_serial_port *port, const unsigned char *buf,
-			int count)
+static int pl2303_write(struct tty_struct *tty, struct usb_serial_port *port,
+				const unsigned char *buf, int count)
 {
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -478,8 +478,9 @@ static int pl2303_write(struct usb_serial_port *port, const unsigned char *buf,
 	return count;
 }
 
-static int pl2303_write_room(struct usb_serial_port *port)
+static int pl2303_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	int room = 0;
 	unsigned long flags;
@@ -494,8 +495,9 @@ static int pl2303_write_room(struct usb_serial_port *port)
 	return room;
 }
 
-static int pl2303_chars_in_buffer(struct usb_serial_port *port)
+static int pl2303_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	int chars = 0;
 	unsigned long flags;
@@ -510,8 +512,8 @@ static int pl2303_chars_in_buffer(struct usb_serial_port *port)
 	return chars;
 }
 
-static void pl2303_set_termios(struct usb_serial_port *port,
-			       struct ktermios *old_termios)
+static void pl2303_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct usb_serial *serial = port->serial;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
@@ -526,11 +528,10 @@ static void pl2303_set_termios(struct usb_serial_port *port,
 
 	spin_lock_irqsave(&priv->lock, flags);
 	if (!priv->termios_initialized) {
-		*(port->tty->termios) = tty_std_termios;
-		port->tty->termios->c_cflag = B9600 | CS8 | CREAD |
-					      HUPCL | CLOCAL;
-		port->tty->termios->c_ispeed = 9600;
-		port->tty->termios->c_ospeed = 9600;
+		*(tty->termios) = tty_std_termios;
+		tty->termios->c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL;
+		tty->termios->c_ispeed = 9600;
+		tty->termios->c_ospeed = 9600;
 		priv->termios_initialized = 1;
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
@@ -539,16 +540,16 @@ static void pl2303_set_termios(struct usb_serial_port *port,
 	   serial settings even to the same values as before. Thus
 	   we actually need to filter in this specific case */
 
-	if (!tty_termios_hw_change(port->tty->termios, old_termios))
+	if (!tty_termios_hw_change(tty->termios, old_termios))
 		return;
 
-	cflag = port->tty->termios->c_cflag;
+	cflag = tty->termios->c_cflag;
 
 	buf = kzalloc(7, GFP_KERNEL);
 	if (!buf) {
 		dev_err(&port->dev, "%s - out of memory.\n", __func__);
 		/* Report back no change occurred */
-		*port->tty->termios = *old_termios;
+		*tty->termios = *old_termios;
 		return;
 	}
 
@@ -569,7 +570,7 @@ static void pl2303_set_termios(struct usb_serial_port *port,
 		dbg("%s - data bits = %d", __func__, buf[6]);
 	}
 
-	baud = tty_get_baud_rate(port->tty);;
+	baud = tty_get_baud_rate(tty);
 	dbg("%s - baud = %d", __func__, baud);
 	if (baud) {
 		buf[0] = baud & 0xff;
@@ -646,12 +647,13 @@ static void pl2303_set_termios(struct usb_serial_port *port,
 
 	/* FIXME: Need to read back resulting baud rate */
 	if (baud)
-		tty_encode_baud_rate(port->tty, baud, baud);
+		tty_encode_baud_rate(tty, baud, baud);
 
 	kfree(buf);
 }
 
-static void pl2303_close(struct usb_serial_port *port, struct file *filp)
+static void pl2303_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -666,7 +668,7 @@ static void pl2303_close(struct usb_serial_port *port, struct file *filp)
 	spin_lock_irqsave(&priv->lock, flags);
 	timeout = PL2303_CLOSING_WAIT;
 	init_waitqueue_entry(&wait, current);
-	add_wait_queue(&port->tty->write_wait, &wait);
+	add_wait_queue(&tty->write_wait, &wait);
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (pl2303_buf_data_avail(priv->buf) == 0 ||
@@ -678,7 +680,7 @@ static void pl2303_close(struct usb_serial_port *port, struct file *filp)
 		spin_lock_irqsave(&priv->lock, flags);
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->tty->write_wait, &wait);
+	remove_wait_queue(&tty->write_wait, &wait);
 	/* clear out any remaining data in the buffer */
 	pl2303_buf_clear(priv->buf);
 	spin_unlock_irqrestore(&priv->lock, flags);
@@ -690,7 +692,7 @@ static void pl2303_close(struct usb_serial_port *port, struct file *filp)
 	/* for lower rates we should really know how much */
 	/* data is in the buffer to compute a delay */
 	/* that is not unnecessarily long) */
-	bps = tty_get_baud_rate(port->tty);
+	bps = tty_get_baud_rate(tty);
 	if (bps > 1200)
 		timeout = max((HZ*2560)/bps,HZ/10);
 	else
@@ -703,8 +705,8 @@ static void pl2303_close(struct usb_serial_port *port, struct file *filp)
 	usb_kill_urb(port->read_urb);
 	usb_kill_urb(port->interrupt_in_urb);
 
-	if (port->tty) {
-		c_cflag = port->tty->termios->c_cflag;
+	if (tty) {
+		c_cflag = tty->termios->c_cflag;
 		if (c_cflag & HUPCL) {
 			/* drop DTR and RTS */
 			spin_lock_irqsave(&priv->lock, flags);
@@ -715,7 +717,8 @@ static void pl2303_close(struct usb_serial_port *port, struct file *filp)
 	}
 }
 
-static int pl2303_open(struct usb_serial_port *port, struct file *filp)
+static int pl2303_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct ktermios tmp_termios;
 	struct usb_serial *serial = port->serial;
@@ -734,9 +737,8 @@ static int pl2303_open(struct usb_serial_port *port, struct file *filp)
 	}
 
 	/* Setup termios */
-	if (port->tty) {
-		pl2303_set_termios(port, &tmp_termios);
-	}
+	if (tty)
+		pl2303_set_termios(tty, port, &tmp_termios);
 
 	//FIXME: need to assert RTS and DTR if CRTSCTS off
 
@@ -746,7 +748,7 @@ static int pl2303_open(struct usb_serial_port *port, struct file *filp)
 	if (result) {
 		dev_err(&port->dev, "%s - failed submitting read urb,"
 			" error %d\n", __func__, result);
-		pl2303_close(port, NULL);
+		pl2303_close(tty, port, NULL);
 		return -EPROTO;
 	}
 
@@ -756,15 +758,16 @@ static int pl2303_open(struct usb_serial_port *port, struct file *filp)
 	if (result) {
 		dev_err(&port->dev, "%s - failed submitting interrupt urb,"
 			" error %d\n", __func__, result);
-		pl2303_close(port, NULL);
+		pl2303_close(tty, port, NULL);
 		return -EPROTO;
 	}
 	return 0;
 }
 
-static int pl2303_tiocmset(struct usb_serial_port *port, struct file *file,
+static int pl2303_tiocmset(struct tty_struct *tty, struct file *file,
 			   unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	u8 control;
@@ -787,8 +790,9 @@ static int pl2303_tiocmset(struct usb_serial_port *port, struct file *file,
 	return set_control_lines(port->serial->dev, control);
 }
 
-static int pl2303_tiocmget(struct usb_serial_port *port, struct file *file)
+static int pl2303_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	unsigned int mcr;
@@ -853,9 +857,10 @@ static int wait_modem_info(struct usb_serial_port *port, unsigned int arg)
 	return 0;
 }
 
-static int pl2303_ioctl(struct usb_serial_port *port, struct file *file,
+static int pl2303_ioctl(struct tty_struct *tty, struct file *file,
 			unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s (%d) cmd = 0x%04x", __func__, port->number, cmd);
 
 	switch (cmd) {
@@ -871,8 +876,9 @@ static int pl2303_ioctl(struct usb_serial_port *port, struct file *file,
 	return -ENOIOCTLCMD;
 }
 
-static void pl2303_break_ctl(struct usb_serial_port *port, int break_state)
+static void pl2303_break_ctl(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct usb_serial *serial = port->serial;
 	u16 state;
 	int result;
@@ -1001,7 +1007,7 @@ static void pl2303_read_bulk_callback(struct urb *urb)
 
 	if (status) {
 		dbg("%s - urb status = %d", __func__, status);
-		if (!port->open_count) {
+		if (!port->port.count) {
 			dbg("%s - port is closed, exiting.", __func__);
 			return;
 		}
@@ -1044,7 +1050,7 @@ static void pl2303_read_bulk_callback(struct urb *urb)
 		tty_flag = TTY_FRAME;
 	dbg("%s - tty_flag = %d", __func__, tty_flag);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length + 1);
 		/* overrun is special, not associated with a char */
@@ -1056,7 +1062,7 @@ static void pl2303_read_bulk_callback(struct urb *urb)
 	}
 
 	/* Schedule the next read _if_ we are still open */
-	if (port->open_count) {
+	if (port->port.count) {
 		urb->dev = port->serial->dev;
 		result = usb_submit_urb(urb, GFP_ATOMIC);
 		if (result)
diff --git a/drivers/usb/serial/safe_serial.c b/drivers/usb/serial/safe_serial.c
index 94bddf06ea4f..f823e4dcea1e 100644
--- a/drivers/usb/serial/safe_serial.c
+++ b/drivers/usb/serial/safe_serial.c
@@ -229,8 +229,8 @@ static void safe_read_bulk_callback (struct urb *urb)
 			int actual_length = data[length - 2] >> 2;
 			if (actual_length <= (length - 2)) {
 				info ("%s - actual: %d", __func__, actual_length);
-				tty_insert_flip_string(port->tty, data, actual_length);
-				tty_flip_buffer_push (port->tty);
+				tty_insert_flip_string(port->port.tty, data, actual_length);
+				tty_flip_buffer_push (port->port.tty);
 			} else {
 				err ("%s - inconsistent lengths %d:%d", __func__,
 				     actual_length, length);
@@ -239,8 +239,8 @@ static void safe_read_bulk_callback (struct urb *urb)
 			err ("%s - bad CRC %x", __func__, fcs);
 		}
 	} else {
-		tty_insert_flip_string(port->tty, data, length);
-		tty_flip_buffer_push (port->tty);
+		tty_insert_flip_string(port->port.tty, data, length);
+		tty_flip_buffer_push (port->port.tty);
 	}
 
 	/* Continue trying to always read  */
@@ -255,7 +255,8 @@ static void safe_read_bulk_callback (struct urb *urb)
 	}
 }
 
-static int safe_write (struct usb_serial_port *port, const unsigned char *buf, int count)
+static int safe_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 	unsigned char *data;
 	int result;
@@ -349,8 +350,9 @@ static int safe_write (struct usb_serial_port *port, const unsigned char *buf, i
 	return (count);
 }
 
-static int safe_write_room (struct usb_serial_port *port)
+static int safe_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	int room = 0;		/* Default: no room */
 	unsigned long flags;
 
diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c
index 29074c1ba22b..892020d48555 100644
--- a/drivers/usb/serial/sierra.c
+++ b/drivers/usb/serial/sierra.c
@@ -250,7 +250,8 @@ struct sierra_port_private {
 	int ri_state;
 };
 
-static int sierra_send_setup(struct usb_serial_port *port)
+static int sierra_send_setup(struct tty_struct *tty,
+						struct usb_serial_port *port)
 {
 	struct usb_serial *serial = port->serial;
 	struct sierra_port_private *portdata;
@@ -260,7 +261,7 @@ static int sierra_send_setup(struct usb_serial_port *port)
 
 	portdata = usb_get_serial_port_data(port);
 
-	if (port->tty) {
+	if (tty) {
 		int val = 0;
 		if (portdata->dtr_state)
 			val |= 0x01;
@@ -284,32 +285,17 @@ static int sierra_send_setup(struct usb_serial_port *port)
 	return 0;
 }
 
-static void sierra_rx_throttle(struct usb_serial_port *port)
+static void sierra_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	dbg("%s", __func__);
+	tty_termios_copy_hw(tty->termios, old_termios);
+	sierra_send_setup(tty, port);
 }
 
-static void sierra_rx_unthrottle(struct usb_serial_port *port)
-{
-	dbg("%s", __func__);
-}
-
-static void sierra_break_ctl(struct usb_serial_port *port, int break_state)
-{
-	/* Unfortunately, I don't know how to send a break */
-	dbg("%s", __func__);
-}
-
-static void sierra_set_termios(struct usb_serial_port *port,
-			struct ktermios *old_termios)
-{
-	dbg("%s", __func__);
-	tty_termios_copy_hw(port->tty->termios, old_termios);
-	sierra_send_setup(port);
-}
-
-static int sierra_tiocmget(struct usb_serial_port *port, struct file *file)
+static int sierra_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	unsigned int value;
 	struct sierra_port_private *portdata;
 
@@ -325,9 +311,10 @@ static int sierra_tiocmget(struct usb_serial_port *port, struct file *file)
 	return value;
 }
 
-static int sierra_tiocmset(struct usb_serial_port *port, struct file *file,
+static int sierra_tiocmset(struct tty_struct *tty, struct file *file,
 			unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct sierra_port_private *portdata;
 
 	portdata = usb_get_serial_port_data(port);
@@ -341,13 +328,7 @@ static int sierra_tiocmset(struct usb_serial_port *port, struct file *file,
 		portdata->rts_state = 0;
 	if (clear & TIOCM_DTR)
 		portdata->dtr_state = 0;
-	return sierra_send_setup(port);
-}
-
-static int sierra_ioctl(struct usb_serial_port *port, struct file *file,
-			unsigned int cmd, unsigned long arg)
-{
-	return -ENOIOCTLCMD;
+	return sierra_send_setup(tty, port);
 }
 
 static void sierra_outdat_callback(struct urb *urb)
@@ -374,8 +355,8 @@ static void sierra_outdat_callback(struct urb *urb)
 }
 
 /* Write */
-static int sierra_write(struct usb_serial_port *port,
-			const unsigned char *buf, int count)
+static int sierra_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 	struct sierra_port_private *portdata = usb_get_serial_port_data(port);
 	struct usb_serial *serial = port->serial;
@@ -463,7 +444,7 @@ static void sierra_indat_callback(struct urb *urb)
 		dbg("%s: nonzero status: %d on endpoint %02x.",
 		    __func__, status, endpoint);
 	} else {
-		tty = port->tty;
+		tty = port->port.tty;
 		if (urb->actual_length) {
 			tty_buffer_request_room(tty, urb->actual_length);
 			tty_insert_flip_string(tty, data, urb->actual_length);
@@ -473,7 +454,7 @@ static void sierra_indat_callback(struct urb *urb)
 		}
 
 		/* Resubmit urb so we continue receiving */
-		if (port->open_count && status != -ESHUTDOWN) {
+		if (port->port.count && status != -ESHUTDOWN) {
 			err = usb_submit_urb(urb, GFP_ATOMIC);
 			if (err)
 				dev_err(&port->dev, "resubmit read urb failed."
@@ -517,9 +498,9 @@ static void sierra_instat_callback(struct urb *urb)
 			portdata->dsr_state = ((signals & 0x02) ? 1 : 0);
 			portdata->ri_state = ((signals & 0x08) ? 1 : 0);
 
-			if (port->tty && !C_CLOCAL(port->tty) &&
+			if (port->port.tty && !C_CLOCAL(port->port.tty) &&
 					old_dcd_state && !portdata->dcd_state)
-				tty_hangup(port->tty);
+				tty_hangup(port->port.tty);
 		} else {
 			dbg("%s: type %x req %x", __func__,
 				req_pkt->bRequestType, req_pkt->bRequest);
@@ -537,8 +518,9 @@ static void sierra_instat_callback(struct urb *urb)
 	}
 }
 
-static int sierra_write_room(struct usb_serial_port *port)
+static int sierra_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct sierra_port_private *portdata = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -557,22 +539,8 @@ static int sierra_write_room(struct usb_serial_port *port)
 	return 2048;
 }
 
-static int sierra_chars_in_buffer(struct usb_serial_port *port)
-{
-	dbg("%s - port %d", __func__, port->number);
-
-	/*
-	 * We can't really account for how much data we
-	 * have sent out, but hasn't made it through to the
-	 * device as we can't see the backend here, so just
-	 * tell the tty layer that everything is flushed.
-	 *
-	 * FIXME: should walk the outstanding urbs info
-	 */
-	return 0;
-}
-
-static int sierra_open(struct usb_serial_port *port, struct file *filp)
+static int sierra_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct sierra_port_private *portdata;
 	struct usb_serial *serial = port->serial;
@@ -612,9 +580,10 @@ static int sierra_open(struct usb_serial_port *port, struct file *filp)
 		}
 	}
 
-	port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
-	sierra_send_setup(port);
+	sierra_send_setup(tty, port);
 
 	/* start up the interrupt endpoint if we have one */
 	if (port->interrupt_in_urb) {
@@ -626,7 +595,8 @@ static int sierra_open(struct usb_serial_port *port, struct file *filp)
 	return 0;
 }
 
-static void sierra_close(struct usb_serial_port *port, struct file *filp)
+static void sierra_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	int i;
 	struct usb_serial *serial = port->serial;
@@ -641,7 +611,7 @@ static void sierra_close(struct usb_serial_port *port, struct file *filp)
 	if (serial->dev) {
 		mutex_lock(&serial->disc_mutex);
 		if (!serial->disconnected)
-			sierra_send_setup(port);
+			sierra_send_setup(tty, port);
 		mutex_unlock(&serial->disc_mutex);
 
 		/* Stop reading/writing urbs */
@@ -651,7 +621,7 @@ static void sierra_close(struct usb_serial_port *port, struct file *filp)
 
 	usb_kill_urb(port->interrupt_in_urb);
 
-	port->tty = NULL;
+	port->port.tty = NULL;	/* FIXME */
 }
 
 static int sierra_startup(struct usb_serial *serial)
@@ -754,12 +724,7 @@ static struct usb_serial_driver sierra_device = {
 	.close             = sierra_close,
 	.write             = sierra_write,
 	.write_room        = sierra_write_room,
-	.chars_in_buffer   = sierra_chars_in_buffer,
-	.throttle          = sierra_rx_throttle,
-	.unthrottle        = sierra_rx_unthrottle,
-	.ioctl             = sierra_ioctl,
 	.set_termios       = sierra_set_termios,
-	.break_ctl         = sierra_break_ctl,
 	.tiocmget          = sierra_tiocmget,
 	.tiocmset          = sierra_tiocmset,
 	.attach            = sierra_startup,
diff --git a/drivers/usb/serial/spcp8x5.c b/drivers/usb/serial/spcp8x5.c
index 55b2570b8b8b..58495f5cca1f 100644
--- a/drivers/usb/serial/spcp8x5.c
+++ b/drivers/usb/serial/spcp8x5.c
@@ -448,7 +448,8 @@ static void spcp8x5_set_workMode(struct usb_device *dev, u16 value,
 
 /* close the serial port. We should wait for data sending to device 1st and
  * then kill all urb. */
-static void spcp8x5_close(struct usb_serial_port *port, struct file *filp)
+static void spcp8x5_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct spcp8x5_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -464,7 +465,7 @@ static void spcp8x5_close(struct usb_serial_port *port, struct file *filp)
 	spin_lock_irqsave(&priv->lock, flags);
 	timeout = SPCP8x5_CLOSING_WAIT;
 	init_waitqueue_entry(&wait, current);
-	add_wait_queue(&port->tty->write_wait, &wait);
+	add_wait_queue(&tty->write_wait, &wait);
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (ringbuf_avail_data(priv->buf) == 0 ||
@@ -475,7 +476,7 @@ static void spcp8x5_close(struct usb_serial_port *port, struct file *filp)
 		spin_lock_irqsave(&priv->lock, flags);
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->tty->write_wait, &wait);
+	remove_wait_queue(&tty->write_wait, &wait);
 
 	/* clear out any remaining data in the buffer */
 	clear_ringbuf(priv->buf);
@@ -486,7 +487,7 @@ static void spcp8x5_close(struct usb_serial_port *port, struct file *filp)
 	 * flow control for data rates of 1200 bps or more, for lower rates we
 	 * should really know how much data is in the buffer to compute a delay
 	 * that is not unnecessarily long) */
-	bps = tty_get_baud_rate(port->tty);
+	bps = tty_get_baud_rate(tty);
 	if (bps > 1200)
 		timeout = max((HZ*2560) / bps, HZ/10);
 	else
@@ -495,8 +496,8 @@ static void spcp8x5_close(struct usb_serial_port *port, struct file *filp)
 	schedule_timeout(timeout);
 
 	/* clear control lines */
-	if (port->tty) {
-		c_cflag = port->tty->termios->c_cflag;
+	if (tty) {
+		c_cflag = tty->termios->c_cflag;
 		if (c_cflag & HUPCL) {
 			spin_lock_irqsave(&priv->lock, flags);
 			priv->line_control = 0;
@@ -518,14 +519,14 @@ static void spcp8x5_close(struct usb_serial_port *port, struct file *filp)
 }
 
 /* set the serial param for transfer. we should check if we really need to
- * transfer. then if be set flow contorl we should do this too. */
-static void spcp8x5_set_termios(struct usb_serial_port *port,
-				struct ktermios *old_termios)
+ * transfer. if we set flow control we should do this too. */
+static void spcp8x5_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct usb_serial *serial = port->serial;
 	struct spcp8x5_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
-	unsigned int cflag = port->tty->termios->c_cflag;
+	unsigned int cflag = tty->termios->c_cflag;
 	unsigned int old_cflag = old_termios->c_cflag;
 	unsigned short uartdata;
 	unsigned char buf[2] = {0, 0};
@@ -533,21 +534,19 @@ static void spcp8x5_set_termios(struct usb_serial_port *port,
 	int i;
 	u8 control;
 
-	if ((!port->tty) || (!port->tty->termios))
-		return;
-
 	/* for the 1st time call this function */
 	spin_lock_irqsave(&priv->lock, flags);
 	if (!priv->termios_initialized) {
-		*(port->tty->termios) = tty_std_termios;
-		port->tty->termios->c_cflag = B115200 | CS8 | CREAD |
-					      HUPCL | CLOCAL;
+		*(tty->termios) = tty_std_termios;
+		tty->termios->c_cflag = B115200 | CS8 | CREAD | HUPCL | CLOCAL;
+		tty->termios->c_ispeed = 115200;
+		tty->termios->c_ospeed = 115200;
 		priv->termios_initialized = 1;
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	/* check that they really want us to change something */
-	if (!tty_termios_hw_change(port->tty->termios, old_termios))
+	if (!tty_termios_hw_change(tty->termios, old_termios))
 		return;
 
 	/* set DTR/RTS active */
@@ -567,7 +566,7 @@ static void spcp8x5_set_termios(struct usb_serial_port *port,
 	}
 
 	/* Set Baud Rate */
-	baud = tty_get_baud_rate(port->tty);;
+	baud = tty_get_baud_rate(tty);;
 	switch (baud) {
 	case 300:	buf[0] = 0x00;	break;
 	case 600:	buf[0] = 0x01;	break;
@@ -643,7 +642,8 @@ static void spcp8x5_set_termios(struct usb_serial_port *port,
 
 /* open the serial port. do some usb system call. set termios and get the line
  * status of the device. then submit the read urb */
-static int spcp8x5_open(struct usb_serial_port *port, struct file *filp)
+static int spcp8x5_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	struct ktermios tmp_termios;
 	struct usb_serial *serial = port->serial;
@@ -665,7 +665,7 @@ static int spcp8x5_open(struct usb_serial_port *port, struct file *filp)
 		return ret;
 
 	spin_lock_irqsave(&priv->lock, flags);
-	if (port->tty->termios->c_cflag & CBAUD)
+	if (tty && (tty->termios->c_cflag & CBAUD))
 		priv->line_control = MCR_DTR | MCR_RTS;
 	else
 		priv->line_control = 0;
@@ -674,8 +674,8 @@ static int spcp8x5_open(struct usb_serial_port *port, struct file *filp)
 	spcp8x5_set_ctrlLine(serial->dev, priv->line_control , priv->type);
 
 	/* Setup termios */
-	if (port->tty)
-		spcp8x5_set_termios(port, &tmp_termios);
+	if (tty)
+		spcp8x5_set_termios(tty, port, &tmp_termios);
 
 	spcp8x5_get_msr(serial->dev, &status, priv->type);
 
@@ -690,7 +690,7 @@ static int spcp8x5_open(struct usb_serial_port *port, struct file *filp)
 	port->read_urb->dev = serial->dev;
 	ret = usb_submit_urb(port->read_urb, GFP_KERNEL);
 	if (ret) {
-		spcp8x5_close(port, NULL);
+		spcp8x5_close(tty, port, NULL);
 		return -EPROTO;
 	}
 	return 0;
@@ -717,7 +717,7 @@ static void spcp8x5_read_bulk_callback(struct urb *urb)
 
 	/* check the urb status */
 	if (urb->status) {
-		if (!port->open_count)
+		if (!port->port.count)
 			return;
 		if (urb->status == -EPROTO) {
 			/* spcp8x5 mysteriously fails with -EPROTO */
@@ -755,7 +755,7 @@ static void spcp8x5_read_bulk_callback(struct urb *urb)
 		tty_flag = TTY_FRAME;
 	dev_dbg(&port->dev, "tty_flag = %d\n", tty_flag);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length + 1);
 		/* overrun is special, not associated with a char */
@@ -767,7 +767,7 @@ static void spcp8x5_read_bulk_callback(struct urb *urb)
 	}
 
 	/* Schedule the next read _if_ we are still open */
-	if (port->open_count) {
+	if (port->port.count) {
 		urb->dev = port->serial->dev;
 		result = usb_submit_urb(urb , GFP_ATOMIC);
 		if (result)
@@ -866,7 +866,7 @@ static void spcp8x5_write_bulk_callback(struct urb *urb)
 }
 
 /* write data to ring buffer. and then start the write transfer */
-static int spcp8x5_write(struct usb_serial_port *port,
+static int spcp8x5_write(struct tty_struct *tty, struct usb_serial_port *port,
 			 const unsigned char *buf, int count)
 {
 	struct spcp8x5_private *priv = usb_get_serial_port_data(port);
@@ -925,9 +925,10 @@ static int spcp8x5_wait_modem_info(struct usb_serial_port *port,
 	return 0;
 }
 
-static int spcp8x5_ioctl(struct usb_serial_port *port, struct file *file,
+static int spcp8x5_ioctl(struct tty_struct *tty, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s (%d) cmd = 0x%04x", __func__, port->number, cmd);
 
 	switch (cmd) {
@@ -943,9 +944,10 @@ static int spcp8x5_ioctl(struct usb_serial_port *port, struct file *file,
 	return -ENOIOCTLCMD;
 }
 
-static int spcp8x5_tiocmset(struct usb_serial_port *port, struct file *file,
+static int spcp8x5_tiocmset(struct tty_struct *tty, struct file *file,
 			    unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct spcp8x5_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	u8 control;
@@ -965,8 +967,9 @@ static int spcp8x5_tiocmset(struct usb_serial_port *port, struct file *file,
 	return spcp8x5_set_ctrlLine(port->serial->dev, control , priv->type);
 }
 
-static int spcp8x5_tiocmget(struct usb_serial_port *port, struct file *file)
+static int spcp8x5_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct spcp8x5_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	unsigned int mcr;
@@ -989,8 +992,9 @@ static int spcp8x5_tiocmget(struct usb_serial_port *port, struct file *file)
 }
 
 /* get the avail space room in ring buffer */
-static int spcp8x5_write_room(struct usb_serial_port *port)
+static int spcp8x5_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct spcp8x5_private *priv = usb_get_serial_port_data(port);
 	int room = 0;
 	unsigned long flags;
@@ -1003,8 +1007,9 @@ static int spcp8x5_write_room(struct usb_serial_port *port)
 }
 
 /* get the number of avail data in write ring buffer */
-static int spcp8x5_chars_in_buffer(struct usb_serial_port *port)
+static int spcp8x5_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct spcp8x5_private *priv = usb_get_serial_port_data(port);
 	int chars = 0;
 	unsigned long flags;
diff --git a/drivers/usb/serial/ti_usb_3410_5052.c b/drivers/usb/serial/ti_usb_3410_5052.c
index a26a629dfc4f..48831a755fc1 100644
--- a/drivers/usb/serial/ti_usb_3410_5052.c
+++ b/drivers/usb/serial/ti_usb_3410_5052.c
@@ -70,6 +70,7 @@
 
 #include <linux/kernel.h>
 #include <linux/errno.h>
+#include <linux/firmware.h>
 #include <linux/init.h>
 #include <linux/slab.h>
 #include <linux/tty.h>
@@ -149,21 +150,23 @@ struct ti_device {
 
 static int ti_startup(struct usb_serial *serial);
 static void ti_shutdown(struct usb_serial *serial);
-static int ti_open(struct usb_serial_port *port, struct file *file);
-static void ti_close(struct usb_serial_port *port, struct file *file);
-static int ti_write(struct usb_serial_port *port, const unsigned char *data,
-	int count);
-static int ti_write_room(struct usb_serial_port *port);
-static int ti_chars_in_buffer(struct usb_serial_port *port);
-static void ti_throttle(struct usb_serial_port *port);
-static void ti_unthrottle(struct usb_serial_port *port);
-static int ti_ioctl(struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg);
-static void ti_set_termios(struct usb_serial_port *port,
+static int ti_open(struct tty_struct *tty, struct usb_serial_port *port,
+				struct file *file);
+static void ti_close(struct tty_struct *tty, struct usb_serial_port *port,
+				struct file *file);
+static int ti_write(struct tty_struct *tty, struct usb_serial_port *port,
+				const unsigned char *data, int count);
+static int ti_write_room(struct tty_struct *tty);
+static int ti_chars_in_buffer(struct tty_struct *tty);
+static void ti_throttle(struct tty_struct *tty);
+static void ti_unthrottle(struct tty_struct *tty);
+static int ti_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg);
+static void ti_set_termios(struct tty_struct *tty, struct usb_serial_port *port,
 	struct ktermios *old_termios);
-static int ti_tiocmget(struct usb_serial_port *port, struct file *file);
-static int ti_tiocmset(struct usb_serial_port *port, struct file *file,
+static int ti_tiocmget(struct tty_struct *tty, struct file *file);
+static int ti_tiocmset(struct tty_struct *tty, struct file *file,
 	unsigned int set, unsigned int clear);
-static void ti_break(struct usb_serial_port *port, int break_state);
+static void ti_break(struct tty_struct *tty, int break_state);
 static void ti_interrupt_callback(struct urb *urb);
 static void ti_bulk_in_callback(struct urb *urb);
 static void ti_bulk_out_callback(struct urb *urb);
@@ -192,8 +195,7 @@ static int ti_command_in_sync(struct ti_device *tdev, __u8 command,
 static int ti_write_byte(struct ti_device *tdev, unsigned long addr,
 	__u8 mask, __u8 byte);
 
-static int ti_download_firmware(struct ti_device *tdev, char *fw_name);
-
+static int ti_download_firmware(struct ti_device *tdev, int type);
 
 /* circular buffer */
 static struct circ_buf *ti_buf_alloc(void);
@@ -430,11 +432,10 @@ static int ti_startup(struct usb_serial *serial)
 
 	/* if we have only 1 configuration, download firmware */
 	if (dev->descriptor.bNumConfigurations == 1) {
-
 		if (tdev->td_is_3410)
-			status = ti_download_firmware(tdev, "ti_3410.fw");
+			status = ti_download_firmware(tdev, 3410);
 		else
-			status = ti_download_firmware(tdev, "ti_5052.fw");
+			status = ti_download_firmware(tdev, 5052);
 		if (status)
 			goto free_tdev;
 
@@ -519,7 +520,8 @@ static void ti_shutdown(struct usb_serial *serial)
 }
 
 
-static int ti_open(struct usb_serial_port *port, struct file *file)
+static int ti_open(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *file)
 {
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	struct ti_device *tdev;
@@ -543,9 +545,9 @@ static int ti_open(struct usb_serial_port *port, struct file *file)
 	if (mutex_lock_interruptible(&tdev->td_open_close_lock))
 		return -ERESTARTSYS;
 
-	if (port->tty)
-		port->tty->low_latency = 
-			(tport->tp_flags & ASYNC_LOW_LATENCY) ? 1 : 0;
+	if (tty)
+		tty->low_latency =
+				(tport->tp_flags & ASYNC_LOW_LATENCY) ? 1 : 0;
 
 	port_number = port->number - port->serial->minor;
 
@@ -573,7 +575,8 @@ static int ti_open(struct usb_serial_port *port, struct file *file)
 		}
 	}
 
-	ti_set_termios(port, port->tty->termios);
+	if (tty)
+		ti_set_termios(tty, port, tty->termios);
 
 	dbg("%s - sending TI_OPEN_PORT", __func__);
 	status = ti_command_out_sync(tdev, TI_OPEN_PORT,
@@ -610,7 +613,8 @@ static int ti_open(struct usb_serial_port *port, struct file *file)
 	usb_clear_halt(dev, port->write_urb->pipe);
 	usb_clear_halt(dev, port->read_urb->pipe);
 
-	ti_set_termios(port, port->tty->termios);
+	if (tty)
+		ti_set_termios(tty, port, tty->termios);
 
 	dbg("%s - sending TI_OPEN_PORT (2)", __func__);
 	status = ti_command_out_sync(tdev, TI_OPEN_PORT,
@@ -661,7 +665,8 @@ release_lock:
 }
 
 
-static void ti_close(struct usb_serial_port *port, struct file *file)
+static void ti_close(struct tty_struct *tty, struct usb_serial_port *port,
+							struct file *file)
 {
 	struct ti_device *tdev;
 	struct ti_port *tport;
@@ -707,8 +712,8 @@ static void ti_close(struct usb_serial_port *port, struct file *file)
 }
 
 
-static int ti_write(struct usb_serial_port *port, const unsigned char *data,
-	int count)
+static int ti_write(struct tty_struct *tty, struct usb_serial_port *port,
+			const unsigned char *data, int count)
 {
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -733,8 +738,9 @@ static int ti_write(struct usb_serial_port *port, const unsigned char *data,
 }
 
 
-static int ti_write_room(struct usb_serial_port *port)
+static int ti_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	int room = 0;
 	unsigned long flags;
@@ -753,8 +759,9 @@ static int ti_write_room(struct usb_serial_port *port)
 }
 
 
-static int ti_chars_in_buffer(struct usb_serial_port *port)
+static int ti_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	int chars = 0;
 	unsigned long flags;
@@ -773,32 +780,26 @@ static int ti_chars_in_buffer(struct usb_serial_port *port)
 }
 
 
-static void ti_throttle(struct usb_serial_port *port)
+static void ti_throttle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
-	struct tty_struct *tty;
 
 	dbg("%s - port %d", __func__, port->number);
 
 	if (tport == NULL)
 		return;
 
-	tty = port->tty;
-	if (!tty) {
-		dbg("%s - no tty", __func__);
-		return;
-	}
-
 	if (I_IXOFF(tty) || C_CRTSCTS(tty))
 		ti_stop_read(tport, tty);
 
 }
 
 
-static void ti_unthrottle(struct usb_serial_port *port)
+static void ti_unthrottle(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
-	struct tty_struct *tty;
 	int status;
 
 	dbg("%s - port %d", __func__, port->number);
@@ -806,12 +807,6 @@ static void ti_unthrottle(struct usb_serial_port *port)
 	if (tport == NULL)
 		return;
 
-	tty = port->tty;
-	if (!tty) {
-		dbg("%s - no tty", __func__);
-		return;
-	}
-
 	if (I_IXOFF(tty) || C_CRTSCTS(tty)) {
 		status = ti_restart_read(tport, tty);
 		if (status)
@@ -820,9 +815,10 @@ static void ti_unthrottle(struct usb_serial_port *port)
 }
 
 
-static int ti_ioctl(struct usb_serial_port *port, struct file *file,
+static int ti_ioctl(struct tty_struct *tty, struct file *file,
 	unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	struct async_icount cnow;
 	struct async_icount cprev;
@@ -875,11 +871,10 @@ static int ti_ioctl(struct usb_serial_port *port, struct file *file,
 }
 
 
-static void ti_set_termios(struct usb_serial_port *port,
-	struct ktermios *old_termios)
+static void ti_set_termios(struct tty_struct *tty,
+		struct usb_serial_port *port, struct ktermios *old_termios)
 {
 	struct ti_port *tport = usb_get_serial_port_data(port);
-	struct tty_struct *tty = port->tty;
 	struct ti_uart_config *config;
 	tcflag_t cflag,iflag;
 	int baud;
@@ -1008,8 +1003,9 @@ static void ti_set_termios(struct usb_serial_port *port,
 }
 
 
-static int ti_tiocmget(struct usb_serial_port *port, struct file *file)
+static int ti_tiocmget(struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	unsigned int result;
 	unsigned int msr;
@@ -1040,9 +1036,10 @@ static int ti_tiocmget(struct usb_serial_port *port, struct file *file)
 }
 
 
-static int ti_tiocmset(struct usb_serial_port *port, struct file *file,
+static int ti_tiocmset(struct tty_struct *tty, struct file *file,
 	unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	unsigned int mcr;
 	unsigned long flags;
@@ -1074,8 +1071,9 @@ static int ti_tiocmset(struct usb_serial_port *port, struct file *file,
 }
 
 
-static void ti_break(struct usb_serial_port *port, int break_state)
+static void ti_break(struct tty_struct *tty, int break_state)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct ti_port *tport = usb_get_serial_port_data(port);
 	int status;
 
@@ -1213,14 +1211,14 @@ static void ti_bulk_in_callback(struct urb *urb)
 		return;
 	}
 
-	if (port->tty && urb->actual_length) {
+	if (port->port.tty && urb->actual_length) {
 		usb_serial_debug_data(debug, dev, __func__,
 			urb->actual_length, urb->transfer_buffer);
 
 		if (!tport->tp_is_open)
 			dbg("%s - port closed, dropping data", __func__);
 		else
-			ti_recv(&urb->dev->dev, port->tty, urb->transfer_buffer,
+			ti_recv(&urb->dev->dev, port->port.tty, urb->transfer_buffer,
 				urb->actual_length);
 
 		spin_lock(&tport->tp_lock);
@@ -1302,7 +1300,7 @@ static void ti_send(struct ti_port *tport)
 {
 	int count, result;
 	struct usb_serial_port *port = tport->tp_port;
-	struct tty_struct *tty = port->tty;
+	struct tty_struct *tty = port->port.tty;	/* FIXME */
 	unsigned long flags;
 
 
@@ -1442,8 +1440,9 @@ static int ti_set_serial_info(struct ti_port *tport,
 		return -EFAULT;
 
 	tport->tp_flags = new_serial.flags & TI_SET_SERIAL_FLAGS;
-	if (port->tty)
-		port->tty->low_latency =
+	/* FIXME */
+	if (port->port.tty)
+		port->port.tty->low_latency =
 			(tport->tp_flags & ASYNC_LOW_LATENCY) ? 1 : 0;
 	tport->tp_closing_wait = new_serial.closing_wait;
 
@@ -1477,7 +1476,7 @@ static void ti_handle_new_msr(struct ti_port *tport, __u8 msr)
 	tport->tp_msr = msr & TI_MSR_MASK;
 
 	/* handle CTS flow control */
-	tty = tport->tp_port->tty;
+	tty = tport->tp_port->port.tty;
 	if (tty && C_CRTSCTS(tty)) {
 		if (msr & TI_MSR_CTS) {
 			tty->hw_stopped = 0;
@@ -1655,65 +1654,65 @@ static int ti_write_byte(struct ti_device *tdev, unsigned long addr,
 	return status;
 }
 
-
-static int ti_download_firmware(struct ti_device *tdev,
-				char *fw_name)
+static int ti_do_download(struct usb_device *dev, int pipe,
+						u8 *buffer, int size)
 {
-	const struct firmware *fw;
-	int status = 0;
-	int buffer_size;
 	int pos;
-	int len;
+	u8 cs = 0;
 	int done;
-	__u8 cs = 0;
-	__u8 *buffer;
-	struct usb_device *dev = tdev->td_serial->dev;
 	struct ti_firmware_header *header;
-	unsigned int pipe = usb_sndbulkpipe(dev,
-		tdev->td_serial->port[0]->bulk_out_endpointAddress);
-
-	buffer_size = TI_FIRMWARE_BUF_SIZE + sizeof(struct ti_firmware_header);
-
-	if (request_firmware(&fw, fw_name, &dev->dev)) {
-		dev_err(&dev->dev, "%s - failed to load firmware \"%s\"\n",
-			__func__, fw_name);
-		return -ENOENT;
-	}
-	if (fw->size > buffer_size) {
-		dev_err(&dev->dev, "%s - firmware \"%s\" is too large\n",
-			__func__, fw_name);
-		release_firmware(fw);
-		return -EINVAL;
-	}
-
-	buffer = kmalloc(buffer_size, GFP_KERNEL);
-	if (!buffer) {
-		dev_err(&dev->dev, "%s - out of memory\n", __func__);
-		release_firmware(fw);
-		return -ENOMEM;
-	}
-
-	memcpy(buffer, fw->data, fw->size);
-	memset(buffer+fw->size, 0xff, buffer_size-fw->size);
-
-	for(pos = sizeof(struct ti_firmware_header); pos < buffer_size; pos++)
+	int status;
+	int len;
+	
+	for(pos = sizeof(struct ti_firmware_header); pos < size; pos++)
 		cs = (__u8)(cs + buffer[pos]);
 
 	header = (struct ti_firmware_header *)buffer;
-	header->wLength = cpu_to_le16((__u16)(buffer_size - sizeof(struct ti_firmware_header)));
+	header->wLength = cpu_to_le16((__u16)(size 
+					- sizeof(struct ti_firmware_header)));
 	header->bCheckSum = cs;
 
 	dbg("%s - downloading firmware", __func__);
-	for (pos = 0; pos < buffer_size; pos += done) {
-		len = min(buffer_size - pos, TI_DOWNLOAD_MAX_PACKET_SIZE);
-		status = usb_bulk_msg(dev, pipe, buffer+pos, len, &done, 1000);
+	for (pos = 0; pos < size; pos += done) {
+		len = min(size - pos, TI_DOWNLOAD_MAX_PACKET_SIZE);
+		status = usb_bulk_msg(dev, pipe, buffer + pos, len,
+								&done, 1000);
 		if (status)
 			break;
 	}
+	return status;
+}
 
-	kfree(buffer);
-	release_firmware(fw);
+static int ti_download_firmware(struct ti_device *tdev, int type)
+{
+	int status = -ENOMEM;
+	int buffer_size;
+	__u8 *buffer;
+	struct usb_device *dev = tdev->td_serial->dev;
+	unsigned int pipe = usb_sndbulkpipe(dev,
+		tdev->td_serial->port[0]->bulk_out_endpointAddress);
+	const struct firmware *fw_p;
+	char buf[32];
+	sprintf(buf, "ti_usb-%d.bin", type);
 
+	if (request_firmware(&fw_p, buf, &dev->dev)) {
+		dev_err(&dev->dev, "%s - firmware not found\n", __func__);
+		return -ENOENT;
+	}
+	if (fw_p->size > TI_FIRMWARE_BUF_SIZE) {
+		dev_err(&dev->dev, "%s - firmware too large\n", __func__);
+		return -ENOENT;
+	}
+
+	buffer_size = TI_FIRMWARE_BUF_SIZE + sizeof(struct ti_firmware_header);
+	buffer = kmalloc(buffer_size, GFP_KERNEL);
+	if (buffer) {
+		memcpy(buffer, fw_p->data, fw_p->size);
+		memset(buffer + fw_p->size, 0xff, buffer_size - fw_p->size);
+		ti_do_download(dev, pipe, buffer, fw_p->size);
+		kfree(buffer);
+	}
+	release_firmware(fw_p);
 	if (status) {
 		dev_err(&dev->dev, "%s - error downloading firmware, %d\n", __func__, status);
 		return status;
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 353798631903..ffaed8ace066 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -143,7 +143,7 @@ static void destroy_serial(struct kref *kref)
 	return_serial(serial);
 
 	for (i = 0; i < serial->num_ports; ++i)
-		serial->port[i]->open_count = 0;
+		serial->port[i]->port.count = 0;
 
 	/* the ports are cleaned up and released in port_release() */
 	for (i = 0; i < serial->num_ports; ++i)
@@ -208,14 +208,14 @@ static int serial_open (struct tty_struct *tty, struct file * filp)
 		goto bailout_kref_put;
 	}
 	 
-	++port->open_count;
+	++port->port.count;
 
 	/* set up our port structure making the tty driver
 	 * remember our port object, and us it */
 	tty->driver_data = port;
-	port->tty = tty;
+	port->port.tty = tty;
 
-	if (port->open_count == 1) {
+	if (port->port.count == 1) {
 
 		/* lock this module before we call it
 		 * this may fail, which means we must bail out,
@@ -230,7 +230,7 @@ static int serial_open (struct tty_struct *tty, struct file * filp)
 			goto bailout_module_put;
 		/* only call the device specific open if this 
 		 * is the first time the port is opened */
-		retval = serial->type->open(port, filp);
+		retval = serial->type->open(tty, port, filp);
 		if (retval)
 			goto bailout_interface_put;
 	}
@@ -243,9 +243,9 @@ bailout_interface_put:
 bailout_module_put:
 	module_put(serial->type->driver.owner);
 bailout_mutex_unlock:
-	port->open_count = 0;
+	port->port.count = 0;
 	tty->driver_data = NULL;
-	port->tty = NULL;
+	port->port.tty = NULL;
 	mutex_unlock(&port->mutex);
 bailout_kref_put:
 	usb_serial_put(serial);
@@ -263,26 +263,26 @@ static void serial_close(struct tty_struct *tty, struct file * filp)
 
 	mutex_lock(&port->mutex);
 
-	if (port->open_count == 0) {
+	if (port->port.count == 0) {
 		mutex_unlock(&port->mutex);
 		return;
 	}
 
-	--port->open_count;
-	if (port->open_count == 0)
+	--port->port.count;
+	if (port->port.count == 0)
 		/* only call the device specific close if this 
 		 * port is being closed by the last owner */
-		port->serial->type->close(port, filp);
+		port->serial->type->close(tty, port, filp);
 
-	if (port->open_count == (port->console? 1 : 0)) {
-		if (port->tty) {
-			if (port->tty->driver_data)
-				port->tty->driver_data = NULL;
-			port->tty = NULL;
+	if (port->port.count == (port->console? 1 : 0)) {
+		if (port->port.tty) {
+			if (port->port.tty->driver_data)
+				port->port.tty->driver_data = NULL;
+			port->port.tty = NULL;
 		}
 	}
 
-	if (port->open_count == 0) {
+	if (port->port.count == 0) {
 		mutex_lock(&port->serial->disc_mutex);
 		if (!port->serial->disconnected)
 			usb_autopm_put_interface(port->serial->interface);
@@ -304,12 +304,12 @@ static int serial_write (struct tty_struct * tty, const unsigned char *buf, int
 
 	dbg("%s - port %d, %d byte(s)", __func__, port->number, count);
 
-	/* open_count is managed under the mutex lock for the tty so cannot
+	/* count is managed under the mutex lock for the tty so cannot
            drop to zero until after the last close completes */
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 
 	/* pass on to the driver specific version of this function */
-	retval = port->serial->type->write(port, buf, count);
+	retval = port->serial->type->write(tty, port, buf, count);
 
 exit:
 	return retval;
@@ -319,9 +319,9 @@ static int serial_write_room (struct tty_struct *tty)
 {
 	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s - port %d", __func__, port->number);
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	/* pass on to the driver specific version of this function */
-	return port->serial->type->write_room(port);
+	return port->serial->type->write_room(tty);
 }
 
 static int serial_chars_in_buffer (struct tty_struct *tty) 
@@ -329,9 +329,9 @@ static int serial_chars_in_buffer (struct tty_struct *tty)
 	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s = port %d", __func__, port->number);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	/* pass on to the driver specific version of this function */
-	return port->serial->type->chars_in_buffer(port);
+	return port->serial->type->chars_in_buffer(tty);
 }
 
 static void serial_throttle (struct tty_struct * tty)
@@ -339,10 +339,10 @@ static void serial_throttle (struct tty_struct * tty)
 	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s - port %d", __func__, port->number);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	/* pass on to the driver specific version of this function */
 	if (port->serial->type->throttle)
-		port->serial->type->throttle(port);
+		port->serial->type->throttle(tty);
 }
 
 static void serial_unthrottle (struct tty_struct * tty)
@@ -350,10 +350,10 @@ static void serial_unthrottle (struct tty_struct * tty)
 	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s - port %d", __func__, port->number);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	/* pass on to the driver specific version of this function */
 	if (port->serial->type->unthrottle)
-		port->serial->type->unthrottle(port);
+		port->serial->type->unthrottle(tty);
 }
 
 static int serial_ioctl (struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg)
@@ -363,12 +363,12 @@ static int serial_ioctl (struct tty_struct *tty, struct file * file, unsigned in
 
 	dbg("%s - port %d, cmd 0x%.4x", __func__, port->number, cmd);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 
 	/* pass on to the driver specific version of this function if it is available */
 	if (port->serial->type->ioctl) {
 		lock_kernel();
-		retval = port->serial->type->ioctl(port, file, cmd, arg);
+		retval = port->serial->type->ioctl(tty, file, cmd, arg);
 		unlock_kernel();
 	}
 	else
@@ -381,10 +381,10 @@ static void serial_set_termios (struct tty_struct *tty, struct ktermios * old)
 	struct usb_serial_port *port = tty->driver_data;
 	dbg("%s - port %d", __func__, port->number);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	/* pass on to the driver specific version of this function if it is available */
 	if (port->serial->type->set_termios)
-		port->serial->type->set_termios(port, old);
+		port->serial->type->set_termios(tty, port, old);
 	else
 		tty_termios_copy_hw(tty->termios, old);
 }
@@ -395,11 +395,11 @@ static void serial_break (struct tty_struct *tty, int break_state)
 
 	dbg("%s - port %d", __func__, port->number);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	/* pass on to the driver specific version of this function if it is available */
 	if (port->serial->type->break_ctl) {
 		lock_kernel();
-		port->serial->type->break_ctl(port, break_state);
+		port->serial->type->break_ctl(tty, break_state);
 		unlock_kernel();
 	}
 }
@@ -457,9 +457,9 @@ static int serial_tiocmget (struct tty_struct *tty, struct file *file)
 
 	dbg("%s - port %d", __func__, port->number);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	if (port->serial->type->tiocmget)
-		return port->serial->type->tiocmget(port, file);
+		return port->serial->type->tiocmget(tty, file);
 	return -EINVAL;
 }
 
@@ -470,9 +470,9 @@ static int serial_tiocmset (struct tty_struct *tty, struct file *file,
 
 	dbg("%s - port %d", __func__, port->number);
 
-	WARN_ON(!port->open_count);
+	WARN_ON(!port->port.count);
 	if (port->serial->type->tiocmset)
-		return port->serial->type->tiocmset(port, file, set, clear);
+		return port->serial->type->tiocmset(tty, file, set, clear);
 	return -EINVAL;
 }
 
@@ -497,7 +497,7 @@ static void usb_serial_port_work(struct work_struct *work)
 	if (!port)
 		return;
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (!tty)
 		return;
 
@@ -1010,8 +1010,8 @@ void usb_serial_disconnect(struct usb_interface *interface)
 	for (i = 0; i < serial->num_ports; ++i) {
 		port = serial->port[i];
 		if (port) {
-			if (port->tty)
-				tty_hangup(port->tty);
+			if (port->port.tty)
+				tty_hangup(port->port.tty);
 			kill_traffic(port);
 		}
 	}
diff --git a/drivers/usb/serial/usb_debug.c b/drivers/usb/serial/usb_debug.c
index 9ca4d4db1ddd..fc5d9952b03b 100644
--- a/drivers/usb/serial/usb_debug.c
+++ b/drivers/usb/serial/usb_debug.c
@@ -31,10 +31,11 @@ static struct usb_driver debug_driver = {
 	.no_dynamic_id = 	1,
 };
 
-int usb_debug_open(struct usb_serial_port *port, struct file *filp)
+int usb_debug_open(struct tty_struct *tty, struct usb_serial_port *port,
+							struct file *filp)
 {
 	port->bulk_out_size = USB_DEBUG_MAX_PACKET_SIZE;
-	return usb_serial_generic_open(port, filp);
+	return usb_serial_generic_open(tty, port, filp);
 }
 
 static struct usb_serial_driver debug_device = {
diff --git a/drivers/usb/serial/visor.c b/drivers/usb/serial/visor.c
index 5fc20122145f..373a3c7ea77b 100644
--- a/drivers/usb/serial/visor.c
+++ b/drivers/usb/serial/visor.c
@@ -35,17 +35,15 @@
 #define DRIVER_DESC "USB HandSpring Visor / Palm OS driver"
 
 /* function prototypes for a handspring visor */
-static int  visor_open		(struct usb_serial_port *port, struct file *filp);
-static void visor_close		(struct usb_serial_port *port, struct file *filp);
-static int  visor_write		(struct usb_serial_port *port, const unsigned char *buf, int count);
-static int  visor_write_room		(struct usb_serial_port *port);
-static int  visor_chars_in_buffer	(struct usb_serial_port *port);
-static void visor_throttle	(struct usb_serial_port *port);
-static void visor_unthrottle	(struct usb_serial_port *port);
+static int  visor_open		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void visor_close		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static int  visor_write		(struct tty_struct *tty, struct usb_serial_port *port, const unsigned char *buf, int count);
+static int  visor_write_room	(struct tty_struct *tty);
+static void visor_throttle	(struct tty_struct *tty);
+static void visor_unthrottle	(struct tty_struct *tty);
 static int  visor_probe		(struct usb_serial *serial, const struct usb_device_id *id);
 static int  visor_calc_num_ports(struct usb_serial *serial);
 static void visor_shutdown	(struct usb_serial *serial);
-static int  visor_ioctl		(struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg);
 static void visor_write_bulk_callback	(struct urb *urb);
 static void visor_read_bulk_callback	(struct urb *urb);
 static void visor_read_int_callback	(struct urb *urb);
@@ -198,10 +196,8 @@ static struct usb_serial_driver handspring_device = {
 	.probe =		visor_probe,
 	.calc_num_ports =	visor_calc_num_ports,
 	.shutdown =		visor_shutdown,
-	.ioctl =		visor_ioctl,
 	.write =		visor_write,
 	.write_room =		visor_write_room,
-	.chars_in_buffer =	visor_chars_in_buffer,
 	.write_bulk_callback =	visor_write_bulk_callback,
 	.read_bulk_callback =	visor_read_bulk_callback,
 	.read_int_callback =	visor_read_int_callback,
@@ -225,10 +221,8 @@ static struct usb_serial_driver clie_5_device = {
 	.probe =		visor_probe,
 	.calc_num_ports =	visor_calc_num_ports,
 	.shutdown =		visor_shutdown,
-	.ioctl =		visor_ioctl,
 	.write =		visor_write,
 	.write_room =		visor_write_room,
-	.chars_in_buffer =	visor_chars_in_buffer,
 	.write_bulk_callback =	visor_write_bulk_callback,
 	.read_bulk_callback =	visor_read_bulk_callback,
 	.read_int_callback =	visor_read_int_callback,
@@ -249,10 +243,8 @@ static struct usb_serial_driver clie_3_5_device = {
 	.throttle =		visor_throttle,
 	.unthrottle =		visor_unthrottle,
 	.attach =		clie_3_5_startup,
-	.ioctl =		visor_ioctl,
 	.write =		visor_write,
 	.write_room =		visor_write_room,
-	.chars_in_buffer =	visor_chars_in_buffer,
 	.write_bulk_callback =	visor_write_bulk_callback,
 	.read_bulk_callback =	visor_read_bulk_callback,
 };
@@ -274,7 +266,7 @@ static int stats;
 /******************************************************************************
  * Handspring Visor specific driver functions
  ******************************************************************************/
-static int visor_open (struct usb_serial_port *port, struct file *filp)
+static int visor_open (struct tty_struct *tty, struct usb_serial_port *port, struct file *filp)
 {
 	struct usb_serial *serial = port->serial;
 	struct visor_private *priv = usb_get_serial_port_data(port);
@@ -300,8 +292,8 @@ static int visor_open (struct usb_serial_port *port, struct file *filp)
 	 * through, otherwise it is scheduled, and with high data rates (like
 	 * with OHCI) data can get lost.
 	 */
-	if (port->tty)
-		port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	/* Start reading from the device */
 	usb_fill_bulk_urb (port->read_urb, serial->dev,
@@ -329,7 +321,8 @@ exit:
 }
 
 
-static void visor_close (struct usb_serial_port *port, struct file * filp)
+static void visor_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct visor_private *priv = usb_get_serial_port_data(port);
 	unsigned char *transfer_buffer;
@@ -361,7 +354,8 @@ static void visor_close (struct usb_serial_port *port, struct file * filp)
 }
 
 
-static int visor_write (struct usb_serial_port *port, const unsigned char *buf, int count)
+static int visor_write(struct tty_struct *tty, struct usb_serial_port *port,
+					const unsigned char *buf, int count)
 {
 	struct visor_private *priv = usb_get_serial_port_data(port);
 	struct usb_serial *serial = port->serial;
@@ -435,8 +429,9 @@ error_no_buffer:
 }
 
 
-static int visor_write_room (struct usb_serial_port *port)
+static int visor_write_room (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct visor_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -460,22 +455,6 @@ static int visor_write_room (struct usb_serial_port *port)
 }
 
 
-static int visor_chars_in_buffer (struct usb_serial_port *port)
-{
-	dbg("%s - port %d", __func__, port->number);
-
-	/* 
-	 * We can't really account for how much data we
-	 * have sent out, but hasn't made it through to the
-	 * device, so just tell the tty layer that everything
-	 * is flushed.
-	 *
-	 * FIXME: Should walk outstanding_urbs
-	 */
-	return 0;
-}
-
-
 static void visor_write_bulk_callback (struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
@@ -520,7 +499,7 @@ static void visor_read_bulk_callback (struct urb *urb)
 
 	usb_serial_debug_data(debug, &port->dev, __func__, urb->actual_length, data);
 
-	tty = port->tty;
+	tty = port->port.tty;
 	if (tty && urb->actual_length) {
 		available_room = tty_buffer_request_room(tty, urb->actual_length);
 		if (available_room) {
@@ -591,8 +570,9 @@ exit:
 			__func__, result);
 }
 
-static void visor_throttle (struct usb_serial_port *port)
+static void visor_throttle (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct visor_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -603,8 +583,9 @@ static void visor_throttle (struct usb_serial_port *port)
 }
 
 
-static void visor_unthrottle (struct usb_serial_port *port)
+static void visor_unthrottle (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct visor_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	int result;
@@ -922,13 +903,6 @@ static void visor_shutdown (struct usb_serial *serial)
 	}
 }
 
-static int visor_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg)
-{
-	dbg("%s - port %d, cmd 0x%.4x", __func__, port->number, cmd);
-
-	return -ENOIOCTLCMD;
-}
-
 static int __init visor_init (void)
 {
 	int i, retval;
diff --git a/drivers/usb/serial/whiteheat.c b/drivers/usb/serial/whiteheat.c
index 665aa77a917b..b07d6a5cac31 100644
--- a/drivers/usb/serial/whiteheat.c
+++ b/drivers/usb/serial/whiteheat.c
@@ -142,18 +142,18 @@ static int  whiteheat_firmware_attach	(struct usb_serial *serial);
 /* function prototypes for the Connect Tech WhiteHEAT serial converter */
 static int  whiteheat_attach		(struct usb_serial *serial);
 static void whiteheat_shutdown		(struct usb_serial *serial);
-static int  whiteheat_open		(struct usb_serial_port *port, struct file *filp);
-static void whiteheat_close		(struct usb_serial_port *port, struct file *filp);
-static int  whiteheat_write		(struct usb_serial_port *port, const unsigned char *buf, int count);
-static int  whiteheat_write_room	(struct usb_serial_port *port);
-static int  whiteheat_ioctl		(struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg);
-static void whiteheat_set_termios	(struct usb_serial_port *port, struct ktermios * old);
-static int  whiteheat_tiocmget		(struct usb_serial_port *port, struct file *file);
-static int  whiteheat_tiocmset		(struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear);
-static void whiteheat_break_ctl		(struct usb_serial_port *port, int break_state);
-static int  whiteheat_chars_in_buffer	(struct usb_serial_port *port);
-static void whiteheat_throttle		(struct usb_serial_port *port);
-static void whiteheat_unthrottle	(struct usb_serial_port *port);
+static int  whiteheat_open		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static void whiteheat_close		(struct tty_struct *tty, struct usb_serial_port *port, struct file *filp);
+static int  whiteheat_write		(struct tty_struct *tty, struct usb_serial_port *port, const unsigned char *buf, int count);
+static int  whiteheat_write_room	(struct tty_struct *tty);
+static int  whiteheat_ioctl		(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg);
+static void whiteheat_set_termios	(struct tty_struct *tty, struct usb_serial_port *port, struct ktermios * old);
+static int  whiteheat_tiocmget		(struct tty_struct *tty, struct file *file);
+static int  whiteheat_tiocmset		(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear);
+static void whiteheat_break_ctl		(struct tty_struct *tty, int break_state);
+static int  whiteheat_chars_in_buffer	(struct tty_struct *tty);
+static void whiteheat_throttle		(struct tty_struct *tty);
+static void whiteheat_unthrottle	(struct tty_struct *tty);
 static void whiteheat_read_callback	(struct urb *urb);
 static void whiteheat_write_callback	(struct urb *urb);
 
@@ -246,7 +246,7 @@ static void rx_data_softint(struct work_struct *work);
 static int firm_send_command(struct usb_serial_port *port, __u8 command, __u8 *data, __u8 datasize);
 static int firm_open(struct usb_serial_port *port);
 static int firm_close(struct usb_serial_port *port);
-static int firm_setup_port(struct usb_serial_port *port);
+static int firm_setup_port(struct tty_struct *tty);
 static int firm_set_rts(struct usb_serial_port *port, __u8 onoff);
 static int firm_set_dtr(struct usb_serial_port *port, __u8 onoff);
 static int firm_set_break(struct usb_serial_port *port, __u8 onoff);
@@ -613,7 +613,8 @@ static void whiteheat_shutdown (struct usb_serial *serial)
 }
 
 
-static int whiteheat_open (struct usb_serial_port *port, struct file *filp)
+static int whiteheat_open (struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp)
 {
 	int		retval = 0;
 	struct ktermios	old_term;
@@ -624,7 +625,8 @@ static int whiteheat_open (struct usb_serial_port *port, struct file *filp)
 	if (retval)
 		goto exit;
 
-	port->tty->low_latency = 1;
+	if (tty)
+		tty->low_latency = 1;
 
 	/* send an open port command */
 	retval = firm_open(port);
@@ -640,9 +642,11 @@ static int whiteheat_open (struct usb_serial_port *port, struct file *filp)
 		goto exit;
 	}
 
-	old_term.c_cflag = ~port->tty->termios->c_cflag;
-	old_term.c_iflag = ~port->tty->termios->c_iflag;
-	whiteheat_set_termios(port, &old_term);
+	if (tty) {
+		old_term.c_cflag = ~tty->termios->c_cflag;
+		old_term.c_iflag = ~tty->termios->c_iflag;
+		whiteheat_set_termios(tty, port, &old_term);
+	}
 
 	/* Work around HCD bugs */
 	usb_clear_halt(port->serial->dev, port->read_urb->pipe);
@@ -663,7 +667,8 @@ exit:
 }
 
 
-static void whiteheat_close(struct usb_serial_port *port, struct file * filp)
+static void whiteheat_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file * filp)
 {
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
 	struct whiteheat_urb_wrap *wrap;
@@ -681,7 +686,7 @@ static void whiteheat_close(struct usb_serial_port *port, struct file * filp)
 	}
 	mutex_unlock(&port->serial->disc_mutex);
 
-	port->tty->closing = 1;
+	tty->closing = 1;
 
 /*
  * Not currently in use; tty_wait_until_sent() calls
@@ -689,12 +694,12 @@ static void whiteheat_close(struct usb_serial_port *port, struct file * filp)
  * acquisition. This should be fixed at some point. Greg's been
  * notified.
 	if ((filp->f_flags & (O_NDELAY | O_NONBLOCK)) == 0) {
-		tty_wait_until_sent(port->tty, CLOSING_DELAY);
+		tty_wait_until_sent(tty, CLOSING_DELAY);
 	}
 */
 
-	tty_driver_flush_buffer(port->tty);
-	tty_ldisc_flush(port->tty);
+	tty_driver_flush_buffer(tty);
+	tty_ldisc_flush(tty);
 
 	firm_report_tx_done(port);
 
@@ -728,11 +733,12 @@ static void whiteheat_close(struct usb_serial_port *port, struct file * filp)
 
 	stop_command_port(port->serial);
 
-	port->tty->closing = 0;
+	tty->closing = 0;
 }
 
 
-static int whiteheat_write(struct usb_serial_port *port, const unsigned char *buf, int count)
+static int whiteheat_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count)
 {
 	struct usb_serial *serial = port->serial;
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
@@ -791,8 +797,9 @@ static int whiteheat_write(struct usb_serial_port *port, const unsigned char *bu
 }
 
 
-static int whiteheat_write_room(struct usb_serial_port *port)
+static int whiteheat_write_room(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
 	struct list_head *tmp;
 	int room = 0;
@@ -811,8 +818,9 @@ static int whiteheat_write_room(struct usb_serial_port *port)
 }
 
 
-static int whiteheat_tiocmget (struct usb_serial_port *port, struct file *file)
+static int whiteheat_tiocmget (struct tty_struct *tty, struct file *file)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
 	unsigned int modem_signals = 0;
 
@@ -828,9 +836,10 @@ static int whiteheat_tiocmget (struct usb_serial_port *port, struct file *file)
 }
 
 
-static int whiteheat_tiocmset (struct usb_serial_port *port, struct file *file,
+static int whiteheat_tiocmset (struct tty_struct *tty, struct file *file,
 			       unsigned int set, unsigned int clear)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
 
 	dbg("%s - port %d", __func__, port->number);
@@ -851,8 +860,9 @@ static int whiteheat_tiocmset (struct usb_serial_port *port, struct file *file,
 }
 
 
-static int whiteheat_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg)
+static int whiteheat_ioctl (struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct serial_struct serstruct;
 	void __user *user_arg = (void __user *)arg;
 
@@ -896,20 +906,21 @@ static int whiteheat_ioctl (struct usb_serial_port *port, struct file * file, un
 }
 
 
-static void whiteheat_set_termios(struct usb_serial_port *port, struct ktermios *old_termios)
+static void whiteheat_set_termios(struct tty_struct *tty, struct usb_serial_port *port, struct ktermios *old_termios)
 {
-	dbg("%s -port %d", __func__, port->number);
-	firm_setup_port(port);
+	firm_setup_port(tty);
 }
 
 
-static void whiteheat_break_ctl(struct usb_serial_port *port, int break_state) {
+static void whiteheat_break_ctl(struct tty_struct *tty, int break_state) {
+	struct usb_serial_port *port = tty->driver_data;
 	firm_set_break(port, break_state);
 }
 
 
-static int whiteheat_chars_in_buffer(struct usb_serial_port *port)
+static int whiteheat_chars_in_buffer(struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
 	struct list_head *tmp;
 	struct whiteheat_urb_wrap *wrap;
@@ -930,8 +941,9 @@ static int whiteheat_chars_in_buffer(struct usb_serial_port *port)
 }
 
 
-static void whiteheat_throttle (struct usb_serial_port *port)
+static void whiteheat_throttle (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
 	unsigned long flags;
 
@@ -945,8 +957,9 @@ static void whiteheat_throttle (struct usb_serial_port *port)
 }
 
 
-static void whiteheat_unthrottle (struct usb_serial_port *port)
+static void whiteheat_unthrottle (struct tty_struct *tty)
 {
+	struct usb_serial_port *port = tty->driver_data;
 	struct whiteheat_private *info = usb_get_serial_port_data(port);
 	int actually_throttled;
 	unsigned long flags;
@@ -1184,9 +1197,10 @@ static int firm_close(struct usb_serial_port *port) {
 }
 
 
-static int firm_setup_port(struct usb_serial_port *port) {
+static int firm_setup_port(struct tty_struct *tty) {
+	struct usb_serial_port *port = tty->driver_data;
 	struct whiteheat_port_settings port_settings;
-	unsigned int cflag = port->tty->termios->c_cflag;
+	unsigned int cflag = tty->termios->c_cflag;
 
 	port_settings.port = port->number + 1;
 
@@ -1235,22 +1249,22 @@ static int firm_setup_port(struct usb_serial_port *port) {
 	    (port_settings.hflow & WHITEHEAT_HFLOW_DTR) ? "DTR" : "");
 	
 	/* determine software flow control */
-	if (I_IXOFF(port->tty))
+	if (I_IXOFF(tty))
 		port_settings.sflow = WHITEHEAT_SFLOW_RXTX;
 	else
 		port_settings.sflow = WHITEHEAT_SFLOW_NONE;
 	dbg("%s - software flow control = %c", __func__, port_settings.sflow);
 	
-	port_settings.xon = START_CHAR(port->tty);
-	port_settings.xoff = STOP_CHAR(port->tty);
+	port_settings.xon = START_CHAR(tty);
+	port_settings.xoff = STOP_CHAR(tty);
 	dbg("%s - XON = %2x, XOFF = %2x", __func__, port_settings.xon, port_settings.xoff);
 
 	/* get the baud rate wanted */
-	port_settings.baud = tty_get_baud_rate(port->tty);
+	port_settings.baud = tty_get_baud_rate(tty);
 	dbg("%s - baud rate = %d", __func__, port_settings.baud);
 
 	/* fixme: should set validated settings */
-	tty_encode_baud_rate(port->tty, port_settings.baud, port_settings.baud);
+	tty_encode_baud_rate(tty, port_settings.baud, port_settings.baud);
 	/* handle any settings that aren't specified in the tty structure */
 	port_settings.lloop = 0;
 	
@@ -1426,7 +1440,7 @@ static void rx_data_softint(struct work_struct *work)
 	struct whiteheat_private *info =
 		container_of(work, struct whiteheat_private, rx_work);
 	struct usb_serial_port *port = info->port;
-	struct tty_struct *tty = port->tty;
+	struct tty_struct *tty = port->port.tty;
 	struct whiteheat_urb_wrap *wrap;
 	struct urb *urb;
 	unsigned long flags;
diff --git a/include/linux/usb/serial.h b/include/linux/usb/serial.h
index 8f891cbaf9ab..09a3e6a7518f 100644
--- a/include/linux/usb/serial.h
+++ b/include/linux/usb/serial.h
@@ -62,7 +62,7 @@
  */
 struct usb_serial_port {
 	struct usb_serial	*serial;
-	struct tty_struct	*tty;
+	struct tty_port		port;
 	spinlock_t		lock;
 	struct mutex            mutex;
 	unsigned char		number;
@@ -89,7 +89,6 @@ struct usb_serial_port {
 
 	wait_queue_head_t	write_wait;
 	struct work_struct	work;
-	int			open_count;
 	char			throttled;
 	char			throttle_req;
 	char			console;
@@ -217,22 +216,27 @@ struct usb_serial_driver {
 	int (*resume)(struct usb_serial *serial);
 
 	/* serial function calls */
-	int  (*open)(struct usb_serial_port *port, struct file *filp);
-	void (*close)(struct usb_serial_port *port, struct file *filp);
-	int  (*write)(struct usb_serial_port *port, const unsigned char *buf,
-		      int count);
-	int  (*write_room)(struct usb_serial_port *port);
-	int  (*ioctl)(struct usb_serial_port *port, struct file *file,
+	/* Called by console with tty = NULL and by tty */
+	int  (*open)(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+	void (*close)(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
+	int  (*write)(struct tty_struct *tty, struct usb_serial_port *port,
+			const unsigned char *buf, int count);
+	/* Called only by the tty layer */
+	int  (*write_room)(struct tty_struct *tty);
+	int  (*ioctl)(struct tty_struct *tty, struct file *file,
 		      unsigned int cmd, unsigned long arg);
-	void (*set_termios)(struct usb_serial_port *port, struct ktermios *old);
-	void (*break_ctl)(struct usb_serial_port *port, int break_state);
-	int  (*chars_in_buffer)(struct usb_serial_port *port);
-	void (*throttle)(struct usb_serial_port *port);
-	void (*unthrottle)(struct usb_serial_port *port);
-	int  (*tiocmget)(struct usb_serial_port *port, struct file *file);
-	int  (*tiocmset)(struct usb_serial_port *port, struct file *file,
+	void (*set_termios)(struct tty_struct *tty,
+			struct usb_serial_port *port, struct ktermios *old);
+	void (*break_ctl)(struct tty_struct *tty, int break_state);
+	int  (*chars_in_buffer)(struct tty_struct *tty);
+	void (*throttle)(struct tty_struct *tty);
+	void (*unthrottle)(struct tty_struct *tty);
+	int  (*tiocmget)(struct tty_struct *tty, struct file *file);
+	int  (*tiocmset)(struct tty_struct *tty, struct file *file,
 			 unsigned int set, unsigned int clear);
-
+	/* USB events */
 	void (*read_int_callback)(struct urb *urb);
 	void (*write_int_callback)(struct urb *urb);
 	void (*read_bulk_callback)(struct urb *urb);
@@ -270,19 +274,19 @@ static inline void usb_serial_console_disconnect(struct usb_serial *serial) {}
 /* Functions needed by other parts of the usbserial core */
 extern struct usb_serial *usb_serial_get_by_index(unsigned int minor);
 extern void usb_serial_put(struct usb_serial *serial);
-extern int usb_serial_generic_open(struct usb_serial_port *port,
-				   struct file *filp);
-extern int usb_serial_generic_write(struct usb_serial_port *port,
-				    const unsigned char *buf, int count);
-extern void usb_serial_generic_close(struct usb_serial_port *port,
-				     struct file *filp);
+extern int usb_serial_generic_open(struct tty_struct *tty,
+		struct usb_serial_port *port, struct file *filp);
+extern int usb_serial_generic_write(struct tty_struct *tty,
+	struct usb_serial_port *port, const unsigned char *buf, int count);
+extern void usb_serial_generic_close(struct tty_struct *tty,
+			struct usb_serial_port *port, struct file *filp);
 extern int usb_serial_generic_resume(struct usb_serial *serial);
-extern int usb_serial_generic_write_room(struct usb_serial_port *port);
-extern int usb_serial_generic_chars_in_buffer(struct usb_serial_port *port);
+extern int usb_serial_generic_write_room(struct tty_struct *tty);
+extern int usb_serial_generic_chars_in_buffer(struct tty_struct *tty);
 extern void usb_serial_generic_read_bulk_callback(struct urb *urb);
 extern void usb_serial_generic_write_bulk_callback(struct urb *urb);
-extern void usb_serial_generic_throttle(struct usb_serial_port *port);
-extern void usb_serial_generic_unthrottle(struct usb_serial_port *port);
+extern void usb_serial_generic_throttle(struct tty_struct *tty);
+extern void usb_serial_generic_unthrottle(struct tty_struct *tty);
 extern void usb_serial_generic_shutdown(struct usb_serial *serial);
 extern int usb_serial_generic_register(int debug);
 extern void usb_serial_generic_deregister(void);
-- 
cgit v1.2.3


From 01e1abb2c27e43339b8829a2e3b1c6f53806b77a Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Tue, 22 Jul 2008 11:16:55 +0100
Subject: tty: Split ldisc code into its own file

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/Makefile    |   2 +-
 drivers/char/tty_io.c    | 636 +----------------------------------------
 drivers/char/tty_ldisc.c | 714 +++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/tty.h      |  11 +-
 4 files changed, 733 insertions(+), 630 deletions(-)
 create mode 100644 drivers/char/tty_ldisc.c

(limited to 'include/linux')

diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index 0e0d12a06462..dc5a327d72d5 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -7,7 +7,7 @@
 #
 FONTMAPFILE = cp437.uni
 
-obj-y	 += mem.o random.o tty_io.o n_tty.o tty_ioctl.o
+obj-y	 += mem.o random.o tty_io.o n_tty.o tty_ioctl.o tty_ldisc.o 
 
 obj-$(CONFIG_LEGACY_PTYS)	+= pty.o
 obj-$(CONFIG_UNIX98_PTYS)	+= pty.o
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 82f6a8c86332..d27a08b374d0 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -655,558 +655,6 @@ EXPORT_SYMBOL_GPL(tty_prepare_flip_string_flags);
 
 
 
-/**
- *	tty_set_termios_ldisc		-	set ldisc field
- *	@tty: tty structure
- *	@num: line discipline number
- *
- *	This is probably overkill for real world processors but
- *	they are not on hot paths so a little discipline won't do
- *	any harm.
- *
- *	Locking: takes termios_mutex
- */
-
-static void tty_set_termios_ldisc(struct tty_struct *tty, int num)
-{
-	mutex_lock(&tty->termios_mutex);
-	tty->termios->c_line = num;
-	mutex_unlock(&tty->termios_mutex);
-}
-
-/*
- *	This guards the refcounted line discipline lists. The lock
- *	must be taken with irqs off because there are hangup path
- *	callers who will do ldisc lookups and cannot sleep.
- */
-
-static DEFINE_SPINLOCK(tty_ldisc_lock);
-static DECLARE_WAIT_QUEUE_HEAD(tty_ldisc_wait);
-/* Line disc dispatch table */
-static struct tty_ldisc_ops *tty_ldiscs[NR_LDISCS];
-
-/**
- *	tty_register_ldisc	-	install a line discipline
- *	@disc: ldisc number
- *	@new_ldisc: pointer to the ldisc object
- *
- *	Installs a new line discipline into the kernel. The discipline
- *	is set up as unreferenced and then made available to the kernel
- *	from this point onwards.
- *
- *	Locking:
- *		takes tty_ldisc_lock to guard against ldisc races
- */
-
-int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	if (disc < N_TTY || disc >= NR_LDISCS)
-		return -EINVAL;
-
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	tty_ldiscs[disc] = new_ldisc;
-	new_ldisc->num = disc;
-	new_ldisc->refcount = 0;
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-
-	return ret;
-}
-EXPORT_SYMBOL(tty_register_ldisc);
-
-/**
- *	tty_unregister_ldisc	-	unload a line discipline
- *	@disc: ldisc number
- *	@new_ldisc: pointer to the ldisc object
- *
- *	Remove a line discipline from the kernel providing it is not
- *	currently in use.
- *
- *	Locking:
- *		takes tty_ldisc_lock to guard against ldisc races
- */
-
-int tty_unregister_ldisc(int disc)
-{
-	unsigned long flags;
-	int ret = 0;
-
-	if (disc < N_TTY || disc >= NR_LDISCS)
-		return -EINVAL;
-
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	if (tty_ldiscs[disc]->refcount)
-		ret = -EBUSY;
-	else
-		tty_ldiscs[disc] = NULL;
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-
-	return ret;
-}
-EXPORT_SYMBOL(tty_unregister_ldisc);
-
-
-/**
- *	tty_ldisc_try_get	-	try and reference an ldisc
- *	@disc: ldisc number
- *	@ld: tty ldisc structure to complete
- *
- *	Attempt to open and lock a line discipline into place. Return
- *	the line discipline refcounted and assigned in ld. On an error
- *	report the error code back
- */
-
-static int tty_ldisc_try_get(int disc, struct tty_ldisc *ld)
-{
-	unsigned long flags;
-	struct tty_ldisc_ops *ldops;
-	int err = -EINVAL;
-	
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	ld->ops = NULL;
-	ldops = tty_ldiscs[disc];
-	/* Check the entry is defined */
-	if (ldops) {
-		/* If the module is being unloaded we can't use it */
-		if (!try_module_get(ldops->owner))
-			err = -EAGAIN;
-		else {
-			/* lock it */
-			ldops->refcount++;
-			ld->ops = ldops;
-			err = 0;
-		}
-	}
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-	return err;
-}
-
-/**
- *	tty_ldisc_get		-	take a reference to an ldisc
- *	@disc: ldisc number
- *	@ld: tty line discipline structure to use
- *
- *	Takes a reference to a line discipline. Deals with refcounts and
- *	module locking counts. Returns NULL if the discipline is not available.
- *	Returns a pointer to the discipline and bumps the ref count if it is
- *	available
- *
- *	Locking:
- *		takes tty_ldisc_lock to guard against ldisc races
- */
-
-static int tty_ldisc_get(int disc, struct tty_ldisc *ld)
-{
-	int err;
-
-	if (disc < N_TTY || disc >= NR_LDISCS)
-		return -EINVAL;
-	err = tty_ldisc_try_get(disc, ld);
-	if (err == -EAGAIN) {
-		request_module("tty-ldisc-%d", disc);
-		err = tty_ldisc_try_get(disc, ld);
-	}
-	return err;
-}
-
-/**
- *	tty_ldisc_put		-	drop ldisc reference
- *	@disc: ldisc number
- *
- *	Drop a reference to a line discipline. Manage refcounts and
- *	module usage counts
- *
- *	Locking:
- *		takes tty_ldisc_lock to guard against ldisc races
- */
-
-static void tty_ldisc_put(struct tty_ldisc_ops *ld)
-{
-	unsigned long flags;
-	int disc = ld->num;
-
-	BUG_ON(disc < N_TTY || disc >= NR_LDISCS);
-
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	ld = tty_ldiscs[disc];
-	BUG_ON(ld->refcount == 0);
-	ld->refcount--;
-	module_put(ld->owner);
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-}
-
-static void * tty_ldiscs_seq_start(struct seq_file *m, loff_t *pos)
-{
-	return (*pos < NR_LDISCS) ? pos : NULL;
-}
-
-static void * tty_ldiscs_seq_next(struct seq_file *m, void *v, loff_t *pos)
-{
-	(*pos)++;
-	return (*pos < NR_LDISCS) ? pos : NULL;
-}
-
-static void tty_ldiscs_seq_stop(struct seq_file *m, void *v)
-{
-}
-
-static int tty_ldiscs_seq_show(struct seq_file *m, void *v)
-{
-	int i = *(loff_t *)v;
-	struct tty_ldisc ld;
-	
-	if (tty_ldisc_get(i, &ld) < 0)
-		return 0;
-	seq_printf(m, "%-10s %2d\n", ld.ops->name ? ld.ops->name : "???", i);
-	tty_ldisc_put(ld.ops);
-	return 0;
-}
-
-static const struct seq_operations tty_ldiscs_seq_ops = {
-	.start	= tty_ldiscs_seq_start,
-	.next	= tty_ldiscs_seq_next,
-	.stop	= tty_ldiscs_seq_stop,
-	.show	= tty_ldiscs_seq_show,
-};
-
-static int proc_tty_ldiscs_open(struct inode *inode, struct file *file)
-{
-	return seq_open(file, &tty_ldiscs_seq_ops);
-}
-
-const struct file_operations tty_ldiscs_proc_fops = {
-	.owner		= THIS_MODULE,
-	.open		= proc_tty_ldiscs_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= seq_release,
-};
-
-/**
- *	tty_ldisc_assign	-	set ldisc on a tty
- *	@tty: tty to assign
- *	@ld: line discipline
- *
- *	Install an instance of a line discipline into a tty structure. The
- *	ldisc must have a reference count above zero to ensure it remains/
- *	The tty instance refcount starts at zero.
- *
- *	Locking:
- *		Caller must hold references
- */
-
-static void tty_ldisc_assign(struct tty_struct *tty, struct tty_ldisc *ld)
-{
-	ld->refcount = 0;
-	tty->ldisc = *ld;
-}
-
-/**
- *	tty_ldisc_try		-	internal helper
- *	@tty: the tty
- *
- *	Make a single attempt to grab and bump the refcount on
- *	the tty ldisc. Return 0 on failure or 1 on success. This is
- *	used to implement both the waiting and non waiting versions
- *	of tty_ldisc_ref
- *
- *	Locking: takes tty_ldisc_lock
- */
-
-static int tty_ldisc_try(struct tty_struct *tty)
-{
-	unsigned long flags;
-	struct tty_ldisc *ld;
-	int ret = 0;
-
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	ld = &tty->ldisc;
-	if (test_bit(TTY_LDISC, &tty->flags)) {
-		ld->refcount++;
-		ret = 1;
-	}
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-	return ret;
-}
-
-/**
- *	tty_ldisc_ref_wait	-	wait for the tty ldisc
- *	@tty: tty device
- *
- *	Dereference the line discipline for the terminal and take a
- *	reference to it. If the line discipline is in flux then
- *	wait patiently until it changes.
- *
- *	Note: Must not be called from an IRQ/timer context. The caller
- *	must also be careful not to hold other locks that will deadlock
- *	against a discipline change, such as an existing ldisc reference
- *	(which we check for)
- *
- *	Locking: call functions take tty_ldisc_lock
- */
-
-struct tty_ldisc *tty_ldisc_ref_wait(struct tty_struct *tty)
-{
-	/* wait_event is a macro */
-	wait_event(tty_ldisc_wait, tty_ldisc_try(tty));
-	if (tty->ldisc.refcount == 0)
-		printk(KERN_ERR "tty_ldisc_ref_wait\n");
-	return &tty->ldisc;
-}
-
-EXPORT_SYMBOL_GPL(tty_ldisc_ref_wait);
-
-/**
- *	tty_ldisc_ref		-	get the tty ldisc
- *	@tty: tty device
- *
- *	Dereference the line discipline for the terminal and take a
- *	reference to it. If the line discipline is in flux then
- *	return NULL. Can be called from IRQ and timer functions.
- *
- *	Locking: called functions take tty_ldisc_lock
- */
-
-struct tty_ldisc *tty_ldisc_ref(struct tty_struct *tty)
-{
-	if (tty_ldisc_try(tty))
-		return &tty->ldisc;
-	return NULL;
-}
-
-EXPORT_SYMBOL_GPL(tty_ldisc_ref);
-
-/**
- *	tty_ldisc_deref		-	free a tty ldisc reference
- *	@ld: reference to free up
- *
- *	Undoes the effect of tty_ldisc_ref or tty_ldisc_ref_wait. May
- *	be called in IRQ context.
- *
- *	Locking: takes tty_ldisc_lock
- */
-
-void tty_ldisc_deref(struct tty_ldisc *ld)
-{
-	unsigned long flags;
-
-	BUG_ON(ld == NULL);
-
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	if (ld->refcount == 0)
-		printk(KERN_ERR "tty_ldisc_deref: no references.\n");
-	else
-		ld->refcount--;
-	if (ld->refcount == 0)
-		wake_up(&tty_ldisc_wait);
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-}
-
-EXPORT_SYMBOL_GPL(tty_ldisc_deref);
-
-/**
- *	tty_ldisc_enable	-	allow ldisc use
- *	@tty: terminal to activate ldisc on
- *
- *	Set the TTY_LDISC flag when the line discipline can be called
- *	again. Do necessary wakeups for existing sleepers.
- *
- *	Note: nobody should set this bit except via this function. Clearing
- *	directly is allowed.
- */
-
-static void tty_ldisc_enable(struct tty_struct *tty)
-{
-	set_bit(TTY_LDISC, &tty->flags);
-	wake_up(&tty_ldisc_wait);
-}
-
-/**
- *	tty_ldisc_restore	-	helper for tty ldisc change
- *	@tty: tty to recover
- *	@old: previous ldisc
- *
- *	Restore the previous line discipline or N_TTY when a line discipline
- *	change fails due to an open error
- */
-
-static void tty_ldisc_restore(struct tty_struct *tty, struct tty_ldisc *old)
-{
-	char buf[64];
-	struct tty_ldisc new_ldisc;
-
-	/* There is an outstanding reference here so this is safe */
-	tty_ldisc_get(old->ops->num, old);
-	tty_ldisc_assign(tty, old);
-	tty_set_termios_ldisc(tty, old->ops->num);
-	if (old->ops->open && (old->ops->open(tty) < 0)) {
-		tty_ldisc_put(old->ops);
-		/* This driver is always present */
-		if (tty_ldisc_get(N_TTY, &new_ldisc) < 0)
-			panic("n_tty: get");
-		tty_ldisc_assign(tty, &new_ldisc);
-		tty_set_termios_ldisc(tty, N_TTY);
-		if (new_ldisc.ops->open) {
-			int r = new_ldisc.ops->open(tty);
-				if (r < 0)
-				panic("Couldn't open N_TTY ldisc for "
-				      "%s --- error %d.",
-				      tty_name(tty, buf), r);
-		}
-	}
-}
-
-/**
- *	tty_set_ldisc		-	set line discipline
- *	@tty: the terminal to set
- *	@ldisc: the line discipline
- *
- *	Set the discipline of a tty line. Must be called from a process
- *	context.
- *
- *	Locking: takes tty_ldisc_lock.
- *		 called functions take termios_mutex
- */
-
-static int tty_set_ldisc(struct tty_struct *tty, int ldisc)
-{
-	int retval;
-	struct tty_ldisc o_ldisc, new_ldisc;
-	int work;
-	unsigned long flags;
-	struct tty_struct *o_tty;
-
-restart:
-	/* This is a bit ugly for now but means we can break the 'ldisc
-	   is part of the tty struct' assumption later */
-	retval = tty_ldisc_get(ldisc, &new_ldisc);
-	if (retval)
-		return retval;
-
-	/*
-	 *	Problem: What do we do if this blocks ?
-	 */
-
-	tty_wait_until_sent(tty, 0);
-
-	if (tty->ldisc.ops->num == ldisc) {
-		tty_ldisc_put(new_ldisc.ops);
-		return 0;
-	}
-
-	/*
-	 *	No more input please, we are switching. The new ldisc
-	 *	will update this value in the ldisc open function
-	 */
-
-	tty->receive_room = 0;
-
-	o_ldisc = tty->ldisc;
-	o_tty = tty->link;
-
-	/*
-	 *	Make sure we don't change while someone holds a
-	 *	reference to the line discipline. The TTY_LDISC bit
-	 *	prevents anyone taking a reference once it is clear.
-	 *	We need the lock to avoid racing reference takers.
-	 */
-
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	if (tty->ldisc.refcount || (o_tty && o_tty->ldisc.refcount)) {
-		if (tty->ldisc.refcount) {
-			/* Free the new ldisc we grabbed. Must drop the lock
-			   first. */
-			spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-			tty_ldisc_put(o_ldisc.ops);
-			/*
-			 * There are several reasons we may be busy, including
-			 * random momentary I/O traffic. We must therefore
-			 * retry. We could distinguish between blocking ops
-			 * and retries if we made tty_ldisc_wait() smarter.
-			 * That is up for discussion.
-			 */
-			if (wait_event_interruptible(tty_ldisc_wait, tty->ldisc.refcount == 0) < 0)
-				return -ERESTARTSYS;
-			goto restart;
-		}
-		if (o_tty && o_tty->ldisc.refcount) {
-			spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-			tty_ldisc_put(o_tty->ldisc.ops);
-			if (wait_event_interruptible(tty_ldisc_wait, o_tty->ldisc.refcount == 0) < 0)
-				return -ERESTARTSYS;
-			goto restart;
-		}
-	}
-	/*
-	 *	If the TTY_LDISC bit is set, then we are racing against
-	 *	another ldisc change
-	 */
-	if (!test_bit(TTY_LDISC, &tty->flags)) {
-		struct tty_ldisc *ld;
-		spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-		tty_ldisc_put(new_ldisc.ops);
-		ld = tty_ldisc_ref_wait(tty);
-		tty_ldisc_deref(ld);
-		goto restart;
-	}
-
-	clear_bit(TTY_LDISC, &tty->flags);
-	if (o_tty)
-		clear_bit(TTY_LDISC, &o_tty->flags);
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-	
-	/*
-	 *	From this point on we know nobody has an ldisc
-	 *	usage reference, nor can they obtain one until
-	 *	we say so later on.
-	 */
-
-	work = cancel_delayed_work(&tty->buf.work);
-	/*
-	 * Wait for ->hangup_work and ->buf.work handlers to terminate
-	 * MUST NOT hold locks here.
-	 */
-	flush_scheduled_work();
-	/* Shutdown the current discipline. */
-	if (o_ldisc.ops->close)
-		(o_ldisc.ops->close)(tty);
-
-	/* Now set up the new line discipline. */
-	tty_ldisc_assign(tty, &new_ldisc);
-	tty_set_termios_ldisc(tty, ldisc);
-	if (new_ldisc.ops->open)
-		retval = (new_ldisc.ops->open)(tty);
-	if (retval < 0) {
-		tty_ldisc_put(new_ldisc.ops);
-		tty_ldisc_restore(tty, &o_ldisc);
-	}
-	/* At this point we hold a reference to the new ldisc and a
-	   a reference to the old ldisc. If we ended up flipping back
-	   to the existing ldisc we have two references to it */
-
-	if (tty->ldisc.ops->num != o_ldisc.ops->num && tty->ops->set_ldisc)
-		tty->ops->set_ldisc(tty);
-
-	tty_ldisc_put(o_ldisc.ops);
-
-	/*
-	 *	Allow ldisc referencing to occur as soon as the driver
-	 *	ldisc callback completes.
-	 */
-
-	tty_ldisc_enable(tty);
-	if (o_tty)
-		tty_ldisc_enable(o_tty);
-
-	/* Restart it in case no characters kick it off. Safe if
-	   already running */
-	if (work)
-		schedule_delayed_work(&tty->buf.work, 1);
-	return retval;
-}
-
 /**
  *	get_tty_driver		-	find device of a tty
  *	@dev_t: device identifier
@@ -2193,7 +1641,6 @@ static int init_dev(struct tty_driver *driver, int idx,
 	struct ktermios *tp, **tp_loc, *o_tp, **o_tp_loc;
 	struct ktermios *ltp, **ltp_loc, *o_ltp, **o_ltp_loc;
 	int retval = 0;
-	struct tty_ldisc *ld;
 
 	/* check whether we're reopening an existing tty */
 	if (driver->flags & TTY_DRIVER_DEVPTS_MEM) {
@@ -2342,25 +1789,12 @@ static int init_dev(struct tty_driver *driver, int idx,
 	 * If we fail here just call release_tty to clean up.  No need
 	 * to decrement the use counts, as release_tty doesn't care.
 	 */
-	 
-	ld = &tty->ldisc;
 
-	if (ld->ops->open) {
-		retval = (ld->ops->open)(tty);
-		if (retval)
-			goto release_mem_out;
-	}
-	if (o_tty && o_tty->ldisc.ops->open) {
-		retval = (o_tty->ldisc.ops->open)(o_tty);
-		if (retval) {
-			if (ld->ops->close)
-				(ld->ops->close)(tty);
-			goto release_mem_out;
-		}
-		tty_ldisc_enable(o_tty);
-	}
-	tty_ldisc_enable(tty);
-	goto success;
+	retval = tty_ldisc_setup(tty, o_tty);
+
+	if (retval)
+		goto release_mem_out;
+	 goto success;
 
 	/*
 	 * This fast open can be used if the tty is already open.
@@ -2498,12 +1932,10 @@ static void release_tty(struct tty_struct *tty, int idx)
 static void release_dev(struct file *filp)
 {
 	struct tty_struct *tty, *o_tty;
-	struct tty_ldisc ld;
 	int	pty_master, tty_closing, o_tty_closing, do_sleep;
 	int	devpts;
 	int	idx;
 	char	buf[64];
-	unsigned long flags;
 
 	tty = (struct tty_struct *)filp->private_data;
 	if (tty_paranoia_check(tty, filp->f_path.dentry->d_inode,
@@ -2705,56 +2137,9 @@ static void release_dev(struct file *filp)
 	printk(KERN_DEBUG "freeing tty structure...");
 #endif
 	/*
-	 * Prevent flush_to_ldisc() from rescheduling the work for later.  Then
-	 * kill any delayed work. As this is the final close it does not
-	 * race with the set_ldisc code path.
-	 */
-	clear_bit(TTY_LDISC, &tty->flags);
-	cancel_delayed_work(&tty->buf.work);
-
-	/*
-	 * Wait for ->hangup_work and ->buf.work handlers to terminate
-	 */
-
-	flush_scheduled_work();
-
-	/*
-	 * Wait for any short term users (we know they are just driver
-	 * side waiters as the file is closing so user count on the file
-	 * side is zero.
+	 * Ask the line discipline code to release its structures
 	 */
-	spin_lock_irqsave(&tty_ldisc_lock, flags);
-	while (tty->ldisc.refcount) {
-		spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-		wait_event(tty_ldisc_wait, tty->ldisc.refcount == 0);
-		spin_lock_irqsave(&tty_ldisc_lock, flags);
-	}
-	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
-	/*
-	 * Shutdown the current line discipline, and reset it to N_TTY.
-	 *
-	 * FIXME: this MUST get fixed for the new reflocking
-	 */
-	if (tty->ldisc.ops->close)
-		(tty->ldisc.ops->close)(tty);
-	tty_ldisc_put(tty->ldisc.ops);
-
-	/*
-	 *	Switch the line discipline back
-	 */
-	WARN_ON(tty_ldisc_get(N_TTY, &ld));
-	tty_ldisc_assign(tty, &ld);
-	tty_set_termios_ldisc(tty, N_TTY);
-	if (o_tty) {
-		/* FIXME: could o_tty be in setldisc here ? */
-		clear_bit(TTY_LDISC, &o_tty->flags);
-		if (o_tty->ldisc.ops->close)
-			(o_tty->ldisc.ops->close)(o_tty);
-		tty_ldisc_put(o_tty->ldisc.ops);
-		WARN_ON(tty_ldisc_get(N_TTY, &ld));
-		tty_ldisc_assign(o_tty, &ld);
-		tty_set_termios_ldisc(o_tty, N_TTY);
-	}
+	tty_ldisc_release(tty, o_tty);
 	/*
 	 * The release_tty function takes care of the details of clearing
 	 * the slots and preserving the termios structure.
@@ -3962,12 +3347,9 @@ EXPORT_SYMBOL(tty_flip_buffer_push);
 
 static void initialize_tty_struct(struct tty_struct *tty)
 {
-	struct tty_ldisc ld;
 	memset(tty, 0, sizeof(struct tty_struct));
 	tty->magic = TTY_MAGIC;
-	if (tty_ldisc_get(N_TTY, &ld) < 0)
-		panic("n_tty: init_tty");
-	tty_ldisc_assign(tty, &ld);
+	tty_ldisc_init(tty);
 	tty->session = NULL;
 	tty->pgrp = NULL;
 	tty->overrun_time = jiffies;
@@ -4280,7 +3662,7 @@ void __init console_init(void)
 	initcall_t *call;
 
 	/* Setup the default TTY line discipline. */
-	(void) tty_register_ldisc(N_TTY, &tty_ldisc_N_TTY);
+	tty_ldisc_begin();
 
 	/*
 	 * set up the console device so that later boot sequences can
diff --git a/drivers/char/tty_ldisc.c b/drivers/char/tty_ldisc.c
new file mode 100644
index 000000000000..241cbdea65ab
--- /dev/null
+++ b/drivers/char/tty_ldisc.c
@@ -0,0 +1,714 @@
+#include <linux/types.h>
+#include <linux/major.h>
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/fcntl.h>
+#include <linux/sched.h>
+#include <linux/interrupt.h>
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+#include <linux/tty_flip.h>
+#include <linux/devpts_fs.h>
+#include <linux/file.h>
+#include <linux/fdtable.h>
+#include <linux/console.h>
+#include <linux/timer.h>
+#include <linux/ctype.h>
+#include <linux/kd.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/poll.h>
+#include <linux/proc_fs.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/smp_lock.h>
+#include <linux/device.h>
+#include <linux/wait.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/seq_file.h>
+
+#include <linux/uaccess.h>
+#include <asm/system.h>
+
+#include <linux/kbd_kern.h>
+#include <linux/vt_kern.h>
+#include <linux/selection.h>
+
+#include <linux/kmod.h>
+#include <linux/nsproxy.h>
+
+/*
+ *	This guards the refcounted line discipline lists. The lock
+ *	must be taken with irqs off because there are hangup path
+ *	callers who will do ldisc lookups and cannot sleep.
+ */
+
+static DEFINE_SPINLOCK(tty_ldisc_lock);
+static DECLARE_WAIT_QUEUE_HEAD(tty_ldisc_wait);
+/* Line disc dispatch table */
+static struct tty_ldisc_ops *tty_ldiscs[NR_LDISCS];
+
+/**
+ *	tty_register_ldisc	-	install a line discipline
+ *	@disc: ldisc number
+ *	@new_ldisc: pointer to the ldisc object
+ *
+ *	Installs a new line discipline into the kernel. The discipline
+ *	is set up as unreferenced and then made available to the kernel
+ *	from this point onwards.
+ *
+ *	Locking:
+ *		takes tty_ldisc_lock to guard against ldisc races
+ */
+
+int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	if (disc < N_TTY || disc >= NR_LDISCS)
+		return -EINVAL;
+
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	tty_ldiscs[disc] = new_ldisc;
+	new_ldisc->num = disc;
+	new_ldisc->refcount = 0;
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(tty_register_ldisc);
+
+/**
+ *	tty_unregister_ldisc	-	unload a line discipline
+ *	@disc: ldisc number
+ *	@new_ldisc: pointer to the ldisc object
+ *
+ *	Remove a line discipline from the kernel providing it is not
+ *	currently in use.
+ *
+ *	Locking:
+ *		takes tty_ldisc_lock to guard against ldisc races
+ */
+
+int tty_unregister_ldisc(int disc)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	if (disc < N_TTY || disc >= NR_LDISCS)
+		return -EINVAL;
+
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	if (tty_ldiscs[disc]->refcount)
+		ret = -EBUSY;
+	else
+		tty_ldiscs[disc] = NULL;
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(tty_unregister_ldisc);
+
+
+/**
+ *	tty_ldisc_try_get	-	try and reference an ldisc
+ *	@disc: ldisc number
+ *	@ld: tty ldisc structure to complete
+ *
+ *	Attempt to open and lock a line discipline into place. Return
+ *	the line discipline refcounted and assigned in ld. On an error
+ *	report the error code back
+ */
+
+static int tty_ldisc_try_get(int disc, struct tty_ldisc *ld)
+{
+	unsigned long flags;
+	struct tty_ldisc_ops *ldops;
+	int err = -EINVAL;
+	
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	ld->ops = NULL;
+	ldops = tty_ldiscs[disc];
+	/* Check the entry is defined */
+	if (ldops) {
+		/* If the module is being unloaded we can't use it */
+		if (!try_module_get(ldops->owner))
+			err = -EAGAIN;
+		else {
+			/* lock it */
+			ldops->refcount++;
+			ld->ops = ldops;
+			err = 0;
+		}
+	}
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+	return err;
+}
+
+/**
+ *	tty_ldisc_get		-	take a reference to an ldisc
+ *	@disc: ldisc number
+ *	@ld: tty line discipline structure to use
+ *
+ *	Takes a reference to a line discipline. Deals with refcounts and
+ *	module locking counts. Returns NULL if the discipline is not available.
+ *	Returns a pointer to the discipline and bumps the ref count if it is
+ *	available
+ *
+ *	Locking:
+ *		takes tty_ldisc_lock to guard against ldisc races
+ */
+
+static int tty_ldisc_get(int disc, struct tty_ldisc *ld)
+{
+	int err;
+
+	if (disc < N_TTY || disc >= NR_LDISCS)
+		return -EINVAL;
+	err = tty_ldisc_try_get(disc, ld);
+	if (err == -EAGAIN) {
+		request_module("tty-ldisc-%d", disc);
+		err = tty_ldisc_try_get(disc, ld);
+	}
+	return err;
+}
+
+/**
+ *	tty_ldisc_put		-	drop ldisc reference
+ *	@disc: ldisc number
+ *
+ *	Drop a reference to a line discipline. Manage refcounts and
+ *	module usage counts
+ *
+ *	Locking:
+ *		takes tty_ldisc_lock to guard against ldisc races
+ */
+
+static void tty_ldisc_put(struct tty_ldisc_ops *ld)
+{
+	unsigned long flags;
+	int disc = ld->num;
+
+	BUG_ON(disc < N_TTY || disc >= NR_LDISCS);
+
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	ld = tty_ldiscs[disc];
+	BUG_ON(ld->refcount == 0);
+	ld->refcount--;
+	module_put(ld->owner);
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+}
+
+static void * tty_ldiscs_seq_start(struct seq_file *m, loff_t *pos)
+{
+	return (*pos < NR_LDISCS) ? pos : NULL;
+}
+
+static void * tty_ldiscs_seq_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	(*pos)++;
+	return (*pos < NR_LDISCS) ? pos : NULL;
+}
+
+static void tty_ldiscs_seq_stop(struct seq_file *m, void *v)
+{
+}
+
+static int tty_ldiscs_seq_show(struct seq_file *m, void *v)
+{
+	int i = *(loff_t *)v;
+	struct tty_ldisc ld;
+	
+	if (tty_ldisc_get(i, &ld) < 0)
+		return 0;
+	seq_printf(m, "%-10s %2d\n", ld.ops->name ? ld.ops->name : "???", i);
+	tty_ldisc_put(ld.ops);
+	return 0;
+}
+
+static const struct seq_operations tty_ldiscs_seq_ops = {
+	.start	= tty_ldiscs_seq_start,
+	.next	= tty_ldiscs_seq_next,
+	.stop	= tty_ldiscs_seq_stop,
+	.show	= tty_ldiscs_seq_show,
+};
+
+static int proc_tty_ldiscs_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &tty_ldiscs_seq_ops);
+}
+
+const struct file_operations tty_ldiscs_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= proc_tty_ldiscs_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+/**
+ *	tty_ldisc_assign	-	set ldisc on a tty
+ *	@tty: tty to assign
+ *	@ld: line discipline
+ *
+ *	Install an instance of a line discipline into a tty structure. The
+ *	ldisc must have a reference count above zero to ensure it remains/
+ *	The tty instance refcount starts at zero.
+ *
+ *	Locking:
+ *		Caller must hold references
+ */
+
+static void tty_ldisc_assign(struct tty_struct *tty, struct tty_ldisc *ld)
+{
+	ld->refcount = 0;
+	tty->ldisc = *ld;
+}
+
+/**
+ *	tty_ldisc_try		-	internal helper
+ *	@tty: the tty
+ *
+ *	Make a single attempt to grab and bump the refcount on
+ *	the tty ldisc. Return 0 on failure or 1 on success. This is
+ *	used to implement both the waiting and non waiting versions
+ *	of tty_ldisc_ref
+ *
+ *	Locking: takes tty_ldisc_lock
+ */
+
+static int tty_ldisc_try(struct tty_struct *tty)
+{
+	unsigned long flags;
+	struct tty_ldisc *ld;
+	int ret = 0;
+
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	ld = &tty->ldisc;
+	if (test_bit(TTY_LDISC, &tty->flags)) {
+		ld->refcount++;
+		ret = 1;
+	}
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+	return ret;
+}
+
+/**
+ *	tty_ldisc_ref_wait	-	wait for the tty ldisc
+ *	@tty: tty device
+ *
+ *	Dereference the line discipline for the terminal and take a
+ *	reference to it. If the line discipline is in flux then
+ *	wait patiently until it changes.
+ *
+ *	Note: Must not be called from an IRQ/timer context. The caller
+ *	must also be careful not to hold other locks that will deadlock
+ *	against a discipline change, such as an existing ldisc reference
+ *	(which we check for)
+ *
+ *	Locking: call functions take tty_ldisc_lock
+ */
+
+struct tty_ldisc *tty_ldisc_ref_wait(struct tty_struct *tty)
+{
+	/* wait_event is a macro */
+	wait_event(tty_ldisc_wait, tty_ldisc_try(tty));
+	if (tty->ldisc.refcount == 0)
+		printk(KERN_ERR "tty_ldisc_ref_wait\n");
+	return &tty->ldisc;
+}
+
+EXPORT_SYMBOL_GPL(tty_ldisc_ref_wait);
+
+/**
+ *	tty_ldisc_ref		-	get the tty ldisc
+ *	@tty: tty device
+ *
+ *	Dereference the line discipline for the terminal and take a
+ *	reference to it. If the line discipline is in flux then
+ *	return NULL. Can be called from IRQ and timer functions.
+ *
+ *	Locking: called functions take tty_ldisc_lock
+ */
+
+struct tty_ldisc *tty_ldisc_ref(struct tty_struct *tty)
+{
+	if (tty_ldisc_try(tty))
+		return &tty->ldisc;
+	return NULL;
+}
+
+EXPORT_SYMBOL_GPL(tty_ldisc_ref);
+
+/**
+ *	tty_ldisc_deref		-	free a tty ldisc reference
+ *	@ld: reference to free up
+ *
+ *	Undoes the effect of tty_ldisc_ref or tty_ldisc_ref_wait. May
+ *	be called in IRQ context.
+ *
+ *	Locking: takes tty_ldisc_lock
+ */
+
+void tty_ldisc_deref(struct tty_ldisc *ld)
+{
+	unsigned long flags;
+
+	BUG_ON(ld == NULL);
+
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	if (ld->refcount == 0)
+		printk(KERN_ERR "tty_ldisc_deref: no references.\n");
+	else
+		ld->refcount--;
+	if (ld->refcount == 0)
+		wake_up(&tty_ldisc_wait);
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+}
+
+EXPORT_SYMBOL_GPL(tty_ldisc_deref);
+
+/**
+ *	tty_ldisc_enable	-	allow ldisc use
+ *	@tty: terminal to activate ldisc on
+ *
+ *	Set the TTY_LDISC flag when the line discipline can be called
+ *	again. Do necessary wakeups for existing sleepers.
+ *
+ *	Note: nobody should set this bit except via this function. Clearing
+ *	directly is allowed.
+ */
+
+void tty_ldisc_enable(struct tty_struct *tty)
+{
+	set_bit(TTY_LDISC, &tty->flags);
+	wake_up(&tty_ldisc_wait);
+}
+
+/**
+ *	tty_set_termios_ldisc		-	set ldisc field
+ *	@tty: tty structure
+ *	@num: line discipline number
+ *
+ *	This is probably overkill for real world processors but
+ *	they are not on hot paths so a little discipline won't do
+ *	any harm.
+ *
+ *	Locking: takes termios_mutex
+ */
+
+static void tty_set_termios_ldisc(struct tty_struct *tty, int num)
+{
+	mutex_lock(&tty->termios_mutex);
+	tty->termios->c_line = num;
+	mutex_unlock(&tty->termios_mutex);
+}
+
+
+/**
+ *	tty_ldisc_restore	-	helper for tty ldisc change
+ *	@tty: tty to recover
+ *	@old: previous ldisc
+ *
+ *	Restore the previous line discipline or N_TTY when a line discipline
+ *	change fails due to an open error
+ */
+
+static void tty_ldisc_restore(struct tty_struct *tty, struct tty_ldisc *old)
+{
+	char buf[64];
+	struct tty_ldisc new_ldisc;
+
+	/* There is an outstanding reference here so this is safe */
+	tty_ldisc_get(old->ops->num, old);
+	tty_ldisc_assign(tty, old);
+	tty_set_termios_ldisc(tty, old->ops->num);
+	if (old->ops->open && (old->ops->open(tty) < 0)) {
+		tty_ldisc_put(old->ops);
+		/* This driver is always present */
+		if (tty_ldisc_get(N_TTY, &new_ldisc) < 0)
+			panic("n_tty: get");
+		tty_ldisc_assign(tty, &new_ldisc);
+		tty_set_termios_ldisc(tty, N_TTY);
+		if (new_ldisc.ops->open) {
+			int r = new_ldisc.ops->open(tty);
+				if (r < 0)
+				panic("Couldn't open N_TTY ldisc for "
+				      "%s --- error %d.",
+				      tty_name(tty, buf), r);
+		}
+	}
+}
+
+/**
+ *	tty_set_ldisc		-	set line discipline
+ *	@tty: the terminal to set
+ *	@ldisc: the line discipline
+ *
+ *	Set the discipline of a tty line. Must be called from a process
+ *	context.
+ *
+ *	Locking: takes tty_ldisc_lock.
+ *		 called functions take termios_mutex
+ */
+
+int tty_set_ldisc(struct tty_struct *tty, int ldisc)
+{
+	int retval;
+	struct tty_ldisc o_ldisc, new_ldisc;
+	int work;
+	unsigned long flags;
+	struct tty_struct *o_tty;
+
+restart:
+	/* This is a bit ugly for now but means we can break the 'ldisc
+	   is part of the tty struct' assumption later */
+	retval = tty_ldisc_get(ldisc, &new_ldisc);
+	if (retval)
+		return retval;
+
+	/*
+	 *	Problem: What do we do if this blocks ?
+	 */
+
+	tty_wait_until_sent(tty, 0);
+
+	if (tty->ldisc.ops->num == ldisc) {
+		tty_ldisc_put(new_ldisc.ops);
+		return 0;
+	}
+
+	/*
+	 *	No more input please, we are switching. The new ldisc
+	 *	will update this value in the ldisc open function
+	 */
+
+	tty->receive_room = 0;
+
+	o_ldisc = tty->ldisc;
+	o_tty = tty->link;
+
+	/*
+	 *	Make sure we don't change while someone holds a
+	 *	reference to the line discipline. The TTY_LDISC bit
+	 *	prevents anyone taking a reference once it is clear.
+	 *	We need the lock to avoid racing reference takers.
+	 */
+
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	if (tty->ldisc.refcount || (o_tty && o_tty->ldisc.refcount)) {
+		if (tty->ldisc.refcount) {
+			/* Free the new ldisc we grabbed. Must drop the lock
+			   first. */
+			spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+			tty_ldisc_put(o_ldisc.ops);
+			/*
+			 * There are several reasons we may be busy, including
+			 * random momentary I/O traffic. We must therefore
+			 * retry. We could distinguish between blocking ops
+			 * and retries if we made tty_ldisc_wait() smarter.
+			 * That is up for discussion.
+			 */
+			if (wait_event_interruptible(tty_ldisc_wait, tty->ldisc.refcount == 0) < 0)
+				return -ERESTARTSYS;
+			goto restart;
+		}
+		if (o_tty && o_tty->ldisc.refcount) {
+			spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+			tty_ldisc_put(o_tty->ldisc.ops);
+			if (wait_event_interruptible(tty_ldisc_wait, o_tty->ldisc.refcount == 0) < 0)
+				return -ERESTARTSYS;
+			goto restart;
+		}
+	}
+	/*
+	 *	If the TTY_LDISC bit is set, then we are racing against
+	 *	another ldisc change
+	 */
+	if (!test_bit(TTY_LDISC, &tty->flags)) {
+		struct tty_ldisc *ld;
+		spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+		tty_ldisc_put(new_ldisc.ops);
+		ld = tty_ldisc_ref_wait(tty);
+		tty_ldisc_deref(ld);
+		goto restart;
+	}
+
+	clear_bit(TTY_LDISC, &tty->flags);
+	if (o_tty)
+		clear_bit(TTY_LDISC, &o_tty->flags);
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+	
+	/*
+	 *	From this point on we know nobody has an ldisc
+	 *	usage reference, nor can they obtain one until
+	 *	we say so later on.
+	 */
+
+	work = cancel_delayed_work(&tty->buf.work);
+	/*
+	 * Wait for ->hangup_work and ->buf.work handlers to terminate
+	 * MUST NOT hold locks here.
+	 */
+	flush_scheduled_work();
+	/* Shutdown the current discipline. */
+	if (o_ldisc.ops->close)
+		(o_ldisc.ops->close)(tty);
+
+	/* Now set up the new line discipline. */
+	tty_ldisc_assign(tty, &new_ldisc);
+	tty_set_termios_ldisc(tty, ldisc);
+	if (new_ldisc.ops->open)
+		retval = (new_ldisc.ops->open)(tty);
+	if (retval < 0) {
+		tty_ldisc_put(new_ldisc.ops);
+		tty_ldisc_restore(tty, &o_ldisc);
+	}
+	/* At this point we hold a reference to the new ldisc and a
+	   a reference to the old ldisc. If we ended up flipping back
+	   to the existing ldisc we have two references to it */
+
+	if (tty->ldisc.ops->num != o_ldisc.ops->num && tty->ops->set_ldisc)
+		tty->ops->set_ldisc(tty);
+
+	tty_ldisc_put(o_ldisc.ops);
+
+	/*
+	 *	Allow ldisc referencing to occur as soon as the driver
+	 *	ldisc callback completes.
+	 */
+
+	tty_ldisc_enable(tty);
+	if (o_tty)
+		tty_ldisc_enable(o_tty);
+
+	/* Restart it in case no characters kick it off. Safe if
+	   already running */
+	if (work)
+		schedule_delayed_work(&tty->buf.work, 1);
+	return retval;
+}
+
+
+/**
+ *	tty_ldisc_setup			-	open line discipline
+ *	@tty: tty being shut down
+ *	@o_tty: pair tty for pty/tty pairs
+ *
+ *	Called during the initial open of a tty/pty pair in order to set up the
+ *	line discplines and bind them to the tty.
+ */
+
+int tty_ldisc_setup(struct tty_struct *tty, struct tty_struct *o_tty)
+{
+	struct tty_ldisc *ld = &tty->ldisc;
+	int retval;
+
+	if (ld->ops->open) {
+		retval = (ld->ops->open)(tty);
+		if (retval)
+			return retval;
+	}
+	if (o_tty && o_tty->ldisc.ops->open) {
+		retval = (o_tty->ldisc.ops->open)(o_tty);
+		if (retval) {
+			if (ld->ops->close)
+				(ld->ops->close)(tty);
+			return retval;
+		}
+		tty_ldisc_enable(o_tty);
+	}
+	tty_ldisc_enable(tty);
+	return 0;
+}
+
+/**
+ *	tty_ldisc_release		-	release line discipline
+ *	@tty: tty being shut down
+ *	@o_tty: pair tty for pty/tty pairs
+ *
+ *	Called during the final close of a tty/pty pair in order to shut down the
+ *	line discpline layer.
+ */
+
+void tty_ldisc_release(struct tty_struct *tty, struct tty_struct *o_tty)
+{
+	unsigned long flags;
+	struct tty_ldisc ld;
+	/*
+	 * Prevent flush_to_ldisc() from rescheduling the work for later.  Then
+	 * kill any delayed work. As this is the final close it does not
+	 * race with the set_ldisc code path.
+	 */
+	clear_bit(TTY_LDISC, &tty->flags);
+	cancel_delayed_work(&tty->buf.work);
+
+	/*
+	 * Wait for ->hangup_work and ->buf.work handlers to terminate
+	 */
+
+	flush_scheduled_work();
+
+	/*
+	 * Wait for any short term users (we know they are just driver
+	 * side waiters as the file is closing so user count on the file
+	 * side is zero.
+	 */
+	spin_lock_irqsave(&tty_ldisc_lock, flags);
+	while (tty->ldisc.refcount) {
+		spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+		wait_event(tty_ldisc_wait, tty->ldisc.refcount == 0);
+		spin_lock_irqsave(&tty_ldisc_lock, flags);
+	}
+	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
+	/*
+	 * Shutdown the current line discipline, and reset it to N_TTY.
+	 *
+	 * FIXME: this MUST get fixed for the new reflocking
+	 */
+	if (tty->ldisc.ops->close)
+		(tty->ldisc.ops->close)(tty);
+	tty_ldisc_put(tty->ldisc.ops);
+
+	/*
+	 *	Switch the line discipline back
+	 */
+	WARN_ON(tty_ldisc_get(N_TTY, &ld));
+	tty_ldisc_assign(tty, &ld);
+	tty_set_termios_ldisc(tty, N_TTY);
+	if (o_tty) {
+		/* FIXME: could o_tty be in setldisc here ? */
+		clear_bit(TTY_LDISC, &o_tty->flags);
+		if (o_tty->ldisc.ops->close)
+			(o_tty->ldisc.ops->close)(o_tty);
+		tty_ldisc_put(o_tty->ldisc.ops);
+		WARN_ON(tty_ldisc_get(N_TTY, &ld));
+		tty_ldisc_assign(o_tty, &ld);
+		tty_set_termios_ldisc(o_tty, N_TTY);
+	}
+}
+
+/**
+ *	tty_ldisc_init		-	ldisc setup for new tty
+ *	@tty: tty being allocated
+ *
+ *	Set up the line discipline objects for a newly allocated tty. Note that
+ *	the tty structure is not completely set up when this call is made.
+ */
+
+void tty_ldisc_init(struct tty_struct *tty)
+{
+	struct tty_ldisc ld;
+	if (tty_ldisc_get(N_TTY, &ld) < 0)
+		panic("n_tty: init_tty");
+	tty_ldisc_assign(tty, &ld);
+}
+
+void tty_ldisc_begin(void)
+{
+	/* Setup the default TTY line discipline. */
+	(void) tty_register_ldisc(N_TTY, &tty_ldisc_N_TTY);
+}
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 4e5833073aa6..e3579cb086e0 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -317,8 +317,6 @@ extern void tty_wait_until_sent(struct tty_struct *tty, long timeout);
 extern int tty_check_change(struct tty_struct *tty);
 extern void stop_tty(struct tty_struct *tty);
 extern void start_tty(struct tty_struct *tty);
-extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
-extern int tty_unregister_ldisc(int disc);
 extern int tty_register_driver(struct tty_driver *driver);
 extern int tty_unregister_driver(struct tty_driver *driver);
 extern struct device *tty_register_device(struct tty_driver *driver,
@@ -383,6 +381,15 @@ extern void tty_port_init(struct tty_port *port);
 extern int tty_port_alloc_xmit_buf(struct tty_port *port);
 extern void tty_port_free_xmit_buf(struct tty_port *port);
 
+extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
+extern int tty_unregister_ldisc(int disc);
+extern int tty_set_ldisc(struct tty_struct *tty, int ldisc);
+extern int tty_ldisc_setup(struct tty_struct *tty, struct tty_struct *o_tty);
+extern void tty_ldisc_release(struct tty_struct *tty, struct tty_struct *o_tty);
+extern void tty_ldisc_init(struct tty_struct *tty);
+extern void tty_ldisc_begin(void);
+/* This last one is just for the tty layer internals and shouldn't be used elsewhere */
+extern void tty_ldisc_enable(struct tty_struct *tty);
 
 
 /* n_tty.c */
-- 
cgit v1.2.3


From 9e98966c7bb94355689478bc84cc3e0c190f977e Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Tue, 22 Jul 2008 11:18:03 +0100
Subject: tty: rework break handling

Some hardware needs to do break handling itself and may have partial
support only. Make break_ctl return an error code. Add a tty driver flag
so you can indicate driver hardware side break support.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/Kconfig              |  2 +-
 drivers/char/amiserial.c          |  3 +-
 drivers/char/cyclades.c           |  8 ++---
 drivers/char/esp.c                |  5 +--
 drivers/char/istallion.c          | 11 +++---
 drivers/char/moxa.c               |  3 +-
 drivers/char/mxser.c              |  3 +-
 drivers/char/pcmcia/synclink_cs.c |  5 +--
 drivers/char/rocket.c             |  5 +--
 drivers/char/sx.c                 |  3 +-
 drivers/char/synclink.c           |  7 ++--
 drivers/char/synclink_gt.c        |  9 ++---
 drivers/char/synclinkmp.c         |  9 ++---
 drivers/char/tty_io.c             | 71 +++++++++++++++------------------------
 drivers/char/vme_scc.c            |  5 +--
 drivers/isdn/capi/capi.c          |  3 +-
 drivers/serial/serial_core.c      |  3 +-
 drivers/usb/class/cdc-acm.c       |  9 +++--
 drivers/usb/serial/usb-serial.c   |  3 +-
 include/linux/tty_driver.h        | 14 ++++++--
 20 files changed, 97 insertions(+), 84 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index ba8782b9c217..a185263b5862 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -218,7 +218,7 @@ config MOXA_SMARTIO
 
 config ISI
 	tristate "Multi-Tech multiport card support (EXPERIMENTAL)"
-	depends on SERIAL_NONSTANDARD && PCI
+	depends on SERIAL_NONSTANDARD && PCI && BROKEN
 	select FW_LOADER
 	help
 	  This is a driver for the Multi-Tech cards which provide several
diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c
index 37457e5a4f2b..3530ff417a51 100644
--- a/drivers/char/amiserial.c
+++ b/drivers/char/amiserial.c
@@ -1248,7 +1248,7 @@ static int rs_tiocmset(struct tty_struct *tty, struct file *file,
 /*
  * rs_break() --- routine which turns the break handling on or off
  */
-static void rs_break(struct tty_struct *tty, int break_state)
+static int rs_break(struct tty_struct *tty, int break_state)
 {
 	struct async_struct * info = (struct async_struct *)tty->driver_data;
 	unsigned long flags;
@@ -1263,6 +1263,7 @@ static void rs_break(struct tty_struct *tty, int break_state)
 	  custom.adkcon = AC_UARTBRK;
 	mb();
 	local_irq_restore(flags);
+	return 0;
 }
 
 
diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c
index e991dc85f2fb..fe6d774fe2e4 100644
--- a/drivers/char/cyclades.c
+++ b/drivers/char/cyclades.c
@@ -3700,14 +3700,15 @@ cy_tiocmset(struct tty_struct *tty, struct file *file,
 /*
  * cy_break() --- routine which turns the break handling on or off
  */
-static void cy_break(struct tty_struct *tty, int break_state)
+static int cy_break(struct tty_struct *tty, int break_state)
 {
 	struct cyclades_port *info = tty->driver_data;
 	struct cyclades_card *card;
 	unsigned long flags;
+	int retval = 0;
 
 	if (serial_paranoia_check(info, tty->name, "cy_break"))
-		return;
+		return -EINVAL;
 
 	card = info->card;
 
@@ -3736,8 +3737,6 @@ static void cy_break(struct tty_struct *tty, int break_state)
 			}
 		}
 	} else {
-		int retval;
-
 		if (break_state == -1) {
 			retval = cyz_issue_cmd(card,
 				info->line - card->first_line,
@@ -3758,6 +3757,7 @@ static void cy_break(struct tty_struct *tty, int break_state)
 		}
 	}
 	spin_unlock_irqrestore(&card->card_lock, flags);
+	return retval;
 }				/* cy_break */
 
 static int get_mon_info(struct cyclades_port *info,
diff --git a/drivers/char/esp.c b/drivers/char/esp.c
index 2eaf09f93e3d..7f077c0097f6 100644
--- a/drivers/char/esp.c
+++ b/drivers/char/esp.c
@@ -1725,13 +1725,13 @@ static int esp_tiocmset(struct tty_struct *tty, struct file *file,
 /*
  * rs_break() --- routine which turns the break handling on or off
  */
-static void esp_break(struct tty_struct *tty, int break_state)
+static int esp_break(struct tty_struct *tty, int break_state)
 {
 	struct esp_struct *info = tty->driver_data;
 	unsigned long flags;
 
 	if (serial_paranoia_check(info, tty->name, "esp_break"))
-		return;
+		return -EINVAL;
 
 	if (break_state == -1) {
 		spin_lock_irqsave(&info->lock, flags);
@@ -1747,6 +1747,7 @@ static void esp_break(struct tty_struct *tty, int break_state)
 		serial_out(info, UART_ESI_CMD2, 0x00);
 		spin_unlock_irqrestore(&info->lock, flags);
 	}
+	return 0;
 }
 
 static int rs_ioctl(struct tty_struct *tty, struct file *file,
diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 7930fba4bafc..63d22b5ebc0d 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -609,7 +609,7 @@ static void	stli_unthrottle(struct tty_struct *tty);
 static void	stli_stop(struct tty_struct *tty);
 static void	stli_start(struct tty_struct *tty);
 static void	stli_flushbuffer(struct tty_struct *tty);
-static void	stli_breakctl(struct tty_struct *tty, int state);
+static int	stli_breakctl(struct tty_struct *tty, int state);
 static void	stli_waituntilsent(struct tty_struct *tty, int timeout);
 static void	stli_sendxchar(struct tty_struct *tty, char ch);
 static void	stli_hangup(struct tty_struct *tty);
@@ -1909,7 +1909,7 @@ static void stli_flushbuffer(struct tty_struct *tty)
 
 /*****************************************************************************/
 
-static void stli_breakctl(struct tty_struct *tty, int state)
+static int stli_breakctl(struct tty_struct *tty, int state)
 {
 	struct stlibrd	*brdp;
 	struct stliport	*portp;
@@ -1917,15 +1917,16 @@ static void stli_breakctl(struct tty_struct *tty, int state)
 
 	portp = tty->driver_data;
 	if (portp == NULL)
-		return;
+		return -EINVAL;
 	if (portp->brdnr >= stli_nrbrds)
-		return;
+		return -EINVAL;
 	brdp = stli_brds[portp->brdnr];
 	if (brdp == NULL)
-		return;
+		return -EINVAL;
 
 	arg = (state == -1) ? BREAKON : BREAKOFF;
 	stli_cmdwait(brdp, portp, A_BREAK, &arg, sizeof(long), 0);
+	return 0;
 }
 
 /*****************************************************************************/
diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c
index 2bba250ffc8e..d3d7864e0c1e 100644
--- a/drivers/char/moxa.c
+++ b/drivers/char/moxa.c
@@ -374,12 +374,13 @@ copy:
 	return ret;
 }
 
-static void moxa_break_ctl(struct tty_struct *tty, int state)
+static int moxa_break_ctl(struct tty_struct *tty, int state)
 {
 	struct moxa_port *port = tty->driver_data;
 
 	moxafunc(port->tableAddr, state ? FC_SendBreak : FC_StopBreak,
 			Magic_code);
+	return 0;
 }
 
 static const struct tty_operations moxa_ops = {
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 1fb25571bf85..f04c3c58a05a 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -2183,7 +2183,7 @@ static void mxser_hangup(struct tty_struct *tty)
 /*
  * mxser_rs_break() --- routine which turns the break handling on or off
  */
-static void mxser_rs_break(struct tty_struct *tty, int break_state)
+static int mxser_rs_break(struct tty_struct *tty, int break_state)
 {
 	struct mxser_port *info = tty->driver_data;
 	unsigned long flags;
@@ -2196,6 +2196,7 @@ static void mxser_rs_break(struct tty_struct *tty, int break_state)
 		outb(inb(info->ioaddr + UART_LCR) & ~UART_LCR_SBC,
 			info->ioaddr + UART_LCR);
 	spin_unlock_irqrestore(&info->slock, flags);
+	return 0;
 }
 
 static void mxser_receive_chars(struct mxser_port *port, int *status)
diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index b694d430f10e..d1fceabe3aef 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -2230,7 +2230,7 @@ static int tiocmset(struct tty_struct *tty, struct file *file,
  * Arguments:		tty		pointer to tty instance data
  *			break_state	-1=set break condition, 0=clear
  */
-static void mgslpc_break(struct tty_struct *tty, int break_state)
+static int mgslpc_break(struct tty_struct *tty, int break_state)
 {
 	MGSLPC_INFO * info = (MGSLPC_INFO *)tty->driver_data;
 	unsigned long flags;
@@ -2240,7 +2240,7 @@ static void mgslpc_break(struct tty_struct *tty, int break_state)
 			 __FILE__,__LINE__, info->device_name, break_state);
 
 	if (mgslpc_paranoia_check(info, tty->name, "mgslpc_break"))
-		return;
+		return -EINVAL;
 
 	spin_lock_irqsave(&info->lock,flags);
  	if (break_state == -1)
@@ -2248,6 +2248,7 @@ static void mgslpc_break(struct tty_struct *tty, int break_state)
 	else
 		clear_reg_bits(info, CHA+DAFO, BIT6);
 	spin_unlock_irqrestore(&info->lock,flags);
+	return 0;
 }
 
 /* Service an IOCTL request
diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c
index e670eae2f510..584d791e84a6 100644
--- a/drivers/char/rocket.c
+++ b/drivers/char/rocket.c
@@ -1236,13 +1236,13 @@ static void rp_set_termios(struct tty_struct *tty,
 	}
 }
 
-static void rp_break(struct tty_struct *tty, int break_state)
+static int rp_break(struct tty_struct *tty, int break_state)
 {
 	struct r_port *info = (struct r_port *) tty->driver_data;
 	unsigned long flags;
 
 	if (rocket_paranoia_check(info, "rp_break"))
-		return;
+		return -EINVAL;
 
 	spin_lock_irqsave(&info->slock, flags);
 	if (break_state == -1)
@@ -1250,6 +1250,7 @@ static void rp_break(struct tty_struct *tty, int break_state)
 	else
 		sClrBreak(&info->channel);
 	spin_unlock_irqrestore(&info->slock, flags);
+	return 0;
 }
 
 /*
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index d5cffcd6a572..2162439bbe48 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -1840,7 +1840,7 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 	return rc;
 }
 
-static void sx_break(struct tty_struct *tty, int flag)
+static int sx_break(struct tty_struct *tty, int flag)
 {
 	struct sx_port *port = tty->driver_data;
 	int rv;
@@ -1857,6 +1857,7 @@ static void sx_break(struct tty_struct *tty, int flag)
 			read_sx_byte(port->board, CHAN_OFFSET(port, hi_hstat)));
 	unlock_kernel();
 	func_exit();
+	return 0;
 }
 
 static int sx_tiocmget(struct tty_struct *tty, struct file *file)
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index 527d220aa4aa..ef6706f09061 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -2897,9 +2897,9 @@ static int tiocmset(struct tty_struct *tty, struct file *file,
  *
  * Arguments:		tty		pointer to tty instance data
  *			break_state	-1=set break condition, 0=clear
- * Return Value:	None
+ * Return Value:	error code
  */
-static void mgsl_break(struct tty_struct *tty, int break_state)
+static int mgsl_break(struct tty_struct *tty, int break_state)
 {
 	struct mgsl_struct * info = (struct mgsl_struct *)tty->driver_data;
 	unsigned long flags;
@@ -2909,7 +2909,7 @@ static void mgsl_break(struct tty_struct *tty, int break_state)
 			 __FILE__,__LINE__, info->device_name, break_state);
 			 
 	if (mgsl_paranoia_check(info, tty->name, "mgsl_break"))
-		return;
+		return -EINVAL;
 
 	spin_lock_irqsave(&info->irq_spinlock,flags);
  	if (break_state == -1)
@@ -2917,6 +2917,7 @@ static void mgsl_break(struct tty_struct *tty, int break_state)
 	else 
 		usc_OutReg(info,IOCR,(u16)(usc_InReg(info,IOCR) & ~BIT7));
 	spin_unlock_irqrestore(&info->irq_spinlock,flags);
+	return 0;
 	
 }	/* end of mgsl_break() */
 
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 2c3e43bb2cc9..cf87bb89a77d 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -165,7 +165,7 @@ static int  read_proc(char *page, char **start, off_t off, int count,int *eof, v
 static int  chars_in_buffer(struct tty_struct *tty);
 static void throttle(struct tty_struct * tty);
 static void unthrottle(struct tty_struct * tty);
-static void set_break(struct tty_struct *tty, int break_state);
+static int set_break(struct tty_struct *tty, int break_state);
 
 /*
  * generic HDLC support and callbacks
@@ -513,7 +513,7 @@ static int  wait_mgsl_event(struct slgt_info *info, int __user *mask_ptr);
 static int  tiocmget(struct tty_struct *tty, struct file *file);
 static int  tiocmset(struct tty_struct *tty, struct file *file,
 		     unsigned int set, unsigned int clear);
-static void set_break(struct tty_struct *tty, int break_state);
+static int set_break(struct tty_struct *tty, int break_state);
 static int  get_interface(struct slgt_info *info, int __user *if_mode);
 static int  set_interface(struct slgt_info *info, int if_mode);
 static int  set_gpio(struct slgt_info *info, struct gpio_desc __user *gpio);
@@ -1452,14 +1452,14 @@ static void unthrottle(struct tty_struct * tty)
  * set or clear transmit break condition
  * break_state	-1=set break condition, 0=clear
  */
-static void set_break(struct tty_struct *tty, int break_state)
+static int set_break(struct tty_struct *tty, int break_state)
 {
 	struct slgt_info *info = tty->driver_data;
 	unsigned short value;
 	unsigned long flags;
 
 	if (sanity_check(info, tty->name, "set_break"))
-		return;
+		return -EINVAL;
 	DBGINFO(("%s set_break(%d)\n", info->device_name, break_state));
 
 	spin_lock_irqsave(&info->lock,flags);
@@ -1470,6 +1470,7 @@ static void set_break(struct tty_struct *tty, int break_state)
 		value &= ~BIT6;
 	wr_reg16(info, TCR, value);
 	spin_unlock_irqrestore(&info->lock,flags);
+	return 0;
 }
 
 #if SYNCLINK_GENERIC_HDLC
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index 5768c4136342..c0490cbd0db2 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -527,7 +527,7 @@ static int  read_proc(char *page, char **start, off_t off, int count,int *eof, v
 static int  chars_in_buffer(struct tty_struct *tty);
 static void throttle(struct tty_struct * tty);
 static void unthrottle(struct tty_struct * tty);
-static void set_break(struct tty_struct *tty, int break_state);
+static int set_break(struct tty_struct *tty, int break_state);
 
 #if SYNCLINK_GENERIC_HDLC
 #define dev_to_port(D) (dev_to_hdlc(D)->priv)
@@ -552,7 +552,7 @@ static int  wait_mgsl_event(SLMP_INFO *info, int __user *mask_ptr);
 static int  tiocmget(struct tty_struct *tty, struct file *file);
 static int  tiocmset(struct tty_struct *tty, struct file *file,
 		     unsigned int set, unsigned int clear);
-static void set_break(struct tty_struct *tty, int break_state);
+static int  set_break(struct tty_struct *tty, int break_state);
 
 static void add_device(SLMP_INFO *info);
 static void device_init(int adapter_num, struct pci_dev *pdev);
@@ -1587,7 +1587,7 @@ static void unthrottle(struct tty_struct * tty)
 /* set or clear transmit break condition
  * break_state	-1=set break condition, 0=clear
  */
-static void set_break(struct tty_struct *tty, int break_state)
+static int set_break(struct tty_struct *tty, int break_state)
 {
 	unsigned char RegValue;
 	SLMP_INFO * info = (SLMP_INFO *)tty->driver_data;
@@ -1598,7 +1598,7 @@ static void set_break(struct tty_struct *tty, int break_state)
 			 __FILE__,__LINE__, info->device_name, break_state);
 
 	if (sanity_check(info, tty->name, "set_break"))
-		return;
+		return -EINVAL;
 
 	spin_lock_irqsave(&info->lock,flags);
 	RegValue = read_reg(info, CTL);
@@ -1608,6 +1608,7 @@ static void set_break(struct tty_struct *tty, int break_state)
 		RegValue &= ~BIT3;
 	write_reg(info, CTL, RegValue);
 	spin_unlock_irqrestore(&info->lock,flags);
+	return 0;
 }
 
 #if SYNCLINK_GENERIC_HDLC
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index d27a08b374d0..d94cd8410c53 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -2849,16 +2849,29 @@ static int tiocsetd(struct tty_struct *tty, int __user *p)
 
 static int send_break(struct tty_struct *tty, unsigned int duration)
 {
-	if (tty_write_lock(tty, 0) < 0)
-		return -EINTR;
-	tty->ops->break_ctl(tty, -1);
-	if (!signal_pending(current))
-		msleep_interruptible(duration);
-	tty->ops->break_ctl(tty, 0);
-	tty_write_unlock(tty);
-	if (signal_pending(current))
-		return -EINTR;
-	return 0;
+	int retval;
+
+	if (tty->ops->break_ctl == NULL)
+		return 0;
+
+	if (tty->driver->flags & TTY_DRIVER_HARDWARE_BREAK)
+		retval = tty->ops->break_ctl(tty, duration);
+	else {
+		/* Do the work ourselves */
+		if (tty_write_lock(tty, 0) < 0)
+			return -EINTR;
+		retval = tty->ops->break_ctl(tty, -1);
+		if (retval)
+			goto out;
+		if (!signal_pending(current))
+			msleep_interruptible(duration);
+		retval = tty->ops->break_ctl(tty, 0);
+out:
+		tty_write_unlock(tty);
+		if (signal_pending(current))
+			retval = -EINTR;
+	}
+	return retval;
 }
 
 /**
@@ -2949,36 +2962,6 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	    tty->driver->subtype == PTY_TYPE_MASTER)
 		real_tty = tty->link;
 
-	/*
-	 * Break handling by driver
-	 */
-
-	retval = -EINVAL;
-
-	if (!tty->ops->break_ctl) {
-		switch (cmd) {
-		case TIOCSBRK:
-		case TIOCCBRK:
-			if (tty->ops->ioctl)
-				retval = tty->ops->ioctl(tty, file, cmd, arg);
-			if (retval != -EINVAL && retval != -ENOIOCTLCMD)
-				printk(KERN_WARNING "tty: driver %s needs updating to use break_ctl\n", tty->driver->name);
-			return retval;
-
-		/* These two ioctl's always return success; even if */
-		/* the driver doesn't support them. */
-		case TCSBRK:
-		case TCSBRKP:
-			if (!tty->ops->ioctl)
-				return 0;
-			retval = tty->ops->ioctl(tty, file, cmd, arg);
-			if (retval != -EINVAL && retval != -ENOIOCTLCMD)
-				printk(KERN_WARNING "tty: driver %s needs updating to use break_ctl\n", tty->driver->name);
-			if (retval == -ENOIOCTLCMD)
-				retval = 0;
-			return retval;
-		}
-	}
 
 	/*
 	 * Factor out some common prep work
@@ -3000,6 +2983,9 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 		break;
 	}
 
+	/*
+	 *	Now do the stuff.
+	 */
 	switch (cmd) {
 	case TIOCSTI:
 		return tiocsti(tty, p);
@@ -3043,12 +3029,11 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	 */
 	case TIOCSBRK:	/* Turn break on, unconditionally */
 		if (tty->ops->break_ctl)
-			tty->ops->break_ctl(tty, -1);
+			return tty->ops->break_ctl(tty, -1);
 		return 0;
-
 	case TIOCCBRK:	/* Turn break off, unconditionally */
 		if (tty->ops->break_ctl)
-			tty->ops->break_ctl(tty, 0);
+			return tty->ops->break_ctl(tty, 0);
 		return 0;
 	case TCSBRK:   /* SVID version: non-zero arg --> no break */
 		/* non-zero arg means wait for all output data
diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c
index f17ac043b551..69c5afe97f19 100644
--- a/drivers/char/vme_scc.c
+++ b/drivers/char/vme_scc.c
@@ -85,7 +85,7 @@ static irqreturn_t scc_rx_int(int irq, void *data);
 static irqreturn_t scc_stat_int(int irq, void *data);
 static irqreturn_t scc_spcond_int(int irq, void *data);
 static void scc_setsignals(struct scc_port *port, int dtr, int rts);
-static void scc_break_ctl(struct tty_struct *tty, int break_state);
+static int scc_break_ctl(struct tty_struct *tty, int break_state);
 
 static struct tty_driver *scc_driver;
 
@@ -942,7 +942,7 @@ static int scc_ioctl(struct tty_struct *tty, struct file *file,
 }
 
 
-static void scc_break_ctl(struct tty_struct *tty, int break_state)
+static int scc_break_ctl(struct tty_struct *tty, int break_state)
 {
 	struct scc_port *port = (struct scc_port *)tty->driver_data;
 	unsigned long	flags;
@@ -952,6 +952,7 @@ static void scc_break_ctl(struct tty_struct *tty, int break_state)
 	SCCmod(TX_CTRL_REG, ~TCR_SEND_BREAK, 
 			break_state ? TCR_SEND_BREAK : 0);
 	local_irq_restore(flags);
+	return 0;
 }
 
 
diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c
index 8a35029caca0..19e005e81fef 100644
--- a/drivers/isdn/capi/capi.c
+++ b/drivers/isdn/capi/capi.c
@@ -1302,11 +1302,12 @@ static void capinc_tty_hangup(struct tty_struct *tty)
 #endif
 }
 
-static void capinc_tty_break_ctl(struct tty_struct *tty, int state)
+static int capinc_tty_break_ctl(struct tty_struct *tty, int state)
 {
 #ifdef _DEBUG_TTYFUNCS
 	printk(KERN_DEBUG "capinc_tty_break_ctl(%d)\n", state);
 #endif
+	return 0;
 }
 
 static void capinc_tty_flush_buffer(struct tty_struct *tty)
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 0bce1fe2c62a..f977c98cfa95 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -934,7 +934,7 @@ uart_tiocmset(struct tty_struct *tty, struct file *file,
 	return ret;
 }
 
-static void uart_break_ctl(struct tty_struct *tty, int break_state)
+static int uart_break_ctl(struct tty_struct *tty, int break_state)
 {
 	struct uart_state *state = tty->driver_data;
 	struct uart_port *port = state->port;
@@ -945,6 +945,7 @@ static void uart_break_ctl(struct tty_struct *tty, int break_state)
 		port->ops->break_ctl(port, break_state);
 
 	mutex_unlock(&state->mutex);
+	return 0;
 }
 
 static int uart_do_autoconfig(struct uart_state *state)
diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
index 95ae6377d7e5..0725b1871f23 100644
--- a/drivers/usb/class/cdc-acm.c
+++ b/drivers/usb/class/cdc-acm.c
@@ -732,13 +732,16 @@ static void acm_tty_unthrottle(struct tty_struct *tty)
 	tasklet_schedule(&acm->urb_task);
 }
 
-static void acm_tty_break_ctl(struct tty_struct *tty, int state)
+static int acm_tty_break_ctl(struct tty_struct *tty, int state)
 {
 	struct acm *acm = tty->driver_data;
+	int retval;
 	if (!ACM_READY(acm))
-		return;
-	if (acm_send_break(acm, state ? 0xffff : 0))
+		return -EINVAL;
+	retval = acm_send_break(acm, state ? 0xffff : 0);
+	if (retval < 0)
 		dbg("send break failed");
+	return retval;
 }
 
 static int acm_tty_tiocmget(struct tty_struct *tty, struct file *file)
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 51917b0f079a..8c2d531eedea 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -395,7 +395,7 @@ static void serial_set_termios(struct tty_struct *tty, struct ktermios *old)
 		tty_termios_copy_hw(tty->termios, old);
 }
 
-static void serial_break(struct tty_struct *tty, int break_state)
+static int serial_break(struct tty_struct *tty, int break_state)
 {
 	struct usb_serial_port *port = tty->driver_data;
 
@@ -409,6 +409,7 @@ static void serial_break(struct tty_struct *tty, int break_state)
 		port->serial->type->break_ctl(tty, break_state);
 		unlock_kernel();
 	}
+	return 0;
 }
 
 static int serial_read_proc(char *page, char **start, off_t off, int count,
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index d2a003586761..e1065ac0d922 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -135,7 +135,7 @@
  *
  *	Optional:
  *
- * void (*break_ctl)(struct tty_stuct *tty, int state);
+ * int (*break_ctl)(struct tty_stuct *tty, int state);
  *
  * 	This optional routine requests the tty driver to turn on or
  * 	off BREAK status on the RS-232 port.  If state is -1,
@@ -146,6 +146,10 @@
  * 	handle the following ioctls: TCSBRK, TCSBRKP, TIOCSBRK,
  * 	TIOCCBRK.
  *
+ *	If the driver sets TTY_DRIVER_HARDWARE_BREAK then the interface
+ *	will also be called with actual times and the hardware is expected
+ *	to do the delay work itself. 0 and -1 are still used for on/off.
+ *
  *	Optional: Required for TCSBRK/BRKP/etc handling.
  *
  * void (*wait_until_sent)(struct tty_struct *tty, int timeout);
@@ -192,7 +196,7 @@ struct tty_operations {
 	void (*stop)(struct tty_struct *tty);
 	void (*start)(struct tty_struct *tty);
 	void (*hangup)(struct tty_struct *tty);
-	void (*break_ctl)(struct tty_struct *tty, int state);
+	int (*break_ctl)(struct tty_struct *tty, int state);
 	void (*flush_buffer)(struct tty_struct *tty);
 	void (*set_ldisc)(struct tty_struct *tty);
 	void (*wait_until_sent)(struct tty_struct *tty, int timeout);
@@ -285,12 +289,18 @@ extern struct tty_driver *tty_find_polling_driver(char *name, int *line);
  * TTY_DRIVER_DEVPTS_MEM -- don't use the standard arrays, instead
  *	use dynamic memory keyed through the devpts filesystem.  This
  *	is only applicable to the pty driver.
+ *
+ * TTY_DRIVER_HARDWARE_BREAK -- hardware handles break signals. Pass
+ *	the requested timeout to the caller instead of using a simple
+ *	on/off interface.
+ *
  */
 #define TTY_DRIVER_INSTALLED		0x0001
 #define TTY_DRIVER_RESET_TERMIOS	0x0002
 #define TTY_DRIVER_REAL_RAW		0x0004
 #define TTY_DRIVER_DYNAMIC_DEV		0x0008
 #define TTY_DRIVER_DEVPTS_MEM		0x0010
+#define TTY_DRIVER_HARDWARE_BREAK	0x0020
 
 /* tty driver types */
 #define TTY_DRIVER_TYPE_SYSTEM		0x0001
-- 
cgit v1.2.3


From e5590717afd5fb6f494323206a1a35ea25610c2d Mon Sep 17 00:00:00 2001
From: Paul Fulghum <paulkf@microgate.com>
Date: Tue, 22 Jul 2008 11:21:39 +0100
Subject: synclink_gt: add serial bit order control

Add control of hardware serial bit order between LSB first
(default/standard) and MSB first.

Signed-off-by: Paul Fulghum <paulkf@microgate.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/synclink_gt.c | 2 ++
 include/linux/synclink.h   | 1 +
 2 files changed, 3 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 0e59cf54adaf..3cfc9e1f8882 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -4403,6 +4403,8 @@ static void msc_set_vcr(struct slgt_info *info)
 		break;
 	}
 
+	if (info->if_mode & MGSL_INTERFACE_MSB_FIRST)
+		val |= BIT4;
 	if (info->signals & SerialSignal_DTR)
 		val |= BIT3;
 	if (info->signals & SerialSignal_RTS)
diff --git a/include/linux/synclink.h b/include/linux/synclink.h
index 45f6bc82d317..c844a229acc9 100644
--- a/include/linux/synclink.h
+++ b/include/linux/synclink.h
@@ -136,6 +136,7 @@
 #define MGSL_INTERFACE_RTS_EN   0x10
 #define MGSL_INTERFACE_LL       0x20
 #define MGSL_INTERFACE_RL       0x40
+#define MGSL_INTERFACE_MSB_FIRST 0x80
 
 typedef struct _MGSL_PARAMS
 {
-- 
cgit v1.2.3


From 217d5a51953143046cf2972eebdefe2b2db05718 Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurentp@cse-semaphore.com>
Date: Tue, 22 Jul 2008 17:43:42 +0200
Subject: fs_enet: Remove unused fields in the fs_mii_bb_platform_info
 structure.

The mdio_port, mdio_bit, mdc_port and mdc_bit fields in the
fs_mii_bb_platform_info structure are left-overs from the move to the Phy
Abstraction Layer subsystem. They are not used anymore and can be safely
removed.

Signed-off-by: Laurent Pinchart <laurentp@cse-semaphore.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/fs_enet_pd.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/fs_enet_pd.h b/include/linux/fs_enet_pd.h
index 9bc045b8c478..decc4b56ccf9 100644
--- a/include/linux/fs_enet_pd.h
+++ b/include/linux/fs_enet_pd.h
@@ -103,10 +103,6 @@ struct fs_mii_bb_platform_info {
 	struct fs_mii_bit 	mdio_dir;
 	struct fs_mii_bit 	mdio_dat;
 	struct fs_mii_bit	mdc_dat;
-	int mdio_port;	/* port & bit for MDIO */
-	int mdio_bit;
-	int mdc_port;	/* port & bit for MDC  */
-	int mdc_bit;
 	int delay;	/* delay in us         */
 	int irq[32]; 	/* irqs per phy's */
 };
-- 
cgit v1.2.3


From d29f749e252bcdbfe7a75a58f0ee92da16f127c0 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Tue, 22 Jul 2008 14:09:06 -0700
Subject: net: Fix build failure with 'make mandocs'.

The function header comments have to go with the functions
they are documenting, or things go horribly wrong when we
try to process them with the docbook tools.

Warning(include/linux/netdevice.h:1006): No description found for parameter 'dev_queue'
Warning(include/linux/netdevice.h:1033): No description found for parameter 'dev_queue'
Warning(include/linux/netdevice.h:1067): No description found for parameter 'dev_queue'
Warning(include/linux/netdevice.h:1093): No description found for parameter 'dev_queue'
Warning(include/linux/netdevice.h:1474): No description found for parameter 'txq'
Error(net/core/dev.c:1674): cannot understand prototype: 'u32 simple_tx_hashrnd; '

Signed-off-by: Dave Jones <davej@redhat.com>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 58 +++++++++++++++++++++++------------------------
 net/core/dev.c            | 51 ++++++++++++++++++++---------------------
 2 files changed, 54 insertions(+), 55 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f5ea445f89f0..b4d056ceab96 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -996,17 +996,17 @@ static inline void netif_tx_schedule_all(struct net_device *dev)
 		netif_schedule_queue(netdev_get_tx_queue(dev, i));
 }
 
+static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
+{
+	clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 /**
  *	netif_start_queue - allow transmit
  *	@dev: network device
  *
  *	Allow upper layers to call the device hard_start_xmit routine.
  */
-static inline void netif_tx_start_queue(struct netdev_queue *dev_queue)
-{
-	clear_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
-}
-
 static inline void netif_start_queue(struct net_device *dev)
 {
 	netif_tx_start_queue(netdev_get_tx_queue(dev, 0));
@@ -1022,13 +1022,6 @@ static inline void netif_tx_start_all_queues(struct net_device *dev)
 	}
 }
 
-/**
- *	netif_wake_queue - restart transmit
- *	@dev: network device
- *
- *	Allow upper layers to call the device hard_start_xmit routine.
- *	Used for flow control when transmit resources are available.
- */
 static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 {
 #ifdef CONFIG_NETPOLL_TRAP
@@ -1041,6 +1034,13 @@ static inline void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 		__netif_schedule(dev_queue->qdisc);
 }
 
+/**
+ *	netif_wake_queue - restart transmit
+ *	@dev: network device
+ *
+ *	Allow upper layers to call the device hard_start_xmit routine.
+ *	Used for flow control when transmit resources are available.
+ */
 static inline void netif_wake_queue(struct net_device *dev)
 {
 	netif_tx_wake_queue(netdev_get_tx_queue(dev, 0));
@@ -1056,6 +1056,11 @@ static inline void netif_tx_wake_all_queues(struct net_device *dev)
 	}
 }
 
+static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
+{
+	set_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 /**
  *	netif_stop_queue - stop transmitted packets
  *	@dev: network device
@@ -1063,11 +1068,6 @@ static inline void netif_tx_wake_all_queues(struct net_device *dev)
  *	Stop upper layers calling the device hard_start_xmit routine.
  *	Used for flow control when transmit resources are unavailable.
  */
-static inline void netif_tx_stop_queue(struct netdev_queue *dev_queue)
-{
-	set_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
-}
-
 static inline void netif_stop_queue(struct net_device *dev)
 {
 	netif_tx_stop_queue(netdev_get_tx_queue(dev, 0));
@@ -1083,17 +1083,17 @@ static inline void netif_tx_stop_all_queues(struct net_device *dev)
 	}
 }
 
+static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
+{
+	return test_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
+}
+
 /**
  *	netif_queue_stopped - test if transmit queue is flowblocked
  *	@dev: network device
  *
  *	Test if transmit queue on device is currently unable to send.
  */
-static inline int netif_tx_queue_stopped(const struct netdev_queue *dev_queue)
-{
-	return test_bit(__QUEUE_STATE_XOFF, &dev_queue->state);
-}
-
 static inline int netif_queue_stopped(const struct net_device *dev)
 {
 	return netif_tx_queue_stopped(netdev_get_tx_queue(dev, 0));
@@ -1463,13 +1463,6 @@ static inline void netif_rx_complete(struct net_device *dev,
 	local_irq_restore(flags);
 }
 
-/**
- *	netif_tx_lock - grab network device transmit lock
- *	@dev: network device
- *	@cpu: cpu number of lock owner
- *
- * Get network device transmit lock
- */
 static inline void __netif_tx_lock(struct netdev_queue *txq, int cpu)
 {
 	spin_lock(&txq->_xmit_lock);
@@ -1482,6 +1475,13 @@ static inline void __netif_tx_lock_bh(struct netdev_queue *txq)
 	txq->xmit_lock_owner = smp_processor_id();
 }
 
+/**
+ *	netif_tx_lock - grab network device transmit lock
+ *	@dev: network device
+ *	@cpu: cpu number of lock owner
+ *
+ * Get network device transmit lock
+ */
 static inline void netif_tx_lock(struct net_device *dev)
 {
 	int cpu = smp_processor_id();
diff --git a/net/core/dev.c b/net/core/dev.c
index ad5598d2bb37..65eea83613ef 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1645,32 +1645,6 @@ out_kfree_skb:
 	return 0;
 }
 
-/**
- *	dev_queue_xmit - transmit a buffer
- *	@skb: buffer to transmit
- *
- *	Queue a buffer for transmission to a network device. The caller must
- *	have set the device and priority and built the buffer before calling
- *	this function. The function can be called from an interrupt.
- *
- *	A negative errno code is returned on a failure. A success does not
- *	guarantee the frame will be transmitted as it may be dropped due
- *	to congestion or traffic shaping.
- *
- * -----------------------------------------------------------------------------------
- *      I notice this method can also return errors from the queue disciplines,
- *      including NET_XMIT_DROP, which is a positive value.  So, errors can also
- *      be positive.
- *
- *      Regardless of the return value, the skb is consumed, so it is currently
- *      difficult to retry a send to this method.  (You can bump the ref count
- *      before sending to hold a reference for retry if you are careful.)
- *
- *      When calling this method, interrupts MUST be enabled.  This is because
- *      the BH enable code must have IRQs enabled so that it will not deadlock.
- *          --BLG
- */
-
 static u32 simple_tx_hashrnd;
 static int simple_tx_hashrnd_initialized = 0;
 
@@ -1738,6 +1712,31 @@ static struct netdev_queue *dev_pick_tx(struct net_device *dev,
 	return netdev_get_tx_queue(dev, queue_index);
 }
 
+/**
+ *	dev_queue_xmit - transmit a buffer
+ *	@skb: buffer to transmit
+ *
+ *	Queue a buffer for transmission to a network device. The caller must
+ *	have set the device and priority and built the buffer before calling
+ *	this function. The function can be called from an interrupt.
+ *
+ *	A negative errno code is returned on a failure. A success does not
+ *	guarantee the frame will be transmitted as it may be dropped due
+ *	to congestion or traffic shaping.
+ *
+ * -----------------------------------------------------------------------------------
+ *      I notice this method can also return errors from the queue disciplines,
+ *      including NET_XMIT_DROP, which is a positive value.  So, errors can also
+ *      be positive.
+ *
+ *      Regardless of the return value, the skb is consumed, so it is currently
+ *      difficult to retry a send to this method.  (You can bump the ref count
+ *      before sending to hold a reference for retry if you are careful.)
+ *
+ *      When calling this method, interrupts MUST be enabled.  This is because
+ *      the BH enable code must have IRQs enabled so that it will not deadlock.
+ *          --BLG
+ */
 int dev_queue_xmit(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
-- 
cgit v1.2.3


From 8086cd451f08f4c0f9693fc66d87754bbd18cfba Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 22 Jul 2008 14:19:19 -0700
Subject: netns: make get_proc_net() static

get_proc_net() can now become static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 fs/proc/proc_net.c      | 11 +++++------
 include/linux/proc_fs.h |  2 --
 2 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c
index b224a28e0c15..7bc296f424ae 100644
--- a/fs/proc/proc_net.c
+++ b/fs/proc/proc_net.c
@@ -27,6 +27,11 @@
 #include "internal.h"
 
 
+static struct net *get_proc_net(const struct inode *inode)
+{
+	return maybe_get_net(PDE_NET(PDE(inode)));
+}
+
 int seq_open_net(struct inode *ino, struct file *f,
 		 const struct seq_operations *ops, int size)
 {
@@ -185,12 +190,6 @@ void proc_net_remove(struct net *net, const char *name)
 }
 EXPORT_SYMBOL_GPL(proc_net_remove);
 
-struct net *get_proc_net(const struct inode *inode)
-{
-	return maybe_get_net(PDE_NET(PDE(inode)));
-}
-EXPORT_SYMBOL_GPL(get_proc_net);
-
 static __net_init int proc_net_ns_init(struct net *net)
 {
 	struct proc_dir_entry *netd, *net_statd;
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index fff1d27ddb4c..15a9eaf4a802 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -305,8 +305,6 @@ static inline struct net *PDE_NET(struct proc_dir_entry *pde)
 	return pde->parent->data;
 }
 
-struct net *get_proc_net(const struct inode *inode);
-
 struct proc_maps_private {
 	struct pid *pid;
 	struct task_struct *task;
-- 
cgit v1.2.3


From 7f1b358a236ee9c19657a619ac6f2dcabcaa0924 Mon Sep 17 00:00:00 2001
From: Maciej Sosnowski <maciej.sosnowski@intel.com>
Date: Tue, 22 Jul 2008 17:30:57 -0700
Subject: I/OAT: I/OAT version 3.0 support

This patch adds to ioatdma and dca modules
support for Intel I/OAT DMA engine ver.3 (aka CB3 device).
The main features of I/OAT ver.3 are:
 * 8 single channel DMA devices (8 channels total)
 * 8 DCA providers, each can accept 2 requesters
 * 8-bit TAG values and 32-bit extended APIC IDs

Signed-off-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dca/dca-core.c          | 131 ++++++++++++++++-----
 drivers/dca/dca-sysfs.c         |   3 +-
 drivers/dma/ioat.c              |  15 +++
 drivers/dma/ioat_dca.c          | 244 ++++++++++++++++++++++++++++++++++++++--
 drivers/dma/ioat_dma.c          |  96 ++++++++++++++--
 drivers/dma/ioatdma.h           |   5 +-
 drivers/dma/ioatdma_hw.h        |   1 +
 drivers/dma/ioatdma_registers.h |  20 ++++
 include/linux/dca.h             |   7 +-
 include/linux/pci_ids.h         |   8 ++
 10 files changed, 481 insertions(+), 49 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dca/dca-core.c b/drivers/dca/dca-core.c
index bf5b92f86df7..ec249d2db633 100644
--- a/drivers/dca/dca-core.c
+++ b/drivers/dca/dca-core.c
@@ -28,13 +28,29 @@
 #include <linux/device.h>
 #include <linux/dca.h>
 
-MODULE_LICENSE("GPL");
+#define DCA_VERSION "1.4"
 
-/* For now we're assuming a single, global, DCA provider for the system. */
+MODULE_VERSION(DCA_VERSION);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Intel Corporation");
 
 static DEFINE_SPINLOCK(dca_lock);
 
-static struct dca_provider *global_dca = NULL;
+static LIST_HEAD(dca_providers);
+
+static struct dca_provider *dca_find_provider_by_dev(struct device *dev)
+{
+	struct dca_provider *dca, *ret = NULL;
+
+	list_for_each_entry(dca, &dca_providers, node) {
+		if ((!dev) || (dca->ops->dev_managed(dca, dev))) {
+			ret = dca;
+			break;
+		}
+	}
+
+	return ret;
+}
 
 /**
  * dca_add_requester - add a dca client to the list
@@ -42,25 +58,39 @@ static struct dca_provider *global_dca = NULL;
  */
 int dca_add_requester(struct device *dev)
 {
-	int err, slot;
+	struct dca_provider *dca;
+	int err, slot = -ENODEV;
 
-	if (!global_dca)
-		return -ENODEV;
+	if (!dev)
+		return -EFAULT;
 
 	spin_lock(&dca_lock);
-	slot = global_dca->ops->add_requester(global_dca, dev);
-	spin_unlock(&dca_lock);
-	if (slot < 0)
+
+	/* check if the requester has not been added already */
+	dca = dca_find_provider_by_dev(dev);
+	if (dca) {
+		spin_unlock(&dca_lock);
+		return -EEXIST;
+	}
+
+	list_for_each_entry(dca, &dca_providers, node) {
+		slot = dca->ops->add_requester(dca, dev);
+		if (slot >= 0)
+			break;
+	}
+	if (slot < 0) {
+		spin_unlock(&dca_lock);
 		return slot;
+	}
 
-	err = dca_sysfs_add_req(global_dca, dev, slot);
+	err = dca_sysfs_add_req(dca, dev, slot);
 	if (err) {
-		spin_lock(&dca_lock);
-		global_dca->ops->remove_requester(global_dca, dev);
+		dca->ops->remove_requester(dca, dev);
 		spin_unlock(&dca_lock);
 		return err;
 	}
 
+	spin_unlock(&dca_lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(dca_add_requester);
@@ -71,30 +101,78 @@ EXPORT_SYMBOL_GPL(dca_add_requester);
  */
 int dca_remove_requester(struct device *dev)
 {
+	struct dca_provider *dca;
 	int slot;
-	if (!global_dca)
-		return -ENODEV;
+
+	if (!dev)
+		return -EFAULT;
 
 	spin_lock(&dca_lock);
-	slot = global_dca->ops->remove_requester(global_dca, dev);
-	spin_unlock(&dca_lock);
-	if (slot < 0)
+	dca = dca_find_provider_by_dev(dev);
+	if (!dca) {
+		spin_unlock(&dca_lock);
+		return -ENODEV;
+	}
+	slot = dca->ops->remove_requester(dca, dev);
+	if (slot < 0) {
+		spin_unlock(&dca_lock);
 		return slot;
+	}
 
-	dca_sysfs_remove_req(global_dca, slot);
+	dca_sysfs_remove_req(dca, slot);
+
+	spin_unlock(&dca_lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(dca_remove_requester);
 
 /**
- * dca_get_tag - return the dca tag for the given cpu
+ * dca_common_get_tag - return the dca tag (serves both new and old api)
+ * @dev - the device that wants dca service
  * @cpu - the cpuid as returned by get_cpu()
  */
-u8 dca_get_tag(int cpu)
+u8 dca_common_get_tag(struct device *dev, int cpu)
 {
-	if (!global_dca)
+	struct dca_provider *dca;
+	u8 tag;
+
+	spin_lock(&dca_lock);
+
+	dca = dca_find_provider_by_dev(dev);
+	if (!dca) {
+		spin_unlock(&dca_lock);
 		return -ENODEV;
-	return global_dca->ops->get_tag(global_dca, cpu);
+	}
+	tag = dca->ops->get_tag(dca, dev, cpu);
+
+	spin_unlock(&dca_lock);
+	return tag;
+}
+
+/**
+ * dca3_get_tag - return the dca tag to the requester device
+ *                for the given cpu (new api)
+ * @dev - the device that wants dca service
+ * @cpu - the cpuid as returned by get_cpu()
+ */
+u8 dca3_get_tag(struct device *dev, int cpu)
+{
+	if (!dev)
+		return -EFAULT;
+
+	return dca_common_get_tag(dev, cpu);
+}
+EXPORT_SYMBOL_GPL(dca3_get_tag);
+
+/**
+ * dca_get_tag - return the dca tag for the given cpu (old api)
+ * @cpu - the cpuid as returned by get_cpu()
+ */
+u8 dca_get_tag(int cpu)
+{
+	struct device *dev = NULL;
+
+	return dca_common_get_tag(dev, cpu);
 }
 EXPORT_SYMBOL_GPL(dca_get_tag);
 
@@ -140,12 +218,10 @@ int register_dca_provider(struct dca_provider *dca, struct device *dev)
 {
 	int err;
 
-	if (global_dca)
-		return -EEXIST;
 	err = dca_sysfs_add_provider(dca, dev);
 	if (err)
 		return err;
-	global_dca = dca;
+	list_add(&dca->node, &dca_providers);
 	blocking_notifier_call_chain(&dca_provider_chain,
 				     DCA_PROVIDER_ADD, NULL);
 	return 0;
@@ -158,11 +234,9 @@ EXPORT_SYMBOL_GPL(register_dca_provider);
  */
 void unregister_dca_provider(struct dca_provider *dca)
 {
-	if (!global_dca)
-		return;
 	blocking_notifier_call_chain(&dca_provider_chain,
 				     DCA_PROVIDER_REMOVE, NULL);
-	global_dca = NULL;
+	list_del(&dca->node);
 	dca_sysfs_remove_provider(dca);
 }
 EXPORT_SYMBOL_GPL(unregister_dca_provider);
@@ -187,6 +261,7 @@ EXPORT_SYMBOL_GPL(dca_unregister_notify);
 
 static int __init dca_init(void)
 {
+	printk(KERN_ERR "dca service started, version %s\n", DCA_VERSION);
 	return dca_sysfs_init();
 }
 
diff --git a/drivers/dca/dca-sysfs.c b/drivers/dca/dca-sysfs.c
index 011328faa5f2..3d47e9d8e34f 100644
--- a/drivers/dca/dca-sysfs.c
+++ b/drivers/dca/dca-sysfs.c
@@ -13,9 +13,10 @@ static spinlock_t dca_idr_lock;
 int dca_sysfs_add_req(struct dca_provider *dca, struct device *dev, int slot)
 {
 	struct device *cd;
+	static int req_count;
 
 	cd = device_create(dca_class, dca->cd, MKDEV(0, slot + 1),
-			   "requester%d", slot);
+			   "requester%d", req_count++);
 	if (IS_ERR(cd))
 		return PTR_ERR(cd);
 	return 0;
diff --git a/drivers/dma/ioat.c b/drivers/dma/ioat.c
index 16e0fd8facfb..9b16a3af9a0a 100644
--- a/drivers/dma/ioat.c
+++ b/drivers/dma/ioat.c
@@ -47,6 +47,16 @@ static struct pci_device_id ioat_pci_tbl[] = {
 
 	/* I/OAT v2 platforms */
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB) },
+
+	/* I/OAT v3 platforms */
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG0) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG1) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG2) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG3) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG4) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG5) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG6) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG7) },
 	{ 0, }
 };
 
@@ -83,6 +93,11 @@ static int ioat_setup_functionality(struct pci_dev *pdev, void __iomem *iobase)
 		if (device->dma && ioat_dca_enabled)
 			device->dca = ioat2_dca_init(pdev, iobase);
 		break;
+	case IOAT_VER_3_0:
+		device->dma = ioat_dma_probe(pdev, iobase);
+		if (device->dma && ioat_dca_enabled)
+			device->dca = ioat3_dca_init(pdev, iobase);
+		break;
 	default:
 		err = -ENODEV;
 		break;
diff --git a/drivers/dma/ioat_dca.c b/drivers/dma/ioat_dca.c
index 9e922760b7ff..6cf622da0286 100644
--- a/drivers/dma/ioat_dca.c
+++ b/drivers/dma/ioat_dca.c
@@ -37,12 +37,18 @@
 #include "ioatdma_registers.h"
 
 /*
- * Bit 16 of a tag map entry is the "valid" bit, if it is set then bits 0:15
+ * Bit 7 of a tag map entry is the "valid" bit, if it is set then bits 0:6
  * contain the bit number of the APIC ID to map into the DCA tag.  If the valid
  * bit is not set, then the value must be 0 or 1 and defines the bit in the tag.
  */
 #define DCA_TAG_MAP_VALID 0x80
 
+#define DCA3_TAG_MAP_BIT_TO_INV 0x80
+#define DCA3_TAG_MAP_BIT_TO_SEL 0x40
+#define DCA3_TAG_MAP_LITERAL_VAL 0x1
+
+#define DCA_TAG_MAP_MASK 0xDF
+
 /*
  * "Legacy" DCA systems do not implement the DCA register set in the
  * I/OAT device.  Software needs direct support for their tag mappings.
@@ -95,6 +101,7 @@ struct ioat_dca_slot {
 };
 
 #define IOAT_DCA_MAX_REQ 6
+#define IOAT3_DCA_MAX_REQ 2
 
 struct ioat_dca_priv {
 	void __iomem		*iobase;
@@ -171,7 +178,9 @@ static int ioat_dca_remove_requester(struct dca_provider *dca,
 	return -ENODEV;
 }
 
-static u8 ioat_dca_get_tag(struct dca_provider *dca, int cpu)
+static u8 ioat_dca_get_tag(struct dca_provider *dca,
+			   struct device *dev,
+			   int cpu)
 {
 	struct ioat_dca_priv *ioatdca = dca_priv(dca);
 	int i, apic_id, bit, value;
@@ -193,10 +202,26 @@ static u8 ioat_dca_get_tag(struct dca_provider *dca, int cpu)
 	return tag;
 }
 
+static int ioat_dca_dev_managed(struct dca_provider *dca,
+				struct device *dev)
+{
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	struct pci_dev *pdev;
+	int i;
+
+	pdev = to_pci_dev(dev);
+	for (i = 0; i < ioatdca->max_requesters; i++) {
+		if (ioatdca->req_slots[i].pdev == pdev)
+			return 1;
+	}
+	return 0;
+}
+
 static struct dca_ops ioat_dca_ops = {
 	.add_requester		= ioat_dca_add_requester,
 	.remove_requester	= ioat_dca_remove_requester,
 	.get_tag		= ioat_dca_get_tag,
+	.dev_managed		= ioat_dca_dev_managed,
 };
 
 
@@ -207,6 +232,8 @@ struct dca_provider *ioat_dca_init(struct pci_dev *pdev, void __iomem *iobase)
 	u8 *tag_map = NULL;
 	int i;
 	int err;
+	u8 version;
+	u8 max_requesters;
 
 	if (!system_has_dca_enabled(pdev))
 		return NULL;
@@ -237,15 +264,20 @@ struct dca_provider *ioat_dca_init(struct pci_dev *pdev, void __iomem *iobase)
 	if (tag_map == NULL)
 		return NULL;
 
+	version = readb(iobase + IOAT_VER_OFFSET);
+	if (version == IOAT_VER_3_0)
+		max_requesters = IOAT3_DCA_MAX_REQ;
+	else
+		max_requesters = IOAT_DCA_MAX_REQ;
+
 	dca = alloc_dca_provider(&ioat_dca_ops,
 			sizeof(*ioatdca) +
-			(sizeof(struct ioat_dca_slot) * IOAT_DCA_MAX_REQ));
+			(sizeof(struct ioat_dca_slot) * max_requesters));
 	if (!dca)
 		return NULL;
 
 	ioatdca = dca_priv(dca);
-	ioatdca->max_requesters = IOAT_DCA_MAX_REQ;
-
+	ioatdca->max_requesters = max_requesters;
 	ioatdca->dca_base = iobase + 0x54;
 
 	/* copy over the APIC ID to DCA tag mapping */
@@ -323,11 +355,13 @@ static int ioat2_dca_remove_requester(struct dca_provider *dca,
 	return -ENODEV;
 }
 
-static u8 ioat2_dca_get_tag(struct dca_provider *dca, int cpu)
+static u8 ioat2_dca_get_tag(struct dca_provider *dca,
+			    struct device *dev,
+			    int cpu)
 {
 	u8 tag;
 
-	tag = ioat_dca_get_tag(dca, cpu);
+	tag = ioat_dca_get_tag(dca, dev, cpu);
 	tag = (~tag) & 0x1F;
 	return tag;
 }
@@ -336,6 +370,7 @@ static struct dca_ops ioat2_dca_ops = {
 	.add_requester		= ioat2_dca_add_requester,
 	.remove_requester	= ioat2_dca_remove_requester,
 	.get_tag		= ioat2_dca_get_tag,
+	.dev_managed		= ioat_dca_dev_managed,
 };
 
 static int ioat2_dca_count_dca_slots(void __iomem *iobase, u16 dca_offset)
@@ -425,3 +460,198 @@ struct dca_provider *ioat2_dca_init(struct pci_dev *pdev, void __iomem *iobase)
 
 	return dca;
 }
+
+static int ioat3_dca_add_requester(struct dca_provider *dca, struct device *dev)
+{
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	struct pci_dev *pdev;
+	int i;
+	u16 id;
+	u16 global_req_table;
+
+	/* This implementation only supports PCI-Express */
+	if (dev->bus != &pci_bus_type)
+		return -ENODEV;
+	pdev = to_pci_dev(dev);
+	id = dcaid_from_pcidev(pdev);
+
+	if (ioatdca->requester_count == ioatdca->max_requesters)
+		return -ENODEV;
+
+	for (i = 0; i < ioatdca->max_requesters; i++) {
+		if (ioatdca->req_slots[i].pdev == NULL) {
+			/* found an empty slot */
+			ioatdca->requester_count++;
+			ioatdca->req_slots[i].pdev = pdev;
+			ioatdca->req_slots[i].rid = id;
+			global_req_table =
+			      readw(ioatdca->dca_base + IOAT3_DCA_GREQID_OFFSET);
+			writel(id | IOAT_DCA_GREQID_VALID,
+			       ioatdca->iobase + global_req_table + (i * 4));
+			return i;
+		}
+	}
+	/* Error, ioatdma->requester_count is out of whack */
+	return -EFAULT;
+}
+
+static int ioat3_dca_remove_requester(struct dca_provider *dca,
+				      struct device *dev)
+{
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	struct pci_dev *pdev;
+	int i;
+	u16 global_req_table;
+
+	/* This implementation only supports PCI-Express */
+	if (dev->bus != &pci_bus_type)
+		return -ENODEV;
+	pdev = to_pci_dev(dev);
+
+	for (i = 0; i < ioatdca->max_requesters; i++) {
+		if (ioatdca->req_slots[i].pdev == pdev) {
+			global_req_table =
+			      readw(ioatdca->dca_base + IOAT3_DCA_GREQID_OFFSET);
+			writel(0, ioatdca->iobase + global_req_table + (i * 4));
+			ioatdca->req_slots[i].pdev = NULL;
+			ioatdca->req_slots[i].rid = 0;
+			ioatdca->requester_count--;
+			return i;
+		}
+	}
+	return -ENODEV;
+}
+
+static u8 ioat3_dca_get_tag(struct dca_provider *dca,
+			    struct device *dev,
+			    int cpu)
+{
+	u8 tag;
+
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	int i, apic_id, bit, value;
+	u8 entry;
+
+	tag = 0;
+	apic_id = cpu_physical_id(cpu);
+
+	for (i = 0; i < IOAT_TAG_MAP_LEN; i++) {
+		entry = ioatdca->tag_map[i];
+		if (entry & DCA3_TAG_MAP_BIT_TO_SEL) {
+			bit = entry &
+				~(DCA3_TAG_MAP_BIT_TO_SEL | DCA3_TAG_MAP_BIT_TO_INV);
+			value = (apic_id & (1 << bit)) ? 1 : 0;
+		} else if (entry & DCA3_TAG_MAP_BIT_TO_INV) {
+			bit = entry & ~DCA3_TAG_MAP_BIT_TO_INV;
+			value = (apic_id & (1 << bit)) ? 0 : 1;
+		} else {
+			value = (entry & DCA3_TAG_MAP_LITERAL_VAL) ? 1 : 0;
+		}
+		tag |= (value << i);
+	}
+
+	return tag;
+}
+
+static struct dca_ops ioat3_dca_ops = {
+	.add_requester		= ioat3_dca_add_requester,
+	.remove_requester	= ioat3_dca_remove_requester,
+	.get_tag		= ioat3_dca_get_tag,
+	.dev_managed		= ioat_dca_dev_managed,
+};
+
+static int ioat3_dca_count_dca_slots(void *iobase, u16 dca_offset)
+{
+	int slots = 0;
+	u32 req;
+	u16 global_req_table;
+
+	global_req_table = readw(iobase + dca_offset + IOAT3_DCA_GREQID_OFFSET);
+	if (global_req_table == 0)
+		return 0;
+
+	do {
+		req = readl(iobase + global_req_table + (slots * sizeof(u32)));
+		slots++;
+	} while ((req & IOAT_DCA_GREQID_LASTID) == 0);
+
+	return slots;
+}
+
+struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase)
+{
+	struct dca_provider *dca;
+	struct ioat_dca_priv *ioatdca;
+	int slots;
+	int i;
+	int err;
+	u16 dca_offset;
+	u16 csi_fsb_control;
+	u16 pcie_control;
+	u8 bit;
+
+	union {
+		u64 full;
+		struct {
+			u32 low;
+			u32 high;
+		};
+	} tag_map;
+
+	if (!system_has_dca_enabled(pdev))
+		return NULL;
+
+	dca_offset = readw(iobase + IOAT_DCAOFFSET_OFFSET);
+	if (dca_offset == 0)
+		return NULL;
+
+	slots = ioat3_dca_count_dca_slots(iobase, dca_offset);
+	if (slots == 0)
+		return NULL;
+
+	dca = alloc_dca_provider(&ioat3_dca_ops,
+				 sizeof(*ioatdca)
+				      + (sizeof(struct ioat_dca_slot) * slots));
+	if (!dca)
+		return NULL;
+
+	ioatdca = dca_priv(dca);
+	ioatdca->iobase = iobase;
+	ioatdca->dca_base = iobase + dca_offset;
+	ioatdca->max_requesters = slots;
+
+	/* some bios might not know to turn these on */
+	csi_fsb_control = readw(ioatdca->dca_base + IOAT3_CSI_CONTROL_OFFSET);
+	if ((csi_fsb_control & IOAT3_CSI_CONTROL_PREFETCH) == 0) {
+		csi_fsb_control |= IOAT3_CSI_CONTROL_PREFETCH;
+		writew(csi_fsb_control,
+		       ioatdca->dca_base + IOAT3_CSI_CONTROL_OFFSET);
+	}
+	pcie_control = readw(ioatdca->dca_base + IOAT3_PCI_CONTROL_OFFSET);
+	if ((pcie_control & IOAT3_PCI_CONTROL_MEMWR) == 0) {
+		pcie_control |= IOAT3_PCI_CONTROL_MEMWR;
+		writew(pcie_control,
+		       ioatdca->dca_base + IOAT3_PCI_CONTROL_OFFSET);
+	}
+
+
+	/* TODO version, compatibility and configuration checks */
+
+	/* copy out the APIC to DCA tag map */
+	tag_map.low =
+		readl(ioatdca->dca_base + IOAT3_APICID_TAG_MAP_OFFSET_LOW);
+	tag_map.high =
+		readl(ioatdca->dca_base + IOAT3_APICID_TAG_MAP_OFFSET_HIGH);
+	for (i = 0; i < 8; i++) {
+		bit = tag_map.full >> (8 * i);
+		ioatdca->tag_map[i] = bit & DCA_TAG_MAP_MASK;
+	}
+
+	err = register_dca_provider(dca, &pdev->dev);
+	if (err) {
+		free_dca_provider(dca);
+		return NULL;
+	}
+
+	return dca;
+}
diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
index ece5a0e3a335..a52156e56886 100644
--- a/drivers/dma/ioat_dma.c
+++ b/drivers/dma/ioat_dma.c
@@ -53,6 +53,12 @@ MODULE_PARM_DESC(ioat_pending_level,
 static void ioat_dma_chan_reset_part2(struct work_struct *work);
 static void ioat_dma_chan_watchdog(struct work_struct *work);
 
+/*
+ * workaround for IOAT ver.3.0 null descriptor issue
+ * (channel returns error when size is 0)
+ */
+#define NULL_DESC_BUFFER_SIZE 1
+
 /* internal functions */
 static void ioat_dma_start_null_desc(struct ioat_dma_chan *ioat_chan);
 static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan);
@@ -129,6 +135,38 @@ static int ioat_dma_enumerate_channels(struct ioatdma_device *device)
 	int i;
 	struct ioat_dma_chan *ioat_chan;
 
+	/*
+	 * IOAT ver.3 workarounds
+	 */
+	if (device->version == IOAT_VER_3_0) {
+		u32 chan_err_mask;
+		u16 dev_id;
+		u32 dmauncerrsts;
+
+		/*
+		 * Write CHANERRMSK_INT with 3E07h to mask out the errors
+		 * that can cause stability issues for IOAT ver.3
+		 */
+		chan_err_mask = 0x3E07;
+		pci_write_config_dword(device->pdev,
+			IOAT_PCI_CHANERRMASK_INT_OFFSET,
+			chan_err_mask);
+
+		/*
+		 * Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
+		 * (workaround for spurious config parity error after restart)
+		 */
+		pci_read_config_word(device->pdev,
+			IOAT_PCI_DEVICE_ID_OFFSET,
+			&dev_id);
+		if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0) {
+			dmauncerrsts = 0x10;
+			pci_write_config_dword(device->pdev,
+				IOAT_PCI_DMAUNCERRSTS_OFFSET,
+				dmauncerrsts);
+		}
+	}
+
 	device->common.chancnt = readb(device->reg_base + IOAT_CHANCNT_OFFSET);
 	xfercap_scale = readb(device->reg_base + IOAT_XFERCAP_OFFSET);
 	xfercap = (xfercap_scale == 0 ? -1 : (1UL << xfercap_scale));
@@ -473,6 +511,13 @@ static dma_cookie_t ioat1_tx_submit(struct dma_async_tx_descriptor *tx)
 		prev = new;
 	} while (len && (new = ioat1_dma_get_next_descriptor(ioat_chan)));
 
+	if (!new) {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"tx submit failed\n");
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		return -ENOMEM;
+	}
+
 	hw->ctl = IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
 	if (new->async_tx.callback) {
 		hw->ctl |= IOAT_DMA_DESCRIPTOR_CTL_INT_GN;
@@ -558,7 +603,14 @@ static dma_cookie_t ioat2_tx_submit(struct dma_async_tx_descriptor *tx)
 		desc_count++;
 	} while (len && (new = ioat2_dma_get_next_descriptor(ioat_chan)));
 
-	hw->ctl = IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
+	if (!new) {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"tx submit failed\n");
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		return -ENOMEM;
+	}
+
+	hw->ctl |= IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
 	if (new->async_tx.callback) {
 		hw->ctl |= IOAT_DMA_DESCRIPTOR_CTL_INT_GN;
 		if (first != new) {
@@ -629,6 +681,7 @@ static struct ioat_desc_sw *ioat_dma_alloc_descriptor(
 		desc_sw->async_tx.tx_submit = ioat1_tx_submit;
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		desc_sw->async_tx.tx_submit = ioat2_tx_submit;
 		break;
 	}
@@ -779,6 +832,7 @@ static void ioat_dma_free_chan_resources(struct dma_chan *chan)
 		}
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		list_for_each_entry_safe(desc, _desc,
 					 ioat_chan->free_desc.next, node) {
 			list_del(&desc->node);
@@ -868,7 +922,8 @@ ioat2_dma_get_next_descriptor(struct ioat_dma_chan *ioat_chan)
 
 		/* set up the noop descriptor */
 		noop_desc = to_ioat_desc(ioat_chan->used_desc.next);
-		noop_desc->hw->size = 0;
+		/* set size to non-zero value (channel returns error when size is 0) */
+		noop_desc->hw->size = NULL_DESC_BUFFER_SIZE;
 		noop_desc->hw->ctl = IOAT_DMA_DESCRIPTOR_NUL;
 		noop_desc->hw->src_addr = 0;
 		noop_desc->hw->dst_addr = 0;
@@ -918,6 +973,7 @@ static struct ioat_desc_sw *ioat_dma_get_next_descriptor(
 		return ioat1_dma_get_next_descriptor(ioat_chan);
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		return ioat2_dma_get_next_descriptor(ioat_chan);
 		break;
 	}
@@ -1061,10 +1117,12 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 		 * perhaps we're stuck so hard that the watchdog can't go off?
 		 * try to catch it after 2 seconds
 		 */
-		if (time_after(jiffies,
-			       ioat_chan->last_completion_time + HZ*WATCHDOG_DELAY)) {
-			ioat_dma_chan_watchdog(&(ioat_chan->device->work.work));
-			ioat_chan->last_completion_time = jiffies;
+		if (ioat_chan->device->version != IOAT_VER_3_0) {
+			if (time_after(jiffies,
+				       ioat_chan->last_completion_time + HZ*WATCHDOG_DELAY)) {
+				ioat_dma_chan_watchdog(&(ioat_chan->device->work.work));
+				ioat_chan->last_completion_time = jiffies;
+			}
 		}
 		return;
 	}
@@ -1120,6 +1178,7 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 		}
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		/* has some other thread has already cleaned up? */
 		if (ioat_chan->used_desc.prev == NULL)
 			break;
@@ -1223,10 +1282,19 @@ static void ioat_dma_start_null_desc(struct ioat_dma_chan *ioat_chan)
 	spin_lock_bh(&ioat_chan->desc_lock);
 
 	desc = ioat_dma_get_next_descriptor(ioat_chan);
+
+	if (!desc) {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"Unable to start null desc - get next desc failed\n");
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		return;
+	}
+
 	desc->hw->ctl = IOAT_DMA_DESCRIPTOR_NUL
 				| IOAT_DMA_DESCRIPTOR_CTL_INT_GN
 				| IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
-	desc->hw->size = 0;
+	/* set size to non-zero value (channel returns error when size is 0) */
+	desc->hw->size = NULL_DESC_BUFFER_SIZE;
 	desc->hw->src_addr = 0;
 	desc->hw->dst_addr = 0;
 	async_tx_ack(&desc->async_tx);
@@ -1244,6 +1312,7 @@ static void ioat_dma_start_null_desc(struct ioat_dma_chan *ioat_chan)
 			+ IOAT_CHANCMD_OFFSET(ioat_chan->device->version));
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		writel(((u64) desc->async_tx.phys) & 0x00000000FFFFFFFF,
 		       ioat_chan->reg_base + IOAT2_CHAINADDR_OFFSET_LOW);
 		writel(((u64) desc->async_tx.phys) >> 32,
@@ -1562,6 +1631,7 @@ struct ioatdma_device *ioat_dma_probe(struct pci_dev *pdev,
 						ioat1_dma_memcpy_issue_pending;
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		device->common.device_prep_dma_memcpy = ioat2_dma_prep_memcpy;
 		device->common.device_issue_pending =
 						ioat2_dma_memcpy_issue_pending;
@@ -1585,9 +1655,11 @@ struct ioatdma_device *ioat_dma_probe(struct pci_dev *pdev,
 
 	dma_async_device_register(&device->common);
 
-	INIT_DELAYED_WORK(&device->work, ioat_dma_chan_watchdog);
-	schedule_delayed_work(&device->work,
-			      WATCHDOG_DELAY);
+	if (device->version != IOAT_VER_3_0) {
+		INIT_DELAYED_WORK(&device->work, ioat_dma_chan_watchdog);
+		schedule_delayed_work(&device->work,
+				      WATCHDOG_DELAY);
+	}
 
 	return device;
 
@@ -1621,7 +1693,9 @@ void ioat_dma_remove(struct ioatdma_device *device)
 	pci_release_regions(device->pdev);
 	pci_disable_device(device->pdev);
 
-	cancel_delayed_work(&device->work);
+	if (device->version != IOAT_VER_3_0) {
+		cancel_delayed_work(&device->work);
+	}
 
 	list_for_each_entry_safe(chan, _chan,
 				 &device->common.channels, device_node) {
diff --git a/drivers/dma/ioatdma.h b/drivers/dma/ioatdma.h
index 685adb62aa5a..a3306d0e1372 100644
--- a/drivers/dma/ioatdma.h
+++ b/drivers/dma/ioatdma.h
@@ -29,7 +29,7 @@
 #include <linux/pci_ids.h>
 #include <net/tcp.h>
 
-#define IOAT_DMA_VERSION  "2.18"
+#define IOAT_DMA_VERSION  "3.30"
 
 enum ioat_interrupt {
 	none = 0,
@@ -135,6 +135,7 @@ static inline void ioat_set_tcp_copy_break(struct ioatdma_device *dev)
 	#ifdef CONFIG_NET_DMA
 	switch (dev->version) {
 	case IOAT_VER_1_2:
+	case IOAT_VER_3_0:
 		sysctl_tcp_dma_copybreak = 4096;
 		break;
 	case IOAT_VER_2_0:
@@ -150,11 +151,13 @@ struct ioatdma_device *ioat_dma_probe(struct pci_dev *pdev,
 void ioat_dma_remove(struct ioatdma_device *device);
 struct dca_provider *ioat_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 struct dca_provider *ioat2_dca_init(struct pci_dev *pdev, void __iomem *iobase);
+struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 #else
 #define ioat_dma_probe(pdev, iobase)    NULL
 #define ioat_dma_remove(device)         do { } while (0)
 #define ioat_dca_init(pdev, iobase)	NULL
 #define ioat2_dca_init(pdev, iobase)	NULL
+#define ioat3_dca_init(pdev, iobase)	NULL
 #endif
 
 #endif /* IOATDMA_H */
diff --git a/drivers/dma/ioatdma_hw.h b/drivers/dma/ioatdma_hw.h
index dd470fa91d86..f1ae2c776f74 100644
--- a/drivers/dma/ioatdma_hw.h
+++ b/drivers/dma/ioatdma_hw.h
@@ -35,6 +35,7 @@
 #define IOAT_PCI_SID            0x8086
 #define IOAT_VER_1_2            0x12    /* Version 1.2 */
 #define IOAT_VER_2_0            0x20    /* Version 2.0 */
+#define IOAT_VER_3_0            0x30    /* Version 3.0 */
 
 struct ioat_dma_descriptor {
 	uint32_t	size;
diff --git a/drivers/dma/ioatdma_registers.h b/drivers/dma/ioatdma_registers.h
index 9832d7ebd931..827cb503cac6 100644
--- a/drivers/dma/ioatdma_registers.h
+++ b/drivers/dma/ioatdma_registers.h
@@ -25,6 +25,10 @@
 #define IOAT_PCI_DMACTRL_DMA_EN			0x00000001
 #define IOAT_PCI_DMACTRL_MSI_EN			0x00000002
 
+#define IOAT_PCI_DEVICE_ID_OFFSET		0x02
+#define IOAT_PCI_DMAUNCERRSTS_OFFSET		0x148
+#define IOAT_PCI_CHANERRMASK_INT_OFFSET		0x184
+
 /* MMIO Device Registers */
 #define IOAT_CHANCNT_OFFSET			0x00	/*  8-bit */
 
@@ -149,7 +153,23 @@
 #define IOAT_DCA_GREQID_VALID       0x20000000
 #define IOAT_DCA_GREQID_LASTID      0x80000000
 
+#define IOAT3_CSI_CAPABILITY_OFFSET 0x08
+#define IOAT3_CSI_CAPABILITY_PREFETCH    0x1
+
+#define IOAT3_PCI_CAPABILITY_OFFSET 0x0A
+#define IOAT3_PCI_CAPABILITY_MEMWR  0x1
+
+#define IOAT3_CSI_CONTROL_OFFSET    0x0C
+#define IOAT3_CSI_CONTROL_PREFETCH  0x1
+
+#define IOAT3_PCI_CONTROL_OFFSET    0x0E
+#define IOAT3_PCI_CONTROL_MEMWR     0x1
+
+#define IOAT3_APICID_TAG_MAP_OFFSET 0x10
+#define IOAT3_APICID_TAG_MAP_OFFSET_LOW  0x10
+#define IOAT3_APICID_TAG_MAP_OFFSET_HIGH 0x14
 
+#define IOAT3_DCA_GREQID_OFFSET     0x02
 
 #define IOAT1_CHAINADDR_OFFSET		0x0C	/* 64-bit Descriptor Chain Address Register */
 #define IOAT2_CHAINADDR_OFFSET		0x10	/* 64-bit Descriptor Chain Address Register */
diff --git a/include/linux/dca.h b/include/linux/dca.h
index af61cd1f37e9..b00a753eda53 100644
--- a/include/linux/dca.h
+++ b/include/linux/dca.h
@@ -10,6 +10,7 @@ void dca_unregister_notify(struct notifier_block *nb);
 #define DCA_PROVIDER_REMOVE  0x0002
 
 struct dca_provider {
+	struct list_head	node;
 	struct dca_ops		*ops;
 	struct device 		*cd;
 	int			 id;
@@ -18,7 +19,9 @@ struct dca_provider {
 struct dca_ops {
 	int	(*add_requester)    (struct dca_provider *, struct device *);
 	int	(*remove_requester) (struct dca_provider *, struct device *);
-	u8	(*get_tag)	    (struct dca_provider *, int cpu);
+	u8	(*get_tag)	    (struct dca_provider *, struct device *,
+				     int cpu);
+	int	(*dev_managed)      (struct dca_provider *, struct device *);
 };
 
 struct dca_provider *alloc_dca_provider(struct dca_ops *ops, int priv_size);
@@ -32,9 +35,11 @@ static inline void *dca_priv(struct dca_provider *dca)
 }
 
 /* Requester API */
+#define DCA_GET_TAG_TWO_ARGS
 int dca_add_requester(struct device *dev);
 int dca_remove_requester(struct device *dev);
 u8 dca_get_tag(int cpu);
+u8 dca3_get_tag(struct device *dev, int cpu);
 
 /* internal stuff */
 int __init dca_sysfs_init(void);
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 9b940e644179..06a5b7ae79be 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2363,6 +2363,14 @@
 #define PCI_DEVICE_ID_INTEL_ICH9_7	0x2916
 #define PCI_DEVICE_ID_INTEL_ICH9_8	0x2918
 #define PCI_DEVICE_ID_INTEL_82855PM_HB	0x3340
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG4	0x3429
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG5	0x342a
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG6	0x342b
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG7	0x342c
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG0	0x3430
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG1	0x3431
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG2	0x3432
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG3	0x3433
 #define PCI_DEVICE_ID_INTEL_82830_HB	0x3575
 #define PCI_DEVICE_ID_INTEL_82830_CGC	0x3577
 #define PCI_DEVICE_ID_INTEL_82855GM_HB	0x3580
-- 
cgit v1.2.3


From 137d3edb48425f82a6a4226b664f90ed5e42eea5 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Sat, 19 Jul 2008 23:03:35 +0900
Subject: sg: reimplement sg mapping iterator

This is alternative implementation of sg content iterator introduced
by commit 83e7d317... from Pierre Ossman in next-20080716.  As there's
already an sg iterator which iterates over sg entries themselves, name
this sg_mapping_iterator.

Slightly edited description from the original implementation follows.

Iteration over a sg list is not that trivial when you take into
account that memory pages might have to be mapped before being used.
Unfortunately, that means that some parts of the kernel restrict
themselves to directly accesible memory just to not have to deal with
the mess.

This patch adds a simple iterator system that allows any code to
easily traverse an sg list and not have to deal with all the details.
The user can decide to consume part of the iteration.  Also, iteration
can be stopped and resumed later if releasing the kmap between
iteration steps is necessary.  These features are useful to implement
piecemeal sg copying for interrupt drive PIO for example.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 include/linux/scatterlist.h |  38 ++++++++++
 lib/scatterlist.c           | 176 ++++++++++++++++++++++++++++++++------------
 2 files changed, 168 insertions(+), 46 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h
index 71fc81360048..e5996984ddd0 100644
--- a/include/linux/scatterlist.h
+++ b/include/linux/scatterlist.h
@@ -224,4 +224,42 @@ size_t sg_copy_to_buffer(struct scatterlist *sgl, unsigned int nents,
  */
 #define SG_MAX_SINGLE_ALLOC		(PAGE_SIZE / sizeof(struct scatterlist))
 
+
+/*
+ * Mapping sg iterator
+ *
+ * Iterates over sg entries mapping page-by-page.  On each successful
+ * iteration, @miter->page points to the mapped page and
+ * @miter->length bytes of data can be accessed at @miter->addr.  As
+ * long as an interation is enclosed between start and stop, the user
+ * is free to choose control structure and when to stop.
+ *
+ * @miter->consumed is set to @miter->length on each iteration.  It
+ * can be adjusted if the user can't consume all the bytes in one go.
+ * Also, a stopped iteration can be resumed by calling next on it.
+ * This is useful when iteration needs to release all resources and
+ * continue later (e.g. at the next interrupt).
+ */
+
+#define SG_MITER_ATOMIC		(1 << 0)	 /* use kmap_atomic */
+
+struct sg_mapping_iter {
+	/* the following three fields can be accessed directly */
+	struct page		*page;		/* currently mapped page */
+	void			*addr;		/* pointer to the mapped area */
+	size_t			length;		/* length of the mapped area */
+	size_t			consumed;	/* number of consumed bytes */
+
+	/* these are internal states, keep away */
+	struct scatterlist	*__sg;		/* current entry */
+	unsigned int		__nents;	/* nr of remaining entries */
+	unsigned int		__offset;	/* offset within sg */
+	unsigned int		__flags;
+};
+
+void sg_miter_start(struct sg_mapping_iter *miter, struct scatterlist *sgl,
+		    unsigned int nents, unsigned int flags);
+bool sg_miter_next(struct sg_mapping_iter *miter);
+void sg_miter_stop(struct sg_mapping_iter *miter);
+
 #endif /* _LINUX_SCATTERLIST_H */
diff --git a/lib/scatterlist.c b/lib/scatterlist.c
index b80c21100d78..876ba6d5b670 100644
--- a/lib/scatterlist.c
+++ b/lib/scatterlist.c
@@ -294,6 +294,117 @@ int sg_alloc_table(struct sg_table *table, unsigned int nents, gfp_t gfp_mask)
 }
 EXPORT_SYMBOL(sg_alloc_table);
 
+/**
+ * sg_miter_start - start mapping iteration over a sg list
+ * @miter: sg mapping iter to be started
+ * @sgl: sg list to iterate over
+ * @nents: number of sg entries
+ *
+ * Description:
+ *   Starts mapping iterator @miter.
+ *
+ * Context:
+ *   Don't care.
+ */
+void sg_miter_start(struct sg_mapping_iter *miter, struct scatterlist *sgl,
+		    unsigned int nents, unsigned int flags)
+{
+	memset(miter, 0, sizeof(struct sg_mapping_iter));
+
+	miter->__sg = sgl;
+	miter->__nents = nents;
+	miter->__offset = 0;
+	miter->__flags = flags;
+}
+EXPORT_SYMBOL(sg_miter_start);
+
+/**
+ * sg_miter_next - proceed mapping iterator to the next mapping
+ * @miter: sg mapping iter to proceed
+ *
+ * Description:
+ *   Proceeds @miter@ to the next mapping.  @miter@ should have been
+ *   started using sg_miter_start().  On successful return,
+ *   @miter@->page, @miter@->addr and @miter@->length point to the
+ *   current mapping.
+ *
+ * Context:
+ *   IRQ disabled if SG_MITER_ATOMIC.  IRQ must stay disabled till
+ *   @miter@ is stopped.  May sleep if !SG_MITER_ATOMIC.
+ *
+ * Returns:
+ *   true if @miter contains the next mapping.  false if end of sg
+ *   list is reached.
+ */
+bool sg_miter_next(struct sg_mapping_iter *miter)
+{
+	unsigned int off, len;
+
+	/* check for end and drop resources from the last iteration */
+	if (!miter->__nents)
+		return false;
+
+	sg_miter_stop(miter);
+
+	/* get to the next sg if necessary.  __offset is adjusted by stop */
+	if (miter->__offset == miter->__sg->length && --miter->__nents) {
+		miter->__sg = sg_next(miter->__sg);
+		miter->__offset = 0;
+	}
+
+	/* map the next page */
+	off = miter->__sg->offset + miter->__offset;
+	len = miter->__sg->length - miter->__offset;
+
+	miter->page = nth_page(sg_page(miter->__sg), off >> PAGE_SHIFT);
+	off &= ~PAGE_MASK;
+	miter->length = min_t(unsigned int, len, PAGE_SIZE - off);
+	miter->consumed = miter->length;
+
+	if (miter->__flags & SG_MITER_ATOMIC)
+		miter->addr = kmap_atomic(miter->page, KM_BIO_SRC_IRQ) + off;
+	else
+		miter->addr = kmap(miter->page) + off;
+
+	return true;
+}
+EXPORT_SYMBOL(sg_miter_next);
+
+/**
+ * sg_miter_stop - stop mapping iteration
+ * @miter: sg mapping iter to be stopped
+ *
+ * Description:
+ *   Stops mapping iterator @miter.  @miter should have been started
+ *   started using sg_miter_start().  A stopped iteration can be
+ *   resumed by calling sg_miter_next() on it.  This is useful when
+ *   resources (kmap) need to be released during iteration.
+ *
+ * Context:
+ *   IRQ disabled if the SG_MITER_ATOMIC is set.  Don't care otherwise.
+ */
+void sg_miter_stop(struct sg_mapping_iter *miter)
+{
+	WARN_ON(miter->consumed > miter->length);
+
+	/* drop resources from the last iteration */
+	if (miter->addr) {
+		miter->__offset += miter->consumed;
+
+		if (miter->__flags & SG_MITER_ATOMIC) {
+			WARN_ON(!irqs_disabled());
+			kunmap_atomic(miter->addr, KM_BIO_SRC_IRQ);
+		} else
+			kunmap(miter->addr);
+
+		miter->page = NULL;
+		miter->addr = NULL;
+		miter->length = 0;
+		miter->consumed = 0;
+	}
+}
+EXPORT_SYMBOL(sg_miter_stop);
+
 /**
  * sg_copy_buffer - Copy data between a linear buffer and an SG list
  * @sgl:		 The SG list
@@ -309,56 +420,29 @@ EXPORT_SYMBOL(sg_alloc_table);
 static size_t sg_copy_buffer(struct scatterlist *sgl, unsigned int nents,
 			     void *buf, size_t buflen, int to_buffer)
 {
-	struct scatterlist *sg;
-	size_t buf_off = 0;
-	int i;
-
-	WARN_ON(!irqs_disabled());
-
-	for_each_sg(sgl, sg, nents, i) {
-		struct page *page;
-		int n = 0;
-		unsigned int sg_off = sg->offset;
-		unsigned int sg_copy = sg->length;
-
-		if (sg_copy > buflen)
-			sg_copy = buflen;
-		buflen -= sg_copy;
-
-		while (sg_copy > 0) {
-			unsigned int page_copy;
-			void *p;
-
-			page_copy = PAGE_SIZE - sg_off;
-			if (page_copy > sg_copy)
-				page_copy = sg_copy;
-
-			page = nth_page(sg_page(sg), n);
-			p = kmap_atomic(page, KM_BIO_SRC_IRQ);
-
-			if (to_buffer)
-				memcpy(buf + buf_off, p + sg_off, page_copy);
-			else {
-				memcpy(p + sg_off, buf + buf_off, page_copy);
-				flush_kernel_dcache_page(page);
-			}
-
-			kunmap_atomic(p, KM_BIO_SRC_IRQ);
-
-			buf_off += page_copy;
-			sg_off += page_copy;
-			if (sg_off == PAGE_SIZE) {
-				sg_off = 0;
-				n++;
-			}
-			sg_copy -= page_copy;
+	unsigned int offset = 0;
+	struct sg_mapping_iter miter;
+
+	sg_miter_start(&miter, sgl, nents, SG_MITER_ATOMIC);
+
+	while (sg_miter_next(&miter) && offset < buflen) {
+		unsigned int len;
+
+		len = min(miter.length, buflen - offset);
+
+		if (to_buffer)
+			memcpy(buf + offset, miter.addr, len);
+		else {
+			memcpy(miter.addr, buf + offset, len);
+			flush_kernel_dcache_page(miter.page);
 		}
 
-		if (!buflen)
-			break;
+		offset += len;
 	}
 
-	return buf_off;
+	sg_miter_stop(&miter);
+
+	return offset;
 }
 
 /**
-- 
cgit v1.2.3


From f472f80034ae5d0be7d7196c7e65d0ca8890afb4 Mon Sep 17 00:00:00 2001
From: Rafi Rubin <rafi@seas.upenn.edu>
Date: Thu, 19 Jun 2008 11:51:04 +0200
Subject: HID: add n-trig digitizer usage

This adds a hid usage that is reported by the N-Trig digitizer in the Dell
Latitude XT screen.

Signed-off-by: Rafi Rubin <rafi@seas.upenn.edu>
Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 include/linux/hid.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/hid.h b/include/linux/hid.h
index fe56b86f2c67..ac4e678a04ed 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -512,7 +512,7 @@ struct hid_descriptor {
 
 /* Applications from HID Usage Tables 4/8/99 Version 1.1 */
 /* We ignore a few input applications that are not widely used */
-#define IS_INPUT_APPLICATION(a) (((a >= 0x00010000) && (a <= 0x00010008)) || (a == 0x00010080) || (a == 0x000c0001))
+#define IS_INPUT_APPLICATION(a) (((a >= 0x00010000) && (a <= 0x00010008)) || (a == 0x00010080) || (a == 0x000c0001) || (a == 0x000d0002))
 
 /* HID core API */
 
-- 
cgit v1.2.3


From c97c6aca75fd5f718056fde7cff798b8cbdb07c0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:50 +0200
Subject: ide: pass hw_regs_t-s to ide_device_add[_all]() (take 3)

* Add 'hw_regs_t **hws' argument to ide_device_add[_all]() and convert
  host drivers + ide_legacy_init_one() + ide_setup_pci_device[s]() to use
  it instead of calling ide_init_port_hw() directly.

  [ However if host has > 1 port we must still set hwif->chipset to hint
    consecutive ide_find_port() call that the previous slot is occupied. ]

* Unexport ide_init_port_hw().

v2:
* Use defines instead of hard-coded values in buddha.c, gayle.c and q40ide.c.
  (Suggested by Geert Uytterhoeven)

* Better patch description.

v3:
* Fix build problem in ide-cs.c. (Noticed by Stephen Rothwell)

There should be no functional changes caused by this patch.

Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c          | 13 ++++++-------
 drivers/ide/arm/ide_arm.c         |  5 ++---
 drivers/ide/arm/palm_bk3710.c     |  6 ++----
 drivers/ide/arm/rapide.c          |  5 ++---
 drivers/ide/h8300/ide-h8300.c     |  5 ++---
 drivers/ide/ide-generic.c         | 23 ++++++++++++-----------
 drivers/ide/ide-pnp.c             |  6 ++----
 drivers/ide/ide-probe.c           | 29 ++++++++++++++++++-----------
 drivers/ide/ide.c                 |  1 -
 drivers/ide/legacy/buddha.c       | 16 +++++++++-------
 drivers/ide/legacy/falconide.c    |  6 ++----
 drivers/ide/legacy/gayle.c        | 13 ++++++-------
 drivers/ide/legacy/ide-4drives.c  | 10 ++++++----
 drivers/ide/legacy/ide-cs.c       | 10 +++-------
 drivers/ide/legacy/ide_platform.c |  9 +++------
 drivers/ide/legacy/macide.c       |  6 ++----
 drivers/ide/legacy/q40ide.c       | 14 ++++++--------
 drivers/ide/mips/au1xxx-ide.c     |  6 ++----
 drivers/ide/mips/swarm.c          |  6 ++----
 drivers/ide/pci/cmd640.c          | 10 ++++++----
 drivers/ide/pci/cs5520.c          |  5 +++--
 drivers/ide/pci/delkin_cb.c       |  6 ++----
 drivers/ide/pci/scc_pata.c        |  5 ++---
 drivers/ide/pci/sgiioc4.c         |  6 ++----
 drivers/ide/ppc/pmac.c            |  5 ++---
 drivers/ide/setup-pci.c           | 38 +++++++++++++++++++++++---------------
 include/linux/ide.h               |  7 ++++---
 27 files changed, 131 insertions(+), 140 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index 52f58c885783..850fe9342a1f 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -442,8 +442,8 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 {
 	ide_hwif_t *hwif;
 	void __iomem *base;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw;
 
 	base = ecardm_iomap(ec, ECARD_RES_MEMC, 0, 0);
 	if (!base)
@@ -467,7 +467,6 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 	if (!hwif)
 		return -ENODEV;
 
-	ide_init_port_hw(hwif, &hw);
 	default_hwif_mmiops(hwif);
 
 	state->hwif[0] = hwif;
@@ -476,7 +475,7 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, NULL);
+	ide_device_add(idx, NULL, hws);
 
 	return 0;
 }
@@ -497,9 +496,9 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 	void __iomem *ioc_base, *easi_base;
 	unsigned int sel = 0;
 	int ret;
+	hw_regs_t hw[2], *hws[] = { &hw[0], NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	struct ide_port_info d = icside_v6_port_info;
-	hw_regs_t hw[2];
 
 	ioc_base = ecardm_iomap(ec, ECARD_RES_IOCFAST, 0, 0);
 	if (!ioc_base) {
@@ -545,16 +544,16 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 	if (hwif == NULL)
 		return -ENODEV;
 
-	ide_init_port_hw(hwif, &hw[0]);
+	hwif->chipset = ide_acorn;
 	default_hwif_mmiops(hwif);
 
 	idx[0] = hwif->index;
 
 	mate = ide_find_port();
 	if (mate) {
-		ide_init_port_hw(mate, &hw[1]);
 		default_hwif_mmiops(mate);
 
+		hws[1] = &hw[1];
 		idx[1] = mate->index;
 	}
 
@@ -569,7 +568,7 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 		d.dma_ops = NULL;
 	}
 
-	ide_device_add(idx, &d);
+	ide_device_add(idx, &d, hws);
 
 	return 0;
 
diff --git a/drivers/ide/arm/ide_arm.c b/drivers/ide/arm/ide_arm.c
index 2f311da4c963..e9831bbd988a 100644
--- a/drivers/ide/arm/ide_arm.c
+++ b/drivers/ide/arm/ide_arm.c
@@ -29,8 +29,8 @@
 static int __init ide_arm_init(void)
 {
 	ide_hwif_t *hwif;
-	hw_regs_t hw;
 	unsigned long base = IDE_ARM_IO, ctl = IDE_ARM_IO + 0x206;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	if (!request_region(base, 8, DRV_NAME)) {
@@ -53,10 +53,9 @@ static int __init ide_arm_init(void)
 
 	hwif = ide_find_port();
 	if (hwif) {
-		ide_init_port_hw(hwif, &hw);
 		idx[0] = hwif->index;
 
-		ide_device_add(idx, NULL);
+		ide_device_add(idx, NULL, hws);
 	}
 
 	return 0;
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index c79b85b6e4a3..023c10753f15 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -351,7 +351,7 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	ide_hwif_t *hwif;
 	unsigned long base, rate;
 	int i;
-	hw_regs_t hw;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	clk = clk_get(NULL, "IDECLK");
@@ -400,13 +400,11 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 
 	i = hwif->index;
 
-	ide_init_port_hw(hwif, &hw);
-
 	default_hwif_mmiops(hwif);
 
 	idx[0] = i;
 
-	ide_device_add(idx, &palm_bk3710_port_info);
+	ide_device_add(idx, &palm_bk3710_port_info, hws);
 
 	return 0;
 out:
diff --git a/drivers/ide/arm/rapide.c b/drivers/ide/arm/rapide.c
index 43057e0303c8..01896f6e8acf 100644
--- a/drivers/ide/arm/rapide.c
+++ b/drivers/ide/arm/rapide.c
@@ -35,8 +35,8 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 	ide_hwif_t *hwif;
 	void __iomem *base;
 	int ret;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw;
 
 	ret = ecard_request_resources(ec);
 	if (ret)
@@ -59,12 +59,11 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 		goto release;
 	}
 
-	ide_init_port_hw(hwif, &hw);
 	default_hwif_mmiops(hwif);
 
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, &rapide_port_info);
+	ide_device_add(idx, &rapide_port_info, hws);
 
 	ecard_set_drvdata(ec, hwif);
 	goto out;
diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index 20fad6d542cc..a71433b3d0f3 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -184,9 +184,9 @@ static const struct ide_port_info h8300_port_info = {
 
 static int __init h8300_ide_init(void)
 {
-	hw_regs_t hw;
 	ide_hwif_t *hwif;
 	int index;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	printk(KERN_INFO DRV_NAME ": H8/300 generic IDE interface\n");
@@ -205,12 +205,11 @@ static int __init h8300_ide_init(void)
 		return -ENOENT;
 
 	index = hwif->index;
-	ide_init_port_hw(hwif, &hw);
 	hwif_setup(hwif);
 
 	idx[0] = index;
 
-	ide_device_add(idx, &h8300_port_info);
+	ide_device_add(idx, &h8300_port_info, hws);
 
 	return 0;
 
diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index 2d92214096ab..bb9fc905d60c 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -31,7 +31,7 @@ static ssize_t store_add(struct class *cls, const char *buf, size_t n)
 	ide_hwif_t *hwif;
 	unsigned int base, ctl;
 	int irq;
-	hw_regs_t hw;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[] = { 0xff, 0xff, 0xff, 0xff };
 
 	if (sscanf(buf, "%x:%x:%d", &base, &ctl, &irq) != 3)
@@ -46,11 +46,9 @@ static ssize_t store_add(struct class *cls, const char *buf, size_t n)
 	hw.irq = irq;
 	hw.chipset = ide_generic;
 
-	ide_init_port_hw(hwif, &hw);
-
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, NULL);
+	ide_device_add(idx, NULL, hws);
 
 	return n;
 };
@@ -90,6 +88,7 @@ static int __init ide_generic_sysfs_init(void)
 
 static int __init ide_generic_init(void)
 {
+	hw_regs_t hw[MAX_HWIFS], *hws[MAX_HWIFS];
 	u8 idx[MAX_HWIFS];
 	int i;
 
@@ -99,8 +98,8 @@ static int __init ide_generic_init(void)
 	for (i = 0; i < MAX_HWIFS; i++) {
 		ide_hwif_t *hwif;
 		unsigned long io_addr = ide_default_io_base(i);
-		hw_regs_t hw;
 
+		hws[i] = NULL;
 		idx[i] = 0xff;
 
 		if ((probe_mask & (1 << i)) && io_addr) {
@@ -129,17 +128,19 @@ static int __init ide_generic_init(void)
 				continue;
 			}
 
-			memset(&hw, 0, sizeof(hw));
-			ide_std_init_ports(&hw, io_addr, io_addr + 0x206);
-			hw.irq = ide_default_irq(io_addr);
-			hw.chipset = ide_generic;
-			ide_init_port_hw(hwif, &hw);
+			hwif->chipset = ide_generic;
+
+			memset(&hw[i], 0, sizeof(hw[i]));
+			ide_std_init_ports(&hw[i], io_addr, io_addr + 0x206);
+			hw[i].irq = ide_default_irq(io_addr);
+			hw[i].chipset = ide_generic;
 
+			hws[i] = &hw[i];
 			idx[i] = i;
 		}
 	}
 
-	ide_device_add_all(idx, NULL);
+	ide_device_add_all(idx, NULL, hws);
 
 	if (ide_generic_sysfs_init())
 		printk(KERN_ERR DRV_NAME ": failed to create ide_generic "
diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c
index 03f2ef5470a3..89cd5cbe8573 100644
--- a/drivers/ide/ide-pnp.c
+++ b/drivers/ide/ide-pnp.c
@@ -29,9 +29,9 @@ static struct pnp_device_id idepnp_devices[] = {
 
 static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id)
 {
-	hw_regs_t hw;
 	ide_hwif_t *hwif;
 	unsigned long base, ctl;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	printk(KERN_INFO DRV_NAME ": generic PnP IDE interface\n");
 
@@ -64,11 +64,9 @@ static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id)
 		u8 index = hwif->index;
 		u8 idx[4] = { index, 0xff, 0xff, 0xff };
 
-		ide_init_port_hw(hwif, &hw);
-
 		pnp_set_drvdata(dev, hwif);
 
-		ide_device_add(idx, NULL);
+		ide_device_add(idx, NULL, hws);
 
 		return 0;
 	}
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 235ebdb29b28..9d8686a49fcf 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1501,7 +1501,7 @@ out_found:
 }
 EXPORT_SYMBOL_GPL(ide_find_port_slot);
 
-int ide_device_add_all(u8 *idx, const struct ide_port_info *d)
+int ide_device_add_all(u8 *idx, const struct ide_port_info *d, hw_regs_t **hws)
 {
 	ide_hwif_t *hwif, *mate = NULL;
 	int i, rc = 0;
@@ -1514,6 +1514,7 @@ int ide_device_add_all(u8 *idx, const struct ide_port_info *d)
 
 		hwif = &ide_hwifs[idx[i]];
 
+		ide_init_port_hw(hwif, hws[i]);
 		ide_port_apply_params(hwif);
 
 		if (d == NULL) {
@@ -1603,15 +1604,18 @@ int ide_device_add_all(u8 *idx, const struct ide_port_info *d)
 }
 EXPORT_SYMBOL_GPL(ide_device_add_all);
 
-int ide_device_add(u8 idx[4], const struct ide_port_info *d)
+int ide_device_add(u8 *idx, const struct ide_port_info *d, hw_regs_t **hws)
 {
+	hw_regs_t *hws_all[MAX_HWIFS];
 	u8 idx_all[MAX_HWIFS];
 	int i;
 
-	for (i = 0; i < MAX_HWIFS; i++)
+	for (i = 0; i < MAX_HWIFS; i++) {
+		hws_all[i] = (i < 4) ? hws[i] : NULL;
 		idx_all[i] = (i < 4) ? idx[i] : 0xff;
+	}
 
-	return ide_device_add_all(idx_all, d);
+	return ide_device_add_all(idx_all, d, hws_all);
 }
 EXPORT_SYMBOL_GPL(ide_device_add);
 
@@ -1634,8 +1638,8 @@ void ide_port_scan(ide_hwif_t *hwif)
 }
 EXPORT_SYMBOL_GPL(ide_port_scan);
 
-static void ide_legacy_init_one(u8 *idx, hw_regs_t *hw, u8 port_no,
-				const struct ide_port_info *d,
+static void ide_legacy_init_one(u8 *idx, hw_regs_t **hws, hw_regs_t *hw,
+				u8 port_no, const struct ide_port_info *d,
 				unsigned long config)
 {
 	ide_hwif_t *hwif;
@@ -1671,9 +1675,12 @@ static void ide_legacy_init_one(u8 *idx, hw_regs_t *hw, u8 port_no,
 
 	hwif = ide_find_port_slot(d);
 	if (hwif) {
-		ide_init_port_hw(hwif, hw);
+		hwif->chipset = hw->chipset;
+
 		if (config)
 			hwif->config_data = config;
+
+		hws[port_no] = hw;
 		idx[port_no] = hwif->index;
 	}
 }
@@ -1681,19 +1688,19 @@ static void ide_legacy_init_one(u8 *idx, hw_regs_t *hw, u8 port_no,
 int ide_legacy_device_add(const struct ide_port_info *d, unsigned long config)
 {
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw[2];
+	hw_regs_t hw[2], *hws[] = { NULL, NULL, NULL, NULL };
 
 	memset(&hw, 0, sizeof(hw));
 
 	if ((d->host_flags & IDE_HFLAG_QD_2ND_PORT) == 0)
-		ide_legacy_init_one(idx, &hw[0], 0, d, config);
-	ide_legacy_init_one(idx, &hw[1], 1, d, config);
+		ide_legacy_init_one(idx, hws, &hw[0], 0, d, config);
+	ide_legacy_init_one(idx, hws, &hw[1], 1, d, config);
 
 	if (idx[0] == 0xff && idx[1] == 0xff &&
 	    (d->host_flags & IDE_HFLAG_SINGLE))
 		return -ENOENT;
 
-	ide_device_add(idx, d);
+	ide_device_add(idx, d, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index d4a6b102a772..b6018f7b0907 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -288,7 +288,6 @@ void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
 	hwif->gendev.parent = hw->parent ? hw->parent : hw->dev;
 	hwif->ack_intr = hw->ack_intr;
 }
-EXPORT_SYMBOL_GPL(ide_init_port_hw);
 
 /*
  *	Locks for IDE setting functionality
diff --git a/drivers/ide/legacy/buddha.c b/drivers/ide/legacy/buddha.c
index 0497e7f85b09..c61bc6a1db36 100644
--- a/drivers/ide/legacy/buddha.c
+++ b/drivers/ide/legacy/buddha.c
@@ -37,6 +37,8 @@
 #define CATWEASEL_NUM_HWIFS	3
 #define XSURF_NUM_HWIFS         2
 
+#define MAX_NUM_HWIFS		3
+
     /*
      *  Bases of the IDE interfaces (relative to the board address)
      */
@@ -148,7 +150,6 @@ static void __init buddha_setup_ports(hw_regs_t *hw, unsigned long base,
 
 static int __init buddha_init(void)
 {
-	hw_regs_t hw;
 	ide_hwif_t *hwif;
 	int i;
 
@@ -159,6 +160,7 @@ static int __init buddha_init(void)
 
 	while ((z = zorro_find_device(ZORRO_WILDCARD, z))) {
 		unsigned long board;
+		hw_regs_t hw[MAX_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
 		u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 		if (z->id == ZORRO_PROD_INDIVIDUAL_COMPUTERS_BUDDHA) {
@@ -221,19 +223,19 @@ fail_base2:
 				ack_intr = xsurf_ack_intr;
 			}
 
-			buddha_setup_ports(&hw, base, ctl, irq_port, ack_intr);
+			buddha_setup_ports(&hw[i], base, ctl, irq_port,
+					   ack_intr);
 
 			hwif = ide_find_port();
 			if (hwif) {
-				u8 index = hwif->index;
-
-				ide_init_port_hw(hwif, &hw);
+				hwif->chipset = ide_generic;
 
-				idx[i] = index;
+				hws[i] = &hw[i];
+				idx[i] = hwif->index;
 			}
 		}
 
-		ide_device_add(idx, NULL);
+		ide_device_add(idx, NULL, hws);
 	}
 
 	return 0;
diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c
index 129a812bb57f..1bb2aa72cc7f 100644
--- a/drivers/ide/legacy/falconide.c
+++ b/drivers/ide/legacy/falconide.c
@@ -91,8 +91,8 @@ static void __init falconide_setup_ports(hw_regs_t *hw)
 
 static int __init falconide_init(void)
 {
-	hw_regs_t hw;
 	ide_hwif_t *hwif;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	if (!MACH_IS_ATARI || !ATARIHW_PRESENT(IDE))
 		return 0;
@@ -111,14 +111,12 @@ static int __init falconide_init(void)
 		u8 index = hwif->index;
 		u8 idx[4] = { index, 0xff, 0xff, 0xff };
 
-		ide_init_port_hw(hwif, &hw);
-
 		/* Atari has a byte-swapped IDE interface */
 		hwif->input_data  = falconide_input_data;
 		hwif->output_data = falconide_output_data;
 
 		ide_get_lock(NULL, NULL);
-		ide_device_add(idx, NULL);
+		ide_device_add(idx, NULL, hws);
 		ide_release_lock();
 	}
 
diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index 7e74b20202df..e45c7341186c 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -125,6 +125,7 @@ static void __init gayle_setup_ports(hw_regs_t *hw, unsigned long base,
 static int __init gayle_init(void)
 {
     int a4000, i;
+    hw_regs_t hw[GAYLE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
     u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!MACH_IS_AMIGA)
@@ -151,7 +152,6 @@ found:
     for (i = 0; i < GAYLE_NUM_PROBE_HWIFS; i++) {
 	unsigned long base, ctrlport, irqport;
 	ide_ack_intr_t *ack_intr;
-	hw_regs_t hw;
 	ide_hwif_t *hwif;
 	unsigned long phys_base, res_start, res_n;
 
@@ -179,20 +179,19 @@ found:
 	base = (unsigned long)ZTWO_VADDR(phys_base);
 	ctrlport = GAYLE_HAS_CONTROL_REG ? (base + GAYLE_CONTROL) : 0;
 
-	gayle_setup_ports(&hw, base, ctrlport, irqport, ack_intr);
+	gayle_setup_ports(&hw[i], base, ctrlport, irqport, ack_intr);
 
 	hwif = ide_find_port();
 	if (hwif) {
-	    u8 index = hwif->index;
+	    hwif->chipset = ide_generic;
 
-	    ide_init_port_hw(hwif, &hw);
-
-	    idx[i] = index;
+	    hws[i] = &hw[i];
+	    idx[i] = hwif->index;
 	} else
 	    release_mem_region(res_start, res_n);
     }
 
-    ide_device_add(idx, NULL);
+    ide_device_add(idx, NULL, hws);
 
     return 0;
 }
diff --git a/drivers/ide/legacy/ide-4drives.c b/drivers/ide/legacy/ide-4drives.c
index 89c8ff0a4d08..6310dc50e3c5 100644
--- a/drivers/ide/legacy/ide-4drives.c
+++ b/drivers/ide/legacy/ide-4drives.c
@@ -30,8 +30,8 @@ static int __init ide_4drives_init(void)
 {
 	ide_hwif_t *hwif, *mate;
 	unsigned long base = 0x1f0, ctl = 0x3f6;
+	hw_regs_t hw, *hws[] = { NULL, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw;
 
 	if (probe_4drives == 0)
 		return -ENODEV;
@@ -57,17 +57,19 @@ static int __init ide_4drives_init(void)
 
 	hwif = ide_find_port();
 	if (hwif) {
-		ide_init_port_hw(hwif, &hw);
+		hwif->chipset = ide_4drives;
+
+		hws[0] = &hw;
 		idx[0] = hwif->index;
 	}
 
 	mate = ide_find_port();
 	if (mate) {
-		ide_init_port_hw(mate, &hw);
+		hws[1] = &hw;
 		idx[1] = mate->index;
 	}
 
-	ide_device_add(idx, &ide_4drives_port_info);
+	ide_device_add(idx, &ide_4drives_port_info, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index 27b1e0b7ecb4..f93d5454ebf8 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -161,8 +161,8 @@ static ide_hwif_t *idecs_register(unsigned long io, unsigned long ctl,
 				unsigned long irq, struct pcmcia_device *handle)
 {
     ide_hwif_t *hwif;
-    hw_regs_t hw;
     int i;
+    hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
     u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!request_region(io, 8, DRV_NAME)) {
@@ -188,13 +188,9 @@ static ide_hwif_t *idecs_register(unsigned long io, unsigned long ctl,
     if (hwif == NULL)
 	goto out_release;
 
-    i = hwif->index;
+    idx[0] = hwif->index;
 
-    ide_init_port_hw(hwif, &hw);
-
-    idx[0] = i;
-
-    ide_device_add(idx, &idecs_port_info);
+    ide_device_add(idx, &idecs_port_info, hws);
 
     if (hwif->present)
 	return hwif;
diff --git a/drivers/ide/legacy/ide_platform.c b/drivers/ide/legacy/ide_platform.c
index a249562b34b5..609da0d43196 100644
--- a/drivers/ide/legacy/ide_platform.c
+++ b/drivers/ide/legacy/ide_platform.c
@@ -54,10 +54,9 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 	void __iomem *base, *alt_base;
 	ide_hwif_t *hwif;
 	struct pata_platform_info *pdata;
+	int ret = 0, mmio = 0;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	int ret = 0;
-	int mmio = 0;
-	hw_regs_t hw;
 	struct ide_port_info d = platform_ide_port_info;
 
 	pdata = pdev->dev.platform_data;
@@ -104,8 +103,6 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 	plat_ide_setup_ports(&hw, base, alt_base, pdata, res_irq->start);
 	hw.dev = &pdev->dev;
 
-	ide_init_port_hw(hwif, &hw);
-
 	if (mmio) {
 		d.host_flags |= IDE_HFLAG_MMIO;
 		default_hwif_mmiops(hwif);
@@ -113,7 +110,7 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, &d);
+	ide_device_add(idx, &d, hws);
 
 	platform_set_drvdata(pdev, hwif);
 
diff --git a/drivers/ide/legacy/macide.c b/drivers/ide/legacy/macide.c
index 0a6195bcfeda..d839df2239fc 100644
--- a/drivers/ide/legacy/macide.c
+++ b/drivers/ide/legacy/macide.c
@@ -95,7 +95,7 @@ static int __init macide_init(void)
 	ide_ack_intr_t *ack_intr;
 	unsigned long base;
 	int irq;
-	hw_regs_t hw;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	if (!MACH_IS_MAC)
 		return -ENODEV;
@@ -130,9 +130,7 @@ static int __init macide_init(void)
 		u8 index = hwif->index;
 		u8 idx[4] = { index, 0xff, 0xff, 0xff };
 
-		ide_init_port_hw(hwif, &hw);
-
-		ide_device_add(idx, NULL);
+		ide_device_add(idx, NULL, hws);
 	}
 
 	return 0;
diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c
index 9c2b9d078f69..fcb04b8b0238 100644
--- a/drivers/ide/legacy/q40ide.c
+++ b/drivers/ide/legacy/q40ide.c
@@ -112,7 +112,7 @@ static int __init q40ide_init(void)
 {
     int i;
     ide_hwif_t *hwif;
-    const char *name;
+    hw_regs_t hw[Q40IDE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
     u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!MACH_IS_Q40)
@@ -121,9 +121,8 @@ static int __init q40ide_init(void)
     printk(KERN_INFO "ide: Q40 IDE controller\n");
 
     for (i = 0; i < Q40IDE_NUM_HWIFS; i++) {
-	hw_regs_t hw;
+	const char *name = q40_ide_names[i];
 
-	name = q40_ide_names[i];
 	if (!request_region(pcide_bases[i], 8, name)) {
 		printk("could not reserve ports %lx-%lx for %s\n",
 		       pcide_bases[i],pcide_bases[i]+8,name);
@@ -135,24 +134,23 @@ static int __init q40ide_init(void)
 		release_region(pcide_bases[i], 8);
 		continue;
 	}
-	q40_ide_setup_ports(&hw, pcide_bases[i],
-			NULL,
-//			m68kide_iops,
+	q40_ide_setup_ports(&hw[i], pcide_bases[i], NULL,
 			q40ide_default_irq(pcide_bases[i]));
 
 	hwif = ide_find_port();
 	if (hwif) {
-		ide_init_port_hw(hwif, &hw);
+		hwif->chipset = ide_generic;
 
 		/* Q40 has a byte-swapped IDE interface */
 		hwif->input_data  = q40ide_input_data;
 		hwif->output_data = q40ide_output_data;
 
+		hws[i] = &hw[i];
 		idx[i] = hwif->index;
 	}
     }
 
-    ide_device_add(idx, NULL);
+    ide_device_add(idx, NULL, hws);
 
     return 0;
 }
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 48d57cae63c6..475da582fd89 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -546,8 +546,8 @@ static int au_ide_probe(struct device *dev)
 	ide_hwif_t *hwif;
 	struct resource *res;
 	int ret = 0;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw;
 
 #if defined(CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA)
 	char *mode = "MWDMA2";
@@ -596,8 +596,6 @@ static int au_ide_probe(struct device *dev)
 	hw.dev = dev;
 	hw.chipset = ide_au1xxx;
 
-	ide_init_port_hw(hwif, &hw);
-
 	/* If the user has selected DDMA assisted copies,
 	   then set up a few local I/O function entry points 
 	*/
@@ -611,7 +609,7 @@ static int au_ide_probe(struct device *dev)
 
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, &au1xxx_port_info);
+	ide_device_add(idx, &au1xxx_port_info, hws);
 
 	dev_set_drvdata(dev, hwif);
 
diff --git a/drivers/ide/mips/swarm.c b/drivers/ide/mips/swarm.c
index 9f1212cc4aed..6da6844d2d8c 100644
--- a/drivers/ide/mips/swarm.c
+++ b/drivers/ide/mips/swarm.c
@@ -75,8 +75,8 @@ static int __devinit swarm_ide_probe(struct device *dev)
 	ide_hwif_t *hwif;
 	u8 __iomem *base;
 	phys_t offset, size;
-	hw_regs_t hw;
 	int i;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[] = { 0xff, 0xff, 0xff, 0xff };
 
 	if (!SIBYTE_HAVE_IDE)
@@ -120,14 +120,12 @@ static int __devinit swarm_ide_probe(struct device *dev)
 	if (hwif == NULL)
 		goto err;
 
-	ide_init_port_hw(hwif, &hw);
-
 	/* Setup MMIO ops. */
 	default_hwif_mmiops(hwif);
 
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, &swarm_port_info);
+	ide_device_add(idx, &swarm_port_info, hws);
 
 	dev_set_drvdata(dev, hwif);
 
diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c
index 1ad1e23e3105..ccde1e444e13 100644
--- a/drivers/ide/pci/cmd640.c
+++ b/drivers/ide/pci/cmd640.c
@@ -717,8 +717,8 @@ static int __init cmd640x_init(void)
 	int second_port_cmd640 = 0, rc;
 	const char *bus_type, *port2;
 	u8 b, cfr;
+	hw_regs_t hw[2], *hws[] = { NULL, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw[2];
 
 	if (cmd640_vlb && probe_for_cmd640_vlb()) {
 		bus_type = "VLB";
@@ -787,7 +787,9 @@ static int __init cmd640x_init(void)
 	 * Initialize data for primary port
 	 */
 	if (cmd_hwif0) {
-		ide_init_port_hw(cmd_hwif0, &hw[0]);
+		cmd_hwif0->chipset = ide_cmd640;
+
+		hws[0] = &hw[0];
 		idx[0] = cmd_hwif0->index;
 	}
 
@@ -832,7 +834,7 @@ static int __init cmd640x_init(void)
 	if (second_port_cmd640) {
 		cmd_hwif1 = ide_find_port();
 		if (cmd_hwif1) {
-			ide_init_port_hw(cmd_hwif1, &hw[1]);
+			hws[1] = &hw[1];
 			idx[1] = cmd_hwif1->index;
 		}
 	}
@@ -843,7 +845,7 @@ static int __init cmd640x_init(void)
 	cmd640_dump_regs();
 #endif
 
-	ide_device_add(idx, &cmd640_port_info);
+	ide_device_add(idx, &cmd640_port_info, hws);
 
 	return 1;
 }
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index 992b1cf8db69..a13f2bc0665c 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -146,6 +146,7 @@ static const struct ide_port_info cyrix_chipsets[] __devinitdata = {
 static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	const struct ide_port_info *d = &cyrix_chipsets[id->driver_data];
+	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	ide_setup_pci_noise(dev, d);
@@ -168,9 +169,9 @@ static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_devic
 	 *	do all the device setup for us
 	 */
 
-	ide_pci_setup_ports(dev, d, 14, &idx[0]);
+	ide_pci_setup_ports(dev, d, 14, &idx[0], &hw[0], &hws[0]);
 
-	ide_device_add(idx, d);
+	ide_device_add(idx, d, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index 0106e2a2df77..33fe15db408a 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -57,9 +57,9 @@ static int __devinit
 delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 {
 	unsigned long base;
-	hw_regs_t hw;
 	ide_hwif_t *hwif = NULL;
 	int i, rc;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	rc = pci_enable_device(dev);
@@ -93,11 +93,9 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 
 	i = hwif->index;
 
-	ide_init_port_hw(hwif, &hw);
-
 	idx[0] = i;
 
-	ide_device_add(idx, &delkin_cb_port_info);
+	ide_device_add(idx, &delkin_cb_port_info, hws);
 
 	pci_set_drvdata(dev, hwif);
 
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 789c66dfbde5..328e2df66550 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -554,7 +554,7 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev,
 {
 	struct scc_ports *ports = pci_get_drvdata(dev);
 	ide_hwif_t *hwif = NULL;
-	hw_regs_t hw;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	int i;
 
@@ -568,11 +568,10 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev,
 	hw.irq = dev->irq;
 	hw.dev = &dev->dev;
 	hw.chipset = ide_pci;
-	ide_init_port_hw(hwif, &hw);
 
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, d);
+	ide_device_add(idx, d, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index c79ff5b41088..76afa1f9c599 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -584,8 +584,8 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	unsigned long bar0, cmd_phys_base, ctl;
 	void __iomem *virt_base;
 	ide_hwif_t *hwif;
+	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
-	hw_regs_t hw;
 	struct ide_port_info d = sgiioc4_port_info;
 
 	/*  Get the CmdBlk and CtrlBlk Base Registers */
@@ -622,8 +622,6 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	if (hwif == NULL)
 		goto err;
 
-	ide_init_port_hw(hwif, &hw);
-
 	/* The IOC4 uses MMIO rather than Port IO. */
 	default_hwif_mmiops(hwif);
 
@@ -634,7 +632,7 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 
 	idx[0] = hwif->index;
 
-	if (ide_device_add(idx, &d))
+	if (ide_device_add(idx, &d, hws))
 		return -EIO;
 
 	return 0;
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 4e2944ee76d8..e68e33bb2c35 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1010,6 +1010,7 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 	struct device_node *np = pmif->node;
 	const int *bidp;
 	ide_hwif_t *hwif;
+	hw_regs_t *hws[] = { hw, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	struct ide_port_info d = pmac_port_info;
 
@@ -1095,11 +1096,9 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 	default_hwif_mmiops(hwif);
        	hwif->OUTBSYNC = pmac_outbsync;
 
-	ide_init_port_hw(hwif, hw);
-
 	idx[0] = hwif->index;
 
-	ide_device_add(idx, &d);
+	ide_device_add(idx, &d, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 7ae6ae45331f..acb467c6f345 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -289,6 +289,7 @@ static int ide_pci_check_iomem(struct pci_dev *dev, const struct ide_port_info *
  *	@d: IDE port info
  *	@port: port number
  *	@irq: PCI IRQ
+ *	@hw: hw_regs_t instance corresponding to this port
  *
  *	Perform the initial set up for the hardware interface structure. This
  *	is done per interface port rather than per PCI device. There may be
@@ -299,11 +300,11 @@ static int ide_pci_check_iomem(struct pci_dev *dev, const struct ide_port_info *
 
 static ide_hwif_t *ide_hwif_configure(struct pci_dev *dev,
 				      const struct ide_port_info *d,
-				      unsigned int port, int irq)
+				      unsigned int port, int irq,
+				      hw_regs_t *hw)
 {
 	unsigned long ctl = 0, base = 0;
 	ide_hwif_t *hwif;
-	struct hw_regs_s hw;
 
 	if ((d->host_flags & IDE_HFLAG_ISA_PORTS) == 0) {
 		if (ide_pci_check_iomem(dev, d, 2 * port) ||
@@ -327,17 +328,17 @@ static ide_hwif_t *ide_hwif_configure(struct pci_dev *dev,
 		return NULL;
 	}
 
+	memset(hw, 0, sizeof(*hw));
+	hw->irq = irq;
+	hw->dev = &dev->dev;
+	hw->chipset = d->chipset ? d->chipset : ide_pci;
+	ide_std_init_ports(hw, base, ctl | 2);
+
 	hwif = ide_find_port_slot(d);
 	if (hwif == NULL)
 		return NULL;
 
-	memset(&hw, 0, sizeof(hw));
-	hw.irq = irq;
-	hw.dev = &dev->dev;
-	hw.chipset = d->chipset ? d->chipset : ide_pci;
-	ide_std_init_ports(&hw, base, ctl | 2);
-
-	ide_init_port_hw(hwif, &hw);
+	hwif->chipset = hw->chipset;
 
 	return hwif;
 }
@@ -430,6 +431,8 @@ out:
  *	@d: IDE port info
  *	@pciirq: IRQ line
  *	@idx: ATA index table to update
+ *	@hw: hw_regs_t instances corresponding to this PCI IDE device
+ *	@hws: hw_regs_t pointers table to update
  *
  *	Scan the interfaces attached to this device and do any
  *	necessary per port setup. Attach the devices and ask the
@@ -440,7 +443,8 @@ out:
  *	where the chipset setup is not the default PCI IDE one.
  */
 
-void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d, int pciirq, u8 *idx)
+void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d,
+			 int pciirq, u8 *idx, hw_regs_t *hw, hw_regs_t **hws)
 {
 	int channels = (d->host_flags & IDE_HFLAG_SINGLE) ? 1 : 2, port;
 	ide_hwif_t *hwif;
@@ -459,10 +463,11 @@ void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d, int
 			continue;	/* port not enabled */
 		}
 
-		hwif = ide_hwif_configure(dev, d, port, pciirq);
+		hwif = ide_hwif_configure(dev, d, port, pciirq, hw + port);
 		if (hwif == NULL)
 			continue;
 
+		*(hws + port) = hw + port;
 		*(idx + port) = hwif->index;
 	}
 }
@@ -537,15 +542,16 @@ out:
 int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 {
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
+	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 	int ret;
 
 	ret = do_ide_setup_pci_device(dev, d, 1);
 
 	if (ret >= 0) {
 		/* FIXME: silent failure can happen */
-		ide_pci_setup_ports(dev, d, ret, &idx[0]);
+		ide_pci_setup_ports(dev, d, ret, &idx[0], &hw[0], &hws[0]);
 
-		ide_device_add(idx, d);
+		ide_device_add(idx, d, hws);
 	}
 
 	return ret;
@@ -557,6 +563,7 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 {
 	struct pci_dev *pdev[] = { dev1, dev2 };
 	int ret, i;
+	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	for (i = 0; i < 2; i++) {
@@ -570,10 +577,11 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 			goto out;
 
 		/* FIXME: silent failure can happen */
-		ide_pci_setup_ports(pdev[i], d, ret, &idx[i*2]);
+		ide_pci_setup_ports(pdev[i], d, ret, &idx[i*2], &hw[i*2],
+				    &hws[i*2]);
 	}
 
-	ide_device_add(idx, d);
+	ide_device_add(idx, d, hws);
 out:
 	return ret;
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 4726126f5a59..f58548becac0 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1000,7 +1000,8 @@ extern int __ide_pci_register_driver(struct pci_driver *driver, struct module *o
 #define ide_pci_register_driver(d) pci_register_driver(d)
 #endif
 
-void ide_pci_setup_ports(struct pci_dev *, const struct ide_port_info *, int, u8 *);
+void ide_pci_setup_ports(struct pci_dev *, const struct ide_port_info *, int,
+			 u8 *, hw_regs_t *, hw_regs_t **);
 void ide_setup_pci_noise(struct pci_dev *, const struct ide_port_info *);
 
 #ifdef CONFIG_BLK_DEV_IDEDMA_PCI
@@ -1217,8 +1218,8 @@ void ide_undecoded_slave(ide_drive_t *);
 
 void ide_port_apply_params(ide_hwif_t *);
 
-int ide_device_add_all(u8 *idx, const struct ide_port_info *);
-int ide_device_add(u8 idx[4], const struct ide_port_info *);
+int ide_device_add_all(u8 *, const struct ide_port_info *, hw_regs_t **);
+int ide_device_add(u8 *, const struct ide_port_info *, hw_regs_t **);
 int ide_legacy_device_add(const struct ide_port_info *, unsigned long);
 void ide_port_unregister_devices(ide_hwif_t *);
 void ide_port_scan(ide_hwif_t *);
-- 
cgit v1.2.3


From b2f951aabc9cc7d5fb987aeec9aef96ccce618a5 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:50 +0200
Subject: ide: add ->read_sff_dma_status method

Add ->read_sff_dma_status method for reading DMA Status register
and use it instead of ->INB.

While at it:

* Use inb() directly in ns87415.c::ns87415_dma_end().

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-dma.c      | 12 ++++++------
 drivers/ide/ide-iops.c     | 10 ++++++++++
 drivers/ide/pci/ns87415.c  | 13 ++++++++++---
 drivers/ide/pci/scc_pata.c |  7 +++++++
 include/linux/ide.h        |  2 ++
 5 files changed, 35 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 7ee44f86bc54..ebddedde24af 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -334,7 +334,7 @@ static int config_drive_for_dma (ide_drive_t *drive)
 static int dma_timer_expiry (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	u8 dma_stat		= hwif->INB(hwif->dma_status);
+	u8 dma_stat		= hwif->read_sff_dma_status(hwif);
 
 	printk(KERN_WARNING "%s: dma_timer_expiry: dma status == 0x%02x\n",
 		drive->name, dma_stat);
@@ -369,7 +369,7 @@ void ide_dma_host_set(ide_drive_t *drive, int on)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	u8 unit			= (drive->select.b.unit & 0x01);
-	u8 dma_stat		= hwif->INB(hwif->dma_status);
+	u8 dma_stat		= hwif->read_sff_dma_status(hwif);
 
 	if (on)
 		dma_stat |= (1 << (5 + unit));
@@ -472,8 +472,8 @@ int ide_dma_setup(ide_drive_t *drive)
 	/* specify r/w */
 	hwif->OUTB(reading, hwif->dma_command);
 
-	/* read dma_status for INTR & ERROR flags */
-	dma_stat = hwif->INB(hwif->dma_status);
+	/* read DMA status for INTR & ERROR flags */
+	dma_stat = hwif->read_sff_dma_status(hwif);
 
 	/* clear INTR & ERROR flags */
 	hwif->OUTB(dma_stat|6, hwif->dma_status);
@@ -520,7 +520,7 @@ int __ide_dma_end (ide_drive_t *drive)
 	/* stop DMA */
 	hwif->OUTB(dma_cmd&~1, hwif->dma_command);
 	/* get DMA status */
-	dma_stat = hwif->INB(hwif->dma_status);
+	dma_stat = hwif->read_sff_dma_status(hwif);
 	/* clear the INTR & ERROR bits */
 	hwif->OUTB(dma_stat|6, hwif->dma_status);
 	/* purge DMA mappings */
@@ -537,7 +537,7 @@ EXPORT_SYMBOL(__ide_dma_end);
 int ide_dma_test_irq(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	u8 dma_stat		= hwif->INB(hwif->dma_status);
+	u8 dma_stat		= hwif->read_sff_dma_status(hwif);
 
 	/* return 1 if INTR asserted */
 	if ((dma_stat & 4) == 4)
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 44aaec256a30..a09bf4369ed8 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -103,6 +103,14 @@ void SELECT_MASK(ide_drive_t *drive, int mask)
 		port_ops->maskproc(drive, mask);
 }
 
+static u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
+{
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
+		return readb((void __iomem *)hwif->dma_status);
+	else
+		return inb(hwif->dma_status);
+}
+
 static void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif = drive->hwif;
@@ -323,6 +331,8 @@ static void ata_output_data(ide_drive_t *drive, struct request *rq,
 
 void default_hwif_transport(ide_hwif_t *hwif)
 {
+	hwif->read_sff_dma_status = ide_read_sff_dma_status;
+
 	hwif->tf_load	  = ide_tf_load;
 	hwif->tf_read	  = ide_tf_read;
 
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index 45ba71a7182f..9ffdbb89df5c 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -63,6 +63,11 @@ static u8 superio_ide_inb (unsigned long port)
 	return inb(port);
 }
 
+static u8 superio_read_sff_dma_status(ide_hwif_t *hwif)
+{
+	return superio_ide_inb(hwif->dma_status);
+}
+
 static void superio_tf_read(ide_drive_t *drive, ide_task_t *task)
 {
 	struct ide_io_ports *io_ports = &drive->hwif->io_ports;
@@ -122,6 +127,8 @@ static void __devinit superio_ide_init_iops (struct hwif_s *hwif)
 	tmp = superio_ide_inb(superio_ide_dma_status[port]);
 	outb(tmp | 0x66, superio_ide_dma_status[port]);
 
+	hwif->read_sff_dma_status = superio_read_sff_dma_status;
+
 	hwif->tf_read = superio_tf_read;
 
 	/* We need to override inb to workaround a SuperIO errata */
@@ -200,13 +207,13 @@ static int ns87415_dma_end(ide_drive_t *drive)
 	u8 dma_stat = 0, dma_cmd = 0;
 
 	drive->waiting_for_dma = 0;
-	dma_stat = hwif->INB(hwif->dma_status);
+	dma_stat = hwif->read_sff_dma_status(hwif);
 	/* get dma command mode */
-	dma_cmd = hwif->INB(hwif->dma_command);
+	dma_cmd = inb(hwif->dma_command);
 	/* stop DMA */
 	outb(dma_cmd & ~1, hwif->dma_command);
 	/* from ERRATA: clear the INTR & ERROR bits */
-	dma_cmd = hwif->INB(hwif->dma_command);
+	dma_cmd = inb(hwif->dma_command);
 	outb(dma_cmd | 6, hwif->dma_command);
 	/* and free any DMA resources */
 	ide_destroy_dmatable(drive);
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 328e2df66550..7a2a7b2a319a 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -126,6 +126,11 @@ static u8 scc_ide_inb(unsigned long port)
 	return (u8)data;
 }
 
+static u8 scc_read_sff_dma_status(ide_hwif_t *hwif)
+{
+	return (u8)in_be32((void *)hwif->dma_status);
+}
+
 static void scc_ide_insw(unsigned long port, void *addr, u32 count)
 {
 	u16 *ptr = (u16 *)addr;
@@ -773,6 +778,8 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 
 	ide_set_hwifdata(hwif, ports);
 
+	hwif->read_sff_dma_status = scc_read_sff_dma_status;
+
 	hwif->tf_load = scc_tf_load;
 	hwif->tf_read = scc_tf_read;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f58548becac0..ca0efbb0a8b4 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -489,6 +489,8 @@ typedef struct hwif_s {
 	const struct ide_port_ops	*port_ops;
 	const struct ide_dma_ops	*dma_ops;
 
+	u8 (*read_sff_dma_status)(struct hwif_s *);
+
 	void (*tf_load)(ide_drive_t *, struct ide_task_s *);
 	void (*tf_read)(ide_drive_t *, struct ide_task_s *);
 
-- 
cgit v1.2.3


From cab7f8eda40d3e3e16b137c67cdddc2cf893c5d7 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:51 +0200
Subject: ide: remove ->dma_{status,command} fields from ide_hwif_t

* Use ->dma_base + offset instead of ->dma_{status,command}
  and remove no longer needed ->dma_{status,command}.

While at it:

* Use ATA_DMA_* defines.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-dma.c          | 42 +++++++++++++++++++++---------------------
 drivers/ide/ide-iops.c         |  4 ++--
 drivers/ide/pci/cmd64x.c       | 12 ++++++------
 drivers/ide/pci/hpt366.c       | 10 +++++-----
 drivers/ide/pci/ns87415.c      | 14 +++++++-------
 drivers/ide/pci/pdc202xx_old.c |  2 +-
 drivers/ide/pci/piix.c         |  4 ++--
 drivers/ide/pci/scc_pata.c     | 38 ++++++++++++++++++--------------------
 drivers/ide/pci/siimage.c      |  4 ++--
 drivers/ide/pci/sl82c105.c     |  4 ++--
 drivers/ide/pci/tc86c001.c     | 13 ++++++++-----
 include/linux/ide.h            |  2 --
 12 files changed, 74 insertions(+), 75 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index f3229642c052..d98a9da2699c 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -377,9 +377,10 @@ void ide_dma_host_set(ide_drive_t *drive, int on)
 		dma_stat &= ~(1 << (5 + unit));
 
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
-		writeb(dma_stat, (void __iomem *)hwif->dma_status);
+		writeb(dma_stat,
+		       (void __iomem *)(hwif->dma_base + ATA_DMA_STATUS));
 	else
-		outb(dma_stat, hwif->dma_status);
+		outb(dma_stat, hwif->dma_base + ATA_DMA_STATUS);
 }
 
 EXPORT_SYMBOL_GPL(ide_dma_host_set);
@@ -475,18 +476,19 @@ int ide_dma_setup(ide_drive_t *drive)
 
 	/* specify r/w */
 	if (mmio)
-		writeb(reading, (void __iomem *)hwif->dma_command);
+		writeb(reading, (void __iomem *)(hwif->dma_base + ATA_DMA_CMD));
 	else
-		outb(reading, hwif->dma_command);
+		outb(reading, hwif->dma_base + ATA_DMA_CMD);
 
 	/* read DMA status for INTR & ERROR flags */
 	dma_stat = hwif->read_sff_dma_status(hwif);
 
 	/* clear INTR & ERROR flags */
 	if (mmio)
-		writeb(dma_stat | 6, (void __iomem *)hwif->dma_status);
+		writeb(dma_stat | 6,
+		       (void __iomem *)(hwif->dma_base + ATA_DMA_STATUS));
 	else
-		outb(dma_stat | 6, hwif->dma_status);
+		outb(dma_stat | 6, hwif->dma_base + ATA_DMA_STATUS);
 
 	drive->waiting_for_dma = 1;
 	return 0;
@@ -512,12 +514,13 @@ void ide_dma_start(ide_drive_t *drive)
 	 * we do this part before issuing the drive cmd.
 	 */
 	if (hwif->host_flags & IDE_HFLAG_MMIO) {
-		dma_cmd = readb((void __iomem *)hwif->dma_command);
+		dma_cmd = readb((void __iomem *)(hwif->dma_base + ATA_DMA_CMD));
 		/* start DMA */
-		writeb(dma_cmd | 1, (void __iomem *)hwif->dma_command);
+		writeb(dma_cmd | 1,
+		       (void __iomem *)(hwif->dma_base + ATA_DMA_CMD));
 	} else {
-		dma_cmd = inb(hwif->dma_command);
-		outb(dma_cmd | 1, hwif->dma_command);
+		dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
+		outb(dma_cmd | 1, hwif->dma_base + ATA_DMA_CMD);
 	}
 
 	hwif->dma = 1;
@@ -537,12 +540,13 @@ int __ide_dma_end (ide_drive_t *drive)
 
 	if (mmio) {
 		/* get DMA command mode */
-		dma_cmd = readb((void __iomem *)hwif->dma_command);
+		dma_cmd = readb((void __iomem *)(hwif->dma_base + ATA_DMA_CMD));
 		/* stop DMA */
-		writeb(dma_cmd & ~1, (void __iomem *)hwif->dma_command);
+		writeb(dma_cmd & ~1,
+		       (void __iomem *)(hwif->dma_base + ATA_DMA_CMD));
 	} else {
-		dma_cmd = inb(hwif->dma_command);
-		outb(dma_cmd & ~1, hwif->dma_command);
+		dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
+		outb(dma_cmd & ~1, hwif->dma_base + ATA_DMA_CMD);
 	}
 
 	/* get DMA status */
@@ -550,9 +554,10 @@ int __ide_dma_end (ide_drive_t *drive)
 
 	if (mmio)
 		/* clear the INTR & ERROR bits */
-		writeb(dma_stat | 6, (void __iomem *)hwif->dma_status);
+		writeb(dma_stat | 6,
+		       (void __iomem *)(hwif->dma_base + ATA_DMA_STATUS));
 	else
-		outb(dma_stat | 6, hwif->dma_status);
+		outb(dma_stat | 6, hwif->dma_base + ATA_DMA_STATUS);
 
 	/* purge DMA mappings */
 	ide_destroy_dmatable(drive);
@@ -888,11 +893,6 @@ void ide_setup_dma(ide_hwif_t *hwif, unsigned long base)
 {
 	hwif->dma_base = base;
 
-	if (!hwif->dma_command)
-		hwif->dma_command	= hwif->dma_base + 0;
-	if (!hwif->dma_status)
-		hwif->dma_status	= hwif->dma_base + 2;
-
 	hwif->dma_ops = &sff_dma_ops;
 }
 
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index a09bf4369ed8..17cad6c39ee3 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -106,9 +106,9 @@ void SELECT_MASK(ide_drive_t *drive, int mask)
 static u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
-		return readb((void __iomem *)hwif->dma_status);
+		return readb((void __iomem *)(hwif->dma_base + ATA_DMA_STATUS));
 	else
-		return inb(hwif->dma_status);
+		return inb(hwif->dma_base + ATA_DMA_STATUS);
 }
 
 static void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index cfa784bacf48..ce58bfcdb3c6 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -262,7 +262,7 @@ static int cmd648_dma_test_irq(ide_drive_t *drive)
 	unsigned long base	= hwif->dma_base - (hwif->channel * 8);
 	u8 irq_mask		= hwif->channel ? MRDMODE_INTR_CH1 :
 						  MRDMODE_INTR_CH0;
-	u8 dma_stat		= inb(hwif->dma_status);
+	u8 dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 	u8 mrdmode		= inb(base + 1);
 
 #ifdef DEBUG
@@ -286,7 +286,7 @@ static int cmd64x_dma_test_irq(ide_drive_t *drive)
 	int irq_reg		= hwif->channel ? ARTTIM23 : CFR;
 	u8  irq_mask		= hwif->channel ? ARTTIM23_INTR_CH1 :
 						  CFR_INTR_CH0;
-	u8  dma_stat		= inb(hwif->dma_status);
+	u8  dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 	u8  irq_stat		= 0;
 
 	(void) pci_read_config_byte(dev, irq_reg, &irq_stat);
@@ -317,13 +317,13 @@ static int cmd646_1_dma_end(ide_drive_t *drive)
 
 	drive->waiting_for_dma = 0;
 	/* get DMA status */
-	dma_stat = inb(hwif->dma_status);
+	dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS);
 	/* read DMA command state */
-	dma_cmd = inb(hwif->dma_command);
+	dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
 	/* stop DMA */
-	outb(dma_cmd & ~1, hwif->dma_command);
+	outb(dma_cmd & ~1, hwif->dma_base + ATA_DMA_CMD);
 	/* clear the INTR & ERROR bits */
-	outb(dma_stat | 6, hwif->dma_status);
+	outb(dma_stat | 6, hwif->dma_base + ATA_DMA_STATUS);
 	/* and free any DMA resources */
 	ide_destroy_dmatable(drive);
 	/* verify good DMA status */
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 397c6cbe953c..d2f470ec8055 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -801,9 +801,9 @@ static void hpt370_irq_timeout(ide_drive_t *drive)
 	printk(KERN_DEBUG "%s: %d bytes in FIFO\n", drive->name, bfifo & 0x1ff);
 
 	/* get DMA command mode */
-	dma_cmd = inb(hwif->dma_command);
+	dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
 	/* stop DMA */
-	outb(dma_cmd & ~0x1, hwif->dma_command);
+	outb(dma_cmd & ~0x1, hwif->dma_base + ATA_DMA_CMD);
 	hpt370_clear_engine(drive);
 }
 
@@ -818,12 +818,12 @@ static void hpt370_dma_start(ide_drive_t *drive)
 static int hpt370_dma_end(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	u8  dma_stat		= inb(hwif->dma_status);
+	u8  dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 
 	if (dma_stat & 0x01) {
 		/* wait a little */
 		udelay(20);
-		dma_stat = inb(hwif->dma_status);
+		dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS);
 		if (dma_stat & 0x01)
 			hpt370_irq_timeout(drive);
 	}
@@ -850,7 +850,7 @@ static int hpt374_dma_test_irq(ide_drive_t *drive)
 		return 0;
 	}
 
-	dma_stat = inb(hwif->dma_status);
+	dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS);
 	/* return 1 if INTR asserted */
 	if (dma_stat & 4)
 		return 1;
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index 9ffdbb89df5c..76ce112fd857 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -65,7 +65,7 @@ static u8 superio_ide_inb (unsigned long port)
 
 static u8 superio_read_sff_dma_status(ide_hwif_t *hwif)
 {
-	return superio_ide_inb(hwif->dma_status);
+	return superio_ide_inb(hwif->dma_base + ATA_DMA_STATUS);
 }
 
 static void superio_tf_read(ide_drive_t *drive, ide_task_t *task)
@@ -208,13 +208,13 @@ static int ns87415_dma_end(ide_drive_t *drive)
 
 	drive->waiting_for_dma = 0;
 	dma_stat = hwif->read_sff_dma_status(hwif);
-	/* get dma command mode */
-	dma_cmd = inb(hwif->dma_command);
+	/* get DMA command mode */
+	dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
 	/* stop DMA */
-	outb(dma_cmd & ~1, hwif->dma_command);
+	outb(dma_cmd & ~1, hwif->dma_base + ATA_DMA_CMD);
 	/* from ERRATA: clear the INTR & ERROR bits */
-	dma_cmd = inb(hwif->dma_command);
-	outb(dma_cmd | 6, hwif->dma_command);
+	dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
+	outb(dma_cmd | 6, hwif->dma_base + ATA_DMA_CMD);
 	/* and free any DMA resources */
 	ide_destroy_dmatable(drive);
 	/* verify good DMA status */
@@ -298,7 +298,7 @@ static void __devinit init_hwif_ns87415 (ide_hwif_t *hwif)
 	if (!hwif->dma_base)
 		return;
 
-	outb(0x60, hwif->dma_status);
+	outb(0x60, hwif->dma_base + ATA_DMA_STATUS);
 }
 
 static const struct ide_port_ops ns87415_port_ops = {
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index fca89eda5c02..1c2f9df31129 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -206,7 +206,7 @@ static int pdc202xx_dma_test_irq(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	unsigned long high_16	= hwif->extra_base - 16;
-	u8 dma_stat		= inb(hwif->dma_status);
+	u8 dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 	u8 sc1d			= inb(high_16 + 0x001d);
 
 	if (hwif->channel) {
diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c
index f04738d14a6f..0ce41b4dddaf 100644
--- a/drivers/ide/pci/piix.c
+++ b/drivers/ide/pci/piix.c
@@ -227,9 +227,9 @@ static void piix_dma_clear_irq(ide_drive_t *drive)
 	u8 dma_stat;
 
 	/* clear the INTR & ERROR bits */
-	dma_stat = inb(hwif->dma_status);
+	dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS);
 	/* Should we force the bit as well ? */
-	outb(dma_stat, hwif->dma_status);
+	outb(dma_stat, hwif->dma_base + ATA_DMA_STATUS);
 }
 
 struct ich_laptop {
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 7a2a7b2a319a..fc163a7772a7 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -128,7 +128,7 @@ static u8 scc_ide_inb(unsigned long port)
 
 static u8 scc_read_sff_dma_status(ide_hwif_t *hwif)
 {
-	return (u8)in_be32((void *)hwif->dma_status);
+	return (u8)in_be32((void *)(hwif->dma_base + 4));
 }
 
 static void scc_ide_insw(unsigned long port, void *addr, u32 count)
@@ -266,14 +266,14 @@ static void scc_dma_host_set(ide_drive_t *drive, int on)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	u8 unit = (drive->select.b.unit & 0x01);
-	u8 dma_stat = scc_ide_inb(hwif->dma_status);
+	u8 dma_stat = scc_ide_inb(hwif->dma_base + 4);
 
 	if (on)
 		dma_stat |= (1 << (5 + unit));
 	else
 		dma_stat &= ~(1 << (5 + unit));
 
-	scc_ide_outb(dma_stat, hwif->dma_status);
+	scc_ide_outb(dma_stat, hwif->dma_base + 4);
 }
 
 /**
@@ -309,13 +309,13 @@ static int scc_dma_setup(ide_drive_t *drive)
 	out_be32((void __iomem *)(hwif->dma_base + 8), hwif->dmatable_dma);
 
 	/* specify r/w */
-	out_be32((void __iomem *)hwif->dma_command, reading);
+	out_be32((void __iomem *)hwif->dma_base, reading);
 
-	/* read dma_status for INTR & ERROR flags */
-	dma_stat = in_be32((void __iomem *)hwif->dma_status);
+	/* read DMA status for INTR & ERROR flags */
+	dma_stat = in_be32((void __iomem *)(hwif->dma_base + 4));
 
 	/* clear INTR & ERROR flags */
-	out_be32((void __iomem *)hwif->dma_status, dma_stat|6);
+	out_be32((void __iomem *)(hwif->dma_base + 4), dma_stat | 6);
 	drive->waiting_for_dma = 1;
 	return 0;
 }
@@ -323,10 +323,10 @@ static int scc_dma_setup(ide_drive_t *drive)
 static void scc_dma_start(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	u8 dma_cmd = scc_ide_inb(hwif->dma_command);
+	u8 dma_cmd = scc_ide_inb(hwif->dma_base);
 
 	/* start DMA */
-	scc_ide_outb(dma_cmd | 1, hwif->dma_command);
+	scc_ide_outb(dma_cmd | 1, hwif->dma_base);
 	hwif->dma = 1;
 	wmb();
 }
@@ -338,13 +338,13 @@ static int __scc_dma_end(ide_drive_t *drive)
 
 	drive->waiting_for_dma = 0;
 	/* get DMA command mode */
-	dma_cmd = scc_ide_inb(hwif->dma_command);
+	dma_cmd = scc_ide_inb(hwif->dma_base);
 	/* stop DMA */
-	scc_ide_outb(dma_cmd & ~1, hwif->dma_command);
+	scc_ide_outb(dma_cmd & ~1, hwif->dma_base);
 	/* get DMA status */
-	dma_stat = scc_ide_inb(hwif->dma_status);
+	dma_stat = scc_ide_inb(hwif->dma_base + 4);
 	/* clear the INTR & ERROR bits */
-	scc_ide_outb(dma_stat | 6, hwif->dma_status);
+	scc_ide_outb(dma_stat | 6, hwif->dma_base + 4);
 	/* purge DMA mappings */
 	ide_destroy_dmatable(drive);
 	/* verify good DMA status */
@@ -364,6 +364,7 @@ static int __scc_dma_end(ide_drive_t *drive)
 static int scc_dma_end(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
+	void __iomem *dma_base = (void __iomem *)hwif->dma_base;
 	unsigned long intsts_port = hwif->dma_base + 0x014;
 	u32 reg;
 	int dma_stat, data_loss = 0;
@@ -402,7 +403,7 @@ static int scc_dma_end(ide_drive_t *drive)
 			printk(KERN_WARNING "%s: SERROR\n", SCC_PATA_NAME);
 			out_be32((void __iomem *)intsts_port, INTSTS_SERROR|INTSTS_BMSINT);
 
-			out_be32((void __iomem *)hwif->dma_command, in_be32((void __iomem *)hwif->dma_command) & ~QCHCD_IOS_SS);
+			out_be32(dma_base, in_be32(dma_base) & ~QCHCD_IOS_SS);
 			continue;
 		}
 
@@ -417,7 +418,7 @@ static int scc_dma_end(ide_drive_t *drive)
 
 			out_be32((void __iomem *)intsts_port, INTSTS_PRERR|INTSTS_BMSINT);
 
-			out_be32((void __iomem *)hwif->dma_command, in_be32((void __iomem *)hwif->dma_command) & ~QCHCD_IOS_SS);
+			out_be32(dma_base, in_be32(dma_base) & ~QCHCD_IOS_SS);
 			continue;
 		}
 
@@ -425,12 +426,12 @@ static int scc_dma_end(ide_drive_t *drive)
 			printk(KERN_WARNING "%s: Response Error\n", SCC_PATA_NAME);
 			out_be32((void __iomem *)intsts_port, INTSTS_RERR|INTSTS_BMSINT);
 
-			out_be32((void __iomem *)hwif->dma_command, in_be32((void __iomem *)hwif->dma_command) & ~QCHCD_IOS_SS);
+			out_be32(dma_base, in_be32(dma_base) & ~QCHCD_IOS_SS);
 			continue;
 		}
 
 		if (reg & INTSTS_ICERR) {
-			out_be32((void __iomem *)hwif->dma_command, in_be32((void __iomem *)hwif->dma_command) & ~QCHCD_IOS_SS);
+			out_be32(dma_base, in_be32(dma_base) & ~QCHCD_IOS_SS);
 
 			printk(KERN_WARNING "%s: Illegal Configuration\n", SCC_PATA_NAME);
 			out_be32((void __iomem *)intsts_port, INTSTS_ICERR|INTSTS_BMSINT);
@@ -832,9 +833,6 @@ static void __devinit init_hwif_scc(ide_hwif_t *hwif)
 
 	ports->hwif = hwif;
 
-	hwif->dma_command = hwif->dma_base;
-	hwif->dma_status = hwif->dma_base + 0x04;
-
 	/* PTERADD */
 	out_be32((void __iomem *)(hwif->dma_base + 0x018), hwif->dmatable_dma);
 
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 6e9d7655d89c..21d7137f7d6c 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -334,7 +334,7 @@ static int siimage_io_dma_test_irq(ide_drive_t *drive)
 	unsigned long addr	= siimage_selreg(hwif, 1);
 
 	/* return 1 if INTR asserted */
-	if (hwif->INB(hwif->dma_status) & 4)
+	if (inb(hwif->dma_base + ATA_DMA_STATUS) & 4)
 		return 1;
 
 	/* return 1 if Device INTR asserted */
@@ -382,7 +382,7 @@ static int siimage_mmio_dma_test_irq(ide_drive_t *drive)
 	}
 
 	/* return 1 if INTR asserted */
-	if (readb((void __iomem *)hwif->dma_status) & 0x04)
+	if (readb((void __iomem *)(hwif->dma_base + ATA_DMA_STATUS)) & 4)
 		return 1;
 
 	/* return 1 if Device INTR asserted */
diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c
index 6efbde297174..f82a6502c1b7 100644
--- a/drivers/ide/pci/sl82c105.c
+++ b/drivers/ide/pci/sl82c105.c
@@ -157,9 +157,9 @@ static void sl82c105_dma_lost_irq(ide_drive_t *drive)
 	 * Was DMA enabled?  If so, disable it - we're resetting the
 	 * host.  The IDE layer will be handling the drive for us.
 	 */
-	dma_cmd = inb(hwif->dma_command);
+	dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
 	if (dma_cmd & 1) {
-		outb(dma_cmd & ~1, hwif->dma_command);
+		outb(dma_cmd & ~1, hwif->dma_base + ATA_DMA_CMD);
 		printk("sl82c105: DMA was enabled\n");
 	}
 
diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index 9b4b27a4c711..a81d47c55ce1 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -63,7 +63,7 @@ static int tc86c001_timer_expiry(ide_drive_t *drive)
 	ide_hwif_t *hwif	= HWIF(drive);
 	ide_expiry_t *expiry	= ide_get_hwifdata(hwif);
 	ide_hwgroup_t *hwgroup	= HWGROUP(drive);
-	u8 dma_stat		= inb(hwif->dma_status);
+	u8 dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 
 	/* Restore a higher level driver's expiry handler first. */
 	hwgroup->expiry	= expiry;
@@ -71,21 +71,24 @@ static int tc86c001_timer_expiry(ide_drive_t *drive)
 	if ((dma_stat & 5) == 1) {	/* DMA active and no interrupt */
 		unsigned long sc_base	= hwif->config_data;
 		unsigned long twcr_port	= sc_base + (drive->dn ? 0x06 : 0x04);
-		u8 dma_cmd		= inb(hwif->dma_command);
+		u8 dma_cmd		= inb(hwif->dma_base + ATA_DMA_CMD);
 
 		printk(KERN_WARNING "%s: DMA interrupt possibly stuck, "
 		       "attempting recovery...\n", drive->name);
 
 		/* Stop DMA */
-		outb(dma_cmd & ~0x01, hwif->dma_command);
+		outb(dma_cmd & ~0x01, hwif->dma_base + ATA_DMA_CMD);
 
 		/* Setup the dummy DMA transfer */
 		outw(0, sc_base + 0x0a);	/* Sector Count */
 		outw(0, twcr_port);	/* Transfer Word Count 1 or 2 */
 
 		/* Start the dummy DMA transfer */
-		outb(0x00, hwif->dma_command); /* clear R_OR_WCTR for write */
-		outb(0x01, hwif->dma_command); /* set START_STOPBM */
+
+		/* clear R_OR_WCTR for write */
+		outb(0x00, hwif->dma_base + ATA_DMA_CMD);
+		/* set START_STOPBM */
+		outb(0x01, hwif->dma_base + ATA_DMA_CMD);
 
 		/*
 		 * If an interrupt was pending, it should come thru shortly.
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ca0efbb0a8b4..85a32f472ef0 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -526,8 +526,6 @@ typedef struct hwif_s {
 	int		irq;		/* our irq number */
 
 	unsigned long	dma_base;	/* base addr for dma ports */
-	unsigned long	dma_command;	/* dma command register */
-	unsigned long	dma_status;	/* dma status register */
 
 	unsigned long	config_data;	/* for use by chipset-specific code */
 	unsigned long	select_data;	/* for use by chipset-specific code */
-- 
cgit v1.2.3


From 81e8d5a34f7d2a2acbe309cfa5810a9699a63239 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:51 +0200
Subject: ide: remove ide_setup_dma()

Export sff_dma_ops and then remove ide_setup_dma().

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/palm_bk3710.c |  7 +++----
 drivers/ide/ide-dma.c         | 12 ++----------
 drivers/ide/pci/alim15x3.c    |  4 +++-
 drivers/ide/pci/hpt366.c      |  4 +++-
 drivers/ide/setup-pci.c       |  4 +++-
 include/linux/ide.h           |  2 +-
 6 files changed, 15 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 023c10753f15..0229d794d909 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -316,15 +316,14 @@ static u8 __devinit palm_bk3710_cable_detect(ide_hwif_t *hwif)
 static int __devinit palm_bk3710_init_dma(ide_hwif_t *hwif,
 					  const struct ide_port_info *d)
 {
-	unsigned long base =
-		hwif->io_ports.data_addr - IDE_PALM_ATA_PRI_REG_OFFSET;
-
 	printk(KERN_INFO "    %s: MMIO-DMA\n", hwif->name);
 
 	if (ide_allocate_dma_engine(hwif))
 		return -1;
 
-	ide_setup_dma(hwif, base);
+	hwif->dma_base = hwif->io_ports.data_addr - IDE_PALM_ATA_PRI_REG_OFFSET;
+
+	hwif->dma_ops = &sff_dma_ops;
 
 	return 0;
 }
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index d98a9da2699c..ac342ebf6c54 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -878,7 +878,7 @@ int ide_allocate_dma_engine(ide_hwif_t *hwif)
 }
 EXPORT_SYMBOL_GPL(ide_allocate_dma_engine);
 
-static const struct ide_dma_ops sff_dma_ops = {
+const struct ide_dma_ops sff_dma_ops = {
 	.dma_host_set		= ide_dma_host_set,
 	.dma_setup		= ide_dma_setup,
 	.dma_exec_cmd		= ide_dma_exec_cmd,
@@ -888,13 +888,5 @@ static const struct ide_dma_ops sff_dma_ops = {
 	.dma_timeout		= ide_dma_timeout,
 	.dma_lost_irq		= ide_dma_lost_irq,
 };
-
-void ide_setup_dma(ide_hwif_t *hwif, unsigned long base)
-{
-	hwif->dma_base = base;
-
-	hwif->dma_ops = &sff_dma_ops;
-}
-
-EXPORT_SYMBOL_GPL(ide_setup_dma);
+EXPORT_SYMBOL_GPL(sff_dma_ops);
 #endif /* CONFIG_BLK_DEV_IDEDMA_SFF */
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index 80d19c0eb780..8015f6f65488 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -483,7 +483,9 @@ static int __devinit init_dma_ali15x3(ide_hwif_t *hwif,
 	if (ide_allocate_dma_engine(hwif))
 		return -1;
 
-	ide_setup_dma(hwif, base);
+	hwif->dma_base = base;
+
+	hwif->dma_ops = &sff_dma_ops;
 
 	return 0;
 }
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index d2f470ec8055..201e5ddae921 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -1346,7 +1346,9 @@ static int __devinit init_dma_hpt366(ide_hwif_t *hwif,
 	if (ide_allocate_dma_engine(hwif))
 		return -1;
 
-	ide_setup_dma(hwif, base);
+	hwif->dma_base = base;
+
+	hwif->dma_ops = &sff_dma_ops;
 
 	return 0;
 }
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index acb467c6f345..b047013f3652 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -377,7 +377,9 @@ int ide_hwif_setup_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
 		if (ide_allocate_dma_engine(hwif))
 			return -1;
 
-		ide_setup_dma(hwif, base);
+		hwif->dma_base = base;
+
+		hwif->dma_ops = &sff_dma_ops;
 	}
 
 	return 0;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 85a32f472ef0..6d774607e32a 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1007,6 +1007,7 @@ void ide_setup_pci_noise(struct pci_dev *, const struct ide_port_info *);
 #ifdef CONFIG_BLK_DEV_IDEDMA_PCI
 int ide_pci_set_master(struct pci_dev *, const char *);
 unsigned long ide_pci_dma_base(ide_hwif_t *, const struct ide_port_info *);
+extern const struct ide_dma_ops sff_dma_ops;
 int ide_hwif_setup_dma(ide_hwif_t *, const struct ide_port_info *);
 #else
 static inline int ide_hwif_setup_dma(ide_hwif_t *hwif,
@@ -1164,7 +1165,6 @@ void ide_destroy_dmatable(ide_drive_t *);
 extern int ide_build_dmatable(ide_drive_t *, struct request *);
 int ide_allocate_dma_engine(ide_hwif_t *);
 void ide_release_dma_engine(ide_hwif_t *);
-void ide_setup_dma(ide_hwif_t *, unsigned long);
 
 void ide_dma_host_set(ide_drive_t *, int);
 extern int ide_dma_setup(ide_drive_t *);
-- 
cgit v1.2.3


From ebb00fb55d0566bb3e81518122a57b4b3bedf1e4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:51 +0200
Subject: ide: factor out simplex handling from ide_pci_dma_base()

* Factor out simplex handling from ide_pci_dma_base() to
  ide_pci_check_simplex().

* Set hwif->dma_base early in ->init_dma method / ide_hwif_setup_dma()
  and reset it in ide_init_port() if DMA initialization fails.

* Use ->read_sff_dma_status instead of ->INB in ide_pci_dma_base().

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c    |  1 +
 drivers/ide/pci/alim15x3.c | 12 +++++++++---
 drivers/ide/pci/hpt366.c   | 12 +++++++++---
 drivers/ide/setup-pci.c    | 35 +++++++++++++++++++++++------------
 include/linux/ide.h        |  1 +
 5 files changed, 43 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 9d8686a49fcf..ec5f58c6f19a 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1363,6 +1363,7 @@ static void ide_init_port(ide_hwif_t *hwif, unsigned int port,
 
 		if (rc < 0) {
 			printk(KERN_INFO "%s: DMA disabled\n", hwif->name);
+			hwif->dma_base = 0;
 			hwif->swdma_mask = 0;
 			hwif->mwdma_mask = 0;
 			hwif->ultra_mask = 0;
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index 8015f6f65488..5ef7817ac64f 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -471,7 +471,15 @@ static int __devinit init_dma_ali15x3(ide_hwif_t *hwif,
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	unsigned long base = ide_pci_dma_base(hwif, d);
 
-	if (base == 0 || ide_pci_set_master(dev, d->name) < 0)
+	if (base == 0)
+		return -1;
+
+	hwif->dma_base = base;
+
+	if (ide_pci_check_simplex(hwif, d) < 0)
+		return -1;
+
+	if (ide_pci_set_master(dev, d->name) < 0)
 		return -1;
 
 	if (!hwif->channel)
@@ -483,8 +491,6 @@ static int __devinit init_dma_ali15x3(ide_hwif_t *hwif,
 	if (ide_allocate_dma_engine(hwif))
 		return -1;
 
-	hwif->dma_base = base;
-
 	hwif->dma_ops = &sff_dma_ops;
 
 	return 0;
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 201e5ddae921..e5651cefa395 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -1320,7 +1320,15 @@ static int __devinit init_dma_hpt366(ide_hwif_t *hwif,
 	unsigned long flags, base = ide_pci_dma_base(hwif, d);
 	u8 dma_old, dma_new, masterdma = 0, slavedma = 0;
 
-	if (base == 0 || ide_pci_set_master(dev, d->name) < 0)
+	if (base == 0)
+		return -1;
+
+	hwif->dma_base = base;
+
+	if (ide_pci_check_simplex(hwif, d) < 0)
+		return -1;
+
+	if (ide_pci_set_master(dev, d->name) < 0)
 		return -1;
 
 	dma_old = inb(base + 2);
@@ -1346,8 +1354,6 @@ static int __devinit init_dma_hpt366(ide_hwif_t *hwif,
 	if (ide_allocate_dma_engine(hwif))
 		return -1;
 
-	hwif->dma_base = base;
-
 	hwif->dma_ops = &sff_dma_ops;
 
 	return 0;
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index b047013f3652..c1b609d9cb28 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -73,15 +73,12 @@ static void ide_pci_clear_simplex(unsigned long dma_base, const char *name)
  *	@d: IDE port info
  *
  *	Fetch the DMA Bus-Master-I/O-Base-Address (BMIBA) from PCI space.
- *	Where a device has a partner that is already in DMA mode we check
- *	and enforce IDE simplex rules.
  */
 
 unsigned long ide_pci_dma_base(ide_hwif_t *hwif, const struct ide_port_info *d)
 {
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	unsigned long dma_base = 0;
-	u8 dma_stat = 0;
 
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		return hwif->dma_base;
@@ -102,11 +99,19 @@ unsigned long ide_pci_dma_base(ide_hwif_t *hwif, const struct ide_port_info *d)
 	if (hwif->channel)
 		dma_base += 8;
 
-	if (d->host_flags & IDE_HFLAG_CS5520)
+	return dma_base;
+}
+EXPORT_SYMBOL_GPL(ide_pci_dma_base);
+
+int ide_pci_check_simplex(ide_hwif_t *hwif, const struct ide_port_info *d)
+{
+	u8 dma_stat;
+
+	if (d->host_flags & (IDE_HFLAG_MMIO | IDE_HFLAG_CS5520))
 		goto out;
 
 	if (d->host_flags & IDE_HFLAG_CLEAR_SIMPLEX) {
-		ide_pci_clear_simplex(dma_base, d->name);
+		ide_pci_clear_simplex(hwif->dma_base, d->name);
 		goto out;
 	}
 
@@ -120,15 +125,15 @@ unsigned long ide_pci_dma_base(ide_hwif_t *hwif, const struct ide_port_info *d)
 	 * we tune the drive then try to grab DMA ownership if we want to be
 	 * the DMA end.  This has to be become dynamic to handle hot-plug.
 	 */
-	dma_stat = hwif->INB(dma_base + 2);
+	dma_stat = hwif->read_sff_dma_status(hwif);
 	if ((dma_stat & 0x80) && hwif->mate && hwif->mate->dma_base) {
 		printk(KERN_INFO "%s: simplex device: DMA disabled\n", d->name);
-		dma_base = 0;
+		return -1;
 	}
 out:
-	return dma_base;
+	return 0;
 }
-EXPORT_SYMBOL_GPL(ide_pci_dma_base);
+EXPORT_SYMBOL_GPL(ide_pci_check_simplex);
 
 /*
  * Set up BM-DMA capability (PnP BIOS should have done this)
@@ -363,7 +368,15 @@ int ide_hwif_setup_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
 	     (dev->class & 0x80))) {
 		unsigned long base = ide_pci_dma_base(hwif, d);
 
-		if (base == 0 || ide_pci_set_master(dev, d->name) < 0)
+		if (base == 0)
+			return -1;
+
+		hwif->dma_base = base;
+
+		if (ide_pci_check_simplex(hwif, d) < 0)
+			return -1;
+
+		if (ide_pci_set_master(dev, d->name) < 0)
 			return -1;
 
 		if (hwif->host_flags & IDE_HFLAG_MMIO)
@@ -377,8 +390,6 @@ int ide_hwif_setup_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
 		if (ide_allocate_dma_engine(hwif))
 			return -1;
 
-		hwif->dma_base = base;
-
 		hwif->dma_ops = &sff_dma_ops;
 	}
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 6d774607e32a..a179a7f6e444 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1008,6 +1008,7 @@ void ide_setup_pci_noise(struct pci_dev *, const struct ide_port_info *);
 int ide_pci_set_master(struct pci_dev *, const char *);
 unsigned long ide_pci_dma_base(ide_hwif_t *, const struct ide_port_info *);
 extern const struct ide_dma_ops sff_dma_ops;
+int ide_pci_check_simplex(ide_hwif_t *, const struct ide_port_info *);
 int ide_hwif_setup_dma(ide_hwif_t *, const struct ide_port_info *);
 #else
 static inline int ide_hwif_setup_dma(ide_hwif_t *hwif,
-- 
cgit v1.2.3


From c6dfa867bb45f4bff2e48f3bc89ab1d6a7ab4c21 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:51 +0200
Subject: ide: add ->exec_command method

Add ->exec_command method for writing ATA Command register
and use it instead of ->OUTBSYNC.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       |  3 +--
 drivers/ide/ide-iops.c     | 19 ++++++++++++++-----
 drivers/ide/ide-probe.c    |  6 +++---
 drivers/ide/ide-taskfile.c |  2 +-
 drivers/ide/pci/scc_pata.c |  9 +++++++++
 drivers/ide/ppc/pmac.c     |  9 +++++++++
 drivers/scsi/ide-scsi.c    |  3 +--
 include/linux/ide.h        |  3 ++-
 8 files changed, 40 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 661b75a89d4d..ca0c04a919cb 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -437,8 +437,7 @@ static ide_startstop_t ide_atapi_error(ide_drive_t *drive, struct request *rq, u
 
 	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		hwif->OUTBSYNC(hwif, WIN_IDLEIMMEDIATE,
-			       hwif->io_ports.command_addr);
+		hwif->exec_command(hwif, WIN_IDLEIMMEDIATE);
 
 	if (rq->errors >= ERROR_MAX) {
 		ide_kill_rq(drive, rq);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 17cad6c39ee3..d8db25581909 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -103,6 +103,14 @@ void SELECT_MASK(ide_drive_t *drive, int mask)
 		port_ops->maskproc(drive, mask);
 }
 
+static void ide_exec_command(ide_hwif_t *hwif, u8 cmd)
+{
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
+		writeb(cmd, (void __iomem *)hwif->io_ports.command_addr);
+	else
+		outb(cmd, hwif->io_ports.command_addr);
+}
+
 static u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
@@ -331,6 +339,7 @@ static void ata_output_data(ide_drive_t *drive, struct request *rq,
 
 void default_hwif_transport(ide_hwif_t *hwif)
 {
+	hwif->exec_command	  = ide_exec_command;
 	hwif->read_sff_dma_status = ide_read_sff_dma_status;
 
 	hwif->tf_load	  = ide_tf_load;
@@ -696,7 +705,7 @@ int ide_driveid_update(ide_drive_t *drive)
 	SELECT_MASK(drive, 1);
 	ide_set_irq(drive, 0);
 	msleep(50);
-	hwif->OUTBSYNC(hwif, WIN_IDENTIFY, hwif->io_ports.command_addr);
+	hwif->exec_command(hwif, WIN_IDENTIFY);
 	timeout = jiffies + WAIT_WORSTCASE;
 	do {
 		if (time_after(jiffies, timeout)) {
@@ -783,7 +792,7 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 	ide_set_irq(drive, 0);
 	hwif->OUTB(speed, io_ports->nsect_addr);
 	hwif->OUTB(SETFEATURES_XFER, io_ports->feature_addr);
-	hwif->OUTBSYNC(hwif, WIN_SETFEATURES, io_ports->command_addr);
+	hwif->exec_command(hwif, WIN_SETFEATURES);
 	if (drive->quirk_list == 2)
 		ide_set_irq(drive, 1);
 
@@ -891,7 +900,7 @@ void ide_execute_command(ide_drive_t *drive, u8 cmd, ide_handler_t *handler,
 
 	spin_lock_irqsave(&ide_lock, flags);
 	__ide_set_handler(drive, handler, timeout, expiry);
-	hwif->OUTBSYNC(hwif, cmd, hwif->io_ports.command_addr);
+	hwif->exec_command(hwif, cmd);
 	/*
 	 * Drive takes 400nS to respond, we must avoid the IRQ being
 	 * serviced before that.
@@ -909,7 +918,7 @@ void ide_execute_pkt_cmd(ide_drive_t *drive)
 	unsigned long flags;
 
 	spin_lock_irqsave(&ide_lock, flags);
-	hwif->OUTBSYNC(hwif, WIN_PACKETCMD, hwif->io_ports.command_addr);
+	hwif->exec_command(hwif, WIN_PACKETCMD);
 	ndelay(400);
 	spin_unlock_irqrestore(&ide_lock, flags);
 }
@@ -1116,7 +1125,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		pre_reset(drive);
 		SELECT_DRIVE(drive);
 		udelay (20);
-		hwif->OUTBSYNC(hwif, WIN_SRST, io_ports->command_addr);
+		hwif->exec_command(hwif, WIN_SRST);
 		ndelay(400);
 		hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
 		hwgroup->polling = 1;
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index ec5f58c6f19a..d1e834a1c5cd 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -295,7 +295,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		hwif->OUTB(0, io_ports->feature_addr);
 
 	/* ask drive for ID */
-	hwif->OUTBSYNC(hwif, cmd, hwif->io_ports.command_addr);
+	hwif->exec_command(hwif, cmd);
 
 	timeout = ((cmd == WIN_IDENTIFY) ? WAIT_WORSTCASE : WAIT_PIDENTIFY) / 2;
 	timeout += jiffies;
@@ -482,7 +482,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 			msleep(50);
 			SELECT_DRIVE(drive);
 			msleep(50);
-			hwif->OUTBSYNC(hwif, WIN_SRST, io_ports->command_addr);
+			hwif->exec_command(hwif, WIN_SRST);
 			(void)ide_busy_sleep(hwif);
 			rc = try_to_identify(drive, cmd);
 		}
@@ -518,7 +518,7 @@ static void enable_nest (ide_drive_t *drive)
 	printk("%s: enabling %s -- ", hwif->name, drive->id->model);
 	SELECT_DRIVE(drive);
 	msleep(50);
-	hwif->OUTBSYNC(hwif, EXABYTE_ENABLE_NEST, hwif->io_ports.command_addr);
+	hwif->exec_command(hwif, EXABYTE_ENABLE_NEST);
 
 	if (ide_busy_sleep(hwif)) {
 		printk(KERN_CONT "failed (timeout)\n");
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 1fbdb746dc88..c56be289417e 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -88,7 +88,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	switch (task->data_phase) {
 	case TASKFILE_MULTI_OUT:
 	case TASKFILE_OUT:
-		hwif->OUTBSYNC(hwif, tf->command, hwif->io_ports.command_addr);
+		hwif->exec_command(hwif, tf->command);
 		ndelay(400);	/* FIXME */
 		return pre_task_out_intr(drive, task->rq);
 	case TASKFILE_MULTI_IN:
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index fc163a7772a7..d140dfd565a4 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -126,6 +126,14 @@ static u8 scc_ide_inb(unsigned long port)
 	return (u8)data;
 }
 
+static void scc_exec_command(ide_hwif_t *hwif, u8 cmd)
+{
+	out_be32((void *)hwif->io_ports.command_addr, cmd);
+	eieio();
+	in_be32((void *)(hwif->dma_base + 0x01c));
+	eieio();
+}
+
 static u8 scc_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	return (u8)in_be32((void *)(hwif->dma_base + 4));
@@ -779,6 +787,7 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 
 	ide_set_hwifdata(hwif, ports);
 
+	hwif->exec_command	  = scc_exec_command;
 	hwif->read_sff_dma_status = scc_read_sff_dma_status;
 
 	hwif->tf_load = scc_tf_load;
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index e68e33bb2c35..7c3a84f8fbed 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -495,6 +495,13 @@ static void pmac_outbsync(ide_hwif_t *hwif, u8 value, unsigned long port)
 				     + IDE_TIMING_CONFIG));
 }
 
+static void pmac_exec_command(ide_hwif_t *hwif, u8 cmd)
+{
+	writeb(cmd, (void __iomem *)hwif->io_ports.command_addr);
+	(void)readl((void __iomem *)(hwif->io_ports.data_addr
+				     + IDE_TIMING_CONFIG));
+}
+
 /*
  * Old tuning functions (called on hdparm -p), sets up drive PIO timings
  */
@@ -1092,6 +1099,8 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 	if (hwif == NULL)
 		return -ENOENT;
 
+	hwif->exec_command = pmac_exec_command;
+
 	/* Setup MMIO ops */
 	default_hwif_mmiops(hwif);
        	hwif->OUTBSYNC = pmac_outbsync;
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index f843c1383a4b..80123890ced0 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -248,8 +248,7 @@ idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 
 	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		hwif->OUTBSYNC(hwif, WIN_IDLEIMMEDIATE,
-			       hwif->io_ports.command_addr);
+		hwif->exec_command(hwif, WIN_IDLEIMMEDIATE);
 
 	rq->errors++;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index a179a7f6e444..d77c1994d0e3 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -489,7 +489,8 @@ typedef struct hwif_s {
 	const struct ide_port_ops	*port_ops;
 	const struct ide_dma_ops	*dma_ops;
 
-	u8 (*read_sff_dma_status)(struct hwif_s *);
+	void	(*exec_command)(struct hwif_s *, u8);
+	u8	(*read_sff_dma_status)(struct hwif_s *);
 
 	void (*tf_load)(ide_drive_t *, struct ide_task_s *);
 	void (*tf_read)(ide_drive_t *, struct ide_task_s *);
-- 
cgit v1.2.3


From b73c7ee25da6133f97f47ffd3557288417da7c76 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:52 +0200
Subject: ide: add ->read_status method

* Remove ide_read_status() inline helper.

* Add ->read_status method for reading ATA Status register
  and use it instead of ->INB.

While at it:

* Don't use HWGROUP() macro.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c   |  4 +++-
 drivers/ide/ide-atapi.c    |  2 +-
 drivers/ide/ide-cd.c       | 12 ++++++++----
 drivers/ide/ide-dma.c      |  5 +++--
 drivers/ide/ide-floppy.c   |  3 ++-
 drivers/ide/ide-io.c       | 19 ++++++++++---------
 drivers/ide/ide-iops.c     | 33 ++++++++++++++++++++++-----------
 drivers/ide/ide-probe.c    | 22 +++++++++++-----------
 drivers/ide/ide-tape.c     |  6 ++++--
 drivers/ide/ide-taskfile.c | 23 ++++++++++++++---------
 drivers/ide/pci/ns87415.c  |  8 +++++++-
 drivers/ide/pci/scc_pata.c |  6 ++++++
 drivers/ide/pci/sgiioc4.c  | 19 ++++++++++---------
 drivers/scsi/ide-scsi.c    |  2 +-
 include/linux/ide.h        |  8 +-------
 15 files changed, 103 insertions(+), 69 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index 850fe9342a1f..f73422dd8474 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -375,12 +375,14 @@ static int icside_dma_test_irq(ide_drive_t *drive)
 
 static void icside_dma_timeout(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
+
 	printk(KERN_ERR "%s: DMA timeout occurred: ", drive->name);
 
 	if (icside_dma_test_irq(drive))
 		return;
 
-	ide_dump_status(drive, "DMA timeout", ide_read_status(drive));
+	ide_dump_status(drive, "DMA timeout", hwif->read_status(hwif));
 
 	icside_dma_end(drive);
 }
diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 2802031de670..6554c4225a05 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -35,7 +35,7 @@ ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	}
 
 	/* Clear the interrupt */
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
 		if (hwif->dma_ops->dma_end(drive) ||
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 6e29dd532090..82879a1a89e5 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -280,11 +280,12 @@ static void ide_dump_status_no_sense(ide_drive_t *drive, const char *msg, u8 st)
  */
 static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 {
-	struct request *rq = HWGROUP(drive)->rq;
+	ide_hwif_t *hwif = drive->hwif;
+	struct request *rq = hwif->hwgroup->rq;
 	int stat, err, sense_key;
 
 	/* check for errors */
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (stat_ret)
 		*stat_ret = stat;
@@ -606,6 +607,8 @@ static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive,
 static int ide_cd_check_ireason(ide_drive_t *drive, struct request *rq,
 				int len, int ireason, int rw)
 {
+	ide_hwif_t *hwif = drive->hwif;
+
 	/*
 	 * ireason == 0: the drive wants to receive data from us
 	 * ireason == 2: the drive is expecting to transfer data to us
@@ -624,7 +627,7 @@ static int ide_cd_check_ireason(ide_drive_t *drive, struct request *rq,
 		 * Some drives (ASUS) seem to tell us that status info is
 		 * available.  Just get it and ignore.
 		 */
-		(void)ide_read_status(drive);
+		(void)hwif->read_status(hwif);
 		return 0;
 	} else {
 		/* drive wants a command packet, or invalid ireason... */
@@ -1199,8 +1202,9 @@ static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 
 	if (blk_fs_request(rq)) {
 		if (info->cd_flags & IDE_CD_FLAG_SEEKING) {
+			ide_hwif_t *hwif = drive->hwif;
 			unsigned long elapsed = jiffies - info->start_seek;
-			int stat = ide_read_status(drive);
+			int stat = hwif->read_status(hwif);
 
 			if ((stat & SEEK_STAT) != SEEK_STAT) {
 				if (elapsed < IDECD_SEEK_TIMEOUT) {
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index ac342ebf6c54..ecf60dbbe3e9 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -100,10 +100,11 @@ static const struct drive_list_entry drive_blacklist [] = {
  
 ide_startstop_t ide_dma_intr (ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	u8 stat = 0, dma_stat = 0;
 
-	dma_stat = drive->hwif->dma_ops->dma_end(drive);
-	stat = ide_read_status(drive);
+	dma_stat = hwif->dma_ops->dma_end(drive);
+	stat = hwif->read_status(hwif);
 
 	if (OK_STAT(stat,DRIVE_READY,drive->bad_wstat|DRQ_STAT)) {
 		if (!dma_stat) {
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 011d72011cc4..6f5294cfff23 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -949,11 +949,12 @@ static int idefloppy_get_format_progress(ide_drive_t *drive, int __user *arg)
 
 		/* Else assume format_unit has finished, and we're at 0x10000 */
 	} else {
+		ide_hwif_t *hwif = drive->hwif;
 		unsigned long flags;
 		u8 stat;
 
 		local_irq_save(flags);
-		stat = ide_read_status(drive);
+		stat = hwif->read_status(hwif);
 		local_irq_restore(flags);
 
 		progress_indication = ((stat & SEEK_STAT) == 0) ? 0 : 0x10000;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index ca0c04a919cb..fdc221ce9920 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -381,8 +381,7 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8
 		if (err == ABRT_ERR) {
 			if (drive->select.b.lba &&
 			    /* some newer drives don't support WIN_SPECIFY */
-			    hwif->INB(hwif->io_ports.command_addr) ==
-				WIN_SPECIFY)
+			    hwif->read_status(hwif) == WIN_SPECIFY)
 				return ide_stopped;
 		} else if ((err & BAD_CRC) == BAD_CRC) {
 			/* UDMA crc error, just retry the operation */
@@ -408,7 +407,7 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8
 		return ide_stopped;
 	}
 
-	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
 		rq->errors |= ERROR_RESET;
 
 	if ((rq->errors & ERROR_RESET) == ERROR_RESET) {
@@ -435,7 +434,7 @@ static ide_startstop_t ide_atapi_error(ide_drive_t *drive, struct request *rq, u
 		/* add decoding error stuff */
 	}
 
-	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
 		hwif->exec_command(hwif, WIN_IDLEIMMEDIATE);
 
@@ -711,7 +710,8 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 #ifdef DEBUG
  	printk("%s: DRIVE_CMD (null)\n", drive->name);
 #endif
-	ide_end_drive_cmd(drive, ide_read_status(drive), ide_read_error(drive));
+	ide_end_drive_cmd(drive, hwif->read_status(hwif),
+			  ide_read_error(drive));
 
  	return ide_stopped;
 }
@@ -1141,7 +1141,7 @@ static ide_startstop_t ide_dma_timeout_retry(ide_drive_t *drive, int error)
 		printk(KERN_WARNING "%s: DMA timeout error\n", drive->name);
 		(void)hwif->dma_ops->dma_end(drive);
 		ret = ide_error(drive, "dma timeout error",
-				ide_read_status(drive));
+				hwif->read_status(hwif));
 	} else {
 		printk(KERN_WARNING "%s: DMA timeout retry\n", drive->name);
 		hwif->dma_ops->dma_timeout(drive);
@@ -1266,7 +1266,7 @@ void ide_timer_expiry (unsigned long data)
 				} else
 					startstop =
 					ide_error(drive, "irq timeout",
-						  ide_read_status(drive));
+						  hwif->read_status(hwif));
 			}
 			drive->service_time = jiffies - drive->service_start;
 			spin_lock_irq(&ide_lock);
@@ -1322,7 +1322,8 @@ static void unexpected_intr (int irq, ide_hwgroup_t *hwgroup)
 	 */
 	do {
 		if (hwif->irq == irq) {
-			stat = hwif->INB(hwif->io_ports.status_addr);
+			stat = hwif->read_status(hwif);
+
 			if (!OK_STAT(stat, READY_STAT, BAD_STAT)) {
 				/* Try to not flood the console with msgs */
 				static unsigned long last_msgtime, count;
@@ -1412,7 +1413,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 			 * Whack the status register, just in case
 			 * we have a leftover pending IRQ.
 			 */
-			(void) hwif->INB(hwif->io_ports.status_addr);
+			(void)hwif->read_status(hwif);
 #endif /* CONFIG_BLK_DEV_IDEPCI */
 		}
 		spin_unlock_irqrestore(&ide_lock, flags);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index d8db25581909..086eceaeeafd 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -111,6 +111,14 @@ static void ide_exec_command(ide_hwif_t *hwif, u8 cmd)
 		outb(cmd, hwif->io_ports.command_addr);
 }
 
+static u8 ide_read_status(ide_hwif_t *hwif)
+{
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
+		return readb((void __iomem *)hwif->io_ports.status_addr);
+	else
+		return inb(hwif->io_ports.status_addr);
+}
+
 static u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
@@ -340,6 +348,7 @@ static void ata_output_data(ide_drive_t *drive, struct request *rq,
 void default_hwif_transport(ide_hwif_t *hwif)
 {
 	hwif->exec_command	  = ide_exec_command;
+	hwif->read_status	  = ide_read_status;
 	hwif->read_sff_dma_status = ide_read_sff_dma_status;
 
 	hwif->tf_load	  = ide_tf_load;
@@ -505,7 +514,7 @@ int drive_is_ready (ide_drive_t *drive)
 		stat = ide_read_altstatus(drive);
 	else
 		/* Note: this may clear a pending IRQ!! */
-		stat = ide_read_status(drive);
+		stat = hwif->read_status(hwif);
 
 	if (stat & BUSY_STAT)
 		/* drive busy:  definitely not interrupting */
@@ -530,24 +539,25 @@ EXPORT_SYMBOL(drive_is_ready);
  */
 static int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad, unsigned long timeout, u8 *rstat)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long flags;
 	int i;
 	u8 stat;
 
 	udelay(1);	/* spec allows drive 400ns to assert "BUSY" */
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (stat & BUSY_STAT) {
 		local_irq_set(flags);
 		timeout += jiffies;
-		while ((stat = ide_read_status(drive)) & BUSY_STAT) {
+		while ((stat = hwif->read_status(hwif)) & BUSY_STAT) {
 			if (time_after(jiffies, timeout)) {
 				/*
 				 * One last read after the timeout in case
 				 * heavy interrupt load made us not make any
 				 * progress during the timeout..
 				 */
-				stat = ide_read_status(drive);
+				stat = hwif->read_status(hwif);
 				if (!(stat & BUSY_STAT))
 					break;
 
@@ -567,7 +577,7 @@ static int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad, unsigned long ti
 	 */
 	for (i = 0; i < 10; i++) {
 		udelay(1);
-		stat = ide_read_status(drive);
+		stat = hwif->read_status(hwif);
 
 		if (OK_STAT(stat, good, bad)) {
 			*rstat = stat;
@@ -718,7 +728,7 @@ int ide_driveid_update(ide_drive_t *drive)
 	} while (stat & BUSY_STAT);
 
 	msleep(50);	/* wait for IRQ and DRQ_STAT */
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (!OK_STAT(stat, DRQ_STAT, BAD_R_STAT)) {
 		SELECT_MASK(drive, 0);
@@ -733,7 +743,7 @@ int ide_driveid_update(ide_drive_t *drive)
 		return 0;
 	}
 	hwif->input_data(drive, NULL, id, SECTOR_SIZE);
-	(void)ide_read_status(drive);	/* clear drive IRQ */
+	(void)hwif->read_status(hwif);	/* clear drive IRQ */
 	local_irq_enable();
 	local_irq_restore(flags);
 	ide_fix_driveid(id);
@@ -943,12 +953,13 @@ static ide_startstop_t do_reset1 (ide_drive_t *, int);
  */
 static ide_startstop_t atapi_reset_pollfunc (ide_drive_t *drive)
 {
-	ide_hwgroup_t *hwgroup	= HWGROUP(drive);
+	ide_hwif_t *hwif = drive->hwif;
+	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	u8 stat;
 
 	SELECT_DRIVE(drive);
 	udelay (10);
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (OK_STAT(stat, 0, BUSY_STAT))
 		printk("%s: ATAPI reset complete\n", drive->name);
@@ -994,7 +1005,7 @@ static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 		}
 	}
 
-	tmp = ide_read_status(drive);
+	tmp = hwif->read_status(hwif);
 
 	if (!OK_STAT(tmp, 0, BUSY_STAT)) {
 		if (time_before(jiffies, hwgroup->poll_timeout)) {
@@ -1208,7 +1219,7 @@ int ide_wait_not_busy(ide_hwif_t *hwif, unsigned long timeout)
 		 * about locking issues (2.5 work ?).
 		 */
 		mdelay(1);
-		stat = hwif->INB(hwif->io_ports.status_addr);
+		stat = hwif->read_status(hwif);
 		if ((stat & BUSY_STAT) == 0)
 			return 0;
 		/*
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index d1e834a1c5cd..c42fcfedcbf6 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -276,7 +276,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 
 	if (io_ports->ctl_addr) {
 		a = ide_read_altstatus(drive);
-		s = ide_read_status(drive);
+		s = hwif->read_status(hwif);
 		if ((a ^ s) & ~INDEX_STAT)
 			/* ancient Seagate drives, broken interfaces */
 			printk(KERN_INFO "%s: probing with STATUS(0x%02x) "
@@ -307,12 +307,12 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		/* give drive a breather */
 		msleep(50);
 		s = use_altstatus ? ide_read_altstatus(drive)
-				  : ide_read_status(drive);
+				  : hwif->read_status(hwif);
 	} while (s & BUSY_STAT);
 
 	/* wait for IRQ and DRQ_STAT */
 	msleep(50);
-	s = ide_read_status(drive);
+	s = hwif->read_status(hwif);
 
 	if (OK_STAT(s, DRQ_STAT, BAD_R_STAT)) {
 		unsigned long flags;
@@ -324,7 +324,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		/* drive responded with ID */
 		rc = 0;
 		/* clear drive IRQ */
-		(void)ide_read_status(drive);
+		(void)hwif->read_status(hwif);
 		local_irq_restore(flags);
 	} else {
 		/* drive refused ID */
@@ -371,7 +371,7 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 
 		ide_set_irq(drive, 0);
 		/* clear drive IRQ */
-		(void)ide_read_status(drive);
+		(void)hwif->read_status(hwif);
 		udelay(5);
 		irq = probe_irq_off(cookie);
 		if (!hwif->irq) {
@@ -396,7 +396,7 @@ static int ide_busy_sleep(ide_hwif_t *hwif)
 
 	do {
 		msleep(50);
-		stat = hwif->INB(hwif->io_ports.status_addr);
+		stat = hwif->read_status(hwif);
 		if ((stat & BUSY_STAT) == 0)
 			return 0;
 	} while (time_before(jiffies, timeout));
@@ -461,7 +461,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 		return 3;
 	}
 
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (OK_STAT(stat, READY_STAT, BUSY_STAT) ||
 	    drive->present || cmd == WIN_PIDENTIFY) {
@@ -471,7 +471,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 			rc = try_to_identify(drive,cmd);
 		}
 
-		stat = ide_read_status(drive);
+		stat = hwif->read_status(hwif);
 
 		if (stat == (BUSY_STAT | READY_STAT))
 			return 4;
@@ -488,7 +488,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 		}
 
 		/* ensure drive IRQ is clear */
-		stat = ide_read_status(drive);
+		stat = hwif->read_status(hwif);
 
 		if (rc == 1)
 			printk(KERN_ERR "%s: no response (status = 0x%02x)\n",
@@ -502,7 +502,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 		SELECT_DRIVE(&hwif->drives[0]);
 		msleep(50);
 		/* ensure drive irq is clear */
-		(void)ide_read_status(drive);
+		(void)hwif->read_status(hwif);
 	}
 	return rc;
 }
@@ -527,7 +527,7 @@ static void enable_nest (ide_drive_t *drive)
 
 	msleep(50);
 
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (!OK_STAT(stat, 0, BAD_STAT))
 		printk(KERN_CONT "failed (status = 0x%02x)\n", stat);
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 353dd11b9283..ef54728a74b0 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -927,11 +927,12 @@ static void idetape_create_mode_sense_cmd(struct ide_atapi_pc *pc, u8 page_code)
 
 static ide_startstop_t idetape_media_access_finished(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
 	struct ide_atapi_pc *pc = tape->pc;
 	u8 stat;
 
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (stat & SEEK_STAT) {
 		if (stat & ERR_STAT) {
@@ -980,6 +981,7 @@ static void ide_tape_create_rw_cmd(idetape_tape_t *tape,
 static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 					  struct request *rq, sector_t block)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	idetape_tape_t *tape = drive->driver_data;
 	struct ide_atapi_pc *pc = NULL;
 	struct request *postponed_rq = tape->postponed_rq;
@@ -1017,7 +1019,7 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	 * If the tape is still busy, postpone our request and service
 	 * the other device meanwhile.
 	 */
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (!drive->dsc_overlap && !(rq->cmd[0] & REQ_IDETAPE_PC2))
 		set_bit(IDETAPE_FLAG_IGNORE_DSC, &tape->flags);
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index c56be289417e..fc2b3957afac 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -124,7 +124,8 @@ EXPORT_SYMBOL_GPL(do_rw_taskfile);
  */
 static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
 {
-	u8 stat = ide_read_status(drive);
+	ide_hwif_t *hwif = drive->hwif;
+	u8 stat = hwif->read_status(hwif);
 
 	if (OK_STAT(stat, READY_STAT, BAD_STAT))
 		drive->mult_count = drive->mult_req;
@@ -141,10 +142,11 @@ static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
  */
 static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	int retries = 5;
 	u8 stat;
 
-	while (((stat = ide_read_status(drive)) & BUSY_STAT) && retries--)
+	while (((stat = hwif->read_status(hwif)) & BUSY_STAT) && retries--)
 		udelay(10);
 
 	if (OK_STAT(stat, READY_STAT, BAD_STAT))
@@ -162,7 +164,8 @@ static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
  */
 static ide_startstop_t recal_intr(ide_drive_t *drive)
 {
-	u8 stat = ide_read_status(drive);
+	ide_hwif_t *hwif = drive->hwif;
+	u8 stat = hwif->read_status(hwif);
 
 	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
 		return ide_error(drive, "recal_intr", stat);
@@ -174,11 +177,12 @@ static ide_startstop_t recal_intr(ide_drive_t *drive)
  */
 static ide_startstop_t task_no_data_intr(ide_drive_t *drive)
 {
-	ide_task_t *args	= HWGROUP(drive)->rq->special;
+	ide_hwif_t *hwif = drive->hwif;
+	ide_task_t *args = hwif->hwgroup->rq->special;
 	u8 stat;
 
 	local_irq_enable_in_hardirq();
-	stat = ide_read_status(drive);
+	stat = hwif->read_status(hwif);
 
 	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
 		return ide_error(drive, "task_no_data_intr", stat);
@@ -192,6 +196,7 @@ static ide_startstop_t task_no_data_intr(ide_drive_t *drive)
 
 static u8 wait_drive_not_busy(ide_drive_t *drive)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	int retries;
 	u8 stat;
 
@@ -200,7 +205,7 @@ static u8 wait_drive_not_busy(ide_drive_t *drive)
 	 * take up to 6 ms on some ATAPI devices, so we will wait max 10 ms.
 	 */
 	for (retries = 0; retries < 1000; retries++) {
-		stat = ide_read_status(drive);
+		stat = hwif->read_status(hwif);
 
 		if (stat & BUSY_STAT)
 			udelay(10);
@@ -383,8 +388,8 @@ static ide_startstop_t task_in_unexpected(ide_drive_t *drive, struct request *rq
 static ide_startstop_t task_in_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = HWGROUP(drive)->rq;
-	u8 stat = ide_read_status(drive);
+	struct request *rq = hwif->hwgroup->rq;
+	u8 stat = hwif->read_status(hwif);
 
 	/* Error? */
 	if (stat & ERR_STAT)
@@ -418,7 +423,7 @@ static ide_startstop_t task_out_intr (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq = HWGROUP(drive)->rq;
-	u8 stat = ide_read_status(drive);
+	u8 stat = hwif->read_status(hwif);
 
 	if (!OK_STAT(stat, DRIVE_READY, drive->bad_wstat))
 		return task_error(drive, rq, __func__, stat);
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index 76ce112fd857..b20e5f01ac89 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -63,6 +63,11 @@ static u8 superio_ide_inb (unsigned long port)
 	return inb(port);
 }
 
+static u8 superio_read_status(ide_hwif_t *hwif)
+{
+	return superio_ide_inb(hwif->io_ports.status_addr);
+}
+
 static u8 superio_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	return superio_ide_inb(hwif->dma_base + ATA_DMA_STATUS);
@@ -127,6 +132,7 @@ static void __devinit superio_ide_init_iops (struct hwif_s *hwif)
 	tmp = superio_ide_inb(superio_ide_dma_status[port]);
 	outb(tmp | 0x66, superio_ide_dma_status[port]);
 
+	hwif->read_status	  = superio_read_status;
 	hwif->read_sff_dma_status = superio_read_sff_dma_status;
 
 	hwif->tf_read = superio_tf_read;
@@ -283,7 +289,7 @@ static void __devinit init_hwif_ns87415 (ide_hwif_t *hwif)
 		outb(8, hwif->io_ports.ctl_addr);
 		do {
 			udelay(50);
-			stat = hwif->INB(hwif->io_ports.status_addr);
+			stat = hwif->read_status(hwif);
                 	if (stat == 0xff)
                         	break;
         	} while ((stat & BUSY_STAT) && --timeout);
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index d140dfd565a4..3d72a5e03f3d 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -134,6 +134,11 @@ static void scc_exec_command(ide_hwif_t *hwif, u8 cmd)
 	eieio();
 }
 
+static u8 scc_read_status(ide_hwif_t *hwif)
+{
+	return (u8)in_be32((void *)hwif->io_ports.status_addr);
+}
+
 static u8 scc_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	return (u8)in_be32((void *)(hwif->dma_base + 4));
@@ -788,6 +793,7 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 	ide_set_hwifdata(hwif, ports);
 
 	hwif->exec_command	  = scc_exec_command;
+	hwif->read_status	  = scc_read_status;
 	hwif->read_sff_dma_status = scc_read_sff_dma_status;
 
 	hwif->tf_load = scc_tf_load;
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 76afa1f9c599..3005a6334a58 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -127,7 +127,7 @@ sgiioc4_checkirq(ide_hwif_t * hwif)
 	return 0;
 }
 
-static u8 sgiioc4_INB(unsigned long);
+static u8 sgiioc4_read_status(ide_hwif_t *);
 
 static int
 sgiioc4_clearirq(ide_drive_t * drive)
@@ -141,18 +141,19 @@ sgiioc4_clearirq(ide_drive_t * drive)
 	intr_reg = readl((void __iomem *)other_ir);
 	if (intr_reg & 0x03) { /* Valid IOC4-IDE interrupt */
 		/*
-		 * Using sgiioc4_INB to read the Status register has a side
-		 * effect of clearing the interrupt.  The first read should
+		 * Using sgiioc4_read_status to read the Status register has a
+		 * side effect of clearing the interrupt.  The first read should
 		 * clear it if it is set.  The second read should return
 		 * a "clear" status if it got cleared.  If not, then spin
 		 * for a bit trying to clear it.
 		 */
-		u8 stat = sgiioc4_INB(io_ports->status_addr);
+		u8 stat = sgiioc4_read_status(hwif);
 		int count = 0;
-		stat = sgiioc4_INB(io_ports->status_addr);
+
+		stat = sgiioc4_read_status(hwif);
 		while ((stat & 0x80) && (count++ < 100)) {
 			udelay(1);
-			stat = sgiioc4_INB(io_ports->status_addr);
+			stat = sgiioc4_read_status(hwif);
 		}
 
 		if (intr_reg & 0x02) {
@@ -304,9 +305,9 @@ sgiioc4_dma_lost_irq(ide_drive_t * drive)
 	ide_dma_lost_irq(drive);
 }
 
-static u8
-sgiioc4_INB(unsigned long port)
+static u8 sgiioc4_read_status(ide_hwif_t *hwif)
 {
+	unsigned long port = hwif->io_ports.status_addr;
 	u8 reg = (u8) readb((void __iomem *) port);
 
 	if ((port & 0xFFF) == 0x11C) {	/* Status register of IOC4 */
@@ -628,7 +629,7 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	/* Initializing chipset IRQ Registers */
 	writel(0x03, (void __iomem *)(irqport + IOC4_INTR_SET * 4));
 
-	hwif->INB = &sgiioc4_INB;
+	hwif->read_status = sgiioc4_read_status;
 
 	idx[0] = hwif->index;
 
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 80123890ced0..2a86af91f64a 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -246,7 +246,7 @@ idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 {
 	ide_hwif_t *hwif = drive->hwif;
 
-	if (ide_read_status(drive) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
 		hwif->exec_command(hwif, WIN_IDLEIMMEDIATE);
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index d77c1994d0e3..a23ae25d7da8 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -490,6 +490,7 @@ typedef struct hwif_s {
 	const struct ide_dma_ops	*dma_ops;
 
 	void	(*exec_command)(struct hwif_s *, u8);
+	u8	(*read_status)(struct hwif_s *);
 	u8	(*read_sff_dma_status)(struct hwif_s *);
 
 	void (*tf_load)(ide_drive_t *, struct ide_task_s *);
@@ -1362,13 +1363,6 @@ static inline void ide_set_irq(ide_drive_t *drive, int on)
 		       hwif->io_ports.ctl_addr);
 }
 
-static inline u8 ide_read_status(ide_drive_t *drive)
-{
-	ide_hwif_t *hwif = drive->hwif;
-
-	return hwif->INB(hwif->io_ports.status_addr);
-}
-
 static inline u8 ide_read_altstatus(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-- 
cgit v1.2.3


From 1f6d8a0fd8f6cc5ee2219a8cf9b2da16dfd67397 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:52 +0200
Subject: ide: add ->read_altstatus method

* Remove ide_read_altstatus() inline helper.

* Add ->read_altstatus method for reading ATA Alternate Status
  register and use it instead of ->INB.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-iops.c     | 13 +++++++++++--
 drivers/ide/ide-probe.c    |  4 ++--
 drivers/ide/pci/scc_pata.c |  6 ++++++
 include/linux/ide.h        |  8 +-------
 4 files changed, 20 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 086eceaeeafd..e106954e13f9 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -119,6 +119,14 @@ static u8 ide_read_status(ide_hwif_t *hwif)
 		return inb(hwif->io_ports.status_addr);
 }
 
+static u8 ide_read_altstatus(ide_hwif_t *hwif)
+{
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
+		return readb((void __iomem *)hwif->io_ports.ctl_addr);
+	else
+		return inb(hwif->io_ports.ctl_addr);
+}
+
 static u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
@@ -349,6 +357,7 @@ void default_hwif_transport(ide_hwif_t *hwif)
 {
 	hwif->exec_command	  = ide_exec_command;
 	hwif->read_status	  = ide_read_status;
+	hwif->read_altstatus	  = ide_read_altstatus;
 	hwif->read_sff_dma_status = ide_read_sff_dma_status;
 
 	hwif->tf_load	  = ide_tf_load;
@@ -511,7 +520,7 @@ int drive_is_ready (ide_drive_t *drive)
 	 * about possible isa-pnp and pci-pnp issues yet.
 	 */
 	if (hwif->io_ports.ctl_addr)
-		stat = ide_read_altstatus(drive);
+		stat = hwif->read_altstatus(hwif);
 	else
 		/* Note: this may clear a pending IRQ!! */
 		stat = hwif->read_status(hwif);
@@ -724,7 +733,7 @@ int ide_driveid_update(ide_drive_t *drive)
 		}
 
 		msleep(50);	/* give drive a breather */
-		stat = ide_read_altstatus(drive);
+		stat = hwif->read_altstatus(hwif);
 	} while (stat & BUSY_STAT);
 
 	msleep(50);	/* wait for IRQ and DRQ_STAT */
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index c42fcfedcbf6..fe14d576ef01 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -275,7 +275,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 	msleep(50);
 
 	if (io_ports->ctl_addr) {
-		a = ide_read_altstatus(drive);
+		a = hwif->read_altstatus(hwif);
 		s = hwif->read_status(hwif);
 		if ((a ^ s) & ~INDEX_STAT)
 			/* ancient Seagate drives, broken interfaces */
@@ -306,7 +306,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		}
 		/* give drive a breather */
 		msleep(50);
-		s = use_altstatus ? ide_read_altstatus(drive)
+		s = use_altstatus ? hwif->read_altstatus(hwif)
 				  : hwif->read_status(hwif);
 	} while (s & BUSY_STAT);
 
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 3d72a5e03f3d..a89dc4780786 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -139,6 +139,11 @@ static u8 scc_read_status(ide_hwif_t *hwif)
 	return (u8)in_be32((void *)hwif->io_ports.status_addr);
 }
 
+static u8 scc_read_altstatus(ide_hwif_t *hwif)
+{
+	return (u8)in_be32((void *)hwif->io_ports.ctl_addr);
+}
+
 static u8 scc_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	return (u8)in_be32((void *)(hwif->dma_base + 4));
@@ -794,6 +799,7 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 
 	hwif->exec_command	  = scc_exec_command;
 	hwif->read_status	  = scc_read_status;
+	hwif->read_altstatus	  = scc_read_altstatus;
 	hwif->read_sff_dma_status = scc_read_sff_dma_status;
 
 	hwif->tf_load = scc_tf_load;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index a23ae25d7da8..e5fa5e868d67 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -491,6 +491,7 @@ typedef struct hwif_s {
 
 	void	(*exec_command)(struct hwif_s *, u8);
 	u8	(*read_status)(struct hwif_s *);
+	u8	(*read_altstatus)(struct hwif_s *);
 	u8	(*read_sff_dma_status)(struct hwif_s *);
 
 	void (*tf_load)(ide_drive_t *, struct ide_task_s *);
@@ -1363,13 +1364,6 @@ static inline void ide_set_irq(ide_drive_t *drive, int on)
 		       hwif->io_ports.ctl_addr);
 }
 
-static inline u8 ide_read_altstatus(ide_drive_t *drive)
-{
-	ide_hwif_t *hwif = drive->hwif;
-
-	return hwif->INB(hwif->io_ports.ctl_addr);
-}
-
 static inline u8 ide_read_error(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-- 
cgit v1.2.3


From 6e6afb3b7401f0181da74a1add57f126946b43e6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:52 +0200
Subject: ide: add ->set_irq method

Add ->set_irq method for setting nIEN bit of ATA Device Control
register and use it instead of ide_set_irq().

While at it:

* Use ->set_irq in init_irq() and do_reset1().

* Don't use HWIF() macro in ide_check_pm_state().

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c       | 14 ++++++++------
 drivers/ide/ide-iops.c     | 37 +++++++++++++++++++++++++++----------
 drivers/ide/ide-probe.c    |  9 ++++-----
 drivers/ide/ide-taskfile.c |  2 +-
 drivers/ide/pci/scc_pata.c | 19 +++++++++++++++++++
 drivers/ide/ppc/pmac.c     | 17 +++++++++++++++++
 include/linux/ide.h        | 10 ++--------
 7 files changed, 78 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index fdc221ce9920..bbd7bd4c48ee 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -746,16 +746,17 @@ static void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
 		 * the bus may be broken enough to walk on our toes at this
 		 * point.
 		 */
+		ide_hwif_t *hwif = drive->hwif;
 		int rc;
 #ifdef DEBUG_PM
 		printk("%s: Wakeup request inited, waiting for !BSY...\n", drive->name);
 #endif
-		rc = ide_wait_not_busy(HWIF(drive), 35000);
+		rc = ide_wait_not_busy(hwif, 35000);
 		if (rc)
 			printk(KERN_WARNING "%s: bus not ready on wakeup\n", drive->name);
 		SELECT_DRIVE(drive);
-		ide_set_irq(drive, 1);
-		rc = ide_wait_not_busy(HWIF(drive), 100000);
+		hwif->set_irq(hwif, 1);
+		rc = ide_wait_not_busy(hwif, 100000);
 		if (rc)
 			printk(KERN_WARNING "%s: drive not ready on wakeup\n", drive->name);
 	}
@@ -1041,7 +1042,7 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 			 * quirk_list may not like intr setups/cleanups
 			 */
 			if (drive->quirk_list != 1)
-				ide_set_irq(drive, 0);
+				hwif->set_irq(hwif, 0);
 		}
 		hwgroup->hwif = hwif;
 		hwgroup->drive = drive;
@@ -1519,6 +1520,7 @@ EXPORT_SYMBOL(ide_do_drive_cmd);
 
 void ide_pktcmd_tf_load(ide_drive_t *drive, u32 tf_flags, u16 bcount, u8 dma)
 {
+	ide_hwif_t *hwif = drive->hwif;
 	ide_task_t task;
 
 	memset(&task, 0, sizeof(task));
@@ -1529,9 +1531,9 @@ void ide_pktcmd_tf_load(ide_drive_t *drive, u32 tf_flags, u16 bcount, u8 dma)
 	task.tf.lbah    = (bcount >> 8) & 0xff;
 
 	ide_tf_dump(drive->name, &task.tf);
-	ide_set_irq(drive, 1);
+	hwif->set_irq(hwif, 1);
 	SELECT_MASK(drive, 0);
-	drive->hwif->tf_load(drive, &task);
+	hwif->tf_load(drive, &task);
 }
 
 EXPORT_SYMBOL_GPL(ide_pktcmd_tf_load);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index e106954e13f9..41ec53f329fa 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -135,6 +135,23 @@ static u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
 		return inb(hwif->dma_base + ATA_DMA_STATUS);
 }
 
+static void ide_set_irq(ide_hwif_t *hwif, int on)
+{
+	u8 ctl = ATA_DEVCTL_OBS;
+
+	if (on == 4) { /* hack for SRST */
+		ctl |= 4;
+		on &= ~4;
+	}
+
+	ctl |= on ? 0 : 2;
+
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
+		writeb(ctl, (void __iomem *)hwif->io_ports.ctl_addr);
+	else
+		outb(ctl, hwif->io_ports.ctl_addr);
+}
+
 static void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif = drive->hwif;
@@ -360,6 +377,8 @@ void default_hwif_transport(ide_hwif_t *hwif)
 	hwif->read_altstatus	  = ide_read_altstatus;
 	hwif->read_sff_dma_status = ide_read_sff_dma_status;
 
+	hwif->set_irq	  = ide_set_irq;
+
 	hwif->tf_load	  = ide_tf_load;
 	hwif->tf_read	  = ide_tf_read;
 
@@ -722,7 +741,7 @@ int ide_driveid_update(ide_drive_t *drive)
 	 */
 
 	SELECT_MASK(drive, 1);
-	ide_set_irq(drive, 0);
+	hwif->set_irq(hwif, 0);
 	msleep(50);
 	hwif->exec_command(hwif, WIN_IDENTIFY);
 	timeout = jiffies + WAIT_WORSTCASE;
@@ -808,12 +827,12 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 	SELECT_DRIVE(drive);
 	SELECT_MASK(drive, 0);
 	udelay(1);
-	ide_set_irq(drive, 0);
+	hwif->set_irq(hwif, 0);
 	hwif->OUTB(speed, io_ports->nsect_addr);
 	hwif->OUTB(SETFEATURES_XFER, io_ports->feature_addr);
 	hwif->exec_command(hwif, WIN_SETFEATURES);
 	if (drive->quirk_list == 2)
-		ide_set_irq(drive, 1);
+		hwif->set_irq(hwif, 1);
 
 	error = __ide_wait_stat(drive, drive->ready_stat,
 				BUSY_STAT|DRQ_STAT|ERR_STAT,
@@ -1129,7 +1148,6 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	ide_hwgroup_t *hwgroup;
 	struct ide_io_ports *io_ports;
 	const struct ide_port_ops *port_ops;
-	u8 ctl;
 
 	spin_lock_irqsave(&ide_lock, flags);
 	hwif = HWIF(drive);
@@ -1174,16 +1192,15 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * immediate interrupt due to the edge transition it produces.
 	 * This single interrupt gives us a "fast poll" for drives that
 	 * recover from reset very quickly, saving us the first 50ms wait time.
+	 *
+	 * TODO: add ->softreset method and stop abusing ->set_irq
 	 */
 	/* set SRST and nIEN */
-	hwif->OUTBSYNC(hwif, ATA_DEVCTL_OBS | 6, io_ports->ctl_addr);
+	hwif->set_irq(hwif, 4);
 	/* more than enough time */
 	udelay(10);
-	if (drive->quirk_list == 2)
-		ctl = ATA_DEVCTL_OBS;		/* clear SRST and nIEN */
-	else
-		ctl = ATA_DEVCTL_OBS | 2;	/* clear SRST, leave nIEN */
-	hwif->OUTBSYNC(hwif, ctl, io_ports->ctl_addr);
+	/* clear SRST, leave nIEN (unless device is on the quirk list) */
+	hwif->set_irq(hwif, drive->quirk_list == 2);
 	/* more than enough time */
 	udelay(10);
 	hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index fe14d576ef01..475bd7263184 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -361,7 +361,7 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 			autoprobe = 1;
 			cookie = probe_irq_on();
 		}
-		ide_set_irq(drive, autoprobe);
+		hwif->set_irq(hwif, autoprobe);
 	}
 
 	retval = actual_try_to_identify(drive, cmd);
@@ -369,7 +369,7 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 	if (autoprobe) {
 		int irq;
 
-		ide_set_irq(drive, 0);
+		hwif->set_irq(hwif, 0);
 		/* clear drive IRQ */
 		(void)hwif->read_status(hwif);
 		udelay(5);
@@ -709,7 +709,7 @@ static int ide_port_wait_ready(ide_hwif_t *hwif)
 		/* Ignore disks that we will not probe for later. */
 		if (!drive->noprobe || drive->present) {
 			SELECT_DRIVE(drive);
-			ide_set_irq(drive, 1);
+			hwif->set_irq(hwif, 1);
 			mdelay(2);
 			rc = ide_wait_not_busy(hwif, 35000);
 			if (rc)
@@ -1066,8 +1066,7 @@ static int init_irq (ide_hwif_t *hwif)
 			sa = IRQF_SHARED;
 
 		if (io_ports->ctl_addr)
-			/* clear nIEN */
-			hwif->OUTBSYNC(hwif, ATA_DEVCTL_OBS, io_ports->ctl_addr);
+			hwif->set_irq(hwif, 1);
 
 		if (request_irq(hwif->irq,&ide_intr,sa,hwif->name,hwgroup))
 	       		goto out_unlink;
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index fc2b3957afac..ea345369553e 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -80,7 +80,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 
 	if ((task->tf_flags & IDE_TFLAG_DMA_PIO_FALLBACK) == 0) {
 		ide_tf_dump(drive->name, tf);
-		ide_set_irq(drive, 1);
+		hwif->set_irq(hwif, 1);
 		SELECT_MASK(drive, 0);
 		hwif->tf_load(drive, task);
 	}
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index a89dc4780786..727eda6db76c 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -149,6 +149,23 @@ static u8 scc_read_sff_dma_status(ide_hwif_t *hwif)
 	return (u8)in_be32((void *)(hwif->dma_base + 4));
 }
 
+static void scc_set_irq(ide_hwif_t *hwif, int on)
+{
+	u8 ctl = ATA_DEVCTL_OBS;
+
+	if (on == 4) { /* hack for SRST */
+		ctl |= 4;
+		on &= ~4;
+	}
+
+	ctl |= on ? 0 : 2;
+
+	out_be32((void *)hwif->io_ports.ctl_addr, ctl);
+	eieio();
+	in_be32((void *)(hwif->dma_base + 0x01c));
+	eieio();
+}
+
 static void scc_ide_insw(unsigned long port, void *addr, u32 count)
 {
 	u16 *ptr = (u16 *)addr;
@@ -802,6 +819,8 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 	hwif->read_altstatus	  = scc_read_altstatus;
 	hwif->read_sff_dma_status = scc_read_sff_dma_status;
 
+	hwif->set_irq = scc_set_irq;
+
 	hwif->tf_load = scc_tf_load;
 	hwif->tf_read = scc_tf_read;
 
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 7c3a84f8fbed..a0d66480a797 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -502,6 +502,22 @@ static void pmac_exec_command(ide_hwif_t *hwif, u8 cmd)
 				     + IDE_TIMING_CONFIG));
 }
 
+static void pmac_set_irq(ide_hwif_t *hwif, int on)
+{
+	u8 ctl = ATA_DEVCTL_OBS;
+
+	if (on == 4) { /* hack for SRST */
+		ctl |= 4;
+		on &= ~4;
+	}
+
+	ctl |= on ? 0 : 2;
+
+	writeb(ctl, (void __iomem *)hwif->io_ports.ctl_addr);
+	(void)readl((void __iomem *)(hwif->io_ports.data_addr
+				     + IDE_TIMING_CONFIG));
+}
+
 /*
  * Old tuning functions (called on hdparm -p), sets up drive PIO timings
  */
@@ -1100,6 +1116,7 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 		return -ENOENT;
 
 	hwif->exec_command = pmac_exec_command;
+	hwif->set_irq	   = pmac_set_irq;
 
 	/* Setup MMIO ops */
 	default_hwif_mmiops(hwif);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e5fa5e868d67..ae93f89e4448 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -494,6 +494,8 @@ typedef struct hwif_s {
 	u8	(*read_altstatus)(struct hwif_s *);
 	u8	(*read_sff_dma_status)(struct hwif_s *);
 
+	void	(*set_irq)(struct hwif_s *, int);
+
 	void (*tf_load)(ide_drive_t *, struct ide_task_s *);
 	void (*tf_read)(ide_drive_t *, struct ide_task_s *);
 
@@ -1356,14 +1358,6 @@ static inline ide_drive_t *ide_get_paired_drive(ide_drive_t *drive)
 	return &hwif->drives[(drive->dn ^ 1) & 1];
 }
 
-static inline void ide_set_irq(ide_drive_t *drive, int on)
-{
-	ide_hwif_t *hwif = drive->hwif;
-
-	hwif->OUTBSYNC(hwif, ATA_DEVCTL_OBS | (on ? 0 : 2),
-		       hwif->io_ports.ctl_addr);
-}
-
 static inline u8 ide_read_error(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-- 
cgit v1.2.3


From 92eb43800a3c1300bd5cb8a2a27e6f2a84f7042e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:53 +0200
Subject: ide: use ->tf_read in ide_read_error()

* Add IDE_TFLAG_IN_FEATURE taskfile flag for reading Feature
  register and handle it in ->tf_read.

* Convert ide_read_error() to use ->tf_read instead of ->INB,
  then uninline and export it.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/h8300/ide-h8300.c |  2 ++
 drivers/ide/ide-iops.c        | 15 +++++++++++++++
 drivers/ide/pci/ns87415.c     |  2 ++
 drivers/ide/pci/scc_pata.c    |  2 ++
 include/linux/ide.h           | 10 +++-------
 5 files changed, 24 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index a71433b3d0f3..548a2bf232c9 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -100,6 +100,8 @@ static void h8300_tf_read(ide_drive_t *drive, ide_task_t *task)
 	/* be sure we're looking at the low order bits */
 	outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
+	if (task->tf_flags & IDE_TFLAG_IN_FEATURE)
+		tf->feature = inb(io_ports->feature_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = inb(io_ports->nsect_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_LBAL)
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 1983b353eb16..113db8744736 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -241,6 +241,8 @@ static void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
 	/* be sure we're looking at the low order bits */
 	tf_outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
+	if (task->tf_flags & IDE_TFLAG_IN_FEATURE)
+		tf->feature = tf_inb(io_ports->feature_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = tf_inb(io_ports->nsect_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_LBAL)
@@ -390,6 +392,19 @@ void default_hwif_transport(ide_hwif_t *hwif)
 	hwif->output_data = ata_output_data;
 }
 
+u8 ide_read_error(ide_drive_t *drive)
+{
+	ide_task_t task;
+
+	memset(&task, 0, sizeof(task));
+	task.tf_flags = IDE_TFLAG_IN_FEATURE;
+
+	drive->hwif->tf_read(drive, &task);
+
+	return task.tf.error;
+}
+EXPORT_SYMBOL_GPL(ide_read_error);
+
 void ide_fix_driveid (struct hd_driveid *id)
 {
 #ifndef __LITTLE_ENDIAN
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index b20e5f01ac89..b9bb8428b35e 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -88,6 +88,8 @@ static void superio_tf_read(ide_drive_t *drive, ide_task_t *task)
 	/* be sure we're looking at the low order bits */
 	outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
+	if (task->tf_flags & IDE_TFLAG_IN_FEATURE)
+		tf->feature = inb(io_ports->feature_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = inb(io_ports->nsect_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_LBAL)
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 727eda6db76c..c110f359b03e 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -741,6 +741,8 @@ static void scc_tf_read(ide_drive_t *drive, ide_task_t *task)
 	/* be sure we're looking at the low order bits */
 	scc_ide_outb(ATA_DEVCTL_OBS & ~0x80, io_ports->ctl_addr);
 
+	if (task->tf_flags & IDE_TFLAG_IN_FEATURE)
+		tf->feature = scc_ide_inb(io_ports->feature_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_NSECT)
 		tf->nsect  = scc_ide_inb(io_ports->nsect_addr);
 	if (task->tf_flags & IDE_TFLAG_IN_LBAL)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ae93f89e4448..7890768d03ed 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -889,6 +889,7 @@ enum {
 	IDE_TFLAG_IN_HOB		= IDE_TFLAG_IN_HOB_FEATURE |
 					  IDE_TFLAG_IN_HOB_NSECT |
 					  IDE_TFLAG_IN_HOB_LBA,
+	IDE_TFLAG_IN_FEATURE		= (1 << 1),
 	IDE_TFLAG_IN_NSECT		= (1 << 25),
 	IDE_TFLAG_IN_LBAL		= (1 << 26),
 	IDE_TFLAG_IN_LBAM		= (1 << 27),
@@ -956,6 +957,8 @@ void ide_tf_dump(const char *, struct ide_taskfile *);
 extern void SELECT_DRIVE(ide_drive_t *);
 void SELECT_MASK(ide_drive_t *, int);
 
+u8 ide_read_error(ide_drive_t *);
+
 extern int drive_is_ready(ide_drive_t *);
 
 void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
@@ -1357,11 +1360,4 @@ static inline ide_drive_t *ide_get_paired_drive(ide_drive_t *drive)
 
 	return &hwif->drives[(drive->dn ^ 1) & 1];
 }
-
-static inline u8 ide_read_error(ide_drive_t *drive)
-{
-	ide_hwif_t *hwif = drive->hwif;
-
-	return hwif->INB(hwif->io_ports.error_addr);
-}
 #endif /* _IDE_H */
-- 
cgit v1.2.3


From 1823649b5abb77ffe638178bc5253249d3ecd17d Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:54 +0200
Subject: ide: add ide_read_bcount_and_ireason() helper

Add ide_read_bcount_and_ireason() helper and use it instead of ->INB
in {cdrom_newpc,ide_pc}_intr().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c |  6 ++----
 drivers/ide/ide-cd.c    | 12 ++++--------
 drivers/ide/ide-iops.c  | 15 +++++++++++++++
 include/linux/ide.h     |  1 +
 4 files changed, 22 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index bfab5c4afc6a..f17a00ccbe96 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -107,11 +107,9 @@ cmd_finished:
 		ide_dma_off(drive);
 		return ide_do_reset(drive);
 	}
-	/* Get the number of bytes to transfer on this interrupt. */
-	bcount = (hwif->INB(hwif->io_ports.lbah_addr) << 8) |
-		  hwif->INB(hwif->io_ports.lbam_addr);
 
-	ireason = hwif->INB(hwif->io_ports.nsect_addr);
+	/* Get the number of bytes to transfer on this interrupt. */
+	ide_read_bcount_and_ireason(drive, &bcount, &ireason);
 
 	if (ireason & CD) {
 		printk(KERN_ERR "%s: CoD != 0 in %s\n", drive->name, __func__);
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 82879a1a89e5..563a380d0e7a 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -895,10 +895,11 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 	struct request *rq = HWGROUP(drive)->rq;
 	xfer_func_t *xferfunc;
 	ide_expiry_t *expiry = NULL;
-	int dma_error = 0, dma, stat, ireason, len, thislen, uptodate = 0;
+	int dma_error = 0, dma, stat, thislen, uptodate = 0;
 	int write = (rq_data_dir(rq) == WRITE) ? 1 : 0;
 	unsigned int timeout;
-	u8 lowcyl, highcyl;
+	u16 len;
+	u8 ireason;
 
 	/* check for errors */
 	dma = info->dma;
@@ -926,12 +927,7 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 		goto end_request;
 	}
 
-	/* ok we fall to pio :/ */
-	ireason = hwif->INB(hwif->io_ports.nsect_addr) & 0x3;
-	lowcyl  = hwif->INB(hwif->io_ports.lbam_addr);
-	highcyl = hwif->INB(hwif->io_ports.lbah_addr);
-
-	len = lowcyl + (256 * highcyl);
+	ide_read_bcount_and_ireason(drive, &len, &ireason);
 
 	thislen = blk_fs_request(rq) ? len : rq->data_len;
 	if (thislen > len)
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 113db8744736..cb11c7861a50 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -405,6 +405,21 @@ u8 ide_read_error(ide_drive_t *drive)
 }
 EXPORT_SYMBOL_GPL(ide_read_error);
 
+void ide_read_bcount_and_ireason(ide_drive_t *drive, u16 *bcount, u8 *ireason)
+{
+	ide_task_t task;
+
+	memset(&task, 0, sizeof(task));
+	task.tf_flags = IDE_TFLAG_IN_LBAH | IDE_TFLAG_IN_LBAM |
+			IDE_TFLAG_IN_NSECT;
+
+	drive->hwif->tf_read(drive, &task);
+
+	*bcount = (task.tf.lbah << 8) | task.tf.lbam;
+	*ireason = task.tf.nsect & 3;
+}
+EXPORT_SYMBOL_GPL(ide_read_bcount_and_ireason);
+
 void ide_fix_driveid (struct hd_driveid *id)
 {
 #ifndef __LITTLE_ENDIAN
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 7890768d03ed..fd05758e6995 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -958,6 +958,7 @@ extern void SELECT_DRIVE(ide_drive_t *);
 void SELECT_MASK(ide_drive_t *, int);
 
 u8 ide_read_error(ide_drive_t *);
+void ide_read_bcount_and_ireason(ide_drive_t *, u16 *, u8 *);
 
 extern int drive_is_ready(ide_drive_t *);
 
-- 
cgit v1.2.3


From 761052e676372465fdeb97c148d5a4b0790fa8a0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:54 +0200
Subject: ide: remove ->INB, ->OUTB and ->OUTBSYNC methods

* Remove no longer needed ->INB, ->OUTB and ->OUTBSYNC methods.

Then:

* Remove no longer used default_hwif_[mm]iops() and ide_[mm_]outbsync().

* Cleanup SuperIO handling in ns87415.c.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c          |  5 -----
 drivers/ide/arm/palm_bk3710.c     |  2 --
 drivers/ide/arm/rapide.c          |  2 --
 drivers/ide/h8300/ide-h8300.c     |  2 --
 drivers/ide/ide-iops.c            | 28 ------------------------
 drivers/ide/ide.c                 |  1 -
 drivers/ide/legacy/ide_platform.c |  4 +---
 drivers/ide/mips/swarm.c          |  3 ---
 drivers/ide/pci/ns87415.c         | 46 +++++++++++----------------------------
 drivers/ide/pci/scc_pata.c        | 12 ----------
 drivers/ide/pci/sgiioc4.c         |  3 ---
 drivers/ide/pci/siimage.c         |  2 +-
 drivers/ide/ppc/pmac.c            | 13 -----------
 include/linux/ide.h               |  7 ------
 14 files changed, 15 insertions(+), 115 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index f73422dd8474..0fd01d630f12 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -469,8 +469,6 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 	if (!hwif)
 		return -ENODEV;
 
-	default_hwif_mmiops(hwif);
-
 	state->hwif[0] = hwif;
 
 	ecard_set_drvdata(ec, state);
@@ -547,14 +545,11 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 		return -ENODEV;
 
 	hwif->chipset = ide_acorn;
-	default_hwif_mmiops(hwif);
 
 	idx[0] = hwif->index;
 
 	mate = ide_find_port();
 	if (mate) {
-		default_hwif_mmiops(mate);
-
 		hws[1] = &hw[1];
 		idx[1] = mate->index;
 	}
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 0229d794d909..545563bc7e23 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -399,8 +399,6 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 
 	i = hwif->index;
 
-	default_hwif_mmiops(hwif);
-
 	idx[0] = i;
 
 	ide_device_add(idx, &palm_bk3710_port_info, hws);
diff --git a/drivers/ide/arm/rapide.c b/drivers/ide/arm/rapide.c
index 01896f6e8acf..a45c2f694949 100644
--- a/drivers/ide/arm/rapide.c
+++ b/drivers/ide/arm/rapide.c
@@ -59,8 +59,6 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 		goto release;
 	}
 
-	default_hwif_mmiops(hwif);
-
 	idx[0] = hwif->index;
 
 	ide_device_add(idx, &rapide_port_info, hws);
diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index 548a2bf232c9..0795d6554913 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -171,8 +171,6 @@ static inline void hw_setup(hw_regs_t *hw)
 
 static inline void hwif_setup(ide_hwif_t *hwif)
 {
-	default_hwif_iops(hwif);
-
 	hwif->tf_load = h8300_tf_load;
 	hwif->tf_read = h8300_tf_read;
 
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index cb11c7861a50..6c54fe1f9841 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -42,18 +42,6 @@ static void ide_outb (u8 val, unsigned long port)
 	outb(val, port);
 }
 
-static void ide_outbsync(ide_hwif_t *hwif, u8 addr, unsigned long port)
-{
-	outb(addr, port);
-}
-
-void default_hwif_iops (ide_hwif_t *hwif)
-{
-	hwif->OUTB	= ide_outb;
-	hwif->OUTBSYNC	= ide_outbsync;
-	hwif->INB	= ide_inb;
-}
-
 /*
  *	MMIO operations, typically used for SATA controllers
  */
@@ -68,22 +56,6 @@ static void ide_mm_outb (u8 value, unsigned long port)
 	writeb(value, (void __iomem *) port);
 }
 
-static void ide_mm_outbsync(ide_hwif_t *hwif, u8 value, unsigned long port)
-{
-	writeb(value, (void __iomem *) port);
-}
-
-void default_hwif_mmiops (ide_hwif_t *hwif)
-{
-	hwif->OUTB	= ide_mm_outb;
-	/* Most systems will need to override OUTBSYNC, alas however
-	   this one is controller specific! */
-	hwif->OUTBSYNC	= ide_mm_outbsync;
-	hwif->INB	= ide_mm_inb;
-}
-
-EXPORT_SYMBOL(default_hwif_mmiops);
-
 void SELECT_DRIVE (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index b6018f7b0907..434dd02a4bdc 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -101,7 +101,6 @@ void ide_init_port_data(ide_hwif_t *hwif, unsigned int index)
 
 	init_completion(&hwif->gendev_rel_comp);
 
-	default_hwif_iops(hwif);
 	default_hwif_transport(hwif);
 
 	ide_port_init_devices_data(hwif);
diff --git a/drivers/ide/legacy/ide_platform.c b/drivers/ide/legacy/ide_platform.c
index 609da0d43196..3d71e336a221 100644
--- a/drivers/ide/legacy/ide_platform.c
+++ b/drivers/ide/legacy/ide_platform.c
@@ -103,10 +103,8 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 	plat_ide_setup_ports(&hw, base, alt_base, pdata, res_irq->start);
 	hw.dev = &pdev->dev;
 
-	if (mmio) {
+	if (mmio)
 		d.host_flags |= IDE_HFLAG_MMIO;
-		default_hwif_mmiops(hwif);
-	}
 
 	idx[0] = hwif->index;
 
diff --git a/drivers/ide/mips/swarm.c b/drivers/ide/mips/swarm.c
index 6da6844d2d8c..c1ffb83a2de7 100644
--- a/drivers/ide/mips/swarm.c
+++ b/drivers/ide/mips/swarm.c
@@ -120,9 +120,6 @@ static int __devinit swarm_ide_probe(struct device *dev)
 	if (hwif == NULL)
 		goto err;
 
-	/* Setup MMIO ops. */
-	default_hwif_mmiops(hwif);
-
 	idx[0] = hwif->index;
 
 	ide_device_add(idx, &swarm_port_info, hws);
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index b9bb8428b35e..02fd3a877afa 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -28,10 +28,6 @@
  */
 #include <asm/superio.h>
 
-static unsigned long superio_ide_status[2];
-static unsigned long superio_ide_select[2];
-static unsigned long superio_ide_dma_status[2];
-
 #define SUPERIO_IDE_MAX_RETRIES 25
 
 /* Because of a defect in Super I/O, all reads of the PCI DMA status 
@@ -40,27 +36,18 @@ static unsigned long superio_ide_dma_status[2];
  */
 static u8 superio_ide_inb (unsigned long port)
 {
-	if (port == superio_ide_status[0] ||
-	    port == superio_ide_status[1] ||
-	    port == superio_ide_select[0] ||
-	    port == superio_ide_select[1] ||
-	    port == superio_ide_dma_status[0] ||
-	    port == superio_ide_dma_status[1]) {
-		u8 tmp;
-		int retries = SUPERIO_IDE_MAX_RETRIES;
-
-		/* printk(" [ reading port 0x%x with retry ] ", port); */
+	u8 tmp;
+	int retries = SUPERIO_IDE_MAX_RETRIES;
 
-		do {
-			tmp = inb(port);
-			if (tmp == 0)
-				udelay(50);
-		} while (tmp == 0 && retries-- > 0);
+	/* printk(" [ reading port 0x%x with retry ] ", port); */
 
-		return tmp;
-	}
+	do {
+		tmp = inb(port);
+		if (tmp == 0)
+			udelay(50);
+	} while (tmp == 0 && retries-- > 0);
 
-	return inb(port);
+	return tmp;
 }
 
 static u8 superio_read_status(ide_hwif_t *hwif)
@@ -120,27 +107,20 @@ static void superio_tf_read(ide_drive_t *drive, ide_task_t *task)
 static void __devinit superio_ide_init_iops (struct hwif_s *hwif)
 {
 	struct pci_dev *pdev = to_pci_dev(hwif->dev);
-	u32 base, dmabase;
+	u32 dma_stat;
 	u8 port = hwif->channel, tmp;
 
-	base = pci_resource_start(pdev, port * 2) & ~3;
-	dmabase = pci_resource_start(pdev, 4) & ~3;
-
-	superio_ide_status[port] = base + 7;
-	superio_ide_select[port] = base + 6;
-	superio_ide_dma_status[port] = dmabase + (!port ? 2 : 0xa);
+	dma_stat = (pci_resource_start(pdev, 4) & ~3) + (!port ? 2 : 0xa);
 
 	/* Clear error/interrupt, enable dma */
-	tmp = superio_ide_inb(superio_ide_dma_status[port]);
-	outb(tmp | 0x66, superio_ide_dma_status[port]);
+	tmp = superio_ide_inb(dma_stat);
+	outb(tmp | 0x66, dma_stat);
 
 	hwif->read_status	  = superio_read_status;
 	hwif->read_sff_dma_status = superio_read_sff_dma_status;
 
 	hwif->tf_read = superio_tf_read;
 
-	/* We need to override inb to workaround a SuperIO errata */
-	hwif->INB = superio_ide_inb;
 }
 
 static void __devinit init_iops_ns87415(ide_hwif_t *hwif)
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index c110f359b03e..38765d9b0314 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -188,14 +188,6 @@ static void scc_ide_outb(u8 addr, unsigned long port)
 	out_be32((void*)port, addr);
 }
 
-static void scc_ide_outbsync(ide_hwif_t *hwif, u8 addr, unsigned long port)
-{
-	out_be32((void*)port, addr);
-	eieio();
-	in_be32((void*)(hwif->dma_base + 0x01c));
-	eieio();
-}
-
 static void
 scc_ide_outsw(unsigned long port, void *addr, u32 count)
 {
@@ -829,10 +821,6 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 	hwif->input_data  = scc_input_data;
 	hwif->output_data = scc_output_data;
 
-	hwif->INB = scc_ide_inb;
-	hwif->OUTB = scc_ide_outb;
-	hwif->OUTBSYNC = scc_ide_outbsync;
-
 	hwif->dma_base = dma_base;
 	hwif->config_data = ports->ctl;
 }
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 3005a6334a58..86f7c4901837 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -623,9 +623,6 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	if (hwif == NULL)
 		goto err;
 
-	/* The IOC4 uses MMIO rather than Port IO. */
-	default_hwif_mmiops(hwif);
-
 	/* Initializing chipset IRQ Registers */
 	writel(0x03, (void __iomem *)(irqport + IOC4_INTR_SET * 4));
 
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 21d7137f7d6c..5965a35d94ae 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -601,7 +601,7 @@ static void __devinit init_mmio_iops_siimage(ide_hwif_t *hwif)
 	 *	Fill in the basic hwif bits
 	 */
 	hwif->host_flags |= IDE_HFLAG_MMIO;
-	default_hwif_mmiops(hwif);
+
 	hwif->hwif_data	= addr;
 
 	/*
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index a0d66480a797..ed073c6635a8 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -486,15 +486,6 @@ pmac_ide_do_update_timings(ide_drive_t *drive)
 		pmac_ide_selectproc(drive);
 }
 
-static void pmac_outbsync(ide_hwif_t *hwif, u8 value, unsigned long port)
-{
-	u32 tmp;
-	
-	writeb(value, (void __iomem *) port);
-	tmp = readl((void __iomem *)(hwif->io_ports.data_addr
-				     + IDE_TIMING_CONFIG));
-}
-
 static void pmac_exec_command(ide_hwif_t *hwif, u8 cmd)
 {
 	writeb(cmd, (void __iomem *)hwif->io_ports.command_addr);
@@ -1118,10 +1109,6 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 	hwif->exec_command = pmac_exec_command;
 	hwif->set_irq	   = pmac_set_irq;
 
-	/* Setup MMIO ops */
-	default_hwif_mmiops(hwif);
-       	hwif->OUTBSYNC = pmac_outbsync;
-
 	idx[0] = hwif->index;
 
 	ide_device_add(idx, &d, hws);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index fd05758e6995..0544ff7d9afc 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -504,11 +504,6 @@ typedef struct hwif_s {
 
 	void (*ide_dma_clear_irq)(ide_drive_t *drive);
 
-	void (*OUTB)(u8 addr, unsigned long port);
-	void (*OUTBSYNC)(struct hwif_s *hwif, u8 addr, unsigned long port);
-
-	u8  (*INB)(unsigned long port);
-
 	/* dma physical region descriptor table (cpu view) */
 	unsigned int	*dmatable_cpu;
 	/* dma physical region descriptor table (dma view) */
@@ -1027,8 +1022,6 @@ static inline int ide_hwif_setup_dma(ide_hwif_t *hwif,
 }
 #endif
 
-extern void default_hwif_iops(ide_hwif_t *);
-extern void default_hwif_mmiops(ide_hwif_t *);
 extern void default_hwif_transport(ide_hwif_t *);
 
 typedef struct ide_pci_enablebit_s {
-- 
cgit v1.2.3


From ba4b2e607e4e9eaa929935325dafd5c86d3b5262 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:55 +0200
Subject: ide: remove dead Virtual DMA support

Lets remove dead Virtual DMA support for now so it doesn't clutter
core IDE code (it can be bring back when there is a need for it):

* Remove IDE_HFLAG_VDMA host flag.

* Remove ide_drive_t.vdma flag.

* cs5520.c: remove stale FIXMEs, cs5520_dma_host_set() and cs5520_dma_ops
  (also there is no longer a need to set IDE_HFLAG_NO_ATAPI_DMA).

There should be no functional changes caused by this patch.

Cc: TAKADA Yoshihito <takada@mbf.nifty.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-disk.c   |  2 +-
 drivers/ide/ide-dma.c    |  5 ++---
 drivers/ide/ide-iops.c   |  3 +--
 drivers/ide/pci/cs5520.c | 30 ------------------------------
 include/linux/ide.h      |  3 ---
 5 files changed, 4 insertions(+), 39 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 3a2e80237c10..df5fe5756871 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -158,7 +158,7 @@ static void ide_tf_set_cmd(ide_drive_t *drive, ide_task_t *task, u8 dma)
 	write = (task->tf_flags & IDE_TFLAG_WRITE) ? 1 : 0;
 
 	if (dma)
-		index = drive->vdma ? 4 : 8;
+		index = 8;
 	else
 		index = drive->mult_count ? 0 : 4;
 
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index ecf60dbbe3e9..e72112efab9a 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -756,9 +756,8 @@ static int ide_tune_dma(ide_drive_t *drive)
 static int ide_dma_check(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	int vdma = (hwif->host_flags & IDE_HFLAG_VDMA)? 1 : 0;
 
-	if (!vdma && ide_tune_dma(drive))
+	if (ide_tune_dma(drive))
 		return 0;
 
 	/* TODO: always do PIO fallback */
@@ -767,7 +766,7 @@ static int ide_dma_check(ide_drive_t *drive)
 
 	ide_set_max_pio(drive);
 
-	return vdma ? 0 : -1;
+	return -1;
 }
 
 int ide_id_dma_bug(ide_drive_t *drive)
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 6c54fe1f9841..c9d15be4c48d 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -866,8 +866,7 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 
  skip:
 #ifdef CONFIG_BLK_DEV_IDEDMA
-	if ((speed >= XFER_SW_DMA_0 || (hwif->host_flags & IDE_HFLAG_VDMA)) &&
-	    drive->using_dma)
+	if (speed >= XFER_SW_DMA_0 && drive->using_dma)
 		hwif->dma_ops->dma_host_set(drive, 1);
 	else if (hwif->dma_ops)	/* check if host supports DMA */
 		ide_dma_off_quietly(drive);
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index a13f2bc0665c..d30cb4d99c45 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -62,8 +62,6 @@ static void cs5520_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	struct pci_dev *pdev = to_pci_dev(hwif->dev);
 	int controller = drive->dn > 1 ? 1 : 0;
 
-	/* FIXME: if DMA = 1 do we need to set the DMA bit here ? */
-
 	/* 8bit CAT/CRT - 8bit command timing for channel */
 	pci_write_config_byte(pdev, 0x62 + controller, 
 		(cs5520_pio_clocks[pio].recovery << 4) |
@@ -89,45 +87,17 @@ static void cs5520_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	cs5520_set_pio_mode(drive, 0);
 }
 
-/*
- *	We wrap the DMA activate to set the vdma flag. This is needed
- *	so that the IDE DMA layer issues PIO not DMA commands over the
- *	DMA channel
- *
- *	ATAPI is harder so disable it for now using IDE_HFLAG_NO_ATAPI_DMA
- */
-
-static void cs5520_dma_host_set(ide_drive_t *drive, int on)
-{
-	drive->vdma = on;
-	ide_dma_host_set(drive, on);
-}
-
 static const struct ide_port_ops cs5520_port_ops = {
 	.set_pio_mode		= cs5520_set_pio_mode,
 	.set_dma_mode		= cs5520_set_dma_mode,
 };
 
-static const struct ide_dma_ops cs5520_dma_ops = {
-	.dma_host_set		= cs5520_dma_host_set,
-	.dma_setup		= ide_dma_setup,
-	.dma_exec_cmd		= ide_dma_exec_cmd,
-	.dma_start		= ide_dma_start,
-	.dma_end		= __ide_dma_end,
-	.dma_test_irq		= ide_dma_test_irq,
-	.dma_lost_irq		= ide_dma_lost_irq,
-	.dma_timeout		= ide_dma_timeout,
-};
-
-/* FIXME: VDMA is disabled because it caused system hangs */
 #define DECLARE_CS_DEV(name_str)				\
 	{							\
 		.name		= name_str,			\
 		.port_ops	= &cs5520_port_ops,		\
-		.dma_ops	= &cs5520_dma_ops,		\
 		.host_flags	= IDE_HFLAG_ISA_PORTS |		\
 				  IDE_HFLAG_CS5520 |		\
-				  IDE_HFLAG_NO_ATAPI_DMA |	\
 				  IDE_HFLAG_ABUSE_SET_DMA_MODE, \
 		.pio_mask	= ATA_PIO4,			\
 	}
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 0544ff7d9afc..3f2889400250 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -355,7 +355,6 @@ typedef struct ide_drive_s {
 	unsigned nodma		: 1;	/* disallow DMA */
 	unsigned remap_0_to_1	: 1;	/* 0=noremap, 1=remap 0->1 (for EZDrive) */
 	unsigned blocked        : 1;	/* 1=powermanagment told us not to do anything, so sleep nicely */
-	unsigned vdma		: 1;	/* 1=doing PIO over DMA 0=doing normal DMA */
 	unsigned scsi		: 1;	/* 0=default, 1=ide-scsi emulation */
 	unsigned sleeping	: 1;	/* 1=sleeping & sleep field valid */
 	unsigned post_reset	: 1;
@@ -1097,8 +1096,6 @@ enum {
 	IDE_HFLAG_NO_IO_32BIT		= (1 << 30),
 	/* never unmask IRQs */
 	IDE_HFLAG_NO_UNMASK_IRQS	= (1 << 31),
-	/* host uses VDMA (disabled for now) */
-	IDE_HFLAG_VDMA			= 0,
 };
 
 #ifdef CONFIG_BLK_DEV_OFFBOARD
-- 
cgit v1.2.3


From 3b2a5c7149ee4af4aff8fee953f66fc846d92cea Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:56 +0200
Subject: ide: filter out "default" transfer mode values in set_xfer_rate()

* Filter out "default" transfer mode values (0x00 - default PIO mode,
  0x01 - default PIO mode w/ IORDY disabled) in write handler for obsoleted
  /proc/ide/hd?/settings:current_speed setting.

  Allowing "default" transfer mode values is a dangerous thing to do as
  we don't support programming controller to the "default" transfer mode
  and devices often use different values for the default and maximum PIO
  mode (i.e. PIO2 default and PIO4 maximum) so the controller will stay
  programmed for higher PIO mode while device will use the lower PIO mode.

  There is no functionality loss as by using special IOCTLs device can
  still be programmed to "default" transfer modes (it is only useful for
  debugging/testing purposes anyway).

* Remove no longer needed IDE_HFLAG_ABUSE_SET_DMA_MODE host flag, it was
  previously used by few host drivers to program the controller to PIO0
  timings for "default" transfer mode == 0x01 (although some host drivers
  would program invalid PIO timings instead).

* Cleanup ide_set_xfer_rate() and add BUG_ON().

Suggested-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-lib.c          | 15 ++-------------
 drivers/ide/ide-proc.c         |  4 ++--
 drivers/ide/pci/aec62xx.c      |  5 -----
 drivers/ide/pci/amd74xx.c      |  1 -
 drivers/ide/pci/cs5520.c       |  3 +--
 drivers/ide/pci/cs5535.c       |  3 +--
 drivers/ide/pci/hpt34x.c       |  1 -
 drivers/ide/pci/hpt366.c       |  1 -
 drivers/ide/pci/pdc202xx_old.c |  1 -
 drivers/ide/pci/serverworks.c  |  4 +---
 drivers/ide/pci/tc86c001.c     |  3 +--
 drivers/ide/pci/via82cxxx.c    |  1 -
 include/linux/ide.h            |  1 -
 13 files changed, 8 insertions(+), 35 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 13af72f09ec4..7ac44d515470 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -266,22 +266,11 @@ int ide_set_xfer_rate(ide_drive_t *drive, u8 rate)
 
 	rate = ide_rate_filter(drive, rate);
 
+	BUG_ON(rate < XFER_PIO_0);
+
 	if (rate >= XFER_PIO_0 && rate <= XFER_PIO_5)
 		return ide_set_pio_mode(drive, rate);
 
-	/*
-	 * TODO: transfer modes 0x00-0x07 passed from the user-space are
-	 * currently handled here which needs fixing (please note that such
-	 * case could happen iff the transfer mode has already been set on
-	 * the device by ide-proc.c::set_xfer_rate()).
-	 */
-	if (rate < XFER_PIO_0) {
-		if (hwif->host_flags & IDE_HFLAG_ABUSE_SET_DMA_MODE)
-			return ide_set_dma_mode(drive, rate);
-		else
-			return ide_config_drive_speed(drive, rate);
-	}
-
 	return ide_set_dma_mode(drive, rate);
 }
 
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index 54d57a15d59e..151c91e933da 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -345,7 +345,7 @@ static int set_xfer_rate (ide_drive_t *drive, int arg)
 	ide_task_t task;
 	int err;
 
-	if (arg < 0 || (arg > 1 && arg < XFER_PIO_0) || arg > XFER_UDMA_6)
+	if (arg < XFER_PIO_0 || arg > XFER_UDMA_6)
 		return -EINVAL;
 
 	memset(&task, 0, sizeof(task));
@@ -357,7 +357,7 @@ static int set_xfer_rate (ide_drive_t *drive, int arg)
 
 	err = ide_no_data_taskfile(drive, &task);
 
-	if (!err && arg) {
+	if (!err) {
 		ide_set_xfer_rate(drive, (u8) arg);
 		ide_driveid_update(drive);
 	}
diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index ae7a4329a581..fbc43e121e6b 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -195,7 +195,6 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.host_flags	= IDE_HFLAG_SERIALIZE |
 				  IDE_HFLAG_NO_ATAPI_DMA |
 				  IDE_HFLAG_NO_DSC |
-				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
@@ -205,7 +204,6 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.init_chipset	= init_chipset_aec62xx,
 		.port_ops	= &atp86x_port_ops,
 		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_NO_AUTODMA |
-				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
@@ -216,7 +214,6 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
 		.port_ops	= &atp86x_port_ops,
 		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
-				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 				  IDE_HFLAG_NON_BOOTABLE,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
@@ -226,7 +223,6 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.init_chipset	= init_chipset_aec62xx,
 		.port_ops	= &atp86x_port_ops,
 		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
-				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
@@ -237,7 +233,6 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
 		.port_ops	= &atp86x_port_ops,
 		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
-				  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 				  IDE_HFLAG_OFF_BOARD,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index 0bfcdd0e77b3..ef7d971031ee 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -218,7 +218,6 @@ static const struct ide_port_ops amd_port_ops = {
 
 #define IDE_HFLAGS_AMD \
 	(IDE_HFLAG_PIO_NO_BLACKLIST | \
-	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
 	 IDE_HFLAG_POST_SET_MODE | \
 	 IDE_HFLAG_IO_32BIT | \
 	 IDE_HFLAG_UNMASK_IRQS)
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index d30cb4d99c45..e8e7df1915bf 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -97,8 +97,7 @@ static const struct ide_port_ops cs5520_port_ops = {
 		.name		= name_str,			\
 		.port_ops	= &cs5520_port_ops,		\
 		.host_flags	= IDE_HFLAG_ISA_PORTS |		\
-				  IDE_HFLAG_CS5520 |		\
-				  IDE_HFLAG_ABUSE_SET_DMA_MODE, \
+				  IDE_HFLAG_CS5520,		\
 		.pio_mask	= ATA_PIO4,			\
 	}
 
diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c
index dc97c48623f3..5404fe4f701d 100644
--- a/drivers/ide/pci/cs5535.c
+++ b/drivers/ide/pci/cs5535.c
@@ -171,8 +171,7 @@ static const struct ide_port_ops cs5535_port_ops = {
 static const struct ide_port_info cs5535_chipset __devinitdata = {
 	.name		= "CS5535",
 	.port_ops	= &cs5535_port_ops,
-	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_POST_SET_MODE |
-			  IDE_HFLAG_ABUSE_SET_DMA_MODE,
+	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_POST_SET_MODE,
 	.pio_mask	= ATA_PIO4,
 	.mwdma_mask	= ATA_MWDMA2,
 	.udma_mask	= ATA_UDMA4,
diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c
index 84c36c117194..9e1d1c4741da 100644
--- a/drivers/ide/pci/hpt34x.c
+++ b/drivers/ide/pci/hpt34x.c
@@ -123,7 +123,6 @@ static const struct ide_port_ops hpt34x_port_ops = {
 #define IDE_HFLAGS_HPT34X \
 	(IDE_HFLAG_NO_ATAPI_DMA | \
 	 IDE_HFLAG_NO_DSC | \
-	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
 	 IDE_HFLAG_NO_AUTODMA)
 
 static const struct ide_port_info hpt34x_chipsets[] __devinitdata = {
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index e5651cefa395..1f1135ce7cd6 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -1409,7 +1409,6 @@ static int __devinit hpt36x_init(struct pci_dev *dev, struct pci_dev *dev2)
 
 #define IDE_HFLAGS_HPT3XX \
 	(IDE_HFLAG_NO_ATAPI_DMA | \
-	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
 	 IDE_HFLAG_OFF_BOARD)
 
 static const struct ide_port_ops hpt3xx_port_ops = {
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index 1c2f9df31129..e54dc653b8c4 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -312,7 +312,6 @@ static void __devinit pdc202ata4_fixup_irq(struct pci_dev *dev,
 
 #define IDE_HFLAGS_PDC202XX \
 	(IDE_HFLAG_ERROR_STOPS_FIFO | \
-	 IDE_HFLAG_ABUSE_SET_DMA_MODE | \
 	 IDE_HFLAG_OFF_BOARD)
 
 static const struct ide_port_ops pdc20246_port_ops = {
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index a1fb20826a5b..127ccb45e261 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -349,9 +349,7 @@ static const struct ide_port_ops svwks_port_ops = {
 	.cable_detect		= svwks_cable_detect,
 };
 
-#define IDE_HFLAGS_SVWKS \
-	(IDE_HFLAG_LEGACY_IRQS | \
-	 IDE_HFLAG_ABUSE_SET_DMA_MODE)
+#define IDE_HFLAGS_SVWKS IDE_HFLAG_LEGACY_IRQS
 
 static const struct ide_port_info serverworks_chipsets[] __devinitdata = {
 	{	/* 0 */
diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index a81d47c55ce1..477e19790102 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -206,8 +206,7 @@ static const struct ide_port_info tc86c001_chipset __devinitdata = {
 	.init_hwif	= init_hwif_tc86c001,
 	.port_ops	= &tc86c001_port_ops,
 	.dma_ops	= &tc86c001_dma_ops,
-	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_OFF_BOARD |
-			  IDE_HFLAG_ABUSE_SET_DMA_MODE,
+	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_OFF_BOARD,
 	.pio_mask	= ATA_PIO4,
 	.mwdma_mask	= ATA_MWDMA2,
 	.udma_mask	= ATA_UDMA4,
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index e47384c70c40..09dc4803ef9d 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -425,7 +425,6 @@ static const struct ide_port_info via82cxxx_chipset __devinitdata = {
 	.enablebits	= { { 0x40, 0x02, 0x02 }, { 0x40, 0x01, 0x01 } },
 	.port_ops	= &via_port_ops,
 	.host_flags	= IDE_HFLAG_PIO_NO_BLACKLIST |
-			  IDE_HFLAG_ABUSE_SET_DMA_MODE |
 			  IDE_HFLAG_POST_SET_MODE |
 			  IDE_HFLAG_IO_32BIT,
 	.pio_mask	= ATA_PIO5,
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 3f2889400250..260e871ae880 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1085,7 +1085,6 @@ enum {
 	IDE_HFLAG_IO_32BIT		= (1 << 24),
 	/* unmask IRQs */
 	IDE_HFLAG_UNMASK_IRQS		= (1 << 25),
-	IDE_HFLAG_ABUSE_SET_DMA_MODE	= (1 << 26),
 	/* serialize ports if DMA is possible (for sl82c105) */
 	IDE_HFLAG_SERIALIZE_DMA		= (1 << 27),
 	/* force host out of "simplex" mode */
-- 
cgit v1.2.3


From d6276b5f5cc7508124de291f3ed59c6945c17ae7 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:56 +0200
Subject: ide: add 'config' field to hw_regs_t

Add 'config' field to hw_regs_t and use it to set hwif->config_data in
ide_init_port_hw(), then convert ide_legacy_init_one() to use hw->config.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c | 4 +---
 drivers/ide/ide.c       | 1 +
 include/linux/ide.h     | 1 +
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 21cea45e9f21..3cc8ade2cc4f 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1689,14 +1689,12 @@ static void ide_legacy_init_one(u8 *idx, hw_regs_t **hws, hw_regs_t *hw,
 	ide_std_init_ports(hw, base, ctl);
 	hw->irq = irq;
 	hw->chipset = d->chipset;
+	hw->config = config;
 
 	hwif = ide_find_port_slot(d);
 	if (hwif) {
 		hwif->chipset = hw->chipset;
 
-		if (config)
-			hwif->config_data = config;
-
 		hws[port_no] = hw;
 		idx[port_no] = hwif->index;
 	}
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 434dd02a4bdc..961f31c648c9 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -286,6 +286,7 @@ void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
 	hwif->dev = hw->dev;
 	hwif->gendev.parent = hw->parent ? hw->parent : hw->dev;
 	hwif->ack_intr = hw->ack_intr;
+	hwif->config_data = hw->config;
 }
 
 /*
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 260e871ae880..e340218b2a5f 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -178,6 +178,7 @@ typedef struct hw_regs_s {
 	ide_ack_intr_t	*ack_intr;		/* acknowledge interrupt */
 	hwif_chipset_t  chipset;
 	struct device	*dev, *parent;
+	unsigned long	config;
 } hw_regs_t;
 
 void ide_init_port_data(struct hwif_s *, unsigned int);
-- 
cgit v1.2.3


From 374e042c3e767ac2e5a40b78529220e0b3de793c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:56 +0200
Subject: ide: add struct ide_tp_ops (take 2)

* Add struct ide_tp_ops for transport methods.

* Add 'const struct ide_tp_ops *tp_ops' to struct ide_port_info
  and ide_hwif_t.

* Set the default hwif->tp_ops in ide_init_port_data().

* Set host driver specific hwif->tp_ops in ide_init_port().

* Export ide_exec_command(), ide_read_status(), ide_read_altstatus(),
  ide_read_sff_dma_status(), ide_set_irq(), ide_tf_{load,read}()
  and ata_{in,out}put_data().

* Convert host drivers and core code to use struct ide_tp_ops.

* Remove no longer needed default_hwif_transport().

* Cleanup ide_hwif_t from methods that are now in struct ide_tp_ops.

While at it:

* Use struct ide_port_info in falconide.c and q40ide.c.

* Rename ata_{in,out}put_data() to ide_{in,out}put_data().

v2:

* Fix missing convertion in ns87415.c.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c       |   2 +-
 drivers/ide/h8300/ide-h8300.c  |  26 +++++----
 drivers/ide/ide-atapi.c        |  13 +++--
 drivers/ide/ide-cd.c           |  12 ++--
 drivers/ide/ide-dma.c          |  12 ++--
 drivers/ide/ide-floppy.c       |   8 +--
 drivers/ide/ide-io.c           |  32 +++++------
 drivers/ide/ide-iops.c         | 122 +++++++++++++++++++++++------------------
 drivers/ide/ide-lib.c          |   2 +-
 drivers/ide/ide-probe.c        |  53 ++++++++++--------
 drivers/ide/ide-tape.c         |   8 +--
 drivers/ide/ide-taskfile.c     |  29 ++++++----
 drivers/ide/ide.c              |   2 +-
 drivers/ide/legacy/falconide.c |  27 +++++++--
 drivers/ide/legacy/q40ide.c    |  27 +++++++--
 drivers/ide/mips/au1xxx-ide.c  |  29 +++++++---
 drivers/ide/pci/ns87415.c      |  50 +++++++++--------
 drivers/ide/pci/scc_pata.c     |  29 +++++-----
 drivers/ide/pci/sgiioc4.c      |  18 +++++-
 drivers/ide/ppc/pmac.c         |  21 +++++--
 drivers/ide/setup-pci.c        |   2 +-
 drivers/scsi/ide-scsi.c        |   7 ++-
 include/linux/ide.h            |  52 ++++++++++++------
 23 files changed, 358 insertions(+), 225 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index 0fd01d630f12..0283d162f7f7 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -382,7 +382,7 @@ static void icside_dma_timeout(ide_drive_t *drive)
 	if (icside_dma_test_irq(drive))
 		return;
 
-	ide_dump_status(drive, "DMA timeout", hwif->read_status(hwif));
+	ide_dump_status(drive, "DMA timeout", hwif->tp_ops->read_status(hwif));
 
 	icside_dma_end(drive);
 }
diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index 0795d6554913..84644e150531 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -155,6 +155,21 @@ static void h8300_output_data(ide_drive_t *drive, struct request *rq,
 	mm_outsw(drive->hwif->io_ports.data_addr, buf, (len + 1) / 2);
 }
 
+static const struct ide_tp_ops h8300_tp_ops = {
+	.exec_command		= ide_exec_command,
+	.read_status		= ide_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= ide_read_sff_dma_status,
+
+	.set_irq		= ide_set_irq,
+
+	.tf_load		= h8300_tf_load,
+	.tf_read		= h8300_tf_read,
+
+	.input_data		= h8300_input_data,
+	.output_data		= h8300_output_data,
+};
+
 #define H8300_IDE_GAP (2)
 
 static inline void hw_setup(hw_regs_t *hw)
@@ -169,16 +184,8 @@ static inline void hw_setup(hw_regs_t *hw)
 	hw->chipset = ide_generic;
 }
 
-static inline void hwif_setup(ide_hwif_t *hwif)
-{
-	hwif->tf_load = h8300_tf_load;
-	hwif->tf_read = h8300_tf_read;
-
-	hwif->input_data  = h8300_input_data;
-	hwif->output_data = h8300_output_data;
-}
-
 static const struct ide_port_info h8300_port_info = {
+	.tp_ops			= &h8300_tp_ops,
 	.host_flags		= IDE_HFLAG_NO_IO_32BIT | IDE_HFLAG_NO_DMA,
 };
 
@@ -205,7 +212,6 @@ static int __init h8300_ide_init(void)
 		return -ENOENT;
 
 	index = hwif->index;
-	hwif_setup(hwif);
 
 	idx[0] = index;
 
diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index f17a00ccbe96..6789b81ea78d 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -22,6 +22,7 @@ ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	void (*io_buffers)(ide_drive_t *, struct ide_atapi_pc *, unsigned, int))
 {
 	ide_hwif_t *hwif = drive->hwif;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	xfer_func_t *xferfunc;
 	unsigned int temp;
 	u16 bcount;
@@ -35,7 +36,7 @@ ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	}
 
 	/* Clear the interrupt */
-	stat = hwif->read_status(hwif);
+	stat = tp_ops->read_status(hwif);
 
 	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
 		if (hwif->dma_ops->dma_end(drive) ||
@@ -140,7 +141,7 @@ cmd_finished:
 					if (pc->sg)
 						io_buffers(drive, pc, temp, 0);
 					else
-						hwif->input_data(drive, NULL,
+						tp_ops->input_data(drive, NULL,
 							pc->cur_pos, temp);
 					printk(KERN_ERR "%s: transferred %d of "
 							"%d bytes\n",
@@ -157,9 +158,9 @@ cmd_finished:
 			debug_log("The device wants to send us more data than "
 				  "expected - allowing transfer\n");
 		}
-		xferfunc = hwif->input_data;
+		xferfunc = tp_ops->input_data;
 	} else
-		xferfunc = hwif->output_data;
+		xferfunc = tp_ops->output_data;
 
 	if ((drive->media == ide_floppy && !scsi && !pc->buf) ||
 	    (drive->media == ide_tape && !scsi && pc->bh) ||
@@ -188,7 +189,7 @@ static u8 ide_read_ireason(ide_drive_t *drive)
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_IN_NSECT;
 
-	drive->hwif->tf_read(drive, &task);
+	drive->hwif->tp_ops->tf_read(drive, &task);
 
 	return task.tf.nsect & 3;
 }
@@ -249,7 +250,7 @@ ide_startstop_t ide_transfer_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 
 	/* Send the actual packet */
 	if ((pc->flags & PC_FLAG_ZIP_DRIVE) == 0)
-		hwif->output_data(drive, NULL, pc->c, 12);
+		hwif->tp_ops->output_data(drive, NULL, pc->c, 12);
 
 	return ide_started;
 }
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 563a380d0e7a..d9798ca433ba 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -285,7 +285,7 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 	int stat, err, sense_key;
 
 	/* check for errors */
-	stat = hwif->read_status(hwif);
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (stat_ret)
 		*stat_ret = stat;
@@ -590,7 +590,7 @@ static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive,
 		cmd_len = ATAPI_MIN_CDB_BYTES;
 
 	/* send the command to the device */
-	hwif->output_data(drive, NULL, rq->cmd, cmd_len);
+	hwif->tp_ops->output_data(drive, NULL, rq->cmd, cmd_len);
 
 	/* start the DMA if need be */
 	if (info->dma)
@@ -627,7 +627,7 @@ static int ide_cd_check_ireason(ide_drive_t *drive, struct request *rq,
 		 * Some drives (ASUS) seem to tell us that status info is
 		 * available.  Just get it and ignore.
 		 */
-		(void)hwif->read_status(hwif);
+		(void)hwif->tp_ops->read_status(hwif);
 		return 0;
 	} else {
 		/* drive wants a command packet, or invalid ireason... */
@@ -990,10 +990,10 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 
 	if (ireason == 0) {
 		write = 1;
-		xferfunc = hwif->output_data;
+		xferfunc = hwif->tp_ops->output_data;
 	} else {
 		write = 0;
-		xferfunc = hwif->input_data;
+		xferfunc = hwif->tp_ops->input_data;
 	}
 
 	/* transfer data */
@@ -1200,7 +1200,7 @@ static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 		if (info->cd_flags & IDE_CD_FLAG_SEEKING) {
 			ide_hwif_t *hwif = drive->hwif;
 			unsigned long elapsed = jiffies - info->start_seek;
-			int stat = hwif->read_status(hwif);
+			int stat = hwif->tp_ops->read_status(hwif);
 
 			if ((stat & SEEK_STAT) != SEEK_STAT) {
 				if (elapsed < IDECD_SEEK_TIMEOUT) {
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index e72112efab9a..be99d463dcc7 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -104,7 +104,7 @@ ide_startstop_t ide_dma_intr (ide_drive_t *drive)
 	u8 stat = 0, dma_stat = 0;
 
 	dma_stat = hwif->dma_ops->dma_end(drive);
-	stat = hwif->read_status(hwif);
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (OK_STAT(stat,DRIVE_READY,drive->bad_wstat|DRQ_STAT)) {
 		if (!dma_stat) {
@@ -335,7 +335,7 @@ static int config_drive_for_dma (ide_drive_t *drive)
 static int dma_timer_expiry (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	u8 dma_stat		= hwif->read_sff_dma_status(hwif);
+	u8 dma_stat		= hwif->tp_ops->read_sff_dma_status(hwif);
 
 	printk(KERN_WARNING "%s: dma_timer_expiry: dma status == 0x%02x\n",
 		drive->name, dma_stat);
@@ -370,7 +370,7 @@ void ide_dma_host_set(ide_drive_t *drive, int on)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	u8 unit			= (drive->select.b.unit & 0x01);
-	u8 dma_stat		= hwif->read_sff_dma_status(hwif);
+	u8 dma_stat		= hwif->tp_ops->read_sff_dma_status(hwif);
 
 	if (on)
 		dma_stat |= (1 << (5 + unit));
@@ -482,7 +482,7 @@ int ide_dma_setup(ide_drive_t *drive)
 		outb(reading, hwif->dma_base + ATA_DMA_CMD);
 
 	/* read DMA status for INTR & ERROR flags */
-	dma_stat = hwif->read_sff_dma_status(hwif);
+	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
 
 	/* clear INTR & ERROR flags */
 	if (mmio)
@@ -551,7 +551,7 @@ int __ide_dma_end (ide_drive_t *drive)
 	}
 
 	/* get DMA status */
-	dma_stat = hwif->read_sff_dma_status(hwif);
+	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
 
 	if (mmio)
 		/* clear the INTR & ERROR bits */
@@ -574,7 +574,7 @@ EXPORT_SYMBOL(__ide_dma_end);
 int ide_dma_test_irq(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	u8 dma_stat		= hwif->read_sff_dma_status(hwif);
+	u8 dma_stat		= hwif->tp_ops->read_sff_dma_status(hwif);
 
 	/* return 1 if INTR asserted */
 	if ((dma_stat & 4) == 4)
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 6f5294cfff23..62be2b27f236 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -247,9 +247,9 @@ static void ide_floppy_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
 
 		data = bvec_kmap_irq(bvec, &flags);
 		if (direction)
-			hwif->output_data(drive, NULL, data, count);
+			hwif->tp_ops->output_data(drive, NULL, data, count);
 		else
-			hwif->input_data(drive, NULL, data, count);
+			hwif->tp_ops->input_data(drive, NULL, data, count);
 		bvec_kunmap_irq(data, &flags);
 
 		bcount -= count;
@@ -402,7 +402,7 @@ static int idefloppy_transfer_pc(ide_drive_t *drive)
 	idefloppy_floppy_t *floppy = drive->driver_data;
 
 	/* Send the actual packet */
-	drive->hwif->output_data(drive, NULL, floppy->pc->c, 12);
+	drive->hwif->tp_ops->output_data(drive, NULL, floppy->pc->c, 12);
 
 	/* Timeout for the packet command */
 	return IDEFLOPPY_WAIT_CMD;
@@ -954,7 +954,7 @@ static int idefloppy_get_format_progress(ide_drive_t *drive, int __user *arg)
 		u8 stat;
 
 		local_irq_save(flags);
-		stat = hwif->read_status(hwif);
+		stat = hwif->tp_ops->read_status(hwif);
 		local_irq_restore(flags);
 
 		progress_indication = ((stat & SEEK_STAT) == 0) ? 0 : 0x10000;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index bbd7bd4c48ee..a896a283f27f 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -330,7 +330,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 			tf->error = err;
 			tf->status = stat;
 
-			drive->hwif->tf_read(drive, task);
+			drive->hwif->tp_ops->tf_read(drive, task);
 
 			if (task->tf_flags & IDE_TFLAG_DYN)
 				kfree(task);
@@ -381,7 +381,7 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8
 		if (err == ABRT_ERR) {
 			if (drive->select.b.lba &&
 			    /* some newer drives don't support WIN_SPECIFY */
-			    hwif->read_status(hwif) == WIN_SPECIFY)
+			    hwif->tp_ops->read_status(hwif) == WIN_SPECIFY)
 				return ide_stopped;
 		} else if ((err & BAD_CRC) == BAD_CRC) {
 			/* UDMA crc error, just retry the operation */
@@ -407,7 +407,7 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8
 		return ide_stopped;
 	}
 
-	if (hwif->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->tp_ops->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
 		rq->errors |= ERROR_RESET;
 
 	if ((rq->errors & ERROR_RESET) == ERROR_RESET) {
@@ -434,9 +434,9 @@ static ide_startstop_t ide_atapi_error(ide_drive_t *drive, struct request *rq, u
 		/* add decoding error stuff */
 	}
 
-	if (hwif->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->tp_ops->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		hwif->exec_command(hwif, WIN_IDLEIMMEDIATE);
+		hwif->tp_ops->exec_command(hwif, WIN_IDLEIMMEDIATE);
 
 	if (rq->errors >= ERROR_MAX) {
 		ide_kill_rq(drive, rq);
@@ -710,7 +710,7 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 #ifdef DEBUG
  	printk("%s: DRIVE_CMD (null)\n", drive->name);
 #endif
-	ide_end_drive_cmd(drive, hwif->read_status(hwif),
+	ide_end_drive_cmd(drive, hwif->tp_ops->read_status(hwif),
 			  ide_read_error(drive));
 
  	return ide_stopped;
@@ -755,7 +755,7 @@ static void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
 		if (rc)
 			printk(KERN_WARNING "%s: bus not ready on wakeup\n", drive->name);
 		SELECT_DRIVE(drive);
-		hwif->set_irq(hwif, 1);
+		hwif->tp_ops->set_irq(hwif, 1);
 		rc = ide_wait_not_busy(hwif, 100000);
 		if (rc)
 			printk(KERN_WARNING "%s: drive not ready on wakeup\n", drive->name);
@@ -1042,7 +1042,7 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 			 * quirk_list may not like intr setups/cleanups
 			 */
 			if (drive->quirk_list != 1)
-				hwif->set_irq(hwif, 0);
+				hwif->tp_ops->set_irq(hwif, 0);
 		}
 		hwgroup->hwif = hwif;
 		hwgroup->drive = drive;
@@ -1142,7 +1142,7 @@ static ide_startstop_t ide_dma_timeout_retry(ide_drive_t *drive, int error)
 		printk(KERN_WARNING "%s: DMA timeout error\n", drive->name);
 		(void)hwif->dma_ops->dma_end(drive);
 		ret = ide_error(drive, "dma timeout error",
-				hwif->read_status(hwif));
+				hwif->tp_ops->read_status(hwif));
 	} else {
 		printk(KERN_WARNING "%s: DMA timeout retry\n", drive->name);
 		hwif->dma_ops->dma_timeout(drive);
@@ -1267,7 +1267,7 @@ void ide_timer_expiry (unsigned long data)
 				} else
 					startstop =
 					ide_error(drive, "irq timeout",
-						  hwif->read_status(hwif));
+						  hwif->tp_ops->read_status(hwif));
 			}
 			drive->service_time = jiffies - drive->service_start;
 			spin_lock_irq(&ide_lock);
@@ -1323,7 +1323,7 @@ static void unexpected_intr (int irq, ide_hwgroup_t *hwgroup)
 	 */
 	do {
 		if (hwif->irq == irq) {
-			stat = hwif->read_status(hwif);
+			stat = hwif->tp_ops->read_status(hwif);
 
 			if (!OK_STAT(stat, READY_STAT, BAD_STAT)) {
 				/* Try to not flood the console with msgs */
@@ -1414,7 +1414,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 			 * Whack the status register, just in case
 			 * we have a leftover pending IRQ.
 			 */
-			(void)hwif->read_status(hwif);
+			(void)hwif->tp_ops->read_status(hwif);
 #endif /* CONFIG_BLK_DEV_IDEPCI */
 		}
 		spin_unlock_irqrestore(&ide_lock, flags);
@@ -1531,9 +1531,9 @@ void ide_pktcmd_tf_load(ide_drive_t *drive, u32 tf_flags, u16 bcount, u8 dma)
 	task.tf.lbah    = (bcount >> 8) & 0xff;
 
 	ide_tf_dump(drive->name, &task.tf);
-	hwif->set_irq(hwif, 1);
+	hwif->tp_ops->set_irq(hwif, 1);
 	SELECT_MASK(drive, 0);
-	hwif->tf_load(drive, &task);
+	hwif->tp_ops->tf_load(drive, &task);
 }
 
 EXPORT_SYMBOL_GPL(ide_pktcmd_tf_load);
@@ -1545,9 +1545,9 @@ void ide_pad_transfer(ide_drive_t *drive, int write, int len)
 
 	while (len > 0) {
 		if (write)
-			hwif->output_data(drive, NULL, buf, min(4, len));
+			hwif->tp_ops->output_data(drive, NULL, buf, min(4, len));
 		else
-			hwif->input_data(drive, NULL, buf, min(4, len));
+			hwif->tp_ops->input_data(drive, NULL, buf, min(4, len));
 		len -= 4;
 	}
 }
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index c9d15be4c48d..07da5fb9eaff 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -68,7 +68,7 @@ void SELECT_DRIVE (ide_drive_t *drive)
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_OUT_DEVICE;
 
-	drive->hwif->tf_load(drive, &task);
+	drive->hwif->tp_ops->tf_load(drive, &task);
 }
 
 void SELECT_MASK(ide_drive_t *drive, int mask)
@@ -79,39 +79,43 @@ void SELECT_MASK(ide_drive_t *drive, int mask)
 		port_ops->maskproc(drive, mask);
 }
 
-static void ide_exec_command(ide_hwif_t *hwif, u8 cmd)
+void ide_exec_command(ide_hwif_t *hwif, u8 cmd)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		writeb(cmd, (void __iomem *)hwif->io_ports.command_addr);
 	else
 		outb(cmd, hwif->io_ports.command_addr);
 }
+EXPORT_SYMBOL_GPL(ide_exec_command);
 
-static u8 ide_read_status(ide_hwif_t *hwif)
+u8 ide_read_status(ide_hwif_t *hwif)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		return readb((void __iomem *)hwif->io_ports.status_addr);
 	else
 		return inb(hwif->io_ports.status_addr);
 }
+EXPORT_SYMBOL_GPL(ide_read_status);
 
-static u8 ide_read_altstatus(ide_hwif_t *hwif)
+u8 ide_read_altstatus(ide_hwif_t *hwif)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		return readb((void __iomem *)hwif->io_ports.ctl_addr);
 	else
 		return inb(hwif->io_ports.ctl_addr);
 }
+EXPORT_SYMBOL_GPL(ide_read_altstatus);
 
-static u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
+u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
 {
 	if (hwif->host_flags & IDE_HFLAG_MMIO)
 		return readb((void __iomem *)(hwif->dma_base + ATA_DMA_STATUS));
 	else
 		return inb(hwif->dma_base + ATA_DMA_STATUS);
 }
+EXPORT_SYMBOL_GPL(ide_read_sff_dma_status);
 
-static void ide_set_irq(ide_hwif_t *hwif, int on)
+void ide_set_irq(ide_hwif_t *hwif, int on)
 {
 	u8 ctl = ATA_DEVCTL_OBS;
 
@@ -127,8 +131,9 @@ static void ide_set_irq(ide_hwif_t *hwif, int on)
 	else
 		outb(ctl, hwif->io_ports.ctl_addr);
 }
+EXPORT_SYMBOL_GPL(ide_set_irq);
 
-static void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
+void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
@@ -180,8 +185,9 @@ static void ide_tf_load(ide_drive_t *drive, ide_task_t *task)
 		tf_outb((tf->device & HIHI) | drive->select.all,
 			 io_ports->device_addr);
 }
+EXPORT_SYMBOL_GPL(ide_tf_load);
 
-static void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
+void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
@@ -241,6 +247,7 @@ static void ide_tf_read(ide_drive_t *drive, ide_task_t *task)
 			tf->hob_lbah    = tf_inb(io_ports->lbah_addr);
 	}
 }
+EXPORT_SYMBOL_GPL(ide_tf_read);
 
 /*
  * Some localbus EIDE interfaces require a special access sequence
@@ -263,8 +270,8 @@ static void ata_vlb_sync(unsigned long port)
  * so if an odd len is specified, be sure that there's at least one
  * extra byte allocated for the buffer.
  */
-static void ata_input_data(ide_drive_t *drive, struct request *rq,
-			   void *buf, unsigned int len)
+void ide_input_data(ide_drive_t *drive, struct request *rq, void *buf,
+		    unsigned int len)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
@@ -304,12 +311,13 @@ static void ata_input_data(ide_drive_t *drive, struct request *rq,
 			insw(data_addr, buf, len / 2);
 	}
 }
+EXPORT_SYMBOL_GPL(ide_input_data);
 
 /*
  * This is used for most PIO data transfers *to* the IDE interface
  */
-static void ata_output_data(ide_drive_t *drive, struct request *rq,
-			    void *buf, unsigned int len)
+void ide_output_data(ide_drive_t *drive, struct request *rq, void *buf,
+		     unsigned int len)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
@@ -347,22 +355,7 @@ static void ata_output_data(ide_drive_t *drive, struct request *rq,
 			outsw(data_addr, buf, len / 2);
 	}
 }
-
-void default_hwif_transport(ide_hwif_t *hwif)
-{
-	hwif->exec_command	  = ide_exec_command;
-	hwif->read_status	  = ide_read_status;
-	hwif->read_altstatus	  = ide_read_altstatus;
-	hwif->read_sff_dma_status = ide_read_sff_dma_status;
-
-	hwif->set_irq	  = ide_set_irq;
-
-	hwif->tf_load	  = ide_tf_load;
-	hwif->tf_read	  = ide_tf_read;
-
-	hwif->input_data  = ata_input_data;
-	hwif->output_data = ata_output_data;
-}
+EXPORT_SYMBOL_GPL(ide_output_data);
 
 u8 ide_read_error(ide_drive_t *drive)
 {
@@ -371,7 +364,7 @@ u8 ide_read_error(ide_drive_t *drive)
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_IN_FEATURE;
 
-	drive->hwif->tf_read(drive, &task);
+	drive->hwif->tp_ops->tf_read(drive, &task);
 
 	return task.tf.error;
 }
@@ -385,13 +378,28 @@ void ide_read_bcount_and_ireason(ide_drive_t *drive, u16 *bcount, u8 *ireason)
 	task.tf_flags = IDE_TFLAG_IN_LBAH | IDE_TFLAG_IN_LBAM |
 			IDE_TFLAG_IN_NSECT;
 
-	drive->hwif->tf_read(drive, &task);
+	drive->hwif->tp_ops->tf_read(drive, &task);
 
 	*bcount = (task.tf.lbah << 8) | task.tf.lbam;
 	*ireason = task.tf.nsect & 3;
 }
 EXPORT_SYMBOL_GPL(ide_read_bcount_and_ireason);
 
+const struct ide_tp_ops default_tp_ops = {
+	.exec_command		= ide_exec_command,
+	.read_status		= ide_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= ide_read_sff_dma_status,
+
+	.set_irq		= ide_set_irq,
+
+	.tf_load		= ide_tf_load,
+	.tf_read		= ide_tf_read,
+
+	.input_data		= ide_input_data,
+	.output_data		= ide_output_data,
+};
+
 void ide_fix_driveid (struct hd_driveid *id)
 {
 #ifndef __LITTLE_ENDIAN
@@ -545,10 +553,10 @@ int drive_is_ready (ide_drive_t *drive)
 	 * about possible isa-pnp and pci-pnp issues yet.
 	 */
 	if (hwif->io_ports.ctl_addr)
-		stat = hwif->read_altstatus(hwif);
+		stat = hwif->tp_ops->read_altstatus(hwif);
 	else
 		/* Note: this may clear a pending IRQ!! */
-		stat = hwif->read_status(hwif);
+		stat = hwif->tp_ops->read_status(hwif);
 
 	if (stat & BUSY_STAT)
 		/* drive busy:  definitely not interrupting */
@@ -574,24 +582,25 @@ EXPORT_SYMBOL(drive_is_ready);
 static int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad, unsigned long timeout, u8 *rstat)
 {
 	ide_hwif_t *hwif = drive->hwif;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	unsigned long flags;
 	int i;
 	u8 stat;
 
 	udelay(1);	/* spec allows drive 400ns to assert "BUSY" */
-	stat = hwif->read_status(hwif);
+	stat = tp_ops->read_status(hwif);
 
 	if (stat & BUSY_STAT) {
 		local_irq_set(flags);
 		timeout += jiffies;
-		while ((stat = hwif->read_status(hwif)) & BUSY_STAT) {
+		while ((stat = tp_ops->read_status(hwif)) & BUSY_STAT) {
 			if (time_after(jiffies, timeout)) {
 				/*
 				 * One last read after the timeout in case
 				 * heavy interrupt load made us not make any
 				 * progress during the timeout..
 				 */
-				stat = hwif->read_status(hwif);
+				stat = tp_ops->read_status(hwif);
 				if (!(stat & BUSY_STAT))
 					break;
 
@@ -611,7 +620,7 @@ static int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad, unsigned long ti
 	 */
 	for (i = 0; i < 10; i++) {
 		udelay(1);
-		stat = hwif->read_status(hwif);
+		stat = tp_ops->read_status(hwif);
 
 		if (OK_STAT(stat, good, bad)) {
 			*rstat = stat;
@@ -737,6 +746,7 @@ no_80w:
 int ide_driveid_update(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	struct hd_driveid *id;
 	unsigned long timeout, flags;
 	u8 stat;
@@ -747,9 +757,9 @@ int ide_driveid_update(ide_drive_t *drive)
 	 */
 
 	SELECT_MASK(drive, 1);
-	hwif->set_irq(hwif, 0);
+	tp_ops->set_irq(hwif, 0);
 	msleep(50);
-	hwif->exec_command(hwif, WIN_IDENTIFY);
+	tp_ops->exec_command(hwif, WIN_IDENTIFY);
 	timeout = jiffies + WAIT_WORSTCASE;
 	do {
 		if (time_after(jiffies, timeout)) {
@@ -758,11 +768,11 @@ int ide_driveid_update(ide_drive_t *drive)
 		}
 
 		msleep(50);	/* give drive a breather */
-		stat = hwif->read_altstatus(hwif);
+		stat = tp_ops->read_altstatus(hwif);
 	} while (stat & BUSY_STAT);
 
 	msleep(50);	/* wait for IRQ and DRQ_STAT */
-	stat = hwif->read_status(hwif);
+	stat = tp_ops->read_status(hwif);
 
 	if (!OK_STAT(stat, DRQ_STAT, BAD_R_STAT)) {
 		SELECT_MASK(drive, 0);
@@ -776,8 +786,8 @@ int ide_driveid_update(ide_drive_t *drive)
 		local_irq_restore(flags);
 		return 0;
 	}
-	hwif->input_data(drive, NULL, id, SECTOR_SIZE);
-	(void)hwif->read_status(hwif);	/* clear drive IRQ */
+	tp_ops->input_data(drive, NULL, id, SECTOR_SIZE);
+	(void)tp_ops->read_status(hwif);	/* clear drive IRQ */
 	local_irq_enable();
 	local_irq_restore(flags);
 	ide_fix_driveid(id);
@@ -798,6 +808,7 @@ int ide_driveid_update(ide_drive_t *drive)
 int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 {
 	ide_hwif_t *hwif = drive->hwif;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	int error = 0;
 	u8 stat;
 	ide_task_t task;
@@ -833,19 +844,19 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 	SELECT_DRIVE(drive);
 	SELECT_MASK(drive, 0);
 	udelay(1);
-	hwif->set_irq(hwif, 0);
+	tp_ops->set_irq(hwif, 0);
 
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_OUT_FEATURE | IDE_TFLAG_OUT_NSECT;
 	task.tf.feature = SETFEATURES_XFER;
 	task.tf.nsect   = speed;
 
-	hwif->tf_load(drive, &task);
+	tp_ops->tf_load(drive, &task);
 
-	hwif->exec_command(hwif, WIN_SETFEATURES);
+	tp_ops->exec_command(hwif, WIN_SETFEATURES);
 
 	if (drive->quirk_list == 2)
-		hwif->set_irq(hwif, 1);
+		tp_ops->set_irq(hwif, 1);
 
 	error = __ide_wait_stat(drive, drive->ready_stat,
 				BUSY_STAT|DRQ_STAT|ERR_STAT,
@@ -950,7 +961,7 @@ void ide_execute_command(ide_drive_t *drive, u8 cmd, ide_handler_t *handler,
 
 	spin_lock_irqsave(&ide_lock, flags);
 	__ide_set_handler(drive, handler, timeout, expiry);
-	hwif->exec_command(hwif, cmd);
+	hwif->tp_ops->exec_command(hwif, cmd);
 	/*
 	 * Drive takes 400nS to respond, we must avoid the IRQ being
 	 * serviced before that.
@@ -968,7 +979,7 @@ void ide_execute_pkt_cmd(ide_drive_t *drive)
 	unsigned long flags;
 
 	spin_lock_irqsave(&ide_lock, flags);
-	hwif->exec_command(hwif, WIN_PACKETCMD);
+	hwif->tp_ops->exec_command(hwif, WIN_PACKETCMD);
 	ndelay(400);
 	spin_unlock_irqrestore(&ide_lock, flags);
 }
@@ -999,7 +1010,7 @@ static ide_startstop_t atapi_reset_pollfunc (ide_drive_t *drive)
 
 	SELECT_DRIVE(drive);
 	udelay (10);
-	stat = hwif->read_status(hwif);
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (OK_STAT(stat, 0, BUSY_STAT))
 		printk("%s: ATAPI reset complete\n", drive->name);
@@ -1045,7 +1056,7 @@ static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 		}
 	}
 
-	tmp = hwif->read_status(hwif);
+	tmp = hwif->tp_ops->read_status(hwif);
 
 	if (!OK_STAT(tmp, 0, BUSY_STAT)) {
 		if (time_before(jiffies, hwgroup->poll_timeout)) {
@@ -1159,6 +1170,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	ide_hwif_t *hwif;
 	ide_hwgroup_t *hwgroup;
 	struct ide_io_ports *io_ports;
+	const struct ide_tp_ops *tp_ops;
 	const struct ide_port_ops *port_ops;
 
 	spin_lock_irqsave(&ide_lock, flags);
@@ -1167,6 +1179,8 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 
 	io_ports = &hwif->io_ports;
 
+	tp_ops = hwif->tp_ops;
+
 	/* We must not reset with running handlers */
 	BUG_ON(hwgroup->handler != NULL);
 
@@ -1175,7 +1189,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		pre_reset(drive);
 		SELECT_DRIVE(drive);
 		udelay (20);
-		hwif->exec_command(hwif, WIN_SRST);
+		tp_ops->exec_command(hwif, WIN_SRST);
 		ndelay(400);
 		hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
 		hwgroup->polling = 1;
@@ -1208,11 +1222,11 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * TODO: add ->softreset method and stop abusing ->set_irq
 	 */
 	/* set SRST and nIEN */
-	hwif->set_irq(hwif, 4);
+	tp_ops->set_irq(hwif, 4);
 	/* more than enough time */
 	udelay(10);
 	/* clear SRST, leave nIEN (unless device is on the quirk list) */
-	hwif->set_irq(hwif, drive->quirk_list == 2);
+	tp_ops->set_irq(hwif, drive->quirk_list == 2);
 	/* more than enough time */
 	udelay(10);
 	hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
@@ -1257,7 +1271,7 @@ int ide_wait_not_busy(ide_hwif_t *hwif, unsigned long timeout)
 		 * about locking issues (2.5 work ?).
 		 */
 		mdelay(1);
-		stat = hwif->read_status(hwif);
+		stat = hwif->tp_ops->read_status(hwif);
 		if ((stat & BUSY_STAT) == 0)
 			return 0;
 		/*
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 7ac44d515470..97fefabea8b8 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -325,7 +325,7 @@ static void ide_dump_sector(ide_drive_t *drive)
 	else
 		task.tf_flags = IDE_TFLAG_IN_LBA | IDE_TFLAG_IN_DEVICE;
 
-	drive->hwif->tf_read(drive, &task);
+	drive->hwif->tp_ops->tf_read(drive, &task);
 
 	if (lba48 || (tf->device & ATA_LBA))
 		printk(", LBAsect=%llu",
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 3cc8ade2cc4f..c588066295db 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -126,7 +126,7 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 
 	id = drive->id;
 	/* read 512 bytes of id info */
-	hwif->input_data(drive, NULL, id, SECTOR_SIZE);
+	hwif->tp_ops->input_data(drive, NULL, id, SECTOR_SIZE);
 
 	drive->id_read = 1;
 	local_irq_enable();
@@ -267,6 +267,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 {
 	ide_hwif_t *hwif = HWIF(drive);
 	struct ide_io_ports *io_ports = &hwif->io_ports;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	int use_altstatus = 0, rc;
 	unsigned long timeout;
 	u8 s = 0, a = 0;
@@ -275,8 +276,8 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 	msleep(50);
 
 	if (io_ports->ctl_addr) {
-		a = hwif->read_altstatus(hwif);
-		s = hwif->read_status(hwif);
+		a = tp_ops->read_altstatus(hwif);
+		s = tp_ops->read_status(hwif);
 		if ((a ^ s) & ~INDEX_STAT)
 			/* ancient Seagate drives, broken interfaces */
 			printk(KERN_INFO "%s: probing with STATUS(0x%02x) "
@@ -297,11 +298,11 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		/* disable DMA & overlap */
 		task.tf_flags = IDE_TFLAG_OUT_FEATURE;
 
-		drive->hwif->tf_load(drive, &task);
+		tp_ops->tf_load(drive, &task);
 	}
 
 	/* ask drive for ID */
-	hwif->exec_command(hwif, cmd);
+	tp_ops->exec_command(hwif, cmd);
 
 	timeout = ((cmd == WIN_IDENTIFY) ? WAIT_WORSTCASE : WAIT_PIDENTIFY) / 2;
 	timeout += jiffies;
@@ -312,13 +313,13 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		}
 		/* give drive a breather */
 		msleep(50);
-		s = use_altstatus ? hwif->read_altstatus(hwif)
-				  : hwif->read_status(hwif);
+		s = use_altstatus ? tp_ops->read_altstatus(hwif)
+				  : tp_ops->read_status(hwif);
 	} while (s & BUSY_STAT);
 
 	/* wait for IRQ and DRQ_STAT */
 	msleep(50);
-	s = hwif->read_status(hwif);
+	s = tp_ops->read_status(hwif);
 
 	if (OK_STAT(s, DRQ_STAT, BAD_R_STAT)) {
 		unsigned long flags;
@@ -330,7 +331,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 		/* drive responded with ID */
 		rc = 0;
 		/* clear drive IRQ */
-		(void)hwif->read_status(hwif);
+		(void)tp_ops->read_status(hwif);
 		local_irq_restore(flags);
 	} else {
 		/* drive refused ID */
@@ -352,6 +353,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 static int try_to_identify (ide_drive_t *drive, u8 cmd)
 {
 	ide_hwif_t *hwif = HWIF(drive);
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	int retval;
 	int autoprobe = 0;
 	unsigned long cookie = 0;
@@ -367,7 +369,7 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 			autoprobe = 1;
 			cookie = probe_irq_on();
 		}
-		hwif->set_irq(hwif, autoprobe);
+		tp_ops->set_irq(hwif, autoprobe);
 	}
 
 	retval = actual_try_to_identify(drive, cmd);
@@ -375,9 +377,9 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 	if (autoprobe) {
 		int irq;
 
-		hwif->set_irq(hwif, 0);
+		tp_ops->set_irq(hwif, 0);
 		/* clear drive IRQ */
-		(void)hwif->read_status(hwif);
+		(void)tp_ops->read_status(hwif);
 		udelay(5);
 		irq = probe_irq_off(cookie);
 		if (!hwif->irq) {
@@ -402,7 +404,7 @@ static int ide_busy_sleep(ide_hwif_t *hwif)
 
 	do {
 		msleep(50);
-		stat = hwif->read_status(hwif);
+		stat = hwif->tp_ops->read_status(hwif);
 		if ((stat & BUSY_STAT) == 0)
 			return 0;
 	} while (time_before(jiffies, timeout));
@@ -417,7 +419,7 @@ static u8 ide_read_device(ide_drive_t *drive)
 	memset(&task, 0, sizeof(task));
 	task.tf_flags = IDE_TFLAG_IN_DEVICE;
 
-	drive->hwif->tf_read(drive, &task);
+	drive->hwif->tp_ops->tf_read(drive, &task);
 
 	return task.tf.device;
 }
@@ -446,6 +448,7 @@ static u8 ide_read_device(ide_drive_t *drive)
 static int do_probe (ide_drive_t *drive, u8 cmd)
 {
 	ide_hwif_t *hwif = HWIF(drive);
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	int rc;
 	u8 stat;
 
@@ -478,7 +481,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 		return 3;
 	}
 
-	stat = hwif->read_status(hwif);
+	stat = tp_ops->read_status(hwif);
 
 	if (OK_STAT(stat, READY_STAT, BUSY_STAT) ||
 	    drive->present || cmd == WIN_PIDENTIFY) {
@@ -488,7 +491,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 			rc = try_to_identify(drive,cmd);
 		}
 
-		stat = hwif->read_status(hwif);
+		stat = tp_ops->read_status(hwif);
 
 		if (stat == (BUSY_STAT | READY_STAT))
 			return 4;
@@ -499,13 +502,13 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 			msleep(50);
 			SELECT_DRIVE(drive);
 			msleep(50);
-			hwif->exec_command(hwif, WIN_SRST);
+			tp_ops->exec_command(hwif, WIN_SRST);
 			(void)ide_busy_sleep(hwif);
 			rc = try_to_identify(drive, cmd);
 		}
 
 		/* ensure drive IRQ is clear */
-		stat = hwif->read_status(hwif);
+		stat = tp_ops->read_status(hwif);
 
 		if (rc == 1)
 			printk(KERN_ERR "%s: no response (status = 0x%02x)\n",
@@ -519,7 +522,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 		SELECT_DRIVE(&hwif->drives[0]);
 		msleep(50);
 		/* ensure drive irq is clear */
-		(void)hwif->read_status(hwif);
+		(void)tp_ops->read_status(hwif);
 	}
 	return rc;
 }
@@ -530,12 +533,13 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 static void enable_nest (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = HWIF(drive);
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	u8 stat;
 
 	printk("%s: enabling %s -- ", hwif->name, drive->id->model);
 	SELECT_DRIVE(drive);
 	msleep(50);
-	hwif->exec_command(hwif, EXABYTE_ENABLE_NEST);
+	tp_ops->exec_command(hwif, EXABYTE_ENABLE_NEST);
 
 	if (ide_busy_sleep(hwif)) {
 		printk(KERN_CONT "failed (timeout)\n");
@@ -544,7 +548,7 @@ static void enable_nest (ide_drive_t *drive)
 
 	msleep(50);
 
-	stat = hwif->read_status(hwif);
+	stat = tp_ops->read_status(hwif);
 
 	if (!OK_STAT(stat, 0, BAD_STAT))
 		printk(KERN_CONT "failed (status = 0x%02x)\n", stat);
@@ -726,7 +730,7 @@ static int ide_port_wait_ready(ide_hwif_t *hwif)
 		/* Ignore disks that we will not probe for later. */
 		if (!drive->noprobe || drive->present) {
 			SELECT_DRIVE(drive);
-			hwif->set_irq(hwif, 1);
+			hwif->tp_ops->set_irq(hwif, 1);
 			mdelay(2);
 			rc = ide_wait_not_busy(hwif, 35000);
 			if (rc)
@@ -1083,7 +1087,7 @@ static int init_irq (ide_hwif_t *hwif)
 			sa = IRQF_SHARED;
 
 		if (io_ports->ctl_addr)
-			hwif->set_irq(hwif, 1);
+			hwif->tp_ops->set_irq(hwif, 1);
 
 		if (request_irq(hwif->irq,&ide_intr,sa,hwif->name,hwgroup))
 	       		goto out_unlink;
@@ -1361,6 +1365,9 @@ static void ide_init_port(ide_hwif_t *hwif, unsigned int port,
 	hwif->host_flags |= d->host_flags;
 	hwif->pio_mask = d->pio_mask;
 
+	if (d->tp_ops)
+		hwif->tp_ops = d->tp_ops;
+
 	/* ->set_pio_mode for DTC2278 is currently limited to port 0 */
 	if (hwif->chipset != ide_dtc2278 || hwif->channel == 0)
 		hwif->port_ops = d->port_ops;
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index ef54728a74b0..0af128826f1e 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -398,7 +398,7 @@ static void idetape_input_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
 		count = min(
 			(unsigned int)(bh->b_size - atomic_read(&bh->b_count)),
 			bcount);
-		drive->hwif->input_data(drive, NULL, bh->b_data +
+		drive->hwif->tp_ops->input_data(drive, NULL, bh->b_data +
 					atomic_read(&bh->b_count), count);
 		bcount -= count;
 		atomic_add(count, &bh->b_count);
@@ -424,7 +424,7 @@ static void idetape_output_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
 			return;
 		}
 		count = min((unsigned int)pc->b_count, (unsigned int)bcount);
-		drive->hwif->output_data(drive, NULL, pc->b_data, count);
+		drive->hwif->tp_ops->output_data(drive, NULL, pc->b_data, count);
 		bcount -= count;
 		pc->b_data += count;
 		pc->b_count -= count;
@@ -932,7 +932,7 @@ static ide_startstop_t idetape_media_access_finished(ide_drive_t *drive)
 	struct ide_atapi_pc *pc = tape->pc;
 	u8 stat;
 
-	stat = hwif->read_status(hwif);
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (stat & SEEK_STAT) {
 		if (stat & ERR_STAT) {
@@ -1019,7 +1019,7 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	 * If the tape is still busy, postpone our request and service
 	 * the other device meanwhile.
 	 */
-	stat = hwif->read_status(hwif);
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (!drive->dsc_overlap && !(rq->cmd[0] & REQ_IDETAPE_PC2))
 		set_bit(IDETAPE_FLAG_IGNORE_DSC, &tape->flags);
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index ea345369553e..aeddbbd69e86 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -64,6 +64,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct ide_taskfile *tf = &task->tf;
 	ide_handler_t *handler = NULL;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	const struct ide_dma_ops *dma_ops = hwif->dma_ops;
 
 	if (task->data_phase == TASKFILE_MULTI_IN ||
@@ -80,15 +81,15 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 
 	if ((task->tf_flags & IDE_TFLAG_DMA_PIO_FALLBACK) == 0) {
 		ide_tf_dump(drive->name, tf);
-		hwif->set_irq(hwif, 1);
+		tp_ops->set_irq(hwif, 1);
 		SELECT_MASK(drive, 0);
-		hwif->tf_load(drive, task);
+		tp_ops->tf_load(drive, task);
 	}
 
 	switch (task->data_phase) {
 	case TASKFILE_MULTI_OUT:
 	case TASKFILE_OUT:
-		hwif->exec_command(hwif, tf->command);
+		tp_ops->exec_command(hwif, tf->command);
 		ndelay(400);	/* FIXME */
 		return pre_task_out_intr(drive, task->rq);
 	case TASKFILE_MULTI_IN:
@@ -125,7 +126,7 @@ EXPORT_SYMBOL_GPL(do_rw_taskfile);
 static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	u8 stat = hwif->read_status(hwif);
+	u8 stat = hwif->tp_ops->read_status(hwif);
 
 	if (OK_STAT(stat, READY_STAT, BAD_STAT))
 		drive->mult_count = drive->mult_req;
@@ -146,8 +147,12 @@ static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 	int retries = 5;
 	u8 stat;
 
-	while (((stat = hwif->read_status(hwif)) & BUSY_STAT) && retries--)
+	while (1) {
+		stat = hwif->tp_ops->read_status(hwif);
+		if ((stat & BUSY_STAT) == 0 || retries-- == 0)
+			break;
 		udelay(10);
+	};
 
 	if (OK_STAT(stat, READY_STAT, BAD_STAT))
 		return ide_stopped;
@@ -165,7 +170,7 @@ static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 static ide_startstop_t recal_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	u8 stat = hwif->read_status(hwif);
+	u8 stat = hwif->tp_ops->read_status(hwif);
 
 	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
 		return ide_error(drive, "recal_intr", stat);
@@ -182,7 +187,7 @@ static ide_startstop_t task_no_data_intr(ide_drive_t *drive)
 	u8 stat;
 
 	local_irq_enable_in_hardirq();
-	stat = hwif->read_status(hwif);
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
 		return ide_error(drive, "task_no_data_intr", stat);
@@ -205,7 +210,7 @@ static u8 wait_drive_not_busy(ide_drive_t *drive)
 	 * take up to 6 ms on some ATAPI devices, so we will wait max 10 ms.
 	 */
 	for (retries = 0; retries < 1000; retries++) {
-		stat = hwif->read_status(hwif);
+		stat = hwif->tp_ops->read_status(hwif);
 
 		if (stat & BUSY_STAT)
 			udelay(10);
@@ -260,9 +265,9 @@ static void ide_pio_sector(ide_drive_t *drive, struct request *rq,
 
 	/* do the actual data transfer */
 	if (write)
-		hwif->output_data(drive, rq, buf, SECTOR_SIZE);
+		hwif->tp_ops->output_data(drive, rq, buf, SECTOR_SIZE);
 	else
-		hwif->input_data(drive, rq, buf, SECTOR_SIZE);
+		hwif->tp_ops->input_data(drive, rq, buf, SECTOR_SIZE);
 
 	kunmap_atomic(buf, KM_BIO_SRC_IRQ);
 #ifdef CONFIG_HIGHMEM
@@ -389,7 +394,7 @@ static ide_startstop_t task_in_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq = hwif->hwgroup->rq;
-	u8 stat = hwif->read_status(hwif);
+	u8 stat = hwif->tp_ops->read_status(hwif);
 
 	/* Error? */
 	if (stat & ERR_STAT)
@@ -423,7 +428,7 @@ static ide_startstop_t task_out_intr (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq = HWGROUP(drive)->rq;
-	u8 stat = hwif->read_status(hwif);
+	u8 stat = hwif->tp_ops->read_status(hwif);
 
 	if (!OK_STAT(stat, DRIVE_READY, drive->bad_wstat))
 		return task_error(drive, rq, __func__, stat);
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 961f31c648c9..132b504168e9 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -101,7 +101,7 @@ void ide_init_port_data(ide_hwif_t *hwif, unsigned int index)
 
 	init_completion(&hwif->gendev_rel_comp);
 
-	default_hwif_transport(hwif);
+	hwif->tp_ops = &default_tp_ops;
 
 	ide_port_init_devices_data(hwif);
 }
diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c
index 1bb2aa72cc7f..3e2c6125f031 100644
--- a/drivers/ide/legacy/falconide.c
+++ b/drivers/ide/legacy/falconide.c
@@ -66,6 +66,27 @@ static void falconide_output_data(ide_drive_t *drive, struct request *rq,
 	outsw_swapw(data_addr, buf, (len + 1) / 2);
 }
 
+/* Atari has a byte-swapped IDE interface */
+static const struct ide_tp_ops falconide_tp_ops = {
+	.exec_command		= ide_exec_command,
+	.read_status		= ide_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= ide_read_sff_dma_status,
+
+	.set_irq		= ide_set_irq,
+
+	.tf_load		= ide_tf_load,
+	.tf_read		= ide_tf_read,
+
+	.input_data		= falconide_input_data,
+	.output_data		= falconide_output_data,
+};
+
+static const struct ide_port_info falconide_port_info = {
+	.tp_ops			= &falconide_tp_ops,
+	.host_flags		= IDE_HFLAG_NO_DMA,
+};
+
 static void __init falconide_setup_ports(hw_regs_t *hw)
 {
 	int i;
@@ -111,12 +132,8 @@ static int __init falconide_init(void)
 		u8 index = hwif->index;
 		u8 idx[4] = { index, 0xff, 0xff, 0xff };
 
-		/* Atari has a byte-swapped IDE interface */
-		hwif->input_data  = falconide_input_data;
-		hwif->output_data = falconide_output_data;
-
 		ide_get_lock(NULL, NULL);
-		ide_device_add(idx, NULL, hws);
+		ide_device_add(idx, &falconide_port_info, hws);
 		ide_release_lock();
 	}
 
diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c
index fcb04b8b0238..2dc306f852a6 100644
--- a/drivers/ide/legacy/q40ide.c
+++ b/drivers/ide/legacy/q40ide.c
@@ -96,6 +96,27 @@ static void q40ide_output_data(ide_drive_t *drive, struct request *rq,
 	outsw_swapw(data_addr, buf, (len + 1) / 2);
 }
 
+/* Q40 has a byte-swapped IDE interface */
+static const struct ide_tp_ops q40ide_tp_ops = {
+	.exec_command		= ide_exec_command,
+	.read_status		= ide_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= ide_read_sff_dma_status,
+
+	.set_irq		= ide_set_irq,
+
+	.tf_load		= ide_tf_load,
+	.tf_read		= ide_tf_read,
+
+	.input_data		= q40ide_input_data,
+	.output_data		= q40ide_output_data,
+};
+
+static const struct ide_port_info q40ide_port_info = {
+	.tp_ops			= &q40ide_tp_ops,
+	.host_flags		= IDE_HFLAG_NO_DMA,
+};
+
 /* 
  * the static array is needed to have the name reported in /proc/ioports,
  * hwif->name unfortunately isn't available yet
@@ -141,16 +162,12 @@ static int __init q40ide_init(void)
 	if (hwif) {
 		hwif->chipset = ide_generic;
 
-		/* Q40 has a byte-swapped IDE interface */
-		hwif->input_data  = q40ide_input_data;
-		hwif->output_data = q40ide_output_data;
-
 		hws[i] = &hw[i];
 		idx[i] = hwif->index;
 	}
     }
 
-    ide_device_add(idx, NULL, hws);
+    ide_device_add(idx, &q40ide_port_info, hws);
 
     return 0;
 }
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 475da582fd89..ed1c9a134079 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -519,6 +519,23 @@ static void auide_setup_ports(hw_regs_t *hw, _auide_hwif *ahwif)
 	*ata_regs = ahwif->regbase + (14 << IDE_REG_SHIFT);
 }
 
+#ifdef CONFIG_BLK_DEV_IDE_AU1XXX_PIO_DBDMA
+static const struct ide_tp_ops au1xxx_tp_ops = {
+	.exec_command		= ide_exec_command,
+	.read_status		= ide_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= ide_read_sff_dma_status,
+
+	.set_irq		= ide_set_irq,
+
+	.tf_load		= ide_tf_load,
+	.tf_read		= ide_tf_read,
+
+	.input_data		= au1xxx_input_data,
+	.output_data		= au1xxx_output_data,
+};
+#endif
+
 static const struct ide_port_ops au1xxx_port_ops = {
 	.set_pio_mode		= au1xxx_set_pio_mode,
 	.set_dma_mode		= auide_set_dma_mode,
@@ -526,6 +543,9 @@ static const struct ide_port_ops au1xxx_port_ops = {
 
 static const struct ide_port_info au1xxx_port_info = {
 	.init_dma		= auide_ddma_init,
+#ifdef CONFIG_BLK_DEV_IDE_AU1XXX_PIO_DBDMA
+	.tp_ops			= &au1xxx_tp_ops,
+#endif
 	.port_ops		= &au1xxx_port_ops,
 #ifdef CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA
 	.dma_ops		= &au1xxx_dma_ops,
@@ -596,15 +616,6 @@ static int au_ide_probe(struct device *dev)
 	hw.dev = dev;
 	hw.chipset = ide_au1xxx;
 
-	/* If the user has selected DDMA assisted copies,
-	   then set up a few local I/O function entry points 
-	*/
-
-#ifdef CONFIG_BLK_DEV_IDE_AU1XXX_PIO_DBDMA	
-	hwif->input_data  = au1xxx_input_data;
-	hwif->output_data = au1xxx_output_data;
-#endif
-
 	auide_hwif.hwif                 = hwif;
 
 	idx[0] = hwif->index;
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index 02fd3a877afa..5cd2b32ff0ef 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -104,7 +104,22 @@ static void superio_tf_read(ide_drive_t *drive, ide_task_t *task)
 	}
 }
 
-static void __devinit superio_ide_init_iops (struct hwif_s *hwif)
+static const struct ide_tp_ops superio_tp_ops = {
+	.exec_command		= ide_exec_command,
+	.read_status		= superio_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= superio_read_sff_dma_status,
+
+	.set_irq		= ide_set_irq,
+
+	.tf_load		= ide_tf_load,
+	.tf_read		= superio_tf_read,
+
+	.input_data		= ide_input_data,
+	.output_data		= ide_output_data,
+};
+
+static void __devinit superio_init_iops(struct hwif_s *hwif)
 {
 	struct pci_dev *pdev = to_pci_dev(hwif->dev);
 	u32 dma_stat;
@@ -115,21 +130,6 @@ static void __devinit superio_ide_init_iops (struct hwif_s *hwif)
 	/* Clear error/interrupt, enable dma */
 	tmp = superio_ide_inb(dma_stat);
 	outb(tmp | 0x66, dma_stat);
-
-	hwif->read_status	  = superio_read_status;
-	hwif->read_sff_dma_status = superio_read_sff_dma_status;
-
-	hwif->tf_read = superio_tf_read;
-
-}
-
-static void __devinit init_iops_ns87415(ide_hwif_t *hwif)
-{
-	struct pci_dev *dev = to_pci_dev(hwif->dev);
-
-	if (PCI_SLOT(dev->devfn) == 0xE)
-		/* Built-in - assume it's under superio. */
-		superio_ide_init_iops(hwif);
 }
 #endif
 
@@ -195,7 +195,7 @@ static int ns87415_dma_end(ide_drive_t *drive)
 	u8 dma_stat = 0, dma_cmd = 0;
 
 	drive->waiting_for_dma = 0;
-	dma_stat = hwif->read_sff_dma_status(hwif);
+	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
 	/* get DMA command mode */
 	dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
 	/* stop DMA */
@@ -271,7 +271,7 @@ static void __devinit init_hwif_ns87415 (ide_hwif_t *hwif)
 		outb(8, hwif->io_ports.ctl_addr);
 		do {
 			udelay(50);
-			stat = hwif->read_status(hwif);
+			stat = hwif->tp_ops->read_status(hwif);
                 	if (stat == 0xff)
                         	break;
         	} while ((stat & BUSY_STAT) && --timeout);
@@ -306,9 +306,6 @@ static const struct ide_dma_ops ns87415_dma_ops = {
 
 static const struct ide_port_info ns87415_chipset __devinitdata = {
 	.name		= "NS87415",
-#ifdef CONFIG_SUPERIO
-	.init_iops	= init_iops_ns87415,
-#endif
 	.init_hwif	= init_hwif_ns87415,
 	.port_ops	= &ns87415_port_ops,
 	.dma_ops	= &ns87415_dma_ops,
@@ -318,7 +315,16 @@ static const struct ide_port_info ns87415_chipset __devinitdata = {
 
 static int __devinit ns87415_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &ns87415_chipset);
+	struct ide_port_info d = ns87415_chipset;
+
+#ifdef CONFIG_SUPERIO
+	if (PCI_SLOT(dev->devfn) == 0xE) {
+		/* Built-in - assume it's under superio. */
+		d.init_iops = superio_init_iops;
+		d.tp_ops = &superio_tp_ops;
+	}
+#endif
+	return ide_setup_pci_device(dev, &d);
 }
 
 static const struct pci_device_id ns87415_pci_tbl[] = {
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 38765d9b0314..5b1a0e950dfd 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -808,19 +808,6 @@ static void __devinit init_mmio_iops_scc(ide_hwif_t *hwif)
 
 	ide_set_hwifdata(hwif, ports);
 
-	hwif->exec_command	  = scc_exec_command;
-	hwif->read_status	  = scc_read_status;
-	hwif->read_altstatus	  = scc_read_altstatus;
-	hwif->read_sff_dma_status = scc_read_sff_dma_status;
-
-	hwif->set_irq = scc_set_irq;
-
-	hwif->tf_load = scc_tf_load;
-	hwif->tf_read = scc_tf_read;
-
-	hwif->input_data  = scc_input_data;
-	hwif->output_data = scc_output_data;
-
 	hwif->dma_base = dma_base;
 	hwif->config_data = ports->ctl;
 }
@@ -872,6 +859,21 @@ static void __devinit init_hwif_scc(ide_hwif_t *hwif)
 		hwif->ultra_mask = ATA_UDMA5; /* 100MHz */
 }
 
+static const struct ide_tp_ops scc_tp_ops = {
+	.exec_command		= scc_exec_command,
+	.read_status		= scc_read_status,
+	.read_altstatus		= scc_read_altstatus,
+	.read_sff_dma_status	= scc_read_sff_dma_status,
+
+	.set_irq		= scc_set_irq,
+
+	.tf_load		= scc_tf_load,
+	.tf_read		= scc_tf_read,
+
+	.input_data		= scc_input_data,
+	.output_data		= scc_output_data,
+};
+
 static const struct ide_port_ops scc_port_ops = {
 	.set_pio_mode		= scc_set_pio_mode,
 	.set_dma_mode		= scc_set_dma_mode,
@@ -895,6 +897,7 @@ static const struct ide_dma_ops scc_dma_ops = {
       .name		= name_str,			\
       .init_iops	= init_iops_scc,		\
       .init_hwif	= init_hwif_scc,		\
+      .tp_ops		= &scc_tp_ops,		\
       .port_ops		= &scc_port_ops,		\
       .dma_ops		= &scc_dma_ops,			\
       .host_flags	= IDE_HFLAG_SINGLE,		\
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 86f7c4901837..5598bd5936d9 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -550,6 +550,21 @@ static int sgiioc4_dma_setup(ide_drive_t *drive)
 	return 0;
 }
 
+static const struct ide_tp_ops sgiioc4_tp_ops = {
+	.exec_command		= ide_exec_command,
+	.read_status		= sgiioc4_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= ide_read_sff_dma_status,
+
+	.set_irq		= ide_set_irq,
+
+	.tf_load		= ide_tf_load,
+	.tf_read		= ide_tf_read,
+
+	.input_data		= ide_input_data,
+	.output_data		= ide_output_data,
+};
+
 static const struct ide_port_ops sgiioc4_port_ops = {
 	.set_dma_mode		= sgiioc4_set_dma_mode,
 	/* reset DMA engine, clear IRQs */
@@ -572,6 +587,7 @@ static const struct ide_port_info sgiioc4_port_info __devinitdata = {
 	.name			= DRV_NAME,
 	.chipset		= ide_pci,
 	.init_dma		= ide_dma_sgiioc4,
+	.tp_ops			= &sgiioc4_tp_ops,
 	.port_ops		= &sgiioc4_port_ops,
 	.dma_ops		= &sgiioc4_dma_ops,
 	.host_flags		= IDE_HFLAG_MMIO,
@@ -626,8 +642,6 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	/* Initializing chipset IRQ Registers */
 	writel(0x03, (void __iomem *)(irqport + IOC4_INTR_SET * 4));
 
-	hwif->read_status = sgiioc4_read_status;
-
 	idx[0] = hwif->index;
 
 	if (ide_device_add(idx, &d, hws))
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index ed073c6635a8..ee557d10a764 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -974,6 +974,21 @@ static void pmac_ide_init_dev(ide_drive_t *drive)
 	}
 }
 
+static const struct ide_tp_ops pmac_tp_ops = {
+	.exec_command		= pmac_exec_command,
+	.read_status		= ide_read_status,
+	.read_altstatus		= ide_read_altstatus,
+	.read_sff_dma_status	= ide_read_sff_dma_status,
+
+	.set_irq		= pmac_set_irq,
+
+	.tf_load		= ide_tf_load,
+	.tf_read		= ide_tf_read,
+
+	.input_data		= ide_input_data,
+	.output_data		= ide_output_data,
+};
+
 static const struct ide_port_ops pmac_ide_ata6_port_ops = {
 	.init_dev		= pmac_ide_init_dev,
 	.set_pio_mode		= pmac_ide_set_pio_mode,
@@ -1003,10 +1018,11 @@ static const struct ide_port_info pmac_port_info = {
 	.name			= DRV_NAME,
 	.init_dma		= pmac_ide_init_dma,
 	.chipset		= ide_pmac,
+	.tp_ops			= &pmac_tp_ops,
+	.port_ops		= &pmac_ide_port_ops,
 #ifdef CONFIG_BLK_DEV_IDEDMA_PMAC
 	.dma_ops		= &pmac_dma_ops,
 #endif
-	.port_ops		= &pmac_ide_port_ops,
 	.host_flags		= IDE_HFLAG_SET_PIO_MODE_KEEP_DMA |
 				  IDE_HFLAG_POST_SET_MODE |
 				  IDE_HFLAG_MMIO |
@@ -1106,9 +1122,6 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 	if (hwif == NULL)
 		return -ENOENT;
 
-	hwif->exec_command = pmac_exec_command;
-	hwif->set_irq	   = pmac_set_irq;
-
 	idx[0] = hwif->index;
 
 	ide_device_add(idx, &d, hws);
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index c1b609d9cb28..804c3ef245f9 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -125,7 +125,7 @@ int ide_pci_check_simplex(ide_hwif_t *hwif, const struct ide_port_info *d)
 	 * we tune the drive then try to grab DMA ownership if we want to be
 	 * the DMA end.  This has to be become dynamic to handle hot-plug.
 	 */
-	dma_stat = hwif->read_sff_dma_status(hwif);
+	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
 	if ((dma_stat & 0x80) && hwif->mate && hwif->mate->dma_base) {
 		printk(KERN_INFO "%s: simplex device: DMA disabled\n", d->name);
 		return -1;
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 2a86af91f64a..659db3f7ae08 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -142,7 +142,8 @@ static void ide_scsi_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
 				unsigned int bcount, int write)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	xfer_func_t *xf = write ? hwif->output_data : hwif->input_data;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
+	xfer_func_t *xf = write ? tp_ops->output_data : tp_ops->input_data;
 	char *buf;
 	int count;
 
@@ -246,9 +247,9 @@ idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 {
 	ide_hwif_t *hwif = drive->hwif;
 
-	if (hwif->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->tp_ops->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
 		/* force an abort */
-		hwif->exec_command(hwif, WIN_IDLEIMMEDIATE);
+		hwif->tp_ops->exec_command(hwif, WIN_IDLEIMMEDIATE);
 
 	rq->errors++;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e340218b2a5f..1286a2275efb 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -408,8 +408,28 @@ typedef struct ide_drive_s {
     ((1<<ide_pci)|(1<<ide_cmd646)|(1<<ide_ali14xx))
 #define IDE_CHIPSET_IS_PCI(c)	((IDE_CHIPSET_PCI_MASK >> (c)) & 1)
 
+struct ide_task_s;
 struct ide_port_info;
 
+struct ide_tp_ops {
+	void	(*exec_command)(struct hwif_s *, u8);
+	u8	(*read_status)(struct hwif_s *);
+	u8	(*read_altstatus)(struct hwif_s *);
+	u8	(*read_sff_dma_status)(struct hwif_s *);
+
+	void	(*set_irq)(struct hwif_s *, int);
+
+	void	(*tf_load)(ide_drive_t *, struct ide_task_s *);
+	void	(*tf_read)(ide_drive_t *, struct ide_task_s *);
+
+	void	(*input_data)(ide_drive_t *, struct request *, void *,
+			      unsigned int);
+	void	(*output_data)(ide_drive_t *, struct request *, void *,
+			       unsigned int);
+};
+
+extern const struct ide_tp_ops default_tp_ops;
+
 struct ide_port_ops {
 	/* host specific initialization of a device */
 	void	(*init_dev)(ide_drive_t *);
@@ -447,8 +467,6 @@ struct ide_dma_ops {
 	void	(*dma_timeout)(struct ide_drive_s *);
 };
 
-struct ide_task_s;
-
 typedef struct hwif_s {
 	struct hwif_s *next;		/* for linked-list in ide_hwgroup_t */
 	struct hwif_s *mate;		/* other hwif from same PCI chip */
@@ -486,22 +504,10 @@ typedef struct hwif_s {
 
 	void (*rw_disk)(ide_drive_t *, struct request *);
 
+	const struct ide_tp_ops		*tp_ops;
 	const struct ide_port_ops	*port_ops;
 	const struct ide_dma_ops	*dma_ops;
 
-	void	(*exec_command)(struct hwif_s *, u8);
-	u8	(*read_status)(struct hwif_s *);
-	u8	(*read_altstatus)(struct hwif_s *);
-	u8	(*read_sff_dma_status)(struct hwif_s *);
-
-	void	(*set_irq)(struct hwif_s *, int);
-
-	void (*tf_load)(ide_drive_t *, struct ide_task_s *);
-	void (*tf_read)(ide_drive_t *, struct ide_task_s *);
-
-	void (*input_data)(ide_drive_t *, struct request *, void *, unsigned);
-	void (*output_data)(ide_drive_t *, struct request *, void *, unsigned);
-
 	void (*ide_dma_clear_irq)(ide_drive_t *drive);
 
 	/* dma physical region descriptor table (cpu view) */
@@ -949,6 +955,19 @@ typedef struct ide_task_s {
 
 void ide_tf_dump(const char *, struct ide_taskfile *);
 
+void ide_exec_command(ide_hwif_t *, u8);
+u8 ide_read_status(ide_hwif_t *);
+u8 ide_read_altstatus(ide_hwif_t *);
+u8 ide_read_sff_dma_status(ide_hwif_t *);
+
+void ide_set_irq(ide_hwif_t *, int);
+
+void ide_tf_load(ide_drive_t *, ide_task_t *);
+void ide_tf_read(ide_drive_t *, ide_task_t *);
+
+void ide_input_data(ide_drive_t *, struct request *, void *, unsigned int);
+void ide_output_data(ide_drive_t *, struct request *, void *, unsigned int);
+
 extern void SELECT_DRIVE(ide_drive_t *);
 void SELECT_MASK(ide_drive_t *, int);
 
@@ -1022,8 +1041,6 @@ static inline int ide_hwif_setup_dma(ide_hwif_t *hwif,
 }
 #endif
 
-extern void default_hwif_transport(ide_hwif_t *);
-
 typedef struct ide_pci_enablebit_s {
 	u8	reg;	/* byte pci reg holding the enable-bit */
 	u8	mask;	/* mask to isolate the enable-bit */
@@ -1112,6 +1129,7 @@ struct ide_port_info {
 	int			(*init_dma)(ide_hwif_t *,
 					    const struct ide_port_info *);
 
+	const struct ide_tp_ops		*tp_ops;
 	const struct ide_port_ops	*port_ops;
 	const struct ide_dma_ops	*dma_ops;
 
-- 
cgit v1.2.3


From 48c3c1072651922ed153bcf0a33ea82cf20df390 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:57 +0200
Subject: ide: add struct ide_host (take 3)

* Add struct ide_host which keeps pointers to host's ports.

* Add ide_host_alloc[_all]() and ide_host_remove() helpers.

* Pass 'struct ide_host *host' instead of 'u8 *idx' to
  ide_device_add[_all]() and rename it to ide_host_register[_all]().

* Convert host drivers and core code to use struct ide_host.

* Remove no longer needed ide_find_port().

* Make ide_find_port_slot() static.

* Unexport ide_unregister().

v2:
* Add missing 'struct ide_host *host' to macide.c.

v3:
* Fix build problem in pmac.c (s/ide_alloc_host/ide_host_alloc/)
  (Noticed by Stephen Rothwell).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c          | 40 +++++------------
 drivers/ide/arm/ide_arm.c         | 12 ++---
 drivers/ide/arm/palm_bk3710.c     | 13 ++----
 drivers/ide/arm/rapide.c          | 17 +++----
 drivers/ide/h8300/ide-h8300.c     | 14 ++----
 drivers/ide/ide-generic.c         | 28 ++++--------
 drivers/ide/ide-pnp.c             | 17 +++----
 drivers/ide/ide-probe.c           | 94 ++++++++++++++++++++++++++++-----------
 drivers/ide/ide.c                 |  2 -
 drivers/ide/legacy/buddha.c       | 19 +++-----
 drivers/ide/legacy/falconide.c    | 11 ++---
 drivers/ide/legacy/gayle.c        | 16 +++----
 drivers/ide/legacy/ide-4drives.c  | 23 +++-------
 drivers/ide/legacy/ide-cs.c       | 46 +++++++++----------
 drivers/ide/legacy/ide_platform.c | 23 +++++-----
 drivers/ide/legacy/macide.c       | 12 ++---
 drivers/ide/legacy/q40ide.c       | 15 +++----
 drivers/ide/mips/au1xxx-ide.c     | 25 +++++------
 drivers/ide/mips/swarm.c          | 13 +++---
 drivers/ide/pci/cmd640.c          | 30 ++++---------
 drivers/ide/pci/cs5520.c          |  8 ++--
 drivers/ide/pci/delkin_cb.c       | 19 +++-----
 drivers/ide/pci/scc_pata.c        | 24 +++++-----
 drivers/ide/pci/sgiioc4.c         | 13 +++---
 drivers/ide/ppc/pmac.c            | 10 ++---
 drivers/ide/setup-pci.c           | 48 ++++++++------------
 include/linux/ide.h               | 21 ++++-----
 27 files changed, 264 insertions(+), 349 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index 0283d162f7f7..6fa58425466a 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -72,7 +72,7 @@ struct icside_state {
 	void __iomem *ioc_base;
 	unsigned int sel;
 	unsigned int type;
-	ide_hwif_t *hwif[2];
+	struct ide_host *host;
 };
 
 #define ICS_TYPE_A3IN	0
@@ -442,10 +442,9 @@ static void icside_setup_ports(hw_regs_t *hw, void __iomem *base,
 static int __init
 icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 {
-	ide_hwif_t *hwif;
 	void __iomem *base;
+	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	base = ecardm_iomap(ec, ECARD_RES_MEMC, 0, 0);
 	if (!base)
@@ -465,17 +464,15 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 
 	icside_setup_ports(&hw, base, &icside_cardinfo_v5, ec);
 
-	hwif = ide_find_port();
-	if (!hwif)
+	host = ide_host_alloc(NULL, hws);
+	if (host == NULL)
 		return -ENODEV;
 
-	state->hwif[0] = hwif;
+	state->host = host;
 
 	ecard_set_drvdata(ec, state);
 
-	idx[0] = hwif->index;
-
-	ide_device_add(idx, NULL, hws);
+	ide_host_register(host, NULL, hws);
 
 	return 0;
 }
@@ -492,12 +489,11 @@ static const struct ide_port_info icside_v6_port_info __initdata = {
 static int __init
 icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 {
-	ide_hwif_t *hwif, *mate;
 	void __iomem *ioc_base, *easi_base;
+	struct ide_host *host;
 	unsigned int sel = 0;
 	int ret;
 	hw_regs_t hw[2], *hws[] = { &hw[0], NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	struct ide_port_info d = icside_v6_port_info;
 
 	ioc_base = ecardm_iomap(ec, ECARD_RES_IOCFAST, 0, 0);
@@ -537,25 +533,11 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 	icside_setup_ports(&hw[0], easi_base, &icside_cardinfo_v6_1, ec);
 	icside_setup_ports(&hw[1], easi_base, &icside_cardinfo_v6_2, ec);
 
-	/*
-	 * Find and register the interfaces.
-	 */
-	hwif = ide_find_port();
-	if (hwif == NULL)
+	host = ide_host_alloc(&d, hws);
+	if (host == NULL)
 		return -ENODEV;
 
-	hwif->chipset = ide_acorn;
-
-	idx[0] = hwif->index;
-
-	mate = ide_find_port();
-	if (mate) {
-		hws[1] = &hw[1];
-		idx[1] = mate->index;
-	}
-
-	state->hwif[0]    = hwif;
-	state->hwif[1]    = mate;
+	state->host = host;
 
 	ecard_set_drvdata(ec, state);
 
@@ -565,7 +547,7 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 		d.dma_ops = NULL;
 	}
 
-	ide_device_add(idx, &d, hws);
+	ide_host_register(host, &d, hws);
 
 	return 0;
 
diff --git a/drivers/ide/arm/ide_arm.c b/drivers/ide/arm/ide_arm.c
index e9831bbd988a..9efd7a86db45 100644
--- a/drivers/ide/arm/ide_arm.c
+++ b/drivers/ide/arm/ide_arm.c
@@ -28,10 +28,9 @@
 
 static int __init ide_arm_init(void)
 {
-	ide_hwif_t *hwif;
+	struct ide_host *host;
 	unsigned long base = IDE_ARM_IO, ctl = IDE_ARM_IO + 0x206;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	if (!request_region(base, 8, DRV_NAME)) {
 		printk(KERN_ERR "%s: I/O resource 0x%lX-0x%lX not free.\n",
@@ -51,12 +50,9 @@ static int __init ide_arm_init(void)
 	hw.irq = IDE_ARM_IRQ;
 	hw.chipset = ide_generic;
 
-	hwif = ide_find_port();
-	if (hwif) {
-		idx[0] = hwif->index;
-
-		ide_device_add(idx, NULL, hws);
-	}
+	host = ide_host_alloc(NULL, hws);
+	if (host)
+		ide_host_register(host, NULL, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 545563bc7e23..24389a571c37 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -347,11 +347,10 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 {
 	struct clk *clk;
 	struct resource *mem, *irq;
-	ide_hwif_t *hwif;
+	struct ide_host *host;
 	unsigned long base, rate;
 	int i;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	clk = clk_get(NULL, "IDECLK");
 	if (IS_ERR(clk))
@@ -393,15 +392,11 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	hw.irq = irq->start;
 	hw.chipset = ide_palm3710;
 
-	hwif = ide_find_port();
-	if (hwif == NULL)
+	host = ide_host_alloc(&palm_bk3710_port_info, hws);
+	if (host == NULL)
 		goto out;
 
-	i = hwif->index;
-
-	idx[0] = i;
-
-	ide_device_add(idx, &palm_bk3710_port_info, hws);
+	ide_host_register(host, &palm_bk3710_port_info, hws);
 
 	return 0;
 out:
diff --git a/drivers/ide/arm/rapide.c b/drivers/ide/arm/rapide.c
index a45c2f694949..11f3307385de 100644
--- a/drivers/ide/arm/rapide.c
+++ b/drivers/ide/arm/rapide.c
@@ -32,11 +32,10 @@ static void rapide_setup_ports(hw_regs_t *hw, void __iomem *base,
 static int __devinit
 rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 {
-	ide_hwif_t *hwif;
 	void __iomem *base;
+	struct ide_host *host;
 	int ret;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	ret = ecard_request_resources(ec);
 	if (ret)
@@ -53,17 +52,15 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 	hw.chipset = ide_generic;
 	hw.dev = &ec->dev;
 
-	hwif = ide_find_port();
-	if (hwif == NULL) {
+	host = ide_host_alloc(&rapide_port_info, hws);
+	if (host == NULL) {
 		ret = -ENOENT;
 		goto release;
 	}
 
-	idx[0] = hwif->index;
+	ide_host_register(host, &rapide_port_info, hws);
 
-	ide_device_add(idx, &rapide_port_info, hws);
-
-	ecard_set_drvdata(ec, hwif);
+	ecard_set_drvdata(ec, host);
 	goto out;
 
  release:
@@ -74,11 +71,11 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 
 static void __devexit rapide_remove(struct expansion_card *ec)
 {
-	ide_hwif_t *hwif = ecard_get_drvdata(ec);
+	struct ide_host *host = ecard_get_drvdata(ec);
 
 	ecard_set_drvdata(ec, NULL);
 
-	ide_unregister(hwif);
+	ide_host_remove(host);
 
 	ecard_release_resources(ec);
 }
diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index 84644e150531..15f76690a48c 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -191,10 +191,8 @@ static const struct ide_port_info h8300_port_info = {
 
 static int __init h8300_ide_init(void)
 {
-	ide_hwif_t *hwif;
-	int index;
+	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	printk(KERN_INFO DRV_NAME ": H8/300 generic IDE interface\n");
 
@@ -207,15 +205,11 @@ static int __init h8300_ide_init(void)
 
 	hw_setup(&hw);
 
-	hwif = ide_find_port_slot(&h8300_port_info);
-	if (hwif == NULL)
+	host = ide_host_alloc(&h8300_port_info, hws);
+	if (host == NULL)
 		return -ENOENT;
 
-	index = hwif->index;
-
-	idx[0] = index;
-
-	ide_device_add(idx, &h8300_port_info, hws);
+	ide_host_register(host, &h8300_port_info, hws);
 
 	return 0;
 
diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index a5c352abff59..e8818362eb46 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -28,27 +28,24 @@ MODULE_PARM_DESC(probe_mask, "probe mask for legacy ISA IDE ports");
 
 static ssize_t store_add(struct class *cls, const char *buf, size_t n)
 {
-	ide_hwif_t *hwif;
+	struct ide_host *host;
 	unsigned int base, ctl;
 	int irq;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[] = { 0xff, 0xff, 0xff, 0xff };
 
 	if (sscanf(buf, "%x:%x:%d", &base, &ctl, &irq) != 3)
 		return -EINVAL;
 
-	hwif = ide_find_port();
-	if (hwif == NULL)
-		return -ENOENT;
-
 	memset(&hw, 0, sizeof(hw));
 	ide_std_init_ports(&hw, base, ctl);
 	hw.irq = irq;
 	hw.chipset = ide_generic;
 
-	idx[0] = hwif->index;
+	host = ide_host_alloc(NULL, hws);
+	if (host == NULL)
+		return -ENOENT;
 
-	ide_device_add(idx, NULL, hws);
+	ide_host_register(host, NULL, hws);
 
 	return n;
 };
@@ -89,18 +86,16 @@ static int __init ide_generic_sysfs_init(void)
 static int __init ide_generic_init(void)
 {
 	hw_regs_t hw[MAX_HWIFS], *hws[MAX_HWIFS];
-	u8 idx[MAX_HWIFS];
+	struct ide_host *host;
 	int i;
 
 	printk(KERN_INFO DRV_NAME ": please use \"probe_mask=0x3f\" module "
 			 "parameter for probing all legacy ISA IDE ports\n");
 
 	for (i = 0; i < MAX_HWIFS; i++) {
-		ide_hwif_t *hwif;
 		unsigned long io_addr = ide_default_io_base(i);
 
 		hws[i] = NULL;
-		idx[i] = 0xff;
 
 		if ((probe_mask & (1 << i)) && io_addr) {
 			if (!request_region(io_addr, 8, DRV_NAME)) {
@@ -118,23 +113,18 @@ static int __init ide_generic_init(void)
 				continue;
 			}
 
-			hwif = ide_find_port();
-			if (hwif == NULL)
-				continue;
-
-			hwif->chipset = ide_generic;
-
 			memset(&hw[i], 0, sizeof(hw[i]));
 			ide_std_init_ports(&hw[i], io_addr, io_addr + 0x206);
 			hw[i].irq = ide_default_irq(io_addr);
 			hw[i].chipset = ide_generic;
 
 			hws[i] = &hw[i];
-			idx[i] = i;
 		}
 	}
 
-	ide_device_add_all(idx, NULL, hws);
+	host = ide_host_alloc_all(NULL, hws);
+	if (host)
+		ide_host_register(host, NULL, hws);
 
 	if (ide_generic_sysfs_init())
 		printk(KERN_ERR DRV_NAME ": failed to create ide_generic "
diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c
index 89cd5cbe8573..4458ca61897a 100644
--- a/drivers/ide/ide-pnp.c
+++ b/drivers/ide/ide-pnp.c
@@ -29,7 +29,7 @@ static struct pnp_device_id idepnp_devices[] = {
 
 static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id)
 {
-	ide_hwif_t *hwif;
+	struct ide_host *host;
 	unsigned long base, ctl;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
@@ -59,14 +59,11 @@ static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id)
 	hw.irq = pnp_irq(dev, 0);
 	hw.chipset = ide_generic;
 
-	hwif = ide_find_port();
-	if (hwif) {
-		u8 index = hwif->index;
-		u8 idx[4] = { index, 0xff, 0xff, 0xff };
+	host = ide_host_alloc(NULL, hws);
+	if (host) {
+		pnp_set_drvdata(dev, host);
 
-		pnp_set_drvdata(dev, hwif);
-
-		ide_device_add(idx, NULL, hws);
+		ide_host_register(host, NULL, hws);
 
 		return 0;
 	}
@@ -79,9 +76,9 @@ static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id)
 
 static void idepnp_remove(struct pnp_dev *dev)
 {
-	ide_hwif_t *hwif = pnp_get_drvdata(dev);
+	struct ide_host *host = pnp_get_drvdata(dev);
 
-	ide_unregister(hwif);
+	ide_host_remove(host);
 
 	release_region(pnp_port_start(dev, 1), 1);
 	release_region(pnp_port_start(dev, 0), 8);
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index c588066295db..84a89561ec0f 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1477,7 +1477,7 @@ static int ide_sysfs_register_port(ide_hwif_t *hwif)
  *	Return the new hwif.  If we are out of free slots return NULL.
  */
 
-ide_hwif_t *ide_find_port_slot(const struct ide_port_info *d)
+static ide_hwif_t *ide_find_port_slot(const struct ide_port_info *d)
 {
 	ide_hwif_t *hwif;
 	int i;
@@ -1523,14 +1523,63 @@ out_found:
 	ide_init_port_data(hwif, i);
 	return hwif;
 }
-EXPORT_SYMBOL_GPL(ide_find_port_slot);
 
-int ide_device_add_all(u8 *idx, const struct ide_port_info *d, hw_regs_t **hws)
+struct ide_host *ide_host_alloc_all(const struct ide_port_info *d,
+				    hw_regs_t **hws)
+{
+	struct ide_host *host;
+	int i;
+
+	host = kzalloc(sizeof(*host), GFP_KERNEL);
+	if (host == NULL)
+		return NULL;
+
+	for (i = 0; i < MAX_HWIFS; i++) {
+		ide_hwif_t *hwif;
+
+		if (hws[i] == NULL)
+			continue;
+
+		hwif = ide_find_port_slot(d);
+		if (hwif) {
+			hwif->chipset = hws[i]->chipset;
+
+			host->ports[i] = hwif;
+			host->n_ports++;
+		}
+	}
+
+	if (host->n_ports == 0) {
+		kfree(host);
+		return NULL;
+	}
+
+	return host;
+}
+EXPORT_SYMBOL_GPL(ide_host_alloc_all);
+
+struct ide_host *ide_host_alloc(const struct ide_port_info *d, hw_regs_t **hws)
+{
+	hw_regs_t *hws_all[MAX_HWIFS];
+	int i;
+
+	for (i = 0; i < MAX_HWIFS; i++)
+		hws_all[i] = (i < 4) ? hws[i] : NULL;
+
+	return ide_host_alloc_all(d, hws_all);
+}
+EXPORT_SYMBOL_GPL(ide_host_alloc);
+
+int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
+		      hw_regs_t **hws)
 {
 	ide_hwif_t *hwif, *mate = NULL;
+	u8 idx[MAX_HWIFS];
 	int i, rc = 0;
 
 	for (i = 0; i < MAX_HWIFS; i++) {
+		idx[i] = host->ports[i] ? host->ports[i]->index : 0xff;
+
 		if (idx[i] == 0xff) {
 			mate = NULL;
 			continue;
@@ -1626,22 +1675,20 @@ int ide_device_add_all(u8 *idx, const struct ide_port_info *d, hw_regs_t **hws)
 
 	return rc;
 }
-EXPORT_SYMBOL_GPL(ide_device_add_all);
+EXPORT_SYMBOL_GPL(ide_host_register);
 
-int ide_device_add(u8 *idx, const struct ide_port_info *d, hw_regs_t **hws)
+void ide_host_remove(struct ide_host *host)
 {
-	hw_regs_t *hws_all[MAX_HWIFS];
-	u8 idx_all[MAX_HWIFS];
 	int i;
 
 	for (i = 0; i < MAX_HWIFS; i++) {
-		hws_all[i] = (i < 4) ? hws[i] : NULL;
-		idx_all[i] = (i < 4) ? idx[i] : 0xff;
+		if (host->ports[i])
+			ide_unregister(host->ports[i]);
 	}
 
-	return ide_device_add_all(idx_all, d, hws_all);
+	kfree(host);
 }
-EXPORT_SYMBOL_GPL(ide_device_add);
+EXPORT_SYMBOL_GPL(ide_host_remove);
 
 void ide_port_scan(ide_hwif_t *hwif)
 {
@@ -1662,11 +1709,10 @@ void ide_port_scan(ide_hwif_t *hwif)
 }
 EXPORT_SYMBOL_GPL(ide_port_scan);
 
-static void ide_legacy_init_one(u8 *idx, hw_regs_t **hws, hw_regs_t *hw,
+static void ide_legacy_init_one(hw_regs_t **hws, hw_regs_t *hw,
 				u8 port_no, const struct ide_port_info *d,
 				unsigned long config)
 {
-	ide_hwif_t *hwif;
 	unsigned long base, ctl;
 	int irq;
 
@@ -1698,31 +1744,29 @@ static void ide_legacy_init_one(u8 *idx, hw_regs_t **hws, hw_regs_t *hw,
 	hw->chipset = d->chipset;
 	hw->config = config;
 
-	hwif = ide_find_port_slot(d);
-	if (hwif) {
-		hwif->chipset = hw->chipset;
-
-		hws[port_no] = hw;
-		idx[port_no] = hwif->index;
-	}
+	hws[port_no] = hw;
 }
 
 int ide_legacy_device_add(const struct ide_port_info *d, unsigned long config)
 {
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
+	struct ide_host *host;
 	hw_regs_t hw[2], *hws[] = { NULL, NULL, NULL, NULL };
 
 	memset(&hw, 0, sizeof(hw));
 
 	if ((d->host_flags & IDE_HFLAG_QD_2ND_PORT) == 0)
-		ide_legacy_init_one(idx, hws, &hw[0], 0, d, config);
-	ide_legacy_init_one(idx, hws, &hw[1], 1, d, config);
+		ide_legacy_init_one(hws, &hw[0], 0, d, config);
+	ide_legacy_init_one(hws, &hw[1], 1, d, config);
 
-	if (idx[0] == 0xff && idx[1] == 0xff &&
+	if (hws[0] == NULL && hws[1] == NULL &&
 	    (d->host_flags & IDE_HFLAG_SINGLE))
 		return -ENOENT;
 
-	ide_device_add(idx, d, hws);
+	host = ide_host_alloc(d, hws);
+	if (host == NULL)
+		return -ENOMEM;
+
+	ide_host_register(host, d, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 132b504168e9..7e9575d1aee3 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -276,8 +276,6 @@ void ide_unregister(ide_hwif_t *hwif)
 	mutex_unlock(&ide_cfg_mtx);
 }
 
-EXPORT_SYMBOL(ide_unregister);
-
 void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
 {
 	memcpy(&hwif->io_ports, &hw->io_ports, sizeof(hwif->io_ports));
diff --git a/drivers/ide/legacy/buddha.c b/drivers/ide/legacy/buddha.c
index c61bc6a1db36..2625667fab4c 100644
--- a/drivers/ide/legacy/buddha.c
+++ b/drivers/ide/legacy/buddha.c
@@ -150,18 +150,15 @@ static void __init buddha_setup_ports(hw_regs_t *hw, unsigned long base,
 
 static int __init buddha_init(void)
 {
-	ide_hwif_t *hwif;
-	int i;
-
 	struct zorro_dev *z = NULL;
+	struct ide_host *host;
 	u_long buddha_board = 0;
 	BuddhaType type;
-	int buddha_num_hwifs;
+	int buddha_num_hwifs, i;
 
 	while ((z = zorro_find_device(ZORRO_WILDCARD, z))) {
 		unsigned long board;
 		hw_regs_t hw[MAX_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
-		u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 		if (z->id == ZORRO_PROD_INDIVIDUAL_COMPUTERS_BUDDHA) {
 			buddha_num_hwifs = BUDDHA_NUM_HWIFS;
@@ -226,16 +223,12 @@ fail_base2:
 			buddha_setup_ports(&hw[i], base, ctl, irq_port,
 					   ack_intr);
 
-			hwif = ide_find_port();
-			if (hwif) {
-				hwif->chipset = ide_generic;
-
-				hws[i] = &hw[i];
-				idx[i] = hwif->index;
-			}
+			hws[i] = &hw[i];
 		}
 
-		ide_device_add(idx, NULL, hws);
+		host = ide_host_alloc(NULL, hws);
+		if (host)
+			ide_host_register(host, NULL, hws);
 	}
 
 	return 0;
diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c
index 3e2c6125f031..4eb5c3f9fecc 100644
--- a/drivers/ide/legacy/falconide.c
+++ b/drivers/ide/legacy/falconide.c
@@ -112,7 +112,7 @@ static void __init falconide_setup_ports(hw_regs_t *hw)
 
 static int __init falconide_init(void)
 {
-	ide_hwif_t *hwif;
+	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	if (!MACH_IS_ATARI || !ATARIHW_PRESENT(IDE))
@@ -127,13 +127,10 @@ static int __init falconide_init(void)
 
 	falconide_setup_ports(&hw);
 
-	hwif = ide_find_port();
-	if (hwif) {
-		u8 index = hwif->index;
-		u8 idx[4] = { index, 0xff, 0xff, 0xff };
-
+	host = ide_host_alloc(&falconide_port_info, hws);
+	if (host) {
 		ide_get_lock(NULL, NULL);
-		ide_device_add(idx, &falconide_port_info, hws);
+		ide_host_register(host, &falconide_port_info, hws);
 		ide_release_lock();
 	}
 
diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index 7baeefa870fa..13d22bded6b4 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -127,9 +127,9 @@ static int __init gayle_init(void)
     unsigned long phys_base, res_start, res_n;
     unsigned long base, ctrlport, irqport;
     ide_ack_intr_t *ack_intr;
+    struct ide_host *host;
     int a4000, i;
     hw_regs_t hw[GAYLE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
-    u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!MACH_IS_AMIGA)
 	return -ENODEV;
@@ -172,23 +172,17 @@ found:
 		return -EBUSY;
 
     for (i = 0; i < GAYLE_NUM_PROBE_HWIFS; i++) {
-	ide_hwif_t *hwif;
-
 	base = (unsigned long)ZTWO_VADDR(phys_base + i * GAYLE_NEXT_PORT);
 	ctrlport = GAYLE_HAS_CONTROL_REG ? (base + GAYLE_CONTROL) : 0;
 
 	gayle_setup_ports(&hw[i], base, ctrlport, irqport, ack_intr);
 
-	hwif = ide_find_port();
-	if (hwif) {
-	    hwif->chipset = ide_generic;
-
-	    hws[i] = &hw[i];
-	    idx[i] = hwif->index;
-	}
+	hws[i] = &hw[i];
     }
 
-    ide_device_add(idx, NULL, hws);
+    host = ide_host_alloc(NULL, hws);
+    if (host)
+	ide_host_register(host, NULL, hws);
 
     return 0;
 }
diff --git a/drivers/ide/legacy/ide-4drives.c b/drivers/ide/legacy/ide-4drives.c
index 6310dc50e3c5..5935153ef2ad 100644
--- a/drivers/ide/legacy/ide-4drives.c
+++ b/drivers/ide/legacy/ide-4drives.c
@@ -28,10 +28,9 @@ static const struct ide_port_info ide_4drives_port_info = {
 
 static int __init ide_4drives_init(void)
 {
-	ide_hwif_t *hwif, *mate;
+	struct ide_host *host;
 	unsigned long base = 0x1f0, ctl = 0x3f6;
-	hw_regs_t hw, *hws[] = { NULL, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
+	hw_regs_t hw, *hws[] = { &hw, &hw, NULL, NULL };
 
 	if (probe_4drives == 0)
 		return -ENODEV;
@@ -55,21 +54,9 @@ static int __init ide_4drives_init(void)
 	hw.irq = 14;
 	hw.chipset = ide_4drives;
 
-	hwif = ide_find_port();
-	if (hwif) {
-		hwif->chipset = ide_4drives;
-
-		hws[0] = &hw;
-		idx[0] = hwif->index;
-	}
-
-	mate = ide_find_port();
-	if (mate) {
-		hws[1] = &hw;
-		idx[1] = mate->index;
-	}
-
-	ide_device_add(idx, &ide_4drives_port_info, hws);
+	host = ide_host_alloc(&ide_4drives_port_info, hws);
+	if (host)
+		ide_host_register(host, &ide_4drives_port_info, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index f93d5454ebf8..1a4b9e6887fa 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -74,7 +74,7 @@ INT_MODULE_PARM(pc_debug, 0);
 
 typedef struct ide_info_t {
 	struct pcmcia_device	*p_dev;
-	ide_hwif_t		*hwif;
+	struct ide_host		*host;
     int		ndev;
     dev_node_t	node;
 } ide_info_t;
@@ -132,7 +132,7 @@ static int ide_probe(struct pcmcia_device *link)
 static void ide_detach(struct pcmcia_device *link)
 {
     ide_info_t *info = link->priv;
-    ide_hwif_t *hwif = info->hwif;
+    ide_hwif_t *hwif = info->host->ports[0];
     unsigned long data_addr, ctl_addr;
 
     DEBUG(0, "ide_detach(0x%p)\n", link);
@@ -157,13 +157,13 @@ static const struct ide_port_info idecs_port_info = {
 	.host_flags		= IDE_HFLAG_NO_DMA,
 };
 
-static ide_hwif_t *idecs_register(unsigned long io, unsigned long ctl,
+static struct ide_host *idecs_register(unsigned long io, unsigned long ctl,
 				unsigned long irq, struct pcmcia_device *handle)
 {
+    struct ide_host *host;
     ide_hwif_t *hwif;
     int i;
     hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-    u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!request_region(io, 8, DRV_NAME)) {
 	printk(KERN_ERR "%s: I/O resource 0x%lX-0x%lX not free.\n",
@@ -184,26 +184,26 @@ static ide_hwif_t *idecs_register(unsigned long io, unsigned long ctl,
     hw.chipset = ide_pci;
     hw.dev = &handle->dev;
 
-    hwif = ide_find_port();
-    if (hwif == NULL)
+    host = ide_host_alloc(&idecs_port_info, hws);
+    if (host == NULL)
 	goto out_release;
 
-    idx[0] = hwif->index;
+    ide_host_register(host, &idecs_port_info, hws);
 
-    ide_device_add(idx, &idecs_port_info, hws);
+    hwif = host->ports[0];
 
     if (hwif->present)
-	return hwif;
+	return host;
 
     /* retry registration in case device is still spinning up */
     for (i = 0; i < 10; i++) {
 	msleep(100);
 	ide_port_scan(hwif);
 	if (hwif->present)
-	    return hwif;
+	    return host;
     }
 
-    return hwif;
+    return host;
 
 out_release:
     release_region(ctl, 1);
@@ -235,7 +235,7 @@ static int ide_config(struct pcmcia_device *link)
     cistpl_cftable_entry_t *cfg;
     int pass, last_ret = 0, last_fn = 0, is_kme = 0;
     unsigned long io_base, ctl_base;
-    ide_hwif_t *hwif;
+    struct ide_host *host;
 
     DEBUG(0, "ide_config(0x%p)\n", link);
 
@@ -330,21 +330,21 @@ static int ide_config(struct pcmcia_device *link)
     if (is_kme)
 	outb(0x81, ctl_base+1);
 
-     hwif = idecs_register(io_base, ctl_base, link->irq.AssignedIRQ, link);
-     if (hwif == NULL && link->io.NumPorts1 == 0x20) {
+     host = idecs_register(io_base, ctl_base, link->irq.AssignedIRQ, link);
+     if (host == NULL && link->io.NumPorts1 == 0x20) {
 	    outb(0x02, ctl_base + 0x10);
-	    hwif = idecs_register(io_base + 0x10, ctl_base + 0x10,
+	    host = idecs_register(io_base + 0x10, ctl_base + 0x10,
 				  link->irq.AssignedIRQ, link);
     }
 
-    if (hwif == NULL)
+    if (host == NULL)
 	goto failed;
 
     info->ndev = 1;
-    sprintf(info->node.dev_name, "hd%c", 'a' + hwif->index * 2);
-    info->node.major = hwif->major;
+    sprintf(info->node.dev_name, "hd%c", 'a' + host->ports[0]->index * 2);
+    info->node.major = host->ports[0]->major;
     info->node.minor = 0;
-    info->hwif = hwif;
+    info->host = host;
     link->dev_node = &info->node;
     printk(KERN_INFO "ide-cs: %s: Vpp = %d.%d\n",
 	   info->node.dev_name, link->conf.Vpp / 10, link->conf.Vpp % 10);
@@ -375,15 +375,15 @@ failed:
 static void ide_release(struct pcmcia_device *link)
 {
     ide_info_t *info = link->priv;
-    ide_hwif_t *hwif = info->hwif;
+    struct ide_host *host = info->host;
 
     DEBUG(0, "ide_release(0x%p)\n", link);
 
-    if (info->ndev) {
+    if (info->ndev)
 	/* FIXME: if this fails we need to queue the cleanup somehow
 	   -- need to investigate the required PCMCIA magic */
-	ide_unregister(hwif);
-    }
+	ide_host_remove(host);
+
     info->ndev = 0;
 
     pcmcia_disable_device(link);
diff --git a/drivers/ide/legacy/ide_platform.c b/drivers/ide/legacy/ide_platform.c
index 3d71e336a221..58a942c6a131 100644
--- a/drivers/ide/legacy/ide_platform.c
+++ b/drivers/ide/legacy/ide_platform.c
@@ -52,11 +52,10 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 {
 	struct resource *res_base, *res_alt, *res_irq;
 	void __iomem *base, *alt_base;
-	ide_hwif_t *hwif;
 	struct pata_platform_info *pdata;
+	struct ide_host *host;
 	int ret = 0, mmio = 0;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	struct ide_port_info d = platform_ide_port_info;
 
 	pdata = pdev->dev.platform_data;
@@ -93,12 +92,6 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 			res_alt->start, res_alt->end - res_alt->start + 1);
 	}
 
-	hwif = ide_find_port();
-	if (!hwif) {
-		ret = -ENODEV;
-		goto out;
-	}
-
 	memset(&hw, 0, sizeof(hw));
 	plat_ide_setup_ports(&hw, base, alt_base, pdata, res_irq->start);
 	hw.dev = &pdev->dev;
@@ -106,11 +99,15 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 	if (mmio)
 		d.host_flags |= IDE_HFLAG_MMIO;
 
-	idx[0] = hwif->index;
+	host = ide_host_alloc(&d, hws);
+	if (host == NULL) {
+		ret = -ENODEV;
+		goto out;
+	}
 
-	ide_device_add(idx, &d, hws);
+	ide_host_register(host, &d, hws);
 
-	platform_set_drvdata(pdev, hwif);
+	platform_set_drvdata(pdev, host);
 
 	return 0;
 
@@ -120,9 +117,9 @@ out:
 
 static int __devexit plat_ide_remove(struct platform_device *pdev)
 {
-	ide_hwif_t *hwif = pdev->dev.driver_data;
+	struct ide_host *host = pdev->dev.driver_data;
 
-	ide_unregister(hwif);
+	ide_host_remove(host);
 
 	return 0;
 }
diff --git a/drivers/ide/legacy/macide.c b/drivers/ide/legacy/macide.c
index d839df2239fc..b49cf8c2b91a 100644
--- a/drivers/ide/legacy/macide.c
+++ b/drivers/ide/legacy/macide.c
@@ -91,8 +91,8 @@ static const char *mac_ide_name[] =
 
 static int __init macide_init(void)
 {
-	ide_hwif_t *hwif;
 	ide_ack_intr_t *ack_intr;
+	struct ide_host *host;
 	unsigned long base;
 	int irq;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
@@ -125,13 +125,9 @@ static int __init macide_init(void)
 
 	macide_setup_ports(&hw, base, irq, ack_intr);
 
-	hwif = ide_find_port();
-	if (hwif) {
-		u8 index = hwif->index;
-		u8 idx[4] = { index, 0xff, 0xff, 0xff };
-
-		ide_device_add(idx, NULL, hws);
-	}
+	host = ide_host_alloc(NULL, hws);
+	if (host)
+		ide_host_register(host, NULL, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c
index 2dc306f852a6..8fb4438a6afd 100644
--- a/drivers/ide/legacy/q40ide.c
+++ b/drivers/ide/legacy/q40ide.c
@@ -131,10 +131,9 @@ static const char *q40_ide_names[Q40IDE_NUM_HWIFS]={
 
 static int __init q40ide_init(void)
 {
+    struct ide_host *host;
     int i;
-    ide_hwif_t *hwif;
     hw_regs_t hw[Q40IDE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
-    u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
     if (!MACH_IS_Q40)
       return -ENODEV;
@@ -158,16 +157,12 @@ static int __init q40ide_init(void)
 	q40_ide_setup_ports(&hw[i], pcide_bases[i], NULL,
 			q40ide_default_irq(pcide_bases[i]));
 
-	hwif = ide_find_port();
-	if (hwif) {
-		hwif->chipset = ide_generic;
-
-		hws[i] = &hw[i];
-		idx[i] = hwif->index;
-	}
+	hws[i] = &hw[i];
     }
 
-    ide_device_add(idx, &q40ide_port_info, hws);
+    host = ide_host_alloc(&q40ide_port_info, hws);
+    if (host)
+	ide_host_register(host, &q40ide_port_info, hws);
 
     return 0;
 }
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index ed1c9a134079..903c628bddd0 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -563,11 +563,10 @@ static int au_ide_probe(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	_auide_hwif *ahwif = &auide_hwif;
-	ide_hwif_t *hwif;
 	struct resource *res;
+	struct ide_host *host;
 	int ret = 0;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 #if defined(CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA)
 	char *mode = "MWDMA2";
@@ -604,25 +603,23 @@ static int au_ide_probe(struct device *dev)
 		goto out;
 	}
 
-	hwif = ide_find_port();
-	if (hwif == NULL) {
-		ret = -ENOENT;
-		goto out;
-	}
-
 	memset(&hw, 0, sizeof(hw));
 	auide_setup_ports(&hw, ahwif);
 	hw.irq = ahwif->irq;
 	hw.dev = dev;
 	hw.chipset = ide_au1xxx;
 
-	auide_hwif.hwif                 = hwif;
+	host = ide_host_alloc(&au1xxx_port_info, hws);
+	if (host == NULL) {
+		ret = -ENOENT;
+		goto out;
+	}
 
-	idx[0] = hwif->index;
+	ide_host_register(host, &au1xxx_port_info, hws);
 
-	ide_device_add(idx, &au1xxx_port_info, hws);
+	auide_hwif.hwif = host->ports[0];
 
-	dev_set_drvdata(dev, hwif);
+	dev_set_drvdata(dev, host);
 
 	printk(KERN_INFO "Au1xxx IDE(builtin) configured for %s\n", mode );
 
@@ -634,10 +631,10 @@ static int au_ide_remove(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct resource *res;
-	ide_hwif_t *hwif = dev_get_drvdata(dev);
+	struct ide_host *host = dev_get_drvdata(dev);
 	_auide_hwif *ahwif = &auide_hwif;
 
-	ide_unregister(hwif);
+	ide_host_remove(host);
 
 	iounmap((void *)ahwif->regbase);
 
diff --git a/drivers/ide/mips/swarm.c b/drivers/ide/mips/swarm.c
index c1ffb83a2de7..b12d9d224831 100644
--- a/drivers/ide/mips/swarm.c
+++ b/drivers/ide/mips/swarm.c
@@ -72,12 +72,11 @@ static const struct ide_port_info swarm_port_info = {
  */
 static int __devinit swarm_ide_probe(struct device *dev)
 {
-	ide_hwif_t *hwif;
 	u8 __iomem *base;
+	struct ide_host *host;
 	phys_t offset, size;
 	int i;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[] = { 0xff, 0xff, 0xff, 0xff };
 
 	if (!SIBYTE_HAVE_IDE)
 		return -ENODEV;
@@ -116,15 +115,13 @@ static int __devinit swarm_ide_probe(struct device *dev)
 	hw.irq = K_INT_GB_IDE;
 	hw.chipset = ide_generic;
 
-	hwif = ide_find_port_slot(&swarm_port_info);
-	if (hwif == NULL)
+	host = ide_host_alloc(&swarm_port_info, hws);
+	if (host == NULL)
 		goto err;
 
-	idx[0] = hwif->index;
+	ide_host_register(host, &swarm_port_info, hws);
 
-	ide_device_add(idx, &swarm_port_info, hws);
-
-	dev_set_drvdata(dev, hwif);
+	dev_set_drvdata(dev, host);
 
 	return 0;
 err:
diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c
index ccde1e444e13..013697b8cef4 100644
--- a/drivers/ide/pci/cmd640.c
+++ b/drivers/ide/pci/cmd640.c
@@ -180,11 +180,6 @@ static u8 recovery_counts[4] = {16, 16, 16, 16}; /* Recovery count (encoded) */
 
 static DEFINE_SPINLOCK(cmd640_lock);
 
-/*
- * These are initialized to point at the devices we control
- */
-static ide_hwif_t  *cmd_hwif0, *cmd_hwif1;
-
 /*
  * Interface to access cmd640x registers
  */
@@ -714,11 +709,11 @@ static int cmd640x_init_one(unsigned long base, unsigned long ctl)
  */
 static int __init cmd640x_init(void)
 {
+	struct ide_host *host;
 	int second_port_cmd640 = 0, rc;
 	const char *bus_type, *port2;
 	u8 b, cfr;
 	hw_regs_t hw[2], *hws[] = { NULL, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	if (cmd640_vlb && probe_for_cmd640_vlb()) {
 		bus_type = "VLB";
@@ -781,17 +776,10 @@ static int __init cmd640x_init(void)
 	printk(KERN_INFO "cmd640: buggy cmd640%c interface on %s, config=0x%02x"
 			 "\n", 'a' + cmd640_chip_version - 1, bus_type, cfr);
 
-	cmd_hwif0 = ide_find_port();
-
 	/*
 	 * Initialize data for primary port
 	 */
-	if (cmd_hwif0) {
-		cmd_hwif0->chipset = ide_cmd640;
-
-		hws[0] = &hw[0];
-		idx[0] = cmd_hwif0->index;
-	}
+	hws[0] = &hw[0];
 
 	/*
 	 * Ensure compatibility by always using the slowest timings
@@ -831,13 +819,9 @@ static int __init cmd640x_init(void)
 	/*
 	 * Initialize data for secondary cmd640 port, if enabled
 	 */
-	if (second_port_cmd640) {
-		cmd_hwif1 = ide_find_port();
-		if (cmd_hwif1) {
-			hws[1] = &hw[1];
-			idx[1] = cmd_hwif1->index;
-		}
-	}
+	if (second_port_cmd640)
+		hws[1] = &hw[1];
+
 	printk(KERN_INFO "cmd640: %sserialized, secondary interface %s\n",
 			 second_port_cmd640 ? "" : "not ", port2);
 
@@ -845,7 +829,9 @@ static int __init cmd640x_init(void)
 	cmd640_dump_regs();
 #endif
 
-	ide_device_add(idx, &cmd640_port_info, hws);
+	host = ide_host_alloc(&cmd640_port_info, hws);
+	if (host)
+		ide_host_register(host, &cmd640_port_info, hws);
 
 	return 1;
 }
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index e8e7df1915bf..b8ec06d22c61 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -114,9 +114,9 @@ static const struct ide_port_info cyrix_chipsets[] __devinitdata = {
  
 static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
+	struct ide_host *host;
 	const struct ide_port_info *d = &cyrix_chipsets[id->driver_data];
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	ide_setup_pci_noise(dev, d);
 
@@ -138,9 +138,11 @@ static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_devic
 	 *	do all the device setup for us
 	 */
 
-	ide_pci_setup_ports(dev, d, 14, &idx[0], &hw[0], &hws[0]);
+	ide_pci_setup_ports(dev, d, 14, &hw[0], &hws[0]);
 
-	ide_device_add(idx, d, hws);
+	host = ide_host_alloc(d, hws);
+	if (host)
+		ide_host_register(host, d, hws);
 
 	return 0;
 }
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index 33fe15db408a..5eb9d9325184 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -56,11 +56,10 @@ static const struct ide_port_info delkin_cb_port_info = {
 static int __devinit
 delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 {
+	struct ide_host *host;
 	unsigned long base;
-	ide_hwif_t *hwif = NULL;
 	int i, rc;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	rc = pci_enable_device(dev);
 	if (rc) {
@@ -87,17 +86,13 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 	hw.dev = &dev->dev;
 	hw.chipset = ide_pci;		/* this enables IRQ sharing */
 
-	hwif = ide_find_port();
-	if (hwif == NULL)
+	host = ide_host_alloc(&delkin_cb_port_info, hws);
+	if (host == NULL)
 		goto out_disable;
 
-	i = hwif->index;
+	ide_host_register(host, &delkin_cb_port_info, hws);
 
-	idx[0] = i;
-
-	ide_device_add(idx, &delkin_cb_port_info, hws);
-
-	pci_set_drvdata(dev, hwif);
+	pci_set_drvdata(dev, host);
 
 	return 0;
 
@@ -110,9 +105,9 @@ out_disable:
 static void
 delkin_cb_remove (struct pci_dev *dev)
 {
-	ide_hwif_t *hwif = pci_get_drvdata(dev);
+	struct ide_host *host = pci_get_drvdata(dev);
 
-	ide_unregister(hwif);
+	ide_host_remove(host);
 
 	pci_release_regions(dev);
 	pci_disable_device(dev);
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 5b1a0e950dfd..d5e2ba6bacd6 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -65,7 +65,7 @@
 
 static struct scc_ports {
 	unsigned long ctl, dma;
-	ide_hwif_t *hwif;  /* for removing port from system */
+	struct ide_host *host;	/* for removing port from system */
 } scc_ports[MAX_HWIFS];
 
 /* PIO transfer mode  table */
@@ -586,15 +586,10 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev,
 				    const struct ide_port_info *d)
 {
 	struct scc_ports *ports = pci_get_drvdata(dev);
-	ide_hwif_t *hwif = NULL;
+	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	int i;
 
-	hwif = ide_find_port_slot(d);
-	if (hwif == NULL)
-		return -ENOMEM;
-
 	memset(&hw, 0, sizeof(hw));
 	for (i = 0; i <= 8; i++)
 		hw.io_ports_array[i] = ports->dma + 0x20 + i * 4;
@@ -602,9 +597,13 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev,
 	hw.dev = &dev->dev;
 	hw.chipset = ide_pci;
 
-	idx[0] = hwif->index;
+	host = ide_host_alloc(d, hws);
+	if (host == NULL)
+		return -ENOMEM;
 
-	ide_device_add(idx, d, hws);
+	ide_host_register(host, d, hws);
+
+	ports->host = host;
 
 	return 0;
 }
@@ -848,8 +847,6 @@ static void __devinit init_hwif_scc(ide_hwif_t *hwif)
 {
 	struct scc_ports *ports = ide_get_hwifdata(hwif);
 
-	ports->hwif = hwif;
-
 	/* PTERADD */
 	out_be32((void __iomem *)(hwif->dma_base + 0x018), hwif->dmatable_dma);
 
@@ -932,7 +929,8 @@ static int __devinit scc_init_one(struct pci_dev *dev, const struct pci_device_i
 static void __devexit scc_remove(struct pci_dev *dev)
 {
 	struct scc_ports *ports = pci_get_drvdata(dev);
-	ide_hwif_t *hwif = ports->hwif;
+	struct ide_host *host = ports->host;
+	ide_hwif_t *hwif = host->ports[0];
 
 	if (hwif->dmatable_cpu) {
 		pci_free_consistent(dev, PRD_ENTRIES * PRD_BYTES,
@@ -940,7 +938,7 @@ static void __devexit scc_remove(struct pci_dev *dev)
 		hwif->dmatable_cpu = NULL;
 	}
 
-	ide_unregister(hwif);
+	ide_host_remove(host);
 
 	iounmap((void*)ports->dma);
 	iounmap((void*)ports->ctl);
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 5598bd5936d9..440f43a86ad3 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -600,9 +600,8 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	unsigned long cmd_base, irqport;
 	unsigned long bar0, cmd_phys_base, ctl;
 	void __iomem *virt_base;
-	ide_hwif_t *hwif;
+	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	struct ide_port_info d = sgiioc4_port_info;
 
 	/*  Get the CmdBlk and CtrlBlk Base Registers */
@@ -635,16 +634,14 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	hw.chipset = ide_pci;
 	hw.dev = &dev->dev;
 
-	hwif = ide_find_port_slot(&d);
-	if (hwif == NULL)
-		goto err;
-
 	/* Initializing chipset IRQ Registers */
 	writel(0x03, (void __iomem *)(irqport + IOC4_INTR_SET * 4));
 
-	idx[0] = hwif->index;
+	host = ide_host_alloc(&d, hws);
+	if (host == NULL)
+		goto err;
 
-	if (ide_device_add(idx, &d, hws))
+	if (ide_host_register(host, &d, hws))
 		return -EIO;
 
 	return 0;
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index ee557d10a764..ecd2f28da1ba 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1039,9 +1039,9 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 {
 	struct device_node *np = pmif->node;
 	const int *bidp;
+	struct ide_host *host;
 	ide_hwif_t *hwif;
 	hw_regs_t *hws[] = { hw, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 	struct ide_port_info d = pmac_port_info;
 
 	pmif->broken_dma = pmif->broken_dma_warn = 0;
@@ -1118,13 +1118,13 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 			 pmif->mdev ? "macio" : "PCI", pmif->aapl_bus_id,
 			 pmif->mediabay ? " (mediabay)" : "", hw->irq);
 
-	hwif = ide_find_port_slot(&d);
-	if (hwif == NULL)
+	host = ide_host_alloc(&d, hws);
+	if (host == NULL)
 		return -ENOENT;
 
-	idx[0] = hwif->index;
+	ide_host_register(host, &d, hws);
 
-	ide_device_add(idx, &d, hws);
+	hwif = host->ports[0];
 
 	return 0;
 }
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 804c3ef245f9..1c0c5570dec8 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -289,7 +289,7 @@ static int ide_pci_check_iomem(struct pci_dev *dev, const struct ide_port_info *
 }
 
 /**
- *	ide_hwif_configure	-	configure an IDE interface
+ *	ide_hw_configure	-	configure a hw_regs_t instance
  *	@dev: PCI device holding interface
  *	@d: IDE port info
  *	@port: port number
@@ -300,23 +300,20 @@ static int ide_pci_check_iomem(struct pci_dev *dev, const struct ide_port_info *
  *	is done per interface port rather than per PCI device. There may be
  *	more than one port per device.
  *
- *	Returns the new hardware interface structure, or NULL on a failure
+ *	Returns zero on success or an error code.
  */
 
-static ide_hwif_t *ide_hwif_configure(struct pci_dev *dev,
-				      const struct ide_port_info *d,
-				      unsigned int port, int irq,
-				      hw_regs_t *hw)
+static int ide_hw_configure(struct pci_dev *dev, const struct ide_port_info *d,
+			    unsigned int port, int irq, hw_regs_t *hw)
 {
 	unsigned long ctl = 0, base = 0;
-	ide_hwif_t *hwif;
 
 	if ((d->host_flags & IDE_HFLAG_ISA_PORTS) == 0) {
 		if (ide_pci_check_iomem(dev, d, 2 * port) ||
 		    ide_pci_check_iomem(dev, d, 2 * port + 1)) {
 			printk(KERN_ERR "%s: I/O baseregs (BIOS) are reported "
 					"as MEM for port %d!\n", d->name, port);
-			return NULL;
+			return -EINVAL;
 		}
 
 		ctl  = pci_resource_start(dev, 2*port+1);
@@ -330,7 +327,7 @@ static ide_hwif_t *ide_hwif_configure(struct pci_dev *dev,
 	if (!base || !ctl) {
 		printk(KERN_ERR "%s: bad PCI BARs for port %d, skipping\n",
 				d->name, port);
-		return NULL;
+		return -EINVAL;
 	}
 
 	memset(hw, 0, sizeof(*hw));
@@ -339,13 +336,7 @@ static ide_hwif_t *ide_hwif_configure(struct pci_dev *dev,
 	hw->chipset = d->chipset ? d->chipset : ide_pci;
 	ide_std_init_ports(hw, base, ctl | 2);
 
-	hwif = ide_find_port_slot(d);
-	if (hwif == NULL)
-		return NULL;
-
-	hwif->chipset = hw->chipset;
-
-	return hwif;
+	return 0;
 }
 
 #ifdef CONFIG_BLK_DEV_IDEDMA_PCI
@@ -443,7 +434,6 @@ out:
  *	@dev: PCI device
  *	@d: IDE port info
  *	@pciirq: IRQ line
- *	@idx: ATA index table to update
  *	@hw: hw_regs_t instances corresponding to this PCI IDE device
  *	@hws: hw_regs_t pointers table to update
  *
@@ -457,10 +447,9 @@ out:
  */
 
 void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d,
-			 int pciirq, u8 *idx, hw_regs_t *hw, hw_regs_t **hws)
+			 int pciirq, hw_regs_t *hw, hw_regs_t **hws)
 {
 	int channels = (d->host_flags & IDE_HFLAG_SINGLE) ? 1 : 2, port;
-	ide_hwif_t *hwif;
 	u8 tmp;
 
 	/*
@@ -476,12 +465,10 @@ void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d,
 			continue;	/* port not enabled */
 		}
 
-		hwif = ide_hwif_configure(dev, d, port, pciirq, hw + port);
-		if (hwif == NULL)
+		if (ide_hw_configure(dev, d, port, pciirq, hw + port))
 			continue;
 
 		*(hws + port) = hw + port;
-		*(idx + port) = hwif->index;
 	}
 }
 EXPORT_SYMBOL_GPL(ide_pci_setup_ports);
@@ -554,7 +541,7 @@ out:
 
 int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 {
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
+	struct ide_host *host;
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 	int ret;
 
@@ -562,9 +549,11 @@ int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 
 	if (ret >= 0) {
 		/* FIXME: silent failure can happen */
-		ide_pci_setup_ports(dev, d, ret, &idx[0], &hw[0], &hws[0]);
+		ide_pci_setup_ports(dev, d, ret, &hw[0], &hws[0]);
 
-		ide_device_add(idx, d, hws);
+		host = ide_host_alloc(d, hws);
+		if (host)
+			ide_host_register(host, d, hws);
 	}
 
 	return ret;
@@ -575,9 +564,9 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 			  const struct ide_port_info *d)
 {
 	struct pci_dev *pdev[] = { dev1, dev2 };
+	struct ide_host *host;
 	int ret, i;
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
-	u8 idx[4] = { 0xff, 0xff, 0xff, 0xff };
 
 	for (i = 0; i < 2; i++) {
 		ret = do_ide_setup_pci_device(pdev[i], d, !i);
@@ -590,11 +579,12 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 			goto out;
 
 		/* FIXME: silent failure can happen */
-		ide_pci_setup_ports(pdev[i], d, ret, &idx[i*2], &hw[i*2],
-				    &hws[i*2]);
+		ide_pci_setup_ports(pdev[i], d, ret, &hw[i*2], &hws[i*2]);
 	}
 
-	ide_device_add(idx, d, hws);
+	host = ide_host_alloc(d, hws);
+	if (host)
+		ide_host_register(host, d, hws);
 out:
 	return ret;
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 1286a2275efb..a41ae57fafc5 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -558,6 +558,11 @@ typedef struct hwif_s {
 #endif
 } ____cacheline_internodealigned_in_smp ide_hwif_t;
 
+struct ide_host {
+	ide_hwif_t	*ports[MAX_HWIFS];
+	unsigned int	n_ports;
+};
+
 /*
  *  internal ide interrupt handler type
  */
@@ -813,13 +818,6 @@ int generic_ide_ioctl(ide_drive_t *, struct file *, struct block_device *, unsig
 extern int ide_vlb_clk;
 extern int ide_pci_clk;
 
-ide_hwif_t *ide_find_port_slot(const struct ide_port_info *);
-
-static inline ide_hwif_t *ide_find_port(void)
-{
-	return ide_find_port_slot(NULL);
-}
-
 extern int ide_end_request (ide_drive_t *drive, int uptodate, int nrsecs);
 int ide_end_dequeued_request(ide_drive_t *drive, struct request *rq,
 			     int uptodate, int nr_sectors);
@@ -1024,7 +1022,7 @@ extern int __ide_pci_register_driver(struct pci_driver *driver, struct module *o
 #endif
 
 void ide_pci_setup_ports(struct pci_dev *, const struct ide_port_info *, int,
-			 u8 *, hw_regs_t *, hw_regs_t **);
+			 hw_regs_t *, hw_regs_t **);
 void ide_setup_pci_noise(struct pci_dev *, const struct ide_port_info *);
 
 #ifdef CONFIG_BLK_DEV_IDEDMA_PCI
@@ -1236,8 +1234,11 @@ void ide_undecoded_slave(ide_drive_t *);
 
 void ide_port_apply_params(ide_hwif_t *);
 
-int ide_device_add_all(u8 *, const struct ide_port_info *, hw_regs_t **);
-int ide_device_add(u8 *, const struct ide_port_info *, hw_regs_t **);
+struct ide_host *ide_host_alloc_all(const struct ide_port_info *, hw_regs_t **);
+struct ide_host *ide_host_alloc(const struct ide_port_info *, hw_regs_t **);
+int ide_host_register(struct ide_host *, const struct ide_port_info *,
+		      hw_regs_t **);
+void ide_host_remove(struct ide_host *);
 int ide_legacy_device_add(const struct ide_port_info *, unsigned long);
 void ide_port_unregister_devices(ide_hwif_t *);
 void ide_port_scan(ide_hwif_t *);
-- 
cgit v1.2.3


From 6f904d015262dfa43eb1cecc00b0998b4c3543f2 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:57 +0200
Subject: ide: add ide_host_add() helper

Add ide_host_add() helper which does ide_host_alloc()+ide_host_register(),
then convert ide_setup_pci_device[s](), ide_legacy_device_add() and some
host drivers to use it.

While at it:

* Fix ide_setup_pci_device[s](), ide_arm.c, gayle.c, ide-4drives.c,
  macide.c, q40ide.c, cmd640.c and cs5520.c to return correct error value.

* -ENOENT -> -ENOMEM in rapide.c, ide-h8300.c, ide-generic.c, au1xxx-ide.c
  and pmac.c

* -ENODEV -> -ENOMEM in palm_bk3710.c, ide_platform.c and delkin_cb.c

* -1 -> -ENOMEM in ide-pnp.c

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/ide_arm.c         |  7 +------
 drivers/ide/arm/palm_bk3710.c     | 10 ++++------
 drivers/ide/arm/rapide.c          |  8 ++------
 drivers/ide/h8300/ide-h8300.c     |  9 +--------
 drivers/ide/ide-generic.c         | 11 ++++-------
 drivers/ide/ide-pnp.c             | 16 ++++++++--------
 drivers/ide/ide-probe.c           | 27 +++++++++++++++++++--------
 drivers/ide/legacy/buddha.c       |  5 +----
 drivers/ide/legacy/gayle.c        |  7 +------
 drivers/ide/legacy/ide-4drives.c  |  7 +------
 drivers/ide/legacy/ide-cs.c       |  8 +++-----
 drivers/ide/legacy/ide_platform.c |  8 ++------
 drivers/ide/legacy/macide.c       |  7 +------
 drivers/ide/legacy/q40ide.c       |  7 +------
 drivers/ide/mips/au1xxx-ide.c     |  8 ++------
 drivers/ide/mips/swarm.c          | 10 ++++------
 drivers/ide/pci/cmd640.c          |  7 +------
 drivers/ide/pci/cs5520.c          |  7 +------
 drivers/ide/pci/delkin_cb.c       |  8 +++-----
 drivers/ide/pci/scc_pata.c        | 10 ++++------
 drivers/ide/ppc/pmac.c            |  9 ++++-----
 drivers/ide/setup-pci.c           | 10 ++--------
 include/linux/ide.h               |  2 ++
 23 files changed, 72 insertions(+), 136 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/ide_arm.c b/drivers/ide/arm/ide_arm.c
index 9efd7a86db45..176532ffae0e 100644
--- a/drivers/ide/arm/ide_arm.c
+++ b/drivers/ide/arm/ide_arm.c
@@ -28,7 +28,6 @@
 
 static int __init ide_arm_init(void)
 {
-	struct ide_host *host;
 	unsigned long base = IDE_ARM_IO, ctl = IDE_ARM_IO + 0x206;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
@@ -50,11 +49,7 @@ static int __init ide_arm_init(void)
 	hw.irq = IDE_ARM_IRQ;
 	hw.chipset = ide_generic;
 
-	host = ide_host_alloc(NULL, hws);
-	if (host)
-		ide_host_register(host, NULL, hws);
-
-	return 0;
+	return ide_host_add(NULL, hws, NULL);
 }
 
 module_init(ide_arm_init);
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 24389a571c37..65bb4b8fd570 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -349,7 +349,7 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	struct resource *mem, *irq;
 	struct ide_host *host;
 	unsigned long base, rate;
-	int i;
+	int i, rc;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	clk = clk_get(NULL, "IDECLK");
@@ -392,16 +392,14 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	hw.irq = irq->start;
 	hw.chipset = ide_palm3710;
 
-	host = ide_host_alloc(&palm_bk3710_port_info, hws);
-	if (host == NULL)
+	rc = ide_host_add(&palm_bk3710_port_info, hws, NULL);
+	if (rc)
 		goto out;
 
-	ide_host_register(host, &palm_bk3710_port_info, hws);
-
 	return 0;
 out:
 	printk(KERN_WARNING "Palm Chip BK3710 IDE Register Fail\n");
-	return -ENODEV;
+	return rc;
 }
 
 /* work with hotplug and coldplug */
diff --git a/drivers/ide/arm/rapide.c b/drivers/ide/arm/rapide.c
index 11f3307385de..2bdd8b734afb 100644
--- a/drivers/ide/arm/rapide.c
+++ b/drivers/ide/arm/rapide.c
@@ -52,13 +52,9 @@ rapide_probe(struct expansion_card *ec, const struct ecard_id *id)
 	hw.chipset = ide_generic;
 	hw.dev = &ec->dev;
 
-	host = ide_host_alloc(&rapide_port_info, hws);
-	if (host == NULL) {
-		ret = -ENOENT;
+	ret = ide_host_add(&rapide_port_info, hws, &host);
+	if (ret)
 		goto release;
-	}
-
-	ide_host_register(host, &rapide_port_info, hws);
 
 	ecard_set_drvdata(ec, host);
 	goto out;
diff --git a/drivers/ide/h8300/ide-h8300.c b/drivers/ide/h8300/ide-h8300.c
index 15f76690a48c..bde7a585f198 100644
--- a/drivers/ide/h8300/ide-h8300.c
+++ b/drivers/ide/h8300/ide-h8300.c
@@ -191,7 +191,6 @@ static const struct ide_port_info h8300_port_info = {
 
 static int __init h8300_ide_init(void)
 {
-	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	printk(KERN_INFO DRV_NAME ": H8/300 generic IDE interface\n");
@@ -205,13 +204,7 @@ static int __init h8300_ide_init(void)
 
 	hw_setup(&hw);
 
-	host = ide_host_alloc(&h8300_port_info, hws);
-	if (host == NULL)
-		return -ENOENT;
-
-	ide_host_register(host, &h8300_port_info, hws);
-
-	return 0;
+	return ide_host_add(&h8300_port_info, hws, NULL);
 
 out_busy:
 	printk(KERN_ERR "ide-h8300: IDE I/F resource already used.\n");
diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index e8818362eb46..a7082c28d06f 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -28,9 +28,8 @@ MODULE_PARM_DESC(probe_mask, "probe mask for legacy ISA IDE ports");
 
 static ssize_t store_add(struct class *cls, const char *buf, size_t n)
 {
-	struct ide_host *host;
 	unsigned int base, ctl;
-	int irq;
+	int irq, rc;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	if (sscanf(buf, "%x:%x:%d", &base, &ctl, &irq) != 3)
@@ -41,11 +40,9 @@ static ssize_t store_add(struct class *cls, const char *buf, size_t n)
 	hw.irq = irq;
 	hw.chipset = ide_generic;
 
-	host = ide_host_alloc(NULL, hws);
-	if (host == NULL)
-		return -ENOENT;
-
-	ide_host_register(host, NULL, hws);
+	rc = ide_host_add(NULL, hws, NULL);
+	if (rc)
+		return rc;
 
 	return n;
 };
diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c
index 4458ca61897a..bac9b392b689 100644
--- a/drivers/ide/ide-pnp.c
+++ b/drivers/ide/ide-pnp.c
@@ -31,6 +31,7 @@ static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id)
 {
 	struct ide_host *host;
 	unsigned long base, ctl;
+	int rc;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	printk(KERN_INFO DRV_NAME ": generic PnP IDE interface\n");
@@ -59,19 +60,18 @@ static int idepnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id)
 	hw.irq = pnp_irq(dev, 0);
 	hw.chipset = ide_generic;
 
-	host = ide_host_alloc(NULL, hws);
-	if (host) {
-		pnp_set_drvdata(dev, host);
+	rc = ide_host_add(NULL, hws, &host);
+	if (rc)
+		goto out;
 
-		ide_host_register(host, NULL, hws);
-
-		return 0;
-	}
+	pnp_set_drvdata(dev, host);
 
+	return 0;
+out:
 	release_region(ctl, 1);
 	release_region(base, 8);
 
-	return -1;
+	return rc;
 }
 
 static void idepnp_remove(struct pnp_dev *dev)
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 84a89561ec0f..17a104b95d54 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1677,6 +1677,24 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
 }
 EXPORT_SYMBOL_GPL(ide_host_register);
 
+int ide_host_add(const struct ide_port_info *d, hw_regs_t **hws,
+		 struct ide_host **hostp)
+{
+	struct ide_host *host;
+
+	host = ide_host_alloc(d, hws);
+	if (host == NULL)
+		return -ENOMEM;
+
+	ide_host_register(host, d, hws);
+
+	if (hostp)
+		*hostp = host;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ide_host_add);
+
 void ide_host_remove(struct ide_host *host)
 {
 	int i;
@@ -1749,7 +1767,6 @@ static void ide_legacy_init_one(hw_regs_t **hws, hw_regs_t *hw,
 
 int ide_legacy_device_add(const struct ide_port_info *d, unsigned long config)
 {
-	struct ide_host *host;
 	hw_regs_t hw[2], *hws[] = { NULL, NULL, NULL, NULL };
 
 	memset(&hw, 0, sizeof(hw));
@@ -1762,12 +1779,6 @@ int ide_legacy_device_add(const struct ide_port_info *d, unsigned long config)
 	    (d->host_flags & IDE_HFLAG_SINGLE))
 		return -ENOENT;
 
-	host = ide_host_alloc(d, hws);
-	if (host == NULL)
-		return -ENOMEM;
-
-	ide_host_register(host, d, hws);
-
-	return 0;
+	return ide_host_add(d, hws, NULL);
 }
 EXPORT_SYMBOL_GPL(ide_legacy_device_add);
diff --git a/drivers/ide/legacy/buddha.c b/drivers/ide/legacy/buddha.c
index 2625667fab4c..7c2afa97f417 100644
--- a/drivers/ide/legacy/buddha.c
+++ b/drivers/ide/legacy/buddha.c
@@ -151,7 +151,6 @@ static void __init buddha_setup_ports(hw_regs_t *hw, unsigned long base,
 static int __init buddha_init(void)
 {
 	struct zorro_dev *z = NULL;
-	struct ide_host *host;
 	u_long buddha_board = 0;
 	BuddhaType type;
 	int buddha_num_hwifs, i;
@@ -226,9 +225,7 @@ fail_base2:
 			hws[i] = &hw[i];
 		}
 
-		host = ide_host_alloc(NULL, hws);
-		if (host)
-			ide_host_register(host, NULL, hws);
+		ide_host_add(NULL, hws, NULL);
 	}
 
 	return 0;
diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index 13d22bded6b4..dd5c467d8dd0 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -127,7 +127,6 @@ static int __init gayle_init(void)
     unsigned long phys_base, res_start, res_n;
     unsigned long base, ctrlport, irqport;
     ide_ack_intr_t *ack_intr;
-    struct ide_host *host;
     int a4000, i;
     hw_regs_t hw[GAYLE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
 
@@ -180,11 +179,7 @@ found:
 	hws[i] = &hw[i];
     }
 
-    host = ide_host_alloc(NULL, hws);
-    if (host)
-	ide_host_register(host, NULL, hws);
-
-    return 0;
+    return ide_host_add(NULL, hws, NULL);
 }
 
 module_init(gayle_init);
diff --git a/drivers/ide/legacy/ide-4drives.c b/drivers/ide/legacy/ide-4drives.c
index 5935153ef2ad..c76d55de6996 100644
--- a/drivers/ide/legacy/ide-4drives.c
+++ b/drivers/ide/legacy/ide-4drives.c
@@ -28,7 +28,6 @@ static const struct ide_port_info ide_4drives_port_info = {
 
 static int __init ide_4drives_init(void)
 {
-	struct ide_host *host;
 	unsigned long base = 0x1f0, ctl = 0x3f6;
 	hw_regs_t hw, *hws[] = { &hw, &hw, NULL, NULL };
 
@@ -54,11 +53,7 @@ static int __init ide_4drives_init(void)
 	hw.irq = 14;
 	hw.chipset = ide_4drives;
 
-	host = ide_host_alloc(&ide_4drives_port_info, hws);
-	if (host)
-		ide_host_register(host, &ide_4drives_port_info, hws);
-
-	return 0;
+	return ide_host_add(&ide_4drives_port_info, hws, NULL);
 }
 
 module_init(ide_4drives_init);
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index 1a4b9e6887fa..21bfac137844 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -162,7 +162,7 @@ static struct ide_host *idecs_register(unsigned long io, unsigned long ctl,
 {
     struct ide_host *host;
     ide_hwif_t *hwif;
-    int i;
+    int i, rc;
     hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
     if (!request_region(io, 8, DRV_NAME)) {
@@ -184,12 +184,10 @@ static struct ide_host *idecs_register(unsigned long io, unsigned long ctl,
     hw.chipset = ide_pci;
     hw.dev = &handle->dev;
 
-    host = ide_host_alloc(&idecs_port_info, hws);
-    if (host == NULL)
+    rc = ide_host_add(&idecs_port_info, hws, &host);
+    if (rc)
 	goto out_release;
 
-    ide_host_register(host, &idecs_port_info, hws);
-
     hwif = host->ports[0];
 
     if (hwif->present)
diff --git a/drivers/ide/legacy/ide_platform.c b/drivers/ide/legacy/ide_platform.c
index 58a942c6a131..051b4ab0f359 100644
--- a/drivers/ide/legacy/ide_platform.c
+++ b/drivers/ide/legacy/ide_platform.c
@@ -99,13 +99,9 @@ static int __devinit plat_ide_probe(struct platform_device *pdev)
 	if (mmio)
 		d.host_flags |= IDE_HFLAG_MMIO;
 
-	host = ide_host_alloc(&d, hws);
-	if (host == NULL) {
-		ret = -ENODEV;
+	ret = ide_host_add(&d, hws, &host);
+	if (ret)
 		goto out;
-	}
-
-	ide_host_register(host, &d, hws);
 
 	platform_set_drvdata(pdev, host);
 
diff --git a/drivers/ide/legacy/macide.c b/drivers/ide/legacy/macide.c
index b49cf8c2b91a..a0bb167980e7 100644
--- a/drivers/ide/legacy/macide.c
+++ b/drivers/ide/legacy/macide.c
@@ -92,7 +92,6 @@ static const char *mac_ide_name[] =
 static int __init macide_init(void)
 {
 	ide_ack_intr_t *ack_intr;
-	struct ide_host *host;
 	unsigned long base;
 	int irq;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
@@ -125,11 +124,7 @@ static int __init macide_init(void)
 
 	macide_setup_ports(&hw, base, irq, ack_intr);
 
-	host = ide_host_alloc(NULL, hws);
-	if (host)
-		ide_host_register(host, NULL, hws);
-
-	return 0;
+	return ide_host_add(NULL, hws, NULL);
 }
 
 module_init(macide_init);
diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c
index 8fb4438a6afd..4abd8fc78197 100644
--- a/drivers/ide/legacy/q40ide.c
+++ b/drivers/ide/legacy/q40ide.c
@@ -131,7 +131,6 @@ static const char *q40_ide_names[Q40IDE_NUM_HWIFS]={
 
 static int __init q40ide_init(void)
 {
-    struct ide_host *host;
     int i;
     hw_regs_t hw[Q40IDE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
 
@@ -160,11 +159,7 @@ static int __init q40ide_init(void)
 	hws[i] = &hw[i];
     }
 
-    host = ide_host_alloc(&q40ide_port_info, hws);
-    if (host)
-	ide_host_register(host, &q40ide_port_info, hws);
-
-    return 0;
+    return ide_host_add(&q40ide_port_info, hws, NULL);
 }
 
 module_init(q40ide_init);
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 903c628bddd0..11b7f61aae40 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -609,13 +609,9 @@ static int au_ide_probe(struct device *dev)
 	hw.dev = dev;
 	hw.chipset = ide_au1xxx;
 
-	host = ide_host_alloc(&au1xxx_port_info, hws);
-	if (host == NULL) {
-		ret = -ENOENT;
+	ret = ide_host_add(&au1xxx_port_info, hws, &host);
+	if (ret)
 		goto out;
-	}
-
-	ide_host_register(host, &au1xxx_port_info, hws);
 
 	auide_hwif.hwif = host->ports[0];
 
diff --git a/drivers/ide/mips/swarm.c b/drivers/ide/mips/swarm.c
index b12d9d224831..badf79fc9e3a 100644
--- a/drivers/ide/mips/swarm.c
+++ b/drivers/ide/mips/swarm.c
@@ -75,7 +75,7 @@ static int __devinit swarm_ide_probe(struct device *dev)
 	u8 __iomem *base;
 	struct ide_host *host;
 	phys_t offset, size;
-	int i;
+	int i, rc;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
 	if (!SIBYTE_HAVE_IDE)
@@ -115,19 +115,17 @@ static int __devinit swarm_ide_probe(struct device *dev)
 	hw.irq = K_INT_GB_IDE;
 	hw.chipset = ide_generic;
 
-	host = ide_host_alloc(&swarm_port_info, hws);
-	if (host == NULL)
+	rc = ide_host_add(&swarm_port_info, hws, &host);
+	if (rc)
 		goto err;
 
-	ide_host_register(host, &swarm_port_info, hws);
-
 	dev_set_drvdata(dev, host);
 
 	return 0;
 err:
 	release_resource(&swarm_ide_resource);
 	iounmap(base);
-	return -ENOMEM;
+	return rc;
 }
 
 static struct device_driver swarm_ide_driver = {
diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c
index 013697b8cef4..e6c62006ca1a 100644
--- a/drivers/ide/pci/cmd640.c
+++ b/drivers/ide/pci/cmd640.c
@@ -709,7 +709,6 @@ static int cmd640x_init_one(unsigned long base, unsigned long ctl)
  */
 static int __init cmd640x_init(void)
 {
-	struct ide_host *host;
 	int second_port_cmd640 = 0, rc;
 	const char *bus_type, *port2;
 	u8 b, cfr;
@@ -829,11 +828,7 @@ static int __init cmd640x_init(void)
 	cmd640_dump_regs();
 #endif
 
-	host = ide_host_alloc(&cmd640_port_info, hws);
-	if (host)
-		ide_host_register(host, &cmd640_port_info, hws);
-
-	return 1;
+	return ide_host_add(&cmd640_port_info, hws, NULL);
 }
 
 module_param_named(probe_vlb, cmd640_vlb, bool, 0);
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index b8ec06d22c61..b03d8ae947e6 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -114,7 +114,6 @@ static const struct ide_port_info cyrix_chipsets[] __devinitdata = {
  
 static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	struct ide_host *host;
 	const struct ide_port_info *d = &cyrix_chipsets[id->driver_data];
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 
@@ -140,11 +139,7 @@ static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_devic
 
 	ide_pci_setup_ports(dev, d, 14, &hw[0], &hws[0]);
 
-	host = ide_host_alloc(d, hws);
-	if (host)
-		ide_host_register(host, d, hws);
-
-	return 0;
+	return ide_host_add(d, hws, NULL);
 }
 
 static const struct pci_device_id cs5520_pci_tbl[] = {
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index 5eb9d9325184..f84bfb4f600f 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -86,12 +86,10 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 	hw.dev = &dev->dev;
 	hw.chipset = ide_pci;		/* this enables IRQ sharing */
 
-	host = ide_host_alloc(&delkin_cb_port_info, hws);
-	if (host == NULL)
+	rc = ide_host_add(&delkin_cb_port_info, hws, &host);
+	if (rc)
 		goto out_disable;
 
-	ide_host_register(host, &delkin_cb_port_info, hws);
-
 	pci_set_drvdata(dev, host);
 
 	return 0;
@@ -99,7 +97,7 @@ delkin_cb_probe (struct pci_dev *dev, const struct pci_device_id *id)
 out_disable:
 	pci_release_regions(dev);
 	pci_disable_device(dev);
-	return -ENODEV;
+	return rc;
 }
 
 static void
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index d5e2ba6bacd6..94a7ab864236 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -588,7 +588,7 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev,
 	struct scc_ports *ports = pci_get_drvdata(dev);
 	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
-	int i;
+	int i, rc;
 
 	memset(&hw, 0, sizeof(hw));
 	for (i = 0; i <= 8; i++)
@@ -597,11 +597,9 @@ static int scc_ide_setup_pci_device(struct pci_dev *dev,
 	hw.dev = &dev->dev;
 	hw.chipset = ide_pci;
 
-	host = ide_host_alloc(d, hws);
-	if (host == NULL)
-		return -ENOMEM;
-
-	ide_host_register(host, d, hws);
+	rc = ide_host_add(d, hws, &host);
+	if (rc)
+		return rc;
 
 	ports->host = host;
 
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index ecd2f28da1ba..c521bf6e1bf2 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1043,6 +1043,7 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 	ide_hwif_t *hwif;
 	hw_regs_t *hws[] = { hw, NULL, NULL, NULL };
 	struct ide_port_info d = pmac_port_info;
+	int rc;
 
 	pmif->broken_dma = pmif->broken_dma_warn = 0;
 	if (of_device_is_compatible(np, "shasta-ata")) {
@@ -1118,11 +1119,9 @@ static int __devinit pmac_ide_setup_device(pmac_ide_hwif_t *pmif, hw_regs_t *hw)
 			 pmif->mdev ? "macio" : "PCI", pmif->aapl_bus_id,
 			 pmif->mediabay ? " (mediabay)" : "", hw->irq);
 
-	host = ide_host_alloc(&d, hws);
-	if (host == NULL)
-		return -ENOENT;
-
-	ide_host_register(host, &d, hws);
+	rc = ide_host_add(&d, hws, &host);
+	if (rc)
+		return rc;
 
 	hwif = host->ports[0];
 
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 1c0c5570dec8..b15cad58dc81 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -541,7 +541,6 @@ out:
 
 int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 {
-	struct ide_host *host;
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 	int ret;
 
@@ -551,9 +550,7 @@ int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 		/* FIXME: silent failure can happen */
 		ide_pci_setup_ports(dev, d, ret, &hw[0], &hws[0]);
 
-		host = ide_host_alloc(d, hws);
-		if (host)
-			ide_host_register(host, d, hws);
+		ret = ide_host_add(d, hws, NULL);
 	}
 
 	return ret;
@@ -564,7 +561,6 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 			  const struct ide_port_info *d)
 {
 	struct pci_dev *pdev[] = { dev1, dev2 };
-	struct ide_host *host;
 	int ret, i;
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 
@@ -582,9 +578,7 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 		ide_pci_setup_ports(pdev[i], d, ret, &hw[i*2], &hws[i*2]);
 	}
 
-	host = ide_host_alloc(d, hws);
-	if (host)
-		ide_host_register(host, d, hws);
+	ret = ide_host_add(d, hws, NULL);
 out:
 	return ret;
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index a41ae57fafc5..764afd94b917 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1238,6 +1238,8 @@ struct ide_host *ide_host_alloc_all(const struct ide_port_info *, hw_regs_t **);
 struct ide_host *ide_host_alloc(const struct ide_port_info *, hw_regs_t **);
 int ide_host_register(struct ide_host *, const struct ide_port_info *,
 		      hw_regs_t **);
+int ide_host_add(const struct ide_port_info *, hw_regs_t **,
+		 struct ide_host **);
 void ide_host_remove(struct ide_host *);
 int ide_legacy_device_add(const struct ide_port_info *, unsigned long);
 void ide_port_unregister_devices(ide_hwif_t *);
-- 
cgit v1.2.3


From 8a69580e1ea9516caada5eed202afd39546e9809 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Wed, 23 Jul 2008 19:55:59 +0200
Subject: ide: add ide_host_free() helper (take 2)

* Add ide_host_free() helper and convert ide_host_remove() to use it.

* Fix handling of ide_host_register() failure in ide_host_add(),
  icside.c, ide-generic.c, falconide.c and sgiioc4.c.

While at it:

* Fix handling of ide_host_alloc_all() failure in ide-generic.c.

* Fix handling of ide_host_alloc() failure in falconide.c
  (also return the correct error value if no device is found).

v2:
* falconide build fix. (From Stephen Rothwell)

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c       | 21 +++++++++++++++++----
 drivers/ide/ide-generic.c      | 27 +++++++++++++++++++++++----
 drivers/ide/ide-probe.c        | 23 ++++++++++++++++++++---
 drivers/ide/legacy/falconide.c | 22 +++++++++++++++++-----
 drivers/ide/pci/sgiioc4.c      | 14 ++++++++++----
 include/linux/ide.h            |  1 +
 6 files changed, 88 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index 6fa58425466a..f575e8341aec 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -445,6 +445,7 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 	void __iomem *base;
 	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
+	int ret;
 
 	base = ecardm_iomap(ec, ECARD_RES_MEMC, 0, 0);
 	if (!base)
@@ -472,9 +473,15 @@ icside_register_v5(struct icside_state *state, struct expansion_card *ec)
 
 	ecard_set_drvdata(ec, state);
 
-	ide_host_register(host, NULL, hws);
+	ret = ide_host_register(host, NULL, hws);
+	if (ret)
+		goto err_free;
 
 	return 0;
+err_free:
+	ide_host_free(host);
+	ecard_set_drvdata(ec, NULL);
+	return ret;
 }
 
 static const struct ide_port_info icside_v6_port_info __initdata = {
@@ -547,11 +554,17 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec)
 		d.dma_ops = NULL;
 	}
 
-	ide_host_register(host, &d, hws);
+	ret = ide_host_register(host, NULL, hws);
+	if (ret)
+		goto err_free;
 
 	return 0;
-
- out:
+err_free:
+	ide_host_free(host);
+	if (d.dma_ops)
+		free_dma(ec->dma);
+	ecard_set_drvdata(ec, NULL);
+out:
 	return ret;
 }
 
diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index a7082c28d06f..31d98fec775f 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -84,13 +84,14 @@ static int __init ide_generic_init(void)
 {
 	hw_regs_t hw[MAX_HWIFS], *hws[MAX_HWIFS];
 	struct ide_host *host;
-	int i;
+	unsigned long io_addr;
+	int i, rc;
 
 	printk(KERN_INFO DRV_NAME ": please use \"probe_mask=0x3f\" module "
 			 "parameter for probing all legacy ISA IDE ports\n");
 
 	for (i = 0; i < MAX_HWIFS; i++) {
-		unsigned long io_addr = ide_default_io_base(i);
+		io_addr = ide_default_io_base(i);
 
 		hws[i] = NULL;
 
@@ -120,14 +121,32 @@ static int __init ide_generic_init(void)
 	}
 
 	host = ide_host_alloc_all(NULL, hws);
-	if (host)
-		ide_host_register(host, NULL, hws);
+	if (host == NULL) {
+		rc = -ENOMEM;
+		goto err;
+	}
+
+	rc = ide_host_register(host, NULL, hws);
+	if (rc)
+		goto err_free;
 
 	if (ide_generic_sysfs_init())
 		printk(KERN_ERR DRV_NAME ": failed to create ide_generic "
 					 "class\n");
 
 	return 0;
+err_free:
+	ide_host_free(host);
+err:
+	for (i = 0; i < MAX_HWIFS; i++) {
+		if (hws[i] == NULL)
+			continue;
+
+		io_addr = hws[i]->io_ports.data_addr;
+		release_region(io_addr + 0x206, 1);
+		release_region(io_addr, 8);
+	}
+	return rc;
 }
 
 module_init(ide_generic_init);
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 6d57b7cd5424..0ead4537fc63 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1729,12 +1729,17 @@ int ide_host_add(const struct ide_port_info *d, hw_regs_t **hws,
 		 struct ide_host **hostp)
 {
 	struct ide_host *host;
+	int rc;
 
 	host = ide_host_alloc(d, hws);
 	if (host == NULL)
 		return -ENOMEM;
 
-	ide_host_register(host, d, hws);
+	rc = ide_host_register(host, d, hws);
+	if (rc) {
+		ide_host_free(host);
+		return rc;
+	}
 
 	if (hostp)
 		*hostp = host;
@@ -1743,7 +1748,7 @@ int ide_host_add(const struct ide_port_info *d, hw_regs_t **hws,
 }
 EXPORT_SYMBOL_GPL(ide_host_add);
 
-void ide_host_remove(struct ide_host *host)
+void ide_host_free(struct ide_host *host)
 {
 	ide_hwif_t *hwif;
 	int i;
@@ -1754,13 +1759,25 @@ void ide_host_remove(struct ide_host *host)
 		if (hwif == NULL)
 			continue;
 
-		ide_unregister(hwif);
 		ide_free_port_slot(hwif->index);
 		kfree(hwif);
 	}
 
 	kfree(host);
 }
+EXPORT_SYMBOL_GPL(ide_host_free);
+
+void ide_host_remove(struct ide_host *host)
+{
+	int i;
+
+	for (i = 0; i < MAX_HWIFS; i++) {
+		if (host->ports[i])
+			ide_unregister(host->ports[i]);
+	}
+
+	ide_host_free(host);
+}
 EXPORT_SYMBOL_GPL(ide_host_remove);
 
 void ide_port_scan(ide_hwif_t *hwif)
diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c
index 4eb5c3f9fecc..724f95073d80 100644
--- a/drivers/ide/legacy/falconide.c
+++ b/drivers/ide/legacy/falconide.c
@@ -114,9 +114,10 @@ static int __init falconide_init(void)
 {
 	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
+	int rc;
 
 	if (!MACH_IS_ATARI || !ATARIHW_PRESENT(IDE))
-		return 0;
+		return -ENODEV;
 
 	printk(KERN_INFO "ide: Falcon IDE controller\n");
 
@@ -128,13 +129,24 @@ static int __init falconide_init(void)
 	falconide_setup_ports(&hw);
 
 	host = ide_host_alloc(&falconide_port_info, hws);
-	if (host) {
-		ide_get_lock(NULL, NULL);
-		ide_host_register(host, &falconide_port_info, hws);
-		ide_release_lock();
+	if (host == NULL) {
+		rc = -ENOMEM;
+		goto err;
 	}
 
+	ide_get_lock(NULL, NULL);
+	rc = ide_host_register(host, &falconide_port_info, hws);
+	ide_release_lock();
+
+	if (rc)
+		goto err_free;
+
 	return 0;
+err_free:
+	ide_host_free(host);
+err:
+	release_mem_region(ATA_HD_BASE, 0x40);
+	return rc;
 }
 
 module_init(falconide_init);
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 440f43a86ad3..42eef19a18f1 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -603,6 +603,7 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	struct ide_host *host;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 	struct ide_port_info d = sgiioc4_port_info;
+	int rc;
 
 	/*  Get the CmdBlk and CtrlBlk Base Registers */
 	bar0 = pci_resource_start(dev, 0);
@@ -638,17 +639,22 @@ sgiioc4_ide_setup_pci_device(struct pci_dev *dev)
 	writel(0x03, (void __iomem *)(irqport + IOC4_INTR_SET * 4));
 
 	host = ide_host_alloc(&d, hws);
-	if (host == NULL)
+	if (host == NULL) {
+		rc = -ENOMEM;
 		goto err;
+	}
 
-	if (ide_host_register(host, &d, hws))
-		return -EIO;
+	rc = ide_host_register(host, &d, hws);
+	if (rc)
+		goto err_free;
 
 	return 0;
+err_free:
+	ide_host_free(host);
 err:
 	release_mem_region(cmd_phys_base, IOC4_CMD_CTL_BLK_SIZE);
 	iounmap(virt_base);
-	return -ENOMEM;
+	return rc;
 }
 
 static unsigned int __devinit
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 764afd94b917..46d5bfe2fefb 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1236,6 +1236,7 @@ void ide_port_apply_params(ide_hwif_t *);
 
 struct ide_host *ide_host_alloc_all(const struct ide_port_info *, hw_regs_t **);
 struct ide_host *ide_host_alloc(const struct ide_port_info *, hw_regs_t **);
+void ide_host_free(struct ide_host *);
 int ide_host_register(struct ide_host *, const struct ide_port_info *,
 		      hw_regs_t **);
 int ide_host_add(const struct ide_port_info *, hw_regs_t **,
-- 
cgit v1.2.3


From d7c26ebb5bca284ece9db7311a9e180aabae1922 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Wed, 23 Jul 2008 19:55:59 +0200
Subject: ide: push pc callback pointer into the ide_drive_t structure

Refrain from carrying the callback ptr with every packet command since the
callback function is only one anyways. ide_drive_t is probably not the most
suitable place for it right now but is the more sane solution. Besides, these
structs are going to be reorganized anyways during the generic ide rewrite.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index 46d5bfe2fefb..a28823297edc 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -308,7 +308,7 @@ struct ide_acpi_drive_link;
 struct ide_acpi_hwif_link;
 #endif
 
-typedef struct ide_drive_s {
+struct ide_drive_s {
 	char		name[4];	/* drive name, such as "hda" */
         char            driver_req[10];	/* requests specific driver */
 
@@ -400,7 +400,12 @@ typedef struct ide_drive_s {
 	struct list_head list;
 	struct device	gendev;
 	struct completion gendev_rel_comp;	/* to deal with device release() */
-} ide_drive_t;
+
+	/* callback for packet commands */
+	void (*pc_callback)(struct ide_drive_s *);
+};
+
+typedef struct ide_drive_s ide_drive_t;
 
 #define to_ide_device(dev)container_of(dev, ide_drive_t, gendev)
 
-- 
cgit v1.2.3


From 8bcda3bc492c56abed33a8cbf824c410d7ce69f0 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Wed, 23 Jul 2008 19:56:00 +0200
Subject: ide: remove pc->callback member from ide_atapi_pc

There should be no functionality change resulting from this patch.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index a28823297edc..d8ec8afb4ef3 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -662,8 +662,6 @@ struct ide_atapi_pc {
 	 */
 	u8 pc_buf[256];
 
-	void (*callback)(ide_drive_t *);
-
 	/* idetape only */
 	struct idetape_bh *bh;
 	char *b_data;
-- 
cgit v1.2.3


From 3b8ac5398ca8137b41aa5da4d9fad89153450aaa Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Wed, 23 Jul 2008 19:56:01 +0200
Subject: ide: add per-device flags

Push device flags up into ide_drive_t.

There should be no functionality change resulting from this patch.

[bart: IDE_FLAG_* -> IDE_AFLAG_*, dev_flags -> atapi_flags]

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 60 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index d8ec8afb4ef3..fcf2d1bede08 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -308,6 +308,64 @@ struct ide_acpi_drive_link;
 struct ide_acpi_hwif_link;
 #endif
 
+/* ATAPI device flags */
+enum {
+	IDE_AFLAG_DRQ_INTERRUPT		= (1 << 0),
+	IDE_AFLAG_MEDIA_CHANGED		= (1 << 1),
+
+	/* ide-cd */
+	/* Drive cannot lock the door. */
+	IDE_AFLAG_NO_DOORLOCK		= (1 << 2),
+	/* Drive cannot eject the disc. */
+	IDE_AFLAG_NO_EJECT		= (1 << 3),
+	/* Drive is a pre ATAPI 1.2 drive. */
+	IDE_AFLAG_PRE_ATAPI12		= (1 << 4),
+	/* TOC addresses are in BCD. */
+	IDE_AFLAG_TOCADDR_AS_BCD	= (1 << 5),
+	/* TOC track numbers are in BCD. */
+	IDE_AFLAG_TOCTRACKS_AS_BCD	= (1 << 6),
+	/*
+	 * Drive does not provide data in multiples of SECTOR_SIZE
+	 * when more than one interrupt is needed.
+	 */
+	IDE_AFLAG_LIMIT_NFRAMES		= (1 << 7),
+	/* Seeking in progress. */
+	IDE_AFLAG_SEEKING		= (1 << 8),
+	/* Saved TOC information is current. */
+	IDE_AFLAG_TOC_VALID		= (1 << 9),
+	/* We think that the drive door is locked. */
+	IDE_AFLAG_DOOR_LOCKED		= (1 << 10),
+	/* SET_CD_SPEED command is unsupported. */
+	IDE_AFLAG_NO_SPEED_SELECT	= (1 << 11),
+	IDE_AFLAG_VERTOS_300_SSD	= (1 << 12),
+	IDE_AFLAG_VERTOS_600_ESD	= (1 << 13),
+	IDE_AFLAG_SANYO_3CD		= (1 << 14),
+	IDE_AFLAG_FULL_CAPS_PAGE	= (1 << 15),
+	IDE_AFLAG_PLAY_AUDIO_OK		= (1 << 16),
+	IDE_AFLAG_LE_SPEED_FIELDS	= (1 << 17),
+
+	/* ide-floppy */
+	/* Format in progress */
+	IDE_AFLAG_FORMAT_IN_PROGRESS	= (1 << 18),
+	/* Avoid commands not supported in Clik drive */
+	IDE_AFLAG_CLIK_DRIVE		= (1 << 19),
+	/* Requires BH algorithm for packets */
+	IDE_AFLAG_ZIP_DRIVE		= (1 << 20),
+
+	/* ide-tape */
+	IDE_AFLAG_IGNORE_DSC		= (1 << 21),
+	/* 0 When the tape position is unknown */
+	IDE_AFLAG_ADDRESS_VALID		= (1 <<	22),
+	/* Device already opened */
+	IDE_AFLAG_BUSY			= (1 << 23),
+	/* Attempt to auto-detect the current user block size */
+	IDE_AFLAG_DETECT_BS		= (1 << 24),
+	/* Currently on a filemark */
+	IDE_AFLAG_FILEMARK		= (1 << 25),
+	/* 0 = no tape is loaded, so we don't rewind after ejecting */
+	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 26)
+};
+
 struct ide_drive_s {
 	char		name[4];	/* drive name, such as "hda" */
         char            driver_req[10];	/* requests specific driver */
@@ -403,6 +461,8 @@ struct ide_drive_s {
 
 	/* callback for packet commands */
 	void (*pc_callback)(struct ide_drive_s *);
+
+	unsigned long atapi_flags;
 };
 
 typedef struct ide_drive_s ide_drive_t;
-- 
cgit v1.2.3


From ea68d270ff55bcdfa5d07697eb68103b5b02c7bb Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Wed, 23 Jul 2008 19:56:01 +0200
Subject: ide-floppy: convert to using the new atapi_flags (take 2)

while at it, remove PC_FLAG_ZIP_DRIVE from the packed command flags altogether
and query the drive type through drive->atapi_flags.

v2:
ide-floppy fix.

There should be no functionality change resulting from this patch.

[bart: IDE_FLAG_* -> IDE_AFLAG_*, dev_flags -> atapi_flags]

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c  |  5 ++--
 drivers/ide/ide-floppy.c | 73 ++++++++++++++++++------------------------------
 include/linux/ide.h      |  3 +-
 3 files changed, 31 insertions(+), 50 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index f848010c15a5..711a5f6d35e4 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -257,7 +257,7 @@ ide_startstop_t ide_transfer_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	}
 
 	/* Send the actual packet */
-	if ((pc->flags & PC_FLAG_ZIP_DRIVE) == 0)
+	if ((drive->atapi_flags & IDE_AFLAG_ZIP_DRIVE) == 0)
 		hwif->tp_ops->output_data(drive, NULL, rq->cmd, 12);
 
 	return ide_started;
@@ -302,7 +302,8 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 			   bcount, dma);
 
 	/* Issue the packet command */
-	if (pc->flags & PC_FLAG_DRQ_INTERRUPT) {
+	if ((pc->flags & PC_FLAG_DRQ_INTERRUPT) ||
+	    (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT)) {
 		ide_execute_command(drive, WIN_PACKETCMD, handler,
 				    timeout, NULL);
 		return ide_started;
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index cbf84f2e9cce..3d8e6dd0f41e 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -125,26 +125,10 @@ typedef struct ide_floppy_obj {
 	int wp;
 	/* Supports format progress report */
 	int srfp;
-	/* Status/Action flags */
-	unsigned long flags;
 } idefloppy_floppy_t;
 
 #define IDEFLOPPY_TICKS_DELAY	HZ/20	/* default delay for ZIP 100 (50ms) */
 
-/* Floppy flag bits values. */
-enum {
-	/* DRQ interrupt device */
-	IDEFLOPPY_FLAG_DRQ_INTERRUPT		= (1 <<	0),
-	/* Media may have changed */
-	IDEFLOPPY_FLAG_MEDIA_CHANGED		= (1 << 1),
-	/* Format in progress */
-	IDEFLOPPY_FLAG_FORMAT_IN_PROGRESS	= (1 << 2),
-	/* Avoid commands not supported in Clik drive */
-	IDEFLOPPY_FLAG_CLIK_DRIVE		= (1 << 3),
-	/* Requires BH algorithm for packets */
-	IDEFLOPPY_FLAG_ZIP_DRIVE		= (1 << 4),
-};
-
 /* Defines for the MODE SENSE command */
 #define MODE_SENSE_CURRENT		0x00
 #define MODE_SENSE_CHANGEABLE		0x01
@@ -429,7 +413,7 @@ static ide_startstop_t idefloppy_start_pc_transfer(ide_drive_t *drive)
 	 * 40 and 50msec work well. idefloppy_pc_intr will not be actually
 	 * used until after the packet is moved in about 50 msec.
 	 */
-	if (pc->flags & PC_FLAG_ZIP_DRIVE) {
+	if (drive->atapi_flags & IDE_AFLAG_ZIP_DRIVE) {
 		timeout = floppy->ticks;
 		expiry = &idefloppy_transfer_pc;
 	} else {
@@ -649,12 +633,6 @@ static ide_startstop_t idefloppy_do_request(ide_drive_t *drive,
 		return ide_stopped;
 	}
 
-	if (floppy->flags & IDEFLOPPY_FLAG_DRQ_INTERRUPT)
-		pc->flags |= PC_FLAG_DRQ_INTERRUPT;
-
-	if (floppy->flags & IDEFLOPPY_FLAG_ZIP_DRIVE)
-		pc->flags |= PC_FLAG_ZIP_DRIVE;
-
 	pc->rq = rq;
 
 	return idefloppy_issue_pc(drive, pc);
@@ -798,7 +776,7 @@ static int ide_floppy_get_capacity(ide_drive_t *drive)
 		switch (pc.buf[desc_start + 4] & 0x03) {
 		/* Clik! drive returns this instead of CAPACITY_CURRENT */
 		case CAPACITY_UNFORMATTED:
-			if (!(floppy->flags & IDEFLOPPY_FLAG_CLIK_DRIVE))
+			if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE))
 				/*
 				 * If it is not a clik drive, break out
 				 * (maintains previous driver behaviour)
@@ -844,7 +822,7 @@ static int ide_floppy_get_capacity(ide_drive_t *drive)
 	}
 
 	/* Clik! disk does not support get_flexible_disk_page */
-	if (!(floppy->flags & IDEFLOPPY_FLAG_CLIK_DRIVE))
+	if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE))
 		(void) ide_floppy_get_flexible_disk_page(drive);
 
 	set_capacity(floppy->disk, floppy->blocks * floppy->bs_factor);
@@ -1046,7 +1024,7 @@ static void idefloppy_setup(ide_drive_t *drive, idefloppy_floppy_t *floppy)
 	drive->pc_callback = ide_floppy_callback;
 
 	if (((gcw[0] & 0x60) >> 5) == 1)
-		floppy->flags |= IDEFLOPPY_FLAG_DRQ_INTERRUPT;
+		drive->atapi_flags |= IDE_AFLAG_DRQ_INTERRUPT;
 	/*
 	 * We used to check revisions here. At this point however I'm giving up.
 	 * Just assume they are all broken, its easier.
@@ -1057,7 +1035,7 @@ static void idefloppy_setup(ide_drive_t *drive, idefloppy_floppy_t *floppy)
 	 * we'll leave the limitation below for the 2.2.x tree.
 	 */
 	if (!strncmp(drive->id->model, "IOMEGA ZIP 100 ATAPI", 20)) {
-		floppy->flags |= IDEFLOPPY_FLAG_ZIP_DRIVE;
+		drive->atapi_flags |= IDE_AFLAG_ZIP_DRIVE;
 		/* This value will be visible in the /proc/ide/hdx/settings */
 		floppy->ticks = IDEFLOPPY_TICKS_DELAY;
 		blk_queue_max_sectors(drive->queue, 64);
@@ -1069,7 +1047,7 @@ static void idefloppy_setup(ide_drive_t *drive, idefloppy_floppy_t *floppy)
 	 */
 	if (strncmp(drive->id->model, "IOMEGA Clik!", 11) == 0) {
 		blk_queue_max_sectors(drive->queue, 64);
-		floppy->flags |= IDEFLOPPY_FLAG_CLIK_DRIVE;
+		drive->atapi_flags |= IDE_AFLAG_CLIK_DRIVE;
 	}
 
 	(void) ide_floppy_get_capacity(drive);
@@ -1158,7 +1136,7 @@ static int idefloppy_open(struct inode *inode, struct file *filp)
 	floppy->openers++;
 
 	if (floppy->openers == 1) {
-		floppy->flags &= ~IDEFLOPPY_FLAG_FORMAT_IN_PROGRESS;
+		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
 		/* Just in case */
 
 		idefloppy_init_pc(&pc);
@@ -1185,14 +1163,14 @@ static int idefloppy_open(struct inode *inode, struct file *filp)
 			ret = -EROFS;
 			goto out_put_floppy;
 		}
-		floppy->flags |= IDEFLOPPY_FLAG_MEDIA_CHANGED;
+		drive->atapi_flags |= IDE_AFLAG_MEDIA_CHANGED;
 		/* IOMEGA Clik! drives do not support lock/unlock commands */
-		if (!(floppy->flags & IDEFLOPPY_FLAG_CLIK_DRIVE)) {
+		if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE)) {
 			idefloppy_create_prevent_cmd(&pc, 1);
 			(void) idefloppy_queue_pc_tail(drive, &pc);
 		}
 		check_disk_change(inode->i_bdev);
-	} else if (floppy->flags & IDEFLOPPY_FLAG_FORMAT_IN_PROGRESS) {
+	} else if (drive->atapi_flags & IDE_AFLAG_FORMAT_IN_PROGRESS) {
 		ret = -EBUSY;
 		goto out_put_floppy;
 	}
@@ -1215,12 +1193,12 @@ static int idefloppy_release(struct inode *inode, struct file *filp)
 
 	if (floppy->openers == 1) {
 		/* IOMEGA Clik! drives do not support lock/unlock commands */
-		if (!(floppy->flags & IDEFLOPPY_FLAG_CLIK_DRIVE)) {
+		if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE)) {
 			idefloppy_create_prevent_cmd(&pc, 0);
 			(void) idefloppy_queue_pc_tail(drive, &pc);
 		}
 
-		floppy->flags &= ~IDEFLOPPY_FLAG_FORMAT_IN_PROGRESS;
+		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
 	}
 
 	floppy->openers--;
@@ -1241,15 +1219,17 @@ static int idefloppy_getgeo(struct block_device *bdev, struct hd_geometry *geo)
 	return 0;
 }
 
-static int ide_floppy_lockdoor(idefloppy_floppy_t *floppy,
-		struct ide_atapi_pc *pc, unsigned long arg, unsigned int cmd)
+static int ide_floppy_lockdoor(ide_drive_t *drive, struct ide_atapi_pc *pc,
+			       unsigned long arg, unsigned int cmd)
 {
+	idefloppy_floppy_t *floppy = drive->driver_data;
+
 	if (floppy->openers > 1)
 		return -EBUSY;
 
 	/* The IOMEGA Clik! Drive doesn't support this command -
 	 * no room for an eject mechanism */
-	if (!(floppy->flags & IDEFLOPPY_FLAG_CLIK_DRIVE)) {
+	if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE)) {
 		int prevent = arg ? 1 : 0;
 
 		if (cmd == CDROMEJECT)
@@ -1270,16 +1250,17 @@ static int ide_floppy_lockdoor(idefloppy_floppy_t *floppy,
 static int ide_floppy_format_unit(idefloppy_floppy_t *floppy,
 				  int __user *arg)
 {
-	int blocks, length, flags, err = 0;
 	struct ide_atapi_pc pc;
+	ide_drive_t *drive = floppy->drive;
+	int blocks, length, flags, err = 0;
 
 	if (floppy->openers > 1) {
 		/* Don't format if someone is using the disk */
-		floppy->flags &= ~IDEFLOPPY_FLAG_FORMAT_IN_PROGRESS;
+		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
 		return -EBUSY;
 	}
 
-	floppy->flags |= IDEFLOPPY_FLAG_FORMAT_IN_PROGRESS;
+	drive->atapi_flags |= IDE_AFLAG_FORMAT_IN_PROGRESS;
 
 	/*
 	 * Send ATAPI_FORMAT_UNIT to the drive.
@@ -1303,15 +1284,15 @@ static int ide_floppy_format_unit(idefloppy_floppy_t *floppy,
 		goto out;
 	}
 
-	(void) idefloppy_get_sfrp_bit(floppy->drive);
+	(void) idefloppy_get_sfrp_bit(drive);
 	idefloppy_create_format_unit_cmd(&pc, blocks, length, flags);
 
-	if (idefloppy_queue_pc_tail(floppy->drive, &pc))
+	if (idefloppy_queue_pc_tail(drive, &pc))
 		err = -EIO;
 
 out:
 	if (err)
-		floppy->flags &= ~IDEFLOPPY_FLAG_FORMAT_IN_PROGRESS;
+		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
 	return err;
 }
 
@@ -1330,7 +1311,7 @@ static int idefloppy_ioctl(struct inode *inode, struct file *file,
 	case CDROMEJECT:
 		/* fall through */
 	case CDROM_LOCKDOOR:
-		return ide_floppy_lockdoor(floppy, &pc, arg, cmd);
+		return ide_floppy_lockdoor(drive, &pc, arg, cmd);
 	case IDEFLOPPY_IOCTL_FORMAT_SUPPORTED:
 		return 0;
 	case IDEFLOPPY_IOCTL_FORMAT_GET_CAPACITY:
@@ -1371,8 +1352,8 @@ static int idefloppy_media_changed(struct gendisk *disk)
 		drive->attach = 0;
 		return 0;
 	}
-	ret = !!(floppy->flags & IDEFLOPPY_FLAG_MEDIA_CHANGED);
-	floppy->flags &= ~IDEFLOPPY_FLAG_MEDIA_CHANGED;
+	ret = !!(drive->atapi_flags & IDE_AFLAG_MEDIA_CHANGED);
+	drive->atapi_flags &= ~IDE_AFLAG_MEDIA_CHANGED;
 	return ret;
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index fcf2d1bede08..147bfee1fe70 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -687,8 +687,7 @@ enum {
 	PC_FLAG_WRITING			= (1 << 6),
 	/* command timed out */
 	PC_FLAG_TIMEDOUT		= (1 << 7),
-	PC_FLAG_ZIP_DRIVE		= (1 << 8),
-	PC_FLAG_DRQ_INTERRUPT		= (1 << 9),
+	PC_FLAG_DRQ_INTERRUPT		= (1 << 8),
 };
 
 struct ide_atapi_pc {
-- 
cgit v1.2.3


From ac77ef8b03677c8ae8afe77bccc5f6a969193a79 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Wed, 23 Jul 2008 19:56:01 +0200
Subject: ide: remove unused PC_FLAG_DRQ_INTERRUPT

There should be no functionality change resulting from this patch.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c | 3 +--
 include/linux/ide.h     | 1 -
 2 files changed, 1 insertion(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 711a5f6d35e4..adf04f99cdeb 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -302,8 +302,7 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 			   bcount, dma);
 
 	/* Issue the packet command */
-	if ((pc->flags & PC_FLAG_DRQ_INTERRUPT) ||
-	    (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT)) {
+	if (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT) {
 		ide_execute_command(drive, WIN_PACKETCMD, handler,
 				    timeout, NULL);
 		return ide_started;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 147bfee1fe70..d67ccca2b964 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -687,7 +687,6 @@ enum {
 	PC_FLAG_WRITING			= (1 << 6),
 	/* command timed out */
 	PC_FLAG_TIMEDOUT		= (1 << 7),
-	PC_FLAG_DRQ_INTERRUPT		= (1 << 8),
 };
 
 struct ide_atapi_pc {
-- 
cgit v1.2.3


From a5bf6190417cbbf80443a9f71c65b653e13e9982 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Thu, 10 Jul 2008 18:38:33 +0300
Subject: UBI: add ubi_sync() interface

To flush MTD device caches.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/kapi.c  | 24 ++++++++++++++++++++++++
 include/linux/mtd/ubi.h |  1 +
 2 files changed, 25 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c
index 51508832566d..e65c8e0bcd5d 100644
--- a/drivers/mtd/ubi/kapi.c
+++ b/drivers/mtd/ubi/kapi.c
@@ -632,3 +632,27 @@ int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum)
 	return vol->eba_tbl[lnum] >= 0;
 }
 EXPORT_SYMBOL_GPL(ubi_is_mapped);
+
+/**
+ * ubi_sync - synchronize UBI device buffers.
+ * @ubi_num: UBI device to synchronize
+ *
+ * The underlying MTD device may cache data in hardware or in software. This
+ * function ensures the caches are flushed. Returns zero in case of success and
+ * a negative error code in case of failure.
+ */
+int ubi_sync(int ubi_num)
+{
+	struct ubi_device *ubi;
+
+	ubi = ubi_get_device(ubi_num);
+	if (!ubi)
+		return -ENODEV;
+
+	if (ubi->mtd->sync)
+		ubi->mtd->sync(ubi->mtd);
+
+	ubi_put_device(ubi);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ubi_sync);
diff --git a/include/linux/mtd/ubi.h b/include/linux/mtd/ubi.h
index f71201d0f3e7..83302bbbddb4 100644
--- a/include/linux/mtd/ubi.h
+++ b/include/linux/mtd/ubi.h
@@ -152,6 +152,7 @@ int ubi_leb_erase(struct ubi_volume_desc *desc, int lnum);
 int ubi_leb_unmap(struct ubi_volume_desc *desc, int lnum);
 int ubi_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype);
 int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum);
+int ubi_sync(int ubi_num);
 
 /*
  * This function is the same as the 'ubi_leb_read()' function, but it does not
-- 
cgit v1.2.3


From 85c6e6e28259e9b58b8984db536c45bc3161f40c Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 16 Jul 2008 10:25:56 +0300
Subject: UBI: amend commentaries

Hch asked not to use "unit" for sub-systems, let it be so.
Also some other commentaries modifications.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c     |  2 +-
 drivers/mtd/ubi/debug.h     |  6 +--
 drivers/mtd/ubi/eba.c       | 22 +++++------
 drivers/mtd/ubi/io.c        | 22 +++++------
 drivers/mtd/ubi/scan.c      | 28 +++++++-------
 drivers/mtd/ubi/scan.h      | 19 +++++----
 drivers/mtd/ubi/ubi-media.h | 23 +++++------
 drivers/mtd/ubi/ubi.h       | 37 +++++++++---------
 drivers/mtd/ubi/wl.c        | 94 ++++++++++++++++++++++-----------------------
 include/linux/mtd/ubi.h     |  4 +-
 10 files changed, 129 insertions(+), 128 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index a5b19944eca8..27271fe32e02 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -524,7 +524,7 @@ out_si:
 }
 
 /**
- * io_init - initialize I/O unit for a given UBI device.
+ * io_init - initialize I/O sub-system for a given UBI device.
  * @ubi: UBI device description object
  *
  * If @ubi->vid_hdr_offset or @ubi->leb_start is zero, default offsets are
diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h
index 8ea99d8c9e1f..7d8d77c31dfe 100644
--- a/drivers/mtd/ubi/debug.h
+++ b/drivers/mtd/ubi/debug.h
@@ -76,21 +76,21 @@ void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req);
 #endif /* CONFIG_MTD_UBI_DEBUG_MSG */
 
 #ifdef CONFIG_MTD_UBI_DEBUG_MSG_EBA
-/* Messages from the eraseblock association unit */
+/* Messages from the eraseblock association sub-system */
 #define dbg_eba(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
 #else
 #define dbg_eba(fmt, ...) ({})
 #endif
 
 #ifdef CONFIG_MTD_UBI_DEBUG_MSG_WL
-/* Messages from the wear-leveling unit */
+/* Messages from the wear-leveling sub-system */
 #define dbg_wl(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
 #else
 #define dbg_wl(fmt, ...) ({})
 #endif
 
 #ifdef CONFIG_MTD_UBI_DEBUG_MSG_IO
-/* Messages from the input/output unit */
+/* Messages from the input/output sub-system */
 #define dbg_io(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
 #else
 #define dbg_io(fmt, ...) ({})
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 8dc488fc0cdf..613cd1e51648 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -19,20 +19,20 @@
  */
 
 /*
- * The UBI Eraseblock Association (EBA) unit.
+ * The UBI Eraseblock Association (EBA) sub-system.
  *
- * This unit is responsible for I/O to/from logical eraseblock.
+ * This sub-system is responsible for I/O to/from logical eraseblock.
  *
  * Although in this implementation the EBA table is fully kept and managed in
  * RAM, which assumes poor scalability, it might be (partially) maintained on
  * flash in future implementations.
  *
- * The EBA unit implements per-logical eraseblock locking. Before accessing a
- * logical eraseblock it is locked for reading or writing. The per-logical
- * eraseblock locking is implemented by means of the lock tree. The lock tree
- * is an RB-tree which refers all the currently locked logical eraseblocks. The
- * lock tree elements are &struct ubi_ltree_entry objects. They are indexed by
- * (@vol_id, @lnum) pairs.
+ * The EBA sub-system implements per-logical eraseblock locking. Before
+ * accessing a logical eraseblock it is locked for reading or writing. The
+ * per-logical eraseblock locking is implemented by means of the lock tree. The
+ * lock tree is an RB-tree which refers all the currently locked logical
+ * eraseblocks. The lock tree elements are &struct ubi_ltree_entry objects.
+ * They are indexed by (@vol_id, @lnum) pairs.
  *
  * EBA also maintains the global sequence counter which is incremented each
  * time a logical eraseblock is mapped to a physical eraseblock and it is
@@ -1128,7 +1128,7 @@ out_unlock_leb:
 }
 
 /**
- * ubi_eba_init_scan - initialize the EBA unit using scanning information.
+ * ubi_eba_init_scan - initialize the EBA sub-system using scanning information.
  * @ubi: UBI device description object
  * @si: scanning information
  *
@@ -1143,7 +1143,7 @@ int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si)
 	struct ubi_scan_leb *seb;
 	struct rb_node *rb;
 
-	dbg_eba("initialize EBA unit");
+	dbg_eba("initialize EBA sub-system");
 
 	spin_lock_init(&ubi->ltree_lock);
 	mutex_init(&ubi->alc_mutex);
@@ -1209,7 +1209,7 @@ int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si)
 		ubi->rsvd_pebs  += ubi->beb_rsvd_pebs;
 	}
 
-	dbg_eba("EBA unit is initialized");
+	dbg_eba("EBA sub-system is initialized");
 	return 0;
 
 out_free:
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 4ac11df7b048..561e7b2f96cb 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -20,15 +20,15 @@
  */
 
 /*
- * UBI input/output unit.
+ * UBI input/output sub-system.
  *
- * This unit provides a uniform way to work with all kinds of the underlying
- * MTD devices. It also implements handy functions for reading and writing UBI
- * headers.
+ * This sub-system provides a uniform way to work with all kinds of the
+ * underlying MTD devices. It also implements handy functions for reading and
+ * writing UBI headers.
  *
  * We are trying to have a paranoid mindset and not to trust to what we read
- * from the flash media in order to be more secure and robust. So this unit
- * validates every single header it reads from the flash media.
+ * from the flash media in order to be more secure and robust. So this
+ * sub-system validates every single header it reads from the flash media.
  *
  * Some words about how the eraseblock headers are stored.
  *
@@ -79,11 +79,11 @@
  * 512-byte chunks, we have to allocate one more buffer and copy our VID header
  * to offset 448 of this buffer.
  *
- * The I/O unit does the following trick in order to avoid this extra copy.
- * It always allocates a @ubi->vid_hdr_alsize bytes buffer for the VID header
- * and returns a pointer to offset @ubi->vid_hdr_shift of this buffer. When the
- * VID header is being written out, it shifts the VID header pointer back and
- * writes the whole sub-page.
+ * The I/O sub-system does the following trick in order to avoid this extra
+ * copy. It always allocates a @ubi->vid_hdr_alsize bytes buffer for the VID
+ * header and returns a pointer to offset @ubi->vid_hdr_shift of this buffer.
+ * When the VID header is being written out, it shifts the VID header pointer
+ * back and writes the whole sub-page.
  */
 
 #include <linux/crc32.h>
diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c
index 96d410e106ab..892c2ba49777 100644
--- a/drivers/mtd/ubi/scan.c
+++ b/drivers/mtd/ubi/scan.c
@@ -19,9 +19,9 @@
  */
 
 /*
- * UBI scanning unit.
+ * UBI scanning sub-system.
  *
- * This unit is responsible for scanning the flash media, checking UBI
+ * This sub-system is responsible for scanning the flash media, checking UBI
  * headers and providing complete information about the UBI flash image.
  *
  * The scanning information is represented by a &struct ubi_scan_info' object.
@@ -103,7 +103,7 @@ static int add_to_list(struct ubi_scan_info *si, int pnum, int ec,
  * non-zero if an inconsistency was found and zero if not.
  *
  * Note, UBI does sanity check of everything it reads from the flash media.
- * Most of the checks are done in the I/O unit. Here we check that the
+ * Most of the checks are done in the I/O sub-system. Here we check that the
  * information in the VID header is consistent to the information in other VID
  * headers of the same volume.
  */
@@ -256,8 +256,8 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb,
 		 * that versions that are close to %0xFFFFFFFF are less then
 		 * versions that are close to %0.
 		 *
-		 * The UBI WL unit guarantees that the number of pending tasks
-		 * is not greater then %0x7FFFFFFF. So, if the difference
+		 * The UBI WL sub-system guarantees that the number of pending
+		 * tasks is not greater then %0x7FFFFFFF. So, if the difference
 		 * between any two versions is greater or equivalent to
 		 * %0x7FFFFFFF, there was an overflow and the logical
 		 * eraseblock with lower version is actually newer then the one
@@ -645,9 +645,9 @@ void ubi_scan_rm_volume(struct ubi_scan_info *si, struct ubi_scan_volume *sv)
  *
  * This function erases physical eraseblock 'pnum', and writes the erase
  * counter header to it. This function should only be used on UBI device
- * initialization stages, when the EBA unit had not been yet initialized. This
- * function returns zero in case of success and a negative error code in case
- * of failure.
+ * initialization stages, when the EBA sub-system had not been yet initialized.
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
  */
 int ubi_scan_erase_peb(struct ubi_device *ubi, const struct ubi_scan_info *si,
 		       int pnum, int ec)
@@ -687,9 +687,10 @@ out_free:
  * @si: scanning information
  *
  * This function returns a free physical eraseblock. It is supposed to be
- * called on the UBI initialization stages when the wear-leveling unit is not
- * initialized yet. This function picks a physical eraseblocks from one of the
- * lists, writes the EC header if it is needed, and removes it from the list.
+ * called on the UBI initialization stages when the wear-leveling sub-system is
+ * not initialized yet. This function picks a physical eraseblocks from one of
+ * the lists, writes the EC header if it is needed, and removes it from the
+ * list.
  *
  * This function returns scanning physical eraseblock information in case of
  * success and an error code in case of failure.
@@ -764,8 +765,9 @@ static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si, int pnum
 		return err;
 	else if (err) {
 		/*
-		 * FIXME: this is actually duty of the I/O unit to initialize
-		 * this, but MTD does not provide enough information.
+		 * FIXME: this is actually duty of the I/O sub-system to
+		 * initialize this, but MTD does not provide enough
+		 * information.
 		 */
 		si->bad_peb_count += 1;
 		return 0;
diff --git a/drivers/mtd/ubi/scan.h b/drivers/mtd/ubi/scan.h
index 966b9b682a42..4e2e3cc0becd 100644
--- a/drivers/mtd/ubi/scan.h
+++ b/drivers/mtd/ubi/scan.h
@@ -59,16 +59,16 @@ struct ubi_scan_leb {
  * @leb_count: number of logical eraseblocks in this volume
  * @vol_type: volume type
  * @used_ebs: number of used logical eraseblocks in this volume (only for
- * static volumes)
+ *            static volumes)
  * @last_data_size: amount of data in the last logical eraseblock of this
- * volume (always equivalent to the usable logical eraseblock size in case of
- * dynamic volumes)
+ *                  volume (always equivalent to the usable logical eraseblock
+ *                  size in case of dynamic volumes)
  * @data_pad: how many bytes at the end of logical eraseblocks of this volume
- * are not used (due to volume alignment)
+ *            are not used (due to volume alignment)
  * @compat: compatibility flags of this volume
  * @rb: link in the volume RB-tree
  * @root: root of the RB-tree containing all the eraseblock belonging to this
- * volume (&struct ubi_scan_leb objects)
+ *        volume (&struct ubi_scan_leb objects)
  *
  * One object of this type is allocated for each volume during scanning.
  */
@@ -92,8 +92,8 @@ struct ubi_scan_volume {
  * @free: list of free physical eraseblocks
  * @erase: list of physical eraseblocks which have to be erased
  * @alien: list of physical eraseblocks which should not be used by UBI (e.g.,
+ *         those belonging to "preserve"-compatible internal volumes)
  * @bad_peb_count: count of bad physical eraseblocks
- * those belonging to "preserve"-compatible internal volumes)
  * @vols_found: number of volumes found during scanning
  * @highest_vol_id: highest volume ID
  * @alien_peb_count: count of physical eraseblocks in the @alien list
@@ -106,8 +106,8 @@ struct ubi_scan_volume {
  * @ec_count: a temporary variable used when calculating @mean_ec
  *
  * This data structure contains the result of scanning and may be used by other
- * UBI units to build final UBI data structures, further error-recovery and so
- * on.
+ * UBI sub-systems to build final UBI data structures, further error-recovery
+ * and so on.
  */
 struct ubi_scan_info {
 	struct rb_root volumes;
@@ -132,8 +132,7 @@ struct ubi_device;
 struct ubi_vid_hdr;
 
 /*
- * ubi_scan_move_to_list - move a physical eraseblock from the volume tree to a
- * list.
+ * ubi_scan_move_to_list - move a PEB from the volume tree to a list.
  *
  * @sv: volume scanning information
  * @seb: scanning eraseblock infprmation
diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h
index c3185d9fd048..26bb7af9787a 100644
--- a/drivers/mtd/ubi/ubi-media.h
+++ b/drivers/mtd/ubi/ubi-media.h
@@ -98,10 +98,11 @@ enum {
  * Compatibility constants used by internal volumes.
  *
  * @UBI_COMPAT_DELETE: delete this internal volume before anything is written
- * to the flash
+ *                     to the flash
  * @UBI_COMPAT_RO: attach this device in read-only mode
  * @UBI_COMPAT_PRESERVE: preserve this internal volume - do not touch its
- * physical eraseblocks, don't allow the wear-leveling unit to move them
+ *                       physical eraseblocks, don't allow the wear-leveling
+ *                       sub-system to move them
  * @UBI_COMPAT_REJECT: reject this UBI image
  */
 enum {
@@ -123,7 +124,7 @@ enum {
  * struct ubi_ec_hdr - UBI erase counter header.
  * @magic: erase counter header magic number (%UBI_EC_HDR_MAGIC)
  * @version: version of UBI implementation which is supposed to accept this
- * UBI image
+ *           UBI image
  * @padding1: reserved for future, zeroes
  * @ec: the erase counter
  * @vid_hdr_offset: where the VID header starts
@@ -159,20 +160,20 @@ struct ubi_ec_hdr {
  * struct ubi_vid_hdr - on-flash UBI volume identifier header.
  * @magic: volume identifier header magic number (%UBI_VID_HDR_MAGIC)
  * @version: UBI implementation version which is supposed to accept this UBI
- * image (%UBI_VERSION)
+ *           image (%UBI_VERSION)
  * @vol_type: volume type (%UBI_VID_DYNAMIC or %UBI_VID_STATIC)
  * @copy_flag: if this logical eraseblock was copied from another physical
- * eraseblock (for wear-leveling reasons)
+ *             eraseblock (for wear-leveling reasons)
  * @compat: compatibility of this volume (%0, %UBI_COMPAT_DELETE,
- * %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT)
+ *          %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT)
  * @vol_id: ID of this volume
  * @lnum: logical eraseblock number
  * @leb_ver: version of this logical eraseblock (IMPORTANT: obsolete, to be
- * removed, kept only for not breaking older UBI users)
+ *           removed, kept only for not breaking older UBI users)
  * @data_size: how many bytes of data this logical eraseblock contains
  * @used_ebs: total number of used logical eraseblocks in this volume
  * @data_pad: how many bytes at the end of this physical eraseblock are not
- * used
+ *            used
  * @data_crc: CRC checksum of the data stored in this logical eraseblock
  * @padding1: reserved for future, zeroes
  * @sqnum: sequence number
@@ -248,9 +249,9 @@ struct ubi_ec_hdr {
  * The @data_crc field contains the CRC checksum of the contents of the logical
  * eraseblock if this is a static volume. In case of dynamic volumes, it does
  * not contain the CRC checksum as a rule. The only exception is when the
- * data of the physical eraseblock was moved by the wear-leveling unit, then
- * the wear-leveling unit calculates the data CRC and stores it in the
- * @data_crc field. And of course, the @copy_flag is %in this case.
+ * data of the physical eraseblock was moved by the wear-leveling sub-system,
+ * then the wear-leveling sub-system calculates the data CRC and stores it in
+ * the @data_crc field. And of course, the @copy_flag is %in this case.
  *
  * The @data_size field is used only for static volumes because UBI has to know
  * how many bytes of data are stored in this eraseblock. For dynamic volumes,
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 940f6b7deec3..1fc32c863b78 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -74,15 +74,15 @@
 #define UBI_IO_RETRIES 3
 
 /*
- * Error codes returned by the I/O unit.
+ * Error codes returned by the I/O sub-system.
  *
  * UBI_IO_PEB_EMPTY: the physical eraseblock is empty, i.e. it contains only
- * 0xFF bytes
+ *                   %0xFF bytes
  * UBI_IO_PEB_FREE: the physical eraseblock is free, i.e. it contains only a
- * valid erase counter header, and the rest are %0xFF bytes
+ *                  valid erase counter header, and the rest are %0xFF bytes
  * UBI_IO_BAD_EC_HDR: the erase counter header is corrupted (bad magic or CRC)
  * UBI_IO_BAD_VID_HDR: the volume identifier header is corrupted (bad magic or
- * CRC)
+ *                     CRC)
  * UBI_IO_BITFLIPS: bit-flips were detected and corrected
  */
 enum {
@@ -99,9 +99,9 @@ enum {
  * @ec: erase counter
  * @pnum: physical eraseblock number
  *
- * This data structure is used in the WL unit. Each physical eraseblock has a
- * corresponding &struct wl_entry object which may be kept in different
- * RB-trees. See WL unit for details.
+ * This data structure is used in the WL sub-system. Each physical eraseblock
+ * has a corresponding &struct wl_entry object which may be kept in different
+ * RB-trees. See WL sub-system for details.
  */
 struct ubi_wl_entry {
 	struct rb_node rb;
@@ -118,10 +118,10 @@ struct ubi_wl_entry {
  * @mutex: read/write mutex to implement read/write access serialization to
  *         the (@vol_id, @lnum) logical eraseblock
  *
- * This data structure is used in the EBA unit to implement per-LEB locking.
- * When a logical eraseblock is being locked - corresponding
+ * This data structure is used in the EBA sub-system to implement per-LEB
+ * locking. When a logical eraseblock is being locked - corresponding
  * &struct ubi_ltree_entry object is inserted to the lock tree (@ubi->ltree).
- * See EBA unit for details.
+ * See EBA sub-system for details.
  */
 struct ubi_ltree_entry {
 	struct rb_node rb;
@@ -225,7 +225,7 @@ struct ubi_volume {
 #ifdef CONFIG_MTD_UBI_GLUEBI
 	/*
 	 * Gluebi-related stuff may be compiled out.
-	 * TODO: this should not be built into UBI but should be a separate
+	 * Note: this should not be built into UBI but should be a separate
 	 * ubimtd driver which works on top of UBI and emulates MTD devices.
 	 */
 	struct ubi_volume_desc *gluebi_desc;
@@ -235,8 +235,7 @@ struct ubi_volume {
 };
 
 /**
- * struct ubi_volume_desc - descriptor of the UBI volume returned when it is
- * opened.
+ * struct ubi_volume_desc - UBI volume descriptor returned when it is opened.
  * @vol: reference to the corresponding volume description object
  * @mode: open mode (%UBI_READONLY, %UBI_READWRITE, or %UBI_EXCLUSIVE)
  */
@@ -316,11 +315,11 @@ struct ubi_wl_entry;
  * @ro_mode: if the UBI device is in read-only mode
  * @leb_size: logical eraseblock size
  * @leb_start: starting offset of logical eraseblocks within physical
- * eraseblocks
+ *             eraseblocks
  * @ec_hdr_alsize: size of the EC header aligned to @hdrs_min_io_size
  * @vid_hdr_alsize: size of the VID header aligned to @hdrs_min_io_size
  * @vid_hdr_offset: starting offset of the volume identifier header (might be
- * unaligned)
+ *                  unaligned)
  * @vid_hdr_aloffset: starting offset of the VID header aligned to
  * @hdrs_min_io_size
  * @vid_hdr_shift: contains @vid_hdr_offset - @vid_hdr_aloffset
@@ -356,16 +355,16 @@ struct ubi_device {
 	struct mutex volumes_mutex;
 
 	int max_ec;
-	/* TODO: mean_ec is not updated run-time, fix */
+	/* Note, mean_ec is not updated run-time - should be fixed */
 	int mean_ec;
 
-	/* EBA unit's stuff */
+	/* EBA sub-system's stuff */
 	unsigned long long global_sqnum;
 	spinlock_t ltree_lock;
 	struct rb_root ltree;
 	struct mutex alc_mutex;
 
-	/* Wear-leveling unit's stuff */
+	/* Wear-leveling sub-system's stuff */
 	struct rb_root used;
 	struct rb_root free;
 	struct rb_root scrub;
@@ -388,7 +387,7 @@ struct ubi_device {
 	int thread_enabled;
 	char bgt_name[sizeof(UBI_BGT_NAME_PATTERN)+2];
 
-	/* I/O unit's stuff */
+	/* I/O sub-system's stuff */
 	long long flash_size;
 	int peb_count;
 	int peb_size;
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index cc8fe2934d2b..761952ba125b 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -19,22 +19,22 @@
  */
 
 /*
- * UBI wear-leveling unit.
+ * UBI wear-leveling sub-system.
  *
- * This unit is responsible for wear-leveling. It works in terms of physical
- * eraseblocks and erase counters and knows nothing about logical eraseblocks,
- * volumes, etc. From this unit's perspective all physical eraseblocks are of
- * two types - used and free. Used physical eraseblocks are those that were
- * "get" by the 'ubi_wl_get_peb()' function, and free physical eraseblocks are
- * those that were put by the 'ubi_wl_put_peb()' function.
+ * This sub-system is responsible for wear-leveling. It works in terms of
+ * physical* eraseblocks and erase counters and knows nothing about logical
+ * eraseblocks, volumes, etc. From this sub-system's perspective all physical
+ * eraseblocks are of two types - used and free. Used physical eraseblocks are
+ * those that were "get" by the 'ubi_wl_get_peb()' function, and free physical
+ * eraseblocks are those that were put by the 'ubi_wl_put_peb()' function.
  *
  * Physical eraseblocks returned by 'ubi_wl_get_peb()' have only erase counter
- * header. The rest of the physical eraseblock contains only 0xFF bytes.
+ * header. The rest of the physical eraseblock contains only %0xFF bytes.
  *
- * When physical eraseblocks are returned to the WL unit by means of the
+ * When physical eraseblocks are returned to the WL sub-system by means of the
  * 'ubi_wl_put_peb()' function, they are scheduled for erasure. The erasure is
  * done asynchronously in context of the per-UBI device background thread,
- * which is also managed by the WL unit.
+ * which is also managed by the WL sub-system.
  *
  * The wear-leveling is ensured by means of moving the contents of used
  * physical eraseblocks with low erase counter to free physical eraseblocks
@@ -43,34 +43,36 @@
  * The 'ubi_wl_get_peb()' function accepts data type hints which help to pick
  * an "optimal" physical eraseblock. For example, when it is known that the
  * physical eraseblock will be "put" soon because it contains short-term data,
- * the WL unit may pick a free physical eraseblock with low erase counter, and
- * so forth.
+ * the WL sub-system may pick a free physical eraseblock with low erase
+ * counter, and so forth.
  *
- * If the WL unit fails to erase a physical eraseblock, it marks it as bad.
+ * If the WL sub-system fails to erase a physical eraseblock, it marks it as
+ * bad.
  *
- * This unit is also responsible for scrubbing. If a bit-flip is detected in a
- * physical eraseblock, it has to be moved. Technically this is the same as
- * moving it for wear-leveling reasons.
+ * This sub-system is also responsible for scrubbing. If a bit-flip is detected
+ * in a physical eraseblock, it has to be moved. Technically this is the same
+ * as moving it for wear-leveling reasons.
  *
- * As it was said, for the UBI unit all physical eraseblocks are either "free"
- * or "used". Free eraseblock are kept in the @wl->free RB-tree, while used
- * eraseblocks are kept in a set of different RB-trees: @wl->used,
+ * As it was said, for the UBI sub-system all physical eraseblocks are either
+ * "free" or "used". Free eraseblock are kept in the @wl->free RB-tree, while
+ * used eraseblocks are kept in a set of different RB-trees: @wl->used,
  * @wl->prot.pnum, @wl->prot.aec, and @wl->scrub.
  *
  * Note, in this implementation, we keep a small in-RAM object for each physical
  * eraseblock. This is surely not a scalable solution. But it appears to be good
  * enough for moderately large flashes and it is simple. In future, one may
- * re-work this unit and make it more scalable.
+ * re-work this sub-system and make it more scalable.
  *
- * At the moment this unit does not utilize the sequence number, which was
- * introduced relatively recently. But it would be wise to do this because the
- * sequence number of a logical eraseblock characterizes how old is it. For
+ * At the moment this sub-system does not utilize the sequence number, which
+ * was introduced relatively recently. But it would be wise to do this because
+ * the sequence number of a logical eraseblock characterizes how old is it. For
  * example, when we move a PEB with low erase counter, and we need to pick the
  * target PEB, we pick a PEB with the highest EC if our PEB is "old" and we
  * pick target PEB with an average EC if our PEB is not very "old". This is a
- * room for future re-works of the WL unit.
+ * room for future re-works of the WL sub-system.
  *
- * FIXME: looks too complex, should be simplified (later).
+ * Note: the stuff with protection trees looks too complex and is difficult to
+ * understand. Should be fixed.
  */
 
 #include <linux/slab.h>
@@ -92,20 +94,21 @@
 
 /*
  * Maximum difference between two erase counters. If this threshold is
- * exceeded, the WL unit starts moving data from used physical eraseblocks with
- * low erase counter to free physical eraseblocks with high erase counter.
+ * exceeded, the WL sub-system starts moving data from used physical
+ * eraseblocks with low erase counter to free physical eraseblocks with high
+ * erase counter.
  */
 #define UBI_WL_THRESHOLD CONFIG_MTD_UBI_WL_THRESHOLD
 
 /*
- * When a physical eraseblock is moved, the WL unit has to pick the target
+ * When a physical eraseblock is moved, the WL sub-system has to pick the target
  * physical eraseblock to move to. The simplest way would be just to pick the
  * one with the highest erase counter. But in certain workloads this could lead
  * to an unlimited wear of one or few physical eraseblock. Indeed, imagine a
  * situation when the picked physical eraseblock is constantly erased after the
  * data is written to it. So, we have a constant which limits the highest erase
- * counter of the free physical eraseblock to pick. Namely, the WL unit does
- * not pick eraseblocks with erase counter greater then the lowest erase
+ * counter of the free physical eraseblock to pick. Namely, the WL sub-system
+ * does not pick eraseblocks with erase counter greater then the lowest erase
  * counter plus %WL_FREE_MAX_DIFF.
  */
 #define WL_FREE_MAX_DIFF (2*UBI_WL_THRESHOLD)
@@ -123,11 +126,11 @@
  * @abs_ec: the absolute erase counter value when the protection ends
  * @e: the wear-leveling entry of the physical eraseblock under protection
  *
- * When the WL unit returns a physical eraseblock, the physical eraseblock is
- * protected from being moved for some "time". For this reason, the physical
- * eraseblock is not directly moved from the @wl->free tree to the @wl->used
- * tree. There is one more tree in between where this physical eraseblock is
- * temporarily stored (@wl->prot).
+ * When the WL sub-system returns a physical eraseblock, the physical
+ * eraseblock is protected from being moved for some "time". For this reason,
+ * the physical eraseblock is not directly moved from the @wl->free tree to the
+ * @wl->used tree. There is one more tree in between where this physical
+ * eraseblock is temporarily stored (@wl->prot).
  *
  * All this protection stuff is needed because:
  *  o we don't want to move physical eraseblocks just after we have given them
@@ -175,7 +178,6 @@ struct ubi_wl_prot_entry {
  * @list: a link in the list of pending works
  * @func: worker function
  * @priv: private data of the worker function
- *
  * @e: physical eraseblock to erase
  * @torture: if the physical eraseblock has to be tortured
  *
@@ -1136,7 +1138,7 @@ out_ro:
 }
 
 /**
- * ubi_wl_put_peb - return a physical eraseblock to the wear-leveling unit.
+ * ubi_wl_put_peb - return a PEB to the wear-leveling sub-system.
  * @ubi: UBI device description object
  * @pnum: physical eraseblock to return
  * @torture: if this physical eraseblock has to be tortured
@@ -1175,11 +1177,11 @@ retry:
 		/*
 		 * User is putting the physical eraseblock which was selected
 		 * as the target the data is moved to. It may happen if the EBA
-		 * unit already re-mapped the LEB in 'ubi_eba_copy_leb()' but
-		 * the WL unit has not put the PEB to the "used" tree yet, but
-		 * it is about to do this. So we just set a flag which will
-		 * tell the WL worker that the PEB is not needed anymore and
-		 * should be scheduled for erasure.
+		 * sub-system already re-mapped the LEB in 'ubi_eba_copy_leb()'
+		 * but the WL sub-system has not put the PEB to the "used" tree
+		 * yet, but it is about to do this. So we just set a flag which
+		 * will tell the WL worker that the PEB is not needed anymore
+		 * and should be scheduled for erasure.
 		 */
 		dbg_wl("PEB %d is the target of data moving", pnum);
 		ubi_assert(!ubi->move_to_put);
@@ -1425,8 +1427,7 @@ static void cancel_pending(struct ubi_device *ubi)
 }
 
 /**
- * ubi_wl_init_scan - initialize the wear-leveling unit using scanning
- * information.
+ * ubi_wl_init_scan - initialize the WL sub-system using scanning information.
  * @ubi: UBI device description object
  * @si: scanning information
  *
@@ -1583,13 +1584,12 @@ static void protection_trees_destroy(struct ubi_device *ubi)
 }
 
 /**
- * ubi_wl_close - close the wear-leveling unit.
+ * ubi_wl_close - close the wear-leveling sub-system.
  * @ubi: UBI device description object
  */
 void ubi_wl_close(struct ubi_device *ubi)
 {
-	dbg_wl("close the UBI wear-leveling unit");
-
+	dbg_wl("close the WL sub-system");
 	cancel_pending(ubi);
 	protection_trees_destroy(ubi);
 	tree_destroy(&ubi->used);
diff --git a/include/linux/mtd/ubi.h b/include/linux/mtd/ubi.h
index 83302bbbddb4..6316fafe5c2a 100644
--- a/include/linux/mtd/ubi.h
+++ b/include/linux/mtd/ubi.h
@@ -45,13 +45,13 @@ enum {
  * @size: how many physical eraseblocks are reserved for this volume
  * @used_bytes: how many bytes of data this volume contains
  * @used_ebs: how many physical eraseblocks of this volume actually contain any
- * data
+ *            data
  * @vol_type: volume type (%UBI_DYNAMIC_VOLUME or %UBI_STATIC_VOLUME)
  * @corrupted: non-zero if the volume is corrupted (static volumes only)
  * @upd_marker: non-zero if the volume has update marker set
  * @alignment: volume alignment
  * @usable_leb_size: how many bytes are available in logical eraseblocks of
- * this volume
+ *                   this volume
  * @name_len: volume name length
  * @name: volume name
  * @cdev: UBI volume character device major and minor numbers
-- 
cgit v1.2.3